Foreign Policy Analysis: Continuity and Change in Its Second Generation

F:relgI1 PI:ic:y ~CJ

-rz r-z -rz Z4- z r-J"

CZ

z-rz Ir-s SecCJ-rza

L«1L1r«1

~e«1c:k

-rz a

~I1c::Llysls ~h

cz -rz,ge

~e-rzerczr-zCJ-rz

Iea1111e

~_

~_

:E----:I:ey

P «1 t r i c:k I _ :E----:I: «111 e y

1?~EJ:'-..T--rIC=E

:I--I:~LL"

EYlglevv-c>c>cl C=liffs" J:'-..Tevv- : J e r s e y 0 7 6 3 2

I !Imlry afCongress Cata/ogJng-lIJ-l'lIh/icatlO/I 1

dr.l

foreign polley analysIs: continuity and change 111 its second gel1er,ltlOlliedited by Laura Nedck, jeanne A. K. 1ky, Patl ick J. I laney. p. em. IncludeS bibiiogrdphiutl references and index. ISBN 0-1 3-060575· I 1. internallonal relations-Re:.earch. 2. internatIOnal relations~ludy and tedLhing. I. NeaLk, Laura. II. Hey, jedflIJe A. K. III. Haney, Patri(.k, Jude. jX129U·663 1995 j27.1'UI-dc20 94-38260

CIP

Chapter 2 was adapted and expllnded from Comparative Politics, Policy, and International Relations edited by William Crotty. Copyright @ 1991 by Northwestem University Press. Reprinted with permission. Chapter 3 was adllpted Ilnd expanded from Defining Power: Influence and Force in the Contemporary International System by John M. Rothgeb, Jr. Copyright @ 1993. Reprinted with permission of St. Mllrtin's Press, Incorpomted.

Editorial/production supervision: Lauren Byrne Editorial director: Charlyce jones Owen Cover design: Maureen Eide Buyer: Bob Anderson

@1995 by Prentice-Hall, Inc. A Simun & Schuster Company Englewood Cliffs, New jersey 07632

All rights reserved. No part of this book may be reproduced, in any form or by any means, without permission in writing from the publisher.

Printed in the United States of America 10 9 8 7 6 5 4 3 2 I

ISBN 0-13-060575-1 PRENTICE-HALL INTERNATIONAL

(UK)

PRENTICE-HALL CA.NADA INC.,

Toronto

PRFNT1CE- I IALL HISPANOAMERICANA, S.A., PRE"ITICE-HAI.1. or INDIA PRIVATE LIMITED,

Mexico New Delhi

Tokyo Sillgnpol'e BRASIL, LrDA., J\Iilhelmine Germany see Kennedy (1980), Lamborn (1991), and Snyder (1991 ). 25. Compared to pre-World War I Germany, cartelization appears to have been more fluid, with factionalism within major actors as well as ties with various civilian players. On Japanese foreign policy making in the 1930s, see Fukui (1977b), Hosoya (1976), and Snyder (-1991-) . 26. The concept of the predominant leader is developed by Margaret Hermann in Hermann, Hermann, and Hagan (1987) and in Hagan, Hermann, and Hermann (n.d.). Historical examples are Germany's Bismarck and Hitler and the Soviet Union's Stalin, while contemporary cases are Cuba's Castro, North Korea's Kim ll-Sung, Iraq's Sadaam Hussein, Syria's Assad, as well as the first generation of many African postcolonial rulers. 27. Political survival is pervasive in foreign policy explanations of war (e.g., Levy 1989; Lebow 1981). The task of retaining power is central to the logic of various general foreign policy models (e.g., Hudson, Sims, and Thomas 1993; Salmore and Salmore 1978; Van Belle 1993), particularly the extensive empirical research on the linkage between domestic conflict and foreign conflict (e.g., James 1988; Wilkenfeld 1973). 28. Other well-documented cases of foreign policy making by similarly unstable coalitions include Syria prior to the June 1967 \>\Iar (Bar-Siman-Tov 1983), Indonesia's confrontation with the \>\Iest under Sukarno (vVeinstein 1976), Argentina's invasion of the Falkland Islands (Levy and Vakili 1990), and China during the Cultural Revolution (Hinton 1972). 29. RissecKappen's (1991) model of public opinion and foreign policy effectively combines the dynamics of political survival and coalition building, showing that the latter is constrained in cohesive governments, which are then able to contain the pressures of public opposition. 30. In still other situations the tasks of building policy coalitions and retaining political power may be different, with the result that leaders may be able to act in ways that either game might separately suggest. Leaders, for example, might pursue a publicly unpopular policy because the regime is sufficiently cohesive to carry out its strong preferences, or because one of the critical members of the regime threatens to defect and bring down the regime immediately if it does not act. Another situation is where a severely fragmented regime engages in unexpectedly strong (though often erratic) foreign policy activit y. This occurs when deadlocked actors within the regime seek to alter the situation by appealing to political groups outside the regime, often with virulent appeals to public nationalism and other unifying nationalist and ideological themes. 31. Accommodation via bargaining underlies statist approaches in the political economy literature (e.g., Ikenberry, Lake, and Mastanduno 1988; Katzenstein 1976; Krasner 1978) and models of economic policymaking (Destler 1986; Vernon, Spar, and Tobin 1991). 32. Compromise is a central theme in the bureaucratic politics case research on U.S. national security policy, as cited in note 3.

142

Sccolld-Gellcralioll Forei)!,11 I'olicy Analysis

33. Much of the international relations theory on pacific democracies rests upon the idea that domestic politics inhibits foreign policy makers (e.g., Doyle 1986). Lamborn (1991) treats domestic politics as a constraint on the risk propensities of both democratic and authoritarian belligerents. 34. This is true of the additional cases of unstable coalitions, as cited in note 29. 35. This is a critical point in any theory that incorporates domestic political phenomena into a broader theory of foreign policy. Any assessment of domestic political constraints must take into account that these pressures are balanced international imperatives and risks. Frameworks in this direction are developed by Lamborn (1991) and by Hudson, Sims, and Thomas (1993). 36. Nincic (1992), Snyder (1991), and Russett (1990) suggest that these dynamics can be contained by democratic systems before they provoke major wars or irreversible overextension of power. 37. These issues were then handled by senior ministerial bureaucrats, although it is important to note that these actors too were divided on the issue and unable to act much more effectively. 38. This definition is broadly based on Callahan (1982). Discussions of overextended commitments in international affairs are found in Snyder (1991) and Kennedy (1987), while Stein (1993) develops the theme of underextension. 39. The concept of risk taking in foreign policy is developed in Bueno de Mesquita (1981) and Lamborn (1991). 40. The articulation of the logic of this approach draws upon Hermann (I 987b). 41. See Hagan (1993) for a complete discussion of the conceptualization and operationalization of organized opposition in terms of the following three properties. Mass-level opposition is the focus of cross-national analyses of domestic and foreign conflict (e.g., Wilkenfdd 1973; James 1988). 42. Of course, the importance of democratic constraints is demonstrated by the extensive literature that shows that democracies rarely fight wars with each other, although they are as war prone as authoritarian systems when fighting nondemocracies. See the literature cited in the first section of this chapter as well as the more extended discussion in Hagan (1994). 43. Several studies point to the importance of institutionalization as a broad foreign policy influence. Snyder's (1991) analysis of five cases of cartelized regimes shows that these pressures were contained in relatively institutionalized systems (mid-eighteenthcentury Britain and the US. and US.S.R. during the cold war) but were not contained in less institutionalized systems (pre-World War I Germany and Japan in the 1930s). Maoz (1989) finds that newly established polities (both democratic and authoritarian) created through revolutionary processes are more prone to international disputes, as compared with more established systems undergoing evolutionary change. Hagan's analysis (1993) finds political influences on foreign policy are most pronounced for regimes in moderately institutionalized political systems, whose leaders have neither the power to suppress opposition nor the norms of consensus building to work with them. Finally, across the case study literature of the politics of foreign policy in non-US. settings, the contextual effects of institutionalization would seem as important as those of democratization (Hagan 1993, chapter 2). 44. It is widely accepted, for example, that in "crisis" situations policymaking becomes concentrated at the highest levels of government and debate is minimized by the pressures of high threat, short decision-making time, and surprise. Others argue that "redistributive" (Lowi 1967) and "institutionally grounded" (Art 1973) issues involve a greater

dispersion of authority and more intense contlict than do others of a less "zero-sum" nature. 45. Formal definitions of the three decision units are as follows. A predominant leader is a single individual with the authority to commit,. irreversibly, natIOnal resources 111 • response to a problem. A single group IS an entlty of two or more people whc: lllte~act directly with all other members and who collectlvely are able to reach a deCISion without consulting outside entities (e.g., cabinets, juntas, and politburos). A coalitIOn of autonomous groups has two or more politically autonomo~s groups (e.g:, p~lrtJes: institutions/bureaucracies), none of which alone has the ability to COlTI1l11t natIOnal resources without the support of all or some of the other actors. See I lagan (1993, . 218-23) for a full discussion of adapting the decision- U11lts approach to understandmg how leaders respond to domestic opposition in foreign policy.

NINE

Event Data in Foreign Policy Analysis Philip A. Schrodt,

UNIVERSITY OF KANSAS

II! Editors'Introduction In this chapter Philip Schrodt reviews the development and use Q[ event data to study joreign policy. Schrodt shows how the use oj event data to study (usually) interstate behavior has been a part oj the study ojjoreign policy jrom the beginnings Q[ the Jield in the 1960s and continues to be important in the second generation Q[ joreign policy analysis. He also discusses how new developments in theories. concepts, and technologies have changed how event data are generated, stored, and used; and how the questions addressed with event data have changed with the field's development. Schrodt discusses data sets thatJocus on actors (e.g., Great Britain, Iraq, the PLO, the UN) and others that Jocus on episodes (e.g., wars, crLses, negotiation). These event data sets have provided the evidence oJ Joreign policy jor many who Jocus their research on the interaction among actors in world politics. Schrodt argues that in the second generation oj Joreign policy analysis, those using event data are trying to keep pace notJust with technological innovations but also with conceptual changes as those in the field try to understand more complex processes and include more and d!1Jerent actors in the data sets. Questions the reader should focus on when reading this chapter include the jollowing: What are event data? What d!1Jerent types Q[ event data are there? What types Q[joreign policy questions can be addressed with event data? What are some oj the shortcomings oj this approach to studying joreign policy? How are the changes in world politics and technology changing this methodology? And what types oj interesting Joreign problems can you think oj that may be able to be studied with this type Q[ method? •

Foreign policy analysis developed at about the same time as the behavioral approach in political science. The objective of the behavioralists was to study political behavior using systematically measured variables, statistical techniques, and unambiguously stated hypotheses. In some areas of political science, the behavioralist studies used

145

147 measurement techniques that had been developed earlier. For example, researchers attempting to model elections found that the traditional questions asked of potential voters in survey research-party affiliation, whether they had voted before, who they were planning to vote for, and so forth-provided a useful foundation for their studies .. While the statistical methodologies and survey methods used in contemporary votmg research are substantially more sophisticated than the voting surveys of the 1920s and 1930s, the basic measurement instrument-the public opinion survey-is the same. No equivalent data existed in the field of foreign policy analysis. Traditional studies of foreign policy primarily used narrative sources such as documents, histories, and memoirs, and there was no way to directly analyze these in a statistical framework. This disjuncture necessitated the development of new methods for generating data. A variety of these methods have been discussed in the other chapters of this volume; this chapter will focus on one of the most commonly used measurement techniques of foreign policy behavior, event data. The basis of many studies of foreign policy is the fundalTlental question "Who did what to whom?" For example, during the Nixon administration (1969-74), the United States and the Soviet Union had a relaxation of diplomatic tensions known as the detente period. This was reflected in a variety of foreign policy actions, including arms control agreements, a decrease in hostile rhetoric, increased trade, and increased cooperation in resolving disputes. A decision maker living during this period would have a general perception that the hostility between the two superpowers had decreased. This perception would be based on a general pattern of cooperative interaction, rather than on a single incident. EVeJZt data are a formal method of measuring the phenomena that contribute to foreign policy perceptions. Event data are generated by examining thousands of newspaper reports on the day-to-day interactions of nation-states and assigning each reported interaction a numerical score or a categorical code. For example, if two countries sign a trade agreement, that interaction might be assigned a numerical score

of +5, whereas if the two countries broke off diplomatic relations, that would be a nU.me.rica.1 score of -8. When these reports are averaged over time, they provIde a rough mdlcatlOn of the level of cooperation and conflict between the two states. Figure 9.1 shows the actions that the United States directed toward the Soviet Union for the period 1948-78 as measured by the Conflict and Peace Data Bank (COPDAB) event data set collected by the late Edward Azar (1980,1982). In the COPDAB coding scheme, negative numbers indicate conflictual behavior; positive numbers indicate cooperation. COPDAB is based on the New York Times and a variety of regional newspaper sources; the data cover the period 1948-78. The COPDAB time series shows three general periods. The early cold war (1948-62) is characterized by uniformly negative relations, though these are more stable in the late 1950s than in the early 1950s. A partial "thaw" occurs in 1962-70 following the Cuban missile crisis, with the relationship being neutral. Finally, the 1970-78 period shows the rise and fall of the detente policy. Other event data sets cov~ring the 1980s. record the "new cold war" of the early Reagan period followed by the Improved relatIons that occur when Gorbachev comes to power in the USSR (see Goldstein and Freeman 1990). The event data record of U.S.-Soviet interactions corresponds closely to the patterns one would expect from a historical study. Moreover, the event data can also be used to fine-tune that chronology. For example, while Nixon clearly intended to implement a detente policy from the beginning of his administration in 1969, there was continued disagreement between the United States and Soviet Union over U.S. involvement in Vietnam, the 1968 Soviet invasion of Czechoslovakia, and other issues, so the interaction pattern is not actually positive until 1971. Positive interactions peak about the time of Nixon's resignation in 1974; the event data scores then decline during the two years of the Ford administration and return to post-Cuban missile crisis levels by 1976. Figure 9.2 shows another example of the use of event data to chart the evolution of a complex international interaction, the Palestinian intifada (uprising) that a~signed

Figure 9.2 Figure 9.1

Israeli-Palestinian Interactions, 1982-92

U.S. Actions Toward tlze Soviet Union, 1948-78

o

100

50

-100

o

-200

-50

-300

-100

-400

------ ISR>PAL ---PAL>ISR

-ISO

-500 -200 -600

-250 00

.".

;:::

0

m

;:::

01

.".

'.0

00

;:::

;:::

~

;:::

m

en

m

en

0

01

a--

;:::

'.0

'.0

"1" '.0

;:::

'.0 '.0

""

00

'.0

;:::

Source: COPDAll scores in Goldstein and I'rccman (J 990,162).

0

"-

'"

01

"-

;:::

.".

'-C

;:::

;:::

"-

,~

00

";:::

'" N 00

"0

0

"0

0

"-

0

0

J., N c0 c0 -J. -J. 00 00 00 00 00 00

"-

'" m

00

Source: Schrodt and Gerner (1994).

0

-0 00

"-

'" '.0

00

0

"0

r:.. 00 r:..

00

0

00 00

"0

'"

"0

0

gg '" a, cO "" 00

00

"0

cO

a--

"-

"-

''"" '" '" '" 0:;

N

a--

01

a--

148

SeCOlln-Cellerntioll Foreign Policy Analysis

began in December 1987. This chart is based on the coding of news stories on IsraeliPalestinian interactions reported by the Reuters international news agency. These reports were automatically coded by a specialized computer program into the World Events Interaction Survey (WEIS) event data categories developed by Charles McClelland (1976). The categorical WETS events were then converted to a monthly numerical score using a scale devised by Goldstein (1992); as in Figure 9.1, negative scores indicate conflict and positive scores indicate cooperation. This time series shows the pattern of interactions-largely uses of force-in considerable detail. The initial increase in conflictual activity in 1982-83 corresponds to Israel's invasion of Lebanon, which was initially directed against Palestine Liberation Organization forces. The invasion is followed by a period of five years of relative quiet, though a separate series of event data on Israel's interactions with Lebanon during this period shows a great deal of conflict as opposition to Israeli forces shifts from the PLO to various Lebanese groups. The intifada begins abruptly in December 1987 and then gradually declines over the next five years, though there is another upsurge in violence following the election of a Labor government in Israel in the summer of 1992. As with the case of the U.S.-Soviet interactions, this time series gives a more exact measure of the patterns of events over time. For example, while the intifada follows a lull in conflict during the summer of 1987, the event data also show a general increase in conflict beginning about eighteen months earlier. This increase may have been a precursor to the larger uprising (see Schrodt and Gerner 1994). As these two figures illustrate, event data can be used to summarize the overall relationship between two countries over time. The patterns shown by event data usually correspond to the narrative summaries of the interactions found in historical sources, but unlike narrative accounts, event data can be subjected to statistical analysis. As a consequence, event data are frequently used to study foreign policy outcomes and some characteristics of the international environment within which foreign policy decisions occur.

III Creating Event Data The creation of event data is basically a process of content analysis (see Krippendorff 1980; Weber 1990) and inyolves three steps. first, a source or sources of news about political interactions is identified. This could be an internationally oriented newspaper such as the New York Times, a set of regional newspapers and news magazines, a news summary such as Facts on File or Deadline Data 011 World Affairs, or a news wire service such as Reuters or the Associated Press. As will be discussed later, the choice of the event source can have a substanti