Major arms imports and the onset of civil and ethnic wars in the postcolonial world, 1956–1998: A preliminary reassessment

Abstract

This study examines the effects of states’ major arms imports from major powers on the likelihood of civil and ethnic wars in the postcolonial world. Focusing on civil and ethnic wars, respectively, in postcolonial states from 1956 through 1998, this investigation reveals that major arms imports from major powers have no statistically significant effect on either civil or ethnic war onset, considering political, economic, demographic, and geographic factors. The results also demonstrate that there are some different causal mechanisms between the onset of civil and ethnic wars. Finally, this article discusses policy implications and suggests some directions of the further research.

Acknowledgements

I would like to thank the Correlates of War 2 Project, James D. Fearon, Kristian S. Gleditsch, David L. Laitin, the State Failure Project, the SIPRI Arms Transfers Project, and Tatu Vanhanen for data availability. I also acknowledge Sandra Davis, Volker Krause, John Sagala, J. David Singer, Jeff Tyner, and Rick Witmer as well as anonymous reviewers and the editors for suggestions and encouragement. This article is sincerely dedicated to the memory of Warren ‘Bud’ Day.

Notes

1 For a detailed discussion of arms–conflict nexus, see Sislin and Pearson (2001).

2 Collier et al. (2003) and Sambanis (2002) present a detailed overview of the recent quantitative studies of civil war.

3 Although there are several disagreements over an appropriate definitions of ethnic conflict or war (e.g., Gurr, 2000; Fearon & Laitin, 2003; Sislin & Pearson, 2001), this inquiry follows the discussion by Sambanis (2001, pp. 261–262). Sambanis (2004) presents an extensive discussion of the complexity of various operational definitions of civil war in the empirical literature.

4 For a discussion of the link between ethno-political groups and ethnic conflict, see Gurr (2000).

5 Unlike other studies (e.g., Blanton, 1999a, 1999b; Craft & Smaldone, 2002), this study employs SIPRI's data on major arms transfers for two reasons. SIPRI (1) identifies each individual exporter and importer, and (2) converts arms transfers figure into millions of constant 1990 U.S. dollars since 1950. For a discussion of sources and methods of the SIPRI arms transfers data, see http://projects.sipri.se/armstrade/atmetods.html. Sislin and Pearson (2001: chapter 2) discuss the limitations of data on arms transfers.

6 It may be noteworthy that Henderson and Singer (2000, p. 287) report that the higher state's military spending has a negative effect on the likelihood of civil war in their bivariate model, while the negative impact no longer holds in their multivariate model.

7 According to Gates et al. (2003), there are some benefits using Vanhanen's Democracy Index; yet, the Index contains some biases, meaning that it is “in favor of extremely fragmented party systems in that political systems with many political parties are considered more democratic than system with few parties” (Gates et al., 2003, p. 13).

8 For several other discussions of the geography of internal armed conflicts, see Buhaug & Gates (2002).

9 Le Billon (2001) and Ross (2004) provide an extensive discussion of links between a variety of natural resources (e.g., oil, drugs, and diamonds) and internal armed conflict.

10 I would like to thank the reviewers on this point and Volker Krause for suggestions.

11 The descriptive statistics of all variables can be obtained from the author upon request.

12 For one of the recent UN small arms reports, see A/58/207 on the August 1, 2003. Also, see Small Arms Survey (2003).

13 This claim should be understood very restrictively since it is possible that arms acquisition/import, militarization, and political instability are endogenous to internal conflict. Increased arms acquisition/import, militarization and political instability may not only raise the likelihood of internal conflict, but internal conflict may also increase arms acquisition/import, militarization and political instability. In the future research, these additional considerations should be examined in more rigorous manners.