1. Perspectives on Work
1. Work and Household in Chinese Culture
Historical Perspectives
Susan Mann
Classical notions of work in China focused on the household as the basic unit of production. Household members were supposed to be interdependent, with women working “inside” and men “outside” to provide for the needs of the whole. In this classical discourse on work, an individual's labor was subsumed by the collective household economy—usually managed by women, whom McDermott (1990) has called “domestic bursars.” The household so conceived was a kin unit and the basic unit of domestic organization, encompassing both production and reproduction of goods and labor. In late imperial times, the ideal household was a joint family, comprising at least three generations under one roof.[1] To be economically successful, the household constructed an elaborate division of labor according to age and sex, based on each member's understanding of his or her roles and responsibilities. The larger the household and the more elaborate the division of labor, the better the chances for success.
As this chapter shows, the classical ideal and its historical transformations differ in important ways from contemporary understandings of work as a job (gongzuo; see chap. 2). China's earliest classical texts recognized work as one of the defining characteristics of free persons—a mark of respectability and a sign of social status. At the same time, work was construed as a collective, not an individual, activity. Philosophers and statesmen imagined work in the context of household production units laboring to serve their own interests, and in so doing, constructing the social division of labor in society as a whole. Writ large, work was a visible manifestation of a harmonious social and cultural system where “those above” labored with their minds and “those below” labored with their hands. Writ small, work was the everyday activity of every person in every respectable household. Individual subjects did not work for the commonweal, nor were they expected to labor
The household itself emerges as a unit of production in the earliest Chinese classics. Imperial statecraft situated each household in the larger context of an occupational division of labor (the “four classes” of people, namely: scholars, farmers, artisans, and merchants). Thus the economy was construed as a complexly interdependent layering of specialized occupational groups, each composed of the basic building blocks organizing all labor: commoner households. To illustrate this in historical perspective, this chapter is divided into three parts. The first explains the earliest philosophical understanding of households as production units within a complex division of labor. The second shows how these early classical notions were implemented in the statecraft and political theory of the early empire (through the thirteenth century). The third part describes the late imperial system of production that flourished on the eve of the twentieth-century revolutions.
CLASSICAL MODELS OF THE HOUSEHOLD AS A UNIT OF PRODUCTION
The organization of work in commoner households is prefigured in the Chinese classics. Early Confucian philosophers and statesmen (551–c. 233 B.C.E. ) understood that people labored primarily to eat—in other words, to feed and clothe themselves—but they never conceptualized labor as an individual human activity. Instead, they talked about work in the context of a division of labor: as part of a household production system, as part of an occupationally differentiated society, and as part of an economy of exchange. Although these same philosophers and statesmen were concerned about mobilizing labor for the state, almost all of them argued that a skillful ruler should mobilize the labor of his subjects only for projects dedicated to the common good. Similarly, rulers were cautioned to exercise restraint when making demands on people's labor. Every household had to be free to work so that it could meet its own needs before taxes.[3]
The same philosophers and statesmen praised human qualities that made good workers: diligence, frugality, and skill. Although there was no classical term for efficiency, strategies to increase labor efficiency continually focused on the division of labor—among the people by occupational groups, and within the household production unit by gender. Classical writers also stressed the need for accessible markets to encourage local and regional
Labor defined and displayed social status, and classical writers deplored its absence (as leisure or idleness). Work meant not simply labor, moreover, but a respectable calling. In the third century B.C.E., Mencius (372–289?) drew his famous distinction between “those who labor with their minds” and “those who labor with their hands”:
Xu Xing, in search of benevolent rule, travels from the state of Chu to Tang, where he pledges his service to Duke Wen. After some time in residence in the Duke's kingdom, Xu Xing is heard to complain that the Duke is not the benevolent ruler he purports to be, because a benevolent ruler is one who “tills the fields and eats together with the people” (yu min bing geng er shi). Plainly, Duke Wen was an oppressor and not a benevolent ruler at all. After all, he maintained granaries, treasuries, and arsenals, all financed from levies he exacted from the people. When a visitor asks Mencius about Xu Xing's complaints, Mencius responds with a series of questions of his own. Does Xu Xing make his own clothes? Does he forge his own tools? Does he fire his own pots? No, concedes the interlocutor, Xu Xing exchanges his grain for these things. Well, says Mencius, there you have it: “Great men have their proper business, and little men have their proper business. … Hence, there is the saying, ‘Some labor with their minds, and some labor with their strength. Those who labor with their minds govern others; those who labor with their strength are governed by others. Those who are governed by others support them; those who govern others are supported by them.’ This is a principle universally recognized.” (Paraphrased from Mencius 3A.4/4–6; translated in Mencius 1960, 249–50)
Noticing that he's made a leap in logic, Mencius goes on to elaborate by noting that the great sage-king Yu could not possibly have gone out to cultivate crops while he was in charge of clearing fields, diverting floods, and opening up land to cultivation for his people. The same was true of the mythic Minister of Agriculture, who taught the people to sow and reap and cultivate the five grains—how could he have done that while working as a farmer? Duke Wen's work as a ruler, Mencius concludes, is not to farm, but to protect his people and provide for their welfare by storing weapons and food, and by filling a treasury to pay for it all.
Mencius and Confucius (551–479 B.C.E. ) return often to this concern about the division of labor between those who rule and those who are ruled. Confucius, like Mencius, understood work to include moral and ritual activities as well as manual or physical labor. He never saw it as an isolated activity, but rather as something done for someone else. He used the term “work” (lao) to talk about labors performed by a filial son for his parents, for example (Analects II:8 and IV:18; see Confucius 1938, 89 and 1979, 74,
Notice how this early work ethic brings to the fore the people's livelihood and checks the state's power to extract labor. Rulers trying to make the people work hard had to be careful that their own demands would not interfere with every household's need to provide for itself. Even Xun Zi (fl. 298–238 B.C.E. ), who is sometimes considered the earliest proponent of “Legalist” thinking (which stressed the absolute power of the state and its need to control labor and production), agreed with Mencius that taxes and labor service extracted from the people should be minimized, for the same reasons (see the discussion in Hsiao 1979, 187).
The classical work ethic, on occasion, even spoofed or expressed skepticism about the value of scholarly or mental work, as opposed to manual labor. The Analects include a story in which one of Confucius' disciples appears as an effete intellectual who is humbled in the presence of an honest laboring farmer. The story concerns Zilu, who loses his way one day and encounters an old man weeding in the fields. When Zilu asks the old man if he has seen Confucius, the reply is dismissive: “Your four limbs are unaccustomed to toil; you cannot distinguish the five kinds of grain (siti buqin, wuliang bufen) —who is your master?” (Analects XVIII.7; see Confucius 1960, 335). Mencius, too, squirmed when he confronted the problem of the scholar's labor and its value, as we see in his conversation with Peng Geng. When Peng Geng asks Mencius about earning a living, he inquires whether the “superior man in his practice of principles” is earning a living just the same way the carpenter and the wheelwright do. What, in other words, is the superior man's purpose? Is it simply to get by like everyone else? Mencius replies, somewhat testily: “What have you to do with his purpose? He is of
Mencius's interest in the division of labor and the mutual obligations of those above and those below is also reflected in his keen interest in markets. No household was expected to be completely self-sufficient, and early statesmen understood that rulers must provide a market economy to stimulate the exchange of goods and services. Mencius himself viewed the orderly and regular exchange of grain for commodities as a hallmark of good government (Mencius IIA.5/2; see Mencius 1960, 199). He was also not concerned about merchant greed or about the trader's quest for profit, but was more worried about the marketplace as a lure for idlers. Like the most famous classical economic theorist, Guan Zi (d. 645 B.C.E. ), Mencius believed it was the ruler's job to balance the interests of farmers against those of merchants so that all of the “four kinds of people” could, in Guan Zi's words, “exchange skills and trade the products of their work, so that the profits of their year-long labors will not allow any one of them to gain unfair advantage over the others. Thereby the people's labors will be combined and they will all gain equitable benefits.”[4]
These early ideas about work also specified gender roles. The most vaunted kinds of work in classical texts—scholarship and farming—excluded women.[5] Significantly, too, men who ruled might labor with their minds, but all women, regardless of status, were expected to perform manual labor, especially spinning, weaving, and needlework. Womanly work (nü gong) was one of the “four virtues” of a respectable woman, first canonized by Ban Zhao (d. 116 C.E. ) in her book, Nü jie (Instructions for women). Early texts make it clear that womanly work means spinning and weaving, and that sericulture (sang) was the counterpart of agriculture (nong) —the former, women's work “inside,” the latter, men's work “outside.” The ideal gender division of labor, with men working in the fields and women weaving in the house, was also described by Mencius, who urged his ruler advisees to exhort every household with five mu of land to diversify production, planting mulberry in the space beneath the walls around family compounds so that women could rear silkworms and the elderly could be clothed in silk (Mencius VIIA.22/2; see Mencius 1960, 461).
In genteel families, a woman working at her loom was the complement not of the farmer but of the scholar poring over his books. The trope of the weaving woman and the studying man may date to a story about Mencius and his mother. Early illustrated copies of the book, Lienü zhuan (Biographies of women) by Liu Xiang (77–6 B.C.E. ), show Meng Mu brandishing her scissors in front of her wayward son. By slashing the cloth on her loom, she shocks Mencius into realizing the cost of his wasted talent. Chastened, Mencius reforms—to become China's greatest philosopher (Liu Xiang 1983, 15–16).
In sum, philosophers and statesmen writing in the classical age developed clear ideas about what kinds of work were legitimate and how the division of labor in society should be organized. They viewed hard work as a natural activity of ordinary commoners striving to feed and clothe their families, care for elders, nurture young, and improve their lives—an activity the government should nurture, encourage, and protect. Finally, they assumed a division of labor in society between the four occupations (scholars, farmers, artisans, merchants), whose mutual dependence would be expedited by trade, and a division of labor within the household between men and women, whose complementary functions would support tax-paying families.
THE RISE OF INTENSIVE HOUSEHOLD PRODUCTION BEFORE THE THIRTEENTH CENTURY
Statecraft writings of the early empire focus on the broad division of labor in society as a whole, rather than on individual households. The oldest economic treatise in the Chinese historical record, the “Treatise on Food and Money” in the History of the Former Han Dynasty, explains that the “four kinds of people” each have their proper occupations (simin you ye): “Those who study to attain their status we call shi (scholars); those who cultivate the soil to grow grain we call nong (farmers); those who by their cleverness produce implements we call gong (artisans); and those who circulate wealth and trade in commodities we call shang (merchants)” (Ban Gu 1983, 532). The hierarchical order of this list echoes Mencius (e.g., Ch'ü 1957, 245–48). The superior form of labor was “mental” labor (study), followed by “manual” labor (plowing, planting, and harvesting; and manufacturing—including tool making, carpentry, woodcarving, ceramics, spinning, weaving, and embroidery). Though trade and business came last, merchants were not always at the bottom of this hierarchy. The Guliang zhuan (Duke Cheng, first year, 589 B.C.E. ) lists merchants right after scholars but before peasants and artisans. Moreover, pejorative judgments about the value of merchants are hard to find in discussions about the four kinds of work. Instead, merchants are simply supposed to limit their profit taking; or, better still, the government is supposed to intervene to prevent undue disparities of wealth between rich and poor. Fierce competition was the norm, as we learn from resigned complaints like the following, written by the Song scholar Ye Shi (1150–1223): “The four kinds of people and the hundred arts flourish by day and subside by night, each vying with all their might, each pursuing their own interests, fearless in the face of danger, insatiable in the face of abundance” (Ye 1961, 1:164). Some early texts propose a hierarchy of occupations. For instance, criticism of merchants and their work begins in
Regardless of occupational hierarchies, work as an essential activity of free persons separated the “four kinds of people” doing respectable work from slaves. During the Han period, slaves labored in mining, manufacturing, farming, handicraft work (including spinning and weaving), domestic and military service, and even entertainment (Ch'ü 1972, 135–59). The Han government also employed slaves, while frequently accusing them of idleness and sloth and of living off the labor and tax revenues supplied by ordinary commoners.[7] The government made clear its preference for laborers conscripted from free commoner households, or even (anticipating a contemporary debate) convicted prisoners.[8] As a result, conscript laborers—or even convicts—and not slaves made up the bulk of the public works labor force, with soldiers playing an important role in peacetime.[9] And despite the obvious importance of corvée labor to the government's vast mining operations and public works projects—including water control, post stations, canal transport, wall construction, and other state-funded construction and service[10]—conscripted labor had to be acquired with minimum cost to individual households whose members were respectably employed. Han political theory, following classical teachings, stressed the need to ease the burden on farm households so that they could perform labor at a level sufficient to support themselves. Dong Zhongshu (179?–104? B.C.E. ) echoed his predecessors when, in his critique of the problems of the first Qin empire (221–206 B.C.E. ), he called on the government to “reduce taxes and levies and lessen labor services, in order to give freer scope to the people's energies” (Hsiao 1979, 502).
A final form of labor organization during the early centuries of empire was monastic production. Like conscript and slave labor, monastic labor was marginalized in policies and political theories about work. Most schools of Buddhism disdained manual labor in any case, especially agricultural labor, which was forbidden to monks by religious law. Monks relied on tenant farmers and on novices for agricultural labor on the Buddhist estates that
In … puqing, all should exert equal effort regardless of whether the task is important or unimportant. No one should sit quietly and go contrary to the wishes of the community. While performing his duties, one should not indulge in ridicule or laughter, or boast about one's talents or ability. Rather, one should concentrate his mind on the Dao, and perform whatever is required by the community. After the task is completed, then one should return to the meditation hall and remain silent as before. One should transcend the two aspects of activity and nonactivity. Thus, though one has worked all day, he has not worked at all.[12]
Devout Chan Buddhists were expected to work without attachment to the action itself or to the rewards it would bring. Nevertheless, the high productivity of monastic estates of all kinds created a vast pool of wealth, much of it tax-exempt, which threatened the state's control over labor and resources. The mid-ninth-century campaigns that destroyed the Buddhist monastic establishment put a permanent end to large-scale monastic production (Dalby 1979, 666–67).
With slave and conscript labor marginalized and monastic production suppressed, farm families and landholding households emerged as the model workers of the early empire. The Mencian household production unit had become more than a philosophical ideal. Records suggest that throughout this early period, most Chinese farm labor was performed by individual households, as tenants or freeholders. In Nishijima Sadao's words:
[T]he typical rural community during Han was the hamlet, consisting in theory of a hundred families, all of which owned small amounts of land. … It must be noted that the rise of great landholdings during Han did not necessarily imply the development of large-scale farming, except in the few cases where slaves were employed to work estates. Tenants of these landowners cultivated their holdings on an individual and small-scale basis and this, due largely to the lack of sufficient slave labor and the intensive nature of farming, continued to be an important feature of Chinese agriculture. (Nishijima 1986, 559)
Similarly, in his study of Han agrarian technology, Hsü concluded that
intensive care of each individual plant had been brought almost to the level of gardening. … Such farming demanded a concentration of labor and was associated with smaller farms. Also the farmer's incentive was crucial to guarantee the thoroughness of the field work, which made extensive use of slaves or contract labor unreliable. All these factors help to explain the maintenance of independent farmers in the Han period and the preference of large landholders for tenancy rather than slavery. (Hsü 1980, 127)
Hsü (127–128) emphasized that even as early as the Han, the most successful farming depended on the combined efforts of all household members, including women and children.
HOUSEHOLD PRODUCTION, SONG (960–1279) THROUGH QING (1644–1911)
The contemporary scholar Kang Chao, while agreeing that the farm household was the most important labor unit in the history of Chinese agriculture (1986, 194), points out that the intensive farming practices developed in late imperial times were probably not “necessary” or even “widespread” in China until the twelfth century, despite the availability of techniques of intensive farming since at least Han times. Chao (1986) locates the shift from extensive to intensive farming in the Song period, identifying three of its most important characteristics: multiple cropping, reclamation of fallow lands, and intensified fertilization along with higher-yield strains and crops. Champa rice, introduced into the Lower Yangzi by Emperor Zhenzong in the early eleventh century, triggered the shift to intensive cultivation, ushering in “a new era of agrarian history in China” by promising “a higher total output by absorbing more labor” (Chao 1986, 200–201). By the end of the Song period—the thirteenth century—the household, whether gentry or peasant, had emerged as the basic unit of agricultural and protoin dustrial production in the Chinese economy.
The Song dynasty was a crucial period in the development of a discourse on work in households. Ebrey has pointed out that instruction books like Yuan Cai's, printed for gentry families seeking to take advantage of the Song's new economic opportunities, paid unusual attention to work:
What most distinguished [Yuan Cai] … was his unconcealed admiration for those who worked honestly and diligently to ensure their families' security. Whatever the importance of ethical strengths, he did not think one should ignore the need for productive property [chan] or wealth [cai], or an occupation or other heritage that brought in income [ye]. (Ebrey 1984, 78)
Ebrey (1984, 121) stresses that Song gentry families conceived of themselves as a “unit of political economy, not simply a group of relatives,” whose survival depended upon the successful use of property and, one might add, on the successful management of labor. Yuan Cai was clear about the importance of an “occupation” (ye) for everyone, regardless of status:
People with no occupation, and those with occupations who are too fond of leisure to be willing to work, will get used to indecent activities if they come from rich families and will become beggars if they come from poor families. (Ebrey 1984, 268)
Yuan offered his own assessment of the most desirable kinds of work for men in respectable gentry families in this order: scholarship (studying for the exams and winning office or, if that fails, teaching), clerkship (handling documents), tutoring children, teaching medicine, practicing Buddhism or Daoism, “gardening” (genteel cultivation), and commerce. All these, he suggested, could “provide support for your family without bringing shame to your ancestors” (Ebrey 1984, 267). At the same time, and in keeping with the philosophical views of early Confucians, Yuan Cai was unapologetic about dodging corvée labor service. The main problem he saw with corvée dodging, in fact, was that if you failed to pull it off, you were caught and punished (Ebrey 1984, 309–10).
After the thirteenth century, households that had intensified production by cultivating double-cropped rice found incentives to increase efficiency still further, as cotton cultivation spread from the southeast coast to the Lower Yangzi region. The labor required to prepare cotton fibers for spinning—including ginning, fluffing, straightening, and sorting—was easily organized in the context of a joint family unit, with its attendant servants and multigenerational workforce. Like hemp and silk, cotton absorbed the labor of women, who could work respectably in the home, where we see them in paintings and drawings from the Song period onward.
Meanwhile, the crowded commercial cities of the Song period (960–1279) and women's prominence in the commercial and service sectors provoked a strident moral discourse on cloistering women that gained wide currency in Song gentry families. A study by Quan Hansheng (1935), which explores expanding work opportunities for Song women, shows women working in all aspects of the urban economy: as proprietors of teashops, fancy restaurants, and herbal medicine stores; as peddlers hawking candy or tea, or selling Buddhist paraphernalia; and as tailors and weavers. Nuns sell jewelry, clothing, caps, hair ornaments, and other kinds of female adornment, all made by hand, at temple fairs. On the farm, women pick mulberry leaves and rear silkworms; in the marketplace, they tell stories, juggle, do magic tricks and acrobatics, sing, dance, and perform opera. Song women worked at all kinds of menial labor as maids—cooking, hauling water, lighting
On the eve of the late imperial period, then, intensive household farming had become the norm in China's most productive regions. During this same period, the boundaries of work also hardened to celebrate and enforce the confinement of female labor inside the home.[13]
THE HOUSEHOLD AS A UNIT OF PRODUCTION IN L ATE IMPERIAL TIMES: THE MING (1368–1644) AND QING (1644–1911) DYNASTIES
The late imperial fisc depended on land and labor taxes collected not from individuals but from registered households. Late Ming and Qing rulers promoted what we would call Confucian family values to support household production through campaigns targeting ordinary commoners. Universal marriage and childbearing (especially bearing sons to continue the patriline), wifely fidelity, filial piety, and ancestor worship were touted and rewarded by the state through government-endowed shrines, imperial certificates of merit, and even cash prizes honoring chaste widows and filial sons. Through laws enforcing partible inheritance, through policies protecting tenants and freeholders from excessive taxation, and through legislation limiting the purchase and sale of slaves, late imperial governments also provided an economic environment that favored free holding peasant households and encouraged them to accumulate resources through labor.
Largely because of these supportive central government policies, the peasant household was the dominant unit of production and reproduction by the seventeenth century, as serfdom and “serf like tenancy” all but disappeared from China's core regions. One hallmark of the success of late imperial policies promoting household labor is the attention paid to “women's work” (nü gong) in the statecraft writings of the late imperial period. Women working respectably at home, as spinners and weavers, began producing for the commercial economy as cotton cultivation spread after the fourteenth century. By the late Ming period (after 1550), cotton's low cost and versatile
In an economy that celebrated workers in households, unattached hired workers—especially single men—were suspect and carefully screened. Landlords looking for hired labor tell us clearly the sorts of qualities they prized—and what they sometimes had to settle for. Zhang Lixiang, a seventeenth-century landlord, advised: “In general, the best is the docile and diligent worker; second best, the skilled and alert one; next, the incompetent but honest one; worst is the one who is cunning, deceitful, talkative, and lazy.” An employer who wanted to keep a good hired worker had to act in loco parentis, providing not only a good wage but also adequate supplies of wine and food (see Wiens 1980, 20, 21, citing observations by the scholar Zhang Lixiang [1611–74]). Meanwhile, local officials strove to anchor every able-bodied person securely in a household economy. The imperial government reserved special praise for officials who expanded opportunities for work in their areas and who looked out for the needs of ordinary peasant householders (Brokaw 1991, 198–99). Officials were expected to encourage and introduce new modes of production, new seed strains, and new technologies whenever possible to improve household productivity. This effort sometimes worked at cross purposes with the government's need to mobilize corvée labor for public works, as peasants loath to divert labor from their own households to the government, sought tax protection from wealthy gentry families, creating a continuing labor shortage for state projects.[14] Dodging corvée service in the interests of one's own household was different from laziness, as Yuan Cai noted. To be lazy was often defined not simply as idleness, but as engaging in unseemly work— and was the mark of a lowly person. Thus women who “neither spin nor weave,” including prostitutes and other women in subethnic groups of “mean people” (jianmin), were stigmatized as pariahs.[15]
In these and other rhetorical signs from late imperial texts, we see how normative ideas of respectable work were firmly lodged in the context of the household. From the imperial court down to the commoner household,
If the Chinese government and Chinese political theory promoted the household as the basic unit of productive labor, and other factors created a hospitable environment for household-based production,[17] we still must ask what made householders so responsive to government policies. Folk sayings attest to the extraordinary work ethic of Chinese farmers. Arkush (1984), surveying hundreds of proverbs from North China commonly in use before the twentieth century, observes that most stress the importance of and returns from hard work. Besides hard work, he notes, other moral qualities are essential: frugality; skill in selecting proper inputs of fertilizer, seed, and so forth; ingenuity to tap all available sources of income, including sideline industries; knowledge (of farming, in particular); self-reliance, or cooperation within a small group; and anticipation of income. A few sayings point to the significance of the household as the production unit. Buck (1937, 307), for instance, recording adages in praise of human farm labor, found one that links it to filial piety: “The filial and obedient couple will be fortunate, the well-tended fields will bear grain” (xiao shun ye niang you fu qin li tianchang you gu). All of these values were recognized in early classical statecraft and philosophy, and all can be traced through records of imperial policy and family practice from earliest times.
Western observers of China in the early twentieth century echoed the sentiments of China's own folk wisdom: Chinese people work hard, even in the face of very low returns, and they are constantly on the lookout for more productive outlets for surplus labor.[18] As Tawney put it, “The first sensation of a visitor to a Chinese city is one of suffocation beneath a torrent of human beings, straining at manual labour or clamouring to be given it” (1932, 119).[19] Studying farm labor at about the same time, Buck remarked: “The amount of man labor in China is almost unlimited and one of the great problems is the discovery of enough productive work to keep this vast human army profitably employed” (1937, 289).
But hard work alone does not explain the success of farm families, as Buck emphasized. The reason why profitable farm employment had been expanded to the maximum was its efficient division of labor. By Buck's estimates, farm work per se occupied the full time of only about two-thirds of the farm population, with sideline industries occupying one-eighth and a combination of farming and sideline work employing another one-fifth. These sideline occupations (called “subsidiary work” in Buck's study) were
This is the same women's work vaunted in late imperial statecraft, which emphasized the need for skill not only in performing tasks but in creating opportunities for productive labor wherever they could be found.[20] At the women's festival on Double Seven (the seventh night of the seventh month), when women and girls used to “beg for skill” (qi qiao) from the stellar Weaving Maid, the goal of every young petitioner was more than skill with the needle. Skill prepared her for entry into a productive household as a valued member. Not only would a talented seamstress make a better marriage; she would also please her mother-in-law, ensuring a less stressful life in the household into which she married.[21]
The importance of skill inculcated through training and disciplined by high motivation, the high returns from intensified labor in the rural workforce, and the centrality of women's work in a gender division of labor where women work “inside” while men folk labor “outside”—all were key aspects of a household production system supported and protected by China's earliest rulers and statesmen. In this production system, workers— male and female—were motivated by complex incentives far more compelling than an individual wage. In late imperial China, political, economic, social, and moral incentives all rewarded couples who kept their married sons together and trained the next generation to work hard in a household economy.
WORK IN CHINESE HOUSEHOLDS AT THE TURN OF THE TWENTIETH CENTURY
What implications does this historic legacy of household labor have for modern China? Recently the traditional gender division of labor has been the subject of attention, as scholars speculate that norms confining women's work to the household reduced the supply of female workers available for China's factories during early industrialization. Huang (P.C.C. 1990, 111), for example, has argued that the “cultural constraints against women venturing outside the home, plus the logistical difficulties of managing female labor that came with those constraints” created a crucial shortage of female labor for light industry, slowing the pace of industrialization in the early twentieth century.[22]Hershatter (1986, 55–56) found that in Tianjin, women were only 9.14 percent of the cotton mill workforce in 1929 (as compared with over 70 percent in Shanghai mills), a factor she attributed partly to the role of women in the rural economy of the North, which confined women almost exclusively to work inside the home. Honig (1986, 64–65), by contrast,
The early Communist government's attempts to reconfigure ideas about work overrode these historical household-based motivational systems. In the Maoist era, work was organized collectively or communally, partly by appealing to classical notions of sharing returns from labor and stressing the mutual obligations between “those above” and “those below”—managers and workers, or technicians and manual laborers. But collective labor was never wholly successful. It violated long-held assumptions about the boundaries of work (women “inside,” men “outside”). It closed down the marketing systems on which households had depended to exchange their surpluses. It overrode the specialization of crops and tasks that maximized the diverse capacities of regional and human resources. And it destroyed the motivational system oriented backward toward the ancestors and forward toward descendants that for centuries had fueled China's renowned work ethic.
The astounding success of the return to household production units after 1978 during the post-Mao reform era at first glance appeared to confirm that household production units still motivated workers better than collectives did. The household responsibility system in China's countryside tapped ideas about gender, household, and work that have a long and remarkably coherent history. Ironically, the conviction that work inside the home is women's work also seems to have survived intact from late imperial times into the contemporary reform era—pretty much unscathed by egalitarian campaigns like the Great Leap Forward and the Cultural Revolution. The boundaries of the household as a production unit appear to be the strongest and the most difficult to penetrate of all the structures of work inherited from the past.
Despite the resilience of the household as a unit of production, however, the modern concept of work as gongzuo (a job) has continued to erode classical boundaries of gender, household, and work, drawing women outside the home into a legitimate public workspace where modern “jobs” abound. In the 1990s, with increased incentives for individuals to pursue their own career goals without regard for their family obligations, the boundaries promised to erode still further, as they have in Taiwan, Hong Kong, and overseas Chinese communities (Whyte 1996). Even less durable, it appears, are other historical ideas about work in China: the long-standing critique of
NOTES
1. See Cohen (1976, 57–85) on the collective character of the family economy.
2. See Harrell (1985). The problem with this reification of the household, of course, is that it ignores the ambiguous relationship of women—especially unmarried women—to the patriline. See, for example, Greenhalgh (1985). Exploration of this topic lies beyond the scope of this chapter, which focuses on the boundaries of the household itself.
3. The so-called “Legalists” took a different view, arguing that the ruler should command the labor of his subjects solely for the purpose of increasing his own power. This line of thinking was discredited in mainstream Chinese political theory, which (rhetorically, at least) deferred to Confucian “benevolence.”
4. Translation, slightly adapted, from Hsiao (1979, 357).
5. Of course women always worked in farming and, especially after the seventeenth century, they were visibly engaged in scholarly activities, but the imperial government's policy and political theory reserved recognition in both arenas for men alone. Farm women were honored for sericulture, spinning, weaving, and needlework, but never for laboring in the fields.
6. Ch'ü T'ung-tsu (1972, 117, 118) comments that “In pre-Ch'in times we do not find any historical evidence to indicate that there was discrimination against merchants. On the contrary, it seems that their social and political status was rather high. … Merchants in Ch'ing times also enjoyed a superior social status.” He dates the earliest discrimination against merchants to a law dated 214 B.C.E.
7. See Ch'ü (1972, 144–145). He quotes the following comment from the Han dynasty “Debates on Salt and Iron” (Yantie lun): “The common people are not free from work from morning to evening, whereas (government) male and female slaves idle about with folded hands” (145).
8. Scholars have generally concluded that the hardest government labor was reserved for convicts and conscripted laborers, not slaves (Ch'ü 1972, 145). This view is shared by Lien-sheng Yang (1969, 29) and C. Martin Wilbur (1943). In fact, Yang emphasizes that “Altogether, government and private slaves did not play any significant role in public works, even in those relatively early periods of Chinese history when slaves are supposed to have been fairly numerous” (210). See also Watson 1980a, b.
9. L. S. Yang (1969, 210) notes that idle soldiers were judged a waste of government resources.
10. Examples of the mobilization of forced labor under the early empire include the 700,000 convicts employed to build the imperial mausoleum and palace of the First Emperor (221–209 B.C.E. ), and the 146,000 men and women drafted to build the city of Chang'an in the early Han (206 B.C.E. –220 C.E. ) (L. S. Yang 1969, 202). Massive water conservancy projects on the Yellow River employed hundreds of thousands of conscripts in Han and Sui (599–618) times (203). Construction and repairs
11. Literally, “everyone is invited,” translated by Ch'en (1973, 148) as “collective participation.”
12. See Ch'en (1973, 148–50). Translation slightly adapted from Ch'en (150). Ch'en (151) views this Chan teaching as an “accommodation to the prevailing Chinese work ethic.”
13. For differing interpretations of the significance of women's textile production in peasant households during late imperial times, see Mann (1997, 143–77) and Bray (1997, 173–272). Both recognize that women's work in the household economy, especially spinning and weaving, was valorized in the late Ming and Qing periods by government policies celebrating women's work in the home. Bray sees this as a shift that marginalized female productive labor by removing it from the higher skilled, higher paying shops in towns and cities dominated by male labor. Mann notes that in many areas the spread of spinning and weaving technologies in peasant households increased female labor productivity. Both agree that norms confining female labor to the home were strikingly effective in removing female labor from the fields in China (as compared with other rice economies, for example in Japan and Southeast Asia).
14. According to Huang Liuhong, the late seventeenth-century magistrate whose advice book became a classic reference for generations of officials, a county official might need conscript labor for the following jobs: repairing the city walls, working on the Yellow River or Grand Canal water conservancies (including damming, dredging, diking, planting willow trees on river banks), towing government barges, repairing guest houses maintained for visiting dignitaries and officials, and so on. (See Huang Liu-hung 1984, 227–33).
15. Cole (1986, 65–72) describes the duomin (“lazy people”) of Shaoxing, a local pariah population whose work marked them as polluted. Duo (“lazy”) is a homophone for duo (“fallen”). Courtesans and prostitutes likewise were scorned as pariahs because they did not do “womanly work” (Mann 1997, 121–142). See also Honig's (1992) study of Subei people in Shanghai, which shows that a reputation for clumsiness, slowness to learn, or lack of skill all barred Subei workers from respectable kinds of employment.
16. Buck (1937, 292). Buck acknowledged that these figures varied by region and that the largest proportion of farm labor done by women was in the double cropping rice areas. He also noticed that hired farm labor was overwhelmingly male.
17. Scholars have noted several features of late imperial society that made it especially conducive to household-based production. These include the rice economy and its capacity to reward intensive labor, a flourishing marketing system, a fluid stratification system, open marriage markets, and a joint family system. On the rice economy, see Bray's (1986) analysis of what she calls “skill-oriented” technologies;
18. Arthur Smith (1894) devoted an entire chapter of his book on Chinese characteristics to praise of “industry,” conceding that “there can be little doubt that casual travelers, and residents of the longest standing, will agree in a profound conviction of the diligence of the Chinese people” (27). King (1973, 16), who was far more impressed with the moral fiber of East Asian farmers than was Smith, professed amazement at “the magnitude of the returns they are getting from their fields, and … at the amount of efficient human labour cheerfully given for a daily wage of five cents and their food, or for fifteen cents, United States currency, without food.”
19. Tawney's list of work includes coal and iron ore mining and iron smelting (“nearly one-half” of the pig-iron produced, he estimated, was made in charcoal furnaces with bellows worked by hand or water power); cotton spinning (some in factories, but some, along with most weaving, still done at home or in small workshops); and finally metalworking, potting, tile-making, building, carpentry, furniture making, painting, shoemaking, hat making, tailoring, tanning, woodcarving, lacquer making, silk reeling and weaving, woolen weaving, tapestry making, rope making, and myriad artisan crafts producing housewares, ornaments, jewelry, and “artistic products”—all of the latter “still much what they were five centuries ago” (1932, 111).
20. Bray (1986) calls rice economies skill-oriented (rather than mechanical) technologies because they require ever-higher levels of skill to produce the rising productivity necessary for supporting more people on scarce land. Bray (155) stresses Thomas Smith's (1959) point that, unlike Western-derived development strategies, the Japanese model “is based on improvement in the application of human skills rather than the substitution of machinery for labour, and requires a low level of capital investment.”
21. As late as 1988, female informants in my interviews in Shanghai recalled vividly (and sometimes painfully) the humiliation of failing to pass their mother-in law's test of their skill in needlework, the first rite de passage of the young bride after her wedding night.
22. Schneider's work on Sicily (1985, 81) posits in a similar vein that “cultural patterns making female labor unavailable for income-producing activities have impeded agrarian transformation and economic development.” Describing the early factory labor force in British India, Morris (1983, 644) remarked that “among the less expected features of the factory labour force—at least, as compared with the early experience of other nations—were the relatively small proportions of female and child labour and the stability of those proportions.” The total proportion of women and children combined never rose above 22 percent or fell below 20 percent between 1892 and 1928. See also Gadgil (1965) and Morris (1965, 65–69). These figures are almost exactly in line with China's, with a similar family system and comparable constraints on women going out to work.
2. Re-Drawing the Boundaries of Work
Views on the Meaning of Work (Gongzuo)
Gail E. Henderson, Barbara Entwisle, Li Ying, Yang Mingliang, Xu Siyuan, and Zhai Fengying
Work is an ambiguous concept everywhere, but perhaps especially so in contemporary China, where economic reforms implemented in the 1980s fundamentally changed the landscape of employment opportunities. Before the reforms, state and collective-run firms accounted for virtually all urban employment. This was still largely true in 1990, when 94 percent of urban employees worked in the state or collective sectors (see table 2.1). The state and collective sectors continue to dominate employment in urban areas, but the degree to which they do so is beginning to slip, together accounting for 83 percent of urban employees in 1995. Employment in the private sector, previously at zero, has mushroomed. By 1995, foreign or jointly run enterprises accounted for 5 percent of the urban labor force, domestic private enterprise (siying) 3 percent, and individually or household run businesses (geti) another 9 percent. Distinctions between what does and does not count as work have likely become muddied with the weakening of employment guarantees, the growth of contract and temporary labor, and the reemergence of the household as a focus of economic activity.
Changes have been even more pronounced in rural areas. Within agriculture, the “household responsibility system” replaced collective agriculture, shifting agricultural activities from the collective to the private sector and under household control. Privatization of sideline and other domestic activities also redefined the boundaries between “inside” and “outside” work. At the same time, industrialization of the country side led over 200 million people to leave agriculture for work in industry, trade, and services during the first 15 years of reform. Table 2.1 shows that the number of employees in TVEs (township-and village-run enterprises, xiangzhen qiye ) doubled between 1985 and 1995, and by the latter date accounted for 29 percent of workers in rural areas. Employment in private enterprises (siying qiye) and
| Percentatge Distibution of Employees by Employment Sector | ||||||
|---|---|---|---|---|---|---|
| Total Millions | State | Collective | Other[a] | Private | Individual | |
| SOURCE: State Statistical Bureau, 1996. Zhongguo tongji nianjian 1995 (Statistical yearbook of China 1995). Beijing: Statistical Publishing House of China, pp. 90–91. | ||||||
[a] a Includes jointly owned economic units, share-holding economic units, foreign funded economic units, and economic units funded by Chinese from Hong Kong, Macao, and Taiwan | ||||||
[b] b Township and village-run enterprises. | ||||||
| Total Urban | ||||||
| 1985 | 128.08 | 70.2 | 26.0 | 0.3 | 3.5 | |
| 1990 | 147.30 | 70.2 | 24.1 | 1.1 | 0.4 | 4.2 |
| 1995 | 173.46 | 64.9 | 18.1 | 5.2 | 2.8 | 9.0 |
| Hubei Urban | ||||||
| 1995 | 8.97 | 65.0 | 15.5 | 4.5 | 2.2 | 13.3 |
| Percentage Distribution of Workers by Employment Sector | ||||||
| Total Millions | TVEs[b] | Private | Individual | |||
| Total Rural | ||||||
| 1985 | 370.65 | 18.8 | —— | —— | ||
| 1990 | 420.10 | 22.1 | 0.3 | 3.5 | ||
| 1995 | 450.42 | 28.6 | 1.0 | 6.8 | ||
| Hubei Rural | ||||||
| 1995 | 18.10 | 36.7 | 1.2 | 12.7 | ||
Within the context of these changes, our goal is to understand how boundaries are drawn between what counts as work and what does not. Our approach is “emic”—to ask people themselves what they think. Because economic transformation has been most dramatic in the countryside, we direct our attention to that context, drawing data from 12 focus group interviews conducted in two villages and two towns. Focus group interviews are guided group discussions on prespecified topics (Stewart and Shamdasani 1990). Developed initially as a marketing tool, focus group interviews have
The particular concept we explore is gongzuo, which can be translated as job or work.[1] We chose this concept because it is frequently used in survey data collection. The China Health and Nutrition Survey (CHNS), for example, a data source used by several contributors to this volume (chaps. 8 and 15), asks whether each person in the household “has gongzuo”; occupation, employment sector, and other information is collected for those household members who do. Gongzuo encapsulates a modern view of work focused on individual activity outside of household contexts (see chap. 1). During the Maoist era (1949–1976), the term gongzuo was restricted to work for wages in the formal sector in a work unit (danwei), for which urban household registration was required. This type of work, especially in a state run work unit, topped the job status hierarchy at that time (Walder 1986), perhaps corresponding to the idea of “a real job.” As we will show, there is considerable ambiguity about the meaning of gongzuo now.
FOCUS GROUP INTERVIEWS: DESIGN
In the focus group interviews, a vignette was used to generate discussion about work (gongzuo). Participants were told that a long-lost relative of the Feng family, Mrs. Chen, was coming for a visit and “we would like you to help us describe their activities to Mrs. Chen.” Then the activities of each of three Feng brothers and their wives were described, followed by two questions: “If Mrs. Chen asks him (her), ‘What do you do?’ (Ni zuo shenma?) what will he (she) answer?” Then, “Does he (she) have a job/work?” (Ta you mei you gongzuo?).Table 2.2 lists the activities of the Feng family members as they were described to the focus group participants in the town and village interviews.
We predicted that in the present time of economic transition, there would be considerable variation in whether or not someone's activities were thought of as gongzuo. Focus group participants' attitudes might depend on characteristics of the activity, such as locus vis-à-vis the household and whether and how it was remunerated (cf. chap. 3). They might depend on the gender of the person performing the activity. Women's contributions might be less visible than men's, even when the work they do is the same. We also hypothesized that response to the Feng family vignette would vary
| Town Interview | Village Interview | |
|---|---|---|
| First brother | Private transport company with 4 men; then seasonal fieldwork | Grows cotton, family garden; then seasonal fieldwork |
| First brother's wife | Cooks, cleans, washes clothes, small garden; then large garden, expensive mushrooms to sell | Cooks, cleans, washes; seasonal fieldwork |
| Second brother | Full time in collective farm tool factory; then only two month's salary | Runs family transport business |
| Second brother's wife | Cooks, cleans, washes clothes, family dry goods shop; then opens bigger store outside home, 6 days/wk | Cooks, cleans, washes clothes, family garden, raises chickens (made 800 yuan ); then chickens die (makes 100 yuan ); then job in local textile factory |
| Third brother | Permanent but only part year in large joint-venture beer factory; then full time, permanent | Helps second brother with transport business; then factory subcontract for transport, good money |
| Third brother's wife | Job at neighborhood-run day care center | Cooks, cleans, washes clothes, family garden, few chickens and pig; then helps in family dry goods shop |
The vignettes were carefully structured to allow us to compare and contrast reactions to characteristics of the activity and to the gender of the worker. The village interviews allow for comparisons between (1) agricultural and nonagricultural activities (a man operating a farm versus a man operating a family transport business; a woman with a chicken business versus a woman working in a family-run shop selling dry goods), (2) year round and seasonal activities (in agriculture), and (3) pay (a woman who
The vignettes used in the town interviews allow for comparisons between (1) activities in different employment sectors (a man working in a collective factory versus a man working in a joint venture factory; a woman employed by a neighborhood-run day care center versus a woman who works in a family store), (2) seasonality (a man with year-round versus a man with seasonal work in a small private transport company; a man with half-year versus a man with full-year employment in a large joint-venture beer factory), and (3) pay (a woman with a small garden versus a woman with a large garden that produces expensive mushrooms to sell; a man with a job in a collective factory versus a man with the same job who only gets two months' salary; a woman who works in a small family shop in the home versus a woman who works in a large family-run store outside of the home). We constructed these comparisons to highlight one particular aspect of work and to control for other aspects, including gender of the worker.
Other comparisons were designed to see whether the gender of the worker makes any difference. The vignettes used in the village interviews allow for comparisons between men and women involved in year-round agricultural activities, seasonal fieldwork, and a family-run business. Those used in the town interviews allow for comparisons between men and women running a family business, and men and women with employment in the state or collective sector. The goal was to be able to examine the gender of the worker independent of the characteristics of the activity.
Our focus group interviews are somewhat more structured than is typically the case. It is not unusual for moderators to be given no more than a list of topics to cover in an interview, leaving up to them the order, exact wording, follow-up questions, and probes. In contrast, we asked moderators to follow particular question sequences and to use specific terms and language when asking about activities. We wanted to examine each interview as a whole, but we also wanted to be able to compare the ways the different focus groups regarded work or job characteristics (e.g., sector of employment, degree of permanence, level of remuneration), especially in relation to the gender of the worker. A common set of “stimuli” was important for making comparisons between interviews. Also, since our interest was in understanding the meaning of gongzuo and the factors affecting people's views of it, controlling how and when moderators used the term in the course of the interview was important.
Twelve focus group interviews were held—the total number determined by the resources available. Separate groups of mothers, fathers, and grandmothers were interviewed in two town neighborhoods and in two rural villages
The sites for the focus group interviews were selected for their potential to generate contrasts. We selected two pairs of town neighborhoods and rural villages. One pair was located in a relatively developed county, Fu Xian, the other in a relatively poor county, Han Shan Xian.[3]
Fu county seat is a town on its way to becoming a city. It is full of commercial and industrial businesses, including department stores, restaurants, hotels, textile factories, a joint-venture feed factory, gravel transport, and lumber companies. A large hospital has also opened recently. In contrast, Han Shan county seat is not marked by the dust and dirt of construction seen in Fu county seat. There is barely a building that rises above three stories. The town is characterized by one resident as having “not much money, not much of an economy.” Whereas the neighborhood in Fu county seat selected for the interviews is very close to the town center, the one in Han Shan county seat is a village annexed by the town where many people continue to farm in the adjacent fields, despite their urban registration.
The two villages also differ, although not as dramatically. Residents of Fu county village are engaged in intensive, profitable vegetable gardening. There are a number of private household businesses, including duck raising, dry goods stores, and a clothing store, but there are no village-run enterprises. (Several were attempted but failed.) Residents who work outside the village go to factories run by other villages or towns. Han Shan county village, however, has its own village-run factories, whose enterprises include stone-cutting, flour grinding, brick-making, and tree-cutting. As in Fu county village, there are family-run businesses, restaurants, and various small vendors. While the Fu county village focuses on growing vegetables, in Han Shan county village agricultural products include rice, wheat, sweet potatoes, peanuts, cotton, tobacco, and soybeans, and many families raise chickens and pigs. Despite this apparent economic diversity, people in Han Shan county village consider themselves and their village to be poor, reporting a per capita income that is only about 25 percent of that of Fu county village. Many of the houses have mud floors and are made of mud bricks. While Fu county village does not look very different, all of the houses have electricity, half have televisions, and a few people are building new homes.
In textbook applications, focus group participants are strangers to one another. Such a plan is not well suited to the social realities of life in the rural villages and town neighborhoods that make up the Chinese countryside.
The focus group interviews were conducted by staff members of the county, provincial, and national Epidemic Disease Prevention Stations. Residents of neighborhoods and villages that were already part of the China Health and Nutrition Survey were recruited with a letter that described the topics of the focus group interviews and requested their signed, informed consent. This consent procedure was repeated verbally at the start of the interviews. The groups consisted of six or seven participants. The interviews took about one and one-half hours. Instead of tape recorders, two trained note-takers were used. The note-takers combined their versions, and translation was done immediately. The analyses described below are based on the Chinese versions and English translations of the focus group interviews, the comments of the moderators, and our own observations.
THE INTERVIEWS: OPINIONS ABOUT GONGZUO
The Feng family vignette elicited a variety of opinions about gongzuo. In some groups, virtually all activities in the vignettes were labeled gongzuo, while other groups were much more conditional or restrictive in their approach. Participants of some groups quickly came to a consensus view, whereas in other groups there was dissension. Our goal here is to describe the focus group interviews in such a way that some general conclusions can be drawn from the details of the discussions. In the following paragraphs, we review the evidence that supports four conclusions about gongzuo and its meaning for the focus group participants: (1) gongzuo has diverse interpretations; (2) differences in views about this term and the general concept that underlies it are not tied in any obvious way to gender or generation of the participants, or to their urban-rural location or its economic context; (3) rather, classification of particular activities as gongzuo or not depends on employment sector, degree of permanence, pay, and nature of the activity; (4) gender of the worker (in the vignettes) matters, but only a little bit.
1. There is no common understanding of gongzuo.
This first point is easily made by considering the totality of the focus group interviews. Many of the specific activities we asked about drew a mixed response. For example, discussing agricultural fieldwork, some focus group participants said, “Farming is gongzuo. It earns money. Whenever you do that, you have gongzuo.” Others said, “It isn't gongzuo. It's not a job. He works on a farm.” Likewise, with respect to agricultural sideline businesses, some participants commented, “She has gongzuo. It's a household sideline. If you plant cash crops, it's gongzuo.” Others said, “Sideline production is not gongzuo.” Regarding family-run nonagricultural businesses, some said, “She's running a store, contributing to the family finances. She has gongzuo.” The contrasting opinion was: “She works for herself, but it's not gongzuo. Getihu (private household business) is not gongzuo.”
Focus group participants did agree that permanent, full-time employment in a danwei (work unit) is gongzuo. In five of the six town focus groups, participants were unanimous in their view that the Feng brother with a job in a collective factory manufacturing farming tools has gongzuo. There was complete agreement about the third Feng brother's wife, who works in a neighborhood-run day care center. All thought that she has gongzuo too. Similarly, in all six village focus groups, participants agreed that when the second Feng brother's wife goes to work in a local factory making cotton clothes, she has a job. One of the fathers from Han Shan county village remarked, “Definitely she has gongzuo, because she works in a unit.” However, there was little agreement about household-based activities in the private sector, including fieldwork, household sidelines, and running or working in a family business.
2. Interpretations do not depend on focus group composition.
Whether or not particular activities are classified as gongzuo does not depend very much on the composition of the focus group. This finding comes from an examination of the focus group interviews one by one, and is particularly well illustrated by the four focus group interviews of fathers. One of these groups expressed a broad concept of work, two groups offered restrictive definitions, and the final group took a conditional approach to the classification of the various activities. As evidence, we provide a brief summary of each of these interviews.
The Fu county village fathers were vocal about their beliefs that everything is gongzuo. Six of the seven fathers had completed lower middle school and one was a high school graduate; four were farmers, two drivers, and one did interior design. They all felt that all of the Fengs' activities should be considered gongzuo. Fieldwork, whether seasonal or not, is gongzuo. “You can earn money from farming.” “Even if it's just in the fields, it's still gongzuo.”
In contrast, the fathers in Han Shan county village were hesitant to say that anything outside the formal sector was gongzuo. The occupations of these fathers were comparable to the fathers in Fu county village (three were farmers; one was a driver; the others did construction or technical work), although their education was a little lower (ranging from less than primary school to middle-level technical school). Most agreed that the first Feng brother, engaged in agricultural fieldwork, does not have gongzuo, although one father dissented: “Whenever you have a thing to do, you have gongzuo.” Opinions did not change with seasonal fieldwork. Running or helping out in a family transport business does not count as gongzuo either. It only becomes gongzuo when the third Feng brother arranges a factory subcontract. For the fathers in Han Shan county village, the Feng wife engaged in fieldwork does not have gongzuo, nor does the wife selling chickens for 800 yuan. They were divided about the wife helping the mother-in-law in the family shop.
The fathers in Han Shan county town were also fairly restrictive in their classification of work activities, though not as restrictive as the fathers in Han Shan county village. The seven fathers in the Han Shan county town group included three farmers, three blue-collar workers, and one in getihu business; two were primary school graduates, four had lower middle school degrees, and one was a high school graduate. This group tended not to rate activities outside the state and collective sectors as gongzuo, although there were exceptions. Work in a getihu transport business is gongzuo, as long as it is not seasonal. These fathers were divided about the second Feng brother's job in a collective factory, particularly when in the vignette the salary is reduced to two months' pay for a year's work. This was the only focus group that questioned whether collective factory employment counts as gongzuo. These fathers were also mixed about a temporary job at the joint-venture factory (as were many of the groups), but agreed that permanent, full-time
| Broad Definition | ||
|---|---|---|
| Fathers | Village | Rich county |
| Mothers | Village | Poor county |
| Grandmothers | Town | Rich county |
| Grandmothers | Village | Rich county |
| Grandmothers | Town | Poor county |
| Restrictive Definition | ||
| Fathers | Village | Rich county |
| Fathers | Town | Poor county |
| Conditional Definition | ||
| Fathers | Town | Rich county |
| Mothers | Town | Rich county |
| Mothers | Town | Poor county |
| Mothers | Village | Rich county |
| Grandmothers | Village | Poor county |
In contrast, the Fu county town fathers tended to label activities as gongzuo when the hours increased from part-to full-time, when the enterprise grew, and when the pay increased. They agreed that the first Feng brother's work in the family transport business qualifies as gongzuo, even when seasonal. Collective factory work is also gongzuo, though according to one father, receiving only two months' salary is not. All but one father said that temporary work at a joint-venture factory is gongzuo, but that father shifted his opinion when the factory work becomes full-time. Wives doing housework and raising animals for family consumption do not have gongzuo, according to the Fu county town fathers. However, when the first brother's wife has a cash crop garden and makes money, “then she has gongzuo —she runs a specialized household.” The fathers' opinions were split about whether running a small family shop is gongzuo, but all agreed that it is gongzuo when the shop turns into a large store with full-time employment.
It would be difficult to identify a “fathers' perspective” from these four interviews. Similarly, evaluating each of the twelve focus group interviews
Two of the fathers groups were the only ones to convey a restrictive definition, while a third fathers group expressed a broad definition and the fourth, a conditional one. Mothers and grandmothers groups expressed both broad and conditional definitions. Town and village groups are evenly distributed among the possibilities. No specific definition of gongzuo is associated with Fu (rich) and Han Shan (poor) county groups. Thus, the meaning of gongzuo expressed by each focus group as a whole appears to be unrelated to the group's gender, generation, urban-rural residence, or economic context.
3. Whether an activity is gongzuo work depends on the nature of the activity, the employment sector, degree of permanence, and pay.
Whereas virtually everyone in every focus group agreed that nonagricultural work in the state or collective sector is gongzuo, there was disagreement about other forms of activity. Take agricultural work, for example. Two of the six village groups felt that fieldwork does not count as gongzuo, and participants in one of the other village groups could not agree among themselves. One of the village groups was unwilling to classify a lucrative chicken business as gongzuo, and there was disagreement in another group. Although none of the town groups was completely unwilling to classify gardening—even growing expensive mushrooms for sale— as gongzuo, there was considerable disagreement in three of the six groups. What is it about these activities that raises doubts in the minds of the focus group participants? Is it agricultural activity per se? Or is it some characteristic of agricultural work, such as its seasonality, or generally low level of remuneration?
In the village focus group interviews, we asked participants to consider the first Feng brother's small farm and the second Feng brother's small transport business. Five groups classified the small transport business as gongzuo, and some participants of the sixth wanted to do so. Only three groups thought that the farm work qualified as gongzuo, two groups thought not, and there was disagreement in the other. We also asked participants in the village interviews to think about the chicken business of the second Feng brother's wife and the small shop in which the third Feng brother's wife worked. Four groups considered activity in either type of business as gongzuo, and there was disagreement about both types of business in the fifth group. But opinions in the sixth group changed: the chicken business was clearly not gongzuo, while the shop clearly was. Evidence from the village interviews suggests that, holding constant the size and household nature of
Initially, we thought that the seasonality of agricultural work might be one reason for ambiguity about its classification. The village focus group interviews asked explicitly about this, inconnection with the first Feng brother and his wife. We were surprised to find that opinions about the fieldwork done by this couple were mostly the same, regardless of whether the fieldwork was seasonal or year-round. Remuneration, however, did make a difference.
In the village interviews, the relevant comparisons begin with the second brother's wife, who works in the family garden, cooks, cleans, washes clothes, and raises chickens for sale. Initially, the chickens bring in 800 yuan. To put this amount in to perspective, according to interviews conducted with village officials in 1995, per capita income in Fu county village was 4000 to 5000 yuan, and in Han Shan county village, 1000 yuan. Four of the six village focus groups described the second brother's wife as having gongzuo, the fifth group said that she does not have gongzuo, and the sixth group could not agree. Then, according to the vignette, her chickens get sick and many die. She now makes only 100 yuan. This change in economic circumstances had little effect on the opinions expressed by the focus groups. A grandmother in Fu county village commented, “It's hard to say. She has gongzuo, but she does not have very good fortune.”
But if we compare opinions about the second brother's wife with opinions about the third brother's wife, the importance of an income is clear. The only difference in the activities of these two wives is that the second brother's wife makes money raising chickens, whereas the third brother's wife raises them for home consumption only. Fu county village fathers thought that the second brother's wife has gongzuo, regardless of the amount of money the chicken business brings in, but their opinions were mixed when chickens are raised for consumption. Similarly, the Han Shan county village fathers began with mixed views about the second brother's wife, but agreed that raising chickens for consumption only is not gongzuo. The opinions of mothers and grandmothers were less influenced by income, with one curious exception. Fu county village mothers said that the third brother's wife has gongzuo but that the second brother's wife does not. It is difficult to know what to make of this. In the debriefing, the moderator wondered about the comprehension level of the mothers in this group.
Thus, in the village interviews it appears that if an activity generates some income, it is more likely to be seen as gongzuo. Level of remuneration does
To sum up the discussion so far, ambiguity about whether agricultural activity counts as gongzuo appears to stem from the nature of the activity per se and also from its level of remuneration. We now turn to private sector activity, asking a similar set of questions. We noted earlier that employment in a danwei clearly is work according to the focus group interviews, whereas there was less agreement about household-based and private-sector activities. Why? Is it because gongzuo is a designation that has been reserved for formal sector—state and collective—employment? Or is it because of the low level of remuneration associated with some jobs in the private sector, or because of their temporary, seasonal, or unstable nature? We need to say at the outset that we can only answer these questions for nonagricultural work. All of the agricultural activities we asked about in the focus group interviews were household-based and in the private sector. Indeed, very few agricultural workers can be found in the state or collective sector nowadays.
In the town focus group interviews, two comparisons are specifically relevant to employment sector. The first is between the first Feng brother's private transport business and the second Feng brother's job in a collective factory that manufactures farming tools. The focus groups responded similarly to these two jobs—that both are gongzuo. The second is between the dry goods store operated by the second Feng brother's wife and the job of the third Feng brother's wife in a neighborhood-run day care center. Here, we see more of a difference in response. Although four of the six groups classified both as gongzuo, the other two groups did not. Mothers in Fu county town could not agree about working in the dry goods store, although most
Doubts about private sector activities become even more evident when the vignette points out their low level of remuneration and temporary, seasonal, or unstable nature. Consider first the town focus group interviews. Whereas the first brother's job in a small private transport company with four other men counts as gongzuo for everyone in five focus groups and for most everyone in the sixth, this is no longer true when the moderator changes the work to seasonal—whenever the harvest comes in. There is now disagreement in two of the groups, and in a third no one is willing to label the activity as gongzuo because it is “unstable.” Similarly, when the third brother's job in a large joint-venture beer factory changes from six to twelve months a year, doubts expressed about calling it gongzuo in two of the six focus group interviews disappear.
Indeed, level of remuneration is important even for jobs in the state and collective sector. When the second brother, who works in a collective factory that manufactures farm tools, receives only two months' salary rather than a full year's salary for his work, opinions shift in three of the six town focus groups. Five of the six groups were unanimous that he has gongzuo on full salary, but only two thought so after the salary cut. Comments in those two groups included: “No. Can't change the answer. He has gongzuo.” “He still has gongzuo, just no wages.” “We think he has gongzuo.” “He has gongzuo, but the benefit is not much.” Comments on the other side included: “It's hard to say.” “He doesn't have a job.” “He doesn't have gongzuo now.”
In summary, in village and town focus groups, whether or not an activity is viewed as gongzuo depends on its characteristics. If it is agricultural, it is less likely to be seen as gongzuo. Agricultural or not, if it is seasonal, temporary, or unstable, it is less likely to be seen as gongzuo. If it is located in the private sector, especially if it is household based, there will be ambiguity about its classification. If the activity does not generate much of an income, it is less likely to be seen as gongzuo.
Consider what these tendencies mean for how the typical activities of men and women are viewed. Men are more likely to operate year-round nonagricultural businesses, especially in rural China, while women tend to oversee seasonal agricultural side lines. We should not be surprised if the men are more likely to be seen as having gongzuo than the women. In urban areas, men are more likely than women to have stable state-sector jobs, and so again we should not be surprised if men are more likely to be viewed as having gongzuo than women. Of course, in addition to the characteristics of the
4. Gender of the worker matters, but only a little bit
The focus group interviews were designed to allow conclusions to be drawn about the gender of the worker independent of the characteristics of the work. In the town interviews it is possible to compare responses to the first Feng brother, who operates a private transport business, with responses to the second brother's wife, who runs a large dry goods store across the street from where the family lives. The responses were the same in five of the six groups. Fathers from Han Shan county town, the sixth group, felt that the brother with the transport business has gongzuo whereas the wife working six days a week in the store does not. “She doesn't have gongzuo. She is running her own store (getihu).” A second comparison is also possible between the second Feng brother's job in a collective factory and the job of the third brother's wife in a neighborhood-run day care center. Virtually everyone agreed that both have gongzuo.
Three comparisons are possible based on the village interviews. The first involves year-round fieldwork. Four of the six groups responded the same way, whether a Feng brother or his wife did this work. A fifth group, grandmothers from Han Shan county village, thought the Feng brother has gongzuo but not his wife. “He has gongzuo. Farming is gongzuo.” “She does not have gongzuo.” “She is a farmer and a housewife.” A sixth group, mothers from Fu county village, held the opposing view. In the debriefing following this interview, the moderator noted that these mothers were emphatic that Feng women doing fieldwork have gongzuo, but Feng men doing fieldwork should “go out and get a real man's job.” A second comparison involves seasonal fieldwork. The results are similar to those for year-round fieldwork, except that one of the fathers from Fu county village was unwilling to describe the seasonal fieldwork of the first Feng brother's wife as gongzuo but expressed no doubts that her husband doing that same work does have gongzuo. A third comparison possible with the village focus group interviews is between the third Feng brother, who helps the second brother with the transport business, and his wife, who helps her mother-in-law run a small shop. One group, the grandmothers in Han Shan county village, distinguished between helping with a transport business and actually running it; they felt that running the business is gongzuo, while helping is not. These same grandmothers thought that the wife who helps in a family dry-goods store does have gongzuo, though. The other five groups did not differentiate between helping with and running a business, whether the comparison was between two brothers or a brother and his wife.
Overall, the focus groups showed a tendency to classify men's activities as gongzuo more often than women's, but the tendency was slight.
IMPLICATIONS
Whichever way we organize our data, we find ambiguity in the meaning of the term gongzuo. Participants in 12 focus group interviews agreed that full-time, permanent, formal sector, danwei employment is a job; they disagreed about other kinds of activities. Interestingly, focus group participants used the same justifications to explain opposing views: “She doesn't have gongzuo —getihu is not gongzuo.” “She has gongzuo; a getihu is gongzuo.” Clearly, and perhaps in contrast to the pre-reform era, the term gongzuo has different meanings for different people. Views depend to some extent on characteristics of the work and to a lesser extent on the gender of the worker, but even so, there is no common definition of the boundary between activities that count as work and those that do not. Nor, given a boundary, is it sharply drawn. At least for focus group participants, the typology presented by Rosenfeld (chap. 3), which distinguishes work activities according to their household basis and remuneration status, identifies only the ends of continua between work and not-work. As Harrell (chap. 4) notes, the middle ground is contested territory. That is, formal-sector, paid work outside the household counts as gongzuo, and unpaid work inside the household does not, but the classification of other kinds of activities generates discussion and disagreement.
How work is viewed depends partly on its characteristics—the nature of the activity (agricultural or not), relationship to the household, employment sector, degree of permanence, and pay. If a job is agricultural, it is less likely to be seen as gongzuo. To some extent, this is because such work is seasonal, is organized by households, and tends not to produce much income. If a job is seasonal, temporary, or unstable, it is less likely to be thought of as gongzuo. If it is household-based, or located in the private sector, there is more disagreement about its classification. If the activity does not generate much income, it will be less likely to be seen as gongzuo. Even if we take these specific characteristics of agriculture into account, however, comparing small farms and agricultural sidelines with other kinds of household businesses, agriculture is still less likely to be viewed as gongzuo. These results suggest that the nature of an activity—whether it is located in the agricultural sector or the industrial or service sectors—affects how it is viewed, independent of remuneration or its location vis-à-vis the household. This is a pattern that may well characterize rural economies around the world (cf. Dixon 1982).
Holding the characteristics of work constant, we find a slight tendency to take women less seriously than men. However, the characteristics that typically
The fact that different kinds of work are differentially visible has implications for monitoring economic change in China. In the cities and towns there is movement away from state and collective sector employment. Contract work and work that is part-time, temporary, and without guaranteed pay are increasing. Households have reemerged as an important base of economic activity. Based on our results, these are the kinds of activities that are less likely to be thought of as gongzuo. In rural areas, industrialization is shifting workers out of agriculture, into jobs that are more likely to be seen as gongzuo. To the extent that attempts to monitor economic change rely on respondent reports (as is the case with social surveys) there is potential for shifting attitudes about work to bias the picture.
One potential source of bias in survey-based monitoring appears not to be a problem, according to the focus group results. Interpretation of gongzuo is not predicted by characteristics of the focus group itself—gender, generation, urban-rural location, or economic level. Based on our results, no systematic bias is introduced by the person who reports on work activities. For example, the picture should be the same whether a husband reports on his and his wife's activities or his wife reports on her own and his activities. Gender differences relate to characteristics of the work and, to a lesser extent, to the gender of the worker, but not the gender or other characteristics of those classifying the activity. Similarly, urban-rural differences relate to the work but not to the views of the persons describing it. This is reassuring from the standpoint of social survey applications. Although
Of course, these conclusions are based on only 12 focus group interviews in two county seats and two rural villages in Hubei province, and it is difficult to know how far the results can be generalized. Strictly speaking, they cannot be generalized at all. Focus group participants were not selected according to any probability method; interviews were conducted in only four sites.[4] However, our failure to find contextual variation in the interpretation of gongzuo may actually be a strength in this regard. We designed the focus group interviews explicitly to explore contextual differences in industrial and commercial development, and their consequences for attitudes about work. We did not find these effects. This is a weak result, but it suggests that where we conducted the interviews did not matter—the findings would have been the same.
NOTES
Funding for this research was provided by the National Institute of Child Health and Development grant P01-HD28076, the National Science Foundation grant #37486, and the Fogarty Center. In addition to the authors, our research team in Hubei included Susan Short, Jill Bouma, Lin Hai, Lu Bing, Li Dan, and members of the Fu and Han Shan County Epidemic Disease Prevention Stations. Susan Short generously made her field notes from additional research work in Fu and Han Shan counties available to us as background for this chapter.
1. In early versions of this chapter we translated gongzuo as “work.” Various readers of the earlier versions pointed out that gongzuo can also be translated as “job,” and that by using “work” we insert a more evaluative tone than is appropriate. Because of this concern, we have chosen to use the term gongzuo in our essay rather than its various translations: “work,” “activity,” “job.”
2. The larger purpose of the interviews was to explore work and child care, and for this reason we selected mothers, fathers, and grandmothers of children or grandchildren less than seven years of age.
3. The county (xian) names are pseudonyms, referring to the economic level of the counties. Fu means wealthy; Han Shan means cold mountain, implying poverty.
4. The focus group interviews were designed in relation to the China Health and Nutrition Survey. The four sites are sampling units within the larger survey, and as such, were selected as part of a probability design.
3. What Is Work?
Comparative Perspectives from the Social Sciences
Rachel A. Rosenfeld
A full understanding of individuals' lives in the context of social change requires a broad definition of work: effort resulting in some product or service for exchange or domestic consumption.[1] Work can be done in the home (broadly defined)[2] or outside it. It can be done for pay (such as a wage, a salary, or profit) or direct exchange or neither. In this chapter I use these two cross-cutting dimensions—work location and returns—to organize discussion about work, gender, and households (see table 3.1). I consider the nature of the work shown within each cell; linkages between the cells for the same individual, the household, or the economy; “movement” of work from one cell to another; and, finally, some of the problems with using these dichotomies for work location and for pay. Examples come mainly from the United States and Europe, although a few from China are drawn from other chapters of this volume. A full application to the China setting is undertaken by Harrell in chapter 4.
WORK LOCATION AND RETURNS
Work for Pay outside the Home
Industrialization brings an increase in formal wage jobs, with the timing and context of this increase related to characteristics of the labor force. In eighteenth-century America, where land was plentiful and labor was not, the first factory workers in New England textile plants were farm daughters, although the typical occupation for employed women then was domestic servant. Women's participation in the labor force increased over time, especially with the growth of service industries and women's education, but
| For Pay or Profit | Unpaid | |
|---|---|---|
| Outside the home | 1. Wage Work | 3. Volunteer work |
| Informal sector | Community action | |
| Self-employment | Family business | |
| Home based | 2. In-home business | 4. Housework, child care |
| Home-based employment | Family business | |
| Informal sector |
Women's wage work outside the home has been higher in state socialist countries and in others with ideologies of gender equality based on equal labor force participation, such as Sweden, the former German Democratic Republic (East Germany), and China. But even in these countries, men's employment rates are somewhat higher. In the former German Democratic Republic, among those born in 1951–53 and 1959–61, 97 percent of men and 88 percent of women held jobs for pay in 1989 (Trappe and Rosenfeld 1998). In urban China, earlier retirement for women than for men explains some of the gender difference in employment (chap. 7). In rural China, gender differences in nonfarm employment are found among the married, although not the unmarried, population (chap. 8).
The boundaries between paid jobs outside the home and other types of
Two additional distinctions relevant in this context are those between contingent and noncontingent work and between informal and formal sector jobs. Polivka and Nardone (1989, 11) define a contingent job as “any job in which an individual does not have an explicit or implicit contract for long-term employment or one in which the minimum hours worked can vary in an unsystematic manner.” American social scientists use the term “contingent work” to describe flexible work arrangements increasingly used by employers since the 1980s to control labor costs in the face of rising international competition, economic downturns, and rapidly changing markets. Migrant agricultural laborers, restaurant servers scheduled day-by-day or week-by-week, and college teachers hired “by the course” are examples. While some workers choose such jobs because they have alternative sources of income or are involved in additional activities (e.g., school, child care), for others it is the choice of these jobs or no jobs. Women seem to predominate among contingent workers, although according to recent U.S. statistics, the gender difference is most evident among the employed between the ages of 20 and 44 (Bureau of Labor Statistics 1995).
The informal sector is a somewhat related concept. Portes and Sassen-Koob (1987, 31) define the informal sector as “all work situations characterized by the absence of (1) a clear separation between capital and labor; (2) a contractual relationship between both; and (3) a labor force that is paid wages and whose conditions of work and pay are legally regulated. … The informal sector is structurally heterogeneous and comprises such activities as direct subsistence, small-scale production and trade, and subcontracting to semiclandestine enterprises and home workers.” Researchers argue that this type of work is growing in contemporary advanced and developing economies as a result of globalization of the division of labor. “Capitalists and TNCs [transnational corporations] … use informal-sector workers, particularly women, instead of formal wage workers to avoid labor legislation and to keep labor cost low” (Ward 1990a, 2). This concept puts more emphasis on the vulnerability of the worker than on job insecurity and variable hours. Zhang (chap. 10) gives examples of exploitation of young unmarried women from rural areas working as wage laborers for small family enterprises in the near suburbs of Beijing.
Home-Based Business or Employment
One feature of industrialization is change in the location of work from farm or homework shop to factory, office, or store. In both industrialized and contemporary less-industrialized countries, however, the home can be a place for paid or income-generating work. The boundary between “work” and “nonwork”—spatially, temporally, and perceptually—can be very fluid in this case. In the early twentieth century in the United States, for example, because of norms and sanctions among many ethnic groups and employers against married women's employment outside the home, women earned money at home by taking in boarders, doing piecework, or producing crafts. Many of these efforts were not officially recognized as work (Bose 1984). In the United States, China, and elsewhere, there is a continuing tradition of agricultural and nonagricultural sidelines run by farm women that bring cash into the family (chap. 6; Entwisle et al. 1995; Sachs 1983). Even if not always recognized as results of “real” work, women's egg, butter, and vegetable sales can be crucial for the family economy (Fink 1986; Rosenfeld 1985). Under some circumstances, this lack of recognition of in-home work can have advantages. Margery Wolf (1985) discovered that because under some Chinese Communist Party (CCP) regimes rural women's sidelines were classified as housework, these women could earn private incomes sometimes greater than those of men working full-time on the farm collective.
Among nonagricultural households also, many small businesses operate from the home, and the informal sector often involves home workers. In industrialized societies, technological advances with respect to equipment size and communications may increase the proportion of workers who are paid for work they do at home. In 1997, 18 percent of workers in the United States did all or some of their work on their primary job in the home, with about a third of these self-employed (U.S. Bureau of the Census 1998, table 663). Some—but not all—of these jobs provide flexibility for workers and employers. Home workers in the United States tend to be either clerical and blue-collar contingent workers or high-status professionals in “regular” jobs to which they telecommute, blurring the line between home and office (Reskin and Padavic 1994; Tomaskovic-Devey and Risman 1993). In China, economic reforms have led to an explosion of family-owned small businesses, and often production activities take place in the same space as domestic activities (chap. 10).
Working for pay within the home seems more compatible with women's domestic work than work away from home. In the United States, however, there is not a large overall difference by sex (U.S. Bureau of the Census 1998, table 663). The proportion of those working at home who were self employed rises with age and is especially high at 65 and over (U.S. Bureau of the Census 1993, 404), suggesting that those in life cycle stages other
Work without Pay outside the Home
As discussed later, it can be difficult to draw the lines for work/nonwork in this cell, given the general definition of work I use. I include here unpaid family workers in businesses located away from home, and also organized volunteer work and other community efforts. Blau, Ferber, and Winkler (1998, 56) define volunteer work as “tasks, performed without direct reward in money or kind, that mainly benefit others rather than the individuals themselves or their immediate family.” In the United States, the proportion of the population doing volunteer or more broadly defined community work seems to have declined over the last two decades. Women are somewhat more involved than men, although the types of volunteer work differ by gender (F. Blau, Ferber, and Winkler 1998). Among American farm families in 1980, men and women were about equally likely to be part of a community organization such as a church, PTA, or Rotary (57 and 61 percent), but men were much more likely to be part of farm-oriented organizations (Rosenfeld 1985). At least some of this unpaid work is done for pay by government or private workers in other countries.
Community action is often studied as political activity rather than as work and tends to be described in community case studies. In China one might put the efforts of “model workers” (e.g., chap. 5), neighborhood vigilance committees, and political activists in this cell, as well as “volunteer” political duties required by the Communist Party. Corvée and other kinds of forced labor might also fit here.
Unpaid Work in the Home
If a regular wage job is the “typical” job for nonfarm men in many settings around the world, unpaid work in the home or compound is “typical” work for women. Here the boundaries between work and nonwork, and among types of work, can be especially cloudy, with consequences for the recognition and value of women's activities in particular. Housework is often trivialized and neither considered “real” work (Hochschild 1989) nor given high priority. Work that can be viewed as domestic work is not valued or rewarded as highly as the same work done elsewhere (chaps. 2 and 10).
Everywhere, even in the most gender-egalitarian countries, women do most of the housework, child care, and elder care (F. Blau, Ferber, and Winkler 1998). There is, however, some variation in the unevenness of the
Unpaid home-based work also includes other subsistence work, such as gardening, canning, sewing, and knitting. According to the 1980 U.S. Farm Women Survey, most farms or ranches had gardens or animals raised for family consumption, and three-quarters of the women said raising vegetables or animals for the table was a regular duty (Rosenfeld 1985, 57). In addition, women often work in family businesses, even though they do not receive direct pay for this and are not listed in surveys as “owner or operator.” This is certainly true for farm and rural women in both the United States and China (Entwisle et al. 1994; Rosenfeld 1985). Zhang's fieldwork shows that among Beijing's “floating” population, women's labor-intensive production work done in the home can be viewed as “chores” rather than “real” work, in contrast with men's part of the production and marketing process, which takes place more often away from home: “In men's narratives of labor division and productivity, the roles of their wives and daughters as the primary producers and supervisors of production often fade away in the domestic background” (chap. 10, page 187).
LINKS AMONG TYPES OF WORK: INDIVIDUALS, HOUSEHOLDS, AND THE ECONOMY
The boundaries between categories do not indicate mutually exclusive activities. The type of work a person does in one category of work can affect the work he or she does in another domain and have interconnections with the work of other household members as well (chap. 15). The most obvious linkage is between unpaid home work and work categorized in the other cells. Women's under representation in market work, especially in “regular” wage or salary jobs, is often attributed to their greater responsibility in the home. As noted, women spend more time on domestic work and less on paid work than men in all countries for which data are available—more and less developed, capitalist and state socialist (F. Blau, Ferber, and Winkler 1998). What varies among countries are gender differences in total work weeks and the balance of paid work outside the home and unpaid work
Causality may also vary: Women may spend less time on domestic work because they are employed, or they may cut down their job hours because of greater domestic responsibility. In most Western countries, being female, married, and a mother make it more likely that one is employed part-time, although countries differ considerably in the proportion of the female workforce with reduced hours (Blossfeld and Hakim 1997; Rosenfeld 1993; Rosenfeld and Birkelund 1995). Reductions in job hours in order to care for young children, for example, are not always possible in the United States but have been part of public policy in Sweden and Norway, as well as in other countries such as West Germany (Kalleberg and Rosenfeld 1990; Trappe and Rosenfeld 1998).[3] Controlling for family configuration, spouse's employment, job characteristics, sex role attitudes, and demographic attributes, Kalleberg and Rosenfeld (1990) examined reciprocal effects between hours employed and participation in domestic work among those employed and married or cohabiting in the United States, Canada, Norway, and Sweden. They found no effects for men in any of these countries, but found that hours employed had a negative effect on housework participation for employed American wives—and effects in the opposite direction for women in Sweden and Norway.
As noted, in the United States the greatest sex differences in contingent work emerge in the childbearing ages. Women's informal work has also been described as allowing women to both care for their families and earn money. “Flexibility” and employment at home, however, do not always lead to this outcome. Piecework in the home at low per-unit rates can lead to very long work days and require the labor of other family members, including children (Ward 1990a). “Flexible” jobs can mean heavy work demands that do not fit with home responsibilities. Reskin and Padavic (1994, 160) cite Costello's (1989) interview with a home worker who processed insurance claims: “If you'd [unexpectedly] get six hours of work someday and you have two little kids at home and you only have three hours worth of TV, when are you going to get the work done?” (see also Rosenfeld 1996).
Implicit in much of the discussion about trade-offs between regular wage jobs and other sorts of work as a result of domestic responsibilities is the idea that there are other earners in a woman's family, especially a husband. Not all households have a husband, however, or a husband who has income, or other people who bring in money (Tiano 1990). Even when there are other earners, women's earnings are often important, especially when household income is low. Tiano's study of northern Mexico shows that women working
More generally, though, paid jobs can support the unpaid work of the same individual or others in the family. Off-farm jobs held by varying combinations of household members—depending on the characteristics of spouses in particular and on local labor demand—can help finance farm production (Rosenfeld 1985). At the same time, unpaid work in the home helps support other kinds of work. In the “two-person” career, a wife who entertains, does the husband's research and clerical tasks, and frees him from household chores can be important to a husband's career success (Pavalko and Elder 1993). Employees' gardening and other subsistence food production can allow Third World employers to offer below-subsistence wages (Ward 1990a). Fink (1986, 56) notes that in the United States “Farm commodity production did not stand on its own or pay its own bills during much of the period from 1880 to 1940; rather, it was subsidized by a subsistence economy that has been ignored because it was women's work.”
Women's paid work outside the home can create opportunities for paid and unpaid work by others, both inside and outside the household. Household members other than the mother care for children. In the United States, in 1993, almost a third of the children five years of age and younger were cared for at least part of the time in their home by someone other than the mother, in some cases by a relative or friend without pay, in other cases by someone who was paid (F. Blau, Ferber, and Winkler 1998, 329). Some families take on shift work, so that one parent is always available to care for the children (Presser 1989). A mother who is a nurse, for example, might work the evening or night shift, while the father has a morning shift. There is a generational aspect to such combinations of paid and unpaid work. If women in the grandmother generation are still employed when their daughters are having their families (as is true among baby boomers in the United States), they may not be available to help with child care. On the other hand, the daughters may have to modify their own employment patterns to care for those in the grandparent generation (Rosenfeld 1996). In China, the changing ratio of children to parents and grandparents with the introduction of the one-child policy may have future consequences for such intergenerational exchanges of women's work especially.
Connelly (1992) demonstrates that in the United States women with children are more likely than others to run home day care centers—work that falls into cell 2. Byerly (1986), however, found that in southern mill towns white women employed as textile factory operatives would spend roughly a day's wages for an African American woman to stay in their homes caring for their children and doing other domestic work—informal sector work in cell 1. Such arrangements often depend on the presence of lower
For China, where there is little part-time employment in nonfarm wage and salary jobs, Entwisle and colleagues (1994, 36) found almost no difference in work patterns between young married women who did and did not report taking care of children. Interviews with two of the women are instructive. Neither mentions cutting down hours of work to care for young children: “Taking care of children is first priority, so you cannot reduce the time spent on this. But you can ask others to help you.” “It is very important to go to work, but a person should not reduce time on child care for this reason. We would not ask relatives for help. We would hire a nanny.” As Hershatter (chap. 5) discusses, however, 1950s labor heroines were not always in a position to be successful in caring for their children as well as fulfilling their duty to display their heroic virtues.
Linkages between demand for unpaid labor in the home and paid work in the labor market are not just at the individual or household level. Policies that try to help women balance home and job responsibilities, such as paid maternity leaves and reduced hours for mothers (and sometimes fathers), can affect the kinds of jobs to which women have access. In Sweden and the former German Democratic Republic, for example, women were steered away from jobs with higher earnings and authority because of the assumption that they would temporarily leave their jobs or reduce their hours when they had children, as allowed by state policy (Rosenfeld, Van Buren, and Kalleberg 1998; Trappe and Rosenfeld 1998).
Elsewhere, women are selected into jobs based on marital status and fertility. Tiano (1990, 217) states, “For many, particularly older, partnered [Mexican] women who occupy the most vulnerable sector of the female labor force …, jobs in the formal service sector are difficult to obtain, as are positions as live-in domestic servants. Thus, their main employment alternatives involve income-generating activities in the informal sector.” Ward (1990a, 12) cites Enloe (1983) as showing, “In Puerto Rico … TNCs [transnational corporations] have encouraged increased investments in sterilization and birth-control programs to ensure a supply of women workers.”
Unpaid work outside the home (cell 3) is another realm of work in which there are important links to other cells at many levels. In the United States, for example, unpaid volunteer work may be very important to one's job, both as a semiformal requirement (to fulfill “community service”) and as a source of political and business contacts. For those not employed (students, men and women out of the labor force), volunteer work may provide experience that improves their chances on the job market later. Many volunteer activities, such as those in churches, schools, and youth groups (PTA,
Participation in unpaid activities outside the home can have a widespread impact on the structure of work and individuals' access to it. Work in cooperatives, political organizations, unions, professional associations, and social movements may lead to either change or preservation of existing economic structures and opportunities. Those taking part in the contemporary American women's movement, for example, generally seek to bring about greater opportunity for women in labor market work, as well as greater support for their unpaid domestic work (e.g., increases in affordable child care). Likewise, labor heroines in China were models for the propagation of new economic arrangements the CCP was trying to impose on rural areas (chap. 5). Many American farmers and their families, on the other hand, belong to organizations or do other community work to try to maintain the family farm and, implicitly, its traditional division of labor.
CHANGES IN WORK LOCATION AND REMUNERATION
Up to this point, I have discussed distributions of individuals among and within the four work categories of the typology I propose and linkages among types of work. There are also linkages among the four categories as changes take place in the relative opportunities for different types of work, with consequences for individual work patterns. Work may shift across the boundaries of household and pay, bringing changes in perceptions, value, and rewards.
As already mentioned, one of the biggest changes in the location and remuneration of work comes with industrialization, as more work is done for pay and away from the home or workshop. This has consequences for the division of labor in the household even beyond the linkages between wage work and domestic work discussed above. With industrialization, many of the products women made with unpaid or paid labor in the home, such as candles, cloth, and clothing, become factory-produced commodities. These factory products may be cheaper overall, but obtaining them requires cash. In some situations, women's ability to earn money at home or in the informal sector decreases as household items are manufactured outside the home. In China, for example, spinning and weaving, including producing
At the same time, with increases in women's paid work away from home, more services that formerly were provided without pay within the home become regular or informal sector jobs. With fewer women providing unpaid work at home, the demand for organized children's activities, prepared or ready-to-prepare food, cleaning services, household appliances, and perhaps certain welfare services are likely to increase. Such demand can be fulfilled through small businesses located in the home and also firms of many sizes outside the home. As noted, where the firm or government has to provide these services for workers, it may discriminate against women.
Although not well documented, jobs involving activities also or formerly performed in the home may be especially likely to employ women (England, Chassie, and McCormack 1982). Overall, women are over represented in service industries and nondurable manufacturing, whether in regular or informal sector jobs. Women are more likely than men to be day care and primary school teachers, for example, or to work in the garment industry in many parts of the world. These industries also tend to pay less than others. Zhang (chap. 10) tells of a migrant woman in the near suburbs of Beijing whose husband made her close a profitable leather goods shop but allowed her to run a large day care center. At the same time, what is “masculine” and what is “feminine” varies over time and across cultures, depending on labor demand, characteristics of potential male and female workers, job structures, and the culture. In Europe, for example, where there is school-based dental care, many dentists are women, while in the United States dentistry has only recently increased its proportion of women (F. Blau, Ferber, and Winkler 1998).
PROBLEMS WITH THE FOURFOLD CLASSIFICATION OF WORK
Starting from the cross-classification of paid versus unpaid work and home based versus non-home-based work highlights often-unrecognized work that women are likely to perform. Discussion of linkages among types of work also leads to consideration of the nature of individual, household, and national work patterns. The problems with this classification are to a large extent created by the very boundaries they demarcate, however, and examining these problems can provide further insights into the complexities of studying work.
One set of problems is that of measurement. The boundaries between categories are not always clear. Certain types of work can be done both inside
The Chinese “inside/outside” distinction illustrates these tensions. Inside work was (and is) women's work: “men farm, women weave” (chap. 1). While this ideal of women's work as “inside” was sometimes taken literally, it masked the reality that women were also in the fields and at the markets (chap. 5). The contrast between the literal ideal and the actual location of activities is important. On the other hand, using the definition of for/ with versus outside the family, rather than physically inside versus outside the home, is very useful, especially in the Chinese context. This dichotomy helps conceptualize work in the last half of the twentieth century in China. Small, family-owned businesses, for example, would be “inside” regardless of where they were located, while new joint-venture employment would be outside. Zhang (chap. 10), though, shows that the physical location of the family business can make a difference in women's involvement. Further, seeing such a distinction as fixed would ignore the changes over time in what is defined as “inside” and therefore suitable for women, as compared with “outside” and therefore inappropriate for women. Zhang stresses how the boundaries of the spaces to which women are allowed free access in the different social strata of the “floating” population change, and how socially constructed spaces shape and are shaped by gender relations. It is not just the site where an activity takes place but also the meaning associated with that site that influences who does the work, in what way, and for what value.
Measurement problems also arise when a person does more than one task at a time. Child care is often combined with other tasks in the home or family business, making it hard to calculate time spent in child care. When there is a family business, determining where production for the family ends and production for pay or profit begins can be difficult. A woman making tamales to sell could also be making her family's dinner. Coleman and Elbert (1983, 3) interviewed one American farm woman who, “when asked to enumerate the hours spent in farm versus home tasks, opened the lid of her washing machine, revealing a common mixture of barn suits, children's jeans, and furniture covers all tumbling around in the soapy water.” Someone with a large garden can sell off the “extra” produce without considering
What is done for “pay or exchange” is also not always clear. In his text on the sociology of work, Hall (1994, 2–3) gives the example of his being a volunteer member of the ski patrol who gets ski lift tickets and discounts on ski equipment. Even more difficult for classification purposes are situations where people provide information, goods, or services without a direct reward. This is true in many community and political efforts, as suggested earlier, but in more private contexts as well. Could “neighborliness” be considered work—for example, helping an elderly neighbor whose family lives in another state? This might be considered unpaid work outside the home, even though there is no direct pay or exchange.
A second set of boundary problems involves the household. When considering different types of work and how they affect and are affected by the work of other household members, what are the boundaries of the household? Does “household” refer to people who are in the same residence, regardless of whether they are family members and regardless of whether they are paid for home-based tasks? Alternatively, is it better to talk about “families” rather than “households”? Davis (chap. 14) emphasizes the importance of distinguishing between families and households for understanding what has been going on recently in Chinese urban housing markets. Family obligations and rights extend beyond coresident units, and household arrangements relate to understandings of such reciprocities. Davis shows how family members redistribute themselves over households with changes in marital status and apartment ownership.
Households may be formed by unrelated individuals. In the Philippines, for example, groups of young factory-worker women living together to save on living expenses often share chores such as cooking or getting water. On the other hand, some households contain paid domestic help, hired hands, or wage laborers (Byerly 1986; Fink 1986). These workers' earnings may help support households other than the ones in which they live. Households can also subsidize the wage work of family members elsewhere. Diane Wolf (1990) reports that in Central Java not only did daughters working in factories and living away from home send back only a small proportion of their wages, but families actually subsidized low factory wages by providing daughters with food, household goods, and sometimes cash. At the same time, savings and gifts from factory wages helped the families survive during a crop failure. Again, simply using a “family/outside” distinction would miss the importance of physical location and its meaning; concurrently, it is important to keep in mind family connections beyond a particular household.
A third set of problems is due to the fact that this classification system does not go far enough in examining patterns and their variations. It sets
On the other hand, this classification scheme deemphasizes combinations of work activities. Entwisle et al. (1994) make a good case for considering sets of work activities and their variations. Women often work double or triple “shifts” made up of formal paid work, unpaid domestic work, and income-generating sidelines. But some women (and men) are likely to have fewer shifts than others, and there is variation among people in the content of each shift. “Indeed, it may be that these combinations rather than individual tasks are the object of negotiation, struggle, choice, and control within households. … Studying the implications of one activity for another runs the risk of confusing causality for joint choice” (Entwisle et al. 1994,38–39).
Looking at the work patterns for the household as a whole is also worthwhile. The New Home Economics (Becker 1981) sees households as decision-making units striving to maximize a household utility function. Households are assumed to allocate their members' labor on the basis of relative skills, wages, opportunities, and substitutability. Men specialize in paid work because their time is more valuable than women's in the labor market (they earn more), and women specialize in home production for family consumption. Moen and Wethington (1992) note that such a theory is often applied after the fact, ignores power differences within the family or household, and misses “nonrational” allocation of family resources that are cultural or personal (as illustrated in D. Wolf 1990). It still emphasizes, however, the agency of the family or household in the face of its immediate or changing economic environment (chap. 15).
In general, households respond to changing circumstances—whether within the family or in the larger economic structure—by changing their consumption and production patterns, directly or indirectly. Changing the household allocation of labor is part of this. Lobao and Meyer (1995), for example, found that when there were work changes during the 1980s farm crisis in the United States, women in Ohio farm households tended to increase time spent on household tasks and off-farm work and decrease their
CONCLUSIONS
I defined work as effort resulting in some product or service for exchange or domestic consumption. Such activities can be done for pay or without pay, be home-based or performed mainly outside the home. Within each cell formed by this cross-classification, there are a number of different kinds of work. Paid work outside the home can be a “regular” office, shop, or factory job; self-employment; an agricultural wage or salary job; or some less regulated, less stable, and less secure work. Paid work within the home can be part of self-employment or can be for someone else. Unpaid work outside the home may be volunteer work, neighboring, or community or political action. Unpaid work within the home includes child care and other domesticwork, but also working in a family business. Work in one category may be linked to work in another category at the individual, household, and national levels, as illustrated by the discussion of child care. Further, the same activity may become work in another category when there are changes in the overall work structure, as when many domestic tasks become paid jobs outside the home. While the cross-classification of location and pay highlights these processes, this model also raises problems of measurement, scope, and integration. At least some of these problems are the result of trying to force boundaries on different types and locations of work, insisting on dichotomies in the face of more complex realities. One value of examining the transition from Maoist to reform-era China is that, as with the “inside/outside” distinction, these boundaries are even more fluid than in many capitalist, advanced industrialized societies, where “work” is more—but certainly not completely—segregated as “household” and “nonhousehold.”
This chapter is more conceptual than theoretical. It describes types of work and the linkages among them without pushing on to generalize about what leads to the development of various kinds of connections and how they affect individuals' lives. Some of this theorizing has been done in the women-in-development and informal-sector literature (e.g., Ward 1990a) and some as reactions to the New Home Economics. The discussion in this chapter has focused on the various dimensions of work and on definitional problems—and suggests the need for a multi dimensional approach to
NOTES
I thank Heike Trappe and the conference participants for their useful comments.
1. One textbook on the sociology of work adds that the effort or activity “is considered by the individual to be work” (Hall 1994, 5), emphasizing the subjective nature of this concept. It is not only the individual doing the task, of course, but also the perceptions of others that set the boundaries of activities considered to be “real” work and that potentially affect their value and rewards. I will not extend the definition to efforts to manage interpersonal relationships, such as “emotion work” or “expressive tasks” (England and Farkas 1986; Hochschild 1983), unless this is in the context of providing a formal service, such as psychotherapy or community mediation. Even this example illustrates how difficult it is to set boundaries on which activities are “work.”
2. Home-based work can include activities that are part of domestic or income generating work done mainly in the home, but that actually take place outside the home, such as shopping, doing errands, or distributing home-produced crafts or food.
3. Reduction in mothers' hours was part of East German family policy, although the regular work week there was 43.75 hours and the reduced hours 40 hours a week. Some women balanced responsibility in the work force and the home by temporarily taking a job for which they were overqualified (Trappe and Rosenfeld 1998).
4. The Changing Meanings
of Work in China
Stevan Harrell
All revolutions, including the industrial one, are in a sense about the division of labor—and, of course, the way it relates to ownership of property. But the Chinese revolution, beginning with the self-strengtheners of the late nineteenth century, continuing through the significant failures of the Republic and the grander and more significant failures of collective socialism to the partial return to household-based labor in the present Reform era,[1] has been particularly concerned with issues of who should do what kinds of work. This chapter is about the ways in which arguments about 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—have evolved as part of the broader discussions about what a post-collectivist Chinese society should be like. In doing so, this chapter draws on the materials and insights provided by the authors of the first three chapters in part 1, as well as on material discussed in later chapters.
As Rosenfeld points out in chapter 3, a useful heuristic device is to classify work done in any society in a two-by-two table, divided into work inside and outside the home, and work for compensation or not (see table 3.1 in chap. 3). This gives us four general types of work: remunerated labor outside the home (cell 1); remunerated labor inside the home (cell 2); unremunerated labor outside the home (cell 3) and unremunerated labor inside the home (cell 4). In any given societal arrangement of work roles, differential values will be placed on the four types of work, and different combinations of the four types will be considered appropriate for particular types of persons: men and women, old and young, noble and commoner, and so on.
A classification like Rosenfeld's is an appropriate starting point for trying to sort out how the proper relationships between work, gender, and household have been defined and contested through Chinese history, and particularly in the transition from the collective to the Reformera. I would make only one refinement, which Rosenfeld herself suggests: we should consider the division between household-based and non-household-based production to be at least as important as that between work that physically takes place inside the house or compound and work that takes place outside. We cannot, for reasons that will become apparent, shift our boundary frame all the way from the house to the household, but in order to understand the discussions about gender and household, we need to talk about the household organizational boundary most of the time, and consider the physical walls as the boundary only a little of the time. With this caveat in mind, let us examine how ideas about work have changed in relation to household and gender through the course of the Chinese revolution, and particularly in the transition from the collectivist to the Reform economy.
THE DIVISION OF L ABOR IN PREINDUSTRIAL CHINA
Our historical baseline is conveniently provided by Mann, who shows in chapter 1 that all through the imperial era of Chinese history, scholars and officials thought of the proper organization of labor as household-based. The ideal empire was an accumulation of households that were primarily self-sufficient, though they might be dependent on markets to one degree or another. The role of the laboring individual was that of a partner in a household division of labor, in which those tasks that could be done without leaving the ideational space of the household itself—its house, fields, gardens, and, for some urban families, also its commercial properties— could be performed by the household's female members. Those tasks that required extensive forays into the wider world—into markets, trading centers, or official circles—were restricted to the males. In this division of labor, then, the “inside,” or female domain, referred sometimes to the physical interior of a family's compound, whereas at other times and places it might also include farmwork. Agriculture stood somewhere between “inside” and “outside,” so it was predominantly but not exclusively male, while such things as transport and politics were supposed to be done by men only.
This preindustrial ideal division of labor would place just about all activities into the two bottom cells of Rosenfeld'sgrid:household-based, whether remunerated or not. Ideally all work was household-based, because all property ideally belonged to households. The only exceptions were corvée, military service, officialdom, and quite unusual (and exclusively male) practices of hiring out for labor (usually in transport) and exploiting resources as a collective band (usually in mining). As Ebrey (1984) has shown, many
Within the family enterprise, the distinction between work for direct compensation and not for direct compensation (between cell 2 and cell 4) was really not very clear. Fieldwork was mostly done by men, though how exclusively male it was depended on the region of the country and the season of the year; women in the south tended to go into the fields more than their northern sisters, and where strictures were relaxed, they were of course most often relaxed at planting or harvest time. Mostly male fieldwork combined with mostly female gardening and animal husbandry and exclusively female cooking, cleaning, and sewing to produce livelihood. One side could not exist without the other, as long as people needed to eat as well as have some place to live and someone to prepare the food. Quantitative analysis might have yielded some figures about whether men's or women's labor products were more or less likely to be exchanged rather than consumed directly, but my hunch is that this, too, varied according to the region. Men might have grown only grain exclusively for household consumption, or they might have also grown vegetables or tobacco or cotton for sale; women might have made cloth either to wear or to sell; they might have either cooked or sold the eggs from the family's chickens or the meat from its slaughtered pigs.
As Mann points out, there was one word, lao, that covered every kind of activity in this household-based production system, from the mind work of the scholar to the field labor of the ordinary male householder to the inside labor performed by women. Semantic distinctions of course were made, but there was no kind of productive activity that could not be accommodated under the general term lao.
CHANGES THAT CAME WITH EARLY INDUSTRIALIZATION
One of the most important changes that comprise the historical transition from agrarian to industrial societies is the transition from an economy in which almost everyone lives off of family property to one in which almost everyone lives off of a salary earned in a non-family-based enterprise. When industry and large-scale wage labor came to China's large cities in the beginning of the twentieth century, China was faced for the first time with the
It was conceptually rather easy to create new rules for male labor outside the household context. After all, even though hired labor was not part of the ideal model set forth in Mann's chapter, there were in fact always a fairly sizable minority of males who worked for wages. For women, the problem was somewhat more difficult. Of course there had always been women who worked for wages as servants (Pruitt 1967) or, among the poorest classes, as transport workers, tea pickers, and so on. But women working outside the household context ran into a double ideological barrier: not only were they, like their menfolk, violating the ideal of the household as a self-sufficient productive unit; they were also violating traditional norms of chastity and family honor that placed women inside the house as well as inside the household. Whereas men who worked for wages had the stigma of being poor, women who worked for wages had the added stigma of being immoral. Hershatter (1986) describes the reluctance of Tianjin families to send their daughters to work in the cotton mills at the beginning of the twentieth century because of the possibility of this stigma.
There were compromises to be made, however, as in so many countries in the early stages of industrialization (D. Wolf 1992; Ong 1987). One of the most important ones was allowing young, unmarried women to take lowwage manufacturing jobs (primarily in textiles) where they could earn supplemental income for their households and do so in a cloistered dormitory atmosphere where their chastity was somewhat less likely to be compromised than if they had been allowed to live on their own or if they had only worked 60 or 70 hours per week instead of the somewhat longer week they actually worked.
Still, by the end of the Republican era in China, the overwhelming majority of the population (probably nearly 90 percent) was still rural (Emerson 1971, 188), and most rural people were household-based farmers— either owners or tenants. Even in the largest and most modern city of Shanghai, in 1957, immediately before the large-scale socialist transformation, the urban work force was reported to consist of over 40 percent in self employment or family-operated businesses, and about 60 percent in wage labor or salaried jobs (Howe 1971, 220). The trend toward wage employment was thus on its way by the time of the founding of the People's Republic, but was by no means complete even in Shanghai, and certainly less advanced elsewhere. The greatest changes in the division of labor still lay ahead in 1949.
THE SOCIALIST REVOLUTION
The Communists set out to change the division of labor fundamentally by making two kinds of changes. Production was to be moved out of the household into the public sphere, and all adults, male or female, were to take part in this public-sphere production. This was accomplished in somewhat different ways in the cities and in the rural areas.
In urban China, household-based production was effectively abolished. Family-owned businesses were confiscated, nationalized, or absorbed into collective enterprises, and everyone with a paid job (which, after 1958, was everyone who was not a child, a student, a disabled person, a pensioner, or an illegal migrant), became part of a state or collective work unit, or danwei. These work units not only paid almost all wages in urban China during this time; they also controlled access to many kinds of services that could not be purchased with cash, such as housing, medical care, and schooling (Walder 1986). In addition, women as well as men were expected to engage in wage labor during this time, so that there was no longer, in the ideology at least, a difference between the woman as the “inside” worker and the man as the “outside” worker. Both worked outside for wages—cell 1 of Rosenfeld's table. Cell 4, unremunerated household-based work, however, seems to have remained primarily the province of women, giving them a kind of twoshift existence (Hochschild 1989) similar to those in other industrial countries. In many cases this also kept women from rising to higher-paid or more responsible positions in their work units, because everybody knew they had to cook, clean, and take care of children. Work in cell 3, unremunerated activities outside the household, also became very important at this time, as people were required to engage in political meetings, neighborhood committees, and intermittently, in political campaigns.
Paradoxically, it was cell 2, household-based work to produce exchange value, that practically disappeared in urban China during the collectivist era. This had been the mainstay of the petty capitalist class that had been so important in the development of Chinese urban life for the previous thousand years and had also provided an important engine for China's early industrialization (Gates 1996). Now it was not petit bourgeois but just plain bourgeois to engage in any kind of family business, be it in manufacturing or sales, and commercial sales were taken over by large state stores and their smaller outlets.
The legitimacy of wage labor in state and collective enterprises was symbolized by the issuance of a gongzuo zheng, or work permit, to every worker or retired worker. This served as a personal identification card until universal shenfen zheng, or identity cards, were introduced in the late 1980s. The little red benzi (booklet), as it was unofficially called, served as the passport to all kinds of social services and even to such life necessities as the ration
The household, of course, continued to exist in the collectivist era, and was in fact the basis for registration of the urban population (Cheng and Selden 1994). In addition, many people saw the family during this time as a “haven in a heartless world” (Lasch 1979), a place where one could retreat at least temporarily from the rigors of the politics that perfused life in public places, including workplaces. So the household in collectivist urban China stood in a paradoxical position. It was no longer the site of production or exchange activities, and its members depended on their work connections not only for income but for legitimate status as citizens. But at the same time, the state attempted to use the household as the lowest level of social control in the collectivist system: without an urban household registration, one could not obtain legitimate work, and without work or officially approved retired status, one could not obtain the ration coupons that gave access to life's material necessities. The household thus changed from a place of independent production and the organization of livelihood to a space for the local implementation of state control. This is perhaps best indicated by the fact that household-based work—the bottom half of Rosenfeld's grid—was no longer allowed (cell 2) or no longer respected (cell 4). That is, housekeeping was still necessary but was not considered enough to earn a woman full citizenship; for this she had to have paid, outside work, or gongzuo.
The household economy and division of labor fared somewhat differently in the rural socialist transformation, in which socialism was implemented through the development of collectivized agriculture. This began with the formation of mutual-aid teams in the early 1950s and proceeded to full-scale collectivization across most of the country in 1956 and the formation of people's communes in 1958. For the first time in Chinese history, agricultural production was not centered around the household.[2] Instead, agricultural workers were organized into production teams, consisting of 20–40 households each, and these teams worked their allocated land collectively, with tasks assigned to individual workers or teams of workers every day or every few days (Huang Shu-min 1989, 66–67; Siu 1989, 229–31). Individual workers were awarded “work points” (gong fen) for their labor contribution.
During the heady days of the summer of 1958, China proclaimed to a skeptical world (with no one more skeptical than N. S. Khrushchev) that it had skipped the intermediate stage of socialism and leapt directly forward into communism; all the activities that otherwise would be distributed among
The greatest paradox of collective agriculture, however, was that it was in one sense still household-based. The work points that were assigned on the basis of each household member's work were in fact allocated not to the individuals themselves, but to the households to which they belonged, and the harvest was shared out each fall according to the number of work points accumulated by the household as a whole. And since, in contrast to the urban situation, members of production teams lived and worked side by side, the production team was in the end a collection of households, not a danwei. The distinction was made clear in the contrast between the two kinds of household registration—nongcun hukou, or farm-village registration, and chengshi hukou, or urban registration.[3]
It was very difficult to move from village to urban household registration because there was an implicit hierarchy. Not only were the farmers poorer and less educated on the whole, but in the system there was a built-in bias against farmers, despite all the ideological ballyhoo about going to the countryside to emulate the poor-and lower-middle peasants (Bernstein 1977). This bias was at least partially because the farmers were not entirely modern or socialist—that is, their work was, in effect, not yet wholly included in cell 1 of Rosenfeld's table (remunerated work outside the household). Farmers did not gongzuo —they did not work at paid jobs in state units. Their economy still bore the stigma of production that was partly household based. Even though their economy was socialist in the sense that the means of production were collectively owned, the fruits of labor were distributed to the household, and in other areas, such as support for education and for the elderly and disabled, farm families had to depend on their own resources. They had labor, laodong, and this was ideologically admirable, but they did not have wage jobs, gongzuo. This compounded with their poverty and ignorance to make them literally second-class citizens,
The socialist era thus produced a series of truly revolutionary changes in the division of labor. For one, the same criteria were applied to women's labor as to men's: if it was pursued outside a household-based context and produced social exchange value, it was compensated and valued. If it either was done in the home or did not produce exchange value, it was uncompensated and devalued. This kind of work (cell 4) had to be done anyway and, despite a few desultory campaigns to the contrary that were later discredited by linking them with the hated Jiang Qing, Mao's widow, everyone assumed that women would do it. This meant that women could rarely come up to the same standard of compensation or prestige in their work as men did, whether in a research institution or on an ordinary agricultural production team. In other words, China moved from a vision of a gendered division of labor, where some tasks were appropriate for women and others for men, to ideological denial of the gendered division of labor. Men and women were expected to do the same things, but women were largely incapable of doing them as well because they still had to do those things formerly defined as women's work but now not defined as anything in particular. Instead of men and women each having their own spheres, they now operated in the same sphere, but with women at a disadvantage.
The other revolutionary change was the radical devaluation and virtual elimination of work categorized on the bottom row of Rosenfeld's table, that is, household-based production. This change was only possible if the shift in the parameters of the gender division of labor occurred at the same time, since taking women out of their former role in household production would give them no socially useful role at all unless they also began doing non-household-based work for compensation.
DILEMMAS OF THE REFORM ERA
All this has been reconsidered, and to an extent repealed, during the Reform era. Collective agriculture was the first to go, with the land redistributed to households, in usufruct if not in fee simple, by the early 1980s. Once again, for a short time at least, most inhabitants of rural China were engaging in household-based agriculture. But the division of labor now is not what it was before the great socialist experiments. For one thing, women have not retreated to an “inside” existence, but have remained full-scale agricultural workers in the new family enterprises. For another, rural China did not remain agricultural. The township and village enterprises (xiangcun qiye) that were a small part of the rural scene during socialism now grew
The cities have changed as much, if not more. Perhaps the most striking thing superficially is the profusion of family-owned stores and restaurants that now line the streets of every major and minor city in China. A large proportion—thoughs till aminority— of labor for exchange value is once again based in the household, and cell 2 (income-generating activities within the household) is once again a prominent sort of labor. Aside from this, there has been a rethinking of the gender division of labor. The activities of cell 4 (housework, child care) continue to be predominantly female, but there is a growing sentiment that this is proper and ought to be recognized as such, that women do not really belong in public life. Women's involvement in the workplace is no longer an aspect of their liberation, as it was considered under socialism, but is a response to household economic needs. Thus we see in Zhang's chapter that among the wealthiest families of the Zhejiang inmigrant community in Beijing, women are once again secluded, if not behind the physical walls of the house, at least behind the conceptual walls of the household, reduced to an existence of mahjong, shopping, and ladies' lunches. Most women, of course, continue to work as a matter of necessity or personal fulfillment or both, but it is no longer assumed; traditional ideas about the gender division of labor, though ripped out of their earlier context, can be used to justify not working, and working for less compensation.
Ideas about household-based enterprise and about the gender division of labor are now in great flux. This is reflected in chapter 2, where Henderson, Entwisle, Li, and associates report on their ethno-semantic attempt to determine what people think of as work in two county towns and two nearby villages in Hubei province in the 1990s. In the socialist era, it was easy to define what was gongzuo: it was something that gave you a benzi, which only came when you worked for wages, and wage work was always, of course, outside the household. Uncompensated housework did not count; in fact it did not count to the extent that women could not get away with doing only that work unless they were retired. Farm work did not count either. It was
But now these distinctions are breaking down. People clearly do different kinds of work (as reflected in the stories of the Feng family that Henderson, Entwisle, Li, and associates told to their focus groups), and they do them to different extents and for different employers, including their own households. There is no longer a clear line between gongzuo and householdbased work; both can bring in money. And there is no longer a clear line between gongzuo and farmwork, especially since farmwork is increasingly commercialized. We could say that there are continua from the ideal types of salaried work or large-scale entrepreneurship on one end, to uncompensated housework on the other end of one scale or subsistence farming on the other end of another scale, but there are no clear divisions. No wonder there is disagreement in the 1990s over what constitutes gongzuo.
Whether or not a new consensus will evolve is unclear. Reform-era China has yet to reach agreement on the rules of enterprise, the division of labor, the nature of gender, or the obligations owed to the household or the family descent line. Socialism imposed a consensus, but it was quickly broken when the socialist economy was dispersed. My own guess is that China will go the way of the United States, where these issues remain in the public arena as points of contention between people with different political, economic, and moral views. Eventually, the term gongzuo may revert to something like its presocialist meaning of any kind of labor or profession. But I doubt this will result in general agreement over who should work where, for whom, or how.
NOTES
1. I use the capitalized term “Reform” or “Reforms” to refer to the specific policies known in Chinese as gaige kaifang, implemented in 1979, and to the era in which these policy changes have been carried out.
2. There had been other attempts by central governments to deprive households of property rights in land, notably the “equal field” system of the early Tang dynasty (Elvin 1973, 59–61). But as far as we can reconstruct the way in which this system actually worked, communal property was still allocated to households, who farmed it and paid a portion of the crop to the local treasury.
3. In small towns in southern Sichuan Province where I have conducted research, the common term is actually danwei hukou (work-unit household regisration).