93

at the expense of a substantial increase in unit cost (to ¥21.70 for dry field and ¥14.84 for paddy plows). Still, the investment was small relative to the value of labor savings, and these plows were evidently well received.

94

While the improved traditional plow was a product of handicraft industry, modern industry showed its willingness to serve agriculture by producing large numbers of double-wheel, double-share (abbreviated DWDS) plows, which were copies of Soviet and Polish models best suited to dry field cultivation, but adaptable to paddy conditions. This animal-powered, all-metal implement was far more costly than traditional or improved plows (initially ¥90).95 Yet, even if one accepts advertised capabilities, it is not clear that the DWDS plow was superior to the improved plow, except in working difficult or unbroken soils, where the greater weight would count in its favor. The depth of plowing was comparable, if more even; the greater breadth of furrow (and hence speed) was offset by the larger pulling requirement-two or three animals, as opposed to one or two for the improved traditional plow.96 Still, in plains areas with well laid out fields and adequate numbers of labor animals (or a shortage of human labor), the greater speed of plowing and the durability of the implement were advantageous. Moreover, continuous price reductions (to ¥61.50 in 1956, ¥27.30 in 1974 97) expanded the market, and the fact that the plow is still produced today attests to its ultimate profitability to at least some north China purchasers.

Nevertheless, promotion of the same plow for use in paddy cultivation in south China proved to be one of the major blunders of Chinese agrotechnical policy. In 1956-58, a vast number of DWDS plows were produced for this purpose, but rice-growing peasants reacted negatively to them. Inventories at distribution centers piled up, and many plows that were sold were later returned unused. In fact, the 40-percent price reduction in 1956 was a direct reaction to the resulting surplus stock (and, ironically, accounts for much of the improved terms of trade of the peasantry during the first 5 year plan).98 While this mistake in policy has commonly been attributed to blind commandism or technical stupidity, these explanations will not suffice.

There were strong political reasons for promoting the DWDS plow: The policy of promoting rapid cooperativization from 1956 on made it necessary to demonstrate to the peasants that the collective organization could improve their livelihood substantially. This could result from the use of the large-scale collective labor force, as in the irrigation projects discussed above, or from the larger accumulation of funds, making more expensive capital equipment accessible for the first time. But at the time the DWDS plow was virtually the only expensive capital item that could be produced and distributed in

93 Price averages, descriptions, and technical characteristics of plows tabulated by the author from PRC, Ministry of Agriculture, Nung-chi t'u-p'u (Peking, 1958) and 2d Ministry of Light Industry, Agricultural Implements Bureau, Nung-yeh chi-chi ts'ung-shu, vol. I (Peking, 1966). On improved implements, see also Amano Motonosuke, "Daiyakushin ki no nōgu," Ajia Keizai 14: 12 (1973).

94 Ministry of Agriculture (1958), op. cit., pp. 25-27.

85 Central Intelligence Agency, Prices of Machinery and Equipment in the P.R.C. (Washington, D.C.: 1975), p. 14.

1957 claims indicated labor savings of 28+ percent, but 1964 materials indicate no labor savings (human plus animal) at a ratio of animal-to-human labor cost of 3 to 1. See Ministry of Agriculture, Nung-yeh chi-chủ (Peking, 1964), p. 40. An article in JMJP, June 16, 1958, suggested that, with only two or weak animals, the DWDS plows could hardly plow more deeply than even traditional plows.

97 Central Intelligence Agency (1975), op. cit., p. 14.

See JMJP, Oct. 7, 1955 and Feb. 5, 1958.

quantity (tractor production began only in 1958). Thus, if benefits could be demonstrated, there was reason to produce and promote this implement.

Experiments in paddy cultivation in 1954-55 did convince the leadership that the DWDS plow could be beneficial, but they erred on the side of optimism in extrapolating from experimental to field conditions: (1) The traditional paddy plow required one buffalo, whereas the DWDS plow required two. Animals had to be trained to "cooperate," and plowmen to manage them. (2) In theory, the DWDS plow required less pulling power than did two traditional plows, but this was not the case in practice. Pulling requirements could be reduced by modifying the DWDS plow, but initially, these modifications were left to the purchasers to make. (3) Paddy fields in practice proved too short for efficient use and the paths between them too narrow to transport the plow. It was optimistically assumed that cooperativization would induce peasants to modify these conditions but the costs involved apparently were not considered. (4) Efficiency improved with use and training: Plowmen who were new to the implement were discouraged when they could not reach the rated speed of plowing. (5) The economic efficiency of DWDS plows was predicated on labor shortages that were expected to result from a rapid expansion of multiple-cropping and fertilizer collection efforts. These changes in technique were not realized in the short run.99 (6) Without modification, the DWDS plow could be used only on a hard-pan paddy, and not in mountainous areas, terraced fields, or muddy (i.e., marshy) fields, nor in areas where animals were scarce. This eliminated, for example, some 55 percent of the cultivated acreage in Chekiang. (7) The experimental results implied increases in yields of more than 14 percent over those due to plowing with traditional plows, but in large-scale comparisons in field conditions, the average increase fell to 5 percent.1

While many of these problems could have been eliminated eventually, it is still true that the DWDS plow offered at best only limited labor savings and a decreased (although more even) depth of plowing compared to the improved paddy plow, despite the much higher cost of the former. In combination with the loss (or decreased strength) of animals and shortages of steel as the Great Leap collapsed, this probably doomed the promotion of this implement for paddy use. By the time recovery was complete, China was in a position to move on to a more advanced stage of mechanization of rice cultivation. In view of the high opportunity-cost of maintaining or expanding the draft animal stock (in terms of grain or meat consumption sacrificed), the direction of advance could only be toward tractorization, but at first this did not seem feasible for paddy cultivation. To be sure, 25-35 horsepower tractors could be modified for use with reasonable efficiency in hard-slab paddy fields if the field size was enlarged to about two-thirds of a hectare per plot, but these conditions, as well as cost considerations, restricted applicability to a small percentage of the acreage in the mountainous south.

JMJP, Feb. 5, 1958.

1 Ibid., and JMJP, Jan. 26, 1958.

3

My calculations imply none at all at a 3:1 ratio of animal-to-human labor cost. Figures drawn from Ting Ying (1961), op. cit., p. 632.

3 Ibid. pp. 626-27.

In response to similar needs, by the late 1950's, rotary power tillers were being produced and distributed in Japan, and China was quick to test their applicability. The initial conclusion was negative, even though labor productivity was three times as great as when the traditional plow was used, quality of work was considerably higher, and the time utilization rate was quite high regardless of the size of the field. The reason for the rejection was clear: Increased plowing depth was considered crucial to the yield-increasing strategy of the Chinese, yet the power tiller cultivated to a depth of only 5 inches on average-just slightly better than the traditional paddy plow and worse than the improved paddy plow (4 and 6 inches, respectively).* Attitudes toward the power tiller changed in the 1960's, however, and by 1966, production was being rapidly expanded; indeed, the spread of the power tiller has been the most notable feature of Chinese tractorization ever since.5 The policy reversal reflected a further emphasis on increased multiple cropping in the south, even at the cost of loss of yields due to a reduced depth of plowing. Double or triple cropping of rice created an overwhelming labor time constraint, which made economical not only the power tiller but also, in a very few areas, the mechanical transplanter, an indigenous innovation that also supposedly saves labor time without directly contributing much to yields.'

The change in attitude toward the power tiller should not be seen as a correction of a previously erroneous policy, however, but rather as the result of a relaxation of the constraints limiting the expansion of multiple cropping in the 1950's, due to the development of seeds with shorter growing seasons, increased fertilizer supplies, improved means of water control, and better organizational techniques. Under these changed conditions, a sacrifice of some plowing depth may be economically efficient.

Despite the rapid growth of the tractor park (for example, Shantung, a relatively advanced province in mechanization, had one tractor for every two brigades by 1974, and hopes to double this ratio by 1980), it is a curious fact that foreign visitors observe tractor cultivation more frequently in the paddy fields of the south than in the dry fields of the north, and in both areas, animal plowing will be seen in the fields even while many tractors are engaged in transport work on the roads. It might be surmised (on the "peasant knows best" theory) that transportation is a more serious production bottleneck (or greater source of profit) than plowing, that is, that tractors can make a greater indirect contribution to yields by freeing labor and draft animals absorbed in moving inputs, crops, and people during the critical crop turnover period than by direct use in plowing. But if this were so, why is the Chinese Government currently chastis

4 Ibid., p. 628; Matsubayashi Minoru et al., op. cit., pp. 382-85. One pass with a power tiller is equivalent to one plowing and two harrowings with traditional implements. But power tillers are also restricted to hard-slab paddy fields. 5 Central Intelligence Agency, Production of Machinery and Equipment in the P.R.C. (Washington, D.C., 1975), p. 10.

Although once-a-year deep-plowing with draft animals or large tractor is still possible.

7 The first transplanter was invented by a Hunan peasant youth in 1958; models approved for widespread manufacture in 1960 were experiment station products, however. These machines, which apparently are not very reliable, at best improve slightly on yields through more consistency in density, orderliness, and depth of transplanting-see Ting Ying (1961) op. cit., pp. 638-42. It is also relevant that they save female, not male labor. Transplanting has not yet been mechanized in Japan, probably because limited multiple cropping imposes no serious time constraint.

8 FBIS, June 2, 1977, p. E3.

ing communes and brigades for this practice, and forcing them to get the tractors back into the fields? 9

A more plausible explanation is that this is a consequence of the separate ownership of tractors at administrative levels higher than the production team, and the state-fixed price at which the plowing services are provided to the teams: Brigades are simply acting as income-maximizing enterprises, to the detriment of the incomes of "member" teams, by using their tractors in work the revenues of which are not limited by the state. Nor is it the only example of a resource allocation problem arising from the hierarchical (and undemocratic) structure of the commune, in which the state has now been forced to intervene in the interests of agricultural production, which is controlled by the team level.10

IMPROVED TECHNIQUES

Of all the agricultural practices recommended in the national agricultural development program, the only one that represented a break with trends in technique elsewhere in Asia was the promotion of "scientific close planting," which once was described as the "main theme" of the technical reform program." While many other aspects of improved technique have proven more important in the long run, the origins, evolution, and eventual fate of this innovation are particularly illuminating to our topic.

The emphasis on greater plant density as a yield-improving technique may derive from Soviet advice and, more specifically, the theories of Lysenko. The latter believed in the lack of intra-species competition, and the advantages of numbers when one species was in competition with another (e.g., crops against weeds). Thus he argued for dense planting. Although, in the Soviet Union, the method and its most famous application (to tree planting for shelter belts) were attacked as failures as early as 1954, these attacks were suppressed by political supports of Lysenko.12

Emphasis on increased plant density in China cannot be explained entirely by Lysenkian Soviet advice, however. Chinese experiments did establish that, in conjunction with deep plowing, increased density not only increased yields, but also significantly raised the marginal response to nitrogen applications (with existing seed varieties).13 The explanation was roughly that deeper plowing and increased fertilizer applications made possible greater vertical root development, thereby permitting closer planting without the decrease in yields that would otherwise result.

Unfortunately, as in the Soviet Union, this change in technique captured the imagination of political cadres, who appreciated the analogy between the "collectivist nature" (ch'ün-t'i-hsing) of plants

See FBIS, March 15, 1978, p. H3: Feb. 10, 1978, pp. E11-12.

10 See, for examples. FBIS, Feb. 17, 1978, p. E12: Feb, 16, 1978, p. H4.

11 Asakawa Kenji, "Four Reforms in Chinese Agriculture," Chugoku Kenkyu Geppo 154 (Jan. 30, 1961), trans. JPRS 9209, p. 43.

12 Loren Graham, Science and Philosophy in the Soviet Union (New York: A. Knopf, 1972), pp. 237-39. Lysenkian doctrine reached China in the early 1950's, where, as in the Soviet Union, it was used to attack theoretical research and promote practical research inspired by advanced peasant practices. See Fang Ts'uinung, "The Road of Service to Agricultural Production by Agricultural Scientists," IIsin chien-she 3 (1954), p. 18.

13 Ting Ying (1961), op. cit., pp. 369 and 372, presents experimental results demonstrating these points. Existing seed varieties presumably had low tillering rates.

and the process of communalization, the apparent refutation of the "bourgeois" law of diminishing returns, and the seeming requirements only of labor and additional seed to implement this reform. In the fever of the Great Leap, the reform was pushed on reluctant peasants with the same "excesses" that caused the debacle in local irrigation work or the promotion of the DWDS plow." For example, the traditional rice planting density in Kwangtung Province was less than 1.5 million seedlings/hectare and differed little from traditional standards elsewhere in Asia.15

But by 1958, this density had been increased by about 50 percent, and with the promotion of this and supporting reforms as the basis for the Great Leap in time for the fall planting, density reached 6-7.5 million seedlings/hectare-higher in some localities. For 1959, the Kwangtung CP Central Committee advocated a further increase to 12-15 million per hectare.16

A severe economic loss was the only clear result of these "excesses." According to one post mortem, an increase in seedling density from 2.3 to 7.5 million/ha did increase yields, but only by 180 kilograms. However, the increased yield was almost canceled by the increased seed requirement of 150 kilograms. Moreover, additional land had to be reserved for seedbeds (at the expense of both preceding and current crops), added labor was required for their preparation, transplanting and care, and heavier applications of fertilizer were required." (The reader may note the parallels and differences between this case and that of F-1 hybrid rice discussed above.)

With the experts again in command, promotion of "scientific dense planting" was subtly modified to a "reasonable range of density" (ho-li mi-chih) and deemphasized. Had the course of Chinese agrotechnical development followed those of other Asian countries, recommended density could then have been expected to decrease. For example, in Japan "there has been a gradual tendency for the density of planting rice. . . to decrease as the quantity of fertilizers used has increased and as rice varieties of a more prolific type have become common.18 That is, seed selection and cultivation emphasized the development of plants bearing a maximum number of tillers (grainbearing offshoots of the main stalk) per seedling. This trend of course saved both seedbed area and labor expended in transplanting. Thus, by the 1950's, standard density in Japanese rice cultivation was down to only 0.84 million seedlings/hectare.19 Improved row planting methods in the Philippines were said to reduce density to 0.99 million/ hectare; experiments with IR-8 seed in west Pakistan showed that a density of 0.87 million/hectare gave highest yields.20

14 The volume of propaganda “refuting" local opinions concerning plant density is a good indication of the extent of peasant resistance. See Kenneth Walker in Eckstein et al., eds., op. cit., p. 421; Ministry of Agriculture, Foodgrains Production Bureau, 1958-nien nung-tso-wu mi chih ching-yon (Peking, 1961).

15 Ibid., p. 5. Density in the Philippines was traditionally about 1.3 million seedlings/ha, according to unpublished IRRI materials.

18 Ibid., p. 5.

17 2.25 million seedlings resulted in 2.85 million effective cars, at 64.24 grains/ear: 7.5 million seedlings led to 5.85 million ears, but only 32.58 grains/ear and with a 10 percent decrease in weight/grain. 0.66 kilograms of seed were required to produce around 10,000 seedlings. Ma Chien-yü, "The Group Concept in Agricultural Production," trans. JPRS 9,398 (Mar. 23, 1961).

18 Takane Matsuo, Rice and Rice Cultivation in Japan (Tokyo: Ministry of Agriculture and Forestry, Government of Japan, 1959), p. 127.

19 Ibid., pp. 127 and 164.

20 Philippines datum from unpublished IRRI materials (1968): Pakistan from Agricultural Department, Government of West Pakistan, Annual Report on Accelerated Rice Research Program (1966), table 7, p. 31.