* The author would like to thank officials in the Hygiene and Family Planning Bureaus of Jinan (Shandong), Chengdu (Sichuan) and Liuzhou (Guangxi), who kindly and patiently answered his numerous questions. He also expresses his thanks to: Gérard Chesnel of the French Embassy in Peking, as well as David Bonavia and the staff of the Far Eastern Economic Review in Hong Kong, both of whom volunteered to open their files to him; John Aird in Washington, D.C., and Claude Aubert in Paris for their careful reading and critical comments and suggestions; and finally Lawrence Schulman, who translated the paper into English.

1. Here are some references for the very high rates of the early 1970s and the surprisingly low rates of the late 1970s.

(a) Xinhuashe, Beijing, 3 September 1980, as quoted in Renmin ribao (hereafter RMRB), 4 September 1980.

(b) RMRB, 12 September 1980, p. 4.

(c) Vice–premier Chen Muhua's article (“In order to realize the four modernizations, it is necessary to control population increase in a planned way”) in RMRB, 11 August 1979. For an English translation of Chen's article, see Pi–chao Ch'en, “ Birth planning in China,” International Family Planning Perspectives, Vol. 5, No. 3 (September 1979), pp. 92–101.CrossRefGoogle Scholar

(d) “Report of the state's Statistical Bureau of the P.R.C. on the implementation of the national economic plan for the year 1979,” dated 30 April 1980 (RMRB, 1 May 1980).

During an interview he granted to me on 11 September 1979 in the People's University (Peking), Professor Liu Zheng, head of Institute for Population Research at the People's University, suggested even higher rates for 1970 (birth rate: 38 per 1,000, death rate: 14·6 per 1,000). He reluctantly advanced those estimates, saying that they were “ very approximate and tentative.” Even if one rejects these extremely high rates (by considering that they are unofficial and were communicated orally) and keeps to the written (and official) word, both the 33·59 birth rate and 25·95 natural increase rate for 1970 mentioned in RMRB, 4 and 12 September 1980 (above (a) and (b)) do represent upward revisions of previous figures or estimates.

2. The national population figure of 970,920,000 at year–end 1979 was mentioned in the Report of the State's Statistical Bureau (see supra, note 1, (d)), RMRB, 1 May 1980. China's population is said to have increased by 167,570,000 inhabitants in 10 years (from year–end 1969 to year–end 1979) and by 429,250,000 inhabitants in 30 years (from year–end 1949 to year–end 1979) Xinhuashe, Beijing, 18 April 1980, quoted in RMRB, 19 April 1980).

3. For example, west Chengdu 1972. One need only add 5,000 inhabitants to the population number (475,000 instead of 470,000) in order for the rates to match the number of births and deaths. This correction is justified by the fact that Mrs Li read: 47 wan (47 x 10,000) without going into the details of the additional thousands of inhabitants. It is likewise necessary to add 7,650 to the population of east Chengdu in 1970 (567,650 instead of 560,000), 9,700 to the population of west Chengdu in 1978 (459,700 against 450,000) in order to make the data coherent.

4. For example, Shuangliu 1976. One must change the population from 726,000 to 718,000 so that birth rates and death rates correspond approximately to the respective numbers of births and deaths.

5. For example, east Chengdu 1971. The absolute figures and rates are not consistent. The implied population derived from each pair of values is quite different.

6. For example, Jinbaijiang 1978. The death rate is 4·68 per 1,000 and not 4·15. The implied population derived from a death rate of 4·15 is 125,301 (instead of 111,000). The birth rate is also underestimated, but by a much smaller proportion.

7. For example, west Chengdu 1974· 2,685 deaths make for a death rate of 5·97 per 1,000 and not 5·55. To obtain a 5·55 rate, we would need to assume that the population of west Chengdu is 483,500 instead of 450,000, an assumption which would be justified if the administrative modification which subtracted the Huangtianba neighbourhood from west Chengdu took place in late 1974. But the 8·79 per 1,000 birth rate corresponds well with a population of 450,000.

8. Jingbaijiang death rate in 1978: 4·68 per 1,000 instead of the 4·15 per 1,000 indicated (cf. supra, note 6).

9. The least satisfying figures are those for 1977: plus 0.29 per 1,000 for the birth rate (18·46 against 18·17); minus 0·17 per 1,000 for the death rate (4·39 against 4·56). Because the two mistakes were added together, the natural increase rate should be reduced from 14·07 per 1,000 to 13·61 per 1,000. The implied population derived from an 18·46 per 1,000 birth rate and an absolute number of 8,486 births is 459,697 inhabitants (7,300 less than the population indicated on Table 5); the implied population derived from a 4· 39 per 1,000 death rate and 2,130 deaths is 485,194 inhabitants (18,000 more).

10. For 1978 the implied population derived from an absolute number of 8,649 births and a birth rate of 17·9 per 1,000 is 483,184 (a population total that approximates the figure of 482,000 inhabitants indicated on Table 5). The implied population derived from an absolute number of 2,122 deaths and a death rate of 4·57 is 464,333 (17,667 less than the figure indicated on Table 5).

11. This population increased by nearly 30% (29·09%) in nine years (1970·79). The increase no doubt applies to the population of the whole agglomeration, including the neighbouring countryside, but it is nevertheless illuminating to compare this progression to that of east Chengdu (3·4% between 1970 and 1978) and the even slower one of west Chengdu. The population of west Chengdu increased by 1·44% between 1970 and 1973; after the administrative revision of 1974, it remained stationary. 1 have been given an alternative figure for Liuzhou population in 1979 (490,000 instead of 497,000). If the first figure is not an approximation (49 wan) for the more precise one of 497,000, which was indicated to me afterwards, along with other year by year figures, then the progression of Liuzhou population within the nine–year period is reduced to 27·27% (instead of 29·09%); and the part of the population growth not accounted for by the natural increase is reduced to 53, 947 (or 53,754), instead of 60,947 (or 60,754), as indicated on tne table (difference (a)–(b)): still slightly over half a total growth of 105,000. Since the difference between the two sets of figures (60,947 out of 112,000 on one hand, 53,947 out of 105,000 on the other hand) is too small to invalidate my argument, I chose, for the sake of simplicity, to ignore the alternative figure of 490,000 inhabitants. The calculations which follow are therefore based on a total population of 497,000 inhabitants.

12. One should remember that the rapid fall in birth is supposed to have been produced for the most part before the period for which annual data are available: it includes these six years of slow growth (1964–70) and end as early as 1971.

13. “ Worse ” means, of course, more births. The latter are more likely to be under reported than over–reported, so that any checking entails the risk of exposing unreported births.

14. For example, population of the xian of Yixing (Jiangsu), visited in August 1979: 1,000,000 inhabitants. Population increase: 20,000 inhabitants per year. As a rule, local authorities mention a 20 per 1,000 natural increase rate not for the present period as in Yixing, but for past years (for example before 1976 or during the 1960s). The case is so frequent one cannot help suspecting a reconstruction a posteriori: an unknown rate would be established belatedly so as to match local trends with the norm, which, according to the central leadership, had hitherto defined the country's demographic evolution.

15. For example, people's commune of Zhuliao (xian of Conghua, Guangdong), also visited in August 1979. Population: 53,000 inhabitants, natural increase rate: 14 per 1,000. But a further question brings the information that the population increased by more than 1,200 inhabitants compared to last year; another one, that the commune did not receive any new zhiqing this year. If we reject the hypothesis of a significant immigration, we are led to assume a natural increase rate on the order of 23 per 1,000 (1,200 inhabitants out of 51,800 = 23·35 per 1,000) instead of the 14 per 1,000 announced. The same commune, where health–care leaves much to be desired, announces an infant mortality rate of 5 per 1,000 live births.

16. Yunnan ranks among the lagging provinces insofar as birth control, a situation local officials attribute to the importance of National Minorities.

17. In the model and often–visited people's commune of Sijiqing (Ever–green Commune), north–west of Peking, the following figures concerning the age of newly married couples were given to me.

No man had been authorized to marry before the age of 24, and practically no woman before 23· 85% or 370 out of the 433 women married between 23 and 27 years of age, 13% at an older age. About four out of five men (345 out of 433) married between 25 and 28 years of age, one out of 12 younger, one out of eight older.

18. Here are three examples: “ For the revolution let us have only one child ”; “ In order to realize the Four Modernizations, we must check in a planned way the growth of population ”; “ By carrying through the family planning work, we can speed up the Four Modernizations.”

19. I will send to interested readers brigade by brigade figures concerning contraception and sterilization in Xi'e.

20. In the main clinic in Liuzhou, out of 530 abortions performed between 1 January and 30 June 1979, 509 were done in the first two months of pregnancy. The 21 others were performed (in a different way) after the end of the third month of pregnancy: three during the fourth month, five during the fifth month and 13 during the sixth month (these figures were communicated to me by the director of the clinic in September 1979).

21. Forced abortions, as well as forced female sterilizations, were mentioned in RMRB, 1 February 1980. In the course of a campaign, 4,000 abortions were practised in Chang'an xian (Shaanxi) within a ten–day period in December 1979. Many women fled to their parents' home, in order to escape abortion. An indirect reference that I.U.D.s might also be inserted under pressure was provided by the trial and condemnation of a former brigade doctor, who illegally removed 26 I.U.D.s in Qishan xian and a neighbouring xian (in Shaanxi province) between the spring of 1976 and the summer of 1978. Out of the 26 women who paid the doctor for having the I.U.D. removed, 21 later became pregnant (Foreign Broadcast Information Service, 79–233, 3 December 1979).

22. As the foot of Wutaishan (Shanxi) the hopelessness of parents who first gave birth to two girls seems to have recently led to an outbreak of cases of infanticide: parents mean to thus preserve their chance of having a boy (from a visitor of Shanxi province, October 1979).

23. The objective of reducing the national increase rate to 5 per 1,000 by 1985 and to 0 by 2,000 has been abandoned since the time the present article was written.