BENJAMIN O.FORDHAM 741 limited theoretical generalizations is useful,especially when doing so can advance a debate in such a closely allied field of study. Thinking Through the Role of Economic and Security Interests One of the striking features of the debate over the relative importance of eco- nomic interests and security concerns is that scholars rarely argue that either set of considerations is entirely irrelevant to policy choice.Even those who strongly emphasize one source of explanation over the other usually acknowledge that many factors contributed to intervention decisions.For example,in spite of the stark differences between revisionist historians of American intervention in World War I and their critics,members of both camps commonly acknowledged the influence of security concerns as well as economic interests in their narra- tives (Fordham 2007,283-84).For most scholars,interpreting the role of eco- nomic and security concerns is a matter of emphasis. The likelihood that both considerations may make a difference complicates efforts to explain intervention in ways that are not always appreciated in the exist- ing research.Most efforts to assess the relative importance of economic and secu- rity concerns have implicitly assumed that these two sources of explanation do not cause one another.This assumption is clearest in the quantitative political sci- ence literature,where the most common approach has been to regress decisions about intervention on indicators of economic and security concerns without pay- ing much attention to the relationships among the independent variables.The assumption is less visible but equally important in qualitative research on inter- vention.Historical researchers usually focus on archival evidence concerning the decision-making process.The assumption that economic and security concerns are independent justifies the inference that,when security concerns dominate policy deliberations,economic interests are relatively unimportant. There are good reasons to suppose that security concerns and economic inter- ests are not causally unrelated,however.Economic interests may help define security concerns,or security-motivated polices could shape economic interests. The fact that the relationship might run in either direction is crucial for assess- ing the relative importance of these two sets of concerns and for discerning their roles in the causal process.Previous research has found evidence that bilateral trade is greater among allies (Gowa and Mansfield 1993,2004;Long 2003;Mans- field and Bronson 1997).Gowa and Mansfield(1993)argue that trade generates security externalities by strengthening the economies,and thus the military potential,of the states involved.They suggest that states seek to manipulate these externalities by encouraging trade with their allies and limiting it with their potential enemies.If trade indeed follows the flag in this way,then part of trade's apparent effect on intervention decisions would really be due to the indi- rect influence of the security commitments on trading patterns. On the other hand,the causal arrow might point in the opposite direction: strong trade relations or other economic interests might make the formation of an alliance more likely.Policy makers should care about the security of economi- cally important markets or sites for investment because the benefits they draw from these economic relationships depend on it.This line of argument follows from liberal claims about the conflict-reducing effects of international commerce on trading partners (e.g.,Oneal and Russett 1997;Polachek 1980).The source of this effect is state leaders'concern about the loss of income to their citizens due to conflict's disruptive effects on trade.The same logic suggests that,to the extent that income from trade depends on the security and stability of a state's trading partners,states should be more likely to make costly commitments to assist those states.Major powers like the United States are especially likely to pro- tect their trading partners because they have the capacity to do so,somethinglimited theoretical generalizations is useful, especially when doing so can advance a debate in such a closely allied field of study. Thinking Through the Role of Economic and Security Interests One of the striking features of the debate over the relative importance of eco￾nomic interests and security concerns is that scholars rarely argue that either set of considerations is entirely irrelevant to policy choice. Even those who strongly emphasize one source of explanation over the other usually acknowledge that many factors contributed to intervention decisions. For example, in spite of the stark differences between revisionist historians of American intervention in World War I and their critics, members of both camps commonly acknowledged the influence of security concerns as well as economic interests in their narra￾tives (Fordham 2007, 283–84). For most scholars, interpreting the role of eco￾nomic and security concerns is a matter of emphasis. The likelihood that both considerations may make a difference complicates efforts to explain intervention in ways that are not always appreciated in the exist￾ing research. Most efforts to assess the relative importance of economic and secu￾rity concerns have implicitly assumed that these two sources of explanation do not cause one another. This assumption is clearest in the quantitative political sci￾ence literature, where the most common approach has been to regress decisions about intervention on indicators of economic and security concerns without pay￾ing much attention to the relationships among the independent variables. The assumption is less visible but equally important in qualitative research on inter￾vention. Historical researchers usually focus on archival evidence concerning the decision-making process. The assumption that economic and security concerns are independent justifies the inference that, when security concerns dominate policy deliberations, economic interests are relatively unimportant. There are good reasons to suppose that security concerns and economic inter￾ests are not causally unrelated, however. Economic interests may help define security concerns, or security-motivated polices could shape economic interests. The fact that the relationship might run in either direction is crucial for assess￾ing the relative importance of these two sets of concerns and for discerning their roles in the causal process. Previous research has found evidence that bilateral trade is greater among allies (Gowa and Mansfield 1993, 2004; Long 2003; Mans- field and Bronson 1997). Gowa and Mansfield (1993) argue that trade generates security externalities by strengthening the economies, and thus the military potential, of the states involved. They suggest that states seek to manipulate these externalities by encouraging trade with their allies and limiting it with their potential enemies. If trade indeed follows the flag in this way, then part of trade’s apparent effect on intervention decisions would really be due to the indi￾rect influence of the security commitments on trading patterns. On the other hand, the causal arrow might point in the opposite direction: strong trade relations or other economic interests might make the formation of an alliance more likely. Policy makers should care about the security of economi￾cally important markets or sites for investment because the benefits they draw from these economic relationships depend on it. This line of argument follows from liberal claims about the conflict-reducing effects of international commerce on trading partners (e.g., Oneal and Russett 1997; Polachek 1980). The source of this effect is state leaders’ concern about the loss of income to their citizens due to conflict’s disruptive effects on trade. The same logic suggests that, to the extent that income from trade depends on the security and stability of a state’s trading partners, states should be more likely to make costly commitments to assist those states. Major powers like the United States are especially likely to pro￾tect their trading partners because they have the capacity to do so, something Benjamin O. Fordham 741