A Post-Cold War Clash of Civilizations?A New Approach to Testing Huntington’s Thesis
1 BackgroundA major area of interest after the Cold War, among academics as well as world leaders, has been to reveal new sources of armed conflict. What will be the causes of war in the post-Cold War era? The “grand theories” of international relations have been applied also in this field, although the debate has been more focused on the extent of war in the new era rather than on specific causes. While liberal institutionalism claims that armed conflict can be avoided the next decade with a “continuous pattern of cooperation” (Keohane 1993), realists like John Mearsheimer, claim that there is no promise of more stability going from a bi-polar world to a multi-polar world (Mearsheimer 1995). Some empirical research has been conducted in order to detect changes when proceeding to a new era. For instance, Oneal & Russet (1999: 36) conclude: “The post-Cold War era is full of affirmations about the importance of democracy, freedom, and prosperity built on interdependent markets.” And further: “Analytically, we are progressing towards a synthesis of Kantian and realist influences and of dyadic and systemic perspectives” (Oneal & Russett 1999:36). Fukuyama (1992) has perhaps the most commonly known contribution claiming that the end of the Cold War marks the “The End of History”. The traditional lines, and eventually all lines of conflict, will disappear, according to Fukuyama, as the nation-states acknowledge the “universal principles” of democracy and liberal markets. It will be the end of all international wars, except minor conflicts between the states that are still “in history”, and the ones that have reached the “end of history”.
On this background, Samuel P. Huntington created great controversy with his article “The Clash of Civilizations?” in 1993 and the follow-up book The Clash of Civilizations and the Remaking of World Order in 1996. Instead of harmony in “the new era”, Huntington focuses on emerging conflicts based on cultural dissimilarities. He makes precise predictions about causes of conflict and wars after the fall of the Iron Curtain: “It is my hypothesis that the fundamental source of conflict in this new world will not be primarily ideological or primarily economic. The great divisions among humankind and the dominating source of conflict will be cultural.” (Huntington 1993: 22). This proposition started a discussion that reaches far beyond academic circles. After the events of September 11 2001, Huntington’s theories are discussed with renewed enthusiasm.
2 Overall Objective and Research QuestionsThe overall objective of this thesis is to investigate into the relationship between culture (religious, ethnic and linguistic differences) and conflict. As a point of departure, I choose to test the validity of Huntington’s “civilizations theory”. This is a theory that emphasizes the importance of culture as a determinant of conflict, and it holds many explicit propositions concerning the relationship between culture and conflict. Specifically, I seek to answer the following questions, derived form Huntington’s theory: Do conflicts occur along the fault lines of civilizations? And if they do – is this a phenomenon strictly evident in the post-Cold War era? Are certain civilizations significantly more prone to conflict than others? Are the Muslim and Western civilizations more prone to conflict than other pairs of civilizations? If the empirical findings suggest that the answer to these questions is yes (hence, support for Huntington’s thesis), this calls for a further inquiry into: What are the mechanisms that explain how cultural dissimilarity may be a cause of armed conflict?
Obviously, these are important questions to decision-makers who deal with foreign affairs and multilateral matters. But also academically, these questions are of high interest. Although Huntington’s civilizations theory has met virtually nothing but criticism in quantitative assessments (Henderson 1998; Henderson & Tucker: 2001; Russett, Oneal & Cox 2000), Huntington claims that their results are due to methodological shortcomings (Huntington 2000). Hence, one objective is to contribute with a more appropriate methodological approach to testing Huntington’s theory. Second, I aim to contribute to the theoretical debate between primordialists, instrumentalists and constructivists on how cultural dissimilarity is linked to armed conflict (see Chapter 2). Through the findings of my analysis I wish to shed more light on this relationship and hopefully contribute to the research in this field (Carment & James 1995; Ellingsen 2000; Fox 2001; Geertz 1973; Glazer and Moynihan 1975; Horowitz 1985; Väyrynen 1994).
3 TheoryHuntington’s civilizations theory is first and foremost an attempt to describe the emerging world order after the collapse of the bi-polarity structure during the Cold War (Huntington 1996: 13-14). Huntington suggests that, as part of an evolution from earlier eras of world history, the post-Cold War era will mark a turning point, following which “culture” is the key to understanding the nature of world trade, world politics, co-operation – and conflict. Huntington claims that, in the new era, the factor determining the nature of conflicts is cultural. Huntington suggests that identity, the question of “who am I?”, will become increasingly important to all individuals after the Cold War. This change at the individual level will eventually reach the level of the broadest cultural identities – civilizations. Thus, civilizations will be dominant actors in world affairs. “Clashes” of these cultural entities will, allegedly, be the primary source of armed conflict. In Huntington’s words: “The conflicts of the future will occur along the cultural fault lines separating civilizations” (Huntington 1993: 25).
In a theoretical setting, one may argue that this theoretical view belongs to the primordial “school” of conflict, although Huntington does not present an entirely mono-causal explanation. In the primordial view, cultural dissimilarity is a traditional feature that is naturally linked to conflict. This perspective is contested by instrumentalists, who claim that elites of society construct cultural differences in order to mobilize the masses and use them as instruments to pursue their personal agenda. The constructivists oppose Huntington from yet a different angle: they agree that cultural dissimilarity may play a role as a cause of armed conflict, but only when facilitated by certain socio-economic background factors.
4 Research Design and MethodologyI choose to explain the research design step by step in accordance with the framework (content) of the thesis. Since I aim to test the validity of the civilizations theory, the first step will be to present and to scrutinize the theory. What are its premises, how does it relate to other theories in this field, what is the main critique against this theory? This will provide a basis for later discussion of the validity of the theory in light of the results. Next, I pinpoint the specific areas of the theory that I would like to test (the causes and nature of conflicts). I then derive verifiable “empirical statements” from Huntington’s specific theoretical statements. This means that I formulate a certain expectation of an empirical phenomenon that occurs, or does not occur, provided that Huntington’s theory corresponds with reality. In other words: “After this logical deduction of empirical statements from the theory, one can make the observations called for in the empirical statements to see whether or not they are true” (Stinchcombe, 1968: 15-17). The formulations of empirical statements are hypotheses that I set out to test empirically in the analysis.
The analysis seeks to test Huntington’s theory, but at the same time it is a design for generalizing about the phenomenon “armed conflict” within a certain period. Since there are many cases of conflict in the relevant period (before and after the Cold War), it is probably not appropriate to apply any sort of case study to test the validity of this theory. The selection bias of a case study and possibility of tendentious relationships calls for a large N-study. I choose as units all armed conflicts from the PRIO/Uppsala dataset ten years of the Cold War and ten years after the Cold War – 1979-2000. From Huntington’s theory I have derived certain expectations (hypotheses) of the features of these conflicts. The analysis tests the existence of a certain relationship between the locations and nature of conflicts, and cultural borders. I do so by expecting certain correlations and properties of the measured variables. One very central variable in the analysis is “distance”, which I have generated (see Chapter 3). This variable is simply a ranking of the conflicts: to what extent does a conflict “occur along the fault lines of civilizations”. Every conflict is measured by its distance to the nearest civilizational fault line, or border. According to the hypotheses, this variable is expected to have certain properties, when controlled for pre- and post-Cold War, and bivariate correlations with other relevant variables in the dataset.The most central question to be answered through the quantitative analysis is whether conflicts are more frequent along the fault lines of civilizations, since this refers to Huntington’s overall claim. Another essential question is whether such a relationship is more evident in the post-Cold War era, like Huntington claims.
I have already pointed out that a case study approach is not suitable for testing the general validity of Huntington’s thesis. However, in order to investigate into the micro foundations behind the existence, or non-existence, of a relationship between cultural difference and armed conflict, such an approach is fruitful. A limited case study will provide a basis upon which to interpret the results in a suitable context and to provide a better understanding of the matter studied. Whereas the initial quantitative analysis tests whether there is such a relationship, i.e. the validity of Huntington’s overall claim, a case study may indicate how cultural differences may, or may not, play a role as a (primary) cause of armed conflict. For this purpose, I choose to study the causes of the war in Bosnia 1992-95 (Chapter 5). I stress, however, that the quantitative test is the main focus of this thesis; the case study is primarily intended to be a basis for discussing the results in an appropriate context. Therefore, the whole of Chapter 3 is dedicated to explaining the methodology of the quantitative test, focusing on the validity and the reliability. Regarding methodological aspects of the case study, I will mention these in the introduction of Chapter 5. The case study is very limited and may just as well be viewed as an extended discussion of the large-N analysis and its results.
5 FindingsThe quantitative analysis provided the following results:
H1: Conflicts more frequently occur along the fault lines of civilizations [Support]
H2: The conflicts of 1990-2000 are generally located closer to the civilizational borders than the conflicts of 1979-89. [Support]
H3: The closer a conflict is located to a civilizational border, the more likely the conflict has a high number of battle deaths. [No support]
H4: The closer a conflict is located to a civilizational border, the more likely the conflict is over territory. [Support]
H5: Intrastate wars are generally closer located to the civilizational borders than interstate and internationalized civil wars. [No support]
H6 a: The Western civilization is more than averagely exposed to conflict. [No support]
H6 b: The Muslim civilization is more than averagely exposed to conflict. [Support]
H6 c: The Buddhist civilization is less than averagely exposed to conflict. [Support]
H6 d: The Western and Muslim civilizations are particularly prone to fight each other. [No Support]
Through the discussion of the results, I pointed out that the support for H1 and H2 should be interpreted in the way that there is a relationship between conflict and difference of cultural identity. This is even more evident after the Cold War. This means that Huntington’s premise that cultural identity is ever more important at the individual level, and consequently affects the pattern of armed conflict, may describe some features of reality. However, it is important to keep in mind that socio-economic background variables, or type of political regime, which are known to correlate with conflict, are not controlled for. H4 gained support and is an expression for cultural conflict often being fought over territorial claims, whereas Huntington is wrong about cultural conflicts being more violent (H3). The result of H5 may give an indication that, methodologically, different types of conflict should perhaps not be studied separately; perhaps they are to some extent the same phenomenon. The discussion of H6 A-D holds that even though there may be some evidence supporting Huntington’s overall claim (H1 & H2), his analysis of inter-civilizational mechanisms holds very little correspondence to reality. On the contrary, I detected many other relevant factors that may explain the variation of exposition of conflict across civilizations just as well, or even better. I also pointed out that an analysis of inter-civilizational relations involves so many possible background variables that it is virtually impossible to say anything certain about this.
As a natural extension of the results from the quantitative analysis, I first reviewed the ten conflicts closest conflict areas to civilizational borderlines as an introduction to looking at the cases qualitatively. Next, I selected a case, the war in Bosnia 1992-95, which follows the general tendency. I attempted to assess the mechanisms that caused the conflict. I found that even though this conflict was fought along ethnic lines, and that ethnic-religious dissimilarity and hatred is one of the causes of this conflict, the picture seems to be much more complex. First, there are indications that the political leaders, on all sides, played on ethnic strings in order to mobilize “their people” and to fight “the others”. At the same time, it is questionable whether the objectives for which the people were mobilized were actually in their own interest. Political leaders may have made use of the psychology of “fear of domination” in order to gain support. This does not, however, prove that potential ethnic tension did not exist before the war. Rather, it urges us to identify what conditions give room for elites to play on ethnic strings in a society. Much evidence points at the economic and political crisis in which Bosnia, and its surroundings, found itself. The transition to a new political system and the ineffectiveness in handling the economic crisis may have swept the grounds for interest-driven politicians to stir up ethnic tension. Lastly, I found it rather fruitful to distinguish between the various types of causes of conflict, as suggested by Dessler (1994) (triggers, targets, channels and catalysts).
A conclusion of my findings may be formulated, very simplified, this way: Through the quantitative analysis I discovered that cultural dissimilarity does play a vital role to the pattern of conflict. This supports the overall claim of Huntington, and challenges previous attempts to test the civilizations thesis. Through the qualitative discussion of Bosnia, however, I confirmed how difficult it is to tell how cultural dissimilarity plays a role as a cause of conflict. I identified other important explanatory factors, and how these may inter-act with cultural factors to cause armed conflict. Thus, futures studies of Huntington’s thesis should take these aspects into account.