Chinese Agricultural Extension and Uneven Economic Growth, 1903-1937: a Case Study of Shandong Province

Home , Chen Han, Endymion Wilkinson, Linqing, Li Guohao, Renshou County, Shandong

Chinese Agricultural Extension and Uneven Economic Growth, 1903-1937: A Case Study of Shandong Province

Thesis

Presented in Partial Fulfillment of the Requirements for the Degree Master of Arts in the Graduate School of The Ohio State University

By Spencer Dean Stewart Graduate Program in East Asian Studies The Ohio State University 2015

Thesis Committee: Karen M. Mancl, Advisor Philip C. Brown Christopher A. Reed

Abstract

In 1929, the Nationalist Party (Guomindang 國民黨) passed a set of regulations to establish a modern agricultural extension system in China. During this time, agricultural extension was viewed as a means to achieve national regeneration amid economic, political, and social uncertainty. The objectives of this thesis are to explore the historical development and economic impact of systematic agricultural extension in Republican

China (1912-1949).

State-run agricultural experiment stations and schools, first established in the late

Qing (1644-1911), laid a foundation for institutional agricultural extension services during the Republican era. The establishment of a centralized system of extension in

1929 represents continuity in state-led development and diffusion of agricultural innovations from imperial China. Taking the Shandong Academy of Agricultural

Sciences (SAAS) as a case study, this thesis argues that institutionalized agricultural research and extension during the Republican era laid a foundation for the People’s

Republic of China (1949-present). Agricultural research and educational institutions established in Shandong in the late Qing and Republican eras made possible the rapid expansion of institutions such as SAAS after 1949.

Agricultural research and extension efforts during the Republican era led to economic growth that was geographically uneven. The diffusion and cultivation of improved cotton varieties in Shandong through the extension system illustrates that ii

extension efforts were largely focused on regions and counties within close geographic proximity to extension centers. Cultivation of improved varieties was concentrated around adjacent counties to the extension centers, leading to economic growth that was spatially uneven.

iii

Acknowledgements

I would first like to acknowledge my advisor, Karen M. Mancl, and my committee members, Philip C. Brown and Christopher A. Reed. They each provided vital feedback on this thesis, and have guided me throughout my two years at The Ohio State

University. I gratefully acknowledge Linda M. Lobao for feedback on early drafts of

Chapters One and Four. Her courses on innovation diffusion and rural poverty were especially instrumental in shaping this research.

This study would have been impossible without the help of faculty at the National

Library of China, the National Agricultural Library of China, and the Shandong

Provincial Library in locating important sources. In particular, Yao Yuzhi at the National

Agricultural Library of China was helpful and patient in finding and photocopying countless documents. Additionally, Chen Yunqian of Nanjing University was kind enough to provide me with access to the digital databases at the Nanjing University

Library, for which I am grateful. I also want to thank my host at the Shandong Academy of Agricultural Sciences, Wang Yanqin, for welcoming me to the academy and showing me around.

I am also grateful for the funding I received to pursue this degree. These included a Foreign Language and Area Studies (FLAS) Fellowship (2013-2014), an International

Graduate Research Associateship through OSU’s Ohio Agricultural Research and

Development Center (OARDC, 2014-2015), and a graduate research appointment through OSU’s Office of International Affairs (OIA, Summer 2015).

Finally, I must acknowledge the love and support I have received from my family.

Special thanks to Teresa for her patience and humor, and to Dean, whose timely arrival has been a source of inspiration and motivation to complete this thesis.

Vita

2013…………………………………...…B.A. Asian Studies, Brigham Young University

2013-2014………………………………………FLAS Fellow, The Ohio State University

2014-2015……………………...International Graduate Research Associateship Program, Ohio Agricultural Research and Development Center, The Ohio State University

2014-2015…………………………………………………...Graduate Research Assistant, Assessing Student Preparation for Global Success at OSU

February 2015…….……………..…… Third Place Award for Presentation in Humanities Edward F. Hayes Graduate Research Forum

Fields of Study

Major Field: East Asian Studies

Table of Contents

Abstract ...... ii Acknowledgements ...... iv Vita...... vi Table of Contents ...... vii List of Tables ...... viii List of Figures ...... ix Introduction ...... 1 1.1 Agricultural Extension: Definition and Literature Review ...... 4 1.2 Historiography of the Republican Era Economy ...... 13 1.3 Overview, Methods, and Sources ...... 18 Agricultural Extension in Republican China ...... 21 2.1 Encouraging Agriculture in Late-Imperial China ...... 23 2.2 Institutionalizing Agricultural Extension, 1900-1937 ...... 28 2.3 Weaknesses of Agricultural Extension in Republican China ...... 36 2.4 Summary ...... 42 Agricultural Experimentation and Extension in Shandong, 1903-2013 ...... 43 3.1 Provincial Agricultural Experiment Stations, 1903-1937 ...... 45 3.2 Experimentation in Wartime Shandong, 1937-1949 ...... 52 3.3 Shandong Academy of Agricultural Sciences, 1946-2013 ...... 56 3.4 Summary ...... 66 Cotton, Extension, and Uneven Economic Growth ...... 67 4.1 Sources and Methods ...... 69 4.2 Historical Overview of Cotton Cultivation in China ...... 75 4.3 Diffusion of U.S. Cotton Varieties in Shandong ...... 87 4.4 Cotton Extension and Spatial Inequality ...... 95 3.5 Summary ...... 97 Conclusions ...... 99 Bibliography ...... 105 vii

List of Tables

Table 4.1 CCMA and NARB Series Coverage and Estimated Annual Output in Shandong ...... 71

Table 4.2 Annual Yields of Foreign and Native Cotton Varieties in Shandong ...... 86

Table 4.3 Output of U.S. Varieties of Cotton in Adjacent and Near-Adjacent Counties ...... 94

Table 4.4 Average Expenses and Income for Crops in Qidong, 1928-1932 ...... 97

viii

List of Figures

Figure 2.1 Republican Era Agricultural Extension System ...... 32

Figure 3.1 Linqing Cotton Experiment Station ...... 49

Figure 3.2 Shandong Regional Capital Markets in Republican China ...... 52

Figure 3.3 Japan’s North China Agricultural Experiment Station Configuration Map ...... 54

Figure 4.1 Generalized Soil Map of Shandong ...... 81

Figure 4.2 Average Yield of U.S. Varieties in Shandong by County ...... 89

Figure 4.3 Average Yield of U.S. Varieties in Shandong by County (% of total output) ...... 90

Chapter 1

Introduction

There is no short cut to national regeneration. However, one can recommend with assurance a cooperative agricultural extension system which is essentially an educational process of teaching, suggestion, and persuasion calculated to change the working and living patterns of Chinese farmers and to bring about a better economic condition which promises the people a better standard of living. -Hsin-Pao Yang, 19451 In 2013, the Shandong Academy of Agricultural Sciences (SAAS, Shandong sheng nongye kexueyuan 山东农业科学院), located in Jinan, celebrated its 110-year anniversary. This celebration marked an important transition from former historical accounts of the academy to recognize its origins dating back to the Qing dynasty (1644-

1911) with the establishment of the Shandong Agricultural Experiment Station

(Shandong nongshi shiyanchang 山東農事試驗場) in 1903.2 In the course of celebration,

1 Hsin-Pao Yang, “Promoting Cooperative Agricultural Extension Service in China,” in Farmers of the World, ed. Edmund deS. Brunner, Irwin T. Sanders, and Douglas Ensminger (New York: Columbia University Press, 1945), 59. 2 This transition contrasts with previous institutional histories, which marked the beginning of SAAS with the establishment of an agricultural experiment station by the Chinese Communist Party (CCP) in Junan 莒南 in 1946. In conjunction with the 110-year celebration in 2013, a website was set up as a platform for information regarding the pre-Communist origins of the academy. According to a number of articles posted on the site, the main reasons for changing the date to the 1903 experiment station was for historical accuracy and recognizing the important research conducted prior to 1949. This change also appears to be influenced by a larger fascination with longevity in the PRC, as many universities and academies are increasingly recognizing their origins in the late-Qing and early Republican eras. A number of articles posted on the website similarly made this point to justify the change. As one author emphasized, “even the Shandong Agricultural University, currently located in Tai’an 泰安, whose predecessor was established three years after the Shandong Agricultural Experiment Station, celebrated their one hundred-year anniversary in 2006” (emphasis added). See Tang Qiming 唐齐鸣, “Shandong sheng nongkeyuan de 1

this experiment station was praised for “commencing the introduction of modern agricultural science and technology into Shandong, indicating a stride forward from experiential agriculture to experimental agriculture, and writing an important page in the history of agricultural science and technology in Shandong.”3

The establishment of the Shandong Agricultural Experiment Station in Jinan was part of a larger trend in late-nineteenth and early-twentieth century China in which science and technology was viewed as the solution to the nation’s various economic, social, and political problems. The New Culture Movement’s (1910s-1920s) “Mr.

Science” (Sai xiansheng 賽先生), for example, was viewed as “a savior of the young

Chinese republic,” with the belief “that scientific attitudes and the scientific method would not only bring wealth and power, they would also transform the people and their stale, ineffectual culture.”4 In conjunction with such efforts, the poor and “backwards” villages of rural China became a target for a number of initiatives, most notably the Rural

Reconstruction Movement (Xiangcun jianshe 鄉村建設) which began in the 1920s.

Experimental counties such as Yan Yangchu’s 晏陽初 (1890-1990) Dingxian 定縣5 and

yanqing ji shengri” 山东省农科院的院庆及生日 (Shandong Academy of Agricultural Sciences Anniversary and Birthday), Shandong Academy of Agricultural Sciences Website, November 10, 2010, accessed July 16, 2015, http://www.saas.ac.cn/saas/?content-1292.html. For example, see the 1994 institutional history, Shandong sheng nongye kexueyuan 山东省农业科学院 (Shandong Academy of Agricultural Sciences), Shandong sheng nongye kexueyuan yuanshi, 1946-1992 山东省农业科学院院史 (Shandong Academy of Agricultural Sciences Academy History, 1946-1992) (Jinan: Jinan shi jianhua yinshuachang 济南市建华印刷厂, 1994). 3 Shandong sheng nongye kexueyuan, Kuayue shiji de tengfei: Shandong nongye kexueyuan chuangjian 110 zhounian (1903-2013) 跨越世纪的腾飞: 山东省农业科学院创建 110 周年 (Souring across the Century: Shandong Academy of Agricultural Sciences’ 110 Year Anniversary) (Jinan: Shandong sheng nongye kexueyuan, 2013), 5. 4 Grace Yen Shen, Unearthing the Nation: Modern Geology and Nationalism in Republican China (Chicago: University of Chicago Press, 2014), 5. 5 See Charles W. Hayford, To the People: James Yen and Village China (New York: Columbia University Press, 1990). For discussion of pig breeding and experimentation in conjunction with this experiment, see 2

Liang Shuming’s 梁漱溟 (1893-1988) Zouping 鄒平6 aimed at reviving the Chinese village in part through means of scientific and technological progression. Although less well known in the English-language scholarship, agricultural research institutions, such as the Jinan experiment station of 1903, were set up for the purpose of both conducting scientific research and promote findings to farmers – a process broadly known as agricultural extension. As exemplified in the statement above by agricultural educator

Hsin-Pao Yang, proponents in China viewed agricultural research and extension as a means of national regeneration during a time of economic, political, and social uncertainty.

This present study seeks to provide a more comprehensive picture of agricultural experimentation and extension in Republican China. The objectives are (1) to understand the historical institutional development of agricultural experimentation and extension during the first half of the twentieth century; (2) to explore how these institutional foundations impacted later development in the People’s Republic of China (PRC, 1949- present); and (3) to assess the economic impact of agricultural extension work during the

Republican era.

This research finds that first, institutionalized agricultural experimentation emerged in the first years of the twentieth century during the Qing dynasty, and expanded throughout the Republican era. A formalized system of extension was subsequently established under the direction of the Nationalist Party (Guomindang 國民黨) in 1929.

Sigrid Schmalzer, “Breeding a Better China: Pigs, Practices, and Place in a Chinese County, 1929-1937,” Geographical Review 92, no. 1 (2002): 1-22. 6 See Guy Alitto, The Last Confucian: Liang Shu-ming and the Chinese Dilemma of Modernity (Berkeley: University of California Press, 1979). 3

Although these institutional forms marked a clear transition from the past, the active role of the government in promoting agricultural experimentation and disseminating new technology was in many ways a continuation of late-imperial efforts by the state to

“encourage agriculture” (quannong 勸農). Second, the agricultural research and extension institutions established during the late-Qing and Republican eras laid an important institutional foundation for the People’s Republic of China, as exemplified in a case study of Shandong. The current agricultural research and extension system and its rapid expansion after 1949 was made possible by pre-Communist institution building.

Finally, by looking at the extension of improved cotton varieties in Shandong, this study finds that agricultural extension work was geographically constrained and spatially uneven, resulting in economic growth that favored regions within close proximity to experiment stations and extension centers. This thesis, therefore, provides insights into a larger body of scholarship on agricultural extension in general and the dearth of scholarship on the historical development of extension in Republican China. Moreover, it also has implications for understanding the Republican era economy by illustrating that state-led economic growth in Shandong was uneven and spatially differentiated.

1.1 Agricultural Extension: Definition and Literature Review

The term “university extension” first came into common use in England in the

1840s. However, the first practical steps to “extend” education beyond the academy were taken in the 1860s by faculty at the University of Cambridge, eventually leading to the

“extension movement” of 1880s. Although this movement had little to do with

agriculture, the idea of university extension was applied to agriculture in the United

States in the early twentieth century. As such, the term “agricultural extension” was popularized in the United States following the formation of the Cooperative Extension

Services in 1914 (Smith-Lever Act) in connection with the Land Grant College system.7

A number of terms are used for “extension” in different languages. For example, the Dutch word for extension (voorlichting) means to clear up or lighten the path, and the

British and Germans use the term “advisory work” (German: beratung).8 The Japanese have used two different terms for extension, one meaning guidance (nōgyō shidō, 農業指

導 ) and the other to popularize or diffuse (nōgyō kairyō fukyū, 農業改良普及).9 One of the earliest translations of the term “agricultural extension” into Chinese (the first, according to one scholar10) appeared in a Chinese-language article published by agronomist Zou Bingwen 鄒秉文 (1893-1985) in 1917 on the agricultural extension system in the United States.11 In this article, Zou translated “agricultural extension” as nongye tuiguang 農業推廣, a direct translation of the English term with tuiguang meaning to popularize, spread, or extend.

7 Anne W. van den Ban and Stuart Hawkins, Agricultural Extension (New York: Longman Scientific & Technical, 1988), 7-8. 8 See discussion in ibid., 8-9. 9 Yang Shimou 杨士谋, Nongye tuiguang jiaoyu gailun 农业推广教育概论 (An Introduction to Agricultural Extension Education) (Beijing: Beijing nongye daxue chubanshe 北京农业大学出版社, 1987), 3. 10 Li Ying 李瑛, Minguo shiqi daxue nongye tuiguang yanjiu 民国时期大学农业推广研究 (The study on the agricultural extension of universities in China) (Hefei: Hefei gongye daxue chubanshe 合肥工业大学出版社, 2012), 7. 11 Zou Bingwen 鄒秉文, “Meiguo nongye tuiguangbu” 美國農業推廣部 (Agricultural Extension Department in the United States), Dongfang zazhi 東方雜誌 14, no. 9 (1917): 160-163. 5

In addition to these varying linguistic understandings, the meaning of agricultural extension is obscured by a number of different institutional arrangements. Agricultural extension is organized in three main ways. First is a state-driven system where extension is conducted directly within a ministry of agriculture, operating at the national or provincial level. This type of organization is found in current extension systems in countries such as the People’s Republic of China, Japan, Holland, Italy, and Thailand.

Second is a system of extension under the direction of agricultural universities, most notably the land-grant university system in the United States. Finally, privatized systems of extension are becoming increasingly popular. In this model, extension work is conducted by private companies or non-government organizations. Countries which have adopted this system include France, England, New Zealand, and Denmark.12

Despite differing institutional arrangements, agricultural extension entails an institutional form dedicated to extension work, as evidenced in definitions of extension in early twentieth century China. Although the use of the term tuiguang (extension) in relation to agriculture dates back to as early as the Song dynasty (960-1279),13 the use of the term nongye tuiguang by Zou and his contemporaries did not simply mean to

“encourage agriculture” as was done in imperial China. Rather, Zou stated:

There are three components to revitalizing agriculture. First is agricultural education. Second is agricultural experimentation. Third is agricultural extension. Agricultural education is meant to train agricultural personnel. Agricultural experimentation is meant to discover new agricultural practices. Where

12 For more detailed discussion, see Hu Ruifa 胡瑞法 and Li Liqiu 李立秋, “Nongye jishu tuiguang de guoji bijiao” 农业技术推广的国际比较 (An International Comparison of Agricultural Technology Extension), Keji daobao 科技导报 (Science and Technology Review), no. 1 (2004): 26-29; and B.E. Swanson and J.B. Claar, “The History and Development of Agricultural Extension” in Agricultural Extension: A Reference Manual, Second Edition, ed. Burton E. Swanson (Rome: Food and Agriculture Organization of the United Nations, 1984), 6-12. 13 Yang Shimou, 25. 6

agricultural extension is to popularize the teachings at the agricultural schools and the discoveries at the agricultural stations to the average farmer to promote the improvement of agriculture.14

As illustrated by Zou, while the popularization of new agricultural practices is an important component, extension work was closely tied to the teachings and discoveries at modern agricultural institutions such as agricultural schools and experiment stations.

Therefore, the idea of agricultural extension entailed the existence of an institutional form dedicated to extension work. As Niels Röling, an eminent specialist in the study of agricultural extension and agricultural innovation, defines it, agricultural extension is “a professional communication intervention deployed by an institution to induce change in voluntary behaviours with a presumed public or collective utility” (emphasis added).15

Agricultural extension is more than the transfer of material technology. It also involves the communication or teaching of human-capital related skills such as management, decision making, leadership, and organization.16 By adopting an anthropogenic understanding of technology, these forms of knowledge and ideas can also be understood as types of technologies. As anthropologist Francesca Bray has emphasized, technology “is not merely material but also ideological.”17 Although agricultural extension in early twentieth century China largely focused in material technologies such as improved seed varieties, pesticides, and fertilizers, it also included non-material technologies such as managerial skills and farm finances. Institutionalized

14 Zou Bingwen, 160. 15 Niels Röling, Extension Science: Information Systems in Agricultural Development (Cambridge: Cambridge University Press, 1988), 49. 16 See, for example, Swanson and Claar, 1-2. 17 Francesca Bray, Technology and Society in Ming China (1368-1644) (American Historical Association, 2000), 65. 7

agricultural extension, therefore, entails the planned and organized transfer of agricultural technology (both material and non-material) as conducted by agricultural institutions such as extension offices, experiment stations and schools.

Agricultural Extension and Innovation Diffusion

The impact of agricultural extension in improving farm performance and human- capital has been contested. Despite a number of studies that confirm the positive impact on agricultural production,18 agricultural extension has been criticized for ignoring larger economic, social, cultural,19 gendered,20 and ecological contexts of farming, and for instead overemphasizing conventional science as the solution to agricultural problems.21

For example, based on their work on agricultural extension in Australia, sociologists

Frank Vanclay and Geoffrey Lawrence concluded that “traditional extension was not malintended but rather socially naive.”22 One of the most common criticisms is that

18 On the positive economic impact of agricultural extension, see Dean Birkhaeuser, Robert E. Evenson, and Gershon Feder, “The Economic Impact of Agricultural Extension: A Review,” Economic Development and Cultural Change 39, no. 3 (1991): 607-650; and Robert E. Evenson, “Economic Impacts of Agricultural Research and Extension,” in Handbook of Agricultural Economics, vol. 1A, Agricultural Production, ed. Bruce L. Gardner and Gordon C. Rausser (Amsterdam: Elsevier Science, 2001). 19For example, renowned educator Paulo Freire, based on his fieldwork in Chile during the early 1970s, criticized agricultural extension as a process of “cultural invasion,” where indigenous farming knowledge is suppressed in order to domesticate, rather than liberate, individual farmers. Moreover, dialogical work was viewed by many extensionists as “time-wasting,” while others believed it was impossible to explain technical information to “ignorant” farmers. See Paulo Freire, Education for Critical Consciousness (New York: Seabury Press, 1973), 85, 101, 104. 20An important growing body of literature has begun to look at gender inequalities in the agricultural extension process. Amy Trauger et al., for example, argued that U.S. agricultural extension work is shaped by traditional roles of men and women in farm work, leaving the needs of women farmers unmet: “the educational programming of extension is generally divided into areas that reflect, reify and reinforce the gendered division of labour on farms in the USA.” See Amy Trauger, Carolyn Sachs, Mary Barbercheck, Nacy Ellen Kiernan, Kathy Brasier and Audrey Schwartzberg, “The Object of Extension: Agricultural Education and Authentic Farmers in Pennsylvania,” Sociologia Ruralis 50 (Apr. 2010): 85. 21 See, for example, Frank Vanclay and Geoffrey Lawrence, The Environmental Imperative: Eco-social Concerns for Australian Agriculture (Rockhampton: Central Queensland University Press, 1995), 116-118. 22 Vanclay and Lawrence, 127. 8

extension services disproportionately benefit specific groups of farmers at the expense of others, leading to greater social stratification.23

This scholarship on agricultural extension is closely tied to a rich body of literature on the study of innovation diffusion, which provides insights into how and why the extension process often disproportionately serves certain groups of farmers and promotes spatially uneven economic growth. Over the years, research on the diffusion of innovation has evolved with the development of new methodological frameworks and models of diffusion. Early models and research were strongly influenced by the “classic model” of diffusion research, characterized as a linear, rational, and planned process.

Research focused primarily on agricultural development from the perspective of the farmer, emphasizing the study of adoption behavior and the individual characteristics of

“innovators” and “laggards.”24 Traces of this classic model are embodied in Everett

Rogers’ 1962 influential book, Diffusion of Innovations. Rogers’ diffusion model was and continues to be highly influential both inside and outside the field of sociology.25

These models, especially their underlying functionalist assumptions and pro- innovation bias, were heavily debated in the 1970s and 1980s, leading to the development of alternative frameworks. One noteworthy critique of Rogers’ model was geographer

23 For example, Jim Hightower’s 1973 study Hard Tomatoes, Hard Times condemned agricultural extension and the land-grant system in the United States as being “committed to the technological and managerial needs of the largest-scale producers and agribusiness corporations,” and omitting “those most in need of research assistance” (p. 25). According to Hightower, the pursuit of efficiency led to large-scale mechanization research and extension where big business interests were consulted in the research and development process rather than the average farmer, leading to the creation of technologies appropriate only for large-scale agribusiness operations. See Jim Hightower, Hard Tomatoes, Hard Times: Failure of the Land Grant College Complex (Cambridge, MA: Shenkman Publishing Company, 1973), 25. 24Frederick C. Fliedel and Peter Korsching, Diffusion Research in Rural Sociology (Middleton: Social Ecology Free press, 2001), 13-16. 25Ibid., 38; Everett M. Rogers, Diffusion of Innovations (New York: The Free Press, 1962). 9

Lawrence Brown’s 1981 book, Innovation Diffusion: A New Perspective. In this study,

Brown’s theoretical framework contrasted with Rogers’ assumption that adoption opportunities are relatively equal, instead emphasizing “that the opportunity to adopt is egregiously and in many cases purposely unequal.”26 In accordance with the literature on agricultural extension, Brown argued that diffusion programs, including extension, are frequently biased towards specific members and social groups in society, enabling them to adopt innovations first, thereby exacerbating economic disparities between social classes. Moreover, Brown also emphasized that opportunities to adopt were limited by infrastructural and geographic constraints, enabling those within closer spatial proximity to diffusion centers to adopt first.27

This innovation diffusion process, therefore, explains one way in which territories become stratified, as some regions are able to reap the benefits of adopting agricultural innovations while others are not. In this regard, agricultural extension serves as a window to understand issues of place-making, or how uneven development creates differentiated regions of prosperity and poverty.28 Contrary to early social and economic theories of poverty and social stratification that emphasize individual characteristics such as human capital and culture,29 Brown’s framework of innovation diffusion highlights the larger

26Lawrence A. Brown, Innovation Diffusion: A New Perspective (New York: Methuen & Co., 1981), 7. 27 Ibid., 232-235. 28 See, for example, the discussion of place-making in Linda Lobao, “Continuity and Change in Place Stratification: Spatial Inequality and Middle-Range Territorial Units,” Rural Sociology 69, no. 1 (2004): 1- 30. 29 The “culture of poverty” thesis has been particularly influential in academia and policy making in the United States. This idea originated with Oscar Lewis’ anthropogenic studies of the 1960s, which assumed that the poor were impoverished largely due to cultural values, attitudes, and practices held by individuals which perpetuated from generation to generation. Therefore, this idea implied that individuals must radically transform inherent cultural values to escape poverty. In more recent years, this idea has been critiqued for “victim blaming” without accounting for larger social and economic structural factors between socioeconomic groups. See Oscar Lewis, The Children of Sanchez: Autobiography of a Mexican Family 10

social, spatial, and political contexts of why stratification occurs. Rather than assume homogenous social and economic structures, more recent scholarship on social stratification emphasizes that power and opportunity are disproportionally distributed across socioeconomic classes and territories, leading to social and regional inequalities.30

Such findings are in conjunction with scholarship on agricultural extension, which shows that extension is often utilized by politicians as a tool of state making to achieve political and economic goals. This is especially apparent in countries where the extension system functions under the direct supervision of a department or ministry of agriculture.

Agricultural extension services were among the agricultural modernization projects criticized by James C. Scott when commenting that such projects often lead to decreased autonomy of the farmer and increased power and control by the state.31 Political scientist

Robert Bates, moreover, critiqued state-sponsored agricultural production schemes in

Africa in the 1980s, pointing out that public services, such as agricultural extension, “that

(New York: Random House, 1961) (Thanks to Christopher A. Reed for pointing me to this earlier work by Lewis). For a general critique of the culture of poverty thesis, see Mario L. Small, David J. Harding, and Michele Lamont, “Reconsidering Culture and Poverty,” The Annals of the American Academy of Political and Social Science (May 2010): 6-27. For further discussion on the individual characteristics of poverty, especially culture and human capital, see Cynthia M. Duncan, Worlds Apart: Poverty and Politics in Rural America – Second Edition (New Haven: Yale University Press, 2014), 233-263. 30 My discussion has been especially influenced by recent scholarship that is part of the spatial turn within sociology. For example, see the collection of essays found in Linda M. Lobao, Gregory Hooks, and Ann R. Tickamyer eds. The Sociology of Spatial Inequality (New York: SUNY Press, 2007). 31 Scott comments that “the unspoken logic behind most of the state projects of agricultural modernization was one of consolidating the power of central institutions and diminishing the autonomy of cultivators and their communities vis-a-vis those institutions. Every new material practice altered in some way the existing distribution of power, wealth, and status.” See James C. Scott, Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed (New Haven: Yale University Press, 1998), 286. James Ferguson has similarly criticized the “development” apparatus in his study of Lesotho, Africa, stating that “it is a machine for reinforcing and expanding the exercise of bureaucratic state power, which incidentally takes ‘poverty’ as its point of entry – launching an intervention that may have no effect on the poverty but does in fact have other concrete effects.” See James Ferguson, The Anti-Politics Machine: “Development,” Depoliticization, and Bureaucratic Power in Lesotho (Minneapolis: University of Minnesota Press, 1994), 255-256. 11

could have been offered to the small-scale farmer are instead put into the service of government schemes. Government-sponsored production units thus often promote the fortunes of a few privileged farmers at the expense of the small farmer in Africa.”32

Similarly, Röling pointed out,

In most countries extension is deployed, first of all, as an instrument to achieve policy goals, i.e., an instrument to make individuals behave in ways which are conducive to achieving the policy goals. Agricultural extension is a typical example. Most governments use agricultural extension as one of the instruments for developing the agricultural industry, not for helping individual farmers. Usually, export goals, national food security, cheap food supplies for urban workers and the efficient use of national resources are given priority above the welfare of individual farmers.33

As these previous studies emphasize, agricultural extension often serves as a tool to promote the needs of the state and expand bureaucratic power, rather than focusing on the needs of the farmers.

A review of this scholarship highlights two important themes addressed in this research. First, agricultural extension services are often geographically and infrastructurally constrained, leading to uneven adoption and economic growth based on access to the provided resources. Second, the state uses agricultural extension efforts to promote the needs of the national economy, even at the expense of the individual farmer.

Therefore, where one lives and the needs of the state are important factors that either allow or restrict adoption of agricultural innovations developed and diffused through the agricultural research and extension apparatus. Moreover, agricultural extension provides

32 Robert H. Bates, Markets and States in Tropical Africa: The Political Basis of Agricultural Policies (Berkeley: University of California Press, 2005), 49. 33 Röling, 38. 12

insights into how aspatial economic policy can lead to spatially differentiated economic growth and stratification.

1.2 Historiography of the Republican Era Economy

An extensive body of scholarship already exists on the Republican era economy, with varied findings and conclusions between scholars sparking important debates in the field.34 The early accepted view was one of overall degeneration, including falling rural incomes and exploitative tenancy relations, which laid the foundation for the eventual

Communist revolution and establishment of the People’s Republic of China in 1949. In particular, both English- and Chinese-language scholarship during the Republican era provided passionate and detailed accounts of chronic poverty and widespread famine.

Perhaps most notably, Walter Mallory’s 1928 book China: Land of Famine led China to become known nationally and internationally as “the land of famine.”35 Explanations for this poor “backward” economy varied greatly between scholars. Distributionist perspectives, such as that of sociologist Chen Hansheng 陳翰笙 (1897-2004), argued that unequal land and income distribution restrained economic development and impoverished the rural population.36 Other scholars, including Fei Xiaotong 費孝通

34The journal Republican China specifically published a series of articles on the Republican economy debate in 1992, which provides a thorough overview and critique of the scholarship. See Republican China 18, no. 1 (1992). 35 A former secretary of the China International Famine Relief Commission, Mallory’s passionate account of famine emphasized the “backwardness” and poverty found in rural China. He further attributed the lack of “modern ideas and practices” to the “backward-looking” nature of the Chinese people. See Walter H. Mallory, China: Land of Famine (New York: American Geographical Society, 1928), 101. For a brief discussion of the influence of this book, see Lillian M. Li, Fighting Famine in North China: State, Market, and Environmental Decline, 1690s-1990s (Stanford: Stanford University Press, 2007), 308-309. 36 Chen Hansheng (Chen Han-seng), The Present Agrarian Problem in China (Shanghai: China Institute of Pacific Relations, 1933). 13

(1910-2005), Richard Henry Tawney (1880-1962), and John Lossing Buck (1890-1975), offered more eclectic explanations for economic stagnation and decline. For example, Fei pointed to imperialism, declining rural industry, and “social erosion,”37 Tawney emphasized natural disadvantages and economic defects,38 and Buck stressed the inadequate access to credit and technological innovations.39

Beginning in the 1980s, however, previous assumptions regarding the Republican economy were called into question as scholarship proposed more complex and in some cases relatively positive views of economic growth. The works of economists Thomas

Rawski and Loren Brandt were especially important in questioning the accepted view of stagnation and decline, arguing instead that economic growth, commercialization in the farm sector, and real wages not only increased, but did so at a much faster rate than previously recognized.40 Brandt’s studies have further argued that the distribution of such

37 Fei’s “social erosion” entailed a process where the educated class became increasingly divided socially and economically from rural China, all within the context of imperialism and the treaty port system of the late-nineteenth and early-twentieth centuries. See Fei Xiaotong (Fei Hsiao-tung), China’s Gentry: Essays in Rural-Urban Relations (Chicago: University of Chicago Press, 1953). 38Tawney’s “natural or inherited disadvantages” included environmental degradation and susceptibility to natural disasters, and “defects of economic organization or in social habit” consisted of civil disorder, exploitation of peasantry, and problems with the land tenure system. These factors combined to create a rural environment where “the position of the rural population is that of a man standing permanently up to the neck in water so that even a ripple is sufficient to drown him.” See Richard H. Tawney, Land and Labour in China (London: George Allen & Unwin LTD, 1932), 77-8. 39Based largely on his extensive statistical surveys conducted in the 1920s and 1930s throughout various parts of China, Buck further concluded that collective farming and redistribution “would probably decrease production per capita,” calling instead for “development of transportation, communication, industry, professions, water control, and various technological practices.” John Lossing Buck, “Fact and Theory about China’s Land” Foreign Affairs 28, no. 1 (1949): 110. See also John Lossing Buck, Chinese Farm Economy (Nanjing: University of Nanking, 1930); and John Lossing Buck, Land Utilization in China (Nanjing: University of Nanking, 1937). 40 Thomas G. Rawski, Economic Growth in Prewar China (Berkeley: University of California Press, 1989); Loren Brandt, Commercialization and Agricultural Development: Central and Eastern China, 1870-1937 (New York: Cambridge University Press, 1989). 14

growth was relatively even, concluding that commercialization was not as “economically polarizing” as emphasized previously.41

These studies by Rawski and Brandt have been received with criticism, however, largely for their neoclassical theoretical assumptions that commercialization translated into economic development and increased rural welfare.42 Most notably, historian Philip

Huang’s studies of the peasant economies of North China and the Yangzi delta challenged Rawski and Brandt, arguing instead for “growth without development,” where population growth led to involutionary development, “in which the total output expands, but at the cost of diminished marginal returns per workday.”43 According to Huang, such intensification and devaluation of agricultural labor discouraged farmers from managing large-scale farms and investing in agriculture as would be expected by models of economic growth outlined by Adam Smith and Karl Marx.44

Despite contrasting views of economic performance, recent revisionist scholarship has continued to recognize several weaknesses in the Republican economy, two of which are particularly relevant to this study: lack of technological innovations and uneven economic growth. As mentioned above, failure to invest in technological

41 Brandt, Commercialization and Agricultural Development, 6; a later study by Brandt and economist Barbara Sands, looking at income per capita rather than by household, found that “income inequality was much lower in rural early twentieth-century China than has been previously inferred on the basis of data on land distribution alone.” See Loren Brandt and Barbara Sands, “Land Concentration and Income Distribution in Republican China,” in Chinese History in Economic Perspective, ed. Thomas G. Rawski and Lillian M. Li (Berkeley: University of California Press, 1992), 180. 42 For one example, see critiques found in Thomas B. Wiens, “Trends in the Late Qing and Republican Rural Economy: Reality or Illusion?,” Republican China 18, no. 1 (1992): 63-76. 43 Philip C.C. Huang, The Peasant Family and Rural Development in the Yangzi Delta, 1350-1988 (Stanford: Stanford University Press, 1990), 11; see also Philip C.C. Huang, The Peasant Economy and Social Change in North China (Stanford: Stanford University Press, 1985), 202-216. 44Ibid. See also Phillip C. C. Huang, “The Study of Rural China’s Economic History,” Republican China 18, no. 1 (1992): 164-176. 15

innovations is considered one reason the agricultural economy did not experience more rapid growth. Economist Ramon Myers’ study of the North China peasant economy concluded that the key rural problem was “the absence of any system generating rapid technological progress in agriculture.”45 Brandt further argued that the lack of technological innovations in the agricultural sector was not due to low demand, but rather insufficient supply. In contrast to other Asian countries during this time, he argued, the

Republican government failed to invest enough in public research and development projects, especially in relation to agricultural research, which was needed to advance important agricultural innovations such as fertilizer-responsive crop varieties.46

Uneven economic growth has been another criticism of the agrarian economy found in revisionist accounts: “the alternative picture being offered is of a spatially- differentiated and uneven growth process.”47 Here, economist Ernest Liang’s study of railways in relation to agricultural production is especially illustrative. With particular emphasis on the 1930s, Liang showed that areas within a given geographic proximity of railways tended to have more productive agricultural economies than areas farther away from rail lines.48 Economic historian Lillian M. Li’s more recent extensive study of

45Myers admitted that this view is similar to that of Buck mentioned above. See Ramon H. Myers, The Chinese Peasant Economy: Agricultural Development in Hopei and Shantung, 1890-1949 (Harvard University Press, 1970), 293. 46 Loren Brandt, “Reflections on China’s Late 19th and Early 20th-Century Economy,” in Reappraising Republican China, ed. Frederic Wakeman, Jr. and Richard Louis Edmonds (New York: Oxford University Press, 2000), 52-53. 47 Ibid., 29. 48 Ernest P. Liang, China: Railways and Agricultural Development, 1875-1935 (Chicago: University of Chicago, Department of Geography, Research Paper No. 203, 1982). 16

famine in North China similarly pointed out that economic growth was uneven based on proximity to railways and major urban centers such as Beijing and Tianjin.49

Of particular relevance is the research of historian Kenneth Pomeranz on

Shandong’s political economy. He argued that state-impact on the rural economy varied across regions, as local officials from the late-nineteenth century to the 1930s became increasingly concerned about defending key economic regions and protecting its autonomy, rather than with subsidizing poor areas, an approach found in traditional Ming and Qing statecraft.50 He stated that “government officials were less interested in raising rural income for its own sake than in ‘national strengthening’… promoting crops that could be used in either import-substituting or export-promoting industries,” especially cotton and mulberries (used in silk production).51 Moreover, “more money was spent on protecting newly strategic areas near the coast and near railroad crossings; once-favored poor areas were abandoned.”52 This new form of statecraft had drastic effects on flooding in specific regions of China, as “flooding was reduced sharply in the places that mattered most to China’s battles with foreign powers. But flooding in inland areas reached its worst levels on record.”53

This present study provides insights into previous scholarship on the agricultural economy of Republican China in order to further emphasize that state-led agricultural

49 Lillian Li, 316-319, 327. 50 Kenneth Pomeranz, The Making of a Hinterland: State, Society, and Economy in Inland North China, 1853-193. (Berkeley: University of California Press, 1993), 19, 274. 51 Ibid., 73. 52 Kenneth Pomeranz, “The Transformation of China’s Environment, 1500-2000,” in The Environment and World History, ed. Edmund Burke III and Kenneth Pomeranz (Berkeley: University of California Press, 2009), 132. 53 Ibid. 17

research and innovations in Shandong led to uneven and spatially differentiated economic growth. Although technological innovations may have been limited in comparison to other Asian nations as mentioned above, a closer look at agricultural research and extension work illustrates both the limits of technological innovations and the impact of the innovations that were introduced.

1.3 Overview, Methods, and Sources

Chapter Two of this thesis provides an overview of the development and institutionalization of agricultural extension in the Republican era from a long-term historical context of state-promoted agricultural experimentation and technology transfer that dates back to imperial China. Institutionalized agricultural extension under the direction of the Nationalist government, therefore, can be seen as a continuation of the role of the state to “encourage agriculture” that dates back to imperial times. Sources for this chapter come from secondary scholarship on late-imperial methods of agricultural development and technology transfer, along with textbooks and periodicals published during the Republican era. Textbooks used for university-level courses on agricultural extension were especially informative in understanding both how extension was conceptualized and how the system function, at least theoretically. Periodicals published throughout the 1920s and 1930s further provided accounts of how experimentation and extension functioned in practice. The periodical Nongye tuiguang 農業推廣 (Agricultural

Extension) was especially informative in publishing the latest information on extension work and the extension system throughout Republican China.

Chapters Three and Four take Shandong as a case study to explore the institutional development of agricultural experiment stations and their economic impact during the Republican era. Previous studies of agricultural research and extension in

Republican China have focused largely on foreign involvement and major universities located in cities such as Nanjing and Guangzhou.54 Therefore, Shandong was chosen for this study to present a more representative analysis of agricultural experimentation and extension works in regions of the Republic of China outside these major cities. Many of the findings and arguments discussed throughout this thesis were not unique to

Shandong. As such, regional variations aside, this case study provides insight into the overall trajectory of agricultural modernization and institutional change in China throughout the twentieth century.

The third chapter follows the historical development of Shandong’s provincial- level experiment stations from the late Qing to the present to highlight their lasting legacies in the People’s Republic of China. It broadly traces the history of the 1903

Shandong Agricultural Experiment Station as it developed and expanded throughout the twentieth century. In addition to the periodicals mentioned above, sources for this chapter consist of publications by the experiment stations and the Shandong Academy of

Agricultural Sciences, along with government reports from Shandong published monthly that included detailed reviews of the research and extension work conducted by the stations.

54 See Li Ying; and Randall E. Stross, The Stubborn Earth: American Agriculturalists on Chinese Soil, 1898-1937 (Berkeley: University of California Press, 1986). 19

Chapter Four looks at the diffusion of improved cotton varieties by the Shandong agricultural and cotton experiment stations to assess the economic impact of extension during the 1920s and 1930s. The diffusion of cotton varieties was spatially uneven and favored regions within close proximity to the stations. Agricultural extension work was geographically and infrastructially constrained, leading to uneven and spatially differentiated economic growth. Statistics for this study come from a dataset on cotton production published throughout the 1920s and 1930s that is introduced in greater detail within the chapter. Geographic information systems (GIS) technology is utilized to visually represent this data and illustrate geographic disparities. This dataset is supplemented with reports and surveys published by the cotton experiment stations in

Shandong. The final chapter (Chapter Five) summarizes the central conclusions of this study.

Chapter 2

Agricultural Extension in Republican China

In a textbook on agricultural extension published in 1936, the authors provided a brief historical overview of the early stages of agricultural extension leading up to institutionalization under the direction of the Nationalist government in Nanjing.

According to these authors, this modern system of extension as it developed in the late

1920s and early 1930s was connected to pre-modern and early-modern methods of agricultural research and technology transfer in imperial China. “Since Shennong55 神農 invented the plow (leisi 耒耜),” the authors remarked, “our country has attached great importance to agriculture and advocated the encouragement of agriculture [by state officials] (quanke 勸課) repeatedly throughout history.” Furthermore, “although this system was not as perfectly equipped as modern agricultural extension, it was nevertheless the forerunner that opened the way for agricultural extension.”56

The connection between early initiatives to “encourage agriculture” in imperial

China and modern institutionalized agricultural extension has been made by a number of

55 Shennong is a pseudo-mythological and legendary ruler of China who is believed to have lived roughly 4,500 years ago. 56 Zhang Zhiwen 章之汶 and Li Xingyu 李醒愚, Nongye tuiguang 農業推廣 (Agricultural Extension) (Shangwu yinshuguan 商務印書館, 1936), 21. 21

scholars.57 Francesca Bray, for example, pointed out that when reviewing pre-modern and early-modern sources, we “find descriptions of Chinese government agricultural policies that are in many respects reminiscent of present-day development strategies.” Further, “it is clear that in many respects Han or Sung policies for agricultural development were similar to those put forward by international development agencies and national government planning offices in Asia today. This is a fascinating area of socio-economic history which deserves further study.”58

Such observations call attention to the long-term historical continuities in agricultural experimentation and extension in China from the pre-modern to modern eras.59 This perspective shows that, although institutional forms of agricultural extension in the early twentieth century marked a transition from the past, clear continuity was exhibited in the types of research being conducted and the active role of the state in promoting agriculture. This chapter is outlined as follows. First, agricultural experimentation and extension work in the twentieth century is placed within a long-term historical framework, dating back to imperial China with particular focus on late-imperial

China (roughly framed as the Ming and Qing dynasties, ca. 1368-1911). Next, agricultural schools, experiment stations, and the extensions system as it developed

57 See, for example, Yang Shimou, 22-29; and Jørgen Delman, Agricultural Extension in Renshou County, China: A Case-Study of Bureaucratic Intervention for Agricultural Innovation and Change (Hamburg: Mitteilungen des Instituts fur Asienkunde, 1993), 88-91. 58 Francesca Bray, Science and Civilision in China, vol. 6, part 2, Agriculture (Cambridge: Cambridge University Press, 1984), 81. 59 The importance of understanding modern development in a pre-modern or early-modern context has been emphasized by a number of scholars, especially the recent edited volume The Environment and World History. In this volume, the authors emphasize the need to pay “particular attention to continuities between the early modern and modern periods,” and “the need to place modern developments in the context of deep histories of human interactions with particular environments.” It is from this perspective that this paper is written. See Edmund Burke III and Kenneth Pomeranz eds., The Environment and World History (Berkeley: University of California, 2009), 4. 22

during the late-Qing and Republican eras is reviewed, emphasizing the continuing role of the state in encouraging agriculture. Finally, it concludes by reviewing over some of the main criticisms of the extension system during the Republican era.

2.1 Encouraging Agriculture in Late-Imperial China

Systematic agricultural experimentation and dissemination of agricultural information in China dates back to as early as the Han dynasty (206 BCE - 220 CE).60 In imperial China, local officials played an important ideological role in promoting agricultural research and transferring agricultural innovations, as has been documented by a number of scholars. The emperors, according to historian Lillian M. Li, were “the symbolic patrons of agriculture,” as the “image of emperor as a leader in agriculture was just as ancient as the image of imperial river-tamer.”61 Similarly, Francesca Bray states:

Encouraging agriculture (quannong 勸農) was a central tenet of policy throughout the imperial era, a core component of a broader statecraft commitment to promoting a prosperous and powerful polity through jiaoyang 教養, educating and nourishing the people. As well as calculating and adjusting taxes, constructing irrigation canals, reclaiming waste land and distributing tools or seeds to migrants or destitute farmers, the documentation, compilation, and dissemination of technical agricultural knowledge was considered an essential technique of government, the duty of state officers at every level from the local magistrate’s Yamen to the imperial court.62

60 Delman, 88-91; Gwyn E. Jones and Chris Garforth, “The History, Development, and Future of Agricultural Extension,” in Improving Agricultural Extension, 3-4. 61 Lillian Li, 89. 62 Francesca Bray, “Chinese Literati and the Transmission of Technological Knowledge: The Case of Agriculture,” in Cultures of Knowledge: Technology in Chinese History, ed. Dagmar Shafer (Leiden: Brill, 2012), 301. 23

Further, historian Peter Perdue illustrated that statecraft during the Ming (1368-1644) and

Qing dynasties consisted of “a pervasive belief in the active role of the imperial state in shaping rural society,” especially in relation to agricultural production.63

Local officials played an active role in promoting agricultural improvement.

According to historian William T. Rowe, the main channels of technological transfer in early modern China involved government officials, mercantile movements, and written texts. Of these, local officials were an especially important source as they authored technological texts and were transferred from region to region, allowing knowledge to be spread more readily.64 In comparison to Europe during the same time, Rowe concluded that the imperial state assumed “a far greater role in the active dissemination of technical knowledge, as well as a far lesser role in assisting private individuals and groups.”65

According to Perdue, local officials often took the initiative to “promote overall advances in agricultural technology by urging peasants to transplant rice seedlings, fertilize them heavily, pay closer attention to water supplies, and work diligently at weeding and pest control.”66 To facilitate improvement, local officials consulted with experienced farmers regarding new innovations, and encouraged experimentation with the guarantee of relief.

Other officials conducted their own research on plots of government-owned land.67 In

Hunan specifically, “officials provided resources, in the form of tax breaks, new seeds,

63 Peter C. Perdue, Exhausting the Earth: State and Peasant in Hunan, 1500-1850 (Cambridge: Harvard University Press, 1987), 11. 64 William T. Rowe, “Political, Social and Economic Factors Affecting the Transmission of Technical Knowledge in Early Modern China,” in Cultures of Knowledge, 26. 65 Ibid., 44. 66 Perdue., 12. 67 Ibid., 14-15. 24

and information, to encourage experimentation. They set up experimental plots to allow farmers to try out new seeds without risk.”68

Timothy Brook’s study of the transfer of rice technology to Zhili (Hebei) from the late Ming to Qing further illustrates some of these late-imperial methods of technology transfer. Brook singled out two main methods, specifically through “agricultural instructors” – southerners brought north to diffuse cultivation practices – and “official instruction” – local officials transmitting knowledge through instruction and writing manuals.69 As Brook summarized, “the transfer of rice technology, in which the north was generally deficient even into the 19th century, took place through the efforts of local gentry and officials who either directly instructed the peasants or hired southerners who could teach by instruction or by example.”70 Local gentry and officials, therefore, were both directly and indirectly involved in the process of innovation diffusion.

Agricultural guidebooks or agricultural treatises (nongshu 農書) further helped circulate knowledge about innovative agricultural practices throughout the empire. As economist Ramon Myers pointed out in his study of north China, in the early Ming and

Qing “officials studied agricultural guidebooks and, after learning the proper methods, encouraged the peasantry to adopt and practice them.”71 Bray distinguished two different types of treatises in imperial China, specifically individually commissioned and state commissioned treatises. Based largely on innovative experimentation and interviews,

68 Ibid., 22. 69 Timothy Brook, "The Spread of Rice Cultivation and Rice Technology into the Hebei Region in the Ming and Qing," in Explorations in the History of Science and Technology in China, ed. Li Guohao (Shanghai: Classics Publishing House, 1982), 684. 70 Ibid., 690. 71 Myers, The Chinese Peasant Economy, 182. 25

some of the most well-known individually commissioned treatises include Jia Sixie’s 賈

思勰 (dates unknown) Qimin yaoshu 齊民要術 (Essential Techniques for the Welfare of the People, 535), Wang Zhen’s 王禎 (1271-1368) Nongshu 農書 (Agricultural Treatise,

1313), and Xu Guangqi’s 徐光啟 (1562-1633) Nongzheng quanshu 農政全書 (Complete

Treatise on Agricultural Administration, 1639). The purpose of these texts were for instructing both farmer and official, and typically included detailed illustrations and descriptions to facilitate imitation. The authors of these texts, especially those listed above, were typically officials interested or experienced in agriculture, and as such had a firm belief in the role of the state and administration in promoting new agricultural practices. As Bray commented on Wang Zhen’s Nongshu, “spurred by pity and a

Confucian concern for the people’s welfare, Wang Chen [sic] wrote a treatise that he hoped would help improve agricultural standards all over China. Informed official encouragement and instruction was the only means of bringing about such improvements,

Wang believed.” Wang’s treatise “was thus intended in the first instance to instruct local officials in best agricultural methods currently available, so that they could pass this information on to the peasants under their jurisdiction.”72 Similarly, the Nongzheng quanshu, according to Bray, was especially innovative in its emphasis on administrative control over agricultural development.73

State commissioned agricultural treatises consisted primarily of citations from other works, rarely containing original material. Being promoted by the state, these texts

72 Bray, Science and Civilision in China, 60. 73 For discussion of these individually commissioned agricultural treatises, see ibid., 55-70. 26

were distributed throughout the empire, having an “influence out of all proportion to their agronomic worth.”74 Such texts were closely tied to government policy to increase specific crops and industries, including cotton and sericulture. Further, they also had ceremonial aspects, aimed at glorifying the emperor. Lillian Li pointed out that “such imperially sponsored publications may have had more symbolic than technological value.”75 For example, the promotion of publications such as the Mianhuatu 棉花圖

(Illustrations of cotton production), focused on cotton production, “demonstrated a bureaucratic as well as imperial interest in promoting this cash crop.”76

It is debatable how effective local officials and agricultural treatises were in facilitating the transfer of technology across the empire. Economist Dwight Perkins argued that “the existence of a centralized government in China did contribute to the six- century [from the fourteenth to twentieth century] rise in grain yields… by preventing or alleviating some of the effects of flood and famine and by accelerating the diffusion of better seeds and improved techniques of farming.”77 Bray, however, argued that although individual officials played a role in improving conditions in the regions they were responsible over, these officials were first and foremost scholars: “their knowledge of farming often came from old books rather than personal experience, and they were prone

74 Ibid., 70. 75 Lillian Li, 90. 76 Ibid., 90. Bray further commented on the symbolic role of agricultural treatises, especially illustrations. She argued illustrations within the texts were both blueprints and icons. State-commissioned texts, such as the Gengzhi tu 耕織圖 (1153 or 1154), were more iconic in nature, emphasizing harmony and order in village life. Illustrations in other texts such as Wang Zhen’s Nongshu functioned more like blueprints for local officials to emulate. See Francesca Bray, “Agricultural Illustrations: Blueprint or Icon?,” in Graphics and Text in the Production of Technical Knowledge in China, in ed. Francesca Bray, Vera Dorofeeva- Lichtmann and George Metailie (Leiden: Brill, 2007), 521-567. 77 Dwight H. Perkins, Agricultural Development in China, 1368-1968 (Chicago: Aldine Publishing Company, 1969), 173. 27

to advocate archaic systems of dubious usefulness, or methods unsuited to local conditions.” Instead, “technical improvements in Ming agriculture owed more to the market and to migration.”78 Moreover, place was important in the transfer of technology, as access to transportation and markets was crucial.79 Myers further emphasized that in the nineteenth century, fewer local resources were devoted to agricultural development due to a rising rural population and the pre-occupation of local officials on taxation and other administrative functions.80 Despite these limitations, however, the perceived role of the state in promoting agricultural development and facilitating technology transfer continued into the twentieth century in the form of state-led agricultural experiment and extension institutions.

2.2 Institutionalizing Agricultural Extension, 1900-1937

Institutionalized agricultural extension began to emerge during the late nineteenth and early twentieth centuries with the establishment of agricultural schools and experiment stations. The first agricultural school was the Hangzhou Sericulture Hall

(Hangzhou canxue guan 杭州蠶學館), established in 1897.81 In the following years, schools were established throughout the Qing empire at the elementary, secondary, and

78 Bray, Technology and Society in Ming China, 29. 79 Ibid., 41. 80 Myers, The Chinese Peasant Economy, 182. 81 There is some debate regarding the first agricultural school to be established in China. Some hold that it was the Jiangxi Gao’an Sericulture School (Jiangxi Gao’an cansang xuetang 江西高安蚕桑学堂) established in 1896. Differences in opinions is due to the fact that there was a period of time between when this school was established and when students first enrollment. For a brief description of the issues surrounding the debate, see Bao Ping 包平, Ershi shiji zhongguo nongye jiaoyu bianqian yanjiu 二十世纪中国农业教育变迁研究 (Study of the Changes in China’s Agricultural Education in the Twentieth Century) (Beijing: Zhongguo sanxia chubanshe 中国三峡出版社, 2007), 12. 28

higher education levels. By 1909, 95 agricultural schools had been established with a total enrolment of 6,028 students.82 Student attendance in agricultural programs increased during the Republican era, helping fill the need for agricultural scientists in agricultural education and experiment work. As such, by 1927, 24 higher-level agricultural institutions and departments of agriculture were established, increasing to 39 by 1937.83

Chinese students studying agricultural-related subjects abroad also contributed to this growth. According to one scholar’s count, from 1901 to 1949, a total of 2,048 Chinese students studied agricultural-related topics abroad.84 The most popular destinations for agricultural studies were Japan and the United States. Japanese agricultural education appears to have been more influential than the United States in the total number of students (although only nominally more influential in the percentage of total students) studying agriculture-related subjects. A survey of students studying abroad during the year 1930, for example, showed that of the 3,064 students studying in Japan, 81 (2.6%) were studying agricultural science,85 while only 26 (2%) of the 1,280 students studying in the United States studied agriculture-related subjects.86

82 More specifically, there were 5 higher-education level agricultural schools with 530 students, 31 secondary-level agricultural schools with 3,226 students, and 59 elementary-level agricultural schools with 2,272 students. See Bao, 21-22. 83 Ibid., 31. 84 The statistics are given as follows: 1901-1911, 312 students; 1912-1927, 367 students; 1928-1937, 875 students; 1938-1945, 350 students; and 1946-1949, 144 students. See Shen Zhizhong 沈志忠, “Nongke liuxuesheng yu Zhongguo jindai nongye keji tizhihua jianshe” 农科留学生与中国近代农业科技体制化建设 (Returned Students Majoring in Agriculture and Institution Construction in the Field of Agricultural Science and Technology in Modern China), Anhui shixue 安徽史学 (Anhui History) 5 (2009): 6. 85 “Zhongguo liuri xuesheng xueke tongji” 中國留日學生學科統計 (Statistics of Chinese Students’ Disciplines Studying in Japan), Zhonghua jiaoyu jie 中華教育界 (China Education World) 18, no. 10 (1930): 4. 86 The statistics provided for study abroad students in the U.S. were more complex than for Japan. Where statistics for Japan only had one category for agriculture – agricultural science (nongke 農科) – the statistics for the U.S. were separated into subfields. The above statistic includes students studying 29

The earliest experiment station was established in Hunan in 1901. Three more experiment stations were established in the two years following, and by 1909, forty had been established nationwide, including a national-level station in Beijing.87 Despite political instabilities and warlordism through much of the Republican era, national and local governments continued to ambitiously establish agricultural experiment institutions.

In 1916, one source counted a total of 113 stations nationwide.88 Between the years of

1927 to 1931 alone, 322 provincial and county stations were established.89 A survey of seven of the 28 provinces in the Republic of China90 published in 1934 and 1935 listed 3 national, 20 provincial, and 271 county experiment stations.91 By 1936, a national survey showed 489 agricultural experiment stations.92

agriculture (nongxue 農學), botany (zhiwu xue 植物學), carpology (guoshi xue 果實學), horticulture (yuanyi 園藝), entomology (kunchong xue 昆蟲學), and forestry (senlin xue 森林學). See “Liumei xuesheng xueke tongji” 留美學生學科統計 (Statistics of Students’ Disciplines Studying in the United States), Jiaoyu zazhi 教育雜誌 (Education Magazine) 22, no. 3 (1930): 1. 87 See Xia Rubing 夏如冰, “Qingmo de nongzheng jigou yu nongye zhengce” 清末的农政机构与农业政策 (Agricultural Administrative Structures and Policies of the late Qing Dynasty), Nanjing nongye daxue xuebao 南京农业大学学报 (Journal of Nanjing Agricultural University) 3 (2002): 47. 88 “Zhongguo nongshi shiyanchang zhi tongji” 中國農事試驗場之統計 (Statistics on China’s Agricultural Experiment Stations), Kexue 科學 (Science) 2, no. 6 (1916): 711. 89 Zhang and Li, 59. 90 Provinces surveyed were Hebei, Shandong, Shanxi, Henan, Shaanxi, Chahar, and Suiyuan. 91 See Li Zhiji 李治楫, “Guonei nongshi shiyan jiguan gaikuang” 國內農事試驗機關概況 (Overview of National Agricultural Experiment Institutes), Nongye tuiguang 農業推廣 (Agricultural Extension), no. 6 (1934): 5-9; Li Zhiji, “Guonei nongshi shiyan jiguan gaikuang (er)” 國內農事試驗機關概況 (二) (Overview of National Agricultural Experiment Institutes, part 2), Nongye tuiguang, no. 7 (1934): 1-12; Li Zhiji, “Guonei nongshi shiyan jiguan gaikuang (san)” 國內農事試驗機關概況 (三) (Overview of National Agricultural Experiment Institutes, part 3), Nongye tuiguang, no. 8 (1935): 1-18; Li Zhiji, “Guonei nongshi shiyan jiguan gaikuang (si)” 國內農事試驗機關概況 (四) (Overview of National Agricultural Experiment Institutes, part 4), Nongye tuiguang, no. 9-10 (1935): 1-7. 92 “Shiye bu guanyu quanguo nongshi shiyanchang diaocha de gexiang tongji” 实业部关于全国农事试验场调查的各项统计 (Department of Industry Regarding the Various Statistical Surveys on National Agricultural Experiment Stations) (1936), in Zhonghua minguoshi dang’an ziliao huibian 中華民國史檔案資料彙編 (Archive Compilation on Republic of China History), 5 ji 楫, 1 bian 編, Caizheng jingji 財政經濟 (Finance and Economy), no. 7, ed. Zhongguo di’er lishi dang’anguan 中國第二歷史檔案館 (Second Historical Archives of China) (Jiangsu: Fenghuang chubanshe 鳳凰出版社, 1994), 392-401. 30

The establishment of agricultural experiment stations and agricultural schools paved the way for a formal centralized system of extension organized shortly after the

Nationalists established governance in Nanjing in 1928. In April 1929, during the Third

Nationalist Party Congress (Guomindang quanguo daibiao dahui 國民黨全國代表大會), the Ministry of Education (Jiaoyubu 教育部) published the Zhonghua Minguo jiaoyu zongzhi jiqi shishi fangzhen 中華民國教育宗旨及其實施方針 (Republic of China education objectives and implementation guidelines). The final article stated that agricultural education institutions were to actively engage in extension work to “improve agricultural production methods, increase farmers technical skills, ameliorate village organization and farmers lives, popularize agricultural scientific knowledge, and promote cooperation for farmer production and consumption.”93

Two months later, on June 14, 1929, the Ministry of Education, the Ministry of

Internal Affairs (Neizhengbu 內政部), and the Ministry of Agriculture and Mines

(Nongkuangbu 農礦部)94 passed the Nongye tuiguang guicheng 農業推廣規程

(Agricultural Extension Regulations), solidifying the work of agricultural extension under the direction of the central and provincial government. These regulations were further modified on March 29, 1933, clarifying some aspects from the original and including

93 Jiaoyubu 教育部 (Ministry of Education), Zhonghua Minguo jiaoyu zongzhi jiqi shishi fangzhen 中華民國教育宗旨及其實施方針 (Republic of China education objectives and implementation guidelines) (April 26, 1929), in Jiangsu sheng zhengfu gongbao 江蘇省政府公報 (Jiangsu Provincial Government Report), no. 139 (1929), tebie yaojian 特別要件 (Important documents): 1-2. 94 In 1930, the Ministry of Agriculture and Mines was consolidated with the Ministry of Industry and Commerce (Gongshangbu 工商部) to form the Ministry of Industry (Shiyebu 實業部). 31

several additional articles, although the changes were relatively minor.95 The basic organization of agricultural extension as set up by these regulations is found in Figure

2.1.96

Figure 2.1 Republican Era Agricultural Extension System

Source: Adapted from Nongye tuiguang guicheng 1929; Nongye tuiguang guicheng 1933; and Zhang and Li, 86. a Changed to the Ministry of Industry in 1930.

95 Jiaoyubu, Neizengbu 內政部 (Ministry of Internal Affairs), and Nongkuangbu 農礦部 (Ministry of Agriculture and Mines), Nongye tuiguang guicheng 農業推廣規程 (Agricultural Extension Regulations) (June 14, 1929), in Nongye tuiguang, no. 1 (1930): 2-7; and Jiaoyubu, Shiyebu 實業部 (Ministry of Industry), and Neizhengbu, Nongye tuiguang guicheng (March 29, 1933), in Nongye tuiguang, no. 4 (1933): 1-8. 96 For a comprehensive introduction to legislation regarding agricultural during the Republican era, see Zhou Xiaoyan 周晓焱 and Zhang Jianhua 张建华, “Nanjing guomin zhengfu de nongye tuiguang lifa yanjiu” 南京国民政府的农业推广立法研究 (A Study on Agricultural Extension Legislation of Nanjing National Government), Xibei nonglin keji daxue xuebao 西北农林科技大学学报 (Journal of Northwest A&F University) 12, no. 4 (2012): 139-145; for a brief overview of these events leading up to the publication of the Agriculture Extension Regulations, see Tang Zhicai 唐志才, Nongye tuiguang 農業推廣 (Agricultural Extension) (Shanghai: Zhengzhong shuju 正中書局, 1944), 8-9 32

These regulations set up extension work under the direction of the Ministry of

Agriculture and Mines (later Ministry of Industry), Ministry of Education, and Ministry of Internal Affairs. As illustrated in Figure 2.1, these ministries were to establish a

Provincial Agricultural Extension Committee (shengli nongye tuiguang weiyuanhui 省立

農業推廣委員會) or Agricultural Extension Office (nongye tuiguang chu 農業推廣處).

According to article 4 of the regulations, this committee was to be made up in part by agricultural specialists including professors from provincial agricultural universities or schools, and senior technicians from provincial agricultural experiment research institutes

(provincial agricultural experiment stations). These experts were to “provide the latest agricultural knowledge and high-quality materials, help in drawing up extension plans, and visit the countryside for work when necessary.”97 Moreover, with these regulations, the agricultural schools and experiment stations were to conduct agricultural research and extension work under the direction of the central or provincial extension organization, and work directly with provincial, county, and village extension centers.98 Article 23 outlined the role of the schools and research institutions:

The general affairs of agricultural extension are as follows: (I) Implement the achievements of the agricultural and forestry experiment stations and agricultural schools, whose main tasks are to: (1) supply high-quality seeds, saplings, and livestock species; (2) popularize high-quality agriculture and forestry management methods; (3) popularize high-quality materials and methods for farmers’ secondary industry; (4) popularize high-quality farm tools and fertilizers; (5) popularize methods for controlling pests and plant diseases; (6) implement any other achievements.99

97 Nongye tuiguang guicheng 1933, no. 4. 98 Nongye tuiguang guicheng 1929, no. 13; Nongye tuiguang guicheng 1933, no. 17. 99 Nongye tuiguang guicheng 1933, no. 23. 33

Both the agricultural universities and experiment stations engaged in this type of research and extension. However, agricultural universities and schools had the primary task of training students to later work in provincial and county extension centers and experiment stations. The agricultural experiment stations primarily worked to improve various crop varieties, distribute seeds, and publish extension leaflets, typically called qianshuo 淺說, to be used in agricultural schools and by farmers. In large part, the work done at the experiment stations was a continuation of former research and technological transfer from imperial China, although it became increasingly institutionalized and centralized in the twentieth century.100 Writing in 1930, one author summed up the relationship between agricultural extension and experimentation, stating “while we emphasize experiment research, we must simultaneously advocate extension in order to ensure that the first goal of experiment research is reaching the village, benefiting the farmer, and thoroughly improving agriculture.”101

Agricultural extension involved more than simply developing and promoting physical agricultural innovations. According to these regulations, the purpose of the agricultural extension system was to “popularize agriculture scientific knowledge,

100 A similar argument has been made for the case of Japan. Edward Pratt, for example, showed that, between 1880 and 1910, a shift took place as the state established agricultural experiment stations, agricultural societies, and other research institutions, which displaced the role of rural elites in promoting agricultural innovations. See Edward E. Pratt, Japan’s Protoindustrial Elite: The Economic Foundations of the Gono (Cambridge: Harvard University Press, 1999), 6, 180; Economist Penelope Francks, while looking at agricultural research and extension in the Saga Plain during the same time period more directly stated that “the nature of these projects reveals the experiment station at this stage simply centralizing and systematizing the kind of applied research which had been going on before.” See Penelope Francks, Technology and Agricultural Development in Pre-War Japan (New Haven: Yale University Press, 1984), 155. 101 Yu Gong 愚公, “Nongye tuiguang gailun” 農業推廣概論 (Overview of Agricultural Extension), Nongye tuiguang, no. 1 (1930): 8. 34

increase farmers’ technical ability, advance agricultural production methods, improve village organization and farmer lives, and promote farmer collaboration.”102 Writing several years later in a textbook used at university-level courses on agricultural extension, the authors wrote:

Agricultural extension is not the promotion 推廣 of agriculture, rather it is the promotion of agricultural knowledge and skills. It is not to make everyone in the country engage in agriculture, rather it is to instill agricultural knowledge and skills in the farmers of today and of the future. Assisted with physical materials, [agricultural extension] makes the farmers trust and follow new agricultural managerial methods, thereby increasing production and also improve the quality of life.103

In a different textbook on agricultural extension published in the 1930s, extension was defined as “a type of education... that takes society as its scope, the people as its target, the farms and farming families as its center, the practical needs of farmers as its provisions, and the improvement of the farmers’ entire life as its ultimate goal.”104 In addition, the authors emphasized that extension was much more than instruction for farmers, but also included home management instruction for women and youth groups for children.105 The meaning of agricultural extension, therefore, was much more than the promotion of new types of seeds and agricultural training. Rather, it aimed to improve all aspects of rural life through proper education.106

102 Nongye tuiguang guicheng 1929, no. 1; Nongye tuiguang guicheng 1933, no. 1. 103 Guan Yida 管義達, Lu Feizhi 陸費執 and Xu Zhen 許振, Nongye tuiguang 農業推廣 (Agricultural Extension) (Shanghai: Zhonghua shuju 中華書局, 1935), 5. 104 Zhang and Li, 15. 105 Ibid., 15. 106 The historical development of agricultural schools, experiment stations, and extension in late Qing and Republican China parallels to some degree earlier developments in the United States. In the United States, the Morrill Act of 1862 led to the establishment of the land-grant colleges focused on agricultural education. In 1887, the Hatch Act led to the creation of agricultural experiment stations under the direction of the land-grant colleges. Finally, the Smith-Lever Act of 1914 established the cooperative extension services, again connected to the land-grant colleges. The main difference between this institutional 35

2.3 Weaknesses of Agricultural Extension in Republican China

Previous reports and scholarship have pointed out a number of weaknesses of agricultural extension during the Republican era, including a narrow scope of extension work, lack of adequate personnel, and an extension system that was both unorganized and closely tied to politics. First, according to a comprehensive report on China’s agricultural situation published by the China-United States Agricultural Mission in 1947, extension work centered primarily on the transmission of physical technologies, stating, “extension work in the minds of too many people in China simply means the physical act of multiplying and distributing seeds, fertilizers, breeding stock, farm machinery, spray materials, and the like. This is not the primary object of extension work.”107 Although

arrangement and what took place in Republican China was that in the latter, education, experimentation, and extension was not all conducted under the umbrella of an agricultural college or a land-grant college system. Although agricultural schools in Republican China often had their own experiment stations or fields to conduct research, the central-, provincial-, and county-level governments also operated experiment stations. Moreover, these schools and state-led experiment stations functioned as supporting institutions to the extension system as established in 1929, rather than being more integrated as found in the United States’ land-grant system. In this regard, the agricultural education, research, and extension apparatus in Republican China was more similar to the system set up in Japan in the late-nineteenth century. Similar to Republican China, local government in Japan was heavily involved in experiment research and extension, and the state-led agricultural experiment stations, schools, and extension system all functioned independent of each other. The various agricultural institutions as they developed in the late-Qing and Republican eras appear to be strongly influenced by Japan, especially following the First Sino-Japanese War (1894-1895). As historian Benjamin Elman pointed out, Chinese scientists increasingly turned to Japan for scientific models following the war. “By 1905, the new Qing ministry of Education was staunchly in favor of science education and textbooks based on the Japanese scientific system. Instead of the West… Japan now mediated the West for Chinese literati and officials.” The transfer of agricultural science and institutional models from Japan was further facilitated through translated text, students studying abroad in Japan, and Japanese scientists serving as advisors in China. See Benjamin A. Elman, “Toward a History of Modern Science in Republican China,” in Science and Technology in Modern China, 1880s-1940s, ed. Jing Tsu and Benjamin A. Elman (Leiden: Brill, 2014), 27. For a detailed overview of the Japanese agricultural research and extension system, see Takekazu Ogura ed., Agricultural Development in Modern Japan (Tokyo: Fuji Publishing Company, 1963); and Francks, 150-160. For U.S. and Japanese influence on agricultural institutions in China, see Stross, 40-46. For a historical overview of the U.S. land-grant college system, see Frederick Rudolph, The American College and University: A History (Athens: University of Georgia Press, 1990), 241-263. 107 China-United States Agricultural Mission, Report of the China-United States Agricultural Mission (Office of Foreign Agricultural Relations, 1947), 78. 36

agricultural extension was frequently defined broadly to include skills such as home and farm management, in practice extension consisted primarily of diffusing physical technologies such as improved seeds, fertilizers, and machinery.

Next, adequately trained personnel engaging in extension work were in short supply. Although agricultural schools were supposed to train students to work in provincial and county extension centers and experiment stations, most graduates, especially those at the university level, seldom sought out such positions, instead looking for jobs in education, government, and business. For example, the China-United States

Agricultural Mission report recorded that many workers in the county extension offices were political appointees with little to no experience in agricultural extension work or farming in general.108 Zhang Fuliang 張福良 (Chang Fu Liang, 1889-1984), former rural secretary of the National Christian Council, argued that the cause of this situation was that higher-level agricultural education attracted mainly urban students, a consequence of charging tuition fees and the nature of the admittance process which placed greater importance on formal education than on actual farm experiences. In Zhang’s words, agricultural education attempted to make farmers out of city boys, and “agricultural scientists out of the sons of Mandarins and Compradores.”109 After graduation, the majority of agricultural-college students did not engage in agricultural work, with most becoming urban-based politicians, bureaucrats and teachers. Not only was this caused by better economic and career opportunities outside the village, graduates also tended to

108 Ibid., 77. 109 Zhang Fuliang (Chang Fu Liang), “Agricultural education and country life,” Educational Review 22, no. 2 (1930), 190. 37

have a Confucianist view of manual labor, which was “beneath his dignity to participate in,” an idea that “the instruction in most agricultural institutions seems to accentuate.”110

Further, even when graduates sought to work in the countryside, they were often alienated culturally and socially. For example, in her study of silk production in Wuxi,

Lynda Bell pointed out that extension workers and farmers viewed one another through the lens of “the other.” That is, farmers saw extension workers as being “foreign” while extension workers viewed farmers as being “backwards” and confined to “tradition.”

Although extension workers perceived themselves as the bearers of scientific truth and modern technique, farmers tended to be largely skeptical.111 Randall Stross similarly argued that agriculturalists were detached from the social conditions of the countryside, and believed in a “technological fix” to the complex problems facing China’s agriculture.

In doing so, they failed to address or even recognize the underlying political, economic, and social changes needed to actually improve rural conditions.112

These findings coincide with what Wen-hsin Yeh referred to as the “alienated academy” of the Republican era. Yeh argued that the beneficiaries of higher education were largely urbanites, focusing on subjects of study that distanced them from the countryside linguistically, culturally, and socially. To the townspeople and villagers, the college-educated elite were both foreign and remote.113 Even agricultural schools at the

110 Ibid., 189. 111 Lynda S. Bell, One Industry, Two Chinas: Silk Filatures and Peasant-Family Production in Wuxi County, 1865-1937 (Stanford: Stanford University Press, 1999), 141-145. 112 Stross, 215. 113 Wen-hsin Yeh, The Alienated Academy: Culture and politics in Republican China, 1919-1937 (Cambridge, MA: Harvard University Press, 1990), 7-48. Yeh does point out a few exceptions to this generalization. In the 1920s, for example, Zhongshan University specifically sought to “break down the barrier between school and society.” (p. 176). 38

village level appear to have suffered from cultural and social alienation. In Stig

Thøgersen’s study of village schools in 20th century Zouping, Shandong, he argues that low enrollment numbers and an utter lack of agricultural schools in local historical memory suggest the limited value that locals placed on such institutions. Not only did they have little economic impact, but those who were educated through such institutions tended to feel as if they were “enlightened” or somehow better than their fellow villagers.

This sense of elitism distanced them from farming and alienated them from their own rural upbringing.114 As Fei Xiaotong similarly pointed out, educated ruralites found it too disgraceful to return to the fields, a sense of shame that was felt by both themselves and their families.115

Finally, others criticized the extension system itself and how it functioned. One of the most apparent weaknesses with the system was disorganization. The China-United

States Agricultural Mission survey recorded:

Extension organization at present is loose and uncorrelated, with many different institutions, agencies, and organizations doing something in the field. The funds are thus badly scattered, personnel is used to poor advantage, administration of a unified program is impossible, and the people are uncertain and confused as to who is their friend and who is trying to take advantage of them.116

This weakness was also voiced by a number of agriculturalists during this time.

Agricultural scientist Shen Zonghan 沈宗涵 (1895-1980), for example, pointed out that many of the government-sponsored extension activities were “uncorrelated with the

114 Stig Thøgersen, A County of Culture: Twentieth-century China Seen from the Village Schools of Zouping, Shandong (Ann Arbor: University of Michigan Press, 2002), 57-90. The first agricultural schools in Zouping was opened in 1910, only to shift focus to sericulture two years later. In 1912, enrollment reached 19 students. (p. 50). 115 Fei Xiaotong, 136. 116 China-United States Agricultural Mission, 77. 39

extension system,” resulting in “inefficient use of funds and of personnel.”117 Moreover, extension was limited to a few regions within each province, failing to diffuse technological innovations nationwide.118 Eminent agricultural economist and scientist

John Lossing Buck similarly emphasized that work conducted at the many agricultural institutions in China was uncorrelated. Limited funding and adequate personnel made it especially difficult to manage these agricultural institutions effectively.119

Another criticism with how the agricultural extension system functioned was its close relationship to the central and local governments. According to the authors of an agricultural extension textbook, the institutional arrangement with extension work under the direction of the central and provincial government made it so that these efforts were easily influenced by “political trends” (zhengchao 政潮).120 This is most readily evidenced in that extension work frequently focused on export-promoting industries such as cotton, tobacco, silk, and tea.

The promotion of agricultural research and extension throughout Republican

China coincided with an “agricultural crisis” from 1920 to 1949, which, according to

Ramon Myers, was caused by an accumulation of war, poor weather conditions, and global market instability. The latter of these proved especially problematic as the

117 Shen Zonghan, Agricultural Resources of China (New York: Cornell University Press, 1951), 357. 118 Ibid., 368. 119 John Lossing Buck, “Some basic agricultural problems of China,” Secretariat Paper No. 1, Tenth Conference of the Institute of Pacific Relations, Stratford-on-Avon, England (New York: International Secretariat, Institute of Pacific Relations, September 1947), 55-62. 120 Instead, the authors called (unsuccessfully) for systematic change modeled after the United States land- grant college system, where the agricultural extension offices and experiment stations are both under the direction of the state (or provincial) agricultural university. See Zhang and Li, 87-88. 40

Republic of China became increasingly dependent on foreign imports while simultaneously witnessing a decline in food grain and handicraft exports.121

Extension work and extension education reflected this decline in exports. The authors of one agricultural extension textbook, for example, dedicated the first fifteen pages of text to a discussion of this issue. They presented tables and charts to show that from 1929 to 1933, China was heavily reliant on foreign countries for agricultural goods, while at the same time exports declined in some of its most important international markets, including tea and silk.122 Agricultural extension, as presented by these authors, was framed in nationalistic terms, with the “urgency” (poqie 迫切) of agricultural improvement and extension as “the only way to national rejuvenation.”123 Chinese educator Guo Bingwen 郭秉文 (1880-1969) similarly commented in 1941 that the

Foreign Trade Commission had “adopted a five year programme to increase the exports of wood-oil, tea, silk, wool, seeds and handicraft products through the use of scientific methods sponsored by the National Agricultural Research Bureau and the agricultural experimental stations in various provinces.”124 As Kenneth Pomeranz summarized, extension efforts “all emphasized crops that could improve China’s balance of trade and

121 Ramon H. Myers, “The Agrarian System” in The Cambridge History of China, vol. 13, Republican China 1912-1949, Part 2, ed. John K. Fairbank and Albert Feurwerker (Cambridge, UK: Cambridge University Press, 1986), 256-66. 122 Zhang and Li, 1-15. 123 Ibid., 14. 124 Gou Bingwen (Ping Wen Kuo), “Promotion and Application of Science in the Government of China” (1941), in Guo Bingwen xiansheng jinianji 郭秉文先生紀念集 (Mr. Guo Bingwen Memorial Collection) (Taipei: Zhonghua xueshuyuan 中華學術院, 1971), 196. 41

thus, it was hoped, help protect its autonomy; rural living standards per se got far less attention.”125

2.4 Summary

This chapter has shown how the state played an active role in agricultural experimentation and technology transfer in late-imperial China. This active role of the state continued into the late-Qing and Republican eras, as agricultural education, experimentation, and extension was institutionalized under the direction of the central government. A modern agricultural extensions system, incorporating existing agricultural schools and experiment stations, was set up by the Nationalists in 1929 to develop and encourage agricultural innovations to promote economic and social development. Despite high expectations for extension during the Republican era, both contemporaries and more recent scholarship have pointed out a number of weaknesses including a narrow scope of extension work, lack of adequate personnel, and an extension system that was both unorganized but also heavily influenced by politics.

125 Pomeranz, The Making of a Hinterland, 74. 42

Chapter 3

Agricultural Experimentation and Extension in Shandong, 1903-2013

For the celebration of the 110th anniversary of the Shandong Academy of

Agricultural Sciences in 2013, an academy history exhibition hall was established to highlight the history and accomplishments of the academy. The hall presented a long- term historical narrative of agricultural research dating back to pre-modern times, closely linking the past and the present. Upon entering the hall, the first section provided a brief overview of Shandong’s farming culture, featuring prominent local figures and agricultural treatises throughout imperial China such as Jia Sixie’s Qimin yaoshu (535) and Wang Zhen’s Nongshu (1313). This was followed with a presentation of the early stages of the academy with the establishment of the Shandong Agricultural Experiment

Station in 1903, and later growth during the Republican era. The largest part of the hall was dedicated to the third section, agricultural research conducted during the People’s

Republic of China, specifically emphasizing research on some of Shandong’s most important agricultural goods such as vegetables, peanuts, and cotton. Finally, before coming to an end, a future vision of SAAS and agricultural science in Shandong was on display and consisted of miniature models of modern research facilities and experimental plots.126

126 The above description is based on the author’s visit to the exhibition hall on May 12, 2014. 43

Despite clear political implications,127 this exhibition hall calls attention to the long-term historical continuities and transformations of agricultural experimentation and extension in China from the early modern to the modern era, as discussed in the previous chapter. Equally important, the exhibition hall highlights institutional continuity from the

Republic of China to the People’s Republic of China. This chapter looks at SAAS and its predecessors to show that agricultural research and extension institutions found in the

PRC today were shaped by institutions put in place during the late-Qing and Republican eras. Although agricultural research and extension was transformed under the Chinese

Communist Party (CCP), many of the research and extension institutions were inherited from the Nationalists and built upon after 1949.128 Additionally, this study further explores how agricultural experimentation and extension was closely influenced by

“political trends.” Since a number of former studies have already provided descriptions of the agricultural extension system129 and agricultural universities130 in China during this time, this chapter focuses primarily on Shandong’s provincial experiment stations and

127 Indeed, the narrative found in this exhibition hall coincides with Kirk Denton’s recent study on museums in the People’s Republic. Denton argued that museums serve as important “sites of memory,” shaping a collective memory and identity of the past, where the fluid nature of this historical memory has been controlled by the state to serve the present needs of economic growth and modernization. See Kirk A. Denton, Exhibiting the Past: Historical Memory and the Politics of Museums in Postsocialist China (Honolulu: University of Hawaii Press, 2014). 128 Several other scholars, especially Julia Strauss and Frank Dikötter, have similarly emphasized continuity in government policy and institutions throughout the twentieth century, with institution building in the late- Qing and Republican eras laying an important institutional foundation for the People’s Republic. See Julia C. Strauss, Strong Institutions in Weak Politics: State Building in Republican China, 1927-1940 (Oxford: Oxford University Press, 1998), 6; Julia C. Strauss, “The Evolution of Republican Government,” in Reappraising Republican China, 75-97; Frank Dikötter, The Age of Openness: China before Mao (Berkeley: University of California Press, 2008), 15. 129 For general studies of agricultural extension during the twentieth century, see, for example, Yang Shimou; Jørgen Delman; and Benedict Stavis, “Agricultural Research and Extension Services in China,” World Development 6, no. 5 (1978): 631-645. 130 For studies of agricultural universities during the Republican era, see Li Ying; and Randall Stross. 44

SAAS to provide a new perspective on agricultural research and extension in twentieth- century China.

3.1 Provincial Agricultural Experiment Stations, 1903-1937

Agricultural research and extension in Shandong province coincided with larger national trends to institutionalize former methods of research and technology transfer.

From 1903 to 1937, seven provincial-level experiment stations were established in

Shandong. The first experiment station in Shandong, and third nationwide, was the

Shandong Agricultural Experiment Station131 established in 1903 by the Jinan

Agriculture and Sericulture Society (Jinan nongsang hui 濟南農桑會). Located on the current site of SAAS in the northeastern region of Jinan, 180 mu 畝 (one mu equals roughly one-sixth of an acre) was purchased to conduct research primarily on local food crops and varieties of beans and cotton from the United States. To facilitate this research, an instructor from Japan, Tanii Kyoukichi 谷井恭吉 (dates unknown), was hired to provide training on agriculture and sericulture. The station’s objective was to promote agriculture and encourage farmers to implement improved methods, all under the direction of the Shandong Bureau of Agriculture, Industry, and Commerce

131 The name of the station changed in 1929 to the Shandong Provincial First Agricultural Experiment Station (Shandong shengli di’yi nongshi shiyanchang 山東省立第一農事試驗場). In 1936, the name once again changed to the Shandong Provincial Agricultural Research Institute (Shandong sheng nongye shiyansuo 山東省農業實驗所). To avoid confusion, this thesis refers to this station as simply the Shandong Agricultural Experiment Station. 45

(Nonggongshangju 農工商局).132 In 1906, the experiment station expanded to 500 mu, with over 50 offices. This same year, the Agriculture and Forestry School (Nonglin xuetang 農林學堂) was added on-site to the experiment station. In 1907, the name of the school changed to the Shandong Higher Agricultural School (Shandong gaodeng nongye xuetang 山東高等農業學堂), subsequently incorporating the experiment station under the direction of the school.133 As such, the experiment station was eventually relocated in

1914 to the western part of Jinan. A summary of the station from roughly 1912 to 1937 shows that the station had an average yearly budget of 12,369.6 yuan, was divided into five divisions – planting, sericulture, agronomic chemistry, pest control, and climate measurement – and had some 19 or so workers including a station director and secretary, along with several technicians, assistants, supervisors, and clerks.134

In addition to this comprehensive experiment station in Jinan, the six other stations focused on specific cash crop goods, two stations each for cotton, sericulture, and tobacco. In 1918, the first of two cotton experiment station was established in Linqing 臨

清, known at the time as the Shandong Provincial Linqing Cotton Experiment Station

(Shandong shengli Linqing mianye shiyanchang 山東省立臨清棉業試驗場) (Figure

132 “Shandong nongshi shiyanchang” 山東農事試驗場 (Shandong Agricultural Experiment Station) (1908), in Zhongguo jindai nongye shi ziliao 中國近代農業史資料 (Sources on the History of Modern Chinese Agriculture), vol. 1, ed. Li Wenzhi 李文治 (Beijing: Sanlian shudian 三联书店, 1957), 874. 133 Tang Qiming 唐齐鸣, “Cong Shandong nongshi shiyanchang dao Shandong nongye kexueyuan” 从山东农事试验场到山东农业科学院 (From Shandong Agricultural Experiment Station to Shandong Academy of Agricultural Science), Lishi zuyin 历史足音 (Historical Footprints) 3 (2013): 8-9. 134 Li Mengzhu 李梦竹, Gao Yuhuan 郜玉环, and Wang Qian 王茜, “Minguo qianqi Shandong guanfang nongye kexue yanjiu de chutan” 民国前期山东官方农业科学研究的初探 (Preliminary Study of Shandong Official Agricultural Science Research in the Early Republican Era), Chuangxin jiaoliu 创新交流 16 (2013): 70-73. 46

3.1).135 Located on 122 mu, and equipped with various instruments including a six horsepower gas engine and fourteen cotton gins, the station was established for the purpose of experimenting and improving various cotton varieties.136 Later in 1926, the second cotton experiment station, the Shandong Provincial Cotton Station (Shandong shengli mianzhongchang 山東省立棉種場), was established in Qidong 齊東.137 Located on 110 mu, the station consisted of 3 divisions – experimentation, extension, and general services – and primarily conducted research on native varieties of cotton and a variety from the United States called Trice. The first of two sericulture stations was established in 1918 in Yidu 益都 (later Qingzhou 青州),138 and the second in Yantai 煙台139 in 1937.

These stations conducted research primarily regarding raising silkworms, spinning silk, and growing mulberry trees. They published a number of qianshuo and trained farmers on these practices.140 In 1935, two tobacco stations were established, the first in Linzi 臨

135 In 1926, the name of the station changed to the Shandong Provincial First Cotton Experiment Station (Shandong shengli di’yi mianye shiyanchang 山東省立第一棉業試驗場). In 1935, the name changed again to the Shandong Provincial Cotton Improvement Branch (Shandong shengli mianzuo gailiang fenchang 山東省立棉作改良分場). 136 Linqing xianzhi 臨清縣志 (Linqing County Gazetteer) (1934), 291-292. 137 In 1930, the name of the station changed to the Shandong Provincial Second Cotton Experiment Station (Shandong shengli mianye shiyancheng 山東省立棉業試驗場) and in 1934 changed to the Shandong Provincial Cotton Improvement Station (Shandong shengli mianzuo gailiangchang 山東省立棉作改良場). 138 Known in 1918 as the Shandong Silkworm Production Office (Shandong canzhong zhizaosuo 山東蠶種製造所). Its name changed to the Shandong Provincial Sericulture Experiment Station (Shandong shengli canye shiyanchang 山東省立蠶業試驗場) in 1930. See “Quanguo nongye tuiguang shishi zhuangkuang diaocha” 全國農業推廣實施狀況調查 (National Agricultural Extension Implementation Survey), Nongye tuiguang, no. 9-10 (1935): 61-62. 139 Known as the Yantai Silk Improvement Station (Yantai cansi gailiangchang 煙台蠶絲改良場). 140 “Quanguo nongye tuiguang shishi zhuangkuang diaocha,” 61-62; “Shandong shengli canye shiyanchang chengji baogao” 山東省立蠶業試驗場成績報告 (Shandong Provincial Sericulture Experiment Station Achievement Report), Shandong sheng banyuekan 山東省半月刊 (Shandong Fortnightly) 2, no. 10 (1937): 112-202; and Zhuang Weimin 庄维民, “Jindai Shandong cansangye de gailiang ji qi yingxiang” 近代山东蚕桑业的改良及其影响 (Modern Shandong Sericulture Improvement and its Influence), Gujin nongye 古今农业 (Ancient and Modern Agriculture), no. 4 (1999): 35-42. 47

淄, the Tobacco Improvement Station (Yancao gailiangchang 煙草改良場), and the second in Qingzhou, the Tobacco Experiment Station (Huangyan shiyanchang 黃煙試驗

場).141 These tobacco stations were actively engaged in research and extension. 142 At the

Linzi station, for example, formal courses were set up running from August of 1935 to

March of 1936 on topics such as cultivation, baking leaves, and pesticides. A series of qianshuo on the same topics were also published in 1935.143 Extension work at these station focused primarily on the Linzi and Qingzhou region,144 as these were two of the most important tobacco production regions and markets in Shandong.145 In addition to

141 General introduction to these stations can all be found in Shandong shengzhi: nongye zhi 山东省志: 农业志 (Shandong Provincial Gazetteer: Agriculture), vol. 2 (Jinan: Shandong renmin chubanshe 山东人民出版社, 2000), 934-936. 142 For examples of types of research conducted, see “Yancao cuiya shiyan baogao” 煙草催芽試驗報告 (Tobacco Germination Experiment Report), Shandong sheng jianshe banyuekan 山东省建设半月刊 (Shandong Construction Fortnightly) 1, no. 15 (1936): 1-8; and “Yancao pinzhong bijiao shiyan baogao” 煙草品種比較試驗報告 (Tobacco Varieties Comparative Text Experiment Report), Shandong sheng jianshe banyuekan 1, no. 15 (1936): 17-29. 143 “Yancao gailiangchang Minguo ershisi nian zhuyao gongzuo baogao” 煙草改良場民國二十四年主要工作報告 (Tobacco Improvement Station Republic of China Year 24 Main Work Report), Shandong sheng jianshe banyuekan 1, no. 15 (1936): 95-100. 144 Ibid. 145 According to a 1937 survey, of 12 counties along the Jinan-Qingdao railway, Linzi ranked number three of twelve in the most amount of tobacco produced and land under cultivation, only slightly behind Linqu 臨朐 and Weixian 濰縣; not surprising, the tobacco market for Linqu was located at Yidu (Qingzhou), the location of the Qingzhou Tobacco Station. Further, the market for Linzi County, Xindian 辛店, sold the largest amount of tobacco, totaling 16,080 tons. Of these, 78 percent were sold to foreign companies, with over half to the British American Tobacco Company (BAT). Similarly, of the 4,800 tons of leaves sold at the Yidu market, 86.5 percent were sold to foreign companies, with 70 percent to BAT. See Li Xintian 李心田, “Shandong Meizhong yancao diaocha baogao” 山東美種煙草調查報告 (Shandong United States’ Varieties of Tobacco Survey Report), Guoji maoyi qingbao 國際貿易情報 (International Trade Information) 2, no. 9 (1937): 1-3; and Li Xintian, “Shandong Meizhong yancao diaocha baogao,” Guoji maoyi qingbao 2, no. 10 (1937): 25-30. For discussion of the rather exploitative and monopolistic nature of BAT, see Huang, The Peasant Economy and Social Change in North China, 130. 48

these stations, Yantai also had an aquiculture station established in 1917,146 and at least one silviculture experiment station was established in Tai’an in 1915.147

Figure 3.1 Linqing Cotton Experiment Station

Source: Linqing xianzhi, 1934.

In addition to these provincial-level experiment stations, county experiment stations were established throughout Shandong. In 1922, 50 county stations were in operation. By 1931, one source showed 58 provincial, city, and county experiment stations with annual expenditures of 75,167 yuan.148 A 1934 report listing 54 county experiment stations, showed that funding between the county stations varied widely, from

4,000 yuan at one station to a mere 58 at another. On average, however, county

146 See Shandong shengli shuichan shiyanchang jikan 山東省立水產試驗場季刊 (Shandong Provincial Aquiculture Experiment Station Quarterly), no. 1 (1924): 1. 147 Zuo Qiong 作瓊, “Shiyebu Shandong linye shiyanchang yange ji gaikuang” 實業部山東林業試驗場沿革及概況 (Department of Industry Shandong Silviculture Experiment Station History and Overview), Xinnong tongxun 新農通訊 (New Farming Newsletter), no. 14 (1933): 9-12. 148 Ibid., 936. 49

experiment stations typically had annual expenditures of around 300 to 400 yuan.149 In

1935, all county experiment stations were closed due to financial reasons. They were consolidated into four experiment districts, each dedicated to researching one crop:

Licheng 歷城, wheat; Huimin 惠民, soybean; Juxian 莒縣, sorghum; and Laiyang 萊陽, fruit trees.150

Research and extension work at the provincial-level agricultural experiment stations, being under the direction of the Shandong provincial government, were easily influenced by various “political trends.” Just as state commissioned agricultural treatises in late-imperial China represented imperial interests in specific crops, the strong emphasis on cash crop research, especially cotton, silk, and tobacco, also illustrates the provincial government’s interests in agricultural research and extension work that increased export- and industry-related goods. As Kenneth Pomeranz pointed out, state- impact on the rural Shandong economy varied across regions, as local officials from the late-nineteenth century to the 1930s became increasingly concerned about defending key economic regions and encouraging export-promoting crops.151 In addition to the

149 See Li Zhiji, “Guonei nongshi shiyan jiguan gaikuang (er)”, 9-12. 150 Shandong shengzhi, 936. The relative financial instability of county experiment stations appears to have been a problem outside of Shandong as well. For example, in his study of Dingxian, Sidney Gamble noted that in a cotton experiment station was established in 1919, only to be shut down in 1928 to be used as a public playground. Meanwhile, a tree nursery was established in 1918 and continued to operate past 1928, as the nursery brought in more direct revenue than cotton experimentation. See S idney D. Gamble, Ting Hsien, A North China Rural Community (New York: International Secretariat Institute of Pacific Relations, 1954), 136. 151 Pomeranz, The Making of a Hinterland, 19, 73, 274. In a related essay, Pomeranz stated directly: “More money was spent on protecting newly strategic areas near the coast and near railroad crossings; once- favored poor areas were abandoned.” This new form of statecraft had drastic effects of flooding in specific regions of China: “flooding was reduced sharply in the places that mattered most to China’s battles with foreign powers. But flooding in inland areas reached its worst levels on record.” See Pomeranz, “The Transformation of China’s Environment, 1500-2000,” 132.

experiment stations in Shandong reflecting an emphasis on import-substituting and export-promoting industries, the geographic location of these stations also coincide with

Pomeranz’s “Heartland” capital market region of Shandong (see Figure 3.2), which, although still restricted by local officials, was “far more involved in outside trade than the

Southwest was, and more a part of the North China regional economy than the North

Coast, which was instead closely tied to Manchuria.”152 Conversely, no provincial stations were located in the southwest region, which was characterized as being poor with incomplete provincial control.153 Discussing cotton improvement initiatives in the 1930s, historian Margherita Zanasi similarly pointed out that “preexisting cotton improvement programs developed around industrial centers,” which catered largely to “Shanghai and

Tianjin factories.”154

152 Kenneth Pomeranz, “Local Interest Story: Political Power and Regional Differences in the Shandong Capital Market, 1900-1937,” in Chinese History in Economic Perspective, 299. 153 Pomeranz, The Making of a Hinterland, 67. 154 Zanasi further stated that “the CCC planned to expand cotton cultivation in more remote areas, especially in the northwest, relying on its interregional marketing network to bring cotton from these more distant provinces to the Shanghai market.” Margherita Zanasi, Saving the Nation: Economic Modernity in Republican China (Chicago: University of Chicago Press, 2006), 134. 51

Figure 3.2 Shandong Regional Capital Markets in Republican China

Source: Adapted from Pomeranz, The Making of a Hinterland, 34.

3.2 Experimentation in Wartime Shandong, 1937-1949

Although Japan had an indirect influence in Shandong agricultural research prior to the Second Sino-Japanese War (1937-1945), its influence became increasingly intensified following the invasion of Shandong beginning in 1937. As historian Aaron

Moore has pointed out, by 1938, Japan’s Asia Development Board developed a number of plans for north China, including an agricultural plan that “prioritized rice and grain to solve China’s food crisis and cotton to help Japan’s declining textile industries.”155

Coinciding with Japan’s vision of “technology for Asian development,” this plan included land development, irrigation projects, and the introduction of agricultural

155 Aaron S. Moore, Constructing East Asia: Technology, Ideology, and Empire in Japan’s Wartime Era 1931-1945 (Stanford: Stanford University Press, 2013), 98. 52

technologies such as pesticides and fertilizers. Overall, the goal of this plan was to increase production of grains on less land in order to grow more cotton.156

To facilitate these plans, Japan consolidated a network of agricultural experiment stations in North China (Figure 3.3). In July 1936, the North China Industrial Science

Research Institute (Huabei chanye kexue yanjiusuo 華北產業科學研究所) was established in Qingdao, and later moved to Beijing in July 1937 after the outbreak of war.

In 1937, the Central Agricultural Experiment Station (Zhongyang nongshi shiyan chang

中央農事試驗場), later renamed the North China Agricultural Experiment Station

(Huabei nongshi shiyan chang 華北農事試驗場) on June 1, 1940, was also established in

Beijing. In 1938, efforts were made to consolidate various experiment stations in North

China (including parts of Hebei, Henan, Shanxi, and Shandong) into a central station in

Beijing, five branch stations (zhichang 支場),157 one sub-station (fenchang 分場),158 two experiment fields (shiyan di 試驗地),159 and 13 seed nurseries (yuanzhong pu 原種圃),160 as shown in Figure 3.3.

156 Ibid., 98-101. 157 Located in Shimen 石門, Junliangcheng 軍糧城, Jinan 濟南, Qingdao 青島, and Kaifeng 開封. 158 Located in Changli 昌黎. 159 Located in Jining 濟寧 and Xuzhou 徐州. 160 Located in Zhangde 彰德, Baoding 保定, Handan 邯鄲, Nanyuan 南苑, Tangshan 唐山, Dexian 德縣, Tai’an 泰安, Zhangdian 張店, Chengyang 城陽, Linqing 臨清, Xinji 辛集, Linfen 臨汾 and Xinxiang 新鄉. 53

Figure 3.3 Japan’s North China Agricultural Experiment Station Configuration Map

Source: Huabei nongshi shiyan chang, Huabei nongshi shiyan chang yaolan, 1943.

These stations, fields, and nurseries were consolidations of previously established agricultural institutions. The Jinan branch station, for example, consisted of the consolidation of a number of institutions including the Shandong Provincial Agricultural

Experiment Station, the Rural Construction Specialized School, and the Agricultural

College at Shandong University.161 Additionally, the Qingdao branch station was a consolidation of the Licun 李村 Experiment Station and Agricultural School in Qingdao, first established by Japan in 1917 and managed by the Nationalist government after

161 Huabei nongshi shiyan chang 華北農事試驗場 (North China Agricultural Experiment Station) ed., Huabei nongshi shiyan chang yaolan 華北農事試驗場要覽 (Overview of the North China Agricultural Experiment Station) (1943), 1-2. 54

1922.162 Eventually, county experiment stations were re-established under the direction of the Japanese government in Shandong, and by 1943, 48 stations were reopened in

Shandong.163 Former institutions were likely consolidated due to disorganization of former institutions and a shortage of personnel.164

Work conducted at these stations included crop experimentation, personnel training, and extension work. Experimentation was conducted on a number of staple crops including wheat, sorghum, corn, and rice, and on more economic crops such as cotton, tobacco, and various types of hemps. The type of experimentation on these crops included breeding, cultivation, soil fertility, fertilizer application, and pesticide use.165

The stations were also influential in training agricultural personnel. By March 1943, for example, a total of 658 personnel were employed by the North China Agricultural

Experiment Station, 364 of which were Japanese and 294 of which were Chinese. The

Beijing main station and the Jinan and Qingdao branch stations also established an agricultural technical training department. According to a 1943 report of the station, the two years prior 479 students graduate from this training; the Beijing station graduated

296 students in general agricultural and 20 students in livestock protection, and the Jinan and Qingdao branch stations graduated 85 students and 78 students in general agriculture

162 Shandong shengzhi, 934-936. 163 Ibid., 936-937. 164 As Japan invaded the northern and eastern parts of China, many Chinese scientists fled west toward Chongqing with the Nationalists. See, for example, the account of the National Southwestern Associated University in John Israel, Lianda: A Chinese University in War and Revolution (Stanford: Stanford University Press, 1998). 165 Huabei nongshi shiyan chang, 18-39; see also “Dongya nongye yanjiu de diantang: Huabei nongshi shiyanchang” 東亞農業研究的殿堂: 華北農事試驗場 (East Asia’s Agricultural Research Palace: The North China Agricultural Experiment Station), Xin nongye 新農業 (New Agriculture) 1, no. 1 (1945): 22- 23. 55

respectively. These graduates went on to work at the North China Agricultural

Experiment Station or in various agricultural societies and levels of government.166

Finally, experimental findings were popularized through extension efforts such as publications, workshops, and lectures, all for the purpose of “improving the agricultural technology of the general public.”167 Moore further commented on efforts to extend agricultural research, especially cotton, in North China:

Japanese “agricultural technicians” who first arrived in the colonies simply set up the same type of research stations they did in Japan and forced nearby peasants to plant their experimental seeds. But after encountering resistance, the technicians then decided to employ other techniques… in north China… they gathered elementary school teachers from 2,300 villages and lectured for several days on purchasing, cultivating, harvesting, and selling cotton in that region and terrain. After the technicians provided the teachers with free cottonseed, the teachers returned to their villages and asked students to plant them168

It is evident that the Japanese invested considerable time and finances to agricultural experimentation and extension in China.

After the war with Japan ended in 1945, the Nationalist government resumed control over agricultural experimentation in Shandong, continuing research and extension work from before 1937. The Nationalists retained control over experimentation and extension work until Jinan was “liberated” by Communist forces in 1948.

3.3 Shandong Academy of Agricultural Sciences, 1946-2013

According to an institutional history of SAAS published in 1994, the history of the academy dated back to an agricultural institute established in Junan in 1946 by the

166 Huabei nongshi shiyan chang, 38-39. 167 Ibid., 18. 168 Moore, 196-197. 56

Chinese Communist Party in conjunction with its Great Production Movement (da shengchan yundong 大生產運動).169 The history of the academy as recorded in this account, therefore, was closely linked to the Communist Party and its revolutionary vigor during a time of civil war with the Nationalists (1946-1949). This history, however, obscures the important historical continuity of agricultural institutions from the late-Qing and Republican eras to the PRC. Great efforts were made both during the war and following the establishment of the PRC to conduct experimental research and engage in extension work. However, these efforts were conducted within a consolidated institutional framework established earlier during the late-Qing and Republican eras. For example, in Shandong province, the leading education and research institutions – the

Shandong Agricultural University and the Shandong Academy of Agricultural Sciences – were both first established in the late Qing: SAAS dates back to the 1903 Shandong

Agricultural Experiment Station, and the Shandong Agricultural University dates back to the 1906 Agriculture and Forestry School.170

Prior to the victory of the CCP over the Nationalists in 1949, the CCP established agricultural improvement, guidance, and experiment stations in “liberated” regions of

Shandong. The CCP began to organize production efforts as early as 1944 in Shandong.

In 1944 and 1945, a number of experimental farms were established or reestablished in

169 Shandong sheng nongye kexueyuan, Shandong sheng nongye kexueyuan yuanshi, 1-3. For overview of the Great Production Movement in Shandong, see Yuan Pengxin 苑朋欣, “Shandong kangri genjudi nongye dashengchan yundong de zhengce yu xiaoguo” 山东抗日根据地农业大生产运动的政策与效果 (The Policies and Effectiveness of the Great Production Movement in Shandong’s Anti-Japanese Base Areas), Jinan daxue xuebao 济南大学学报 (Journal of Jinan University) 21, no. 4 (2011): 63-67. 170 Shandong nongye daxueshi 山东农业大学史 (Shandong Agricultural University History) (Shandong, Tai’an: Shandong nongye daxue dianzi yinxiang chubanshe 山东农业大学电子音像出版社, 2006), 3-12. 57

the Jiaodong 膠東 region. This laid the foundation for the Shandong Agricultural

Guidance Institute (Shandong nongye zhidaosuo 山東農業指導所), organized in January

1946 in Junan with agricultural scientist Zhang Ke 章柯 (dates unknown) as director. The purpose of this institution was to train personnel, and, as Zhang Ke later commented, many important agricultural cadres in the 1980s were students at this institute.171 During fall 1946, the Shandong Agricultural Research Institute (Shandong nongye shiyansuo 山

東農業實驗所) was established in Junan for experimental research. As the CCP expanded its territory, this research institute was relocated. In spring 1948, it moved to

Juxian, taking over a county experiment station established first during the Republican era. Shortly thereafter, the institute was relocated to Qingzhou. While there, a silk experiment station was reestablished at the former location of the Yidu silk improvement station first set up in 1918. Finally, in September of 1948, the station relocated to Jinan, eventually settling in the northwestern part of Jinan where the Shandong Agricultural

Experiment Station was first established in 1903. According to Zhang Ke, after a period of political ideological education and investigations, the institute retained a number of agricultural specialists that remained in Jinan after “liberation.”172 In 1950, the name of the institute was changed to the Shandong Agricultural Science Research Institute

171 This institute later closed in 1948. See Zhang Ke 章柯, “Wo suo liaojie de Shandong jiefangqu de nongye keji gongzuo” 我所了解的山东解放区的农业科技工作 (My Understanding of the Agricultural Science and Technology Work in Shandong’s Liberated Zones) (July 1984), in Shandong sheng nongye kexueyuan yuanshi, 182-183. 172 Ibid., 183. 58

(Shandong nongye kexue yanjiusuo 山东农业科学研究所), and in 1958 it was changed to the Shandong Academy of Agricultural Sciences.173

In addition to what later came to be known as SAAS, the Chinese Communist

Party built on the pre-existing infrastructure of provincial- and county-level agricultural institutions to establish research facilities throughout Shandong. As early as July 1948, meetings were being held to encourage CCP leaders at each administrative unit to establish experiment stations and “restore” (huifu 恢復) pre-existing stations.174 A summary of Shandong agricultural experiment stations in September 1948 reported that the former agricultural experiment stations in Fangzi 坊子 and Qingzhou had been reestablished, and at the county level, 59 experiment stations were being restored or newly built.175 At this time, a district agricultural experiment station was also established on the former location of one of the four experiment districts first established in 1935 in

Huimin. According to a September 1948 report, the station employed two technicians

173 For an overview of this period, see accounts by Zhang Ke, 181-184; and Chen Qiwen 陈启文, “Huiyi Shandong jiefangqu de nongye keyan gongzuo” 回忆山东解放区的农业科研工作 (Recollections of Agricultural Science Work in Shandong’s Liberated Zones) (May 1984), in Shandong sheng nongye kexueyuan yuanshi, 185-189. 174 “Shandong sheng shiye huiyi gexiang jueyi” 山东省实业会议各项决议 (Shandong Province Industrial Conference Resolutions) (July 1948), in Shandong jiefang hou keji shiliao xuanbian 山东解放后科技史料选编 (Selection of Historical Materials on Science and Technology in Shandong after Liberation), ed. Shandong sheng kexue jishu weiyuanhui 山东省科学技术委员会 (Shandong: Shandong sheng kexue jishu weiyuanhui, 1987), 129. 175 According to this report, the biggest challenges they faced at this time were rebuilding destroyed buildings from the war, a shortage of trained personnel, and a shortage of books. The experimentation and extension work conducted at these stations was strikingly similar to work reports from the Republican era. See “Shandong nongchang jianshe qingkuang baogao” 山东农场建设情况报告 (Report on the Construction of Shandong Agricultural Stations) (September 1948), in Shandong jiefang hou keji shiliao xuanbian, 134-135. 59

who previously worked as agricultural specialists in Shandong during the Republican era, one of whom served as the director of the Linqing cotton station for two years.176

Many of these experiment stations were later consolidated under the direction the

Shandong Agricultural Science Research Institute and later SAAS, building on the existing infrastructure of provincial- and county-level experiment stations to establish a network of branch stations throughout the province. By 1952, the former Linqing cotton station and the Yidu tobacco station were consolidated under the direction of the

Shandong Agricultural Science Research Institute in Jinan.177 Further, in 1958, a number of previously existing agricultural stations were consolidated to form research institutes under the direction of the Shandong Agricultural Science Research Institute. The Linqing and Yidu stations were upgraded to cotton and tobacco research institutes, respectively. A silviculture research institute in Tai’an was formed as a consolidation of the previous station first established in 1915. Moreover, a meteorology research institute was reorganized in western Jinan, where weather reports had been collected consistently since

1919.178

In addition to these institutions, the agricultural extension system, and the role of these institutions within that system, remained largely the same from before and after the establishment of the PRC. As pointed out in a number of studies, the agricultural extension system continued to be highly centralized and bureaucratized under the

176 “Shandong sheng Bohaiqu nongye shiyanchang baogao” 山东省渤海区农业试验场报告 (Report on the Shandong Bohai District Agricultural Experiment Station) (September 1948), in Shandong jiefang hou keji shiliao xuanbian, 135-137. 177 Shandong sheng nongye kexueyuan, Shandong sheng nongye kexueyuan yuanshi, 18-20. 178 Ibid., 55-58, 94-95. 60

direction of the Ministry of Agriculture.179 Under this system, academies of agricultural science (successors of the agricultural experiment stations) and agricultural universities have functioned throughout the PRC as affiliated institutions to the extension system, just as they did during the Republican era (see Figure 2.1).180 Even under collectivization, the provincial-level research institutes, agricultural academies, and agricultural universities supported the centralized extension system, although they worked directly with communes, production brigades, and production teams, rather than just county- and village-level extension stations.181

Such institutional arrangements ensured that these agricultural institutions continued to be heavily impacted by politics. Most notably, the Great Leap Forward

(1958-1961) and the Cultural Revolution (1966-1976) had immense impact on research conducted at institutions such as SAAS, as political ideology drove research and agricultural scientists were sent to the countryside for re-education.182 On the eve of the

Great Leap Forward, for example, Mao Zedong 毛泽东 (1893-1976) proposed the “eight- word charter” (bazi xianfa 八字宪法), advocating agricultural improvement in relation to soil (tu 土), fertilizer (fei 肥), water (shui 水), seeds (zhong 种), closeness (mi 密), protection (bao 保), management (guan 管), and implements (gong 工). This “eight-word

179 See, most notably, Jørgen Delman, Agricultural Extension in Renshou County, China. 180 See Delman, 60-61; and Zhang Xiaoyong and Corne Kempenaar, Agricultural Extension System in China (Wageningen: Plant Research International,2009): 3. 181 See, for example, Food and Agriculture Organization of the United Nations, Learning from China: A Report on Agriculture and the Chinese People’s Communes (Rome: Rood and Agriculture Organization of the United Nations, 1978), 55-56. For an extensive overview of extension during the 1950s and 1960s, see Benedict Stavis, Making Green Revolution: The Politics of Agricultural Development in China (Ithaca: Rural Development Committee, Cornell University, 1974), 172-190. 182 See Shandong sheng nongye kexueyuan, Shandong sheng nongye kexueyuan yuanshi, 49-104. 61

charter” had a significant impact on agricultural development and research during the

1950s and 1960s, both positive and negative.183 Two principles that proved to be especially problematic were the concepts of “soil,” which included the idea of deep plowing, and “closeness,” referring to close planting.184 The problem with deep plowing and close planting was not that they were theoretically incorrect, but that in the context of the Great Leap Forward, they were taken to an extreme that violated fundamental principles of ecology. As such, scholars have pointed to these principles as contributing to crop failures that led to widespread famine and the death of millions throughout the

PRC from 1959 to 1962.185

Agricultural research reports from SAAS during this time reflect an obsession with studying the effects of close planting and deep plowing on agricultural output, as research was focused on Mao’s eight-word charter and as many agriculturalists were forced to fabricate results to promote the Great Leap Forward.186 In 1959 and 1960, the

183 See Guo Shengfu 郭圣福, “Nongye ‘bazi xianfa’ pingxi” 农业八字宪法评析 (Analysis of the Agriculture Eight-Word Charter), Dangshi yanjiu yu jiaoyu 党史研究与教育 (Party History Research and Education), no. 6 (2008): 34-39. 184 The ideas of deep plowing and close planting were influenced by Soviet agricultural scientists and botanists, especially Trofim Lysenko (1898-1976). Although not strongly advocated by Chinese agriculturalists, Mao Zedong nonetheless found these ideas appealing and subsequently promoted them. See Kang Chao, Agricultural Production in Communist China, 1949-1965 (Madison: University of Wisconsin Press, 1970), 88. 185 Jasper Becker, Hungry Ghosts: Mao’s Secret Famine (New York: Henry Holt and Company, 1996), 72- 74; and Frank Dikötter, Mao’s Great Famine: The History of China’s Most Devastating Catastrophe, 1958-1962 (New York: Walker & Company, 2010), 37-40, 135. 186 See, for example, Tong Pingya 佟屏亚, “Nongye ‘dayuejin’ kaoliang nongye kexuejia: xiezai 1958nian nongye ‘dayuejin’ 50 zhounian” 农业”大跃进“考量农业科学家: 写在 1958 年农业”大跃进“50 周年 (Agricultural “Great Leap Forward” Examination of Agricultural Scientists: Writing on the 50th Anniversary of the Agricultural “Great Leap Forward”), Nongye kaogu 农业考古 (Agricultural Archaeology), no. 4 (2008): 7-16. 62

academy published a number of books on close planting for crops such as wheat,187 millet,188 corn,189 peanuts,190 and tobacco.191 Such publications were highly political in nature. The study on wheat, for example, opened by glorifying the party and Mao’s eight- word charter which led to a “bumper harvest of wheat in 1959 that is unprecedented,” thereby confirming that the eight-word charter “is our nations fundamental policy for increasing agricultural production.”192 The book on millet further broke down each of the eight charters to show how each aspect led to increased agricultural production. By increasing the number of plants grown per mu from around 20,000 to nearly 70,000, the study showed, output could increase by over 160 percent, and that deep plowing could further increase production and soil fertility. 193 Although some of the research findings showed clear limitations to the principles of close planting and deep plowing,194 these

187 Shandong sheng nongye kexueyuan ed., Shandong sheng 1959 nian xiaomai fengchan dianxing jingyan 山东省 1959 年小麦丰产典型经验 (Typical Experiences of Wheat Bumper Harvests in Shandong Province, 1959) (Jinan: Shandong renmin chubanshe, 1959). 188 Shandong sheng nongyeting 山东省农业厅 (Shandong Office of Agriculture) and Shandong sheng nongye kexueyuan eds., 1959 nian Shandong sheng guzi fengchan jingyan 1959 年山东省谷子丰产经验 (Typical Experiences of Millet Bumper Harvests in Shandong Province, 1959) (Jinan: Shandong renmin chubanshe, 1960). 189 Shandong sheng nongyeting and Shandong sheng nongye kexueyuan eds., 1959 nian Shandong sheng yumi fengchan jingyan 1959 年山东省玉米丰产经验 (Typical Experiences of Corn Bumper Harvests in Shandong Province, 1959) (Jinan: Shandong renmin chubanshe, 1960). 190 Shandong sheng nongyeting and Shandong sheng nongye kexueyuan eds., 1959 nian Shandong sheng huasheng fengchan jingyan 1959 年山东省花生丰产经验 (Typical Experiences of Peanut Bumper Harvests in Shandong Province, 1959) (Jinan: Shandong renmin chubanshe, 1960). 191 Shandong sheng nongyeting and Shandong sheng nongye kexueyuan eds., 1959 nian Shandong sheng huangyan fengchan jingyan 1959 年山东省黄烟丰产经验 (Typical Experiences of Tobacco Bumper Harvests in Shandong Province, 1959) (Jinan: Shandong renmin chubanshe, 1960). 192 Shandong sheng 1959 nian xiaomai fengchan dianxing jingyan, i. 193 1959 nian Shandong sheng guzi fengchan jingyan, 3, 7-9. 194 A study of close planting of cotton at the academy in 1958, for example, clearly showed that under uniform conditions, planting too close stunted growth of the cotton plants, produced less cotton bolls, and higher rates of boll shedding. According to this study, the ideal number of plants per mu was somewhere between 4,000 to 5,000, and was 4,240 according to one study. The authors clarified, however, that these experiments were all conducted under general management conditions, without taking any effective measures to increase light intensity. If better management and cultivation methods were adopted, with improved irrigation, fertilization, soil conditions, and increased light intensity, the authors concluded that 63

publications nonetheless promoted close planting and deep plowing as advocated by

Mao.

This focus by scientist at SAAS on scientific results that were deemed “politically correct” was solidified in part by personal visits from many of the top leaders of the CCP during the Great Leap Forward. Mao, for example, made two visits to the academy, once in August 1958 and again in September of 1959. During both of these trips, Mao was particularly interested in cotton production, visiting the cotton experiment field attached to the academy. Agricultural scientist Qin Jie 秦杰 (1908-1984) later recalled that Mao’s visit and specific interest in cotton prompted additional attention to cotton research at the station.195 In addition to Mao, other CCP leaders such as Zhu De 朱德 (1886-1976) and

Deng Xiaoping 邓小平 (1904-1997) visited the academy in 1959.

A comprehensive institutional history of SAAS is beyond the scope of this study, especially due to the rapid growth and expansion of this institution throughout the

People’s Republic of China. With 22 current research units and over 1,800 employees, the magnitude and scope of research conducted by the academy today is unprecedented by the standards of late Qing and Republican China. In recent years, as CCP leadership is

over 5,000 plants per acre could be planted and still ensure high production of cotton bolls (pp. 103-105). See Shandong sheng nongye kexue yanjiusuo 山东省农业科学研究所 (Shandong Agricultural Science Research Institute), “Guanyu mianhua mizhi zengchan de diaocha yanjiu he jinhou yijian” 关于棉花密植增产的调查研究和今后意见 (On the Investigation of Close Planting Increasing Cotton Yields and Future Recommendations), in 1958 nian nongzuowu mizhi jingyan 1958 年农作物密植经验 (Agricultural Crops Close Planting Experiments, 1958), ed. Nongye bu liangshi zuowu shengchan ju 农业部粮食作物生产局 (Ministry of Agriculture Crop Production Bureau) (Beijing: Nongye chubanshe 农业出版社, 1958), 98- 105. 195 See Qin Jie 秦杰, “Weida de guanhuai, wuqiong de liliang – jinian Mao zhuxi liangci shicha woyuan mianhua shiyan tian” 伟大的关怀, 无穷的力量 – 纪念毛主席两次视察我院棉花试验田 (Great Care, Infinite Power – Remembering Chairman Mao’s Two Visits to our Academy’s Cotton Experiment Fields) (August 1978), in Shandong sheng nongye kexueyuan yuanshi, 192-4. 64

especially concerned with environmental protection and redeveloping rural infrastructure, agricultural institutions such as SAAS are conducting important research and extension work needed to improve the vast environmental issues in China today.196 Recent leaders of the CCP have shown their support as Hu Jintao 胡锦涛 (1942- ) visited in 2009, and

Xi Jinping 习近平 (1953- ) most recently visited in 2013 in celebration of the 110th anniversary of the academy.

The changing historical narrative of SAAS to recognize its pre-Communist roots is significant, but not simply for highlighting a pursuit for historical accuracy or a fascination with longevity. Rather, this shift recognizes the important institutional continuity from the late-Qing and Republican eras to the present day, continuity that was previously overshadowed by Communist rhetoric and political correctness. The exponential expansion of agricultural research and extension during the People’s

Republic of China was not historically isolated, but was made possible by institution building taking place prior to 1949. The late-Qing and Republican eras laid a foundation of agricultural experiment stations, universities, and a centralized extension system that expanded and grew after 1949, but nevertheless remained the same in significant ways.

196 For example, environmental scientist Karen Mancl found that agricultural academies such as SAAS play an especially important role in environmental technology transfer to rural China. See Karen M. Mancl, “Environmental Technology Transfer in Rural China” (master’s thesis, Ohio State University, 2010), 66- 68. For an excellent overview of environmental issues in the People’s Republic, see Robert B. Marks, China: Its Environment and History (Maryland: Rowman & Littlefield, 2012), 265-330. 65

3.4 Summary

This chapter has explored the development of the Shandong Agricultural

Experiment Station and other provincial-level stations in Shandong beginning in the late

Qing, the evolution of these institutions during the Republican era, and the legacies of these stations on later development in the People’s Republic of China. During the late-

Qing and Republican eras, agricultural experiment stations in Shandong engaged in research and extension work on important cash crops such as cotton, tobacco, and silk.

Following Japan’s invasion of Shandong in 1937, the existing agricultural institutions were consolidated to form a network of research and extension under the direction of the

Japanese in North China. These agricultural institutions, especially the provincial- and county-level experiment stations, were again consolidated and expanded under the

People’s Republic of China. The transformation of the 1903 Shandong Agricultural

Experiment Station to the present Shandong Academy of Agricultural Sciences is one example. The rapid expansion of agricultural research and extension in the PRC was made possible by a pre-existing network of agricultural experiment stations, universities, and a centralized extension system developed during the Republican era.

Chapter 4

Cotton, Extension, and Uneven Economic Growth

Writing in 1933, industrialist Rong Zongjing 榮宗敬 (1873-1938) warned that unless cotton production increased, money would increasingly “flow out of China.” For this reason, he encouraged the Nationalist government to “instruct each provincial

Department of Construction 建設部 to actively increase cotton fields, appoint agricultural specialists to effectively persuade farmers to plant cotton,” and even carry out “compulsory production” (qiangpo shengchan 強迫生產). Not only did Rong believe that such efforts by the state would lead to dramatic increases in cotton production within a few years, but also that the subsequent increase in cotton production would be a “key element in reviving the countryside.”197

Rong was not unique in emphasizing the importance of cotton production and the cotton textile industry in rural Republican China. Indeed, during the Republican era, the importance of the cotton industry for the national and rural economy was emphasized generally by a number of other industrialists, economists, and politicians.198 Moreover, by 1933 cotton research and extension efforts were already being conducted at various

197 Rong Zongjing 榮宗敬, “Yijian shu: zengjin mianchan wei fuxing nongcun zhi yaosu” 意見書: 增進棉產為復興農村之要素 (Opinion: Promoting Cotton is a Key Element in Reviving the Countryside), Nongcun fuxing weiyuanhui huibao 農村復興委員會會報 (Reports of the Committee to Revive the Villages), no. 3 (1933): 22. 198 For general discussion, see Zanasi, 136-142. 67

state-run agricultural experiment stations and schools to produce higher quality varieties of cotton in hopes of promoting the local textile industry. Therefore, statements such as

Rong’s regarding the important role of the state in increasing cotton output, while not unique, raise the question: to what degree were provincial governments effective in stimulating the economy through research and extension?

This chapter seeks to answer this question by providing a study of the impact of cotton research and extension efforts in Shandong that finds that its economic impacts were geographically constrained and spatially uneven. Output of improved varieties of cotton was largely concentrated around the two provincial cotton experiment stations located in Linqing, in northwest Shandong, and Qidong, north of Jinan. This impact was caused primarily by extension work which targeted counties near the cotton stations, and limited access to resources by farmers geographically estranged from extension services.

A look at the diffusion of improved varieties of cotton, therefore, illustrates how agricultural extension work in general was conducive to promoting spatially uneven growth that favored regions within close geographic proximity to extension sites.199

Organization of this chapter proceeds as follows. First comes an overview of the sources and methods utilized in this chapter. Second, a brief overview of the history of cotton production in China is presented to contextualize the introduction and spread of new cotton varieties during the first half of the twentieth century. Next, Shandong is

199 With a focus on state-led experimentation and extension, this chapter does not focus on the role of the market in promoting improved cotton varieties. The quantitative data presented in this chapter does not clearly show the varying degrees of impact that the market and extension services each had on increased output. What this chapter does show, however, is the limited geographic impact of extension services in diffusing improved varieties to spatially-estranged parts of Shandong. 68

taken as a case study to assess how cotton research and extension work during the

Republican era contributed to the uneven diffusion of improved cotton varieties. Such findings have implications for understanding spatial economic inequality and economic growth in general during this period, as discussed in the final section.

4.1 Sources and Methods

The two most detailed series of cotton statistics in Republican China were collected by the National Agricultural Research Bureau (Zhongyang nongye shiyansuo

中央農業實驗所, NARB) and by the Chinese Cotton Millowners’ Association

(Huashang shachang lianhehui 華商紗廠聯合會, CCMA). The NARB series comes from the annual crop reports published in Nongqing baogao 農情報告 (Crop Reports; 1934-

1938) with estimated national- and provincial-level output for cotton and other crops from 1933 to 1937. The CCMA series was collected for a much longer period of time, roughly 1918 to 1937, published first in Mianchan diaocha baogao 棉產調查報告

(Chinese Cotton Production Survey Report; 1920-1922) and later renamed to Zhongguo mianchan tongji 中國棉產統計 (Chinese Cotton Production Statistics; 1923, 1928-1938).

The latter was published by the Chinese Cotton Statistics Association (Zhonghua mianye tongjihui 中華棉業統計會) beginning in 1931. From 1924 to 1926, chaotic conditions in the countryside prevented the association from surveying cotton production, however

estimates for these years were later published in 1928 based on questionnaires sent out to the countryside.200

The NARB series likely presents a more accurate picture of national output, as it surveyed more provinces than the CCMA data. The NARB series consists of statistics for

18 to 21 of the 28 provinces in Republican China,201 whereas CCMA at most surveyed 12 provinces in any given year.202 The number of counties surveyed by each set also varied, with NARB surveying a larger number of counties than CCMA, as illustrated for

Shandong in Table 4.1. Although the NARB series surveyed more counties than the

CCMA series, these crop reports did not provide county-level data, and did not differentiate between output for native and foreign varieties, both of which are found in the CCMA series. Further, as shown in Table 4.1, estimated output varied between these two series. This difference was due to varying sample sizes and a more fundamental issue with data collection by NARB. As economist Richard Kraus pointed out, NARB collected data for the acreage of summer crops in July, only collecting data on yield per mu after harvest. Therefore, “in any year when growing conditions improve or decline radically late in the year, the NARB system would likely produce under- or over-

200 For an overview of the two series, see Richard A. Kraus, Cotton and Cotton Goods in China, 1918-1936 (New York: Garland Publishing, Inc., 1980), 13-22. For a thorough review of cotton statistics in China up to 1929, including the CCMA and other series, see Jiang Dixian 蔣迪先, “Zhongguo mianchan tongji zhi guoqu ji jianglai” 中國棉產統計之過去及將來 (The Past and Future of Cotton Production Statistics), Zhongguo mianchan tongji 中國棉產統計 (Chinese Cotton Production Statistics), no. 18 (1929): 1-42. 201 More specifically, 18 provinces were surveyed in 1933; 20 provinces in 1934; 21 provinces in 1935 and 1936; and 20 provinces were surveyed in 1937. 202 The number of provinces surveyed ranged from eight to twelve from 1918 to 1938. The provinces surveyed were Hebei, Shandong, Henan, Shanxi, Shaanxi, Jiangsu, Anhui, Jiangxi, Hunan, Hubei, Zhejiang, Sichuan, and Liaoning. 70

estimates respectively.”203 Conversely, the CCMA series was continually revised, with the last estimate recorded after harvest.204

Table 4.1 CCMA and NARB Series Coverage and Estimated Annual Output in Shandong CCMA Series NARB Series # of Counties # of Counties Year Output (dana) Output (dan) Surveyed Surveyed 1918 18 720,787 - - 1919 18 894,558 - - 1920 21 126,070 - - 1921 32 304,077 - - 1922 38 1,005,230 - - 1923 40 1,387,666 - - 1924 - 937,224 - - 1925 - 995,603 - - 1926 27 518,279 - - 1927 27 709,755 - - 1928 27 620,413 - - 1929 32 1,213,080 - - 1930 36 2,170,658 - - 1931 37 2,154,882 - - 1932 41 1,769,394 - - 1933 40 1,468,932 77 880,000 1934 61 1,334,053 85 2,060,000 1935 61 407,215 93 1,139,000 1936 87 1,790,227 100 2,554,000 1937 82 1,630,357 82 1,852,000 Sources: Data for CCMA series from Huashang shachang lianhehui, Mianchan diaocha baogao, no. 8 (1920), no. 9 (1921), no. 10 (1921), no. 11 (1922); and Zhonghua mianye tongjihui, Zhongguo mianchan tongji, no. 12 (1923), no. 17 (1928), no. 18 (1929), no. 19 (1930), no. 20 (1931), no. 21 (1932), no. 22 (1933), no. 23 (1934), no. 24 (1935), no. 25 (1938); data for the NARB series from Nongqing baogao 農情報告 (Crop Reports) vol. 2, no. 1 (1934), vol. 3, no. 1 (1935), vol. 4 no. 1 (1936), vol. 4, no. 12 (1936), and vol. 6, no. 2 (1938). a One dan 擔 (picul) equals 131.58 pounds, or roughly 60 kilograms.

Despite apparent complications, especially the inconstant number of counties surveyed, the CCMA series serves as the most descriptive source of data for a study of the output of foreign cotton varieties from the United States (hereafter U.S. cotton varieties) in Shandong. Although the total estimated national and provincial output might

203 Kraus, 21. 204 Ibid., 22. 71

be less reliable, the estimated output of U.S. varieties is arguably more accurate. One of the biggest difficulties facing the CCMA was a lack of time and human resources.

Beginning in 1919, ten individuals were sent out to collect cotton statistics, one of which was assigned to Shandong. Due to time restrictions, these individuals primarily visited the main production regions/counties, and for any additional data consulted with local institutions affiliated with agricultural or cotton production.205 Moreover, in 1924 it was proposed that provincial cotton and agricultural stations be in charge of estimating final cotton output.206 As early as 1931 the CCMA statistics nationwide were being collected primarily by provincial governments and agricultural organizations, including cotton experiment stations and agricultural universities. In Shandong statistics for 1931 were collected by individuals at the Linqing (Zhang Zecheng 張則程) and Qidong (Hu

Pingchu 胡平初) provincial cotton stations, along with the Department of Agriculture at

Qingdao University (Ren Jimin 任濟民).207 An extension report from the Qidong cotton station in 1932 further elaborated on the collection of these statistics:

In cooperation with the Chinese Cotton Statistics Association, this station handled the matter of collecting cotton statistics in the counties of Shandong. The station conducted the surveys – separated into the four cotton districts of south Shandong, west Shandong, north Shandong, and the Xiaoqing 小清 river basin – by dispatching personnel to each cotton-producing county to conduct field surveys. The first survey estimation was conducted in July of this year, and a second survey was conducted in November.208

205 Jiang, 14. 206 Ibid., 9-10. 207 Zhonghua mianye tongjihui, Zhongguo mianchan tongji, no. 20 (1931): 3-4. 208 Shandong shengli di’er mianye shiyanchang 山東省立第二棉業試驗場 (Shandong Provincial Second Cotton Experiment Station) ed., Shandong shengli di’er mianye shiyanchang tuiguang baogao 山東省立第二棉業試驗場推廣報告 (Shandong Provincial Second Cotton Experiment Station Extension Report), no. 1 (1932), 36. 72

This close connection between surveying and the provincial cotton experiment stations supports the reliability of the statistics for U.S. cotton varieties, as the diffusion of U.S. cotton seeds and cotton output was closely monitored by these stations and other provincial-level organizations in charge of collecting the data. Historian Kenneth

Pomeranz has further commented on the reporting of U.S. cotton statistics:

there are good reasons for this particular item to have been more accurately reported than others, since… there were relatively centralized county offices that provided the seeds (and threatened to withhold it from those who did not report results and turn in the seeds from their harvest), and collected the cotton; moreover, this cotton, unlike domestic varieties, was not only rarely used on the farm, but was almost always shipped out of the county it was produced in, making it far less likely than domestic cotton (or for that matter, most other crops) to escape detection.209

Additionally, qualitative and quantitative sources seem to confirm that the counties surveyed in the early years represented the majority of U.S. varieties being grown in

Shandong, as only later would these varieties be diffused to counties found in later surveys. Even when surveying additional counties in later years, for example, the bulk amount of U.S. varieties produced in the 1930s continued to be in those counties found in the 1918-1919 reports; the 18 counties found in the 1918 and 1919 surveys produced 81.8 percent of total provincial output of U.S. cotton varieties in 1936, when a total of 86 counties were surveyed. The overall increasing output of U.S. varieties as found in the

CCMA series, therefore, cannot simply be attributed to an increased number of counties

209 Kenneth Pomeranz, “The Making of a Hinterland” (Ph.D. dissertation, Yale University, 1988), 404-405. In his study, Pomeranz also used the CCMA series in his study, although his data source comes from a condensed version of the CCMA series reproduced in Zhongguo Shiyebu 中國實業部, Guoji Maoyiju 國際貿易局, Zhongguo shiye zhi: Shandong sheng 中國實業志: 山東省 (Industrial Gazetteer of China: Shandong Province) (Nanjing: Guiji maoyiju 國際貿易部, 1934), 159-161. 73

surveyed. In summary, although U.S. cotton varieties were almost certainly grown in counties not found in various years of the CCMA series, the CCMA series arguably captured the bulk amount of output of U.S. cotton for the years surveyed, thereby presenting a relatively accurate picture of cultivation and diffusion of U.S. varieties at the county level during this time.

Data from the CCMA series is presented here in the form of tables and maps, the latter of which was created by using geographic information systems (GIS) technology.

Although less commonly used by historians than by other researchers,210 GIS is an important tool that can be used in historical research to draw additional attention to elements of space and environment – both natural and human. ERSI’s ArcGIS software was utilized with data from the CCMA series. Provincial and county information was provided by the China Historical Geographic Information System (CHGIS) project, available for download online [http://www.fas.harvard.edu/~chgis/index.html].211 As seen below, the focus of this study is specifically on the spatial element of innovation diffusion and uneven economic growth. Elements of natural environment, such as topography, are less central to this analysis and await further research. Data used in the maps consist of three to four year averages of U.S. cotton output. The first set of years,

210 The use of GIS software in this study was in many ways inspired by historian Philip C. Brown’s recent book, Cultivating Commons, which used GIS technology to assess the relationship between land redistribution systems (known as warichi, or “dividing the land”) and vulnerability to natural disasters such as floods in early modern Japan. See Philip C. Brown, Cultivating Commons: Joint Ownership of Arable Land in Early Modern Japan (Honolulu: University of Hawai’i Press, 2011), 144. For an overview of some of the advantages and complications with using GIS in historical analysis, see p. 50-52. 211 The county-level information used in this study was based on the CHGIS 1911 Data. In Shandong, county names and boundaries shifted from 1911 to the 1920s and 1930s. Adjusting for changes in county names proved to be far less difficult than geographical boundaries. Despite shifting geographic boundaries, however, the 1911 Data is still useful in mapping the overall spread and concentrated cultivation of improved varieties in the 1920s and 1930s. 74

1921-1923, represent output during the early 1920s. Output of U.S. cotton varieties at the county level are lacking for the years 1924 and 1925, so the next set of years, 1926-1929, show output during the latter half of the 1920s. Data from 1930 to 1937 is split into two four year averages, with 1930-1933 illustrating U.S. cotton yields during the early 1930s, and 1934-1937 representing output on the eve of the Second Sino-Japanese war. It is worth noting that the year 1935 was an especially poor year for total cotton output due to poor weather conditions throughout Shandong. Many studies remove this year from analysis based on the premise that it is clearly a statistical anomaly and thereby obscures the overall average output during this time. However, the year 1935 was retained in most averages for this study as including this year in the analysis had little effect on the average difference in output between native and foreign varieties, and output disparities between cotton regions and counties.

4.2 Historical Overview of Cotton Cultivation in China

Cotton cloth was known in China as early as the Tang dynasty (618-907), if not earlier. Cultivation of cotton, however, did not begin until the Song dynasty (960-1279) as it spread to border regions through two routes: from Indochina, spreading into

Guangdong and Fujian; and from the West via the Silk Road, spreading into Shaanxi and

Gansu. It was from these peripheral regions that cotton cultivation eventually spread to the rest of China, although it was not until the thirteenth century that cotton was cultivated in the Yangzi valley, and even later before cultivated in northern provinces such as Shandong and Zhili (Hebei).

This expansion in the thirteenth century was made possible by the development of a new variety of cotton whose biological character made it more suitable to various climate and soil conditions, thereby capable of cultivation in the north.212 The expansion of cotton cultivation was also encouraged by the Yuan (1271-1368) and the Ming (1368-

1644) empires. The Mongol empire specifically set up a cotton production bureau to instruct local farmers on cultivation and weaving techniques.213 Cotton cultivation further expanded during the Ming largely due to a huge demand by the government to provide cotton for the military, officials’ salaries, and trade. Cultivation and production of cotton further increased in the northern provinces beginning in the early seventeenth century when farmers discovered a method to control humidity and temperature by digging cellars; the dryer climate in the north made cotton spinning and weaving difficult as the fibers would easily dry out and break.214 As such, according to Francesca Bray, by the late Ming cotton was “the most important fibre crop in China.”215 Overall cotton production is believed to have increased considerably during the Qing dynasty, although cultivation geographically did not change significantly.216

Another crucial transformation in cotton cultivation came with the introduction of foreign varieties of cotton, mostly from the United States, during the late nineteenth and early twentieth centuries. The new varieties of cotton had longer fibers, usually between

212 Kang Chao, The Development of Cotton Textile Production in China (Cambridge, MA: Harvard University Press, 1977), 14. Chao also points out that the spread of cotton cultivation and consumption in China was slowed by the strong silk industry, p. 10-11. 213 Bray, Science and Civilision in China, 539. 214 Chao, The Development of Cotton Textile Production in China, 20-21. 215 Bray, Science and Civilision in China, 539. See also Michael Loewe, Early Life in Early Imperial China: During the Han Period 202 BC-AD 220 (London: Batsford, 1968), 20-21. 216 Chao, The Development of Cotton Textile Production in China, 24. 76

1 inch (25 mm) and 1.5 inches (38 mm), thereby making them capable of producing higher quality yarn. In contrast, the length of the native varieties were around 0.75 (19 mm) to 0.8 inches (22 mm) and had low elasticity. This resulted in low tensile strength that became especially problematic with the introduction of power looms in the nineteenth century.217 New high quality varieties, therefore, were introduced through efforts by the national and provincial government, in addition to agricultural universities,218 through establishing experiment stations and extension services to improve upon and diffuse a number of varieties, including Trice, Kings’ Improved,

Stoneville, and Delfose.219

By the beginning of the Republican era, according to Richard Kraus, “cotton growing and its manufacture occupied a central position in the Chinese economy… In the agricultural sector cotton was the leading cash crop and the leading industrial raw material produced by the agricultural sector.”220 According to the national surveys conducted by John Lossing Buck from 1929-1933, cotton was by far the dominant material used for clothing in rural China, consisting of roughly 91 percent of work garments and 79 percent of dress garments. Silk was the second most used material,

217 Ibid. 218 For a rather positive overview of efforts by Nanjing University, see Shen Zonghan, “First Attempts to Transform Chinese Agriculture, 1927-1937,” in The Strenuous Decade: China’s Nation-Building Efforts, 1927-1937, ed. Paul K. T. Sih (New York: St. John’s University Press, 1970), 205-232; and Shen, Agricultural Resources of China, 312-315. See also Stross, 116-142. 219 Chao, The Development of Cotton Textile Production in China, 25-26. 220 Kraus, 5. 77

although it only consisted of a mere 1 percent of work garments and 7 percent of dress garments.221

Whether or not cotton cultivation actually increased during the Republican era, and how much of that increase can be attributed to the active role of the state, has been contested. Economist Kang Chao, for example, singled out the year 1919, with the establishment of the Chinese Cotton Improvement Commission, as a “turning point” in cotton cultivation and production in China. Foreign experts primarily from the U.S. and

Japan were hired, cotton experiment stations were established on the national and provincial level, and scientific experimentation was conducted throughout the 1920s and reached its peak in the 1930s, all for the purpose of breeding and diffusing superior cotton varieties in China.222 Kraus, however, is skeptical of state and private extension policies. By looking at long-term trends, Kraus credited the annual fluctuations in cotton output to weather conditions, rather than the “enlightened and vigorous government and private extension policies.”223 Contrarily, historian Margherita Zanasi has argued that

Kraus’s long-term analysis obscures the impact of state efforts especially that of the

Cotton Control Commission (CCC), a national-level association established in 1932 aimed at developing the cotton industry through extension efforts such as introducing new seeds and cultivation practices. Zanasi argues that “the short-term analysis shows a crisis in the late 1920s and early 1930s that was overcome with the important (if not sole)

221 Buck, Land Utilization in China, 438-439; Kraus for one believes that these statistics for the importance of cotton in clothing “remained essentially stable throughout the Twenties and Thirties and probably throughout the whole of the first four decades of the Twentieth century.” Kraus, 4-5. 222 Chao, The Development of Cotton Textile Production in China, 26. 223 Kraus, 7. 78

contribution of the CCC.”224 While this chapter generally agrees with Zanasi’s conclusion that the national and provincial government contributed to the growth in cotton production, it also emphasizes that such efforts must be understood spatially and in a longer historical context, as efforts by the CCC consisted of a continuation of efforts – although possibly more centralized – dating back to the late nineteenth and early twentieth centuries. The following case study of Shandong attempts to provide a more long-term and spatially sensitive understanding of the economic impacts of the provincial government’s efforts to increase cotton production through experimentation and extension work.

Cotton in Shandong

Cotton cultivation became increasingly widespread throughout Shandong during the sixteenth century with the growing market demand for raw cotton in the lower

Yangzi. As agronomist Xu Guangqi notably commented in the early seventeenth century,

“Nowadays in the north the cotton is cheap while cloth is expensive; in the south the opposite is true. Cotton is therefore shipped to the south, while cloth is shipped to and sold in the north.”225 This demand for raw cotton in the south led to the expansion of cotton cultivation in both the northwest and southwest regions of Shandong. Jing Su and

Luo Lun, for example, concluded that, during the Qing dynasty, Shandong had four main

224 Zanasi, 157; short-term averages of imported cotton further supports in part Zanasi’s claim of the role of the state in increasing cotton production. Averages over time were 2.3 million (1925-1927), 2.6 million (1928-1930), 3.7 million (1931-1933) and 1.4 million (1934-1936). Although cotton consumption in China remained relatively constant from 1924 to 1936, imports tended to increase from 1924, reaching its peak at 4.7 million piculs in 1931, then dropped dramatically to .8 million piculs in 1936. See Kraus, 52. 225 As quoted in Huang, The Peasant Economy and Social Change in North China, 112. 79

cotton production regions: the region south of the Yellow River with Qidong at the center; just north of the Yellow River with Linyi 臨邑 at the center; the northwestern region with Liaocheng 聊城 as the center; and the southwestern region surrounding

Yuncheng 鄆城.226

By the late-Qing and Republican eras, the main cotton production regions had changed slightly, as cotton cultivation was generally separated into three production districts, all located in the western half of Shandong. The west Shandong region consisted of counties west of the Tianjin-Pukou railway and north of the Yellow River, with

Linqing as the highest producing county. Both native and U.S. varieties were grown in this region, with native production predating the Republican era. The north region comprised of counties east of the Tianjin-Pukou railway north of the Yellow River, primarily regions bordering the Xiaoqing river, such as Binxian 賓縣 and Qidong.

Similar to the west region, both native and U.S. varieties were grown. Finally, the south region was made up of counties south of the Yellow River and west of the Tianjin-Pukou railway. Cotton grown in this region was less than the other two regions, almost all was native varieties, and was primarily used within the district due to a lack of integration with the provincial cotton market.227 As we will see further below, a fourth region in eastern Shandong began to emerge in the 1920s and 1930s as counties in the Jiaodong region, such as Gaomi 高密 and Changyi 昌邑, began to cultivate native and U.S.

226 Jing Su and Luo Lun, Landlord and Labor in Late Imperial China: Case Studies from Shandong, trans. Endymion Wilkinson (Cambridge, Mass.: Harvard University Press, 1978), 79-83. For a good overview of the expansion of cotton cultivation during the Ming and Qing, especially in relation to commercialization, see Huang, The Peasant Economy and Social Change in North China, 111-120. 227 Kenneth Pomeranz, The Making of a Hinterland, 67. 80

varieties of cotton.228 The climate and soil conditions in the three main cotton production regions were particularly good for cotton production, especially the alluvial and sandy soils located on both sides of the Yellow River.229 As shown in Figure 4.1, the soil conditions in the three main production districts were relatively uniform.

Cotton cultivation throughout Shandong, especially into the Jiaodong region, expanded with the introduction of improved cotton varieties from the United States, typically referred to as “foreign cotton” (Yangmian 洋棉) or “American cotton”

(Meimian 美棉) in contrast to the native or “Chinese cotton” (Zhongmian 中棉) varieties.230 The new varieties from the United States had three primary advantages: higher yields, producing on average 50 more jin 斤 (or catties; one jin equaled 1.3158 pounds) per mu than local varieties; more valuable, with 100 jin of cotton selling for seven more yuan than local varieties; and more versatile to various soil conditions, allowing U.S. varieties to be grown in other regions of Shandong, including the eastern part of Shandong.231

228 Wu Zhi 吳知, “Shandong sheng mianhua zhi shengchan yu yunxiao” 山東省棉花之生產與運銷 (Cotton Production, Transport, and Marketing in Shandong Province), Zhengzhi jingji xuebao 政治經濟學報 (Journal of Political Economy) 1, no. 5 (1936): 8-9; see also Lin Xiuzhu 林修竹 (Lin Maoquan 林茂泉), Maoquan shiye wenji 茂泉實業文集 (Maoquan’s Collected Works on Industry) (1926), in Jindai Zhongguo shiliao yekan,近代中國史料業刊, 3 bian 編, 47 juan 卷, ed. Shen Yunlong 沈雲龍 (Taipei: Wenhai chubanshe 文海出版社, 1988), 79-80. 229 For detailed descriptions of Shandong’s soil and climate conditions, see James Thorp, Geography of the Soils of China (Beijing: National Geological Survey of China, 1939), 168-175, 216-235; James Thorp and T. Y. Tschau, “Notes on Shantung: A Reconnaissance Soil Survey of Shantung,” Turang zhuanbao 土壤專報 (Soil Bulletin), no. 14 (1936): 1-105; and F.H. King, Farmers of Forty Centuries (Madison, Wisconsin: Democrat Printing Company, 1911), 216-260. 230Locals called the foreign varieties dahua 大花 and local varieties as xiaohua 小花. See Wu Zhi, 16-19. 231 Lin Xiuzhu, 81. Wu Zhi 吳知 from the Nankai Institute of Economics, similarly commented in 1936 that U.S. varieties were more adaptable than native varieties to Shandong’s varying soil conditions, especially to sand and silt loam found in Jiaodong. For a more technical discussion of Shandong soils and cotton production, see Wu Zhi, 10-11. 81

Figure 4.1 Generalized Soil Map of Shandong

Source: “Generalized Soil Map of Shantung Province, China,” in James Thorp and T.Y. Tschau, “Notes on Shantung: A Reconnaissance Soil Survey of Shantung” Turang zhuanbao, no. 14 (1936).232

Before the 1920s, however, U.S. varieties of cotton were diffused with limited success in Shandong by missionaries, cotton mills in Qingdao, and provincial and county governments. This process of diffusion was limited by warlordism and banditry. Ren

Dekuan 任德寬 (dates unknown), former director of the Linqing Cotton Station, further pointed out in 1924 that the U.S. cotton struggled to diffuse because production costs

232 For discussion of soil surveys and the creation of soil maps including Figure 4.1 during the Republican era, see Shen, Unearthing the Nation, 120-123. 82

were high and the price of raw cotton was relatively low internationally.233 Interest in promoting U.S. varieties of cotton increased in the late 1910s and 1920s, especially through the Shandong Office of Encouraging Industry (OEI; Quanyesuo 勸業所), as has been documented in great detail by Kenneth Pomeranz.234 Lin Xiuzhu 林修竹 (1884-

1948; also known as Lin Maoquan 林茂泉), industrial educator and director of the

Shandong Office of Industry (Shiyeting 實業廳) in the 1920s, commented during this time that the promotion of cotton was especially successful after the establishment of OEI offices in various counties: “the policy to encouraging industry 勸業 takes cotton cultivation as one of its central affairs. Consequently, the expansion of cotton fields and the switch to grow American cotton 美棉 has been actively carried out.”235 According to

Lin, this work was done by purchasing U.S. seed varieties by the provincial Office of

Industry and providing the seeds at no cost to each county OEI to be distributed to farmers along with instructions.236

In addition to efforts by OEI, cotton experimentation and extension was promoted by the provincial government from the late 1910s to the 1930s with the establishment of the Linqing cotton experiment station in 1918 and the Qidong cotton experiment station in 1926. These two stations were established for the purpose of improving and diffusing native and foreign varieties. After the establishment of the Qidong station, for example, researchers began to conduct selective breeding of both native and foreign varieties.

233 See Ren Dekuan 任德寬, “Shandong zhi meimian tuiguang wenti” 山东之美棉推广问题 (Problems with Extending U.S. Cotton in Shandong), Nongxue 农学 (Agronomy) 1, no. 5 (1924): 17-19. 234 Pomeranz, The Making of a Hinterland, chap. 2. 235 Lin Xiuzhu, 79. 236 Ibid., 80. 83

Reports from the station in the 1930s show that they experimented primarily with the foreign varieties Trice (Chinese: tuolisi 脫里司 or tuozimian 脫字棉) and Kings’

Improved (Chinese: jinzimian 金字棉), along with several native varieties from Qidong,

Zhengding 正定 (Hebei), and the much praised Million Dollar237 cotton variety (baiwan mian 百萬棉).238 Beginning in 1927, the station collected various cotton varieties from

Qidong, and after five years of seed selection, produced a strain that was able to yield 130 jin of unginned cotton per mu. Further, in 1929, the station began to conduct research on

Trice. After five years of selective breeding, the station bred Trice no. 36 (tuolisi di’sanshiliu hao 脫里司第三十六號), which was capable of yielding 140-150 jin of unginned cotton per mu.239 The length of fiber of Trice no. 36 ranged from 0.83 inches

(21 mm) to 1.25 inches (32 mm) with an average length of about 1 inch (24 mm).

Compared to other native and improved foreign varieties, Trice no. 36 was one of the most superior breeds in China at the time, even surpassing the average length of Million

Dollar cotton at 0.87 inches (22 mm).240

237 This Million Dollar cotton variety was discovered by foreign cotton expert J. B. Griffing in 1919 near Shanghai. He was so impressed with the variety that he stated that this discovery “bids fair to modify the popular belief that none of the Chinese cotton can compare favorably with American Upland.” See J. B. Griffing, “A New Variety of Cotton From China,” Journal of Heredity 18, no. 11 (1927): 496-497. 238 See Shandong shengli di’er mianye shiyanchang ed., Shandong shengli di’er mianye shiyanchang chengji baogao 山東省立第二棉業試驗場成績報告 (Shandong Provincial Second Cotton Experiment Station Achievement Report), no. 2 (1929); and Shandong shengli di’er mianye shiyanchang ed., Shandong shengli di’er mianye shiyanchang shiyan baogao 山東省立第二棉業試驗場試驗報告 (Shandong Provincial Second Cotton Experiment Station Experiment Report), no. 3 (1930), no. 4 (1931), and no. 5 (1932); see also Wu Zhi, 16-19. 239 One source claimed that average output was 150 jin per mu. See “Shandong shengli di’er mianye shiyanchang: yucheng youliang mianzhong 山東省立第二棉業試驗場: 育成優良棉種 (Shandong Provincial Second Cotton Experiment Station: Breeding Superior Cotton), Nongye zhoubao 農業週報 (The Farmers Weekly), no. 59 (1930): 298; Wu Zhi, conversely, stated that the average was 140 jin per mu. See Wu Zhi, 19. 240See table in Wu Zhi, 19. 84

The ultimate goal of such experimentation was extension. A report published in

1932 by the Qidong cotton station recorded that extension work included distributing improved seeds (especially Trice and Trice no. 36), publishing extension leaflets

(qianshuo) for use in village agricultural schools, training farmers on cotton cultivation and cotton ginning techniques, and assisting in cooperative selling initiatives. According to the report, “this station’s purpose of cotton extension is to increase cotton production and improve the lives of cotton farmers.”241

Such efforts by the state to promote cotton research and extension in Shandong certainly contributed to the increase of cotton output during the Republican era. Despite drought, famine, and war which frequently hindered production, provincial statistics of cotton production found in Table 4.2 show that total output generally increased from

1919 to 1937. According to the CCMA series, cotton yields increased from an average of a little under 0.9 million dan 擔 (or picul; one dan equals 131.58 pounds or roughly 60 kilograms) from 1921 to 1923 to over 1.2 million dan from 1934-1937. If the poor harvests from 1935 are excluded, the average of the later years shows an even more dramatic increase, reaching over 1.5 million dan. Overall production in Shandong also rose relative to other provinces surveyed in the CCMA series. Where Shandong roughly produced 5 percent of total output during the first three years of the survey (1919-1921), output relative to other provinces peaked during the early 1930s, reaching over 20 percent for three years in a row from 1930 to 1932 – even reaching 33.7 percent of total national production in 1931 – before dropping to roughly 13.2 percent of total output from 1934 to

241 Shandong shengli di’er mianye shiyanchang ed., Shandong shengli di’er mianye shiyanchang tuiguang baogao, 1. 85

1937 (excluding 1935). Additionally, the increase of foreign varieties grown in

Shandong, which was almost exclusively the U.S. varieties Trice and Trice no. 36, is especially apparent. Some of the first statistics on foreign varieties from the early 1920s show that on average from 1921-1923, 1.8 percent of output was of these foreign varieties. By the years 1936 and 1937, foreign varieties had increased to over 60 percent of total output (see Table 3.2).

Table 4.2 Annual Yields of Foreign and Native Cotton Varieties in Shandong Foreign Varieties 洋棉 Native Varieties 中棉 Year Total (dan) Total (dan) % of Total Total (dan) % of Total 1919a - - - - 894,558 1920a - - - - 126,070 1921 11,768 3.9% 292,309 96.1% 304,077 1922 6,565 0.7% 998,665 99.3% 1,005,230 1923 12,704 0.9% 1,374,962 99.1% 1,387,666 1924a - - - - 937,224 1925a - - - - 995,603 1926 164,938 31.8% 353,341 68.2% 518,279 1927 214,332 30.2% 495,423 69.8% 709,755 1928 257,647 41.5% 362,766 58.5% 620,413 1929 478,281 39.4% 734,799 60.6% 1,213,080 1930 577,172 26.6% 1,593,486 73.4% 2,170,658 1931 1,112,484 51.6% 1,042,398 48.4% 2,154,882 1932 710,105 40.1% 1,059,280 59.9% 1,769,394 1933 677,655 46.1% 791,277 53.9% 1,468,932 1934 620,788 46.5% 713,265 53.5% 1,334,053 1935 191,070 46.9% 216,145 53.1% 407,215 1936 1,085,215 60.6% 705,012 39.4% 1,790,227 1937 1,124,905 69.0% 505,452 31.0% 1,630,357

Year (Average) 1921-1923 10345.7 1.2% 888645.3 98.8% 898991.0 1926-1929 278799.5 36.4% 486582.3 63.6% 765381.8 1930-1933 769354.0 40.7% 1121610.3 59.3% 1890966.5 1934-1937 755494.5 58.5% 534968.5 41.5% 1290463.0 1934-1937 943636.0 59.5% 641243.0 40.5% 1584879.0 (excluding 1935) Sources: Data from Huashang shachang lianhehui, Mianchan diaocha baogao no. 8 (1920), no. 9 (1921), no. 10 (1921), no. 11 (1922); and Zhonghua mianye tongjihui, Zhongguo mianchan tongji no. 12 (1923), no. 17 (1928), no. 18 (1929), no. 19 (1930), no. 20 (1931), no. 21 (1932), no. 22 (1933), no. 23 (1934), no. 24 (1935), no. 25 (1938). a Statistics for these years did not differentiate between foreign and native varieties. 86

4.3 Diffusion of U.S. Cotton Varieties in Shandong

Although annual yields of cotton increased overall during the 1920s and 1930s, especially the U.S. varieties, this increase was spatially uneven. By looking at 1934 alone, the west Shandong region produced 61.3 percent of all total cotton and 76.6 percent of U.S. varieties, and the north Shandong region yielded 27.9 percent of total cotton and 23.2 percent of U.S. varieties. The south Shandong region and the Jiaodong region fared worse. The south region produced 9.7 percent of total output and a mere 0.2 percent of U.S. varieties while the Jiaodong region produced only 1.1 percent of total output and less than 0.1 percent of U.S varieties.242

In explaining this divergence, Kenneth Pomeranz has argued that improved varieties were adopted in the north and west region due to village “permeability” and rejected in the south region in part due to village solidarity, where village elites rejected agricultural innovations to protect political power. Therefore, although the south region tended to be more prosperous than the north and west historically, due to greater soil fertility and lower susceptibility to natural disasters, the close-knit communities of the early twentieth century restricted access to agricultural innovations that helped promote the rural economy in the north and west cotton regions.243 Moreover, as discussed in

Chapter Three, the locations of these stations were in Shandong’s “Heartland” capital market region (Figure 3.2), and reflect an emphasis by the government to promote cotton improvement catering to urban and treaty-port factories.

242 Zhongguo mianchan tongji no. 23 (1934); statistics also reproduced in Wu Zhi, 14. 243 See Pomeranz, The Making of a Hinterland, chap 2; See also discussion in Joseph Esherick, The Origins of the Boxer Uprising (Berkeley: University of California Press, 1987), 27-28. 87

A closer look at county-level data, however, reveals that adoption and cultivation of U.S. cotton varieties was more complex than Pomeranz’s northwest-southwest divergence, as yields were closely related to geographic proximity to the cotton experiment stations. Figure 4.2 and Figure 4.3 illustrate both the diffusion process of U.S. varieties overtime and the limits of diffusion. As these maps illustrate, not all counties in the north and west cotton production regions yielded high amounts of U.S. cotton.

Rather, output tended to be higher in counties adjacent to or near Linqing and Qidong.

Figure 4.3 especially illustrates how output of U.S. cotton, while rising in many counties, was negligible in relation to a few select counties near these experiment stations and extension sites; most counties in both the northwest and especially the southwest region consistently yielded less than 1 percent of total output. Counties near the Linqing and

Qidong cotton stations consistently yielded high amounts of cotton. Linqing alone on average yielded 14.1 percent of all U.S. cotton output in Shandong from 1921 to 1937.244

Moreover, the highest yielding counties for each three to four year averages found on

Figure 4.2 and Figure 4.3 were consistently in the west cotton production region near

Linqing: from 1921 to 1923, Qingping 清平 yielded 15.9 percent of provincial total; from

1926 to 1929 Gaotang 高唐 yielded 20.7 percent of the total; from 1930 to 1933 Linqing yielded 15.8 percent of the total; and from 1934 to 1937 Xiajin 夏津 yielded 14.1 percent of total.

244 More specifically, the percentage of total output for Linqing was 12.6 percent (1921-1923), 18.8 percent (1926-1929), 15.8 percent (1930-1933), and 9 percent (1934-1937). 88

Figure 4.2 Average Yield of U.S. Varieties in Shandong by County

Sources: Data from Huashang shachang lianhehui, Mianchan diaocha baogao no. 8 (1920), no. 9 (1921), no. 10 (1921), no. 11 (1922); and Zhonghua mianye tongjihui, Zhongguo mianchan tongji no. 12 (1923), no. 17 (1928), no. 18 (1929), no. 19 (1930), no. 20 (1931), no. 21 (1932), no. 22 (1933), no. 23 (1934), no. 24 (1935), no. 25 (1938).

Figure 4.3 Average Yield of U.S. Varieties in Shandong by County (% of total output)

A closer look at the work conducted by these provincial cotton stations and supporting institutions further explains the correlation between proximity to stations and adoption, as the extension system was geographically and infrastructurally constrained.

Improved varieties such as Trice were provided to farmers through an extended transportation process from overseas to provincial experiment stations, then on to county extension or experiment stations, and finally to individual farmers.245 Farmers, therefore, were highly reliant on the provincial and county experiment stations to provide improved seeds, the latter of which appears to have been especially unreliable. According to an extension report published by the Qidong cotton station in 1932, the provincial Office of

Industry distributed 2,956 jin of Trice to 34 counties in Shandong, consisting of roughly

60 jin per county. The majority of the counties were located in the north region (15 counties), next the east Shandong or Jiaodong region (9 counties), third the south region

(7 counties) and last the west region (3 counties).246 The cotton seeds were distributed to each county’s Bureau of Construction (Jiansheju 建設局), and due to the relatively low amount of overall seeds distributed, these seeds were likely meant to be sent directly to county agricultural experiment stations, typically under the direction of the Department of Construction. By 1922, 50 county stations were in operation and by 1934 it had increased to at least 54 stations. The provincial government encouraged these county stations to experiment and adapt cotton varieties to meet local soil and climate conditions, as problems were found with seeds bred in one county being less effective in others.247

245 See, for example, discussion in Pomeranz, The Making of a Hinterland, 79-80. 246 See Shandong shengli di’er mianye shiyanchang ed., Shandong shengli di’er mianye shiyanchang tuiguang baogao, 2-4. 247 Pomeranz, The Making of a Hinterland, 97, fn. 127. 91

How successful these county stations were, however, is highly questionable. For one, funding at these stations varied widely between counties, ranging from a couple thousand yuan at some to under a hundred yuan at others, according to a 1934 survey; on average county experiment stations had annual expenditures of around 300 to 400 yuan.

In contrast, provincial cotton stations were better supported financially and with better trained personnel. According to the same survey, the Linqing cotton station had average annual expenditures of 12,684 yuan, with one technician, two assistant technicians, one extension agent, two office workers, two trainees and one secretary. The Qidong cotton station was relatively similar, consisting of 14,980 yuan of annual expenditures with one technician, two assistant technicians, two extension agents, two office workers, four trainees, and one secretary.248 With such financial and resource differences between the provincial and county stations, counties near the provincial cotton stations tended to adopt and produce U.S. cotton at a much higher rate than others, as they had access to these additional resources.

Moreover, higher rates of adoption not only correlate with fewer geographic, sociopolitical, and financial barriers, but also to provincial-level initiatives to explicitly promote cotton in counties bordering the experiment stations. A 1932 extension report by the Qidong cotton station recorded that the station established three Extension

Experiment Districts (Tuiguang shiyanqu 推廣實驗區), located in the southeastern part of Qidong county, the northern part of Zouping, and the western region of Gaoyuan 高苑; all three locations were within close proximity to the Qidong station. Cotton was

248 See Li Zhiji, “Guonei nongshi shiyan jiguan gaikuang (er),” 7-12. 92

disseminated in these districts to farmers who had previously registered with the station for a specific amount of seeds. According to the report, 120 households were given seeds throughout 39 villages in Qidong (29 villages), Zouping (9 villages), and Gaoyuan (1 village), totaling 9,395 jin of Trice; this total was over triple the amount of seeds provided to 34 counties by the Office of Industry in 1932 (see above).249

Further, in 1933, the Shandong Office of Construction (Jiansheting 建設廳) promoted a new plan to establish a series of Cotton Demonstration Districts (Mianzuo shifanqu 棉作示範區), Cotton Local Experiment Districts (Mianzuo difang shiyanqu 棉

作地方試驗區), and Cotton Extension Center Districts (Mianzuo tuiguang zhongxinqu

棉作推廣中心區), with the purpose of promoting Trice. The majority of these districts were all located in the north and west regions of Shandong, specifically eleven of the eighteen demonstration districts, five of the eight local experiment districts, and all of the extension centers. The cotton extension centers were especially explicit in emphasizing the promotion of cotton in counties near the cotton stations, stating that “cotton- producing counties adjacent to the first [Linqing] and second [Qidong] cotton stations are selected as cotton extension center districts, to propagate excellent cotton seeds [i.e. Trice and Trice no. 36] to the farmers in each of these counties.”250 Further, counties such as

Qidong were frequently used as experimental counties for initiatives that would only later

249 Shandong shengli di’er mianye shiyanchang ed., Shandong shengli di’er mianye shiyanchang tuiguang baogao, 4-11. 250The plan further called for each of these districts and counties to conduct work under the direction of the provincial cotton stations. Seeds were to be distributed by the provincial cotton stations, and all seeds produced in the districts were to be collected in conjunction with the provincial cotton stations and the county government. “Lu jianshe ting tuiguang meimian” 魯建設廳推廣美棉 (Shandong Office of Construction Promote U.S. Cotton), Nongcun fuxing weiyuanhui huibao 1, no. 7 (1933): 184-185. 93

be extended to other regions. In 1934, for instance, Qidong county was chosen as an experiment district where improved cotton seeds were distributed to each cotton farmer, cooperatives were organized, and the Shanghai Bank provided loans to poor farmers, all to increase production.251

Such efforts by the provincial government, in conjunction with geographic, infrastructural, and sociopolitical constraints, combined to confine the diffusion of U.S. varieties to counties near the provincial cotton experiment stations. A look at Table 4.3 of the sixteen adjacent counties and near-adjacent counties that were targeted in cotton extension initiatives (mentioned above) shows that these counties consistently yielded over 70 percent of total output of U.S. varieties from 1926 to 1936.

Table 4.3 Output of U.S. Varieties of Cotton in Adjacent and Near-Adjacent Countiesa Output of U.S. Years % of total in Shandongb varieties (dan) 1926-1929 207,831 74.5% 1930-1933 540,246 70.2% 1934-1937 534,799 70.8% Sources: Data from Huashang shachang lianhehui, Mianchan diaocha baogao no. 8 (1920), no. 9 (1921), no. 10 (1921), no. 11 (1922); and Zhonghua mianye tongjihui, Zhongguo mianchan tongji no. 12 (1923), no. 17 (1928), no. 18 (1929), no. 19 (1930), no. 20 (1931), no. 21 (1932), no. 22 (1933), no. 23 (1934), no. 24 (1935), no. 25 (1938). a Counties included in this total are Qidong, Zouping, Zhangqiu 章丘, Gaoyuan, Putai 蒲台, Binxian, Enxian 恩縣, Linqing, Xiajin, Qingping, Tangyi 堂邑, Guantao 館陶, Qiuxian 邱縣, Jiyang 濟陽, Huimin, and Qingcheng 青城. b The high percentage of total output by these counties was not simply due to them comprising a significant portion of the surveyed counties. From 1926 to 1929, only 8 of the 16 counties above were surveyed for U.S. varieties, making up only 25 percent of the 32 counties included in the average for this period. Moreover, from 1930 to 1933, these 16 counties made up 39 percent of the 41 surveyed counties and from 1934 to 1936 only 18 percent of the 87 counties from which data was collected.

251“Lu jiansheting fazhan mianye jihua” 魯建設廳發展棉業計劃 (Shandong Office of Construction Plan to Develop the Cotton Industry), Nongcun fuxing weiyuanhui huibao 2, no. 5 (1934): 109-110. 94

4.4 Cotton Extension and Spatial Inequality

In Philip Huang’s 1985 study of the North China peasant economy, he argued in part that “the spread of cotton cultivation [during the Republican era] led to the social stratification of the small peasantry.” Moreover, “cotton brought far greater returns than food grains, thereby providing the opportunity for some peasants to profit from farming and to move up the socioeconomic ladder.”252 These conclusions are supported by his case study of Michang village, located in northeastern Hebei. Based on surveys of land sold from 1890 to 1936 within the village, Huang pointed out that the majority of the land was bought in the latter 20 years, after cotton was introduced into the village.

Additionally, the majority of the land was bought by managerial farmers and other well- to-do middle and rich farming households, who had profited on cotton cultivation and commercialized agriculture. Conversely, poor downward-moving peasants were forced to sell, becoming tenants and laborers.253 Huang concluded:

the spread of cotton cultivation, with its higher returns and higher risks, unavoidably encouraged social differentiation. A few among the petty proprietors could take advantage of the increased returns to move upward into rich peasant and managerial farming; others slid down the socioeconomic ladder to become tenants, day-laborers, and landless year-laborers.254

In addition to Huang’s conclusions of social differentiation, the findings in this present study points to growing spatial differentiation at the county-level based on proximity to cotton experiment stations. With cotton being one of the most important economic crops during the Republican era, individual farm economies were affected by

252 Huang, The Peasant Economy and Social Change in North China, 106. 253 Ibid., 108-109. 254 Ibid., 111. 95

the cultivation of new improved cotton varieties. According to a longitudinal study of the costs and earnings for several food crops (wheat, soybean, sorghum, and millet) and cotton in Qidong county from 1928 to 1932, U.S. varieties of cotton were consistently the most profitable crop grow. Table 4.4 presents the five year average expenses and income for these crops from 1928 to 1932. During this time, U.S. cotton varieties averaged 7.39 yuan per mu of net profit, while native varieties of cotton averaged 6.28 yuan per mu.

Food crops proved to be far less profitable, with soybeans averaging 3.53 yuan, millet

2.80 yuan, sorghum 1.84 yuan, and wheat 0.72 yuan. The average earnings of cotton in

Qidong relative to other crops is perhaps high in comparison to other counties in North

China. For example, according to a 1937 survey in Michang, Hebei, cotton brought in a net income over two times that of sorghum.255 Although the statistics in Table 3.4 showing that cotton had a net income four times that of sorghum might be high in relation to the rest of Shandong and North China, it nevertheless clearly illustrates how the adoption of U.S. cotton varieties had an immense impact on a farmer’s income.

255 Huang, The Peasant Economy and Social Change in North China, 107. 96

Table 4.4 Average Expenses and Income for Crops in Qidong, 1928-1932 (Unit: Yuan) U.S. Native Wheat Soybean Sorghum Millet Cotton Cotton Expenses per mu 8.28 8.32 5.40 2.11 7.05 7.73 Seeds 0.19 0.18 0.39 0.26 0.07 0.05 Fertilizer 2.83 2.83 2.93 0.00 2.65 3.07 Human Labor 3.86 3.93 1.12 1.15 2.85 3.06 Animal Labor 0.47 0.46 0.50 0.24 0.55 0.62 Taxes 0.93 0.93 0.47 0.46 0.93 0.93 Gross Income per mu 15.67 14.61 6.13 5.64 8.89 10.53 Primary Products 14.63 13.89 5.31 4.78 7.34 8.84 Secondary Products 1.04 0.72 0.82 0.86 1.55 1.70 Net Income per mu 7.39 6.28 0.72 3.53 1.84 2.80 Source: Shandong shengli di’er mianye shiyanchang ed., Shandong shengli di’er mianye shiyanchang tuiguang baogao, no. 1 (1932), 55-64; see also Zhongguo shiye zhi: Shandong sheng, 183-186.

By applying these findings of income per mu to the county level, a general conclusion can be made that counties which cultivated and produced more cotton as a whole profited relative to low-adopting counties. This does not mean that all farmers within high-adopting counties benefited from the introduction of these new varieties, as emphasized by Huang above. Rather, the unequal adoption of U.S. cotton varieties across counties illustrates how economic growth in Shandong during the Republican era was spatially uneven when viewed from the county level. The cotton research and extension system, which favored regions within close geographic proximity to the cotton stations, resulted not only in the uneven diffusion of improved varieties of cotton, but also spatially differentiated economic growth.

3.5 Summary

This chapter began by asking the question: to what degree were local provincial governments effective in stimulating the economy through research and extension? The

research and extension of U.S. varieties of cotton in Shandong has been taken as a case study. Quantitative statistics collected by the Chinese Cotton Millowners Association show that output of U.S. varieties increased dramatically from the early 1920s to 1937.

By looking at the county level, however, this increase in output was spatially uneven, as counties within close proximity to the Linqing and Qidong cotton stations consistently had the highest yields of U.S. varieties. This limited impact was caused primarily by the nature of the extension work which targeted counties near the cotton stations. A look at this diffusion process, therefore, illustrates how agricultural extension work in general is conducive to promoting spatially uneven growth that favors regions within close geographic proximity to extension centers. Since the adoption of improved varieties of cotton translated into increased income for the farmer, these findings also illustrate how state-led agricultural research and innovations in Shandong led to uneven and spatially differentiated economic growth.

Chapter 5

Conclusions

Out of extension education the shape of things to come is emerging. All nations will, through the more systematic application of science, realize ever-increasing agricultural production from more fertile soil, ensure happier families in more comfortable homes, greater participation in community life and greater interest and understanding in local, national, and international problems. Science, wherever it has gone around the world, is responsible for progress in agriculture and rural welfare. Bringing that science to more and more rural people is the job of extension. -Ensminger and Sanders, 1945256

Modern agricultural extension services have had an immense impact on agricultural economies worldwide since it was first developed during the early twentieth century. First institutionalized in the United States and parts of Europe, modern extension services spread globally throughout the early twentieth century to bring science, the impetus for economic, social, and cultural progress, to the average farmer. It was within this context that modern agricultural extension services and experiment institutions were introduced into China during the early twentieth century. Proponents within China and abroad saw extension and scientific experimentation as a means to national regeneration and a more prosperous China. As such, extension services were promoted by the state to

256 Douglas Ensminger and Irwin T. Sanders, “What Extension Is,” in Farmers of the World, ed. Edmund deS. Brunner, Irwin T. Sanders, and Douglas Ensminger (New York: Columbia University Press, 1945), 7. 99

“save the people” (jiumin 救民) through improved economic conditions and standards of living.257

Previous scholarship on agricultural experimentation and extension in Republican

China has focused largely on the weaknesses of extension services, whether it be lack of funding or inadequate personnel. Ultimately, scholarship has pointed out that the extension system failed to introduce sufficient agricultural innovations to “save the people” and transform the rural economy. A discussion of these failures, however, is but one part of the story. Equally important are the transformations that occurred despite or because of such “failures.”

The purpose of this present study, therefore, was to explore both the institutional legacies and economic impact of modern agricultural extension in Republican China.

Systematic agricultural experimentation and dissemination of agricultural information in

China dates back to imperial times, when the state played an active role to “encourage agriculture” and transfer agricultural technologies. The active role of the state in developing agriculture continued into the late-Qing and Republican eras, as agricultural experimentation and extension became institutionalized and centralized under the direction of the central government. Agricultural schools and experiment stations were established in the final years of the nineteenth century and throughout the early years of the twentieth century. They paved the way for an institutionalized form of extension established by the Nationalist Party beginning in 1929.

257 See Yu Gong, 1; and Hsin-Pao Yang, 59. 100

Institutionalized agricultural experimentation and extension in the Republican era further laid a foundation for later development in the People’s Republic of China.

Following 1949, the PRC built upon and expanded many existing institutions to form its current agricultural research network and extension system. Tracing the history of agricultural experimentation in Shandong illustrates the institutional legacies of agricultural experiment stations and schools established in the late-Qing and Republican eras. For example, the Shandong Agricultural Experiment Station, first established in

Jinan in 1903, underwent a number of changes during the Qing dynasty, the Republic of

China, the Second Sino-Japanese War, and finally the People’s Republic of China. The

Shandong Agricultural Experiment Station has developed into the well-supported and impactful Shandong Academy of Agricultural Sciences, which continues to be an important research institution in developing agricultural and environmental technologies.

The rapid expansion of agricultural research and extension in the PRC was made possible by a pre-existing network of agricultural experiment stations, universities, and a centralized extension system developed during the Republican era.

The development and spread of new agricultural technologies through the extension system in Republican China was aimed at promoting the economy and improving the well-being of the individual farmer. Despite high expectations for extension during this time, it had a number of weaknesses, including a narrow scope of extension work, lack of adequate personnel, and a system that was both unorganized and heavily influenced by politics. The link between agricultural extension and the state is a common theme throughout twentieth century China and much of the world. The state’s

101

active role in research and extension made agricultural extension work especially prone to the various “political trends” of the times as extension was employed as a tool of statecraft to achieve the political and economic needs of the state. Experimentation and extension work, therefore, favored economic crops needed to promote provincial and national exports, such as cotton, silk, and tobacco. This emphasis on strategic regions, coupled with the limited geographic impact of agricultural extension services, meant that innovations introduced through the extension system were limited to strategic regions within close proximity to experiment stations and extension centers. The diffusion of improved cotton varieties into Shandong, as found in Chapter Four, illustrates the spatial disparities of adoption at the county level, as counties near the cotton experiment stations had greater access to improved seeds and thereby adopted at a much higher rate than counties farther away.

This study of agricultural extension and experimentation in Shandong during the

Republican era has important implications for understanding economic growth during this time. State-led extension initiatives had clear limitations, as the spread of new technologies and agricultural innovations through this system were constrained by practical restraints, such as lack of personnel and shortage of funding. Moreover, in conjunction with recent economic policy in the PRC to let some areas and farmers “get rich first,” extension services in Shandong focused on key economic markets and regions.

For example, the cultivation of improved and more profitable cotton varieties was promoted through experimentation and extension services, with the benefits of these services limited to counties within close proximity to the cotton experiment stations. The

102

economic growth that these services spurred, therefore, was geographically concentrated around the extension sites. In the context of Republican China, such findings illustrate that rather than reducing our discussion to a debate over decline or growth, the

Republican economy must be understood it the context of geographical access to resources, whether provided by the state or the market.

In addition to economic growth, the process of innovation diffusion and technology transfer also has a larger spatial and political context. Adoption of a new innovation or technology is rarely uniform across geographic space, as regions within closer geographic proximity to centers of diffusion are favored. Moreover, state-led initiatives to diffuse innovations, such as agricultural extension services, are closely tied to the needs of the state, whether it be export goals or food security. This political context explains why the needs of the individual farming family is frequently secondary to the perceived needs of the state in developmental projects worldwide, both historically and today.

In many ways, the findings in this study are not unique to Republican China, and can be found in the more-recent history of agricultural extension throughout much of the developing world. For example, problems with mobility (lack of ability to reach farmers effectively), organization, and inadequately trained personnel have continued to plague agricultural extension work, especially throughout much of Asia.258 In some places, the inadequacies of previous models of extension have led to the reassessment and

258 See V.A. Sigman and B.E. Swanson, “Utilization of Technology: The Corner-Stone of Agricultural Development Policy and Programs,” in Agricultural Extension: A Reference Manual, 227-231. 103

development of more participatory approaches to extension. In others, such as the United

States, Australia, and parts of Europe, agricultural extension has also fallen into “a period of crisis.”259

While systematic agricultural extension in many parts of the world is undergoing a period of crisis, it continues to flourish and grow in the People’s Republic of China.

The rapid expansion of agricultural research and extension, and continued support by top leaders in the PRC for institutions such as the Shandong Academy of Agricultural

Sciences, illustrates a strong state commitment to overcome hunger and feed the people.

As an important legitimizing principle of the Chinese Communist Party, this commitment represents an intensification of imperial efforts to “encourage agriculture,” and an expansion of state-led institutional agricultural extension as established during the

Republican era.

259 Vanclay and Lawrence, 110. 104

Bibliography

Chinese-Language Sources

Bao Ping 包平. Ershi shiji zhongguo nongye jiaoyu bianqian yanjiu 二十世纪中国农业教育变迁研究 (Study of the Changes in China’s Agricultural Education in the Twentieth Century). Beijing: Zhongguo sanxia chubanshe 中国三峡出版社, 2007.

Chen Qiwen 陈启文. “Huiyi Shandong jiefangqu de nongye keyan gongzuo” 回忆山东解放区的农业科研工作 (Recollections of Agricultural Science Work in Shandong’s Liberated Zones). May 1984. In Shandong sheng nongye kexueyuan yuanshi, 1946-1992 山东省农业科学院院史 (Shandong Academy of Agricultural Science Academy History, 1946-1992), 185-189. Jinan: Jinan shi jianhua yinshuachang 济南市建华印刷厂, 1994.

“Dongya nongye yanjiu de diantang: Huabei nongshi shiyanchang” 東亞農業研究的殿堂: 華北農事試驗場 (East Asia’s Agricultural Research Palace: The North China Agricultural Experiment Station). Xin nongye 新農業 (New Agriculture) 1, no. 1 (1945): 22-23.

Guan Yida 管義達, Lu Feizhi 陸費執 and Xu Zhen 許振. Nongye tuiguang 農業推廣 (Agricultural Extension). Shanghai: Zhonghua shuju 中華書局, 1935.

Guo Shengfu 郭圣福. “Nongye ‘bazi xianfa’ pingxi” 农业八字宪法评析 (Analysis of the Agriculture Eight-Word Charter). Dangshi yanjiu yu jiaoyu 党史研究与教育 (Party History Research and Education), no. 6 (2008): 34-39.

Hu Ruifa 胡瑞法 and Li Liqiu 李立秋. “Nongye jishu tuiguang de guoji bijiao” 农业技术推广的国际比较 (An International Comparison of Agricultural Technology Extension). Keji daobao 科技导报 (Science and Technology Review), no. 1 (2004): 26-29.

105

Huabei nongshi shiyan chang 華北農事試驗場 (North China Agricultural Experiment Station). Huabei nongshi shiyan chang yaolan 華北農事試驗場要覽 (Overview of the North China Agricultural Experiment Station). 1943.

Huashang shachang lianhehui 華商紗廠聯合會 (Chinese Cotton Millowners Association). Mianchan diaocha baogao 棉產調查報告 (Chinese Cotton Production Survey Report), 1920-1922.

Jiang Dixian 蔣迪先. “Zhongguo mianchan tongji zhi guoqu ji jianglai” 中國棉產統計之過去及將來 (The Past and Future of Cotton Production Statistics). Zhongguo mianchan tongji 中國棉產統計 (Chinese Cotton Production Statistics), no. 18 (1929): 1-42.

Jiaoyubu 教育部 (Ministry of Education). Zhonghua Minguo jiaoyu zongzhi jiqi shishi fangzhen 中華民國教育宗旨及其實施方針 (Republic of China education objectives and implementation guidelines). April 26, 1929. In Jiangsu sheng zhengfu gongbao 江蘇省政府公報 (Jiangsu Provincial Government Report), no. 139 (1929), tebie yaojian 特別要件 (Important documents): 1-2.

Jiaoyubu, Neizengbu 內政部 (Ministry of Internal Affairs), and Nongkuangbu 農礦部 (Ministry of Agriculture and Mines). Nongye tuiguang guicheng 農業推廣規程 (Agricultural Extension Regulations). June 14, 1929. In Nongye tuiguang 農業推廣 (Agricultural Extension), no. 1 (1930): 2-7.

Jiaoyubu, Shiyebu 實業部 (Ministry of Industry), and Neizhengbu. Nongye tuiguang guicheng. March 29, 1933. In Nongye tuiguang, no. 4 (1933): 1-8.

Li Mengzhu 李梦竹, Gao Yuhuan 郜玉环, and Wang Qian 王茜. “Minguo qianqi Shandong guanfang nongye kexue yanjiu de chutan” 民国前期山东官方农业科学研究的初探 (Preliminary Study of Shandong Official Agricultural Science Research in the Early Republican Era). Chuangxin jiaoliu 创新交流 16 (2013): 70-73.

Li Wenzhi 李文治 ed. Zhongguo jindai nongye shi ziliao 中國近代農業史資料 (Sources on the History of Modern Chinese Agriculture). Vol. 1. Beijing: Sanlian shudian, 1957.

Li Xintian 李心田. “Shandong Meizhong yancao diaocha baogao” 山東美種煙草調查報告 (Shandong United States’ Varieties of Tobacco Survey Report). Guoji maoyi qingbao 國際貿易情報 (International Trade Information) 2, no. 9 (1937): 1-3.

106

——— . “Shandong Meizhong yancao diaocha baogao.” Guoji maoyi qingbao 2, no. 10 (1937): 25-30.

Li Ying 李瑛. Minguo shiqi daxue nongye tuiguang yanjiu 民国时期大学农业推广研究 (Study on the Agricultural Extension of Universities in China). Hefei: Hefei gongye daxue chubanshe 合肥工业大学出版社, 2012.

Li Zhiji 李治楫. “Guonei nongshi shiyan jiguan gaikuang” 國內農事試驗機關概況 (Overview of National Agricultural Experiment Institutes). Nongye tuiguang, no. 6 (1934): 5-9.

———. “Guonei nongshi shiyan jiguan gaikuang (er)” 國內農事試驗機關概況 (二) (Overview of National Agricultural Experiment Institutes, part 2). Nongye tuiguang, no. 7 (1934): 1-12.

———. “Guonei nongshi shiyan jiguan gaikuang (san)” 國內農事試驗機關概況 (三) (Overview of National Agricultural Experiment Institutes, part 3). Nongye tuiguang, no. 8 (1935): 1-18.

———. “Guonei nongshi shiyan jiguan gaikuang (si)” 國內農事試驗機關概況 (四) (Overview of National Agricultural Experiment Institutes, part 4). Nongye tuiguang, no. 9-10 (1935): 1-7.

Linqing xianzhi 臨清縣志 (Linqing County Gazetteer). 1934.

Lin Xiuzhu 林修竹 (Lin Maoquan 林茂泉). Maoquan shiye wenji 茂泉實業文集 (Maoquan’s Collected Works on Industry). 1926. In Jindai Zhongguo shiliao yekan,近代中國史料業刊, 3 bian 編, 47 juan 卷, edited by Shen Yunlong 沈雲龍. Taipei: Wenhai chubanshe 文海出版社, 1988.

“Liumei xuesheng xueke tongji” 留美學生學科統計 (Statistics of Students’ Disciplines Studying in the United States). Jiaoyu zazhi 教育雜誌 22, no. 3 (1930): 1.

“Lu jiansheting fazhan mianye jihua” 魯建設廳發展棉業計劃 (Shandong Office of Construction Plan to Develop the Cotton Industry). Nongcun fuxing weiyuanhui huibao 農村復興委員會會報 (Reports of the Committee to Revive the Villages) 2, no. 5 (1934): 109-110.

“Lu jiansheting tuiguang meimian” 魯建設廳推廣美棉 (Shandong Office of Construction Promote U.S. Cotton). Nongcun fuxing weiyuanhui huibao 1, no. 7 (1933): 184-185.

107

Qin Jie 秦杰. “Weida de guanhuai, wuqiong de liliang – jinian Mao zhuxi liangci shicha woyuan mianhua shiyan tian” 伟大的关怀, 无穷的力量 – 纪念毛主席两次视察我院棉花试验田 (Great Care, Infinite Power – Remembering Chairman Mao’s Two Visits to our Acadmy’s Coton Experiment Fields). August 1978. In Shandong sheng nongye kexueyuan yuanshi, 192-194.

“Quanguo nongye tuiguang shishi zhuangkuang diaocha” 全國農業推廣實施狀況調查 (National Agricultural Extension Implementation Survey). Nongye tuiguang, no. 9-10 (1935): 8-76.

Ren Dekuan 任德寬. “Shandong zhi meimian tuiguang wenti” 山东之美棉推广问题 (Problems with Extending U.S. Cotton in Shandong). Nongxue 农学 (Agronomy) 1, no. 5 (1924): 8-20.

Rong Zongjing 榮宗敬. “Yijian shu: zengjin mianchan wei fuxing nongcun zhi yaosu” 意見書：增進棉產為復興農村之要素 (Opinion: Promoting Cotton is a Key Element in Reviving the Countryside). Nongcun fuxing weiyuanhui huibao, no. 3 (1933): 22.

“Shandong nongchang jianshe qingkuang baogao” 山东农场建设情况报告 (Report on the Construction of Shandong Agricultural Stations). September 1948. In Shandong jiefang hou keji shiliao xuanbian 山东解放后科技史料选编 (Selection of Historical Materials on Science and Technology in Shandong after Liberation), edited by Shandong sheng kexue jishu weiyuanhui 山东省科学技术委员会, 134- 135. Shandong: Shandong sheng kexue jishu weiyuanhui, 1987.

“Shandong nongshi shiyanchang” 山東農事試驗場 (Shandong Agricultural Experiment Station). 1908. In Zhongguo jindai nongye shi ziliao, 874

Shandong nongye daxueshi 山东农业大学史 (Shandong Agricultural University History). Shandong, Tai’an: Shandong nongye daxue dianzi yinxiang chubanshe 山东农业大学电子音像出版社, 2006.

“Shandong sheng Bohaiqu nongye shiyanchang baogao” 山东省渤海区农业试验场报告 (Report on the Shandong Bohai District Agricultural Experiment Station). September 1948. In Shandong jiefang hou keji shiliao xuanbian, 135-137.

Shandong sheng nongye kexue yanjiusuo 山东省农业科学研究所 (Shandong Agricultural Science Research Institute). “Guanyu mianhua mizhi zengchan de diaocha yanjiu he jinhou yijian” 关于棉花密植增产的调查研究和今后意见 (On the Investigation of Close Planting Increasing Cotton Yields and Future Recommendations). In 1958 nian nongzuowu mizhi jingyan 1958 年农作物密植

108

经验 (Agricultural Crops Close Planting Experiments, 1958), edited by Nongye bu liangshi zuowu shengchan ju 农业部粮食作物生产局 (Ministry of Agriculture Crop Production Bureau), 98-105. Beijing: Nongye chubanshe 农业出版社, 1958.

Shandong sheng nongye kexueyuan 山东省农业科学院 (Shandong Academy of Agricultural Sciences). Kuayue shiji de tengfei: Shandong nongye kexueyuan chuangjian 110 zhounian (1903-2013) 跨越世纪的腾飞: 山东省农业科学院创建 110 周年 (Souring across the Century: Shandong Academy of Agricultural Sciences’ 110 Year Anniversary). Jinan: Shandong nongye kexueyuan 山东农业科学院, 2013.

———. Shandong sheng 1959 nian xiaomai fengchan dianxing jingyan 山东省 1959 年小麦丰产典型经验 (Typical Experiences of Wheat Bumper Harvests in Shandong Province, 1959). Jinan: Shandong renmin chubanshe 山东人民出版社, 1959.

———. Shandong sheng nongye kexueyuan yuanshi, 1946-1992 山东省农业科学院院史 (Shandong Academy of Agricultural Science Academy History, 1946-1992). Jinan: Jinan shi jianhua yinshuachang 济南市建华印刷厂, 1994.

Shandong sheng nongyeting 山东省农业厅 (Shandong Office of Agriculture) and Shandong sheng nongye kexueyuan eds. 1959 nian Shandong sheng guzi fengchan jingyan 1959 年山东省谷子丰产经验 (Typical Experiences of Millet Bumper Harvests in Shandong Province, 1959). Jinan: Shandong renmin chubanshe, 1960.

———. 1959 nian Shandong sheng huangyan fengchan jingyan 1959 年山东省黄烟丰产经验 (Typical Experiences of Tobacco Bumper Harvests in Shandong Province, 1959). Jinan: Shandong renmin chubanshe, 1960.

———. 1959 nian Shandong sheng huasheng fengchan jingyan 1959 年山东省花生丰产经验 (Typical Experiences of Peanut Bumper Harvests in Shandong Province, 1959). Jinan: Shandong renmin chubanshe, 1960.

———. 1959 nian Shandong sheng yumi fengchan jingyan 1959 年山东省玉米丰产经验 (Typical Experiences of Corn Bumper Harvests in Shandong Province, 1959). Jinan: Shandong renmin chubanshe, 1960.

“Shandong sheng shiye huiyi gexiang jueyi” 山东省实业会议各项决议 (Shandong Province Industrial Conference Resolutions). July 1948. In Shandong jiefang hou keji shiliao xuanbian, 125-134. 109

“Shandong shengli canye shiyanchang chengji baogao” 山東省立蠶業試驗場成績報告 (Shandong Provincial Sericulture Experiment Station Achievement Report). Shandong sheng banyuekan 山東省半月刊 (Shandong Fortnight) 2, no. 10 (1937): 112-202.

Shandong shengli di’er mianye shiyanchang 山東省立第二棉業試驗場 (Shandong Provincial Second Cotton Experiment Station) ed. Shandong shengli di’er mianye shiyanchang chengji baogao 山東省立第二棉業試驗場成績報告 (Shandong Provincial Second Cotton Experiment Station Achievement Report), no. 2 (1929).

———. Shandong shengli di’er mianye shiyanchang tuiguang baogao 山東省立第二棉業試驗場推廣報告 (Shandong Provincial Second Cotton Experiment Station Extension Report), no. 1 (1932).

———. Shandong shengli di’er mianye shiyanchang shiyan baogao 山東省立第二棉業試驗場試驗報告 (Shandong Provincial Second Cotton Experiment Station Experiment Report), no. 3 (1930).

———. Shandong shengli di’er mianye shiyanchang shiyan baogao 山東省立第二棉業試驗場試驗報告 (Shandong Provincial Second Cotton Experiment Station Experiment Report), no. 4 (1931).

———. Shandong shengli di’er mianye shiyanchang shiyan baogao 山東省立第二棉業試驗場試驗報告 (Shandong Provincial Second Cotton Experiment Station Experiment Report), no. 5 (1932).

“Shandong shengli di’er mianye shiyanchang: yucheng youliang mianzhong 山東省立第二棉業試驗場: 育成優良棉種 (Shandong Provincial Second Cotton Experiment Station: Breeding Superior Cotton). Nongye zhoubao 農業週報 (The Farmers Weekly), no. 59 (1930): 298.

Shandong shengli shuichan shiyanchang jikan 山東省立水產試驗場季刊 (Shandong Provincial Aquiculture Experiment Station Quarterly), no. 1 (1924).

Shandong shengzhi: nongye zhi 山东省志: 农业志 (Shandong Provincial Gazetteer: Agriculture). Vol. 2. Jinan: Shandong renmin chubanshe, 2000.

Shen Zhizhong 沈志忠. “Nongke liuxuesheng yu Zhongguo jindai nongye keji tizhihua jianshe” 农科留学生与中国近代农业科技体制化建设 (Returned Students Majoring in Agriculture and Institution Construction in the Field of Agricultural Science and Technology in Modern China). Anhui shixue 安徽史学 (Anhui History) 5 (2009): 5-11. 110

“Shiye bu guanyu quanguo nongshi shiyanchang diaocha de gexiang tongji” 实业部关于全国农事试验场调查的各项统计 (Department of Industry Regarding the Various Statistical Surveys on National Agricultural Experiment Stations). 1936. In Zhonghua minguoshi dang’an ziliao huibian 中華民國史檔案資料彙編 (Archive Compilation on Republic of China History), 5 ji 楫, 1 bian 編, Caizheng jingji 財政經濟 (Finance and Economy) no. 7, edited by Zhongguo di’er lishi dang’anguan 中國第二歷史檔案館 (Second Historical Archives of China), 392- 401. Jiangsu: Fenghuang chubanshe 鳳凰出版社.

Tang Qiming 唐齐鸣. “Cong Shandong nongshi shiyanchang dao Shandong nongye kexueyuan” 从山东农事试验场到山东农业科学院 (From Shandong Agricultural Experiment Station to Shandong Academy of Agricultural Science). Lishi zuyin 历史足音 (Historical Footprints) 3(2013): 8-11.

Tong Pingya 佟屏亚. “Nongye ‘dayuejin’ kaoliang nongye kexuejia: xiezai 1958nian nongye ‘dayuejin’ 50 zhounian” 农业”大跃进“考量农业科学家: 写在 1958 年农业”大跃进“50 周年 (Agricultural “Great Leap Forward” Examination of Agricultural Scientists: Writing on the 50th Anniversary of the Agricultural “Great Leap Forward”). Nongye kaogu 农业考古 (Agricultural Archaeology), no. 4 (2008): 7-16.

Tang Zhicai 唐志才. Nongye tuiguang 農業推廣 (Agricultural Extension). Shanghai: Zhengzhong shuju 正中書局, 1944.

Wu Zhi 吳知. “Shandong sheng mianhua zhi shengchan yu yunxiao” 山東省棉花之生產與運銷 (Cotton Production, Transport, and Marketing in Shandong Province). Zhengzhi jingji xuebao 政治經濟學報 (Journal of Political Economy) 1, no. 5 (1936): 1-90.

Xia Rubing 夏如冰. “Qingmo de nongzheng jigou yu nongye zhengce” 清末的农政机构与农业政策 (Agricultural Administrative Structures and Policies of the late Qing Dynasty). Nanjing nongye daxue xuebao 南京农业大学学报 (Journal of Nanjing Agricultural University) 3 (2002): 44-50.

“Yancao cuiya shiyan baogao” 煙草催芽試驗報告 (Tobacco Germination Experiment Report). Shandong sheng jianshe banyuekan 山东省建设半月刊 (Shandong Construction Fortnight) 1, no. 15 (1936): 1-8.

“Yancao gailiangchang Minguo ershisi nian zhuyao gongzuo baogao” 煙草改良場民國二十四年主要工作報告 (Tobacco Improvement Station Republic of China Year

111

24 Main Work Report). Shandong sheng jianshe banyuekan 1, no. 15 (1936): 95- 100.

“Yancao pinzhong bijiao shiyan baogao” 煙草品種比較試驗報告 (Tobacco Varieties Comparative Text Experiment Report). Shandong sheng jianshe banyuekan 1, no. 15 (1936): 17-29.

Yang Shimou 杨士谋. Nongye tuiguang jiaoyu gailun 农业推广教育概论 (An Introduction to Agricultural Extension Education). Beijing: Beijing nongye daxue chubanshe 北京农业大学出版社, 1987.

Yuan Pengxin 苑朋欣. “Shandong kangri genjudi nongye dashengchan yundong de zhengce yu xiaoguo” 山东抗日根据地农业大生产运动的政策与效果 (The Policies and Effectiveness of the Great Production Movement in Shandong’s Anti-Japanese Base Areas), Jinan daxue xuebao 济南大学学报 (Journal of Jinan University) 21, no. 4 (2011): 63-67.

Yu Gong 愚公. “Nongye tuiguang gailun” 農業推廣概論 (Overview of Agricultural Extension). Nongye tuiguang, no. 1 (1930): 1-9.

Zhang Ke 章柯. “Wo suo liaojie de Shandong jiefangqu de nongye keji gongzuo” 我所了解的山东解放区的农业科技工作 (My Understanding of the Agricultural Science and Technology Work in Shandong’ Liberate Zones). June 1984. In Shandong sheng nongye kexueyuan yuanshi, 182-183.

Zhuang Weimin 庄维民. “Jindai Shandong cansangye de gailiang jiqi yingxiang” 近代山东蚕桑业的改良及其影响 (Modern Shandong Sericulture Improvement and its Influence). Gujin nongye 古今农业 (Ancient and Modern Agriculture), no. 4 (1999): 35-42.

Zhang Zhiwen 章之汶 and Li Xingyu 李醒愚. Nongye tuiguang 農業推廣 (Agricultural Extension). Shangwu yinshuguan 商務印書館, 1936.

“Zhongguo liuri xuesheng xueke tongji” 中國留日學生學科統計 (Statistics of Chinese Students’ Disciplines Studying in Japan). Zhonghua jiaoyu jie 中華教育界 (China Education World) 18, no. 10 (1930): 4.

“Zhongguo nongshi shiyanchang zhi tongji” 中國農事試驗場之統計 (Statistics on China’s Agricultural Experiment Stations). Kexue 科學 (Science) 2, no. 6 (1916): 711.

112

Zhongyang nongye shiyansuo 中央農業實驗所 (National Agricultural Research Bureau). Nongqing baogao 農情報告 (Crop Reports), 1934-1938.

Zhongguo Shiyebu 中國實業部 (China Ministry of Industry), Guoji Maoyiju 國際貿易局 (Bureau of International Trade). Zhongguo shiye zhi: Shandong sheng 中國實業志: 山東省 (Industrial Gazetteer of China: Shandong Province). Nanjing: Guiji maoyiju 國際貿易部, 1934.

Zhonghua mianye tongjihui 中華棉業統計會 (Chinese Cotton Statistics Association). Zhongguo mianchan tongji 中國棉產統計 (Chinese Cotton Production Statistics). 1923, 1928-1938.

Zhou Xiaoyan 周晓焱 and Zhang Jianhua 张建华. “Nanjing guomin zhengfu de nongye tuiguang lifa yanjiu” 南京国民政府的农业推广立法研究 (A Study on Agricultural Extension Legislation of Nanjing National Government). Xibei nonglin keji daxue xuebao 西北农林科技大学学报 (Journal of Northwest A&F University) 12, no. 4 (2012): 139-145.

Zou Bingwen 鄒秉文. “Meiguo nongye tuiguangbu” 美國農業推廣部 (Agricultural Extension Department in the United States). Dongfang zazhi 東方雜誌 14, no. 9 (1917): 160-163.

Zuo Qiong 作瓊. “Shiye bu Shandong linye shiyanchang yange ji gaikuang” 實業部山東林業試驗場沿革及概況 (Department of Industry Shandong Silviculture Experiment Station History and Overview). Xinnong tongxun 新農通訊 (New Farming Newsletter), no. 14 (1933): 9-12.

English-Language Sources

Alitto, Guy. The Last Confucian: Liang Shu-ming and the Chinese Dilemma of Modernity. Berkeley: University of California Press, 1979.

Bates, Robert H. Markets and States in Tropical Africa: The Political Basis of Agricultural Policies. Berkeley: University of California Press, 2005.

Becker, Jasper. Hungry Ghosts: Mao’s Secret Famine. New York: Henry Holt and Company, 1996.

Bell, Lynda S. One Industry, Two Chinas: Silk Filatures and Peasant-Family Production in Wuxi County, 1865-1937. Stanford University Press, 1999.

113

Birkhaeuser, Dean, Robert E. Evenson, and Gershon Feder. “The Economic Impact of Agricultural Extension: A Review.” Economic Development and Cultural Change 39, no. 3 (1991): 607-650.

Brandt, Loren. Commercialization and Agricultural Development: Central and Eastern China, 1870-1937. New York: Cambridge University Press, 1989.

———. “Reflections on China’s Late 19th and Early 20th-Century Economy.” In Reappraising Republican China, edited by Frederic Wakeman, Jr. and Richard Louis Edmonds, 28-54. New York: Oxford University Press.

Brandt, Loren and Barbara Sands. “Land Concentration and Income Distribution in Republican China.” In Chinese History in Economic Perspective, edited by Thomas G. Rawski and Lillian M. Li, 179-206. Berkeley: University of California Press, 1992.

Bray, Francesca. “Agricultural Illustrations: Blueprint or Icon?” In Graphics and Text in the Production of Technical Knowledge in China, edited by Francesca Bray, Vera Dorofeeva-Lichtmann and George Metailie, 521-568. Leiden: Brill, 2007.

———. “Chinese Literati and the Transmission of Technological Knowledge: The Case of Agriculture.” In Cultures of Knowledge: Technology in Chinese History, edited by Dagmar Shafer, 299-326. Leiden: Brill, 2012.

———. Science and Civilision in China. Vol. 6, part 2, Agriculture. Cambridge: Cambridge University Press, 1984.

———. Technology and Society in Ming China (1368-1644). American Historical Association, 2000.

Brook, Timothy. "The Spread of Rice Cultivation and Rice Technology into the Hebei Region in the Ming and Qing." In Explorations in the History of Science and Technology in China, edited by Li Guohao, 659-690. Shanghai: Classics Publishing House, 1982.

Brown, Lawrence A. Innovation Diffusion: A New Perspective. New York: Methuen & Co., 1981.

Brown, Philip C. Cultivating Commons: Joint Ownership of Arable Land in Early Modern Japan. Honolulu: University of Hawai’i Press, 2011.

Buck, John Lossing. Chinese Farm Economy. Nanjing: University of Nanking, 1930.

———. “Fact and Theory about China’s Land.” Foreign Affairs 28, no. 1 (1949): 92- 101. 114

———. Land Utilization in China. Nanjing: University of Nanking, 1937.

———. “Some Basic Agricultural Problems of China.” Secretariat Paper No. 1, Tenth Conference of the Institute of Pacific Relations, Stratford-on-Avon, England. New York: International Secretariat, Institute of Pacific Relations, September 1947.

Burke III, Edmund and Kenneth Pomeranz eds. The Environment and World History. Berkeley: University of California, 2009.

Chao, Kang. Agricultural Production in Communist China, 1949-1965. Madison: University of Wisconsin Press, 1970.

———. The Development of Cotton Textile Production in China. Cambridge, MA: Harvard University Press, 1977.

Chen Hansheng (Chen Han-seng). The Present Agrarian Problem in China. Shanghai: China Institute of Pacific Relations, 1933.

China-United States Agricultural Mission. Report of the China-United States Agricultural Mission. Office of Foreign Agricultural Relations, 1947.

Delman, Jørgen. Agricultural Extension in Renshou County, China: A Case-Study of Bureaucratic Intervention for Agricultural Innovation and Change. Hamburg: Mitteilungen des Instituts fur Asienkunde, 1993.

Denton, Kirk A. Exhibiting the Past: Historical Memory and the Politics of Museums in Postsocialist China. Honolulu: University of Hawaii Press, 2014.

Dikötter, Frank. Mao’s Great Famine: The History of China’s Most Devastating Catastrophe, 1958-1962. New York: Walker & Company, 2010.

———. The Age of Openness: China before Mao. Berkeley: University of California Press, 2008.

Duncan, Cynthia M. Worlds Apart: Poverty and Politics in Rural America – Second Edition. New Haven: Yale University Press, 2014.

Elman, Benjamin A. “Toward a History of Modern Science in Republican China.” In Science and Technology in Modern China, 1880s-1940s, edited by Jing Tsu and Benjamin A. Elman, 15-38. Leiden: Brill, 2014.

Ensminger, Douglas and Irwin T. Sanders. “What Extension Is.” In Farmers of the World, edited by Edmund deS. Brunner, Irwin T. Sanders, and Douglas Ensminger, 1-7. New York: Columbia University Press, 1945. 115

Esherick, Joseph. The Origins of the Boxer Uprising. Berkeley: University of California, 1987.

Evenson, Robert E. “Economic Impacts of Agricultural Research and Extension.” In Handbook of Agricultural Economics. Vol. 1A, Agricultural Production, edited by Bruce L. Gardner and Gordon C. Rausser. Amsterdam: Elsevier Science, 2001.

Fei Xiaotong (Fei Hsiao-tung). China’s Gentry: Essays in Rural-Urban Relations. Chicago: University of Chicago Press, 1953.

Ferguson, James. The Anti-Politics Machine: “Development,” Depoliticization, and Bureaucratic Power in Lesotho. Minneapolis: University of Minnesota Press, 1994.

Fliedel, Frederick C. and Peter Korsching. Diffusion Research in Rural Sociology. Middleton: Social Ecology Free press, 2001.

Food and Agriculture Organization of the United Nations ed. Learning from China: A Report on Agriculture and the Chinese People’s Communes. Rome: Rood and Agriculture Organization of the United Nations, 1978.

Francks, Penelope. Technology and Agricultural Development in Pre-War Japan. New Haven: Yale University Press, 1984.

Freire, Paulo. Education for Critical Consciousness. New York: Seabury Press, 1973.

Gamble, Sidney D. Ting Hsien: A North China Rural Community. New York: International Secretariat Institute of Pacific Relations, 1954.

Gou Bingwen (Ping Wen Kuo). “Promotion and Application of Science in the Government of China” (1941). In Guo Bingwen xiansheng jinianji 郭秉文先生紀念集 (Mr. Guo Bingwen Memorial Collection). Taipei: Zhonghua xueshuyuan 中華學術院, 1971.

Griffing, J. B. “A New Variety of Cotton from China.” Journal of Heredity 18, no. 11 (1927): 496-497.

Hayford, Charles W. To the People: James Yen and Village China. New York: Columbia University Press, 1990.

Hightower, Jim. Hard Tomatoes, Hard Times: Failure of the Land Grant College Complex. Cabridge, MA: Shenkman Publishing Company, 1973.

116

Huang, Philip C.C. The Peasant Economy and Social Change in North China. Stanford: Stanford University Press, 1985.

———. The Peasant Family and Rural Development in the Yangzi Delta, 1350-1988. Stanford: Stanford University Press, 1990.

———. “The Study of Rural China’s Economic History.” Republican China 18, no. 1 (1992): 164-176.

Israel, John. Lianda: A Chinese University in War and Revolution. Stanford: Stanford University Press, 1998.

Jing Su and Luo Lun. Landlord and Labor in Late Imperial China: Case Studies from Shandong. Translated by Endymion Wilkinson. Cambridge, Mass.: Harvard University Press, 1978.

Jones, Gwyn E. and Chris Garforth. “The History, Development, and Future of Agricultural Extension.” In Improving Agricultural Extension: A Reference Manual, edited by Burton E. Swanson, Robert P. Bentz, and Andrew J. Sofranko, 3-12. Rome: Food and Agricultural Organization of the United Nations, 1997.

King, F.H. Farmers of Forty Centuries. Madison, Wisconsin: Democrat Printing Company, 1911.

Kraus, Richard A. Cotton and Cotton Goods in China, 1918-1936. New York: Garland Publishing, Inc., 1980.

Lewis, Oscar. The Children of Sanchez: Autobiography of a Mexican Family. New York: Random House, 1961.

Li, Lillian M. Fighting Famine in North China: State, Market, and Environmental Decline, 1690s-1990s. Stanford: Stanford University Press, 2007.

Liang, Ernest P. China: Railways and Agricultural Development, 1875-1935. Chicago: University of Chicago, Department of Geography, Research Paper No. 203, 1982.

Lobao, Linda M. “Continuity and Change in Place Stratification: Spatial Inequality and Middle-Range Territorial Units.” Rural Sociology 69, no. 1 (2004): 1-30.

Lobao, Linda M., Gregory Hooks, and Ann R. Tickamyer eds. The Sociology of Spatial Inequality. New York: SUNY Press, 2007.

Loewe, Michael. Early Life in Early Imperial China: During the Han Period 202 BC-AD 220. London: Batsford, 1968.

117

Mallory, Walter H. China: Land of Famine. New York: American Geographical Society, 1928.

Mancl, Karen M. “Environmental Technology Transfer in Rural China.” Master’s thesis, Ohio State University, 2010.

Marks, Robert B. China: Its Environment and History. Maryland: Rowman & Littlefield, 2012.

Moore, Aaron S. Constructing East Asia: Technology, Ideology, and Empire in Japan’s Wartime Era 1931-1945. Stanford: Stanford University Press, 2013.

Myers, Ramon H. “The Agrarian System.” In The Cambridge History of China. Vol. 13, Republican China 1912-1949, Part 2, edited by John K. Fairbank and Albert Feurwerker, 230-256. Cambridge, UK: Cambridge University Press, 1986.

———. The Chinese Peasant Economy: Agricultural Development in Hopei and Shantung, 1890-1949. Harvard University Press, 1970.

Ogura, Takekazu ed. Agricultural Development in Modern Japan. Tokyo: Fuji Publishing Company, 1963.

Perdue, Peter C. Exhausting the Earth: State and Peasant in Hunan, 1500-1850. Cambridge: Harvard University Press, 1987.

Perkins, Dwight H. Agricultural Development in China, 1368-1968. Chicago: Aldine Publishing Company, 1969.

Pomeranz, Kenneth. “Local Interest Story: Political Power and Regional Differences in the Shandong Capital Market, 1900-1937.” In Chinese History in Economic Perspective,295-318.

——— . The Making of a Hinterland: State, Society, and Economy in Inland North China, 1853-1937. University of California Press, 1993.

——— . “The Making of a Hinterland.” Ph.D. dissertation, Yale University, 1988.

——— . “The Transformation of China’s Environment, 1500-2000.” In The Environment and World History, 118-164.

Pratt, Edward E. Japan’s Protoindustrial Elite: The Economic Foundations of the Gono. Cambridge: Harvard University Press, 1999.

Rawski, Thomas G. Economic Growth in Prewar China. Berkeley: University of California Press, 1989. 118

Rogers, Everett M. Diffusion of Innovations. New York: The Free Press, 1962.

Röling, Niels. Extension Science: Information Systems in Agricultural Development. Cambridge: Cambridge University Press, 1988.

Rowe, William T. “Political, Social and Economic Factors Affecting the Transmission of Technical Knowledge in Early Modern China.” In Cultures of Knowledge: Technology in Chinese History, 25-44.

Rudolph, Frederick. The American College and University: A History. Athens: University of Georgia Press, 1990.

Schmalzer, Sigrid. “Breeding a Better China: Pigs, Practices, and Place in a Chinese County, 1929-1937.” Geographical Review 92, no. 1 (2002): 1-22.

Scott, James C. Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. New Haven: Yale University Press, 1998.

Shen, Grace Yen. Unearthing the Nation: Modern Geology and Nationalism in Republican China. Chicago: University of Chicago Press, 2014.

Shen, Zonghan. Agricultural Resources of China. New York: Cornell University Press, 1951.

——— . “First Attempts to Transform Chinese Agriculture, 1927-1937.” In The Strenuous Decade: China’s Nation-Building Efforts, 1927-1937, edited by Paul K. T. Sih, 205-232. New York: St. John’s University Press, 1970.

Small, Mario L., David J. Harding, and Michele Lamont. “Reconsidering Culture and Poverty.” The Annals of the American Academy of Political and Social Science (May 2010): 6-27.

Stavis, Benedict. “Agricultural Research and Extension Services in China.” World Development 6, no. 5 (1978): 631-645.

———. Making Green Revolution: The Politics of Agricultural Development in China. Ithaca: Rural Development Committee, Cornell University, 1974.

Strauss, Julia C. Strong Institutions in Weak Politics: State Building in Republican China, 1927-1940. Oxford: Oxford University Press, 1998.

———. “The Evolution of Republican Government.” In Reappraising Republican China, 75-97.

119

Stross, Randall E. The Stubborn Earth: American Agriculturalists on Chinese Soil, 1898- 1937. Berkeley: University of California Press, 1986.

Swanson, B.E. and J.B. Claar. “The History and Development of Agricultural Extension.” In Agricultural Extension: A Reference Manual, Second Edition, 1- 19.

Tawney, Richard H. Land and Labour in China. London: George Allen & Unwin LTD, 1932.

Thøgersen, Stig. A County of Culture: Twentieth-Century China Seen from the Village Schools of Zouping, Shandong. Ann Arbor: University of Michigan Press, 2002.

Thorp, James. Geography of the Soils of China. Beijing: National Geological Survey of China, 1939.

Thorp, James and T. Y. Tschau. “Notes on Shantung: A Reconnaissance Soil Survey of Shantung.” Turang zhuanbao 土壤專報 (Soil Bulletin), no. 14 (1936): 1-105.

Trauger, Amy, Carolyn Sachs, Mary Barbercheck, Nacy Ellen Kiernan, Kathy Brasier and Audrey Schwartzberg. “The Object of Extension: Agricultural Education and Authentic Farmers in Pennsylvania.” Sociologia Ruralis 50 (Apr. 2010): 85-103. van den Ban, Anne W. and Stuart Hawkins. Agricultural Extension. New York: Longman Scientific & Technical, 1988.

Vanclay, Frank and Geoffrey Lawrence. The Environmental Imperative: Eco-social Concerns for Australian Agriculture. Rockhampton: Central Queensland University Press, 1995.

Wiens, Thomas B. “Trends in the Late Qing and Republican Rural Economy: Reality or Illusion?” Republican China 18, no. 1 (1992): 63-76.

Yang, Hsin-Pao. “Promoting Cooperative Agricultural Extension Service in China.” In Farmers of the World, 46-60.

Yeh, Wen-hsin. The Alienated Academy: Culture and politics in Republican China, 1919- 1937. Cambridge, MA: Harvard University Press, 1990.

Zanasi, Margherita. Saving the Nation: Economic Modernity in Republican China. Chicago: University of Chicago Press, 2006.

Zhang Fuliang (Chang Fu Liang). “Agricultural education and country life.” Educational Review 22, no. 2 (1930): 188-193.

120

Zhang Xiaoyong and Corne Kempenaar, Agricultural Extension System in China. Wageningen: Plant Research International, 2009.

121