《国际政治经济学》文献资料（Use of History and Sociology in International Political Economy）Where do high growth political economies come from

World Development,Vol.22,No.9.pp.1269-1293.1994 Pergamon Copyright1994 Elsevier Science Ltd Printed in Great Britain.All rights reserved 0305-750X/94$7.00+0.00 0305.750X(94)00051-4 Where Do High Growth Political Economies Come From?The Japanese Lineage of Korea's "Developmental State" ATUL KOHLI* Princeton University.New lersey Summary.-While many scholars have sought to analyze South Korea's economic success,not enough attention has been paid to the impact of Japanese colonialism.Japanese colonial influence on Korea in 1905-45.while brutal and humiliating,was also decisive in shaping a political economy that later evolved into the high-growth South Korean path to development.More specifically,three state- society characteristics that we now readily associate as elements of the South Korean "model"originated during the colonial period:Korean state under the Japanese influence was transformed from a relatively corrupt and ineffective social institution into a highly authoritarian,penetrating organization,capable of simultaneously controlling and transforming Korean society:production-oriented alliances involving the state and dominant classes evolved,leading up to considerable expansion of manufacturing,including "exports:"and the lower classes in both the city and the countryside came to be systematically con- trolled by the state and dominant classes.While there were impurtan discontinuities following WWIl. when the dust settled.South Korea under Park Chung-Hee fell back into the grooves of colonial origins and traveled along them,well into the 1980s. Questions about the wide range of (economic)perfor- ment,the current debate bogs down over the interpre- mance of underdeveloped countries today ..belong as tation of this role,i.e.over the extent to which state much to history as (they)do to economic analysis. intervention was "market conforming"versus "mar- Sir Arthur Lewis ket distorting,"or to use a related set of concepts,the [We]turn to history and only to history if what we are extent to which the state "led"rather than"followed" seeking are the actual causes,sources,and conditions of the market.2 overt changes of patterns and structures in society. Interesting and significant as this debate is,it is Robert Nisbet also incomplete,Much of it revolves around unravel- ing the economic role of the South Korean state and,in turn.tracing the impact of this role on economic out- 1.INTRODUCTION comes.The prior question of why the South Korean state was able to do what it did,and the related genetic Three decades of sustained.high economic growth issue of the historical roots of the Korean political has made South Korea a "model of development." economy thus tend to be underemphasized.Since Performance of other developing countries is now there is much to be learned about the Korean"model often judged against that of "East Asian newly indus- of development"by adopting a longer historical per- trializing countries (NICs),"including South Korea. Scholars and policy makers around the world have become curious:"How did South Korea do it?;"Can *I would like to acknowledge the helpful comments of the others learn from the experience?"A large body of lit- anonymous reviewers and the following:Bruce Cumings erature has developed-some of it of rather high Anthony D'Costa,Frank Dobbin,Cary Fraser,Stephan quality-attempting to interpret the Korean political Haggard,Koichi Hamada,Chung-in Moon.James Palais. economy.A central debate in this literature concerns Gustav Ranis,Benjamin Schneider,Robert Wade and John the relative roles of the state and of the market in Waterbury.This paper was written during 1992-93.while I was on sabbatical and,thanks to T.N.Srinivasan.a visitor at explaining South Korea's economic success.While the Economic Growth Center.Yale University.Financial hardly any sensible observer continues to deny the support from Ford Foundation is also acknowledged.Final state's extensive role in Korean economic develop- revision accepted:March 15,1994. 1269

Pergamon World Developpmenr, Vol. 22, No. 9, pp. 1269-l 293, 1994 Copyright 0 1994 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0305-750x/94 $7.00 + 0.00 0305-750x(94)00051-4 Where Do High Growth Political Economies Come From? The Japanese Lineage of Korea’s “Developmental State” ATUL KOHLI* Princeton University, New Jersey Summary. - While many scholars have sought to analyze South Korea’s economic success, not enough attention has been paid to the impact of Japanese colonialism. Japanese colonial influence on Korea in 1905-45, while brutal and humiliating, was also decisive in shaping a political economy that later evolved into the high-growth South Korean path to development. More specifically, three state￾society characteristics that we now readily associate as elements of the South Korean “model” originated during the colonial period: Korean state under the Japanese influence was transformed from a relatively corrupt and ineffective social institution into a highly authoritarian, penetrating organization, capable of simultaneously controlling and transforming Korean society; production-oriented alliances involving the state and dominant classes evolved, leading up to considerable expansion of manufacturing, including “exports;” and the lower classes in both the city and the countryside came to be systematically con￾trolled by the state and dominant classes. While there were important discontinuities following WWII. when the dust settled, South Korea under Park Chung-Hee fell back into the grooves of colonial origins and traveled along them, well into the 1980s Questions about the wide range of (economic) perfor￾mance of underdeveloped countries today . belong as much to history as (they) do to economic analysis. Sir Arthur Lewis [We] turn to history and only to history if what we are seeking are the actual causes, sources, and conditions of overt changes of patterns and structures in society. Robert Nisbet 1, INTRODUCTION Three decades of sustained, high economic growth has made South Korea a “model of development.” Performance of other developing countries is now often judged against that of “East Asian newly indus￾trializing countries (NICs),” including South Korea. Scholars and policy makers around the world have become curious: “How did South Korea do it?; “Can others learn from the experience?” A large body of lit￾erature has developed - some of it of rather high quality - attempting to interpret the Korean political economy.’ A central debate in this literature concerns the relative roles of the state and of the market in explaining South Korea’s economic success. While hardly any sensible observer continues to deny the state’s extensive role in Korean economic develop￾ment, the current debate bogs down over the interpre￾tation of this role, i.e. over the extent to which state intervention was “market conforming” versus “mar￾ket distorting,” or to use a related set of concepts, the extent to which the state “led” rather than “followed” the market.’ Interesting and significant as this debate is, it is also incomplete. Much of it revolves around unravel￾ing the economic role of the South Korean state and, in turn, tracing the impact of this role on economic out￾comes. The prior question of why the South Korean state was able to do what it did, and the related genetic issue of the historical roots of the Korean political economy thus tend to be underemphasized. Since there is much to be learned about the Korean “model of development” by adopting a longer historical per- *I would like to acknowledge the helpful comments of the anonymous reviewers and the following: Bruce Cumings. Anthony D’Costa, Frank Dobbin, Cary Fraser, Stephan Haggard, Koichi Hamada, Chung-in Moon, James Palais, Gustav Ranis, Benjamin Schneider, Robert Wade and John Waterbury. This paper was written during 1992-93, while I was on sabbatical and, thanks to T. N. Srinivasan. a visitor at the Economic Growth Center, Yale University. Financial support from Ford Foundation is also acknowledged. Final revision accepted: March 15, 1994. 1269

1270 WORLD DEVELOPMENT spective,especially tracing its origins back to its Korea had competing historical legacies:e.g.,there Japanese colonial lineage.this neglect is unfortunate. was the distant legacy of Choson (i.e.of Korea under For example,few "developmentalists."if any, the rule of Yi dynasty)of corrupt court politics at the ascribe much significance to the continuities that link apex;then there were indigenous revolutionary ten- colonial and posteolonial Korea.This is certainly so dencies that found expression in North Korea;and among the more strictly economic analysts of South there was the possibility of considerable American Korean growth experience:however,somewhat sur- influence.Moreover.completely new paths could prisingly and unfortunately.this problem also charac- have been charted.Subsequent decisions were thus terizes the works of several institutionally sensitive critical in putting South Korea on a path that reestab- scholars of South Korea.Among the latter.some dis- lished historical continuities.Nevertheless,it is diffi- cuss the colonial period but quickly conclude that the cult to imagine South Korea adopting a growth path impact was not of lasting significance (e.g..Jones and that it did without a deeply influential lapanese colo- Sakong,1980,pp.22-37).others deny the contribu- nial past. tions of this past altogether.and yet others virtually More specifically.I trace below the colonial ori- ignore it.presumably because of a view that signifi- gins of three patterns that we now readily associate as cant changes in South Korean economy began only elements of the South Korean"model."First,I discuss after the adoption of an "export-led model of develop- how the Korean state under the Japanese influence ment"in the early 1960s.5 Korean scholarship on was transformed from a relatively corrupt and ineffec- Korea has its own,albeit understandable,blind spots: tive social institution into a highly authoritarian, the nationalist impulse often leads to a denial of any penetrating organization.capable of simultaneously continuity between colonial and postcolonial periods. controlling and transforming the Korean socicty.This lest the contemporary achievements be viewed as a is followed by an analysis of a second pattern.namely product of a much disliked colonial rule.Only a the new state's production-oriented alliances with the handful of Korean specialists,especially those with a dominant classes,an alliance that buttressed the strong historical bent,have understood and empha- state's capacity to both control and transform. sized the Japanese colonial roots of the more recent. Relatedly,it is also important to take note of the struc- high-growth Korean political cconomy.Building on tural changes in the economy:not only did the colo- the insights of this last group of Korean specialists.I nial economy experience steady growth and industri- attempt in this essay to reinterpret some specific his- alization.but it also became rather heavily torical materials with the hope of deriving general export-oriented,including exports of manufactured lessons of interest to scholars of comparative and products.Finally,there was the third pattern of brutal international development. repression and systematic control of the lower classes The argument below is that Japanese colonial in both the cities and the countryside.The cumulative influence on Korea.in 1905-45,was decisive in shap- impact of these state-class configurations was to help ing a political economy that later evolved into the create a framework for the evolution of a high-growth high-growth South Korean path to development. political economy.I also,toward the end of this dis- Japanese colonialism differed in important respects cussion,briefly suggest-though not develop,leav- from the colonialism of European powers.As late ing that for another essay-how these patterns con- developers,the Japanese made extensive use of state tinued into subsequent periods. power for their own economic development,and they It is important to reiterate that the main task of this used the same state power to pry open and transform paper is not to set the historical record straight.That is Korea in a relatively short period.Japanese colonial for historians of Korea:they are already busy doing so impact was thus more intense.more brutal and deeply and I am only building on some of their work.Given architectonic;it also left Korea with three and a half the importance of the South Korean case in the con- decades of economic growth (the average,annual temporary discourse on development,it is important growth rate in production was more than 3%)and a that developmentalists understand what country spe- relatively advanced level of industrialization (nearly ciafists already know:I thus hope to reinterpret and 35%of Korea's "national production"in 1940 origi- synthesize some specific materials with general impli- nated in mining and manufacturing).*While there cations.Three sets of general ideas will be debated via were important discontinuities in the postcolonial the historical materials.First.there are Korea-related period,the grooves that Japanese colonialism carved comparative questions.For example,how much on the Korean social soil cut deep.The decade and a choice does a developing country really have when half following the departure of the lapanese was at adopting a specific development strategy:ie.to what least chaotic,and often tragic.When the dust settled, extent was South Korea a beneficiary of its historical however.South Korea under Park Chung-Hee fell inheritance,as distinct from creating anew a high- back into the grooves of an earlier origin and traversed growth.export-oriented "model of development?" along them.well into the 1980s.Of course,this was Closely related is the issue of transferability of the not inevitable:historical continuities seldom are. Korean "model"across national boundaries:if the

1270 WORLD DEVELOPMENT spective, especially tracing its origins back to its Japanese colonial lineage. this neglect is unfortunate. For example, few “developmentalists,” if any, ascribe much significance to the continuities that link colonial and postcolonial Korea. This is certainly so among the more strictly economic analysts of South Korean growth experience;? however, somewhat sur￾prisingly and unfortunately. this problem also charac￾terizes the works of several institutionally sensitive scholars of South Korea. Among the latter, some dis￾cuss the colonial period but quickly conclude that the impact was not of lasting significance (e.g., Jones and Sakong, 1980, pp. 22-37). others deny the contribu￾tions of this past altogether? and yet others virtually ignore it, presumably because of a view that signifi￾cant changes in South Korean economy began only after the adoption of an “export-led model of develop￾ment” in the early 1960~.~ Korean scholarship on Korea has its own, albeit understandable, blind spots; the nationalist impulse often leads to a denial of any continuity between colonial and postcolonial periods, lest the contemporary achievements be viewed as a product of a much disliked colonial rule.” Only a handful of Korean specialists, especially those with a strong historical bent, have understood and empha￾sized the Japanese colonial roots of the more recent. high-growth Korean political economy.’ Building on the insights of this last group of Korean specialists, I attempt in this essay to reinterpret some specific his￾torical materials with the hope of derivjing general lessons of interest to scholars of comparative and international development. The argument below is that Japanese colonial influence on Korea, in 1905-45, was decisive in shap￾ing a political economy that later evolved into the high-growth South Korean path to development. Japanese colonialism differed in important respects from the colonialism of European powers. As late developers, the Japanese made extensive use of state power for their own economic development, and they used the same state power to pry open and transform Korea in a relatively short period. Japanese colonial impact was thus more intense, more brutal and deeply architectonic; it also left Korea with three and a half decades of economic growth (the average, annual growth rate in production was more than 3%) and a relatively advanced level of industrialization (nearly 35% of Korea’s “national production” in I940 origi￾nated in mining and manufacturing).* While there were important discontinuities in the postcolonial period, the grooves that Japanese colonialism carved on the Korean social soil cut deep. The decade and a half following the departure of the Japanese was at least chaotic, and often tragic. When the dust settled, however, South Korea under Park Chung-Hee fell back into the grooves of an earlier origin and traversed along them, well into the 1980s. Of course, this was not inevitable: historical continuities seldom are. Korea had competing historical legacies: e.g., there was the distant legacy of Chosen (i.e. of Korea under the rule of Yi dynasty) of corrupt court politics at the apex; then there were indigenous revolutionary ten￾dencies that found expression in North Korea; and there was the possibility of considerable American influence. Moreover, completely new paths could have been charted. Subsequent decisions were thus critical in putting South Korea on a path that reestab￾lished historical continuities. Nevertheless, it is diffi￾cult to imagine South Korea adopting a growth path that it did without a deeply influential Japanese colo￾nial past. More specifically, I trace below the colonial ori￾gins of three patterns that we now readily associate as elements of the South Korean “model.” First, I discuss how the Korean state under the Japanese influence was transformed from a relatively corrupt and ineffec￾tive social institution into a highly authoritarian, penetrating organization, capable of simultaneously controlling and transforming the Korean society. This is followed by an analysis of a second pattern, namely, the new state’s production-oriented alliances with the dominant classes. an alliance that buttressed the state’s capacity to both control and transform. Relatedly, it is also important to take note of the struc￾tural changes in the economy: not only did the colo￾nial economy experience steady growth and industri￾alization, but it also became rather heavily export-oriented, including exports of manufactured products. Finally, there was the third pattern of brutal repression and systematic control of the lower classes in both the cities and the countryside. The cumulative impact of these state-class configurations was to help create a framework for the evolution of a high-growth political economy. I also, toward the end of this dis￾cussion, briefly suggest - though not develop, leav￾ing that for another essay - how these patterns con￾tinued into subsequent periods. It is important to reiterate that the main task of this paper is not to set the historical record straight. That is for historians of Korea; they are already busy doing so and 1 am only building on some of their work. Given the importance of the South Korean case in the con￾temporary discourse on development, it is important that developmentalists understand what country spe￾cialists already know; I thus hope to reinterpret and synthesize some specific materials with general impli￾cations. Three sets of general ideas will be debated via the historical materials. First, there are Korea-related comparative questions. For example, how much choice does a developing country really have when adopting a specific development strategy: i.e. to what extent was South Korea a beneficiary of its historical inheritance. as distinct from creating anew a high￾growth. export-oriented “model of development‘!” Closely related is the issue of transferability of the Korean “model” across national boundaries: if the

JAPANESE LINEAGE OF KOREA'S"DEVELOPMENTAL STATE 1271 roots of contemporary South Korean political econ- officials-cum-aristocrats,the Yangban,mutually omy are indeed as deep as a relatively unique colonial checked each others powers.The power of the experience.can others really emulate the experience? Yangban class rested in part on access to hereditary Second,at a higher level of generality,there are theo- land wealth but also on a close identification with the retical issues revolving around the concept of"devel- centralized bureaucracy,which helped both secure opmental states":what characterizes them and where socioeconomic privileges and was a further source of do they come from?Finally,at the most general level. wealth and power.Royal authority,in turn,was sel- there is at least an implication in this essay that some dom all that great.Being under Chinese suzerainty, of the variations we notice today among the more or Korean emperors did not enjoy the "mandate of less dynamic Third World political economies may heaven"that the Chinese emperors possessed.In addi- have some of their roots in a variable colonial past.If tion,the recruitment of the aristocracy to the bureau- so,a further investigation of this analytical claim cracy via the examination system enabled landed would require reopening the issue of the colonial roots power to be deeply embedded all through the Korean of the contemporary Third World that has unfortu- state,checking the scope of Royal authority vis-a-vis nately been lost in the postdependency scholarship on the Yangban.10 While this balance of power was a development. source of stability for several centuries,as external pressures grew,and along with it the state's need for taxes and other socioeconomic resources,it also 2.THE CONSTRUCTION OF A COLONIAL became a major constraint on monarchial power to STATE initiate reforms:The monarchial state,according to Palais,"could not solve the problem of creating ade- (a)The old,"predatory"state quate political authority for the achievement of national goals."Yi state was thus simultaneously By the time the Japanese gained decisive influence "centralized and weak." over Korea-say around 1905.after the Japanese vic- In addition to the limiting balance of power tory in Russo-Japanese war of 1904-the old state between the monarchy and the Yangban,there were within Choson was already in an advanced state of other factors at work that contributed to the Yi state's disintegration.While it is not necessary to recall his- ineffectiveness.First,it was not merely the presence torical details,a brief understanding of the state-soci- of a powerful land-controlling strata in society that ety links in late Choson are essential to appreciate limited the state's capacity.As I discuss below,the changes wrought by Japanese colonial power.The Yi colonial state in Korea carved out a different type of dynasty had provided continuous and,for the most ruling alliance with the same landowning class,but part,stable rule to Korea for nearly 500 years.The with a vastly more effective state.The key factor at same intricate state and class alliances that were work in Yi Korea was thus the direct control that responsible for this stability,however,also became landed groups exercised on state offices (Fairbank. major constraints on successful adaptation to chang- Reischauer and Craig,1978,p.307).Second,the ing external pressures,especially in the second half of Korean monarchy remained to the end a highly per- the nineteenth century.For example,the clearest man- sonalistic,patrimonial institution.In the words of ifestation of the powerlessness of a centralized monar- Cumings (1981),the Korean monarchs were incap- chial state was the continued inability to collect taxes able of acting along"the modern distinction between owed to the state on agrarian incomes,especially from public and private realms"and thus incapable of the powerful Yangban elite,the landowning-official designing state-led national goals of economic devel- class of Korea(Palais,1975).This recurring inability, opment (p.10).Third,the ruling strata below the in turn.came to be associated with several problem- monarch was highly factionalized,2 Such strife in the atic political trends:First,the state resorted to squeez- ruling strata made it difficult to design cohesive ing the peasantry via"taxation"(e.g.,corvee labor and responses to growing challenges.Finally,it is impor- military service),contributing to brigandage and a tant to note that the reach of the Yi state from the cen- restive peasant population.Second,the state's limited ter to the periphery was rather limited.While provin- resources exacerbated the competition and tensions in cial and county officials were directly appointed from what was already a personalized and factionalized Seoul,each county magistrate was responsible for elite at the apex of the political pyramid.Finally, governing nearly 40,000 people (there being some financial limitations made it difficult to mobilize any 330 magistrates for about 12 million Koreans).3 Since serious military response to growing external pres- these magistrates were rotated frequently,they often sures. depended on the well-entrenched Yangban elite for How does one explain powerlessness in a central- local governance.Moreover,the lower level officials ized polity?The leading historian of late Yi Korea, below the magistrate-were not salaried employ James Palais,traces the roots of this conundrum back ees.They were rather a hereditary group who were to the manner in which the monarchy and the Korean allowed to collect and keep some local taxes as com-

JAPANESE LINEAGE OF KOREA’S “DEVELOPMENTAL STATE” 1271 roots of contemporary South Korean political econ￾omy are indeed as deep as a relatively unique colonial experience, can others really emulate the experience? Second, at a higher level of generality, there are theo￾retical issues revolving around the concept of “devel￾opmental states”: what characterizes them and where do they come from? Finally, at the most general level, there is at least an implication in this essay that some of the variations we notice today among the more or less dynamic Third World political economies may have some of their roots in a variable colonial past. If so, a further investigation of this analytical claim would require reopening the issue of the colonial roots of the contemporary Third World that has unfortu￾nately been lost in the postdependency scholarship on development. 2. THE CONSTRUCTION OF A COLONIAL STATE (a) The old, “predatory” state By the time the Japanese gained decisive influence over Korea-say around 1905, after the Japanese vic￾tory in Russo-Japanese war of 1904 - the old state within Chosbn was already in an advanced state of disintegration. While it is not necessary to recall his￾torical details, a brief understanding of the state-soci￾ety links in late Chosen are essential to appreciate changes wrought by Japanese colonial power.‘The Yi dynasty had provided continuous and, for the most part, stable rule to Korea for nearly 500 years. The same intricate state and class alliances that were responsible for this stability, however, also became major constraints on successful adaptation to chang￾ing external pressures, especially in the second half of the nineteenth century. For example, the clearest man￾ifestation of the powerlessness of a centralized monar￾chial state was the continued inability to collect taxes owed to the state on agrarian incomes, especially from the powerful Yangban elite, the landowning-official class of Korea (Palais, 1975). This recurring inability, in turn, came to be associated with several problem￾atic political trends: First, the state resorted to squeez￾ing the peasantry via “taxation” (e.g., corvee labor and military service), contributing to brigandage and a restive peasant population. Second, the state’s limited resources exacerbated the competition and tensions in what was already a personalized and factionalized elite at the apex of the political pyramid. Finally, financial limitations made it difficult to mobilize any serious military response to growing external pres￾sures. How does one explain powerlessness in a central￾ized polity? The leading historian of late Yi Korea, James Palais, traces the roots of this conundrum back to the manner in which the monarchy and the Korean officials-cum-aristocrats, the Yangban, mutually checked each others powers. The power of the Yangban class rested in part on access to hereditary land wealth but also on a close identification with the centralized bureaucracy, which helped both secure socioeconomic privileges and was a further source of wealth and power. Royal authority, in turn, was sel￾dom all that great. Being under Chinese suzerainty, Korean emperors did not enjoy the “mandate of heaven” that the Chinese emperors possessed. In addi￾tion, the recruitment of the aristocracy to the bureau￾cracy via the examination system enabled landed power to be deeply embedded all through the Korean state, checking the scope of Royal authority vis-d-vis the Yangban. I0 While this balance of power was a source of stability for several centuries, as external pressures grew, and along with it the state’s need for taxes and other socioeconomic resources, it also became a major constraint on monarchial power to initiate reforms: The monarchial state, according to Palais, “could not solve the problem of creating ade￾quate political authority for the achievement of national goals.” Yi state was thus simultaneously “centralized and weak.“” In addition to the limiting balance of power between the monarchy and the Yangban, there were other factors at work that contributed to the Yi state’s ineffectiveness. First, it was not merely the presence of a powerful land-controlling strata in society that limited the state’s capacity. As I discuss below, the colonial state in Korea carved out a different type of ruling alliance with the same landowning class, but with a vastly more effective state. The key factor at work in Yi Korea was thus the direct control that landed groups exercised on state offices (Fairbank, Reischauer and Craig, 1978, p. 307). Second, the Korean monarchy remained to the end a highly per￾sonalistic, patrimonial institution. In the words of Cumings (1981) the Korean monarchs were incap￾able of acting along “the modern distinction between public and private realms” and thus incapable of designing state-led national goals of economic devel￾opment (p. 10). Third, the ruling strata below the monarch was highly factionalized.r2 Such strife in the ruling strata made it difficult to design cohesive responses to growing challenges. Finally, it is impor￾tant to note that the reach of the Yi state from the cen￾ter to the periphery was rather limited. While provin￾cial and county officials were directly appointed from Seoul, each county magistrate was responsible for governing nearly 40,000 people (there being some 330 magistrates for about 12 million Koreans).” Since these magistrates were rotated frequently, they often depended on the well-entrenched Yangban elite for local governance. Moreover, the lower level officials -below the magistrate - were not salaried employ￾ees. They were rather a hereditary group who were allowed to collect and keep some local taxes as com-

1272 WORLD DEVELOPMENT pensation for their services.These petty functionaries regional power competition for quite some time operated virtually as local ezars.not easily influenced (Conroy.1960).Given the mercantilist nature of from above and responsible for the "venality and Japanese political economy.however.it does not exploitation of the peasant population"(Lee in Eckert make much sense to raise the old question on imperi- eral,199.p.111. alism of whether security was more important than In sum.the ineffectiveness of the Yi state was economic interests.More than in the case of most rooted in part in the pattern of state-class linkages and imperial powers.the Meiji oligarchs of Japan readily in part in the design of the state itself.Regarding the associated national power with national wealth and latter.a personalistic apex.a factionalized ruling national wealth with overseas economic opportunities strata,and a limited downward reach of central (Duus.1984.Pp.132-133. authorities.were all significant characteristics con- Certain unique aspects of Japanese imperialism are tributing to the state's powerlessness.This state- essential to note for a full understanding of the colo- weak from the inside and hemmed in by powerful nial impact on Korea (Peattie,1984,pp.3-60).First, social actors from the outside-contributed little,if the Japanese had themselves barely escaped being any.to sustained economic progress.4 Worse,when imperialized.As both late developers and late imperi faced with growing security challenges and related alists.Japan colonized neighboring states with whom fiscal crises,the Yi state turned on its own society. they shared racial and cultural traits:it was as if becoming rapacious and predatory.The views of sev- England had colonized a few.across-the-channcl con- eral historians and observers of the day converge on tinental states.Proximity meant that many more such a perspective on the precolonial Korean state: Japanese ended up playing a direct role in colonial programs of the Yi government became "embezzle- rulc.including a much larger role of military and ment facilities for a rapacious officialdom"(Lew in police,than was ever the case in European overseas Eckert et al..1990,p.179):"maladministration...of colonies.The near geographical contiguity and shared the native Yi dynasty had affected adversely the culural and racial traits also implied that the Japanese whole of Korean public service"(Ireland.1926.p. could realistically consider their rule to be permanent, 92):"one of the strongest and most fixed impressions leading eventually to a full integration of colonies into made (during my travels to Korea)was that of the an expanded Japan.As will be discussed below,this well-nigh hopeless corruption of Korean court:and possibility.in turn,influenced both the economic and the Korean government "takes from the people political strategies of Japan in Korea.especially the directly and indirectly.everything that they carn over Japanese-initiated industrialization of Korea. and above a bare subsistence.and gives them in return Furthermore.Japanese colonial strategy was practically nothing."I deeply informed by their own successful domestic Since corrupt and ineffective states are indeed a reform efforts following the Meiji restoration.Of all common feature in parts of the contemporary Third the colonizing nations,Japan stands out as nearly the World.one may genuinely wonder:how was Korea's only one with a successful record of deliberate,state- "predatory"state historically transformed into what led political and economic transformation.By trial some may describe as a"developmental"state The and error the Meiji oligarchs had designed a political impact of Japanese colonial power was decisive in economy that was well suited for the task of"catching altering both the nature of the Korean state and the up"with advanced Western powers.The essential ele- relationship of this state to various social classes.The ments of this political economy are well known and transformation of the state is discussed immediately can be briefly reiterated:the creation of an effective below and the changing relationship of the state to centralized state capable of both controlling and trans- social classes in subsequent sections. forming Japanese society:deliberate state interven- tion aimed.first at agricultural development.and sec- ond at rapid industrial growth:and production of a (b)Toward a "developmental"state disciplined,obedient and educated work force.It was this model of deliherate development.with its empha- The Japanese military victory over the Russians in sis on state building and on the use of state power to 1904 marked the emergence of Japan as the major facilitate socioeconomic change-in contrast say,to regional power,a power that had been rising steadily the British.who having created a private property since the Meiji restoration in the 1860s.Subsequently. regime,waited in vain for Bengali samtindars in India Japan.with the acquiescence of Western powers,had to turn into a sheep farming gentry-that moved the a relatively free hand in dominating and molding Japanese colonizers.I%In Peattie's words,much of Korea.Japanese motives in Korea,like the motives of what Japan undertook in its colonies "was based upon all imperial powers.were mixed:they sought to con- Meiji experience in domestic reform"(Peattie.1984. trol it politically and to transform it economically for P.29. their own advantage.Security concerns were probably It is not surprising that the earliest Japanese efforts dominant insofar as Korea had been an object of in Korea were focused on reforming the disintegrating

I272 WORLD DEVELOPMENT pensation for their service\. These petty functionaries operated virtually as local czars. not easily influenced from above and responsible for the “venality and exploitation of the peasant population” (Lee in Eckert er LII.. 1990. p. 1 I I). In sum. the ineffectiveness of the Yi state was rooted in part in the pattern of state-class linkages and in part in the design of the state itself. Regarding the latter, a personalistic apex. a factionalized ruling strata, and a limited downward reach of central authorities. were all significant characteristics con￾tributing to the state’s powerlessness. This state - weak from the inside and hemmed in by powerful social actors from the outside - contributed little. if any, to sustained economic progress? Worse, when faced with growing security challenges and related fiscal crises, the Yi state turned on its own society, becoming rapacious and predatory. The views of sev￾eral historians and observers of the day converge on such a perspective on the precolonial Korean state: programs of the Yi government became “embezzle￾ment facilities for a rapacious officialdom” (Lew in Eckert et crl., 1990. p. 179); “maladministration of the native Yi dynasty had affected adversely the whole of Korean public service” (Ireland. 1926. p, 92): “one of the strongest and most fixed impressions made (during my travels to Korea) WI\ that of the well-nigh hopeless corruption of Korean court:“‘i and the Korean govermnent “takes from the people directly and indirectly, everything that they earn ovel and above a bare subsistence. and gives them in return practically nothing.“l(> Since corrupt and ineffecti\;e states arc indeed a common feature in parts of the contemporary Third World. one may genuinely wonder: how ~a\ Korea’s “predatory” state historically transformed into what some may describe as a “developmental” state?” The impact of Japanese colonial pow’er was decisive in altering both the nature of the Korean state and the relationship of this state to various social classes. The transformation of the state is discussed immediately below and the changing relationship of the state to social classes in subsequent sections,. The Japanese military victory over the Russians in 1904 marked the emergence of Japan as the major regional power. a power that had been ri\ing ste&ly since the Meiji restoration in the 1860s. Subsequently. Japan. with the acquiescence of Wcstcrn powers. had a relatively free hand in dominating and molding Korea. Japanese motives in Korea. like the motives of all imperial powers. were mixed; they sought to con￾trol it politically and to transform it economically lip their own advantage. Security concern\ were probably dominant insofar as Korea had been an ob.jcct of regional power competition for quite some time (Conroy. 1960). Given the mercantilist nature of Japanese political economy. however, it does not make much sense to raise the old question on imperi￾alism of whether security was more important than economic interests. More than in the case of most imperial powers. the Meiji oligarchs of Japan readily associated national power with national wealth and national wealth with overseas economic opportunities (Duus. 1984, pp. 132-133). Certain unique aspects of Japanese imperialism are essential to note for a full understanding of the colo￾nial impact on Korea (Peattie. 1984, pp. 3-60). First, the Japanese had themselves barely escaped being imperialized. As both late developers and late imperi￾alists. Japan colonized neighboring states with whom they shared racial and cultural traits; it was as if England had colonized a few, across-the-channel con￾tinental states. Proximity meant that many more Japanese ended up playing a direct role in colonial rule. including a much larger role of military and police. than was ever the case in European overseas colonies. The near geographical contiguity and shared cultural and racial traits also implied that the Japanese could realistically consider their rule to be permanent, leading eventually to a full integration of colonies into an expanded Japan. As will be discussed below. this possibility, in turn, influenced both the economic and political strategies of Japan in Korea. especially the Japanese-initiated industrialization of Korea. Furthermore. Japanese colonial strategy was deeply informed by their own successful domestic reform efforts following the Meiji restoration. Of all the colonizing nations. Japan stands out as nearly the only one with a successful record of deliberate. state￾led political and economic transformation. By trial and error the Meiji oligarchs had designed a political economy that was well suited for the task of “catching up” with advanced Western powers. The essential ele￾ments of this political economy are well known and can be briefly reiterated: the creation of an effective centralized state capable of both controlling and trans￾forming Japanese society: deliberate state interven￾tion aimed. first at agricultural development. and Tee￾trnd at rapid industrial growth: and production of a disciplined. obedient and educated work force. It was this model of deliberate development. with its empha￾si\ on state building and on the use of state power to facilitate socioeconomic change - in contrast say, to the British, who having created a private property regime. waited in vain for Bengali ~ctminclar.s in India to turn into a sheep farming gentry -that moved the Japanejc colonizers. IH In Peattie’s words, much of v, hat Japan undertook in its colonies “was based upon Meiji experience in domestic reform” (Peattie, 1984. p. 2Y). It is not surprising that the earliest Japanese efforts in Korea were focused on reforming the disintegrating

JAPANESE LINEAGE OF KOREA'S"DEVELOPMENTAL STATE 1273 Choson state;both political control and economic General,1909,p.45);scholarly observers have in transformation depended on it.A fair number of polit- retrospect characterized the Japanese colonial civil ical reforms had thus in fact been put into place during service as "outstanding,"composed of"hard working 1905-10.especially 1907-09,even prior to the formal and trusted cadres,"who deserve "high marks as a annexation of Korea in 1910.Subsequently,the group"(Peattie,1984,p.26).Elements of the highly decade of 1910-20 was again critical,whcn,undcr devclopcd,Japancsc stylc of burcaucratic govcrnment very harsh authoritarian circumstances,a highly were thus transferred directly to Korea. bureaucratized and a deeply penetrating state was con- structed. (i)The new civil service A key architect of the early reforms in 1907-09, While other colonial powers in other parts of the whose role helps us trace the origins of the design of world also created a competent civil service (e.g.,the the new Korean state,was the Meiji oligarch and the British in India),the Japanese colonial project was former Meiji era premier of Japan,Ito Hirubumi.Ito qualitatively distinct;both the extent and the intensity as a young man had been one of the handful of leaders of bureaucratic penetration was unique.There were who had led the Meiji"revolution"and who had sub- some 10,000 officials in the Japanese-Korean govern- sequently participated in the reform efforts that fol- ment in 1910:by 1937.this number was up to 87.552 lowed the destruction of Tokugawa Shogunate.Ito More than half of these government officials in 1937, had travelled extensively in Europe and had been fas- 52,270 to be exact,were Japanese.Contrast this with cinated with Prussian bureaucracy as a model for the French in Vietnam (where,by the way,the pres- Japan:the Prussian "model"offered to him a route to ence of the French was already more significant than Western rationality and modernity without"succumb- say,that of the British in Africa),who ruled a nearly ing"to Anglo-American liberalism (Halliday,1975, similar sized colony with some 3,000 French;in other p.37).Within Japan,Ito in 1878 had "led the cam- words.there were nearly 15 Japanese officials in paign to make the bureaucracy the absolutely unas- Korea for every French administrator in Vietnam sailable base and center of political power in the state (Robinson,in Eckert et al.,1990,p.257).The pres- system."Subsequently,Ito helped reorganize Tokyo ence of Korean bureaucrats,trained and employed by University in 1881 as a "school for government the Japanese,was also sizable:nearly 40,000 Koreans bureaucrats"and by 1887,"a basic civil service and qualified as government officials just before WWIl. entrance apprenticeship based on the Prussian model While most of the Koreans did not occupy senior posi- was installed."19 With this experience behind him, tions in the colonial government,there can be little when Ito was appointed in the early 1900s to run the doubt that,over the four decades of colonial rule,they Korean protectorate,and where his powers as became an integral part of a highly bureaucratic form Resident-General were near absolute "The of government.Moreover,during WWIl.as the uncrowned King of Korea"-he was quite self-con- demand for Japanese officials grew elsewhere,many scious of his task:"Korea can hardly be called an Koreans moved higher up in the bureaucratic hierar- organized state in the modern sense:I am trying to chy.I will return below to the issue of continuity:this make it such"(Ladd,1908.pp.435 and 174). sizable cadre of Japanese-trained Korean bureaucrats Ito and his successors set out to deliberately con- virtually took over the day-to-day running of a trun- struct a new Japanese-controlled Korean state.The cated South Korea,first under US military government first task was to gain central control.With superior and eventually when a sovereign state was formed. military power behind them,the Japanese in 1907 dis- One further characteristic of the colonial govern- mantled the Korean army,repressed those who ment that needs to be underlined is the successful links "mutinied."incorporated other army officers into a that the Japanese created between a highly concen- Japanese-controlled gendarmery,and forced the trated power center in Seoul,and a densely bureaucra- Korean monarch to abdicate.Having captured the tized periphery.All bureaucracies face the problem of heart of the state.the colonial rulers sought to system- how to ensure that central commands are faithfully atically create a depersonalized "public arena,"to implemented by the officials at the bottom rung.This, spread their power both wide and deep,and to coopt in turn,requires ensuring that lower level officials and/or repress native Korean political forces.For respond mainly to those above them in the bureau example,the patrimonial elements of the monarchial cratic hierarchy.rather than to personal interests,or to state were destroyed rather early,and replaced by a the interests of societal actors with whom they inter- cabinet-style government run by Japanese burcau- act.The Japanese in Korea were quite self-conscious crats.20 Since the appointments of these and other of this problem and repeatedly experimented till they lower level bureaucrats were governed by "elaborate arrived at arrangements deemed satisfactory. rules and regulations which,in the main follow(ed) Of course,certain circumstances were helpful in the lines of the Imperial Japanese services."the new establishing authority links between the center and the Korean state quickly acquired a "rational"character periphery:ruling arrangements in Seoul were highly (Ireland,1926,p.104;and H.I.J.M.'s Residency authoritarian-the power of the Japanese Governor

JAPANESE LINEAGE OF KOREA’S “DEVELOPMENTAL STATE” 1213 Chosbn state; both political control and economic transformation depended on it. A fair number of polit￾ical reforms had thus in fact been put into place during 1905-10, especially 1907-09, even prior to the formal annexation of Korea in 1910. Subsequently, the decade of 1910-20 was again critical, when, under very harsh authoritarian circumstances, a highly bureaucratized and a deeply penetrating state was con￾structed. A key architect of the early reforms in 1907-09, whose role helps us trace the origins of the design of the new Korean state, was the Meiji oligarch and the former Meiji era premier of Japan, Ito Hirubumi. Ito as a young man had been one of the handful of leaders who had led the Meiji “revolution” and who had sub￾sequently participated in the reform efforts that fol￾lowed the destruction of Tokugawa Shogunate. Ito had travelled extensively in Europe and had been fas￾cinated with Prussian bureaucracy as a model for Japan: the Prussian “model” offered to him a route to Western rationality and modernity without “succumb￾ing” to Anglo-American liberalism (Halliday, 1975, p. 37). Within Japan, Ito in 1878 had “led the cam￾paign to make the bureaucracy the absolutely unas￾sailable base and center of political power in the state system.” Subsequently, Ito helped reorganize Tokyo University in 1881 as a “school for government bureaucrats” and by 1887, “a basic civil service and entrance apprenticeship based on the Prussian model was installed.“19 With this experience behind him, when Ito was appointed in the early 1900s to run the Korean protectorate, and where his powers as Resident-General were near absolute - “The uncrowned King of Korea” -he was quite self-con￾scious of his task: “Korea can hardly be called an organized state in the modem sense; I am trying to make it such” (Ladd, 1908, pp. 435 and 174). Ito and his successors set out to deliberately con￾struct a new Japanese-controlled Korean state. The first task was to gain central control. With superior military power behind them, the Japanese in 1907 dis￾mantled the Korean army, repressed those who “mutinied,” incorporated other army officers into a Japanese-controlled gendarmery, and forced the Korean monarch to abdicate. Having captured the heart of the state, the colonial rulers sought to system￾atically create a depersonalized “public arena,” to spread their power both wide and deep, and to coopt and/or repress native Korean political forces. For example, the patrimonial elements of the monarchial state were destroyed rather early, and replaced by a cabinet-style government run by Japanese bureau￾crats.20 Since the appointments of these and other lower level bureaucrats were governed by “elaborate rules and regulations which, in the main follow(ed) the lines of the Imperial Japanese services,” the new Korean state quickly acquired a “rational” character (Ireland, 1926, p. 104; and H.I.J.M.‘s Residency General, 1909, p. 45); scholarly observers have in retrospect characterized the Japanese colonial civil service as “outstanding,” composed of “hard working and trusted cadres,” who deserve “high marks as a group” (Peattie, 1984, p. 26). Elements of the highly developed, Japanese style of bureaucratic government were thus transferred directly to Korea. (i) The new civil service While other colonial powers in other parts of the world also created a competent civil service (e.g., the British in India), the Japanese colonial project was qualitatively distinct; both the extent and the intensity of bureaucratic penetration was unique. There were some 10,000 officials in the Japanese-Korean govem￾ment in 1910; by 1937. this number was up to 87,552. More than half of these government officials in 1937, 52,270 to be exact, were Japanese. Contrast this with the French in Vietnam (where, by the way, the pres￾ence of the French was already more significant than, say, that of the British in Africa), who ruled a nearly similar sized colony with some 3,000 French; in other words, there were nearly 15 Japanese officials in Korea for every French administrator in Vietnam (Robinson, in Eckert et al., 1990, p. 257). The pres￾ence of Korean bureaucrats, trained and employed by the Japanese, was also sizable: nearly 40,000 Koreans qualified as government officials just before WWII. While most of the Koreans did not occupy senior posi￾tions in the colonial government, there can be little doubt that, over the four decades of colonial rule, they became an integral part of a highly bureaucratic form of government. Moreover, during WWII, as the demand for Japanese officials grew elsewhere, many Koreans moved higher up in the bureaucratic hierar￾chy. I will return below to the issue of continuity: this sizable cadre of Japanese-trained Korean bureaucrats virtually took over the day-to-day running of a trun￾cated South Korea, first under US military government and eventually when a sovereign state was formed. One further characteristic of the colonial govem￾ment that needs to be underlined is the successful links that the Japanese created between a highly concen￾trated power center in Seoul, and a densely bureaucra￾tized periphery. All bureaucracies face the problem of how to ensure that central commands are faithfully implemented by the officials at the bottom rung. This, in turn, requires ensuring that lower level officials respond mainly to those above them in the bureau￾cratic hierarchy, rather than to personal interests, or to the interests of societal actors with whom they inter￾act. The Japanese in Korea were quite self-conscious of this problem and repeatedly experimented till they arrived at arrangements deemed satisfactory. Of course, certain circumstances were helpful in establishing authority links between the center and the periphery: ruling arrangements in Seoul were highly authoritarian - the power of the Japanese Governor

1274 WORLD DEVELOPMENT Generals in both policy making and implementation during 1915-20,about 2,000 policemen-or nearly was absolute and nearly all of them were senior mili- one out of every 10 available policemen-were tary men-and Korea was not a very large country sternly disciplined every year for transgression of (again,for example,note the contrast with the role of police rules (Chen,1984,pp.236-239). the British in India).The Japanese,however,took This extensive and closely supervised police force additional actions2 For example,when confronted that penetrated every Korean village,performed with corrupt regional or local officials,the central numerous functions other than "normal"police duties authorities experimented-in line with"new"insti- of law and order maintenance.Powers granted to tutional cconomics -with paying these officials police included control over"politics,education,reli- higher salaries.especially "entertainment-allow- gion,morals,health and public welfare,and tax col- ance,"with the hope that,if more satisfied,they may lection"(Robinson in Eckert et al.,1990,p.259).The perform public tasks better.When this did not work. police,who presented themselves in military uni- or at least not fully,the colonial authorities further forms,again replete with swords.also had summary centralized,leading up to even less discretion,and powers to judge and punish minor offenders,includ- more rule-governed behavior for lower level officials ing the punishment of whipping.Even in production, These officials in the early colonial period were even local police were known to have"compelled villages required to wear crisp uniforms,replete with swords, to switch from existing food crops"to cash crops and so as to distinguish them sharply from an average citi- to adopt"new techniques"in rice production so as to zen -thus creating a state-society,or a public-pri- facilitate exports to Japan.Moreover,during land sur- vate,distinction through the use of symbolic politics veys (conducted during 1910-18;more on this as well as to convey the will of the state in the far below),as a result of which tenancy and conflicts over reaches of the society.When such efforts also failed to land increased,local police "always intervened in secure full compliance.Korean officials would be favor of landlords."23 It is thus not surprising that even replaced by the more socialized and complying a Japanese observer was led to conclude that Terauchi Japanese officials,at least until more suitable (the first Japanese Governor General of Korea.fol- Koreans,who were likely to comply,could be found lowing Ito and formal annexation)and his successors for the job. had transformed the "entire Korean peninsula into a military camp." (ii)The police force One final aspeet of the police role eoncerns the In addition to the civil bureaucracy,the other links between the police and local society via local essential arm of the new Korean state that the elites.The police successfully utilized the proverbial Japanese helped develop was a well-organized police carrot and stick to incorporate "village elders"and force.Once again,there is nothing unique about colo- others into a ruling "alliance."The police thus but- nial powers developing a police force;what is note- tressed their already extensive powers by working worthy here are both the extensive and the intensive with,rather than against,indigenous authority struc- nature of police supervision in colonial Korea.The tures.So armed,the police used the knowledge and colonial police force was designed on the lines of the influence of the local elites to mold the behavior of Meiji police force insofar as it was highly centralized, average citizens in such diverse matters as,"birth con- well disciplined and played an extensive role in social trol.types of crops grown,count and movement of and economic reforms.22 The police force in colonial people,prevention of spread of diseases,mobilization Korea grew rapidly:from some 6.222 gendarmes and of forced labor and to report on transgressions"(Chen. police in 1910 to 20,777 in 1922 and again to over 1984,p.226).The police and many local elites thus 60,000 in 1941 (Robinson in Eckert et al.,1990,p. came to be viewed and despised by Koreans at large as 259).One scholar suggests that at the height of the "collaborationists";unfortunately for Koreans,while colonial rule,there were enough police so that the many of the landed elite were indeed eventually elim- lowest level policeman knew "every man in the vil- inated as a political force (i.e.via land reforms follow- lage"(Chen.1984,p.225).While senior police offi- ing the Korean War),much of the colonial police force cers were normally Japanese,over half the police was incorporated directly into the new state structure force was made up of Koreans,often lower class of South Korea. Koreans.These Koreans were trained by the Japanese In sum,the personalized and factionalized Yi state in police academies,especially established within with a limited reach in society came to be replaced by Korea for the purpose.Records indicate that for every a colonial state with considerable capacity to penetrate Korean police position there were 10-20 applicants and control the society;this state was simultaneously (Chen.1984,p.236),suggesting a level of coopera- oppressive and efficacious.A highly centralised apex tion between Koreans and Japanese that probably with near absolute powers of legislation and exccution pains the modern Korean nationalist sentiments. -and thus of setting and implementing "national" Beyond formal training,the Japanese maintained very goals-and a pervasive,disciplined civil and police close supervision over their police force;for example, bureaucracies constituted the cores of the new state

1214 WORLD DEVELOPMENT Generals in both policy making and implementation was absolute and nearly all of them were senior mili￾tary men - and Korea was not a very large country (again, for example, note the contrast with the role of the British in India). The Japanese, however, took additional actions.” For example, when confronted with corrupt regional or local officials, the central authorities experimented - in line with “new” insti￾tutional economics - with paying these officials higher salaries, especially “entertainment-allow￾ance,” with the hope that, if more satisfied, they may perform public tasks better. When this did not work, or at least not fully, the colonial authorities further centralized, leading up to even less discretion, and more rule-governed behavior for lower level officials. These officials in the early colonial period were even required to wear crisp uniforms, replete with swords, so as to distinguish them sharply from an average citi￾zen - thus creating a state-society, or a public-pri￾vate, distinction through the use of symbolic politics - as well as to convey the will of the state in the far reaches of the society. When such efforts also failed to secure full compliance, Korean officials would be replaced by the more socialized and complying Japanese officials, at least until more suitable Koreans, who were likely to comply, could be found for the job. (ii) The poiicr jbrce In addition to the civil bureaucracy, the other essential arm of the new Korean state that the Japanese helped develop was a well-organized police force. Once again, there is nothing unique about colo￾nial powers developing a police force; what is note￾worthy here are both the extensive and the intensive nature of police supervision in colonial Korea. The colonial police force was designed on the lines of the Meiji police force insofar as it was highly centralized, well disciplined and played an extensive role in social and economic reforms.-?2 The police force in colonial Korea grew rapidly: from some 6,222 gendarmes and police in I9 IO to 20,777 in 1922 and again to over 60,000 in 1941 (Robinson in Eckert et al., 1990, p. 259). One scholar suggests that at the height of the colonial rule, there were enough police so that the lowest level policeman knew “every man in the vil￾lage” (Chen, 1984, p. 225). While senior police offi￾cers were normally Japanese, over half the police force was made up of Koreans, often lower class Koreans. These Koreans were trained by the Japanese in police academies, especially established within Korea for the purpose. Records indicate that for every Korean police position there were 30-20 applicants (Chen, 1984, p. 236), suggesting a level of coopera￾tion between Koreans and Japanese that probably pains the modern Korean nationalist sentiments. Beyond formal training, the Japanese maintained very close supervision over their police force; for example, during 19 15-20, about 2,000 policemen - or nearly one out of every 10 available policemen - were sternly disciplined every year for transgression of police rules (Chen, 1984, pp. 236-239). This extensive and closely supervised police force, that penetrated every Korean village, performed numerous functions other than “normal” police duties of law and order maintenance. Powers granted to police included control over “politics, education, reli￾gion, morals, health and public welfare, and tax col￾lection” (Robinson in Eckert et cd., 1990, p. 259). The police, who presented themselves in military uni￾forms, again replete with swords, also had summary powers to judge and punish minor offenders, includ￾ing the punishment of whipping. Even in production, local police were known to have “compelled villages to switch from existing food crops” to cash crops and to adopt “new techniques” in rice production so as to facilitate exports to Japan. Moreover, during land sur￾veys (conducted during 1910&l 8; more on this below), as a result of which tenancy and conflicts over land increased, local police “always intervened in favor of landlords.“” It is thus not surprising that even a Japanese observer was led to conclude that Terauchi (the first Japanese Governor General of Korea, fol￾lowing Ito and formal annexation) and his successors had transformed the “entire Korean peninsula into a military camp.“‘” One final aspect of the police role concerns the links between the police and local society via local elites. The police successfully utilized the proverbial carrot and stick to incorporate “village elders” and others into a ruling “alliance.” The police thus but￾tressed their already extensive powers by working with, rather than against, indigenous authority struc￾tures. So armed, the police used the knowledge and influence of the local elites to mold the behavior of average citizens in such diverse matters as, “birth con￾trol. types of crops grown, count and movement of people, prevention of spread of diseases, mobilization of forced labor and to report on transgressions” (Chen. 1984, p. 226). The police and many local elites thus came to be viewed and despised by Koreans at large as “collaborationists”; unfortunately for Koreans, while many of the landed elite were indeed eventually elim￾inated as a political force (i.e. via land reforms follow￾ing the Korean War), much of the colonial police force was incorporated directly into the new state structure of South Korea. In sum, the personalized and factionalized Yi state with a limited reach in society came to be replaced by a colonial state with considerable capacity to penetrate and control the society; this state was simultaneously oppressive and efficacious. A highly centralised apex with near absolute powers of legislation and execution - and thus of setting and implementing “national” goals - and a pervasive, disciplined civil and police bureaucracies constituted the cores of the new state

JAPANESE LINEAGE OF KOREA'S"DEVELOPMENTAL STATE" 1275 (iii)The politics of the new state rated into the new system of colonial rule.Second,and The politics practiced by the new rulers added to relatedly,the colonial state forged numerous implicit the state's capacity to convey its will to the society. and explicit "alliances"with Korean propertied Except for a somewhat liberal interlude in the 1920s, classes.The nature of these turned out to be of critical for the most part,the political practices of the long-term significance.While I return to a detailed Japanese colonial state in Korea were brutally author- discussion of this issue below.it should be noted here itarian.For example,Korean newspapers were either that,on the whole,Korean monied groups-in both suspended or heavily censored,political protest was the city and the countryside-did not oppose colonial met with swift retribution,and political organizations rle.Most of them benefited from this rule and gener- and public gatherings were generally banned.Those ally went along-some even with enthusiasm-with professing Korean nationalist sentiments were thus the colonial project.Third,the Japanese undertook either exiled or remained fragmented:while there was considerable expansion of education,facilitating latent and scattered sympathy for nationalists and for propaganda and political resocialization.Whereas in communists all through the colonial period,a coherent 1910 nearly 10,000 students attended some sort of nationalist movement was never allowed to develop.25 school,by 1941 this number was up to 1.7 million and The Japanese used "thought police"to detect and the rate of literacy by 1945 was nearly 50%.The focus eliminate political dissidence,and also developed a was on primary education and the curricula was "spy system"to buttress the civil and police bureau- designed with the "objcct"of raising "practical men cracy that was "probably better developed in Korea able to meet the requirements of the state."27 than anywhere in the world"(Grajdanzev,1944,p. To conclude this subsection,the Japanese colo- 55). nialists in Korca replaced the decrepit Yi statc with a The colonial authorities were quite self-conscious centralized and powerful state.This was no liberal about their use of repression as a means to instill fear state;it was more statist vis-a-vis the Korean society, in the minds of Koreans and thus to minimize dissi- and considerably more repressive,than even the statist dence and reinforce bureaucratic control:in order to and illiberal Japanese political economy of the period. avoid"restlessness"in the "popular mind,"note gov- Central decision making was highly concentrated in ernment reports of the period,it was "essential"to the office of the Governor General.The Governor "maintain unshakable the dignity of the government" General's will,reflecting the imperial design and and"to impress the people with the weight of the new goals,was translated into implemented policies via regime"(Government-General of Chosen,1914,pp. the use of an extensive,well-designed and disciplined 2-3).When Koreans still resisted.Governor General bureaucracy.The new state also achieved consider- Terauchi Masatake supposedly responded,"I will able downward penetration:both the civil and police whip you with scorpions"(quoted in Peattie,1984,p. bureaucracies reached into the nooks and crannies of 18).and when eventually the Koreans succumbed,the the society,while continuing to respond to central gloating satisfaction is also obvious in official docu- directives:Korean elites in the localities were incor- ments:"they have gradually yielded their obstinate porated into the ruling "alliance";and,when all else prejudices and their disdainful attitude” failed in the Japanese efforts to control and transform (Government-General of Chosen,1935,p.81). Korea,there existed a well-functioning intelligence In spite of the awesome state that the Japanese cre- service to buttress the state's supervisory role.While a ated,it would be a mistake to believe that a thorough fuller understanding of how power was generated in bureaucratic penetration and politics of fear were the this system.and the uses to which it was put,will only ruling instruments in the hands of the colonial- emerge in due course,it should already be evident as ists.There is no doubt that bureaucratic growth to how the precolonial,ineffective state was trans- enabled the new state to undertake many more eco- formed into a state that-for better or for worse- nomic activities that contributed to economic growth could get things done. (more on this below),and that repression enabled the establishment of order,freeing the state elite to focus on other "developmental"matters.Nevertheless, 3.THE COLONIAL STATE,PROPERTIED bureaucratic and repressive power are seldom enough CLASSES AND ECONOMIC CHANGE to elicit a measure of cooperation--from,at least. some groups in society-that is essential for gener- The colonial state in Korea was a busy state.While ating economic dynamism.We must thus also take pursuing the imperial interests of Japan,it evolved a note of some other,nonrepressive ruling instruments full policy agenda,including the goal of economic that the new colonial state put to use. transformation of Korea.The broad strategy of trans- First,a segment of the Korean political elite in the formation was two pronged:the state utilized its precolonial period was quite favourably inclined bureaucratic capacities to directly undertake quitc a toward Japan.26 These Koreans from the political few economic tasks;and,more important,the state class were both officially and unofficially incorpo- involved propertied groups-both in the countryside

JAPANESE LINEAGE OF KOREA’S “DEVELOPMENTAL STATE” 1215 (iii) The politics ofthe new state The politics practiced by the new rulers added to the state’s capacity to convey its will to the society. Except for a somewhat liberal interlude in the 1920s for the most part, the political practices of the Japanese colonial state in Korea were brutally author￾itarian. For example, Korean newspapers were either suspended or heavily censored, political protest was met with swift retribution, and political organizations and public gatherings were generally banned. Those professing Korean nationalist sentiments were thus either exiled or remained fragmented; while there was latent and scattered sympathy for nationalists and for communists all through the colonial period, a coherent nationalist movement was never allowed to develop.25 The Japanese used “thought police” to detect and eliminate political dissidence, and also developed a “spy system” to buttress the civil and police bureau￾cracy that was “probably better developed in Korea than anywhere in the world” (Grajdanzev, 1944, p. 55). The colonial authorities were quite self-conscious about their use of repression as a means to instill fear in the minds of Koreans and thus to minimize dissi￾dence and reinforce bureaucratic control: in order to avoid “restlessness” in the “popular mind,” note gov￾ernment reports of the period, it was “essential” to “maintain unshakable the dignity of the government” and “to impress the people with the weight of the new regime” (Government-General of Chosen, 1914, pp. 2-3). When Koreans still resisted, Governor General Terauchi Masatake supposedly responded, “I will whip you with scorpions” (quoted in Peattie, 1984, p. 18), and when eventually the Koreans succumbed, the gloating satisfaction is also obvious in official docu￾ments: “they have gradually yielded their obstinate prejudices and their disdainful attitude” (Government-General of Chosen, 1935, p. 8 1). In spite of the awesome state that the Japanese cre￾ated, it would be a mistake to believe that a thorough bureaucratic penetration and politics of fear were the only ruling instruments in the hands of the colonial￾ists. There is no doubt that bureaucratic growth enabled the new state to undertake many more eco￾nomic activities that contributed to economic growth (more on this below), and that repression enabled the establishment of order, freeing the state elite to focus on other “developmental” matters. Nevertheless, bureaucratic and repressive power are seldom enough to elicit a measure of cooperation - from, at least, some groups in society -that is essential for gener￾ating economic dynamism. We must thus also take note of some other, nonrepressive ruling instruments that the new colonial state put to use. First, a segment of the Korean political elite in the precolonial period was quite favourably inclined toward Japan.26 These Koreans from the political class were both officially and unofficially incorpo￾rated into the new system of colonial rule. Second, and relatedly, the colonial state forged numerous implicit and explicit “alliances” with Korean propertied classes. The nature of these turned out to be of critical long-term significance. While I return to a detailed discussion of this issue below, it should be noted here that, on the whole, Korean monied groups - in both the city and the countryside-did not oppose colonial rule. Most of them benefited from this rule and gener￾ally went along - some even with enthusiasm-with the colonial project. Third, the Japanese undertook considerable expansion of education, facilitating propaganda and political resocialization. Whereas in 1910 nearly 10,000 students attended some sort of school, by 1941 this number was up to 1.7 million and the rate of literacy by 1945 was nearly 50%. The focus was on primary education and the curricula was designed with the “object” of raising “practical men able to meet the requirements of the state.” 27 To conclude this subsection, the Japanese colo￾nialists in Korea replaced the decrepit Yi state with a centralized and powerful state. This was no liberal state; it was more statist vis-&vis the Korean society, and considerably more repressive, than even the statist and illiberal Japanese political economy of the period. Central decision making was highly concentrated in the office of the Governor General. The Governor General’s will, reflecting the imperial design and goals, was translated into implemented policies via the use of an extensive, well-designed and disciplined bureaucracy. The new state also achieved consider￾able downward penetration: both the civil and police bureaucracies reached into the nooks and crannies of the society, while continuing to respond to central directives: Korean elites in the localities were incor￾porated into the ruling “alliance”; and, when all else failed in the Japanese efforts to control and transform Korea, there existed a well-functioning intelligence service to buttress the state’s supervisory role. While a fuller understanding of how power was generated in this system, and the uses to which it was put, will emerge in due course, it should already be evident as to how the precolonial, ineffective state was trans￾formed into a state that - for better or for worse - could get things done. 3. THE COLONIAL STATE, PROPERTIED CLASSES AND ECONOMIC CHANGE The colonial state in Korea was a busy state. While pursuing the imperial interests of Japan, it evolved a full policy agenda, including the goal of economic transformation of Korea. The broad strategy of trans￾formation was two pronged: the state utilized its bureaucratic capacities to directly undertake quite a few economic tasks; and, more important, the state involved propertied groups -both in the countryside

1276 WORLD DEVELOPMENT and in the cities,and both Japanese and Koreans-in pressure was maintained on respective Governor- production-oriented alliances leading up to sustained Generals to boost the cost-efficiency of public ser- economic change.The results measured by the criteria vices.Unlike many other governments.the colonial of growth and industrialization(though not by such state in Korea did not operate with a"soft budget con- other criteria as human rights,national self-determi- straint."On the contrary,there was consistent pressure nation and fair economic distribution)were a consid- to economize,"hardening"the budget constraint,with erable success.Since successes generally begets emu- significant,positive "trickle-down"effect on the effi- lation and continuity,it is important to analyze the ciency of the bureaucracy,including the economic colonial economic strategy. bureaucracy. Two general observations ought to be noted at the outset.First,while the Governor-General in Korea possessed near absolute powers.he was nevertheless (a)Increased state capacity an agent of the Japanese imperial government.The colonial state in Korea thus pursued,not Korean,but The increased capacity of the new colonial state in Japanese needs and interests that changed over time.28 Korea to directly undertake economic tasks is evident In broad brush strokes,during the early phase,say,the fairly early in the historical record.For example,there first decade of the colonial rule.Japan treated Korea was the issue of state capacity to collect taxes.The old mainly as a strategic gain that could also be exploited Yi state,one may recollect,proved quite incapable of in a fairly classic fashion:exchange of agricultural extracting taxes from society,especially revenues products for manufactured goods.Subsequently,as from landowners.The contrasting performance of the Japanese demand for food outpaced its own supply, colonial state is notable.Land revenue in 1905.the the colonial state aggressively undertook measures to year the Japanese influence in Korea started to grow. increase food production in Korea.Manufacturing was some 4.9 million yen:by 1908.this had jumped to was discouraged in this carly phase,again in a fairly 6.5 million yen.or a real increase of some 30%in classic fashion,to protect Japanese exports to Korea. three years.3 Subsequently,numerous other sources Following WWI,however,with swollen company of revenue were added to that obtained from land- profits,Japan sought opportunities for export of capi- e.g.,railways,post office and customs:and receipts tal and thus relaxed restrictions against production of from the ginseng monopoly and from such public manufactured products in Korea.At the same time. undertakings as salt manufacture,coal mines.timber following the need to coopt nationalistic pressures work and printing bureaus-and the jump in revenue within Korea,the colonial state also involved selected intake was phenomenal.Whereas the total revenue in and prominent Korean businessmen in the growth of 1905 (land and other revenues)was 7.3 million yen. manufacturing.Aggressive industralization of Korea by 191I one year after formal annexation.the total occurred only in the 1930s.This was in part a result of revenue intake was 24 million yen.or an increase of Japan's strategy to cope with the depression-i.e.to more than 300%.32 The factors that help explain this create a protected,high-growth economy on an increased state capacity were two-fold.First.the colo- empire-wide scale-and in part a result of Japan's nial state.backed by superior coercive power.snapped aggressive industrialization,again on an empire-wide the stranglehold landowning groups had on the Yi scale,that reflected national power considerations.2 state.pensioning off the Yangban elite,and replacing Japan was able to switch its imperial policies in Korea them by Japanese career bureaucrats:I will return to frequently and decisively;this,in turn,underlined the this issue below.Second.the colonial elite utilized the highly centralized nature of authority within the newly created civil and police bureaucracy to collect Japanese controlled Korean state. taxes.More specifically.as early as 1906.36 revenue The second related observation concerns the collection officers.again replete with uniforms and pressures on the Governor-General in Korea to simul- swords.were posted all over Korea to identify cult- taneously pursue imperial interests and run a vated land,owners of the land.and the revenue due cost-effective government.Reading through from the land (H.I.J.M.'s Residency General,1908. historical documents of the time,especially the annual Chapter V).While the rate of taxation on land was not reports of the Governor-General in Korea,it becomes increased,it was regularized.In addition.uniformed clear that,among their various achievements.the revenue officers worked in conjunction with local colonial authorities in Korea wanted to emphasize police officers in the process of tax collection.lest any their repeated efforts to enhance revenues and to one forget this newly established separation of state minimize expenditures.especially by rationalizing and society,or the willful presence of the new state in the bureaucracy.30 Since any shortfall between rev- society. enues and expenditures within Korea had to be The successful land survey that the Japanese con- financed by the Japanese Imperial government- ducted in Korea during 1910-18 similarly highlighted and typically,there was a net revenue inflow from the efficacy of the new state.The Yi state had repeat- Japan to Korea-one presumes that constant and firm edly discussed such a comprehensive land survey but

1276 WORLD DEVELOPMENT and in the cities, and both Japanese and Koreans - in production-oriented alliances leading up to sustained economic change. The results measured by the criteria of growth and industrialization (though not by such other criteria as human rights, national self-determi￾nation and fair economic distribution) were a consid￾erable success. Since successes generally begets emu￾lation and continuity, it is important to analyze the colonial economic strategy. Two general observations ought to be noted at the outset. First, while the Governor-General in Korea possessed near absolute powers, he was nevertheless an agent of the Japanese imperial government. The colonial state in Korea thus pursued, not Korean, but Japanese needs and interests that changed over time.?” In broad brush strokes, during the early phase, say, the first decade of the colonial rule, Japan treated Korea mainly as a strategic gain that could also be exploited in a fairly classic fashion: exchange of agricultural products for manufactured goods. Subsequently, as Japanese demand for food outpaced its own supply, the colonial state aggressively undertook measures to increase food production in Korea. Manufacturing was discouraged in this early phase, again in a fairly classic fashion, to protect Japanese exports to Korea. Following WWI, however, with swollen company profits, Japan sought opportunities for export of capi￾tal and thus relaxed restrictions against production of manufactured products in Korea. At the same time, following the need to coopt nationalistic pressures within Korea, the colonial state also involved selected and prominent Korean businessmen in the growth of manufacturing. Aggressive industrialization of Korea occurred only in the 1930s. This was in part a result of Japan’s strategy to cope with the depression - i.e. to create a protected, high-growth economy on an empire-wide scale - and in part a result of Japan’s aggressive industrialization, again on an empire-wide scale, that reflected national power considerations.‘” Japan was able to switch its imperial policies in Korea frequently and decisively; this, in turn, underlined the highly centralized nature of authority within the Japanese controlled Korean state. The second related observation concerns the pressures on the Governor-General in Korea to simul￾taneously pursue imperial interests and run a cost-effective government. Reading through historical documents of the time, especially the annual reports of the Governor-General in Korea, it becomes clear that, among their various achievements, the colonial authorities in Korea wanted to emphasize their repeated efforts to enhance revenues and to minimize expenditures, especially by rationalizing the bureaucracy.30 Since any shortfall between rev￾enues and expenditures within Korea had to be financed by the Japanese Imperial government - and typically, there was a net revenue inflow from Japan to Korea-one presumes that constant and firm pressure was maintained on respective Governor￾Generals to boost the cost-efficiency of public ser￾vices. Unlike many other governments. the colonial state in Korea did not operate with a “soft budget con￾straint.” On the contrary, there was consistent pressure to economize, “hardening” the budget constraint, with significant. positive “trickle-down” effect on the effi￾ciency of the bureaucracy, including the economic bureaucracy. (a) Incrensed state capnci9 The increased capacity of the new colonial state in Korea to directly undertake economic tasks is evident fairly early in the historical record. For example, there was the issue of state capacity to collect taxes. The old Yi state, one may recollect, proved quite incapable of extracting taxes from society, especially revenues from landowners. The contrasting performance of the colonial state is notable. Land revenue in 1905. the year the Japanese influence in Korea started to grow. was some 4.9 million yen; by 1908, this had jumped to 6.5 million yen, or a real increase of some 30% in three years.” Subsequently, numerous other sources of revenue were added to that obtained from land - e.g., railways, post office and customs; and receipts from the ginseng monopoly and from such public undertakings as salt manufacture, coal mines. timber work and printing bureaus - and the jump in revenue intake was phenomenal. Whereas the total revenue in 1905 (land and other revenues) was 7.3 million yen, by 191 1 one year after formal annexation, the total revenue intake was 24 million yen. or an increase of more than 300%.3’ The factors that help explain this increased state capacity were two-fold. First, the colo￾nial state. backed by superior coercive power, snapped the stranglehold landowning groups had on the Yi state, pensioning off the Yangban elite, and replacing them by Japanese career bureaucrats: I will return to this issue below. Second. the colonial elite utilized the newly created civil and police bureaucracy to collect taxes. More specifically. as early as 1906, 36 revenue collection officers, again replete with uniforms and swords. were posted all over Korea to identify culti￾vated land, owners of the land, and the revenue due from the land (H.I.J.M.‘s Residency General, 1908. Chapter V). While the rate oftaxation on land was not increased, it was regularized. In addition. uniformed revenue officers worked in conjunction with local police officers in the process of tax collection, lest any one forget this newly established separation of state and society, or the willful presence of the new state in society. The successful land survey that the Japanese con￾ducted in Korea during I9 IO- I8 similarly highlighted the efficacy of the new state. The Yi state had repeat￾edly discussed such a comprehensive land survey but

JAPANESE LINEAGE OF KOREA'S"DEVELOPMENTAL STATE 1277 never carried it out:the bureaucratic capacity was corresponding sums of deposits being 120,000 yen in absent,as was the power to confront land controlling 1909 and 981.000 yen in 1913)(Government-General groups who wanted to hide the extent of their taxable of Chosen,1914,p.19).Later during the colonial rule, lands.By contrast.the colonial state made an exhaus- the Government-General required Koreans to buy tive land survey a priority.Over a period of eight years government bonds that helped finance the industrial- the Japanese invested some 30 million yen in the pro- ization drive of the 1930s.While capital inflows from ject (compared,say,to the total revenue intake of the Japan remained the dominant source,local capital Government-General in 1911 of 24 million yen).The accumulation also increased considerably.The colo- survey "mapped all plots of land.classified it accord- nial state in Korea.even more than the Japanese Meiji ing to type,graded its productivity and established state on which it was modeled,became heavily and ownership."While Japanese civil servants super- directly involved in economic tasks.and judged vised the entire project.Korean landowners cooper- strictly by economic criteria,performed these tasks ated and eventually benefited;local land investigation rather effectively. committees,for example,who were responsible for More significant than the state's direct economic investigating the "ownership,location,boundaries role was the indirect role that led up to the involve- and class of land"were composed of "land-owners ment of wealthy groups in productive activities.The themselves"(Government-General of Chosen,1912, mechanics of how these state-private sector alliances p.13).As a result of the survey,the colonial state were created are important because similar arrange secured a revenue base and,less obviously,enhanced ments were later central to South Korea's phenomenal its control over the Korean agrarian sector by involv- economic success.The dynamics of change in both ing the landowning classes as ruling partners.What the agrarian and industrial sectors thus deserve our the Korean landlords lost in terms of autonomy from, attention. and influence over,the traditional Yi state,they made up by.first,securing new,Western-style,legal private property rights and later (as discussed below),by (b)The state and the agrarian sector enhanced profits from land(Robinson in Eckert et al., 1990,pp.266-267). The colonial state restructured its relationship with Over time.the colonial state in Korea undertook the Korean landed classes.The highest Yangban elite numerous other projects of economic value.This is no who held offices in the Yi state were pensioned off place for a comprehensive discussion;I simply wish (Government-General of Chosen,1911,pp.18-19). to flag some of the main areas.34 First,the Govern- As career bureaucrats took over official functions,the ment-General invested heavily in infrastructure,so direct control of landed classes on the state weakened. much so that Korea's roads and railways were among The successful land survey further confirmed the the finest that a developing country inherited from supremacy of the new state because,as a result of it. their colonial past.Second,as mentioned above,the the capacity of the landed classes to evade the reach of Japanese made significant investments in Korea in the state had shrunk.In return,however,the state primary education.Given the long gestation period, offered the landowners plenty,so as to not only not however,the returns on this investment were probably alienate them,but to make them active partners in exe- reaped,less by colonial Korea,but more by the two cuting the state's goals.For example,the Japanese sovereign Koreas who inherited a relatively literate introduced a new legal code-based on the Meiji labor force.Third,the colonial government ran a num- legal code-that created Western-style legal private ber of economic enterprises directly:e.g.,railways. property,thus securing the control of Korean landed communications.opium,salt and tobacco.Judged by groups over land in perpetuity.While the Japanese in the regular financial contribution that these public the process ended up owning significant amount of undertakings made to public revenues,they were run agricultural land in Korea,most Koreans who con- relatively efficiently.Finally,the Government- trolled land prior to the arrival of the Japanese main- General played an important role in the overall tained,nay,even expanded their land ownership.35 process of capital accumulation.While I will return to Moreover,as mentioned above,many among the this issue again below,and the direct role of the new landed elite were incorporated into local governance, colonial state in extracting taxes has already been cooperating with and helping local agents of the state noted,a few other points also deserve attention.The maintain control over villages.While students of colo- currency and banking reforms that the new colonial nialism often distinguish direct and indirect colonial state undertook rather early led to a significant jump in rule,the Japanese political arrangement in Korea uti- private,institutional savings:e.g.,deposits in the lized both forms:direct bureaucratic penetration was Bank of Chosen (Korea)doubled from some 18 mil- buttressed by the authority of local influentials.This lion yen in 1911 to 37 million yen in 1913 and the arrangement also suggests that,contrary to some number of depositors in the postal savings bank went recent arguments,the presence of a landowning class up from about 20,000 in 1909 to 420.000 in 1913 (the does not necessarily inhibit the formation of a power-

JAPANESE LINEAGE OF KOREA’S “DEVELOPMENTAL STATE” 1277 never carried it out; the bureaucratic capacity was absent, as was the power to confront land controlling groups who wanted to hide the extent of their taxable lands. By contrast. the colonial state made an exhaus￾tive land survey a priority. Over a period of eight years the Japanese invested some 30 million yen in the pro￾ject (compared, say, to the total revenue intake of the Government-General in 1911 of 24 million yen). The survey “mapped all plots of land, classified it accord￾ing to type, graded its productivity and established ownership.“?? While Japanese civil servants super￾vised the entire project, Korean landowners cooper￾ated and eventually benefited; local land investigation committees, for example, who were responsible for investigating the “ownership, location, boundaries and class of land” were composed of “land-owners themselves” (Government-General of Chosen, 1912, p. 13). As a result of the survey, the colonial state secured a revenue base and, less obviously, enhanced its control over the Korean agrarian sector by involv￾ing the landowning classes as ruling partners. What the Korean landlords lost in terms of autonomy from, and influence over, the traditional Yi state, they made up by, first, securing new, Western-style, legal private property rights and later (as discussed below), by enhanced profits from land (Robinson in Eckert er al., 1990, pp. 266-267). Over time, the colonial state in Korea undertook numerous other projects of economic value. This is no place for a comprehensive discussion; I simply wish to flag some of the main areas.34 First, the Govem￾ment-General invested heavily in infrastructure, so much so that Korea’s roads and railways were among the finest that a developing country inherited from their colonial past. Second, as mentioned above, the Japanese made significant investments in Korea in primary education. Given the long gestation period, however, the returns on this investment were probably reaped, less by colonial Korea, but more by the two sovereign Koreas who inherited a relatively literate labor force. Third, the colonial government ran a num￾ber of economic enterprises directly: e.g., railways, communications, opium, salt and tobacco. Judged by the regular financial contribution that these public undertakings made to public revenues, they were run relatively efficiently. Finally, the Government￾General played an important role in the overall process of capital accumulation. While I will return to this issue again below, and the direct role of the new colonial state in extracting taxes has already been noted, a few other points also deserve attention. The currency and banking reforms that the new colonial state undertook rather early led to a significant jump in private, institutional savings: e.g., deposits in the Bank of Chosen (Korea) doubled from some 18 mil￾lion yen in 1911 to 37 million yen in 1913 and the number of depositors in the postal savings bank went up from about 20,000 in 1909 to 420,000 in 19 13 (the corresponding sums of deposits being 120,000 yen in 1909 and 98 1,000 yen in 19 13) (Government-General of Chosen, 19 14, p. 19). Later during the colonial rule, the Government-General required Koreans to buy government bonds that helped finance the industrial￾ization drive of the 1930s. While capital inflows from Japan remained the dominant source, local capital accumulation also increased considerably. The colo￾nial state in Korea, even more than the Japanese Meiji state on which it was modeled, became heavily and directly involved in economic tasks, and judged strictly by economic criteria, performed these tasks rather effectively. More significant than the state’s direct economic role was the indirect role that led up to the involve￾ment of wealthy groups in productive activities. The mechanics of how these state-private sector alliances were created are important because similar arrange￾ments were later central to South Korea’s phenomenal economic success. The dynamics of change in both the agrarian and industrial sectors thus deserve our attention. (b) The state and the agrarian sector The colonial state restructured its relationship with the Korean landed classes. The highest Yangban elite who held offices in the Yi state were pensioned off (Government-General of Chosen, 1911, pp. 18-19). As career bureaucrats took over official functions, the direct control of landed classes on the state weakened. The successful land survey further confirmed the supremacy of the new state because, as a result of it, the capacity of the landed classes to evade the reach of the state had shrunk. In return, however, the state offered the landowners plenty, so as to not only not alienate them, but to make them active partners in exe￾cuting the state’s goals. For example, the Japanese introduced a new legal code - based on the Meiji legal code - that created Western-style legal private property, thus securing the control of Korean landed groups over land in perpetuity. While the Japanese in the process ended up owning significant amount of agricultural land in Korea, most Koreans who con￾trolled land prior to the arrival of the Japanese main￾tained, nay, even expanded their land ownership.35 Moreover, as mentioned above, many among the landed elite were incorporated into local governance, cooperating with and helping local agents of the state maintain control over villages. While students of colo￾nialism often distinguish direct and indirect colonial rule, the Japanese political arrangement in Korea uti￾lized both forms: direct bureaucratic penetration was buttressed by the authority of local influentials. This arrangement also suggests that, contrary to some recent arguments, the presence of a landowning class does not necessarily inhibit the formation of a power-

1278 WORLD DEVELOPMENT ful "developmental"state;much depends on the spe- grains as a whole declined substantially after the early cific relationship of the state and landowners.3 years of the colonial period."The same author points The Japanese colonial government periodically out that this disjuncture between production and con- made significant efforts to boost agricultural produc- sumption was a result of several causes,but mainly tion,especially Korea's main product,rice.The due to a combination of population growth and few underlying motivation was changing Japanese eco- nonagricultural opportunities that increased the bur- nomic needs:e.g.,prior to 1919,the efforts to boost den on tenants and on small farmers (Suh,1978.pp production were minimal.Following a rice shortage 86-87).If there was steady growth in production but and related riots in Japan in 1918,a major plan to the consumption for the majority of the population expand rice production in Korea was implemented. declined,given the considerable inequality in land The success on this front contributed to "over produc- ownership,it is likely that the incomes of landowning tion"and following a glut and pressures from groups mushroomed.Other available evidence is con- Japanese rice producers,all plans to increase rice pro- sistent with this proposition:the rates of return on duction were cancelled in 1933.Again,however,the agricultural investment were very high for most of the war with China in 1938-39 created food shortages in period;income inequalities widened:and,as noted Japan and Korea was"resuscitated as a granary of the above,there was rapid growth of small depositors in Empire."37 saving institutions.The general point is that Korean During the early phase the Japanese focused their landowning groups did rather well under colonial gov- efforts on land improvement,especially on irrigation, ernment;they became part of an implicit but comfort- drainage and reclamation of arable land.The resulting able ruling alliance. increase in production was not huge and resulted both Three other characteristics of the changing agrar from extensive and intensive efforts;e.g..increase in ian sector are noteworthy.First,Japanese corporations rice production during 1910-24 averaged around and entrepreneurs ended up owning large tracts of 1.5%per annum and land productivity in the samc Korcan agricultural land-anywherc from onc-quar- period improved at about 0.8%per annum (Suh,1978, ter to one-third of all the arable land.This was a result p.73,Table 33).Subsequently,when rapid increase in of a conscious government policy that began with the rice production becamc a goal,Korea's Japanese hope of attracting Japancse immigrants to Korea,but rulers utilized the knowledge acquired during the when that goal met with only limited success. Meiji transformation and concentrated their efforts on Japanese corporations became heavily involved. spreading the use of improved seeds,fertilizer and Especially significant as a landowner was the infa- irrigation.The gains were significant:the percentage mous Oriental Development Company (e.g.,see of paddy land using improved seed doubled during Moskowitz.1974),which,like most other Japanese 1915-40,reaching 85%;fertilizer input expanded 10 landowners,leased lands to tenants,collected rents in times during the same period(Suh,1978.p.77,Table kind,most often rice,and sold the rice in the export 34):and during 1919-38 land under irrigation market back to Japan.The rate of return on such activ- increased annually by nearly 10%(Suh,1978,p.73, ities was high,higher than in Japan,and many a for- Table 33;and Ishikawa,1967,pp.84-109).As a tunes were made (Suh,1978,p.85,Table 39).From result,rice production during 1920-35 grew at nearly our standpoint,the direct involvement of the Japanese 3%per annum and nearly two-thirds of this growth in Korean agriculture helps explicate two points:the resulted from improvements in land productivity.3 mechanics of how the more advanced techniques of The overall rate of increase in rice production per unit agricultural production may have been transferred of land for the colonial period (1910-40)averaged a from Japan to Korea;and the mechanics underlying respectable 2%per annum(compare this,for example. "forced exports,"whereby Japanese landowners sold with India's post-Green Revolution-say,1970 to rice grown in Korea back to Japan directly. present-rates of productivity increase in cereal pro- A second characteristic of the changing agrarian duction:they have been only a little higher than 2 sector was its heavy export orientation.For example. per annum).While some of these improvements may while total Korean rice production during the colonial have been a"spontaneous response to food shortages period nearly doubled,rice exports to Japan during the and higher prices in Japan,it is nevertheless difficult same period increased six times (Suh,1978,p.92. to imagine a relatively quick increase in supply with- Table 43).In addition.while the overall economy of out significant public efforts,especially in providing the Japanese empire was protected,trading within the new seeds and in facilitating the spread of fertilizer. empire was relatively free of tariffs and other restric- It is a sad fact that increases in production in Korea tions.Rapid growth of exports to the metropole with a did not lead to improvement in food consumption. more advanced agricultural sector thus points to an Bulk of the increased production ended up in the additional source-the quintessential source of com- export market and imported goods did not become petition-that must have also contributed to sus- consumption items for the vast majority.As a well tained improvements in agricultural productivity. documented study concludes,"per capita use of food Finally,the geography of the changing agrarian scene

1278 WORLD DEVELOPMENT ful “developmental” state; much depends on the spe￾cific relationship of the state and landowners.‘h The Japanese colonial government periodically made significant efforts to boost agricultural produc￾tion, especially Korea’s main product, rice. The underlying motivation was changing Japanese eco￾nomic needs: e.g., prior to 1919, the efforts to boost production were minimal. Following a rice shortage and related riots in Japan in 1918, a major plan to expand rice production in Korea was implemented. The success on this front contributed to “over produc￾tion” and following a glut and pressures from Japanese rice producers, all plans to increase rice pro￾duction were cancelled in 1933. Again, however, the war with China in 1938-39 created food shortages in Japan and Korea was “resuscitated as a granary of the Empire.“37 During the early phase the Japanese focused their efforts on land improvement, especially on irrigation, drainage and reclamation of arable land. The resulting increase in production was not huge and resulted both from extensive and intensive efforts; e.g., increase in rice production during 1910-24 averaged around 1.5% per annum and land productivity in the same period improved at about 0.8% per annum (Suh, 1978, p. 73, Table 33). Subsequently, when rapid increase in rice production became a goal, Korea’s Japanese rulers utilized the knowledge acquired during the Meiji transformation and concentrated their efforts on spreading the use of improved seeds, fertilizer and irrigation. The gains were significant: the percentage of paddy land using improved seed doubled during 19 15a0, reaching 85%; fertilizer input expanded 10 times during the same period (Suh, 1978, p. 77, Table 34); and during 1919-38 land under irrigation increased annually by nearly 10% (Suh, 1978, p. 73, Table 33; and Ishikawa, 1967, pp. 84-109). As a result, rice production during 1920-35 grew at nearly 3% per annum and nearly two-thirds of this growth resulted from improvements in land productivity.‘x The overall rate of increase in rice production per unit of land for the colonial period (1910-40) averaged a respectable 2% per annum (compare this, for example, with India’s post-Green Revolution - say, 1970 to present-rates of productivity increase in cereal pro￾duction; they have been only a little higher than 2% per annum). While some of these improvements may have been a “spontaneous” response to food shortages and higher prices in Japan, it is nevertheless difficult to imagine a relatively quick increase in supply with￾out significant public efforts, especially in providing new seeds and in facilitating the spread of fertilizer. It is a sad fact that increases in production in Korea did not lead to improvement in food consumption. Bulk of the increased production ended up in the export market and imported goods did not become consumption items for the vast majority. As a well documented study concludes, “per capita use of food grains as a whole declined substantially after the early years of the colonial period.” The same author points out that this disjuncture between production and con￾sumption was a result of several causes, but mainly due to a combination of population growth and few nonagricultural opportunities that increased the bur￾den on tenants and on small farmers (Sub, 1978, pp. 86-87). If there was steady growth in production but the consumption for the majority of the population declined, given the considerable inequality in land ownership, it is likely that the incomes of landowning groups mushroomed. Other available evidence is con￾sistent with this proposition: the rates of return on agricultural investment were very high for most of the period; income inequalities widened; and, as noted above, there was rapid growth of small depositors in saving institutions. The general point is that Korean landowning groups did rather well under colonial gov￾ernment; they became part of an implicit but comfort￾able ruling alliance. Three other characteristics of the changing agrar￾ian sector are noteworthy. First, Japanese corporations and entrepreneurs ended up owning large tracts of Korean agricultural land - anywhere from one-quar￾ter to one-third of all the arable land. This was a result of a conscious government policy that began with the hope of attracting Japanese immigrants to Korea, but when that goal met with only limited success. Japanese corporations became heavily involved. Especially significant as a landowner was the infa￾mous Oriental Development Company (e.g., see Moskowitz, 1974), which, like most other Japanese landowners, leased lands to tenants, collected rents in kind, most often rice, and sold the rice in the export market back to Japan. The rate of return on such activ￾ities was high, higher than in Japan, and many a for￾tunes were made (Suh, 1978, p. 85, Table 39). From our standpoint, the direct involvement of the Japanese in Korean agriculture helps explicate two points: the mechanics of how the more advanced techniques of agricultural production may have been transferred from Japan to Korea: and the mechanics underlying “forced exports,” whereby Japanese landowners sold rice grown in Korea back to Japan directly. A second characteristic of the changing agrarian sector was its heavy export orientation. For example, while total Korean rice production during the colonial period nearly doubled, rice exports to Japan during the same period increased six times (Suh, 1978, p. 92, Table 43). In addition. while the overall economy of the Japanese empire was protected, trading within the empire was relatively free of tariffs and other restric￾tions. Rapid growth of exports to the metropole with a more advanced agricultural sector thus points to an additional source -the quintessential source of com￾petition - that must have also contributed to sus￾tained improvements in agricultural productivity. Finally, the geography of the changing agrarian scene