¹ An excellent Marxist interpretation of how developments in agriculture influenced social protest movements in cities and villages is advanced in Hupei University Political Economy Department of Teaching and Research (edit.), Chung-kuo chin-tai kfio-min ching-chi shih chiang-i [A Commentary on the History of Modern China's National Economy] (Peking: Higher Education Publication House, 1958), pp. 279–300 and pp. 334–354Google Scholar. Even Western scholars like R. H. Tawney have attempted to locate modern China's economic difficulties in agriculture. See Tawney, R. H., Land and Labor in China, (Boston: Beacon Press, 1966), chapters 2 and 3.Google Scholar

² These two schools of thought are capably discussed and examined in Li Li-chung, “Kuan-yu Chung-kuo nung-ts'un she-hui hsing-chih” [Concerning the Characteristics of Chinese Rural Society] Chung-kuo ching-chi 11:2 (December 1936) pp. 113–116.

³ This important question can only be answered when we have an adequate theoretical explanation of how traditional agricultural transformation actually takes place. Certain rural regions in China very definitely benefitted from trade expansion and the rising importance of cities after 1900–10, but we still do not know how the gains from trade were distributed between city and village and between various socioeconomic classes. When price trends are examined and terms of trade analysis made of prices at which peasants purchased and sold their crops, we will be informed to some extent of just how much benefits of trade peasants in various parts of China actually enjoyed.

⁴ Kantōchō, , Rōji jidai ni okeru Kantōshū [The liaotung Peninsula during the Period of Russian Control] (Dairen: Kantōchō, 1931), pp. 17–18Google Scholar. The area of the Kantōshū was roughly 1,337 square miles and was about 100 miles in length and 35 miles at its greatest width. Sec Fig. 1 for principal districts, cities, and the South Manchurian railway. Taiwan, which is 13,881 square miles in size, is roughly ten times the size of the Liaotung Peninsula. In 1935 the total area of farmland in Liaotung was 199,288 hectares or approximately 58% of the total land size of that region. In 1937 the total area of farm land used in Taiwan was 846,243 hectares or about 24% of the total land size of the island. The farmed area of Taiwan of that period was roughly four times the farmed area in Liaotung.

⁵ Kantōchō, , Kantōchō shisei nijūnen shi [A Twenty Year History of the Administration of Kwantung Prefecture] (Dairen: Kantōchō, 1926), p. 19.Google Scholar

⁶ Ibid., p. 22.

⁷ Ibid, p. 48.

⁸ Kantōkyoku, , Kantōkyoku shisei sanjūnen shi [A Thirty Year History of the Administration of the Kwantung Bureau] (Dairen, 1936), pp. 58–59.Google Scholar

⁹ Rōji jidai ni okeru Kantōshū, p. 74. While agriculture occupied a position of similar importance in the north China economy, the crops of wheat, millet, and sorghum were the important food grains. For a description of agriculture and farming conditions in this region, see Myers, Ramon H., The Chinese Peasant Economy, Cambridge: Harvard University Press, 1970, pp. 8–12Google Scholar, and chapters 4–8. For a description of crops and cultivation practices in north Manchuria which resembled the rural conditions in south Manchuria see chōsaka, Minami Manshū tetsudō kabushiki kaisha, Kita Manshū keizai chomacr;sa shiryō [Survey Materials on the North Manchurian Economy] (Dairen: Research Office of the South Manchurian Railway Company, 1911), pp. 1–10Google Scholar. Liaotung experienced the same seasonal temperature and annual rainfall variations as did Mukden and Hsinkyo. See the graph depicting five year average monthly temperatures and rainfall in Kantōkyoku kambō bunshoka (comp.), Kantōkyoku tōkei sanjūnen shi [A Thirty Year Statistical History of the Kwantung Bureau] (Dairen, 1937), graph 2 in appendix.

¹⁰ For a description of farming fertilizer practices, see Roji jidai ni okeru Kantōshū, pp. 74–75.

¹¹ A koku is a grain capacity unit of roughly 180 liters. Per capita food grain consumption in Liaotung as of 1902–03 was estimated to be 2 go of grain (.18 liters) per person per day. “The food-grain harvested was consumed entirely by the people living in Kwantung.” See Rōji jidai ni okeru Kantōshū, p. 75.

¹² This estimate was obtained from data for occupational structure of total population of Liaotung as reported in Kantōkyoku, Kantōcho tōkei ni jūen shi [A Twenty Year History of Statistics for Kwantung Prefecture] (Dairen: Kantōkyoku, 1926), pp. 42–43Google Scholar. This figure approximates that of Taiwan for 1930 when 68% of the workforce was employed in agriculture and forestry. See Barclay, George W., Colonial Development and Population in Taiwan, (Princeton: Princeton University Press, 1954), p. 58.Google Scholar

¹³ Kantōshū Furanten minseichō (Civil Administration Office of P'u Ian tien of Kwantung Province), Kannai Shinajin no nōka keizai [The Chinese Farm Economy within the Jurisdiction of P'u Ian tien] (Dairen: Kantōchō, 1932), p. 23.

¹⁴ Ibid., p. 24. A Japanese survey of Hua chia t'un village in this same district reported that two hundred years before, families with the surname of Hua had cleared the land, but by 1930 this large village of 78 households did not contain one household with the surname of Hua as members of this lineage group had died or migrated elsewhere. The village consisted of households with eight new surnames. This variation in lineage number per village indicates considerable lateral mobility of families and characterized rural social structure of this region.

¹⁵ Data obtained from Temporary Land Commission Survey of Kwantung Prefecture. Kantōchō rinji tochi chōsabu (comp.), Kantōshō jijō [General Conditions of Kwantung Prefecture] (Dairen: Kantōchō, 1923), II, pp. 651–652Google Scholar. This excellent two-volume study provides a vast storehouse of primary information and statistical materials concerning economic and social conditions in this region for the period 1916–1919.

¹⁶ See Myers, Ramon, “Land Distribution in Revolutionary China: 1890–1937” The Chung Chi Journal, 8:2 (May 1969), pp. 65–66Google Scholar. Land distribution in Taiwan of the early 1920's was even more unequally distributed, and the gini coefficient showing land concentration measured a high .76.

¹⁷ Kyohei, Suzuki, “Kantōshū ni okeru ichi nōson nōka shūnyū” [”Family farm earnings of a village in Kwantung Prefecture”] Mantetsu chōsa geppō 16:4 (April 1936) p. 178.Google Scholar

¹⁸ See Kantōshū jijō, II, pp. 653–654.

¹⁹ Kantōshū jijō, I, pp. 245–246. One estimate for the total rent collected from land held outside of Port Arthur administrative area was placed at a value of 32,650 yen. See I, p. 244. This represented a very large income flow for wealth holders living inside the area.

²⁰ Suzuki Kyohei, pp. 192–193; Kantōshū jijō, II, pp. 654–655. In this region, like north China, rent per unit of land was adjusted by landlords according to poor and abundant harvest. Most contracts were oral.

²¹ Kantōshū jijō, I, p. 694.

²² For a discussion of mortgaging and pledging of land for obtaining credit see chōsabu, Kantōchō rinji tochi, Kantōshū tochi chōsa jigy–ō hōkokusho [The Report of the Land Survey Enterprise in Kwantung Prefecture] (Dairen: Kantōchō, 1924), pp. 6–8.Google Scholar

²³ For land inheritance see Ibid., p. 11 and Kannai Shinajin no nōka keizai, pp. 36–37. This source comments as follows on the custom of fen-chia. “When the father dies and household division is to take place, one share of the land (called yang-lao-ti) or money (called yang-lao-fei) is used for supporting the aged widow and for her funeral, and the remainder is then divided between the sons.”

²⁴ For other examples pertaining to the Liaotung region observe the household budget comparisons of large, medium and small farms in P'i tzu huo district contained in Kantōshū jijō, II, pp. 1313–1361.

The differences in income and household wealth by farm size arc also great as revealed in a household budget study of fifteen farms in this region in 1923. sangyōbu, Mantetsu, Manjin nōka hfizai chōsa hōkoku [A Survey Report of the Manchurian Family Farm Economy] (Dairen: South Manchurian Railway Company, 1936)Google Scholar. See also Kannai Shinajin no nōka keizai, pp. 177–178.

²⁵ For studies of Manchuria that show these patterns, see the socioeconomic analysis applied from a Marxist point of view by Kyohei, Suzuki, “Chūbu Manshū ni okeru nōmin bunka” [”Differentiation of peasantry in central Manchuria”] Mantetsu chōsa geppō 15:10 (October 1935) pp. 25–58Google Scholar; same writer, “Ichi shiryo yori mitaru Manshu kakuchi no nōmin bunka” [”Differentiation of peasantry of various areas as seen from one source material”] Mantetsu chōsa geppō 15:9 (September 1935) pp. 41–94Google Scholar; Yavarin, V., “Manshū nōson ni okeru kaikyū kōsei ron” [”An essay concerning class composition in Manchurian villages”] Mantetsu chōsa geppō 15:12 (December 1935) pp. 179–220Google Scholar. For north China see Ramon H. Myers, The Chinese Peasant Economy, chapter 10.

²⁶ Crop and farm price statistics from which TABLES 2, 4, and 5 were constructed are based upon the following source materials. Data for years 1906–1925 were obtained from Kantōchō, Kantōchō tōkei nijūnen shi [A History of Twenty Years of Statistics for Kwantung Prefecture] (Dairen, 1926). Data for years 1925–1930 were obtained from Kantōchō, Shōwa gonen, Kantōchō dai nijūgo tōkeisho [The Twenty-Fifth Statistical Abstract for the Kwantung Prefecture, 1930] (Dairen, 1931). Data for years 1931–35 were obtained from Kantōkyoku, Shōwa jōnen, Kantōkyoku dai sanjū tōkeisho [The Thirtieth Statistical Abstract for the Kwantung Bureau, 1935] (Dairen, 1936). Data for years 1936–1942 were obtained from Kantōkyoku Shōwa jūnananen Kantōkyoku dai sanjūnana tōkeisho [The Thirty-seventh Statistical Abstract for the Kwantung Bureau, 1942] (Dairen, 1943).

²⁷ Kantōshū tochi chōsa jigyō hōkpkfsho, p. 46. “The land survey enterprise in Kwantung commenced in 1914, ended in 1924, and cost 2.3 million yen to complete. This survey made it possible to clarify land rights and made it more convenient for people to transfer land. This survey also made it possible to establish a land tax system, lay the foundations for management and ownership of land, and determine the basis for a general policy to implement and plan industrial enterprise development.”

²⁸ Ibid., p. 459.

²⁹ The method used to construct this index was as follows: For each crop we obtained the annual production by multiplying yield times cultivated area. The commodity unit price, obtained from averaging 1935–37 crop price, was then multiplied by annual production to convert to value. The unit of account was the yen, and crop statistics prior to 1924 were in Chinese measurement units whereas those after 1924 were in Japanese measurement units. Chinese measurement units were converted to Japanese units on the basis of conversion units used during the 1914–1924 land survey, and as already mentioned, cultivated area before 1924 was adjusted upward by 20%.

³⁰ The following estimates of cultivated area for each of the six main crops are presented as percentage of total cultivated area.

³¹ This can be seen below where current value for six crops are expressed as a percentage of total crop value individually and for the group.

³² To determine if price changes over time influenced the index, the average output for 1925–27 and 1930–32 were multipled by their respective value percentage discrepancies (plus or minus) current 1935–37 prices. The value distortions found from this exercise were exceedingly minute so that it did not make any difference which of the three sets of years were used as price weights to compute the output index. For this reason, 1935–37 commodity price averages were used to construct the aggregate agricultural output for the 1906–1941 period.

³³ See Myers, Ramon H. and Ching, Adrienne “Agricultural Development in Taiwan Under Japanese Colonial Rule,” Journal of Asian Studies 23:4 (August 1964) pp. 556–557.CrossRefGoogle Scholar

³⁴ Ohkawa, Kazushi, “Phases of Agricultural Development and Economic Growth,” in Ohkawa, Kazushi, Johnston, Bruce F., and Kaneda, Hiromitsu (ed.), Agriculture and Economic Growth: Japan's Experience, (Tokyo: University Press, 1969), p. 7.Google Scholar

³⁵ We define work animals as horses, mules, and donkeys. An index to represent their total over time was derived from prices obtained for years 1918–1921. District work animal prices for these years were obtained from Kantōshō jijō, I, II.

³⁶ Due to our inability to construct a reliable index to show trend in farm capital stock, we did not attempt to calculate the contribution of land, labor, and capital to increase total farm output. At the beginning of this study we contemplated using a production function approach of the Solo type to estimate the ‘unexplained’ portion of output increase which might be observed to have occurred and be unrelated to the growth of the conventional inputs of land, labor, and capital. We had to abandon this approach for two reasons: first, we could not obtain a reliable long term index to measure fixed and working capital over the period; second, given the paucity of farm management information for Liaotung we could not derive estimates of factor share contribution or input elasticities. While it might have been possible to use factor share estimates derived from farm management data collected from Manchurian farms in 1939–40, we did not adopt these coefficients for Liaotung as farming conditions probably differed by this time.

³⁷ The production of this surplus made it possible by the mid-1920's for the Liaotung peasantry to experience a marked improvement in their living standards. A Japanese study undertaken in 1930/31 commented that “In former times a farmer supplied his own materials when building a home by collecting mud from the river banks and using sorghum stalks to make bricks. Recently the farmer buys some bricks and cement as well as new ceramic tile for building his home, although he will continue to produce some of the native-type bricks by himself.” This same report noted that “Children on their way to school now buy school books, pens, and writing paper. Previously, the fertilizer farmers used consisted of dirt and compost, but now chemical fertilizers are sold, and peasants appear to be discussing their prices in everyday conversation.” See Kannai Shinajin no nōka keizai, p. 17. Further, this surplus undoubtedly played an important role in the rapid industrial development underway in this region after World War 1. A recent study advanced estimates of manufacturing output growth to show that between 1926–1936 a 7.2% per annum growth rate was achieved in the manufacturing sector. We have calculated this growth rate from the table in Kungtu C. Sun (Assisted by Ralph W. Huenemann), The Economic Development of Manchuria in the First Half of the Twentieth Century, (Cambridge: East Asia Research Center of Harvard University, 1969), p. 96.Google Scholar

³⁸ See Myers, Ramon H. and Ching, Adrienne, “Agricultural Development in Taiwan under Japanese Colonial Rule,” Journal of Asian Studies 23:4 (August 1964) pp. 562–569CrossRefGoogle Scholar; Ho, Samuel Pao-San, “Agricultural Transformation under Colonialism: The Case of Taiwan,” Journal of Economic History 28:3 (September 1968), pp. 322–326CrossRefGoogle Scholar; Ho, Yhi-Min, Agricultural Development of Taiwan, 1903–1960, (Nashville: Vanderbilt University Press, 1966), pp. 71–76.Google Scholar

³⁹ This theme has been developed in more rigorous fashion in Myers, Ramon H., “The Commercialization of Agriculture in Modern China,” in Wilmott, W. (ed.) Economic Organization in Modern China, (Stanford: Stanford University Press, 1972)Google Scholar. Increased specialization can operate in the following way to raise yield and output. If two family farms devoted two-thirds of their labor to food grains and one-third to a fiber crop, and increasing returns to labor existed for the industrial crop and constant returns to the food grain crop (perhaps a most unrealistic assumption as it is more likely that increasing returns to labor existed for food grains as well as fiber and other industrial crops) a farm devoting all its time and effort to producing food grains might satisfy the food requirements for both households while the other farm devoted two-thirds of its labor to producing the fiber crop at the cost of only a loss of one-third of the food grain producing farm's labor. The farm producing the fiber crop now pays the food crop producing farm more product than it had forgone with a surplus left over. As such exchange would take place through rural markets, yields of certain crops where specialization had occurred would begin to rise gradually.

⁴⁰ Our assertion, of course, requires further empirical research for confirmation. Such research might be conducted along the following lines. Some areas within Liaotung might have had high growth rates for virtually all crops, whereas some crops might have had high growth rates regardless of area differences. Therefore, it becomes necessary to separate what might be termed area from crop effects by examining crop and area growth rates. If certain areas did register rapid growth rates whereas certain crops did not show high growth rates for all areas, such a finding would suggest that specialization and exchange were taking place and therefore might be contributing greatly to the increase in output.

⁴¹ The basis for this assertion as applied to Liaotung is provided in later pages of this essay, but for evidence of the extensive crop specialization which had taken place see Manjin nōka keizai chōsa hōkoku: Showa kyū nendo-Kantōshūnai—rōdō no bu [Section Concerning Labor in Kwantung Prefecture, 1934] (Dairen, 1936), pp. 7–8. The information cited in this source indicates that the typical family farm used their land intensively for growing fruit, vegetables, and the peanut. The average family farm purchased 20% of its food from the market and sold about half of its crop. See this same source, pp. 10–13. If technological change had become so important to account for this 70% increase in food grain yield as presented in Table VI, we would also expect the return on farm investment to be rather high. From data cited in the above farm management study, we estimated the return on farm investment to be 10%. This was a rate of return which exceeded the average return earned by mainland farms and was equivalent to the rate of return farms in Taiwan obtained during the 1930's. These calculations are contained in unpublished materials compiled by Ramon H. Myers and can be obtained by interested readers upon request.

⁴² Kantōchō shisei nijūncn shi, p. 506.

⁴³ In 1925 the total area devoted to plant and tree experimentation in the Liaotung came to roughly 110 acres of research station land. Between 1906 and 1925 about 592,000 shrubs for fruit were distributed to farmers. These consisted in the main of pears and apples. Ibid., pp. 508–509. For evidence of changing research programs see Kantōchō, , Kantōchō yōran: Taishō jūyonnen (A Survey of the Kwantung Prefecture, 1925) (Dairen, 1925), pp. 495–505.Google Scholar

⁴⁴ Kantōkyoku bunshoka (comp.), Kantōkyoku shisei sanjūnen gyōsekf chōsa shiryō, [Survey Materials Regarding Achievements during the Past Thirty Years of the Kwantung Bureau's Administration] (Dairen, 1937), pp. 277–289.

⁴⁵ Kantōshū jijō, II pp. 273–275.

⁴⁶ Kantōkyoku shisei sanjūnen gyōseki chōsa shiryō, pp. 291–293.

⁴⁷ While it is impossible to distinguish between products produced in Manchuria and Liaotung which were exported from Dairen, the phenomenal increase in trade must have drawn products produced locally into its mainstream. For example the value of grain products shipped from Dairen rose ninefold between 1913 and 1928 whereas the price level merely doubled. The growth in vegetable and fruit export, which most certainly included products grown mainly in Liaotung, increased six and tenfold, respectively. See Kantōkyoku kambō bunshoka (comp.), Kantōkyoku tōkei sanjūnen shi [A Thirty Year History of Statistics of the Kwantung Bureau] (Dairen, 1937), pp. 414–415.Google Scholar

⁴⁸ The evidence showing vegetable crop yield differences in P'u Ian tien district (Furanten) for 1918/19 shows that area and yield devoted to different vegetables varied greatly in villages of this district. As these villages were located fairly close to one another, these large differences suggest that specialization was taking place. Kantōshū jijō, II, pp. 688–693.

⁴⁹ Between 1917/18 and 1927/28, considerable amount of road building was completed in Chinchow and P'u Ian tien prefectures. Kannai Shinajin no nōka keizai, p. 21.

⁵⁰ The difficulties of police household registration in northern Liaotung are discussed in detail for the first decade of Japanese administration (1960–1915) in Kantōkyoku shisei sanjūnen gyōseki chōsa shiryō, pp. 151–154.

⁵¹ Ibid., pp. 154–210 for a lengthy account of Japanese policies to purge the region around Pi tsu huo of banditry.

⁵² Kantōshū tochi chōsa jigyō hōkpkusho, pp. 2–3.

⁵³ For an account of price movements in Liaotung between 1906 and 1927, see chōsaka, Shomubu, Mans hū bukka chōsa: Shōwa sannen, jū-ichi gatsu, jū nichi (A Survey of Manchurian Prices: November 19, 1928), (Dairen, 1929), pp. 42–43Google Scholar; for data showing prices between 1930 and 1941, see Kantōkyoku, , Bukk” chingin chōsa nempō (An Annual Report of a Survey of Prices and Wages) (Dairen, 1942), p. 1.Google Scholar

⁵⁴ For population data of Dairen between 1903 and 1920, see Shinozaki Yoshiro, Talien [Dairen] (Dairen, 1921), p. 9 and pp. 50–51.

⁵⁵ Ibid., pp. 540–543.

⁵⁶ Ibid., pp. 788–789.

⁵⁷ For an excellent analytic study of these other regions and their Treaty Port growth, see Murphey, Rhoads, The Treaty Ports and China's Modernization: What Went Wrong? (Ann Arbor: University of Michigan Papers in Chinese Studies, No. 7, 1970).Google Scholar