300 words essay outline

and now the “one-child” period – we found that the most consistent pattern is one of the “one-child generation” being markedly different from its predecessors, and not just owing to age differences. As interesting, though, was the finding that those who grew up during the

Cultural Revolution period (1966–1976) were not markedly different from the other generations (except the “one-child” group), and were especially similar in most respects to the older generations. As noted in Hung et al., the Cultural Revolution cohort has “often been characterized as the ‘lost’ generation, given the harsh, bitter, violent struggles they endured during their formative years, which did not prepare them for the changing world.”3 During this period, increased communist attacks on Westernism and Confucianism were joined with radical social experiments involving the re-location and “re-education” of nearly 17 million youth. The life circumstances and psychological impacts wrought by such projects have been well-captured in the terms “political chaos,” “civil dis- order,” “extreme poverty,” “‘trauma’ for all perspectives of society,” and yet it was “also an era full of idealism and passion for ‘noble causes’.”4 For the sent- down youth and other adolescents of the period, the traumas and impacts were nothing short of life-changing. “They were told to sacrifice their youth, material comforts and family life for the welfare of the country.”5 But, in spite of spending their adolescence in those harsh and sometimes violent conditions, the “lost” gen- eration does not seem so distinctive after all, at least when compared to their elders, when it comes to political attitudes and behaviour today. In our earlier study, we did not distinguish between those of the Cultural

Revolution generation who actually experienced being “sent down” and those others who observed that phenomenon but did not actually experience it themselves. In other words, an unstated assumption was that all who were socialized during the social experiments of the period would be affected similarly, whether they were sent down or not. In this study, we investigate the veracity of that assumption.

Theory: Direct verus Indirect Experience There are certainly good reasons to expect that being separated from family, friends and the urban environment within which one had been raised would play significantly – perhaps even uniquely – in shaping the rest of one’s life, including orientations and attitudes towards politics and government. Those who did not personally experience these same life-changing events could hardly be affected in exactly the same ways or to the same degree. Literature from the field of psychology dealing with direct versus indirect

experience with a range of types of events/experiences has generally concluded that direct experience is likely to have significantly greater impact on both

3 Hung, Gu and Yin 2007, 839. 4 Egri and Ralston 2004, 212; Yi, Ribbens and Morgan 2010, 604. 5 Sun and Wang 2010, 68.

Attitudinal Differences within the Cultural Revolution Cohort 235

accuracy of recall of the event6 and formation of attitudes related to the event7

than is the case for indirect experience. Niesser et al. suggest this may be due not so much to the stress associated with the event or even its “consequentiality,” but rather to the likely “rehearsal” of telling the narrative of the experience over and over again.8

In the field of history, direct experience is also assigned special significance. Cohen, for instance, notes that “the experienced past is deeply grounded in the senses.”9 He continues by saying that:

Closely related to this sensory aspect of the lived past is the fact that experience encompasses the entire range of human emotions, and the closer our contact with real experience the more peo- ple’s emotional lives – the things that make them sad or angry or nervous or bored, their wor- ries, hatreds, hopes, fears – become foregrounded. We become aware not just of the canal that was built but also of the pain in the backs of men who built it – such awareness being immeas- urably facilitated if we ourselves have experienced a comparable pain.10

Combining these lines of thought, it is reasonable to conclude that longer-term attitudes and behaviour are more likely to be affected by direct experience than by simply observing an event, for two reasons: (1) the heightened emotive aspects of directly participating in the event, and (2) the reinforcement from ongoing “rehearsal” of the experience both in interactions with other participants and even in the telling and retelling of the experience to others who were not themselves participants. The “event” in the case of the sent-down experience in China was – for the

average participant – certainly more long term than the shorter-term duration of an earthquake,11 a classroom experiment12 or even an environmental immer- sion programme,13 and usually spanned multiple years (for some as many as eight years, but probably averaging five to six).14 Given that a youth was literally plucked from his/her family and urban environment and transplanted to a rural peasant village or mountainous region, never knowing whether the new setting would be temporary or permanent, it is certainly reasonable to assume that the impact on the participant’s life would run deeply and broadly and would be sus- tained throughout the rest of his or her life. While those fellow youth who wit- nessed the departures of friends and/or siblings and who would hear the first-hand

6 Murachver et al. 1996; Niesser et al. 1996. 7 Duerden and Witt 2010. 8 Niesser et al. 1996, 338. 9 Cohen 1997, 60. 10 Ibid. 11 Niesser et al. 1996. 12 Murachver et al. 1996. 13 Duerden and Witt 2010. 14 Those who were sent down at the beginning of the Cultural Revolution in 1968 and who remained in

that status until the end of the Cultural Revolution in 1976 would obviously have been sent down for eight years. However, some were sent down in earlier, pre-1968 experiments, and it is impossible to say how long they remained in “sent down” status. Based on one sample of urban residents who had previously been sent down, the average duration was six years (Zhou 2004, 142, fn. 3). Although some voluntarily remained in the countryside beyond the end of the Cultural Revolution, our own study is based on an urban sample from 2006, and hence it is reasonable to think of their sent-down experience as having ended no later than 1976.

236 The China Quarterly, 225, March 2016, pp. 234–252

accounts of the experiences of others – but who were not sent down themselves – may have been impacted as well, it is also reasonable to posit that such impact would have been less traumatic and of less duration than for those who were more than mere observers. The whole “send-down” experiment was premised on the notion that urban

youth could be re-educated by being transplanted into rural, peasant society for a substantial period of time. The re-education would presumably enhance the understanding of the need to develop a classless society,15 and more generally result in acceptance of communist ideology and the communist regime. To the extent that those expectations were met in practice, we could well find that the formerly sent-down youth, as compared to their peers who were not sent down, have developed attitudes that are even more consistent with that ideology and even more supportive of that regime. It is those expectations (or “hypotheses”) that we address in the first set of ana-

lyses below.

Data For our previous comparison of the Cultural Revolution generation with other generations of citizens living in China at the time, we used data from the China Survey of 2008, which included a rich array of items on political attitudes as well as the demographic data needed for identifying the various generations. Unfortunately for our purposes in the remainder of this paper, that survey did not ask respondents whether they had ever been sent down. So, for the following analyses, we rely upon data from the 2006 General Social Survey of China (GSSC).16 This survey included not only a sufficient number of items on political attitudes which could be seen as related to the intended consequences of the rus- tication programme, but also the critical item asking respondents, “did you ever experience being sent down?” Most relevant for our analyses are items asking for respondents’ attitudes on the seriousness of class conflict (on two dimensions: rich versus poor, and workers versus owners) and towards the regime (one item on “blaming” government for poverty and the other on acceptance of the current level of “democracy”17). (See Table 1 for exact wording and descriptive statistics for these four items.) The GSSC of 2006 was a nationwide survey of Chinese citizens. Because the

send-down programme of the Cultural Revolution was focused on urban youth, with most of the sent-down youth eventually returning to urban areas,

15 For discussion of the meaning of class conflict during the Cultural Revolution, see Kraus 1977. 16 The GSSC of 2006 used multi-layered strata sampling and surveyed 10,000 valid respondents. For more

details on the GSSC 2006 survey and associated sampling procedures, see http://www.chinagss.org. 17 We should note that the term “democracy” has different meanings in different contexts. As to what it

means in China today, He and Feng (2008, 166) report that: “The Chinese people have changed their view of democracy from an instrument for producing a strong state to a necessary means for protecting individual liberties … A basic converging point now is rule by the people and rule of law.” See also Lu and Shi 2015 and Lu, Aldrich and Shi 2014.

Attitudinal Differences within the Cultural Revolution Cohort 237

http://www.chinagss.org

Table 1: Wording of Questionnaire Items and Descriptive Statistics

Dependent variable Mean S.D. Measurement N* The main reason for poverty is

due to some inappropriate policies of the government.

2.991 0.719 From strongly disagree (1), disagree (2), agree (3), to strongly agree (4)

1,272

We don’t need to raise the democratic level as long as we have steady economic growth.**

2.619 0.791 From strongly agree (1), agree (2), disagree (3), to strongly disagree(4)

1,247

In your view, how serious is the conflict between rich people and poor people?

2.683 0.932 From no conflict (1), not serious (2), somewhat serious (3), to very serious (4)

1,300

In your view, how serious is the conflict between workers and owners?

2.735 0.847 From no conflict (1), not serious (2), somewhat serious (3), to very serious (4)

1,293

Combined score: poverty + democratic level

5.616 1.070 From 2–8 1,194

Combined score: rich vs. poor + workers vs. owner

5.420 1.424 From 2–8 1,275

Independent variable Sent-down: have you ever been

sent down? 0.183 0.387 Dummy (0 = never; 1 = have) 1,348

Female 0.567 0.496 Dummy (0 = male; 1 = female)

1,348

Duration: returning year minus sent-down year

4.259 3.775 Continuous variable 247

College 0.080 0.272 Dummy (0 = no college education; 1 = have some college education)

1,348

Party member 0.115 0.319 Dummy (0 = non-member; 1 = member)

1,348

Age of the first marriage 24.685 4.639 Continuous variable 1,348 Income 1076.943 1529.479 Continuous variable 999 Retired 0.338 0.473 Dummy (0 = have not retired;

1 = have retired) 1,348

Life satisfaction: in general, how satisfied do you feel about your life?

3.367 0.744 From very unsatisfied (1), unsatisfied (2), neutral (3), satisfied (4), to very satisfied (5)

1,348

Age 49.018 3.222 Continuous variable 1,348 Social status: people are saying

that society can be divided into four different classes; which of the following do you belong to?

1.861 0.879 From peasant class (1), working class (2), middle class (3), to entrepreneur class (4)

1,332

Source: General Social Survey of China 2006.

Notes: *N is for Cultural Revolution generation only (N = 1,348); **see fn. 17 concerning the meaning of “democratic.”

238 The China Quarterly, 225, March 2016, pp. 234–252

and because the send-down question was only put to urban respondents, we ana- lyse only the data for the urban subsample of 6,013 respondents.18

Cultural Revolution Generation: Sent-down versus Not-sent-down Using data from the urban subsample, we find – as anticipated – that those who had been sent down are substantially more likely to buy into the “class struggle” dimension of communist ideology than those who were not sent down. This is reflected in Table 2, in the items dealing with the perceived seriousness of class conflict between workers and owners (p < .05) and between poor and rich (with the direction as expected, although insignificantly so with p = .115).19 To this extent, Mao’s experiment may have succeeded, at least in convincing its sub- jects of the extent to which class struggle was a reality.20

However, our findings pertaining to “loyalty to the regime” are markedly inconsistent with Mao’s intentions. Rather than being more supportive of the sta- tus quo regime, the formerly sent-down youth – as compared to their non-sent-down peers – are today more likely to agree that “the main reason for poverty is due to inappropriate government policy” (p < .01) and less likely to agree that “we don’t need to raise our democratic level as long as we have stable economic growth” (p < .05). In other words, the sent-down portion of the Cultural Revolution generation can actually be seen as less blindly and com- pletely loyal to the communist regime than their non-sent-down peers. But, might these differences be owing less to the sent-down experience itself

than to systematic differences in background of those who were sent down versus those who were not? Zhou and Hou investigated a number of possible differences in the backgrounds of these two groups, and concluded that only the education level of the father was significantly related to the likelihood of being sent down.21

Of our sample of 1,348 from the Cultural Revolution generation (of whom 247 were sent down), three fathers of sent-down youth had a college education, com- pared to four fathers of non-sent-downs.22 As would be expected, removing those seven individuals from the sample makes no significant difference in our findings.

18 Although using a different dataset, different items and just the urban subsample, our findings pertaining to the comparison of the Cultural Revolution generation to other generations are similar to those of our earlier study in that marked differences occur between the Cultural Revolution generation and the younger generation, but not between the Cultural Revolution generation and older generations as a whole.

19 This is also true for perceived conflict between government officials and masses (p<.10). As for the dis- tinction in results between the “workers/owners” and “poor/rich” conceptions of class conflict, it should be noted that in the Chinese context, “class conflict” is traditionally seen as involving different positions in the economic hierarchy rather than the distribution of wealth per se. See, e.g., Kelley and Evans 1995, 158.

20 See Bernstein 1977, 19; Young 1986, 41. 21 Zhou and Hou 1999, 21–22. 22 We should note that our “sent-down” subsample of 247 respondents was being compared to all other

youth who were socialized during the entire Cultural Revolution period, meaning that the latter group is substantially larger than the group who would actually have been sent down. Hence, why the subsample of only 247/1,348 may seem a smaller proportion than might otherwise be expected!

Attitudinal Differences within the Cultural Revolution Cohort 239

Overall, then, the findings above attest to the influence of the sent-down experi- ence, as separable from socialization during the Cultural Revolution, which was shared by both peer groups.23

Even if the consequences of the “Cultural Revolution experience,” per se, var- ied between sent-downs and not-sent-downs, it should not be assumed that the sent-down experience itself was invariable. Indeed, recent literature has suggested that different durations of that experience had long-term life consequences, and

Table 2: Comparison between “Sent-downs” and “Non-sent-downs” (Urban Sample

Sent down and socialized during

the Cultural Revolution

Socialized during the Cultural Revolution but not sent down

p-value Total N

Agree: poverty due to inappropriate government policy

85.78 (232) 77.17 (1,038) 0.004 1,270

Disagree: we don’t need to raise our democratic level as long as we have stable economic growth

65.35 (228) 57.33 (1,017) 0.026 1,245

Seriousness: conflict between poor and rich

64.85 (239) 59.34 (1,060) 0.115 1,299

Seriousness: conflict between workers and owners

68.91 (238) 61.67 (1,054) 0.037 1,292

Source: General Social Survey of China 2006.

Notes: Numbers in parentheses are the number of observations in each respective cohort.

23 In order to assess the independent influence of the Cultural Revolution per se, we might first compare between (1) sent-downs who were socialized during the Cultural Revolution and (2) a second set of sent- downs who were socialized prior to the Cultural Revolution (i.e. during the consolidation phase). Having done so, it would appear that the sent-down experience trumps the period of socialization, in that the only significant difference of attitude that we have found between the two generations of sent- down youth is that sent-downs socialized during the Cultural Revolution are significantly (at .05 level) more likely to blame poverty on “inappropriate government policy.”

It can be argued, though, that the more meaningful distinction is between just sub-categories of the two generational groups: for instance, between those who went down voluntarily or were sent down as exuberant leaders of the Red Guard movement, and those who went voluntarily or were sent down as children of “bad families.” It might reasonably be theorized that such different subsets of sent-down youth would not only have had different backgrounds, but also different experiences in the countryside as well as after returning to urban areas, resulting in different attitudinal consequences today. While this suggests another very interesting research question, addressing it directly is beyond the scope and cap- abilities of this study. Operationalization of the critical distinction between “Red Guard leaders” and those from “bad family backgrounds” would be very indirect and depend exclusively on intersecting Party membership and period of socialization. Those operationalizations would be so indirect as to ren- der suspect any resulting findings.

240 The China Quarterly, 225, March 2016, pp. 234–252

that men and women were affected differently.24 Our purpose in the following sections is to determine whether differential consequences, along both duration and gender lines, apply to political attitudes as well.

Duration of Sent-down Experience Zhou and Hou, using data on “a representative sample of urban residents drawn from a multi-stage scheme in 20 cities in China in 1993 and 1994,”25 conclude that “the longer a respondent stayed in the rural area, the more severe its impact on the life course.”26 Using General Social Survey data for 2003, Qian and Hodson find that the sent-down experience was particularly “traumatic” and dis- ruptive of “life course developments and process of socioeconomic attainment” for “those who stayed in the countryside for an extended period of time.”27

Among the differences, those with shorter periods of being sent down “had more personal income than late returnees” and “were happier about their life!”28

But, whether those different life experiences and attitudes about life in general would translate into – or at least be joined by – different political attitudes remains, until now, an open question. There may be reasons to anticipate such consequences. One might think, for instance, that since longer duration has resulted in greater negative life consequences, greater resentment and thus greater willingness to criticize the government would follow. Alternatively, though, one might think that the longer the sent-down experience, the more one could be socialized into the rural, relatively passive political culture, with the result of les- sened willingness to criticize government or express positive attitudes towards a regime alternative. Our evidence from the 2006 GSSC data (see Table 3), while more consistent with the second of those arguments, does not include a signifi- cant relationship at the .05 level between duration and either willingness to blame the government for poverty or positive attitudes towards democratic reform, although the latter would be significant at the .10 level.29 That is, those who returned after a longer period experiencing peasant life (i.e. more than five years for Table 3) are slightly less likely today to blame government for poverty and to prefer raising the “democratic level,” but not significantly so.

24 Zhou and Hou 1999; Qian and Hodson 2011. 25 Zhou and Hou 1999, 17. 26 Ibid., 28. 27 Qian and Hodson 2011, 205. 28 Ibid., 213. 29 One might at first surmise that the greater “liberalism” of those with a short sent-down experience could

be owing to the fact that this group tends now to have significantly higher education levels than those with a longer sent-down experience, and thus that the relationship between duration and attitudes is actually spurious. However, when we divide those with a short sent-down experience into two groups – those sent down early and those sent down later – the education level of the former group is almost double the level of those in the second group, yet there are virtually no differences between these two subgroups on the relevant political attitudes. A more direct test of the possible effects of education on political attitudes of sent-downs is to divide all sent-downs into two groups: those with and without a college education today. Doing so, we again find no significant differences on the attitudes under investigation here.

Attitudinal Differences within the Cultural Revolution Cohort 241

Similarly, there could be two alternative, and again, contradictory, expecta- tions relating the length of sent-down experience to perception of strong class conflict. One might think that the longer the sent-down experience, the longer the person had to build up resentment of their rural peasant life, as contrasted with recollections of what life had been like in the city, and thus to accept the class-conflict basis of communist ideology. Alternatively, one might argue that the longer the sent-down experience, the more time there was to be fully socia- lized into and accept the peasant lifestyle, while for those whose time in the coun- tryside was too short for that socialization process, the contrast between their brief experience as a peasant and the return to urban life would be even more stark and thus result in a stronger perception of class conflict. Once again, while our evidence (shown in Table 3) tends in the direction of supporting the sec- ond argument, neither of the relationships between duration and perceptions of class conflict are significant, even at the .10 level.30

Our overall conclusion, then, must be that duration of the sent-down experi- ence in one’s youth does not significantly impact political attitudes three decades later.

Table 3: Comparison by Duration of Sent-down Experience (Urban Sample; Sent-downs Only)

Sent down for <6 years

Sent down for >/= 6 years

p-value Total N

Agree: poverty due to inappropriate government policy

86.86 (175) 78.26 (57) 0.409 232

Disagree: we don’t need to raise our democratic level as long as we have stable economic growth

68.39 (174) 55.56 (54) 0.083 228

Seriousness: conflict between poor and rich

66.30 (181) 60.34 (58) 0.409 239

Seriousness: conflict between workers and owners

70.00 (180) 65.52 (58) 0.521 238

Source: General Social Survey of China 2006.

Notes: Numbers in parentheses are the number of observations in each respective cohort.

30 As tests of robustness of the finding of lack of significance for duration of the sent-down experience, we used a number of alternative cut-off points (with “longer duration” operationalized as more than 4 years, more than 3 years, and more than 2 years), and in no case was duration found to be significant (even at the .10 level). We also conducted multivariate regression analyses controlling for the respon- dent’s gender, current income, age at first marriage, whether college-educated, whether retired (as of 2006), perceived social status, reported life satisfaction, age, and Party membership. Again, the contri- butions of duration to explanation of political attitudes and perceived class conflict were statistically insignificant (.10 level). For discussion of the selection of control variables, see the section titled “The ‘experience’ versus later consequences.”

242 The China Quarterly, 225, March 2016, pp. 234–252

Men versus Women Although young sent-down men may well have experienced unwelcome shocks to their educational and career aspirations, women suffered even broader and dee- per consequences that had the potential to affect virtually all aspects of their future lives, including social, economic, and perhaps even political. The young, educated, urban women who were sent down to the countryside, and then were forced to experience life as a traditional peasant woman, were likely to have the most shocking, discomforting, humiliating experience of all. This is, at least partly, because Mao’s programme of equalization for women and men was already beginning to bear fruit in the urban areas but not in the rural areas, where women (and men) were still expected to play traditional roles.31

As Honig notes: “Many sent-down youth had to confront not only the unexpect- ed reality of women’s subordination, but their first experiences of gender inequal- ity as well.”32 Adding to the humiliation was the fact that many young sent-down females were seen as ill-equipped for hard work in the fields, and thus were assessed more as burdens than as helpers, and were treated accordingly.33

Following this line of argument, it would not be surprising to find that women’s attitudes would change even more so than those of men as a result of the sent- down experience during the Cultural Revolution. To address these expectations, we have broken the Cultural Revolution cohort

into four groups: sent-down men, sent-down women, not-sent-down men and not-sent-down women. Comparing first across genders, we find no significant dif- ferences between sent-down men and women (see Table 4), although there are inter-gender differences within the not-sent-down group (see Table 5). Not-sent-down women were more likely than their male counterparts to agree

31 See Jin 2006, 613–14. 32 Honig 2000, 104. 33 Although some of the formerly sent-down girls now recount their impression that sent-down girls and

boys were treated equally, with only local women treated subserviently (see Rae Yang 1995, cited in Honig 2000), others recall things quite differently, and probably more realistically. (Note that Rae Yang had been a leader of the Red Guards.) According to Honig (2000, 104–05): “Zhai Zhenhua, for example, reports her dismay at the discovery that women in the rural area New Yan’an where she had been sent ‘weren’t a major force in production.’ To prepare meals they worked far fewer hours than men, and it appeared to Zhai that during the winter women did not work outside the home at all. The real surprise for Zhai, though, was finding that when women did work the same hours as men, they earned less … ‘Women in the city weren’t discriminated against so overtly. All uni- versity graduates earned the same salary, as far as I knew. To make a living in the countryside brains didn’t really matter. What mattered was strength and sex’ (Zhai 1992, 170–71).” A few paragraphs later, Honig alleges that “sent-down youth were shocked by the persistence of gender inequality in the countryside.”

What is important for our argument is not so much whether equality was the reality in the urban areas, but rather that equality was the perception that many sent-down girls took with them. According to Honig (2000, 107): “Too young to have had work experience themselves, they assumed that urban women were indeed liberated and enjoyed full equality in the workforce.”

As for the so-called “iron girls” – who “did it all” – that was a very limited phenomenon and most sent-down girls would not have been iron girls themselves. As noted in Honig (2000, 99), “The Iron Girls represented only a very tiny percentage of the female population and left unaltered (and uncontested) the status of the vast majority of women.”

Attitudinal Differences within the Cultural Revolution Cohort 243

that the democracy level does not need to be raised (p < .10) and were less likely than the men to see substantial conflict between workers and owners (p < .05). Comparing within gender sent-down versus not-sent-down, we find no signifi-

cant attitudinal differences between the two groups of men, but significant differ- ences between the two groups of women (see Table 6). The sent-down women were significantly more likely than their not-sent-down counterparts to blame poverty on government policy (p < .01), to disagree that the democratic level does not need to be raised as long as economic growth remains stable (p < .05), and to perceive substantial conflict between workers and owners (p = .01). Drawing the findings together, it seems reasonable to speculate that the differ-

ences that exist between sent-down and not-sent-down groups overall – with sent-downs more likely to perceive class conflict but less “blindly loyal” to the regime – are owing primarily to changes that occurred in the women’s attitudes (but not the men’s) because of their sent-down experiences. We have now found,

Table 4: Comparison between Male and Female Sent-downs (Urban Sample)

Male Female p-value Total N

Agree: poverty due to inappropriate government policy

83.17 (101) 87.79 (131) 0.318 232

Disagree: we don’t need to raise our democratic level as long as we have stable economic growth

65.00 (100) 65.63 (128) 0.922 228

Seriousness: conflict between poor and rich 65.71 (105) 64.18 (134) 0.805 239 Seriousness: conflict between workers and

owners 66.99 (103) 70.37 (135) 0.577 238

Source: General Social Survey of China 2006.

Notes: Numbers in parentheses are the number of observations in each respective cohort.

Table 5: Comparison between Male and Female Non-sent-downs (Urban Sample)

Male Female p-value Total N

Agree: poverty due to inappropriate government policy

77.36 (455) 77.02 (583) 0.895 1,038

Disagree: we don’t need to raise our democratic level as long as we have stable economic growth

60.27 (448) 55.01 (569) 0.092 1,017

Seriousness: conflict between poor and rich 58.46 (455) 60.00 (605) 0.614 1,060 Seriousness: conflict between workers and

owners 66.00 (450) 58.44 (604) 0.013 1,054

Source: General Social Survey of China 2006.

Notes: Numbers in parentheses are the number of observations in each respective cohort.

244 The China Quarterly, 225, March 2016, pp. 234–252

Table 6: Comparison between Sent-downs and Non-sent-downs by Gender (Urban Sample)

Male Female

Sent-downs Non-sent-downs p-value Sent-downs Non-sent-downs p-value Agree: poverty due to inappropriate government policy 83.17 (101) 77.36 (455) 0.199 87.79 (131) 77.02 (583) 0.006 Disagree: we don’t need to raise our democratic level as long as we have

stable economic growth 65.00 (100) 60.27 (448) 0.380 65.63 (128) 55.01 (569) 0.028

Seriousness: conflict between poor and rich 65.71 (105) 58.46 (455) 0.172 64.18 (134) 60.00 (605) 0.370 Seriousness: conflict between workers and owners 66.99 (103) 66.00 (450) 0.848 70.37 (135) 58.44 (604) 0.010

Source: General Social Survey of China 2006.

Notes: Numbers in parentheses are the number of observations in each respective cohort.

A ttitudinalD

ifferences w ithin

the C ulturalRevolution

Cohort 245

term s o

f u se, availab

le at h ttp

s://w w

w .cam

b rid

g e.o

rg /co

re/term s. h

ttp s://d

o i.o

rg /10.1017/S0305741015001678

D o

w n

lo ad

ed fro

m h

ttp s://w

w w

.cam b

rid g

e.o rg

/co re. U

n iversity o

f Liverp o

o l Lib

rary, o n

30 Sep 2017 at 15:34:56, su

b ject to

th e C

am b

rid g

e C o

after all, that sent-down and not-sent-down men do not differ on the items being analysed, while the two groups of women do differ, with the sent-down women being more willing to criticize the government, to see merit in democratization and more likely to see the class struggle as being relevant today. We also find that women and men in the not-sent-down group did differ significantly on two of the items. We find it plausible that it is the change in attitudes among sent- down women as a result of that experience that resulted, ultimately, in women having essentially the same attitudes as men among the group with the sent-down experience. We recognize the limitations of developing what is in effect a longitudinal

argument to explain what we are seeing in cross-sectional data. Nonetheless, we find the argument to be compelling, logical and certainly consistent with the empirical findings.

The “Experience” versus Later Consequences Thus far, we have found evidence to support a conclusion of long-term attitu- dinal consequences of having been sent down, at least for some of the former sent-down youth, but we have not yet addressed whether those consequences are owing to the experience itself or rather attributable to other, social/economic impacts of the experience. If the formerly sent-down youth suffered negative life course consequences as a result of the forced experience in the countryside, it is possible that it is those life consequences, rather than psychological trauma asso- ciated with the sent-down experience itself, that are responsible for any differ- ences in political attitudes found to exist today. Qian and Hodson have noted that, while in the countryside for as many as ten

years, the sent-down youths’ “lives and careers were significantly delayed or dis- rupted, including delayed marriage, disrupted education, and lost job experi- ence.”34 Qian and Hodson found from a 2003 survey that those who had been sent down were not significantly less well-off financially, but they were more likely to have faced forced early retirements than was true for others of their generation.35

This burden represents the convergence of lasting ripple effects of the send-down experience: the years spent in rural areas did not contribute to accumulated job skills or enterprise specific seni- ority setting up the send-down generation for heightened risk of layoffs during the downsizing and privatization periods starting in the 1990s and beyond. These findings are consistent with those of Zhao and Zhou (2004) who found that the generation of the Cultural Revolution had lower promotion rates than adjacent cohorts. In aggregate, these negative outcomes for both men and women reinforce the image of the sent-down youth as “the unlucky generation” that faced privation both in their youth and again at the end of their careers.36

34 Qian and Hodson 2011, 205. 35 Ibid., 216. 36 Ibid. What may have been true generally for the formerly sent-downs was especially true for those who

spent longer periods in the countryside. According to Zhou and Hou (1999, 32): “The send-down epi- sode had an especially large and negative effect on those who experienced extended rural durations.

246 The China Quarterly, 225, March 2016, pp. 234–252

Qian and Hodson sum up the reality for many of the formerly sent-down youth: “They had to work harder, get more education, start their families later, and face early retirement.”37

With regard to gender differences, Qian and Hodson report that “young women during the Cultural Revolution indeed held up more than half of the sky – women were more likely to be sent down in the first place and less likely to return home within 5 years.” Hence, women of the “unlucky generation” are even more likely than men to have experienced the dual privation.38

Given the later social and economic consequences of their sent-down experi- ence, it is certainly reasonable to think that atypical disillusionment with the sta- tus quo regime and continued perceptions of class struggle might stem at least as much – if not more, perhaps even exclusively – from these later life consequences of the sent-down experience than from lingering unhappiness over the experience itself. After all, extant literature on China has reported a significant relationship between material life satisfaction and support for the current regime.39 To assess whether the sent-down experience itself has a significant impact on attitudes today pertaining to the regime and to ongoing class struggle, net influence of sub- sequent social and economic consequences, we have conducted multivariate ana- lyses which include as independent variables not only whether the subject had been sent down and the respondent’s gender, but also indicators of post-sent-down life course: current income, age at first marriage, whether college-educated, whether retired (as of 2006),40 perceived social status, and reported life satisfaction. Party membership is also controlled because CCP mem- bership could be important in shaping respondents’ attitudes towards govern- ment, democracy and social class. The results of these analyses are reported in Tables 7 and 8. In the most important respects, the findings of these multivariate analyses are

consistent with findings from cross-tabulation analyses as reported above. None of the control variables is significant in any of the models. Having been sent

footnote continued

Their marriage and childbearing were significantly delayed. When they returned to the cities, they had less advantageous locations in the urban labor force (with respect to type of occupations and work orga- nizations) compared with those who had a shorter rural duration.” In spite of Zhou and Hou’s finding regarding delay of first marriage, Qian and Hodson (2011, 213) report finding no significant difference, by duration, on this variable. Because “duration of sent-down experience” is applicable only to the sent- down respondents, it is not included in these multivariate analyses comparing sent-downs to not-sent-downs. However, the results of multiple regression analyses for just the sent-down subsample, including duration and all of the other control variables listed here, are discussed above in fn. 30.

37 Qian and Hodson 2011, 216. 38 Ibid., 212. 39 Chen, Zhong and Hillard 1997, 59. 40 The oldest respondents in our dataset covering the Cultural Revolution generation were 54 years old at

the time of the survey in 2006. Given that the normal retirement age in China is 60, those retired at the time of the survey would be considered subject to early retirement; whether retirement was “forced” or not cannot be determined from the existing data.

Attitudinal Differences within the Cultural Revolution Cohort 247

Table 7: Regression Estimates for Political Attitudes

Agree: poverty due to inappropriate government policy

Disagree: we don’t need to raise our democratic level as long as we have stable economic growth

Coeff. s.e. Coeff. s.e. Coeff. s.e. Coeff. s.e. Sent-down 0.370 (0.172) 0.244 (0.228) 0.318 (0.196) 0.281 (0.311) Female 0.250 (0.175) 0.203 (0.199) −0.076 (0.174) −0.088 (0.188) Sent-down X Female 0.680 (0.262) 0.260 (0.268) College education 0.228 (0.301) 0.236 (0.302) 0.472 (0.328) 0.474 (0.329) Party member −0.110 (0.225) −0.124 (0.225) 0.285 (0.252) 0.281 (0.252) Age at marriage 0.004 (0.017) 0.004 (0.017) 0.001 (0.018) 0.001 (0.018) Income 0.010 (0.100) 0.010 (0.099) −0.044 (0.096) −0.044 (0.096) Retired 0.071 (0.169) 0.064 (0.168) −0.246 (0.172) −0.249 (0.172) Life satisfaction −0.190 (0.125) −0.196 (0.125) −0.034 (0.129) −0.036 (0.129) Social status −0.141 (0.098) −0.142 (0.098) 0.002 (0.094) 0.002 (0.094) Intercept Intercept (1/2) −4.026 (0.624) −4.075 (0.633) −2.803 (0.615) −2.817 (0.621) Intercept (2/3) −1.769 (0.599) −1.819 (0.606) −0.447 (0.615) −0.460 (0.621) Intercept (3/4) 0.776 (0.599) 0.727 (0.605) 2.067 (0.612) 2.053 (0.618) Number of obs. 944 924

Source: General Social Survey of China 2006.

Notes: Variables that are significant at the .05 level are highlighted in bold. The models (DV = poverty and democratic level) are estimated by the ordered logistic regression with survey weights because these dependent variables

are ordinal with four-point scale.

248 The

C hina

Q uarterly,

225, M arch

2016, pp.

234 –252

term s o

f u se, availab

le at h ttp

s://w w

w .cam

b rid

g e.o

rg /co

re/term s. h

ttp s://d

o i.o

rg /10.1017/S0305741015001678

D o

w n

lo ad

ed fro

m h

ttp s://w

w w

.cam b

rid g

e.o rg

/co re. U

n iversity o

f Liverp o

o l Lib

rary, o n

30 Sep 2017 at 15:34:56, su

b ject to

th e C

am b

rid g

e C o

Table 8: Regression Estimates for Perceived Class Conflict

Seriousness: conflict between poor and rich Seriousness: conflict between workers and owners

Coeff. s.e. Coeff. s.e. Coeff. s.e. Coeff. s.e. Sent-down 0.113 (0.186) −0.117 (0.271) 0.329 (0.180) −0.087 (0.265) Female 0.018 (0.153) −0.062 (0.170) 0.062 (0.156) −0.091 (0.179) Sent-down X Female 0.256 (0.264) 0.617 (0.253) College education −0.012 (0.245) 0.001 (0.244) 0.131 (0.244) 0.148 (0.243) Party member 0.289 (0.211) 0.264 (0.211) 0.245 (0.175) 0.191 (0.180) Age at marriage −0.023 (0.018) −0.023 (0.018) 0.006 (0.016) 0.005 (0.016) Income −0.008 (0.042) −0.008 (0.042) −0.026 (0.066) −0.025 (0.062) Retired 0.092 (0.153) 0.076 (0.154) 0.145 (0.154) 0.110 (0.157) Life satisfaction −0.088 (0.113) −0.103 (0.115) −0.066 (0.115) −0.088 (0.117) Social status 0.003 (0.092) 0.000 (0.092) −0.107 (0.096) −0.111 (0.095) Intercept Intercept (1/2) −2.812 (0.663) −2.919 (0.683) −2.548 (0.664) −2.751 (0.689) Intercept (2/3) −1.167 (0.634) −1.272 (0.651) −0.576 (0.608) −0.774 (0.629) Intercept (3/4) 0.733 (0.635) 0.632 (0.651) 1.616 (0.600) 1.431 (0.619) Number of obs. 953 950

Source: General Social Survey of China 2006.

Note: Variables that are significant at the .05 level are highlighted in bold. The models (DV = rich vs. poor, and workers vs. owners) are estimated by the ordered logistic regression with survey weights because these dependent

variables are ordinal with four-point scale.

A ttitudinalD

ifferences w ithin

the C ulturalRevolution

Cohort 249

term s o

f u se, availab

le at h ttp

s://w w

w .cam

b rid

g e.o

rg /co

re/term s. h

ttp s://d

o i.o

rg /10.1017/S0305741015001678

D o

w n

lo ad

ed fro

m h

ttp s://w

w w

.cam b

rid g

e.o rg

/co re. U

n iversity o

f Liverp o

o l Lib

rary, o n

30 Sep 2017 at 15:34:56, su

b ject to

th e C

am b

rid g

e C o

down is by itself significant in the first model for “poverty due to inappropriate government policy.” But, the most significant impacts are owing to the inter- action of having been sent down and being female for both the latter dependent variable and the perception of serious conflict between workers and owners. These findings from multivariate analysis suggest that the sent-down experience itself has affected the attitudes of formerly sent-down youth, and especially for- merly sent-down females, even when controlling for other economic and social consequences of that experience.41

Conclusion If the “send-down” programme resulted in the re-education that was first envi- sioned and intended by Mao’s regime, and if those consequences had a lasting effect, then it might well be expected that the formerly sent-down youth – as com- pared to their peers who were not sent down – would today have attitudes even more consistent with that regime’s class-based ideology and be even more sup- portive of the regime itself. Our findings here are consistent with the expectation that those who were sent down are more likely to buy into the “class struggle” dimension of communist ideology,42 but not with the expectation that they would be more loyal to the current communist regime.43 Indeed, those who were sent down during the Cultural Revolution are actually less blindly and completely loyal to the status quo, communist regime than their non-sent-down peers. From the standpoint of psychological theory positing differential impacts

41 To increase confidence in these findings, given the disproportionally smaller percentage of sent-downs in our sample, we also used matching with propensity scores, which reduces observational biases in non- experimental survey studies. The matching analysis produced similar results to those reported in Tables 6, 7 and 8. There is one notable exception: when compared to the multiple regression results in Table 7, the contribution of “sent-down” is actually significant at the .05 level for “Disagree: we don’t need to raise our democratic level as long as we have stable economic growth” when propensity score matching is used, thus lending even more support for our general conclusions.

42 Although we have until now treated seriousness of the poor/rich conflict and seriousness of the worker/ owner conflict as two separate concepts, an alternative approach would be to treat them as two dimen- sions of a single concept, i.e. class conflict. Under the latter approach, the two indicators would be com- bined into a single measure. Separately from the analyses just reported, we have simply added the two variables together to produce the combined measure, and have subjected that measure to multivariate regression analysis incorporating all of the independent and control variables listed in Table 8. The result is similar to that for “seriousness: conflict between workers and owners” alone for the second model, with “sent-down X female” the only significant contributor. However, none of the independ- ent/control variables (including “sent-down”) is significant in the model that does not include “sent- down X female.”

43 Although we have until now treated willingness to criticize the government and preference for increasing the level of democracy as two separate concepts, an alternative approach would be to treat them as two dimensions of a single concept, i.e. willingness to reconsider the political status quo. Under the latter approach, the two indicators would be combined into a single measure. Separately from the analyses just reported, we have simply added the two variables together to produce the combined measure, and have subjected that measure to multivariate regression analysis incorporating all of the independent and control variables listed in Table 7. The result is similar to that for “poverty due to inappropriate government policy” alone, with the coefficient for “sent-down” being the only significant contributor for the model without “sent-down X female,” and “sent-down X female” the only significant contribu- tor (although only at the .10 level) for the second model.

250 The China Quarterly, 225, March 2016, pp. 234–252

of direct experience versus indirect knowledge, our findings of different political attitudes many years after the sent-down event are largely consistent with that theory. Not content to assume that all sent-down experiences are alike, and hence that

all consequences would be the same, we further explored possible attitudinal dis- tinctions caused by different durations of sent-down experiences and the different experiences of sent-down men versus women. Although the different lengths of stay in the countryside did not result in significantly different attitudes, we do find evidence to support the argument that the attitudinal impact of the sent- down experience was greater for women than for men. While significant differ- ences exist between men and women who were not sent down, there are no such gender differences among those who were sent down. We find it plausible to speculate that the sent-down women’s attitudes became more like their male counterparts – i.e. more willing to criticize the government, to see merit in a higher “democratic level,” and to see the class struggle as relevant today – as a consequence of their particularly shocking and discomforting sent-down experience. The end result is a group of formerly sent-down youth – and particularly sent-

down women – which, as compared to not-sent-down peers, is today more willing to accept the class-struggle foundation of Mao’s communist ideology but which, at the same time, is more willing to assess the performance and structure of the communist regime critically. Furthermore, results of multivariate analysis suggest that, even when controlling for other social and economic consequences of hav- ing been sent down, the sent-down experience itself significantly impacted these attitudinal differences.

摘摘要要: 这篇文章研究在文化大革命时期上山下乡的经验是否会造成民众政治态度的变迁。我们使用 2006 年的中国综合社会调查, 比较经历过上山下乡的群众以及那些同时在文革时期社会化的人。我们的分析发现与那些

在文革时期社会化但没有经历过上山下乡运动的人比较, 经历过上山下乡的人们, 尤其是女性, 比较容易接受毛泽东所谓的阶级冲突下的共产主义的概念; 但也同时发现, 上山下乡的民众更会对共产政体的表现以及其结构进行批判。

关关键键词词: 中国; 文化大革命; 上山下乡运动; 政治态度; 性别差异; 民意调查

References Bernstein, Thomas P. 1977. Up to the Mountains and Down to the Villages: The Transfer of Youth from

Urban to Rural China. New York: Yale University Press. Billing, Chris. 2005. “Up to the mountain, down to the village.” Documentary film. Washington, DC:

Small Handful Productions. Chen, Jie, Yang Zhong and Jan William Hillard. 1997. “The level and sources of popular support for

China’s current political regime.” Communist and Post-Communist Studies 30(2), 45–64.

Attitudinal Differences within the Cultural Revolution Cohort 251

Cohen, Paul A. 1997. History in Three Keys: The Boxers as Event, Experience, and Myth. New York: Columbia University Press.

Duerden, Mat D., and Peter A. Witt. 2010. “The impact of direct and indirect experiences on the development of environmental knowledge, attitudes, and behavior.” Journal of Environmental Psychology 30, 379–392.

Egri, Carolyn P., and David A. Ralston. 2004. “Generation cohorts and personal values: a compari- son of China and the United States.” Organization Science 15(2), 210–220.

Harmel, Robert, and Yao-Yuan Yeh. 2015. “China’s age cohorts: differences in political attitudes and behavior.” Social Science Quarterly 96(1), 214–234.

He, Kai, and Huiyun Feng. 2008. “A path to democracy: in search of China’s democratization model.” Asian Perspective 32(3), 139–169.

Honig, Emily. 2000. “Iron girls revisited: gender and the politics of work in the Cultural Revolution, 1966–76.” In Barbara Entwisle and Gail E. Henderson (eds.), Re-drawing Boundaries: Work, Households, and Gender in China. Los Angeles: University of California Press, 97–110.

Hung, Kineta H., Flora Fang Gu and Chi Kin (Bennett) Yin. 2007. “Social institutional approach to identifying generation cohorts in China with a comparison with American consumers.” Journal of International Business Studies 38, 836–853.

Jin, Yihong. 2006. “Rethinking the ‘iron girls’: gender and labour during the Chinese Cultural Revolution.” Gender & History 18(3), 613–634.

Kelley, Jonathan, and M.D.R. Evans. 1995. “Class and class conflict in six Western nations.” American Sociological Review 60(2), 157–178.

Kraus, Richard Curt. 1977. “Class conflict and the vocabulary of social analysis in China.” The China Quarterly 69, 54–74.

Lu, Jie, John Aldrich and Tianjian Shi. 2014. “Revisiting media effects in authoritarian societies: democratic conceptions, collectivistic norms, and media access in urban China.” Politics and Society 42(2), 253–283.

Lu, Jie, and Tianjian Shi. 2015. “The battle of ideas and discourses before democratic transition: dif- ferent democratic conceptions in authoritarian China.” International Political Science Review 36(1), 20–41.

Murachver, Tamar, Margaret-Ellen Pipe, Rachael Gordon, J. Laurence Owens and Robyn Fivush. 1996. “Do, show, and tell: children’s event memories acquired through direct experience, observa- tion, and stories.” Child Development 67, 3029–44.

Niesser, Ulric, Eugene Winograd, Erik T. Bergman, Charles A. Schreiber, Stephen E. Palmer and Mary Susan Weldon. 1996. “Remembering the earthquake: direct experience vs. hearing the news.” Memory 4(4), 337–357.

Qian, Zhenchao, and Randy Hodson. 2011. “‘Sent down’ in China: stratification challenged but not denied.” Research in Social Stratification and Mobility 29, 205–219.

Sun, Jiaming, and Xun Wang. 2010. “Value differences between generations in China: a study in Shanghai.” Journal of Youth Studies 13(1), 65–81.

Yi, Xiang, Barbara Ribbens and Caryn N. Morgan. 2010. “Generational differences in China: career implications.” Career Development International 15(6), 601–620.

Young, Graham. 1986. “Mao Zedong and the class struggle in socialist society.” The Australian Journal of Chinese Affairs (16), 41–80.

Zhou, Xueguang. 2004. The State and Life Chances in Urban China: Redistribution and Stratification, 1949–1994. New York: Cambridge University Press.

Zhou, Xueguang, and Liren Hou. 1999. “Children of the Cultural Revolution: the state and the life course in the People’s Republic of China.” American Sociological Review 64(1), 12–36.

252 The China Quarterly, 225, March 2016, pp. 234–252