Liberal Peace and the Scramble for the Rocks:

The Dokdo/Takeshima, Senkaku/Diaoyu, and Paracel and Spratly Islands Disputes

Min Gyo Koo

minkoo@usc.edu

Postdoctoral Fellow

Center for International Studies

University of Southern California

Abstract: This paper explores the three most prominent offshore island disputes in East

Asia: the Dokdo/Takeshima dispute between South Korea and Japan, the Senkaku/Diaoyu

dispute between Japan and China, and the Paracel and Spratly dispute between China and

Vietnam. In the context of rising economic interdependence in East Asia, this paper tests

the liberal peace hypothesis that economic interdependence fosters peaceful relations by

giving countries an economic incentive to avoid costly disputes. The OLS regression

results are mixed, but favor the pacific benefits of economic interdependence. In the

Senkaku/Diaoyu dispute, a realist claim finds support for the theory that interdependence

fosters conflict rather than peace. Yet in the cases of the Dokdo/Takeshima and the

Paracel and Spratly disputes, the liberal peace argument is supported with evidence that

higher levels of trade dependence of one party on the other alleviate the degree of hostility.

In view of the rising pressure of territorial nationalism and fluid geopolitics in a post-

September world, even the mere containment of the territorial disputes in East Asia requires

the skillful diplomacy and tireless efforts of political leaders. The findings of this paper

indicate that one of the best strategies to hedge against potentially disruptive behaviors is to

engage more with each other by facilitating greater economic interdependence. In

conclusion, the path to deeper and wider economic interdependence in East Asia is likely to

be a bumpy one, but the institutionalization of economic integration may help to smooth it.

An earlier version of this paper was presented at the 2005 Annual Meeting of the American

Political Science Association, Washington, D.C., September 3, 2005.

1

I. INTRODUCTION

East Asia is home to many of the world’s most vexing territorial disputes.1 The territories

in dispute need not cover the entire soil of a particular country, as in the cases of the two

Chinas and the two Koreas, in order to seriously strain interstate relationships. Even small,

barely habitable offshore islands can serve as the most persistent and explosive bone of

contention. Examples of unresolved island disputes include competing sovereignty claims

to the Dokdo/Takeshima in the East Sea/Sea of Japan, the Senkaku/Diaoyu in the East

China Sea, the Northern Territories/Kuriles in the Northwestern Pacific Ocean, the islands

of Sipadan, Sebatik, and Ligitan in the Celebes Sea, and the Paracels and Spratlys in the

South China Sea.

The energy hungry coastal countries in the region all eye the high potentials of oil

and gas deposits near disputed islands. They also rely on contested offshore areas to

provide a large portion of their marine diets. As has been reported widely, the danger of

conflict escalation looms particularly large during a global shortage of energy and marine

resources. Yet material concerns are hardly the sole drivers of the island disputes in

contemporary East Asia. Most of them also raise questions of national identity and pride,

thereby feeding territorial nationalism following the end of ideology.

In the postwar period, most notably since China adopted its reform policy in 1978,

1 There are at least thirty five territorial dyads in the region contiguous on land or within 400 nautical

miles (the sum of two hypothetical countries’ 200 nautical mile exclusive economic zones) of water

between their undisputed land territories. Most of these dyads have outstanding territorial

disagreements with each other: China-Taiwan, China-Japan, China-Vietnam, China-Philippines,

North Korea-South Korea, South Korea-Japan, Japan-Russia, Philippines-Taiwan, Thailand-

Myanmar, Thailand-Cambodia, and Cambodia-Vietnam, among others.

2

the economic integration of East Asian countries has been truly impressive, as manifested by

soaring intra-regional flows of trade, capital, long-term investments, and technology. Such

a rapid economic integration has led to a burgeoning scholarly interest in the “liberal peace”

debate about whether or not economic interdependence fosters peaceful relations by giving

countries an economic incentive to avoid costly disputes. Liberals argue that the territorial

disputes in East Asia remain a potential source of conflict, but their salience has been in

steady decline due to increasing economic interdependence. By contrast, realists remain

skeptical about the mitigating influence of economic interdependence.

In the context of rising economic interdependence in East Asia, this paper

investigates why and to what extent a given dispute becomes more/less heated over time,

focusing on the three most prominent cases: the Dokdo/Takeshima dispute between South

Korea and Japan, the Senkaku/Diaoyu dispute between Japan and China, and the Paracel

and Spratly dispute between China and Vietnam. These cases are particular interesting not

only because disputant countries have engaged in various patterns of diplomatic and military

behaviors in dealing with their respective island disputes, but also because economic

interdependence of individual dyads has varied widely across time and space, thus providing

an ideal laboratory to test the liberal peace argument.

When and why do disputant countries pursue conflictual policies? Conversely, why

do they at other times seek the containment, if not the resolution, of territorial disputes by

shelving thorny sovereignty issues? In the absence of economic interdependence, one can

find examples of the aggressive use of military force and intransigent bargaining strategies.

It may not be a coincidence that violent clashes over the Paracels and Spratlys marked Sino-

Vietnamese relations during the Cold War period. By contrast, military inaction and

accommodative diplomacy are equally evident, particularly when disputant countries share

3

compelling economic interests. Despite regular outbreak of tension, violent military action

has not been a viable option in the cases of the Dokdo/Takeshima and Senkaku/Diaoyu

disputes, given the high priority currently paid to economic development and cooperation.

In Section II, I present an overview of conventional explanations for East Asian

territorial disputes and argue that the liberal peace thesis is a promising explanatory avenue

to take. In Section III, I define my dependent variable as the level of hostility reached in a

given dyad-year. Then, I identify the independent and control variables. In Section IV, I

test liberal peace hypothesis using ordinary least squares (OLS) regression. In Section V, I

summarize the findings and draw policy implications.

II. CHARACTERISTICS OF EAST ASIAN ISLAND DISPUTES

1. Conventional Explanations

Enduring island disputes in East Asia can be characterized in various ways, but four features

are particularly important: (1) despite the frequent resort to the past to justify their

contemporary claims, most disputes originated from colonial times in the late 19th and early

20th centuries, often exacerbated by the arbitrary map-making of colonial powers and

exploited by post-colonial nationalists; (2) in contrast to other parts of the world, few

territorial disputes have been formally resolved, and very few target countries, whose

territorial claim is challenged by revisionist countries, recognize the existence of disputes at

all, preventing claimant countries from engaging in negotiations; (3) a periodic pattern exists

in the iteration of disputes, repeating between initiation, escalation, and de-escalation; (4)

nevertheless, recurring crises have rarely increased the level of hostility beyond control

except for a few cases. Set against this background, the existing literature on East Asian

4

territorial disputes has largely focused on one of five research tracks as follows.

The first category of work treats territorial disputes as a result of direct competition

for locations of strategic and economic value. A number of studies focus on the critical

sea lanes, disputed islands, and continental shelf with high potentials of oil and gas deposits

in the East and South China Seas. Given the growing economic and strategic value of

disputed maritime areas, these studies often predict that East Asian countries are likely to

pursue more assertive territorial policies.2

The second category of work is based directly or indirectly on historical and cultural

approaches to international relations. From this viewpoint, a number of contested areas

involve questions of national identity and pride, and the centrality of territory in national

identity formation suggests that conflicts over the disputed islands should be highly salient

for the victims of Western and Japanese colonialism.3

In the third category of work, scholars examine territorial disputes from a legal

2 Hamzah Ahmad, ed., The Straits of Malacca: International Cooperation in Trade, Funding, and Navigational

Safety (Selangor Darul Ehsan: Pelanduk Publications, 1997); Kent E. Calder, Pacific Defense: Arms,

Energy, and America’s Future in Asia (New York: William Morrow, 1996); Robert Catley and Makmur

Keliat, Spratlys: The Dispute in the South China Sea (Brookfield: Ashgate, 1997); Choon-ho Park, “Oil

Under Troubled Waters: The Northeast Asia Sea-Bed Controversy,” Harvard International Law Journal

14: 2, (1973), pp. 212-260; Mark J. Valencia, China and the South China Sea Disputes: Conflicting Claims

and Potential Solutions in the South China Sea, Adelphi Paper 298 (London: The International Institute

for Strategic Studies, 1995); Mark J. Valencia, Jon M. Van Dyke, and Noel A. Ludwig, Sharing the

Resources of the South China Sea (Honolulu: University of Hawaii Press, 1997). 3 Andrew Mack, “Island Disputes in Northeast Asia,” Working Paper No. 1997/2 (Canberra:

Australian National University, 1997); Yong-Ha Sin, Korea’s Territorial Rights to Tokdo: An Historical

Study (Seoul: Dokdo Research Association, 1997); Scott Snyder, “The South China Sea Dispute

Prospects for Preventive Diplomacy,” Special Report 18 (Washington, D.C.: United Sates Institute

of Peace, 1996); Unryu Suganuma, Sovereign Rights and Territorial Space in Sino-Japanese Relations:

Irredentism and the Diaoyu/Senkaku Islands (Honolulu: University of Hawaii Press, 2000).

5

perspective, focusing on issues such as third-party involvement in conflict settlement and the

legal validity and enforceability of individual countries’ claims to territorial seas, exclusive

economic zones (EEZs), and continental shelves under the United Nations Convention on

the Law of the Sea (UNCLOS), which came into force in 1994.4

The fourth category of work examines how state elites’ desire to remain in office

allows domestic coalitions and electoral politics to influence their decisions on conflict

behavior. In this category of work, there are at least three related approaches, including the

democratic peace perspective,5 the legitimization strategy approach,6 and the pluralist model

of territorial disputes.7

Fifthly and finally, some systemic/structural accounts highlight that the U.S. as a

hegemonic power balancer has ensured that East Asian territorial disputes take place within

4 Greg Austin, China’s Ocean Frontier: International Law, Military Force and National Development

(Canberra: Allen and Unwin, 1998); Barry Buzan, A Sea of Trouble? Sources of Dispute in the New

Ocean Regime, Adelphi Paper 143 (London: International Institute for Strategic Studies, 1978);

Jonathan I. Charney, “Central East Asian Maritime Boundaries and the Law of the Sea,” The

American Journal of International Law 89 (1995), pp. 724-749; R. Haller-Trost, “The Spratly Islands: A

Study on the Limitations of International Law,” Occasional Paper 14 (Canterbury: Centre of

South-East Asian Studies of University of Kent at Canterbury, 1990). 5 Jennifer M. Lind, “Democratization and Stability in East Asia,” paper presented at the 2004 Annual

Meeting of the American Political Science Association, September 2-5 (Chicago, IL, 2004). 6 Youngshik D. Bong, Flashpoints at Sea? Legitimization Strategy and East Asian Island Disputes, Ph.D.

Dissertation in Political Science (Philadelphia: University of Pennsylvania, 2002); Erica Strecker

Downs and Philip C. Saunders, “Legitimacy and the Limits of Nationalism: China and the Diaoyu

Islands,” International Security 23: 3 (1998/99), pp. 114-146; M. Taylor Fravel, The Long March to Peace:

Explaining China’s Settlement of Territorial Disputes, Ph.D. Dissertation in Political Science (Stanford:

Stanford University, 2003). 7 Chien-Peng Chung, Domestic Politics, International Bargaining and China’s Territorial Disputes (New York:

Routledge, 2004); Phil Deans, “Contending Nationalisms and the Diaoyutai/Senkaku Dispute,”

Security Dialogue 31: 1 (2000), pp. 119-131.

6

certain confines.8 More recently, many note that the biggest source of structural

uncertainty is coming from China, which is a party to many of the thorny maritime and

territorial disputes.9

Despite their explanatory utility, existing studies all share several spatial and thematic

limitations. Most notably, they examine only a fraction of disputes in a particular time and

space, thereby making their conclusions tentative and ungeneralizable. For example, the

historical and legal issues surrounding the Senkaku/Diaoyu dispute in the late 1960s are not

compared with those of the Paracel dispute in the 1970s; the limited use of force over the

Dokdo/Takeshima Islands in the early 1950s is not juxtaposed with the intensive use of

force over the Spratly Islands in the late 1980s.

Relatedly, the arguments in these studies are not integrated in broader international

relations theory. What is most striking is the absence of scholarly attention to the influence

of economic interdependence on repeated clashes over offshore islands. Despite some

burgeoning interests in its pacific effect, few studies on economic interdependence have

been dedicated to territorial disputes as a distinct category of regional conflicts.10

8 Kent E. Calder, “Securing Security through Prosperity: The San Francisco System in Comparative

Perspective,” The Pacific Review 17: 1 (2004), pp. 135-157; Victor D. Cha, “Bridging the Gap: The

Strategic Context of the 1965 Korea-Japan Normalization Treaty,” Korean Studies 20 (1996), pp. 123-

160; Kimie Hara, “50 Years from San Francisco: Re-examining the Peace Treaty and Japan’s

Territorial Problems,” Pacific Affairs 74: 3 (2001), pp. 361-382; Chi-kin Lo, China’s Policy towards

Territorial Disputes: The Case of the South China Sea Islands (New York: Routledge, 1989). 9 Chung 2004; Lo 1989; Valencia 1995; Avery Goldstein, “Balance-of-Power Politics: Consequences

for Asian Security Order,” in Muthiah Alagappa, ed., Asian Security Order: Instrumental and Normative

Features (Stanford: Stanford University Press, 2003) 10 See Stuart Harris and Andrew Mack, eds., Asia-Pacific Security: The Economic-Politics Nexus (St.

Leonards: Allen & Unwin Australia, 1997); Mike M. Mochizuki, “Security and Economic

Interdependence in Northeast Asia,” Asia/Pacific Research Center Working Paper (Stanford:

7

My particular take on these empirical and theoretical issues is to examine the three

island disputes simultaneously through the prism of liberal peace. In what follows, I

summarize the theoretical debate about the question of liberal peace.

2. Liberal Peace Debate

International relations scholars have long been puzzled by the question of liberal peace, as

economic interdependence in fact has both positive and negative influences on international

conflicts. An examination of the empirical literature indicates that one can easily find

support for either positive or negative aspects of economic interdependence.11

Early arguments that connected economic interdependence with less conflict were

based on the hope that international trade would somehow quell the passion for territorial

conquest that seemed so prevalent in the 17th and 18th centuries.12 This liberal view was

restated in the early 20th century by Norman Angell, who argued that war, by destroying

trade ties, is “commercially suicidal.”13 Theories about complex economic interdependence

Stanford University, 1998); Susan L. Shirk and Christopher P. Twomey, eds., Power and Prosperity:

Economics and Security Linkages in Asia-Pacific (New Brunswick: Transaction Publishers, 1996); Ming

Wan, “Economic Interdependence and Economic Cooperation: Mitigating Conflict and

Transforming Security Order in Asia,” in Muthiah Alagappa, ed., Asian Security Order: Instrumental

and Normative Features (Stanford: Stanford University Press, 2003). 11 Of the twenty works reviewed, one study found that ten support liberal hypotheses; six produce

mixed or conditional results; and four support realist hypotheses. See Susan McMillan,

“Interdependence and Conflict,” Mershon International Studies Review 41: 1 (1997), pp. 33-58. 12 Michael Doyle, Ways of War and Peace: Realism, Liberalism, and Socialism (New York: W.W. Norton &

Company, 1997). 13 Norman Angell, The Great Illusion: A study of the Relation of Military Power in Nations to Their Economic

and Social Advantage (New York: Putnam, 1910).

8

carried the interdependence arguments one step further.14 Though not all contemporary

adherents of liberal peace go as far as Kenichi Ohmae,15 who asserts that the world is now

“borderless,” it is rather commonly accepted among liberal scholars that territorial

boundaries are declining in economic significance.16

Realists turn liberal peace theory on its head. They argue that economic

interdependence not only fails to promote peace, but in fact heightens the likelihood of

conflict since interdependence tends to foster asymmetry as a result of dependence and

inequality between trading partners in an anarchic setting.17 As Kenneth Waltz notes,

increased interdependence may also lead to increased conflict since increased contact creates

potential opportunities for discord.18 Furthermore, state actors under anarchy must worry

that others will gain more from cooperation than they do, since those relative gains might

later be turned into military advantage.19 In a realist world, therefore, states have

historically given top priority to the conquest of territory in order to advance their economic,

14 Robert Keohane and Joseph Nye, Power and Interdependence: World Politics in Transition (Boston: Little,

Brown, 1977). 15 Kenichi Ohmae, The Borderless World: Power and Strategies in the Interlinked Economy (New York:

Harper Business, 1993). 16 Susan Strange, The Retreat of the State: The Diffusion of Power in the World Economy (Cambridge:

Cambridge University Press, 1996); Richard Rosecrance, The Rise of the Virtual State: Wealth and

Power in the Coming Century (New York: Basic Books, 2000). 17 Albert Hirschman, National Power and the Structure of Foreign Trade (Berkeley: University of

California Press, 1980 [1945]); Stanley Hoffman, “Rousseau on War and Peace,” American Political

Science Review 57: 2 (1963): pp. 317-333. 18 Kenneth Waltz, “The Myth of Interdependence,” in Charles Kindleberger, ed., The International

Corporation (Cambridge: The MIT Press, 1970). 19 Joseph Grieco, Cooperation among Nations: Europe, America, and Non-tariff Barriers to Trade (Ithaca:

Cornell University Press, 1990); Joanne Gowa, Allies, Adversaries, and International Trade (Princeton:

Princeton University Press, 1994).

9

security, and other interests, regardless of economic interdependence.20

III. RESEARCH STRATEGY

My major challenge in putting together this study is to provide a systematic account of the

varying degrees of hostility involved in each island dispute from a large-N perspective. A

number of statistical studies of international peace and conflict have produced many useful

and important findings.21 Yet few attempts have been made thus far to focus on East Asian

territorial disputes. In this section, I define the dependent and independent variables, and

explain their selection. I also consider four sets of variables to control for spurious

relationships between the dependent and independent variables.

1. Dependent Variable: Dispute Intensity

In defining the dependent variable, I focus on the annual variation in the dispute process

between the three pairs of countries: South Korea-Japan, Japan-China, and China-Vietnam,

20 Robert Gilpin, War and Change in World Politics (Cambridge: Cambridge University Press, 1981), p.

23. 21 Katherine Barbieri, The Liberal Illusion: Does Trade Promote Peace? (Ann Arbor: University of

Michigan Press, 2002); John R. Oneal, Frances H. Oneal, Zeev Maoz, and Bruce M. Russett, “The

Liberal Peace: Interdependence, Democracy, and International Conflict, 1950-1985,” Journal of Peace

Research 33: 1 (1996), pp. 11-28; John R. Oneal and Bruce M. Russett, “The Classical Liberals Were

Right: Democracy, Interdependence, and Conflict, 1950-85,” International Studies Quarterly 41: 2

(1997), pp. 267-293; Bruce Russett, John Oneal, and David Davis, “The Third Leg of the Kantian

Tripod for Peace: International Organizations and Militarized Disputes, 1950-85,” International

Organization 52: 3 (1998), pp. 441-467; Paul D. Senese and John A. Vasquez, “A Unified Explanation

of Territorial Conflict: Testing the Impact of Sampling Bias, 1919–1992,” International Studies

Quarterly 47: 2 (2003), pp. 275-298.

10

respectively. In terms of operationalizing the outcome variable, the crucial question is

where to draw lines among given dyad-years. What degree of behavior do we classify

visible or invisible territorial claims as an act of initiation, escalation, or de-escalation?

Various international conflicts range from verbal disputes (e.g., diplomatic rebukes)

to outright military showdowns and to wars. Many scholars agree that there tends to be a

temporal sequence between mild rhetorical disputes and intensive wars.22 Building upon

existing literature, I use a five-point scale measure to code the highest level of hostility

reached in a given territorial dyad-year.23 In collecting outcome data for the postwar period,

I draw upon Zeev Maoz’s Correlates of War (COW)-revised Dyadic Militarized Interstate

22 Michael Brecher, Crises in World Politics: Theory and Reality (Oxford: Pergamon, 1993); Glenn Snyder

and Paul A. Diesing, Conflict Among Nations: Bargaining, Decision Making, and System Structure in

International Crises (Princeton: Princeton University Press, 1977); Paul K. Huth, Standing Your Ground:

Territorial Disputes and International Conflict (Ann Arbor: University of Michigan Press, 1996); Beth A.

Simmons, “Capacity, Commitment, and Compliance: International Law and the Settlement of

Territorial Disputes,” Journal of Conflict Resolution 46: 6 (2002), pp. 829-856. 23 The scale to be used in this study is as follows:

0 = minimal or no diplomatic/political conflict over disputed territory. The challenger may be

involved in a territorial dispute, but there is very limited evidence of public confrontation over

territory, and the dispute has no apparent adverse effect on bilateral relations between a challenger

and a target.

1 = moderate to high levels of diplomatic and political conflict over disputed territory. A

challenger actively confronts a target over disputed territory, and the actions and counter-actions of

challenger and target countries may include hostile rhetoric, public recriminations, and soliciting of

third-party support in order to pressure a target into making concessions.

2 = Either a challenger or a target country, in addition to active and confrontational diplomatic

actions, may use sanctions on bilateral diplomatic, economic, or military ties; seize civilians in

disputed areas; and threaten to use military force to reinforce its territorial claim.

3 = Mild physical clashes or isolated incidents between police or armed forces may take place.

4 = If the circumstances continue to escalate, a military escalation may follow with an outbreak of

a MID or a full-fledged war.

11

Dispute (MID) dataset (DYMID 1.1) as well as other resources.24

2. Independent Variable: Trade Interdependence

I hypothesize the relationship between territorial disputes and complex economic

interdependence as follows:

Hypothesis 1 If pairs of countries have high levels of economic ties, they are less likely to

initiate and escalate a dispute to higher levels of hostility.

For the sake of analytical simplicity, I choose to use trade as primary proxy for bilateral

economic interdependence. Different studies use a wide variety of indicators of trade

interdependence—such as trade volumes and values, systemic trade levels, trade as a proportion

of GDP, elasticity of supply and demand, and the trade of strategic goods. One of the

simplest ways to capture the dyadic levels of trade interdependence is to construct a trade

dependence index, defined as the share of dyadic trade flow (the sum of imports and exports)

in GDP.25 Although based on the same baseline measure of trade dependence, Oneal and

24 At the bilateral level, unfortunately, there is no single source that lists the evolution process of

East Asian territorial disputes other than the Dyadic MID dataset, which itself requires substantial

updates and enhancements. As a result, I consult a broad range of sources. In classifying

different levels of hostility, I require confirmation of a dispute from at least two different scholarly

sources. The coding of each outcome variable is reported in the following section. In general, it

is not difficult to identify conflicts when public statements are issued by state leaders that a dispute

exists, or when physical conflict takes place. Nevertheless, there will undoubtedly be some debate

and questions concerning the coding of specific cases.

25 Formally,i

ij

i

ijiji

GDPTrade

GDP) Exports (Imports

Dependence Trade =+

= , where Tradeij denotes bilateral trade

between states i and j.

12

Russett (1997) and Barbieri (2002) offer two of the most different constructions of trade

interdependence. Oneal and Russett use the lower state’s trade dependence score as an indicator

of dyadic interdependence. The rationale for this measure is that the least dependent

country is the “weakest link,” in Dixon’s terms26; it is less constrained to refrain from force,

because it needs the relationship less. By contrast, Barbieri argues that it is problematic to

employ the characteristics of only one country when describing the characteristics of a dyad.

Hence she proposes two components of interdependence, namely the salience and symmetry

of trade dependence to include the interactive nature of dyadic trade dependence.27 She

defines trade interdependence as a product of these two measures.

I find that Barbieri’s interaction terms tend to be highly correlated with either one or

both of the two trade dependence scores for a dyad, thereby creating statistical problems of

multicollinearity. As a result, I choose to use each country’s trade dependence score as a

separate indicator of dyadic trade interdependence, as trade dependence scores by definition

already reflect the dyadic nature of trade. Yet, unlike Oneal and Russett, I do not adopt the

weakest link hypothesis, since I believe that both more and less dependent countries in a

26 William J. Dixon, “Democracy and the Management of International Conflict,” Journal of Conflict

Resolution 37: 1 (1993), pp. 42-68. 27 For Barbieri, the salience score for a pair of countries, defined as the geometric mean of trade

dependence measures, captures the extent to which the partners are important to each other

( jiij Dependence Trade Dependence Trade Salience ×= ). For example, higher dyadic-salience score

indicates that the trade relationship is important for both countries. This index is designed to

capture the liberal claim that (salient) trade interdependence can produce a positive impact on peace

by providing the necessary bonds to deter conflict. Barbieri defines the symmetry of dependence as

the difference of trade-dependence measures of a pair of countries (Symmetryij = 1- │Trade Dependencei –

Trade Dependencej│). This measure is designed to capture the realist concern that unequal trade

dependence may exert a negative influence on peace by conferring coercive power to a less

dependent country and/or touching off nationalist sentiments in a more dependent country.

13

dyad constitute the composition of bilateral economic interdependence.

3. Control Variables

Alliances and Relative Capabilities

I hypothesize the relationship between territorial disputes and the geopolitical posture in

which countries are situated as follows:

Hypothesis 2.1 If pairs of countries share common security ties and interests (or indirectly

through superpowers such as the U.S. and the Soviet Union before its collapse), they are less likely

to initiate and escalate a dispute to higher levels of hostility.28

Hypothesis 2.2 If a challenger country experiences faster growth in military capabilities than

its target country, it is more likely to initiate and escalate a dispute to higher levels of hostility.29

With respect to Hypothesis 2.1., a dummy variable is often used to measure whether or not

disputant countries have alliance ties. For instance, a value of 1 is assigned if challenger

and target countries have a defense pact or entente military alliance, or currently share a

common ally; otherwise, a value of 0 is assigned. As for the three island disputes examined

here, however, none of the dispute dyads have formed military alliance thus far, thereby

making such a binomial measure a constant rather than a variable. In the case of the

28 See Gowa 1994 and Joanne Gowa and Edward D. Mansfield, “Power Politics and International

Trade,” American Political Science Review 87 (1993), pp. 408-420. 29 See Charles F. Doran, Systems in Crisis: New Imperatives of High politics at Century’s End (Cambridge:

Cambridge University Press, 1991) and A. F. K. Organski and Jacek Kugler, The War Ledger

(Chicago: University of Chicago Press, 1980).

14

Dokdo/Takeshima dispute, I use the U.S. defense commitment as proxy variable to capture

the quasi-alliance relationship between South Korea and Japan. Other than that, the

alliance variable is dropped out of regression models.

To test Hypothesis 2.2, I construct a variable for a short-term material power balance

that measures the national capabilities of each country relative to its territorial counterpart.

The primary source for data on balance of national capabilities is the COW II-based

National Material Capabilities dataset (Version 3.0), which contains annual values for total

population, urban population, iron and steel production, energy consumption, military

personnel, and military expenditure of all state members, currently from 1816-2001. The

widely-used Composite Index of National Capability (CINC) is based on these six indicators

and included in the dataset.30

Autocracy and Political Legitimacy

The logic of a domestic power politics explanation can be hypothesized as follows:

Hypothesis 3.1 If pairs of countries include an authoritarian regime, they are more likely to

initiate and escalate a dispute to higher levels of hostility.

Hypothesis 3.2 If pairs of countries have incumbent leaders who are in a legitimacy deficit

and/or who are challenged by strong opposition groups, they are more likely to initiate and escalate a

dispute to higher levels of hostility.

For Hypothesis 3.1, I code the degree to which the political system of challenger and target

30 For the years 2002-05, I estimate the CINC scores using the technique of exponential smoothing,

where necessary.

15

countries are democratic as opposed to authoritarian. I utilize the POLITY IV to create a

net-democracy score, which ranges in value from -10 to +10.31 This 21-point net-

democracy score is created by subtracting each country’s autocracy score (which ranges from

0 to 10) from its democracy score (which also ranges from 0 to 10). As values approach -

10, a country is judged to be highly authoritarian, whereas scores approaching +10 indicate

that it is highly democratic. Conventionally, a country is considered democratic if its net-

democracy score is greater than or equal to +6 and is considered non-democratic, or

autocratic, if its net-democracy score is less than or equal to +5.32

For Hypothesis 3.2, I use two measures selectively for the security of incumbent

leaders. The first measure is a dummy variable to capture whether or not the ruling party

or ruling government possesses a majority of legislative seats (1: majority; 0: minority). In

authoritarian political systems, the resort to organized violent challenges by opposition

groups outside of the military is a strong signal that the regime faces a critical problem of

legitimacy. I use a dummy variable coded with a value of 1 if a country has experienced

this type of violent political challenge to the regime within the past year.33

Shared Intergovernmental Organization (IGO) Membership

I hypothesize the role of IGOs in determining the dispute process as follows:

31 Monty G. Marshall and Keith Jaggers, “Political Regime Characteristics and Transitions, 1800-

2002: Dataset Users’ Manual,” Center for International Development and Conflict Management

(College Park: University of Maryland, 2002). 32 Paul K. Huth and Todd L. Allee, The Democratic Peace and Territorial Conflict in the 20th Century

(Cambridge: Cambridge University Press, 2002), pp. 91-92. 33 Huth and Allee 2002, pp. 95-98.

16

Hypothesis 4 If pairs of countries have joint membership in regional and/or global IGOs,

they are less likely to initiate and escalate a dispute to higher levels of hostility.

Realists often dismiss IGOs as unimportant because they typically lack means of

enforcement that are independent of the international distribution of power.34 Yet a

close look at various IGOs indicates that they may serve any of six functions: coercing

norm breakers; mediating among conflicting parties; reducing uncertainty by conveying

information; problem-solving, including expanding states’ conception of their self-

interest to be more inclusive and long-term; socialization and shaping norms; and

generating narratives of mutual identification.35

Given the pronounced formal “organization gap” in East Asia,36 my measure for

joint IGO membership of each dyad focuses on two broad-based, multilateral IGOs: the

General Agreement on Tariffs and Trade (GATT)/World Trade Organization (WTO)

and Asia-Pacific Economic Cooperation (APEC). If pairs of countries share joint

membership of either IGO in a particular dyad-year, the case will be coded with a value

of 1, and the case will be coded with a value of 0 if they share none of the membership.

Temporal Dependence

OLS regression models, which I use to estimate the impact of causal factors on the intensity

of a dispute, assume independence of events. Yet this assumption is likely to be violated

when analyzing disputes, since the occurrence and escalation of one dispute might affect

34 John Mearsheimer, The Tragedy of Great Power Politics (New York: W.W. Norton, 2001). 35 Russett, Oneal, and Davis 1998, pp. 444-445. 36 Kent E. Calder and Min Ye, “Regionalism and Critical Junctures: Explaining the ‘Organization

Gap’ in Northeast Asia,” The Journal of East Asian Studies 4: 2 (2004), pp. 191-226.

17

those of other ones. To address temporal dependence in my time-series regression models

with discrete dependent variables, I include a lagged dependent variable to control for

whether or not a particular dyad experiences a dispute in the previous year. The lagged

dependent variable should be positively related to the dispute intensity in present time.

IV. ESTIMATION RESULTS

In this section, I present a brief chronological account and analyze each of the three island

disputes using OLS regression. For each dyad-year, values are recorded for the dependent

variable at time t and for the explanatory variables at time t-1. I employ a one-year lag for

all the explanatory variables in order to reduce the problems posed by the reciprocal

relationship between the outcome and explanatory variables.

1. The Dokdo/Takeshima Dispute

An Overview of the Dokdo/Takeshima Dispute

South Korea and Japan have forged a deeper diplomatic and economic partnership,

particularly since the landmark joint declaration of 1998.37 A new spirit of partnership

culminated in the successful co-hosting of the World Cup 2002. At the end of 2003 the

two neighbors began to negotiate a free trade agreement (FTA) to further strengthen their

already close economic ties. South Korea’s decades-long embargo on Japanese cultural

products has now been lifted, while a number of South Korean pop stars are currently

37 Jason U. Manosevitz, “Japan and South Korea: Security Relations Reach Adolescence,” Asian

Survey 43: 5 (2003), pp. 801-825.

18

sweeping across Japan, creating the so-called “Korean Wave” fever. A pragmatic

calculation of national interests would suggest more cooperative behavior. Yet beneath the

surface, few years have passed since the end of Japanese colonial rule in 1945 without

dangerous undercurrents capable of causing a diplomatic crisis across the East Sea/Sea of

Japan. The unsettled sovereignty question of the Dokdo/Takeshima Islands—known as

Dokdo (Lone Islands) in Korea and as Takeshima (Bamboo Islands) in Japan—is one of the

most fundamental barriers to better bilateral relations. In the postwar period, the

Dokdo/Takeshima dispute has unfolded in four different rounds of clashes.

The First Round of Dispute (1952-65)

Competing claims to the Dokdo/Takeshima Islands emerged during the San Francisco

Peace Treaty negotiations between the Allies and Japan. Upon its conclusion in 1951, a

series of hostile rhetoric and physical clashes over the island and other bilateral issues

followed between South Korea and Japan, most notably in 1952-54 when South Korea

regained physical control of the islands.38 With grim prospects for mutual gains from

trade, the actual outcomes during the first round of the island clash support the liberal

peace hypothesis. After a brief trade boom during the Korean War (1950-53), the two

countries quickly returned to quasi-enemy status, as indicated by low trade dependence

scores. Yet by the mid-1960s, the island issue took a back seat as the prospect of trade

and investment cooperation between the two quasi-allies became increasingly urgent.39

38 Cha 1996; Hara 2001. 39 In particular, the grim economic situations in South Korea motivated President Park Chung Hee

to “live or die” with the diplomatic normalization with Japan. Indeed the conclusion of the

normalization treaty in 1965 provided a fledgling South Korean economy with much-needed

foreign capital from Japan. The treaty also cleared the way for an extensive expansion of trade

19

The Second Round of Dispute (1977-78)

The second round of the Dokdo/Takeshima dispute was catalyzed by a global trend towards

a 200nm EEZ regime, which made the 1965 fishing agreement between South Korea and

Japan outdated. Most notably, the proclamation of 200nm fishing zone by the Soviet

Union in 1976 ignited fishery disputes between Seoul and Tokyo because South Korea’s

distant water fishing vessels that had lost their fishing ground in the Soviet zone in the

Northwest Pacific began to swamp the coastal areas of Japan’s Hokkaido.40 In response,

Japan unilaterally declared a 12nm territorial waters law and a 200nm fishing zone law in July

1977 in spite of South Korea’s angry reactions.41 Yet the intensity of the dispute remained

moderate at worst, as more pressing economic considerations—particularly for South Korea

during the period of its economic takeoff—overrode the question of sovereign and

maritime rights.42

The Third Round of Dispute (1996-98)

After over fifteen years of relative calm, the extensive confrontation over the maritime and

territorial boundary issues in 1996-98 drove the overall South Korea-Japan relationship to

one of its lowest points in the post-normalization era. Aside from the conclusion of the

UNCLOS in 1994, this period coincided with the beginning of the post-Cold War period

relations. For more details, see Cha 1996 and Kong-dan Oh, Japan-Korea Rapprochement: A Study in

Political, Cultural, and Economic Cooperation in the 1980s, Ph.D. Dissertation in Asian Studies (Berkeley:

University of California, 1986). 40 Choon-ho Park, “South Korea and the Law of the Sea,” in Choon-ho Park, East Asia and the Law

of the Sea (Seoul: Seoul National University Press, 1983), pp. 146-147. 41 Dae Song Hyun, Postwar Korea-Japan Relationship and Territorial Dispute: Discourse of Tokdo Problem and

Image in Korea, Ph.D. Dissertation in Political Science (Tokyo: University of Tokyo, 2004), p. 78. 42 Cha 1996; Oh 1986.

20

and the democratization of South Korea, both of which contributed to a more contentious

diplomacy.43 Yet a full-scale diplomatic crisis was averted when the dire need for

emergency loans from abroad due to the 1997-98 financial crisis forced South Korea to back

down on the territorial and maritime issues with Japan. Despite rising anti-Japanese

sentiments, newly-elected South Korean President Kim Dae Jung used his abundant political

capital generated by his country’s financial emergency to forge a compromise on the

maritime and territorial issues.

The Fourth Round of Dispute (2004-05)

The inauguration of the Roh Moo Hyun administration in 2003 has brought to the fore new

political agendas in South Korea to foster more balanced and equal relationships with its

traditional allies, namely the U.S. and Japan. In Japan, the Koizumi government’s abrasive

diplomacy with its neighbors has made matters even worse.44 If anyone still doubted that

there remained a wide gap between South Korea and Japan, the 2004-05 Dokdo/Takeshima

dispute proved the point. In 2004, the two quasi-allies collided over South Korea’s Dokdo

postage stamps that illustrated the fauna and flora of the disputed islands. In early 2005,

South Korean-Japanese relations went from bad to worse. In spite of violent anti-Japanese

demonstrations in South Korea, a Japanese local government established an annual

“Takeshima Day” and the Japanese Education Ministry authorized the revisions of new

history textbooks that reinforce Japan’s symbolic claim to the islands.

43 Bong 2003; Lind 2004. 44 Chung Min Lee, “Rethinking Future Paths on the Korean Peninsula,” The Pacific Review 17: 2 (2004),

pp. 249-270; Gilbert Rozman, “Japan and Korea: Should the U.S. Be Worried about Their New

Spat in 2001?” The Pacific Review 15: 1 (2002), pp. 1-28.

21

Operationalizations

Building upon Maoz’s Dyadic MID dataset and a comprehensive chronology that I have

collected, I code the highest level of hostility vis-à-vis the Dokdo/Takeshima dispute during

the period of 1953-2005. Figure 1 illustrates the value assigned to each dyad-year, based on

a five-point scale.

0

1

2

3

4

1953 1958 1963 1968 1973 1978 1983 1988 1993 1998 2003

I use South Korea’s trade dependence on Japan and Japan’s trade dependence on

South Korea as two separate independent variables to measure the dyadic level of economic

interdependence. South Korea’s trade dependence on Japan has fluctuated widely over

time (See Figure 2). By contrast, Japan’s trade dependence on South Korea has been on a

steady rise, but its level is much lower than that of South Korea’s in absolute terms (See

Figure 3). These trade dependence scores indicate that trade with Japan has been more

important to South Korea than trade with South Korea has been to Japan.

Figure 1. Intensity of the Dokdo/Takeshima Dispute (1953-2005)

22

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

1953 1958 1963 1968 1973 1978 1983 1988 1993 1998 2003

In addition, I consider five categorical variables to control for the two countries’

geopolitical postures, domestic power dynamics, institutional contexts, and time dependence.

The U.S. defense commitment variable is coded with a value of 1 if the incumbent U.S.

administration maintains strong security commitment to both South Korea and Japan in a

given dyad-year, and coded with a value of 0 otherwise. This indicator is designed to

Figure 2. South Korea’s Trade Dependence on Japan (1953-2004)

0.0%

0.2%

0.4%

0.6%

0.8%

1.0%

1.2%

1.4%

1.6%

1953 1958 1963 1968 1973 1978 1983 1988 1993 1998 2003

Figure 3. Japan’s Trade Dependence on South Korea: 1953-2004

Sources: The International Monetary Fund, Direction of Trade Statistics; The World Bank, World Development Indicators.

23

measure the strength of the security triangle among the U.S., South Korea, and Japan.45

The joint regime strength variable is coded with a value of 1 if the ruling parties of South Korea

and Japan both hold majority seats in the National Assembly and Lower House, respectively,

and coded with a value of 0 otherwise. A dichotomous indicator for joint democracy is coded

with a value of 1 when both countries score above 6 for democratic governance on the

Polity scale in the POLITY IV dataset; otherwise, it is coded with a value of 0. The joint

IGO membership variable is coded with a value of 1 if South Korea and Japan share common

membership in either the WTO or APEC, and coded with a value of 0 if they share none of

the memberships.46 Finally, I consider the lagged dependent variable to control for time

dependence (See Appendix 1 for the coding of these control variables).

Estimation Results

I now turn to the OLS regression analysis that tests whether or not economic

interdependence determines the intensity of the Dokdo/Takeshima dispute. The

estimation results reported in Table 1 provide some supports for liberal peace hypothesis.

South Korea’s trade dependence on Japan has a statistically significant negative impact on the

45 The Kennedy (1961-62), Nixon (1969-73), Carter (1977-80), Clinton (1993-96), and Bush (2001-

present) administrations are coded with a value 0, accordingly. It is not an easy task to measure

the level of defense commitment in a coherent manner. There are many factors that affect U.S.

troop levels, U.S. funding/burden sharing, troop locations, equipment modernization, and the like.

If taken together, they may show a trend, but given the short timeline, my measure is probably the

only expedient method. For more details about the characteristics of the security triangle,

particularly the role of the U.S. in defining the quasi-alliance relationship between South Korea and

Japan, see Victor D. Cha, Alignment Despite Antagonism: The United States-Korea-Japan Security Triangle

(Stanford: Stanford University Press, 1999). 46 Japan joined the GATT in 1955, while South Korea joined the organization in 1967. Both

countries were founding members of APEC in 1989.

24

dispute intensity (coefficient β = -6.514). This indicates that the more dependent on Japan

South Korea becomes for trade, the less likely the Dokdo/Takeshima dispute is to be

escalated. In other words, the level of hostility diminishes when constraints on the hostile

reactions—most likely from South Korea—are kept at bay with the significant level of South

Korea’s dependence on Japan for trade. Yet Japan’s trade dependence on South Korea is not

statistically significant, presumably reflecting Japan’s low level of trade dependence on South

Korea in both absolute and relative terms.

With regard to control variables, joint democracy has a statistically significant negative

impact on the dispute intensity as hypothesized (coefficient β = -.369). Although the level

of statistical significance is a bit low, this result indicates that democracy does have a pacific

influence on the level of hostility involved in the Dokdo/Takeshima dispute. As

hypothesized, the lagged dependent variable (past dispute intensity) has a statistically

significant positive effect on the dispute intensity (coefficient β = .424). Finally, neither U.S.

defense commitment nor joint IGO membership variable is statistically significant.

Table 1. OLS Results: Dokdo/Takeshima Dispute (1953-2005)

Explanatory variables Dispute Intensityt South Korea’s trade dependence on Japant-1 -6.514 (-1.845)* Japan’s trade dependence on South Koreat-1 74.286 (1.317) U.S. defense commitmentt-1 -.099 (-.541) Joint regime strengtht-1 -.369 (-1.449) Joint democracyt-1 -.566 (-1.837)* Joint IGO membershipt-1 -.240 (-.541) Past dispute intensity .424 (2.945)** R2 .790 Durbin-Watson d 1.836

Note: The t-statistics appear in parentheses. * Significant at the .10 level. ** Significant at the .05 level. *** Significant at the .01 level.

25

2. The Senkaku/Diaoyu Dispute

An Overview of the Senkaku/Diaoyu Dispute

The Senkaku/Diaoyu Islands—known as Senkaku Retto (Rocky Hill Islands) in Japan and as

Diaoyutai (Fishing Platform Islands) in China—are effectively controlled by Japan, but its

ownership claim is contested by China and, to a lesser extent, by Taiwan. The dispute over

this small and uninhabited group of islands may appear paradoxical given that China is the

world’s most populous country and the third largest in land area, and that Japan is the

world’s second largest economy and one of the largest maritime countries. The two East

Asian giants have forged closer economic ties since their diplomatic rapprochement in 1972

and a pragmatic consideration of national interests would suggest more cooperative behavior.

Yet the reality in the East China Sea does not bode well for a shared effort to promote

mutual cooperation, as the Senkaku/Diaoyu dispute has served as the most persistent and

explosive bone of bilateral contention. The post-war dispute over the Senkaku/Diaoyu

Islands has unfolded in five different rounds of clashes since its inception in the late 1960s.

The First Round of Dispute (1968-71)

The initial impetus of the first flare-up of the Senkaku/Diaoyu dispute came in late 1968

with the issuance of a geological survey reported by a UN committee, which found a high

probability of oil and gas deposits near the disputed islands. The UN report caused a great

sensation among the oil-hungry coastal countries in Northeast Asia, leading to what might

be dubbed an instant “oil-war” with an unparalleled overlapping of unilateral and arbitrary

claims to maritime areas. The tension was increased a notch further with the conclusion of

the 1971 Okinawa Reversion Agreement between the U.S. and Japan. Yet the island issue

26

gradually took a back seat by the end of 1971, as China and Japan both wanted to defuse the

bilateral tension in the middle of the negotiations for the Sino-Japanese normalization

(concluded in 1972) as well as the Japan-Taiwan de-recognition crisis.47

The Second Round of Dispute (1978)

The Senkaku/Diaoyu question surfaced again in a more serious manner in the spring of

1978. The 1978 crisis was intimately related to the Peace and Friendship Treaty (PFT)

negotiations between Japan and China. An initial impetus came in April 1978 when a

group of anti-PFT Japanese politicians urged that the Senkaku/Diaoyu matter be resolved as

part of the PFT negotiations, thereby causing a virtual collapse of the PFT talks. China’s

immediate reaction was to escalate the issue by sending a flotilla of armed fishing boats to

the disputed area.48 By the summer of 1978, the combination of national pride, yearning

for energy supply, and ultranationalist public pressures pushed both the Chinese and

Japanese governments to depart from a conciliatory approach that had marked the post-

normalization period. Yet broader economic and geopolitical calculations took higher

priority towards the end of 1978. Mutual benefits in concluding the PFT in general, and

shelving the Senkaku/Diaoyu issue in particular, involved an increase in bilateral trade and

investment, an external validation of Deng’s open-door policy by Japan, and the creation of

an informal common front against the possible military threat from the Soviet Union.49

The Third Round of Dispute (1990-91)

47 Chung 2004; Hara 2001; Park 1973; Suganuma 2000. 48 Daniel Tretiak, “The Sino-Japanese Treaty of 1978: The Senkaku Incident Prelude,” Asian Survey

18: 12 (1978), pp. 1242-1243. 49 Chung 2004, pp. 38-39; Tretiak 1978, p. 1237.

27

Contending sovereignty claims to the Senkaku/Diaoyu Islands collided once again in

September 1990 when the Japanese government accepted an application filed by a Japanese

ultranationalist group, which sought official lighthouse status for a beacon that had been

built on one of the islands in 1978. Citizens and students in China who belatedly learned

about the dispute vented their anger towards the Chinese leaders, who were thought to beg

for Japanese loans at the expense of China’s territorial sovereignty and national pride.50

The geopolitical vacuum left by the collapse of the Soviet Union further exacerbated the

situation by motivating Chinese political elites to mount a major propaganda campaign

against Japan to shore up their threatened legitimacy.51 In contrast to their combative

rhetoric, however, Japanese and Chinese leaders sought to defuse the dispute quickly and

quietly in the aftermath of the 1989 Tiananmen incident. Both parties saw a far greater

importance in keeping their bilateral economic relations intact than in standing up for

nationalist values of territorial integrity.52

The Fourth Round of Dispute (1996-97)

The long-running island dispute reared its ugly head again above the East China Sea in the

summer of 1996. Coupled with the introduction of the UNCLOS, an initial impetus for

the 1996-97 flare-up was provided by another lighthouse built on one of the islands by an

ultranationalist Japanese group. The situation became much more complicated by China’s

assertive nuclear tests and its intimidation of Taiwan. Culminating with the death of a pro-

China activist from Hong Kong during his failed attempt to land on one of the islands in

50 Downs and Saunders 1998/99, pp. 131-132. 51 Downs and Saunders 1998/99, p. 127; Chung 2004, p. 42. 52 Bong 2002; Downs and Saunders 1998/99.

28

September, the second lighthouse crisis turned out to be far more serious than was its

predecessor. Alarmed by aggravating situations, however, both governments sought to

defuse the tension by concluding a new bilateral fishery agreement in 1997. Particularly for

China, pushing its sovereign claims to the islands would have negatively affected its efforts to

secure Japan’s support for its deepening economic modernization.53

The Fifth Round of Dispute (2004-05)

In contrast to the previous clashes, where Japanese ultranationalist groups served as principal

initiators, the 2004 flare-up was catalyzed by a group of Chinese activists who landed on one

of the islands in March. In a chain reaction, the landing incident inflamed the Japanese

sense of indignation against China. Such a stormy clash of rival nationalisms was further

amplified by intense resource competition from the spring through the summer of 2004.

Simmering bilateral tensions came to a boiling point in April 2005 when a series of

sometimes violent anti-Japanese rallies broke out in the major cities of China. These

Chinese protests were catalyzed by the Japanese government’s authorization of history

textbooks that have reinforced Japan’s territorial claim to the islands as well as glossing over

its wartime atrocities. China acknowledged that diplomatic relations with Japan were at a

three-decade low.54

Operationalizations

The intensity of the Senkaku/Diaoyu dispute during the period of 1960-2005 is coded on a

five-point scale as illustrated in Figure 4.

53 Bong 2002, p. 77. 54 The Associated Press, April 18, 2005.

29

0

1

2

3

4

1960 1965 1970 1975 1980 1985 1990 1995 2000 2005

I use Japan’s trade dependence on China and China’s trade dependence on Japan as

two separate independent variables to measure the dyadic level of economic

interdependence. Japan’s trade dependence on China fluctuated at levels less than 1.5

percent during the period of 1960-90, but began to take off exponentially after 1991,

reaching 3.59 percent in 2004 (See Figure 5). China’s trade dependence on Japan also

indicates an overall rise, but has fluctuated more widely and with much higher levels of

dependence in absolute terms than did those of Japan’s on China (See Figure 6). These

trade dependence scores indicate that bilateral trade is increasingly becoming important for

both China and Japan, although trade with Japan has traditionally been more important to

China than trade with China has been to Japan until recently.

Figure 4. Intensity of the Senkaku/Diaoyu Dispute (1960-2005)

30

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

3.5%

4.0%

1960 1965 1970 1975 1980 1985 1990 1995 2000

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

1960 1965 1970 1975 1980 1985 1990 1995 2000

In addition, I consider five control variables to capture the two countries’ geopolitical

postures, domestic power dynamics, institutional contexts, and time dependence. First of

all, there is a growing concern about the durability of American strength and willingness to

serve as a power balancer in a region where the balance of power is shifting because of the

rise of China and its perceived threat to American influence. The Sino-U.S. power differentials

Figure 5. Japan’s Trade Dependence on China (1960-2004)

Figure 6. China’s Trade Dependence on Japan (1960-2004)

Sources: The International Monetary Fund, Direction of Trade Statistics; The World Bank, World Development Indicators.

31

variable, defined as the difference of the CINC scores of the U.S. and China, is designed to

measure the potentially destabilizing effect of the changing balance of power between the

two strategic powers. Relatedly, the China’s relative power variable measures China’s share in

the total power equation among the U.S., Japan, and China (China’s CINC/[America’s CINC

+ Japan’s CINC + China’s CINC]).

The joint regime strength variable is coded with a value of 1 if the ruling party in Japan

holds a majority of seats in the Lower House and if the ruling elites in China do not

experience violent political challenges such as insurgent movements, assassination, riots, and

violent demonstrations in a given dyad-year; otherwise, this variable is coded with a value of

0. The joint IGO membership variable is coded with a value of 1 if Japan and China share

common membership in either the WTO or APEC, and coded with a value of 0 if they

share none of the memberships.55 Finally, I consider the lagged dependent variable to

control for time dependence (See Appendix 2 for the coding of the control variables).

Estimation Results

The OLS results reported in Table 2 support the realist claim rather than the liberal peace

argument. Japan’s trade dependence on China has a statistically significant positive impact on the

dispute intensity (coefficient β = 46.754), whereas China’s trade dependence on Japan is

statistically insignificant. This estimation result indicates that the more dependent on China

Japan becomes for trade, the more likely the Senkaku/Diaoyu dispute is to be escalated. It

is plausible that Japan may feel increasingly insecure, as one of its traditional power

advantages—asymmetric trade relations with China in its favor—continues to erode. As

noted above, Japan had always been less dependent on China for trade in both absolute and

55 China joined APEC in 1991 and the GATT/WTO in 2001.

32

relative terms than China had been so on Japan during the Cold War period. Yet the rate

of increase in Japan’s trade dependence on China has dramatically risen since the early 1990s.

Hence Japan’s growing sense of insecurity, coupled with its concern about relative gains

from trade with China, may heighten the probability of the Senkaku/Diaoyu dispute.56

Control variables produced mixed estimation results. China’s relative power has a

statistically significant negative influence on the dispute intensity (coefficient β = -16.389).

It was hypothesized that if a challenger country (i.e., China) experiences faster growth in

capabilities than its target country (i.e., Japan and, to a large extent, the U.S. as well), the

former is more likely to initiate a dispute. By contrast, the estimation result shows that, as

China’s relative power within the Sino-U.S.-Japan triangle increases, the Senkaku/Diaoyu

dispute is less likely to be initiated and escalated.

The hypothesized negative impact of joint regime strength (coefficient β = -.531) is

statistically significant, indicating that regime stability has a pacifying influence on the dispute

intensity. Although the negative coefficient for the Sino-U.S. power differentials indicates that

the U.S. preponderance might have a pacifying influence on the dispute intensity, the level of

statistical confidence is insignificant. Finally, neither joint IGO membership nor the past dispute

intensity has a statistically significant effect on dispute intensity.

56 Conversely, China may feel decreasingly constrained to initiate and escalate the sovereignty dispute

with the improvement of its bargaining position as a result of Japan’s deepening trade dependence

on China. Indeed, territorial and resource competition in the East China Sea has become an even

more serious bilateral irritant than ever before in the context of China’s economic success and

military buildup and the resurgence of Japan’s international assertiveness.

33

Table 2. OLS Results: Senkaku/Diaoyu Dispute (1960-2005)

Explanatory variables Dispute Intensityt Japan’s trade dependence on Chinat-1 46.754 (2.065)** China’s trade dependence on Japant-1 1.834 (.266) Sino-U.S. power differentialst-1 -8.401 (-1.072) China’s relative powert-1 -16.389 (-1.857)* Joint regime strengtht-1 -.531(-2.191)** Joint IGO membershipt-1 -.225 (-.184) Past dispute intensity -.066 (-.065) R2 .474 Durbin-Watson d 2.158

Note: The t-statistics appear in parentheses. * Significant at the .10 level. ** Significant at the .05 level. *** Significant at the .01 level.

3. The Paracel and Spratly Dispute

An Overview of the Paracel and Spratly Dispute

The South China Sea encompasses hundreds of rocks, reefs, and small islands, the majority

of which are located in the Paracel and Spratly archipelagos. The question of who owns

these islands and rocks has become one of the most fundamental barriers to closer relations

in Southeast Asia, particularly since high potentials of oil and gas deposits were discovered

in the late 1960s. By the 1980s, five East Asian countries—including the two Chinas,

Vietnam, the Philippines, and Malaysia—have established some kind of physical presence on

at least one of the islands. While the prohibitive cost of drilling in the deep waters remains

a significant barrier to the utilization of hydrocarbon potentials, the scramble for these

islands looms particularly large during a period of sharply rising energy consumptions in

East Asia, particularly in China. In the South China Sea, China has the most assertive and

controversial claims to these islands. Although Vietnam is not the only party to the many

34

territorial disputes with China in the area, it has had the sharpest differences with China.

As a result, this section focuses on the offshore island dispute between China and Vietnam,

which has unfolded in three distinct rounds of violent clashes in the postwar period.

The First Round of Dispute (1974)

Despite the consistency and firmness of the competing claims of China and (South)

Vietnam vis-à-vis the offshore islands, no military action had been taken until 1974, when

China suddenly asserted its claim to the western Paracel Islands in a forceful manner.

When the smoke cleared within a few days, the South Vietnamese forces were clearly

defeated and forced to withdraw from the Paracel Islands.57 The growing strategic and

economic value of the offshore islands and the decline in U.S. presence created a golden

opportunity for China to act. At the same time, the Soviet Union’s growing interest in the

area greatly alarmed China.58 Unsurprisingly, this period lacked mitigating economic forces

between China and South Vietnam, thereby leading to a more violent clash.

The Second Round of Dispute (1978-79)

The second half of the 1970s marked turbulent Sino-Vietnamese relations. With the

substantial decrease in the Chinese aid after the reunification of Vietnam in 1975, Hanoi

became even closer to the Soviet Union. Deepening ties between Hanoi and Moscow and

their hegemonic ambition in Indochina was far from acceptable to China. It was apparent

that in the immediate aftermath of the 1974 battle for the Paracels, Hanoi departed from its

57 Fravel 2003, pp. 381-382. 58 Catley and Keliat 1997, pp. 75-77.

35

previous position of recognizing China’s sovereignty over the offshore islands.59 Hanoi and

Beijing eventually collided over the island issues as well as other bilateral irritants, resulting in

the 1979 Sino-Vietnamese border war.60

The Third Round of Dispute (1988)

In March 1988, Chinese and Vietnamese naval forces clashed again on Johnson Reef in the

Spratly Islands, following China’s establishment of a physical presence on Fiery Cross Reef

in January. This clash, which cost three Vietnamese ships and seventy four Vietnamese

lives, was China’s first armed conflict in the Spratlys.61 China’s surprise move clearly

demonstrated its growing assertiveness and ambition in dealing with the disputed islands.

The virtual absence of commercial ties between China and Vietnam provided little

deterrence for the violent clash. Yet at the turn of the 1990s, Sino-Vietnamese relations

began to improve dramatically, primarily thanks to Vietnam’s loss of Soviet aid and its dire

need for an alternative source of economic support. Official trade between China and

Vietnam resumed in 1989 and an agreement was reached in November 1991 to normalize

59 Lo 1989, pp. 91-98. 60 From one perspective, the island dispute probably did not play a primary role in the 1979 Sino-

Vietnamese border war. Sino-Vietnamese relations had been dominated by other issues before the

dispute over the islands became a public one. Yet it is important to note the significant secondary

effects of the dispute over the Paracel and Spratly Islands on Sino-Vietnamese relations. Public

breakdown of Sino-Vietnamese relations was foreshadowed by the exchange of hostile rhetoric vis-

à-vis the offshore islands in 1977-78. Given the small patch of border land in dispute, it is likely

that both sides saw the border dispute as a convenient cover under which to express their

dissatisfaction with the other side’s position on the more important island dispute in the South

China Sea. For more details, see Charles McGregor, The Sino-Vietnamese Relationship and the Soviet

Union, Adelphi Paper 232 (London: International Institute for Strategic Studies, 1988), pp. 19-20. 61 Fravel 2003, p. 388-396.

36

their diplomatic relationships.62

Deepening Sino-Vietnamese Rapprochement and Its Prospects

After the 1991 normalization, Vietnam and China continued to engage in heated polemics

over the disputed islands, particularly in 1992 when China issued a concession for oil

exploration within the Vietnamese-claimed western part of the Spratly area to a U.S.

company, Crestone.63 In contrast to previous clashes, however, none of the confrontations

escalated beyond accusations, claims, counterclaims, and minor displays of military force.

On the whole, both China and Vietnam would most likely want to secure a peaceful

management of their territorial disputes because an escalation of the island dispute would

frighten away foreign investors and slow down economic growth, thereby undermining the

basis of their respective regime legitimacy. The signing of the Land Border Treaty in

December 1999, as well as the two treaties (one on fishery cooperation and the other on

maritime delimitation) of the Gulf of Tonkin in December 2000, may stand as evidence of

continued improvement of Sino-Vietnamese relations.64

Operationalizations 62 Catley and Keliat 1997, p. 79; Stein Tønnesson, “Sino-Vietnamese Rapprochement and the South

China Sea Irritant,” Security Dialogue 34: 1 (2003). 63 Valencia 1995, pp. 31-32. 64 Sino-Vietnamese rapprochement reflects a broader trend in the region. At the turn of the new

millennium, engagement between China and the Association of Southeast Asian Nations (ASEAN)

has been truly impressive. At their summit in Phnom Penh, Cambodia in 2002, ASEAN and

China signed a Declaration on the Conduct of Parties in the South China Sea, with the aim of

preventing conflict and promoting cooperation in the region. Although this declaration did not

establish a legally binding code of conduct, it represents a significant step forward (Tønnesson

2003, pp. 55-57).

37

The intensity of the Paracel and Spratly dispute during the period of 1960-2004 is coded on

a five-point scale as illustrated in Figure 7.

0

1

2

3

4

1960 1965 1970 1975 1980 1985 1990 1995 2000

I use China’s trade dependence on Vietnam and Vietnam’s trade dependence on China as

two separate independent variables to measure the dyadic level of economic

interdependence. I use South Vietnam’s figures for the period of 1960-75, since North

Vietnam remained silent on the territorial issues in the South China Sea before the

reunification of Vietnam.65 As shown in Figures 8 and 9, the commercial links between

China and Vietnam were had virtually been severed during the Cold War period except for

the early 1970s. By contrast, mutual trade dependence has been on a dramatic rise since

1991, particularly for Vietnam in both absolute and relative terms.

65 There exists no single source that reports Vietnam’s economic statistics during the Cold War

period. As such, I have relied on a variety of sources including IMF’s Direction of Trade Statistics,

World Bank’s World Development Indicators, WTO’s International Trade Statistics, UN’s Monthly Bulletin of

Statistics, Asian Development Bank’s Key Indicators of Developing Asian and Pacific Countries, and

Vietnam’s official statistics.

Figure 7. Intensity of the Paracel and Spratly Dispute (1960-2004)

38

0.00%

0.05%

0.10%

0.15%

0.20%

0.25%

0.30%

0.35%

1960 1965 1970 1975 1980 1985 1990 1995 2000

0.00%

2.00%

4.00%

6.00%

8.00%

10.00%

12.00%

14.00%

1960 1965 1970 1975 1980 1985 1990 1995 2000

In addition, four categorical and continuous measures are considered as control

variables. The Sino-Vietnamese power differentials variable is defined as the difference of the

CINC scores of the two countries. The joint regime constraints variable is defined as a

geometric mean of the two countries’ Executive Constraints scores in the POLITY IV

dataset.66 The joint IGO membership variable is coded with a value of 1 if China and Vietnam

66 It is hypothesized that this variable is positively related to the initiation and escalation of a dispute

Figure 8. China’s Trade Dependence on Vietnam (1960-2003)

Figure 9. Vietnam’s Trade Dependence on China (1960-2003)

39

share common membership in either the WTO or APEC, and coded with a value of 0 if

they share none of the memberships.67 Finally, I consider the lagged dependent variable to

control for time dependence (See Appendix 3 for the coding of the control variables).

Estimation Results

The OLS results reported in Table 3 support the liberal peace hypothesis. China’s trade

dependence on Vietnam has a statistically significant negative impact on the intensity of the

Paracel and Spratly dispute (coefficient β = -1261.852), although the hypothesized influence

of Vietnam’s trade dependence on China is not statistically significant. This estimation result

indicates that the more dependent on Vietnam China becomes for trade, the less likely the

Paracel and Spratly dispute is to be escalated. Not surprisingly, China and Vietnam

frequently resorted to threat or use of force in dealing with the offshore island disputes in

the 1970s and 80s when they lacked substantial economic ties.

With respect to control variables, estimation results are mixed. The joint regime

constraints variable is statistically significant in its hypothesized causal influence on the

intensity of the Paracel and Spratly dispute. The result for the Sino-Vietnamese power

differentials is not statistically significant, indicating that the intensity of the Paracel and

Spratly dispute has not been affected by the widening gap in national capabilities between

China and Vietnam. Finally, neither the joint IGO membership nor the past dispute occurrence

in accordance with Hypothesis 3.2. Operationally, the Executive Constraints variable in the

POLITY IV refers to the extent of institutionalized constraints on the decision-making power of

chief executives, whether individuals or collectivities. Such limitations may be imposed by any

accountability groups including legislatures, the ruling party in a one-party state, the military in

coup-prone polities, and the like (Marshall and Jaggers 2002, p. 23). 67 Vietnam joined APEC in 1998 and is currently working on its application for WTO admission.

40

variable has had a statistically significant causal impact on the dispute intensity.

Table 3. OLS Results: Paracel and Spratly Dispute (1960-2004)

Explanatory variables Dispute Intensityt China’s trade dependence on Vietnamt-1 -1261.852 (-1.772)* Vietnam’s trade dependence on Chinat-1 15.932 (1.086) Sino-Vietnamese power differentialst-1 40.866 (1.083) Joint regime constraintst-1 .584 (1.813)* Joint IGO membershipt-1 -.392 (-.495) Past dispute intensity .103 (.626) R2 .491 Durbin-Watson d 1.973

Note: The t-statistics appear in parentheses. * Significant at the .10 level. ** Significant at the .05 level. *** Significant at the .01 level.

V. A SUMMARY OF THE FINDINGS AND CONCLUSION

In exploring the three most prominent offshore island disputes in East Asia, I started from

an observation that all of them involve both tangible and intangible values. The fishery

issues have served as a surrogate battlefield for the Dokdo/Takeshima dispute. The

tangible value of the numerous islands in the East and South China Sea had been limited to

fishing recourses until high potentials of oil and gas deposits were discovered at the end of

1960s. The global trend to adopt the UNCLOS further heightened the tangible value of

these otherwise barren islands. More importantly, contested areas raise questions of

national identity and pride, indicating that conflicts over the offshore islands should be

basically non-negotiable and highly salient for the victims of Western and Japanese

colonialism.

After defining the dependent and independent variables, I tested the liberal peace

41

hypothesis using OLS regression. My quantitative evidence of the pacific benefits of

economic interdependence is mixed, but favors the pacific benefits of economic

interdependence. In the Senkaku/Diaoyu dispute, a realist claim finds support for the

theory that interdependence fosters conflict rather than peace. Yet in the cases of the

Dokdo/Takeshima and the Paracel and Spratly disputes, the liberal peace argument is

supported with evidence that higher levels of trade dependence of one party on the other

alleviate the degree of hostility.

I also considered some other causal factors to control for spurious relationships

between the dependent and independent variables. Largely consistent with existing

explanations, I found that joint democracy has a pacifying effect on the process of the

Dokdo/Takeshima dispute, while joint regime weakness has an aggravating influence on the

Senkaku/Diaoyu and the Paracel and Spratly disputes. Yet I discovered little causal

evidence of structural factors—such as alliances and relative power differentials—on the

dispute intensity. Finally, I found that joint membership in IGOs has little determining

impact on the dispute process.68

In a region where the balance of power is shifting because of the rise of China and

the stagnation of Japan, the disputes over the offshore islands have significant regional

repercussions. In a new millennium, China and Japan both present a different landscape of

the future of East Asian territorial disputes since they have both geopolitical and territorial

68 Yet as with any kind of regression analysis with a relatively small N, these statistical findings

require a cautionary interpretation. Also, given the complexity surrounding these island disputes,

there remain stories to be explained in an in-depth, qualitative manner, as I attempted elsewhere.

For more details, see Min Gyo Koo, Scramble for the Rocks: The Disputes over the Dokdo/Takeshima,

Senkaku/Diaoyu, and Paracel and Spratly Islands, Ph.D. Dissertation in Political Science (Berkeley:

University of California, 2005).

42

ambitions. Not all is grim in the world of East Asian island disputes, however. I argue

that economic interdependence can mitigate the intensity of persistent island disputes. For

East Asian countries, the question is now how they institutionalize their deepening and

widening economic interdependence to get more mileage out of it. The burgeoning efforts

to form FTAs amongst themselves generate positive expectations in this regard.69

In view of the rising pressure of territorial nationalism and fluid geopolitics in a

post-September world, even the mere containment of the territorial disputes in East Asia

requires the skillful diplomacy and tireless efforts of political leaders. The findings of this

paper indicate that one of the best strategies to hedge against potentially disruptive

behaviors is to engage more with each other by facilitating greater economic

interdependence. In conclusion, the path to deeper and wider economic interdependence

in East Asia is likely to be a bumpy one, but the institutionalization of economic integration

may help to smooth it.

69 Besides standard welfare calculations, an emerging interest in FTAs provides the two regional

giants with a convenient venue in which to vie for regional economic leadership. This point was

driven home most dramatically with regards to ASEAN. In February 2003, China signed a

surprise agreement with its Southeast Asian neighbors to form an ASEAN-China FTA by 2010.

Challenged to do the same, and to demonstrate its capability to continue to lead within East Asian

economies, Japan proposed a Japan-ASEAN FTA. For more details, see Vinod K. Aggarwal and

Min Gyo Koo, “Beyond Network Power? The Dynamics of Formal Economic Integration in

Northeast Asia,” The Pacific Review 18: 2 (2005), pp. 189-216.

43

Appendix 1. Coding of Control Variables: The Dokdo/Takeshima Dispute

U.S. defense commitment Joint regime strength Joint democracy Joint IGO membership

1953 1 0 0 0 1954 1 0 0 0 1955 1 1 0 0 1956 1 1 0 0 1957 1 1 0 0 1958 1 1 0 0 1959 1 1 0 0 1960 1 0 0 0 1961 0 0 1 0 1962 0 1 0 0 1963 1 1 0 0 1964 1 1 0 0 1965 1 1 0 0 1966 1 1 0 0 1967 1 1 0 1 1968 1 1 0 1 1969 0 1 0 1 1970 0 1 0 1 1971 0 1 0 1 1972 0 1 0 1 1973 0 1 0 1 1974 1 1 0 1 1975 1 1 0 1 1976 1 0 0 1 1977 0 0 0 1 1978 0 0 0 1 1979 0 0 0 1 1980 0 1 0 1 1981 1 1 0 1 1982 1 1 0 1 1983 1 1 0 1 1984 1 1 0 1 1985 1 1 0 1 1986 1 1 0 1 1987 1 1 0 1 1988 1 0 1 1 1989 1 0 1 1 1990 1 0 1 1 1991 1 1 1 1 1992 1 1 1 1 1993 0 0 1 1 1994 0 0 1 1 1995 0 0 1 1 1996 0 0 1 1 1997 0 0 1 1 1998 1 0 1 1 1999 1 0 1 1 2000 1 0 1 1 2001 0 0 1 1 2002 0 0 1 1 2003 0 0 1 1 2004 0 0 1 1 2005 0 0 1 1

44

Appendix 2. Coding of Control Variables: The Senkaku/Diaoyu Dispute

Sino-U.S. power differentials* China’s relative power** Joint regime strength Joint IGO membership

1960 0.18 0.32 1 0 1961 0.17 0.30 1 0 1962 0.17 0.29 1 0 1963 0.17 0.30 1 0 1964 0.16 0.31 1 0 1965 0.16 0.31 1 0 1966 0.17 0.30 1 0 1967 0.16 0.29 1 0 1968 0.15 0.29 1 0 1969 0.15 0.30 1 0 1970 0.13 0.32 1 0 1971 0.12 0.33 1 0 1972 0.11 0.34 1 0 1973 0.10 0.34 1 0 1974 0.10 0.35 1 0 1975 0.09 0.37 1 0 1976 0.09 0.37 0 0 1977 0.09 0.37 0 0 1978 0.09 0.38 0 0 1979 0.08 0.38 0 0 1980 0.08 0.39 1 0 1981 0.09 0.38 1 0 1982 0.08 0.39 1 0 1983 0.08 0.39 1 0 1984 0.09 0.38 1 0 1985 0.09 0.37 1 0 1986 0.09 0.37 1 0 1987 0.09 0.37 1 0 1988 0.09 0.37 1 0 1989 0.10 0.36 0 0 1990 0.09 0.36 0 0 1991 0.08 0.38 1 1 1992 0.10 0.37 1 1 1993 0.10 0.37 0 1 1994 0.10 0.38 0 1 1995 0.09 0.39 0 1 1996 0.09 0.39 0 1 1997 0.09 0.39 0 1 1998 0.10 0.39 0 1 1999 0.10 0.39 0 1 2000 0.10 0.39 0 1 2001 0.10 0.40 0 1 2002 0.10 0.40 0 1 2003 0.10 0.41 0 1 2004 0.10 0.41 0 1 2005 0.10 0.42 0 1

* Sino-U.S. power differentials=America’s CINC-China’s CINC; ** China’s relative power=China’s CINC/(America’s CINC+Japan’s CINC+China’s CINC); Figures for 2002-05 are estimates (Source: National Material Capabilities (V. 3.0)).

45

Appendix 3. Coding of Control Variables: The Paracel and Spratly Dispute

Sino-Vietnamese power differentials* Joint regime constraints Joint IGO membership

1960 0.12 2.45 0 1961 0.10 2.45 0 1962 0.10 2.45 0 1963 0.10 1.41 0 1964 0.11 1.41 0 1965 0.10 1.41 0 1966 0.10 1.00 0 1967 0.10 1.73 0 1968 0.10 1.73 0 1969 0.10 2.45 0 1970 0.10 2.45 0 1971 0.10 2.45 0 1972 0.10 2.45 0 1973 0.10 2.45 0 1974 0.11 2.45 0 1975 0.11 2.45 0 1976 0.11 3.00 0 1977 0.11 3.00 0 1978 0.11 3.00 0 1979 0.11 3.00 0 1980 0.11 3.00 0 1981 0.11 3.00 0 1982 0.10 3.00 0 1983 0.10 3.00 0 1984 0.10 3.00 0 1985 0.10 3.00 0 1986 0.10 3.00 0 1987 0.10 3.00 0 1988 0.10 3.00 0 1989 0.10 3.00 0 1990 0.10 3.00 0 1991 0.10 3.00 0 1992 0.11 3.00 0 1993 0.11 3.00 0 1994 0.12 3.00 0 1995 0.12 3.00 0 1996 0.12 3.00 0 1997 0.12 3.00 0 1998 0.12 3.00 1 1999 0.12 3.00 1 2000 0.12 3.00 1 2001 0.13 3.00 1 2002 0.13 3.00 1 2003 0.13 3.00 1 2004 0.13 3.00 1

* Figures for 2002-04 are estimates.