Printer Friendly

Progressive research on degenerate alliances.

John Vasquez is right to insist that students of international politics should, justify their theories in terms of Imre Lakatos's (1970) criteria for distinguishing progressive research programs from degenerative ones. Indeed, in our article on European alliance patterns (Christensen and Snyder 1990) we paid close attention to these criteria. We are grateful for the opportunity to remind readers of our strategies for theory-building and of the empirical and theoretical puzzles that motivated us to write the original article.

Before responding to Vasquez's (1997) specific criticisms of our arguments, we must clarify the general relationship between our article and the enterprise of neorealism. We are not hard-core neorealist disciples whose main motivation is to "explain away" anomalies in the neorealist research program, as Vasquez contends. In common with neorealists, we believe that considerations of power often play a central role in the calculations of state leaders and that the logic of anarchy often shapes their strategic choices. We also believe, however, that domestic politics, perceptions, ideology, and other factors may play a major role in shaping choices and outcomes. Sometimes these factors may override the considerations that neorealists emphasize. More commonly, we believe that such factors interact with or mediate the effect of neorealist factors, such as the international distribution of power and the degree of military vulnerability. We adopt this perspective in our article on alliances and in many of our other works (for example, Christensen 1996, Snyder 1991). Whether this makes us neorealists is a labeling exercise in which we decline to take part.

We resist being cast in the role of apologists for neorealism. Still less are we interested in trying to reconcile the heterogeneous arguments of the various scholars whom Vasquez calls neorealist. Nonetheless, we do take the view that Waltz and other scholars working in a similar mode have generated insights into international politics that can be employed as tools in our broader, more eclectic research programs. Consequently, we felt we had a stake in sharpening those tools in order to use them more productively to solve a specific problem that concerned us: namely, the conditions that lead to unconditional "chain-gang" alliances as a response to threat in multipolarity, as opposed to "passing the buck" to third parties to parry the threat. We think that our borrowings from neorealism for that specific purpose yielded valid insights, when employed in the context of our overall research design. Moreover, our argument constitutes a progressive problem shift, as judged by Lakatos's criteria.


Before discussing how our article fulfills those criteria, we will briefly restate our central arguments; except for one footnote (11), Vasquez (1997) misses our main point entirely. We began by noting that alliance patterns before the two world wars were very different, in fact opposite, yet the polarity of the international system, as defined by Waltz, was very similar. For the most part, the European powers chain-ganged as a response to the German threat before 1914 and buck-passed in the face of a similar threat in the 1930s, despite the similar multipolar configurations of power. Therefore, Waltz's ultraparsimonious structural analysis, focused exclusively on polarity, cannot explain the variation.

Of course, Waltz (1979, 1996) explicitly and repeatedly states in his work that, in proposing a structural realist theory, he is not interested in explaining foreign policy. "Waltz is interested in showing that a system of two is more stable than a system of many. He therefore evinces no interest in predicting which pathology of multipolarity will appear in particular circumstances" (Christensen and Snyder 1990, 142). Thus, the problem that interested us could be viewed as largely irrelevant to Waltz's concerns. Notwithstanding these differences in objective, we found Waltz's analysis of the dynamics of multipolar and bipolar worlds a useful starting point. This, however, no more makes us slavish neorealists than emphasizing the economic sources of political power makes one an orthodox Marxist. Although multipolarity was a factor in both the quick escalation of the July crisis in 1914 and the lethargic response to Hitler in the late 1930s, identifying which multipolar malady occurs in which circumstances requires looking for additional causal factors.

To unravel this puzzle, we argued that, if leaders in a multipolar setting believe that offensive doctrines are likely to be effective, then they are more likely to form tight alliances; if they believe defenses are hardy, then they are more likely to attempt to pass onto their current or potential future allies the high costs of initial resistance to aggressors. Under multipolar conditions, when leaders believe wars will be decided early, chain-gangs are likely, and when they expect war to be attritional, buck-passing is likely. Strategic realities are filtered through leaders' perceptions, however, and leaders often misperceive the objective strategic environment, as they did before both world wars. So, in our approach, leaders' perceptions of the strategic environment, not objective conditions, constitute the central variable that determines outcomes.

Except for one footnote, Vasquez ignores this perceptual aspect of our argument, misreading it as a purely structural account intended to save Waltzian neorealism's arguments about states' tendency to balance against threats. This misreading is also the source of Vasquez's claim that our argument constitutes a degenerative problem shift for neorealism, on the grounds that it "hides" or "explains away" what he sees as the failure of interwar states to balance. Far from hiding lacunae in the realist account about multipolar instability, we ourselves uncovered and explained them using perceptual variables that orthodox neorealists typically underrate.

Vasquez also fails to recognize that our dependent variable is not "balancing" but alliance maladies that complicate the balancing process. Failures to balance properly occurred before both wars, not just World War II. We wrote: "In multipolarity, two equal and opposite alliance dilemmas impede efficient balancing" (Christensen and Snyder 1990, 140). Before World War I, states were hyperactive; before World War II, they were lethargic. Thus, we are not testing some generic proposition about "the law of balancing." Rather, we are exploring implications of Waltz's more specific arguments about the mechanisms that make multipolar worlds less stable than bipolar ones. This means that, all things being equal, balancing occurs less smoothly in multipolar worlds than in bipolar ones. Vasquez's claim that Waltz should expect multipolar alliances to form smoothly runs counter to one of the major themes, if not the major theme, of Waltz's work. Thus, Vasquez's arguments about alliance dynamics before the world wars are unhelpful as criticisms either of our work or that of Waltz.


Having clarified the objectives of our article, we can now assess whether it satisfies Lakatos's criteria for a progressive problem shift. One criterion is whether the new theoretical formulation can explain phenomena that the original one cannot. We pass that test, since we explain when chain-ganging occurs and when buck-passing occurs, whereas Waltz only explained that both were more common in multipolarity than in bipolarity. A second criterion is whether the new formulation can explain a large number of new observations, or whether each additional observation demands ad hoc adjustments to the theory. As we argue in the historical section below, we pass that test, too, insofar as our formulation adequately explains the alliance choices of all the major European powers before both world wars. A third criterion is whether the new formulation can be generalized successfully to new domains, and whether any theoretical adjustments needed to extend its range can be accomplished parsimoniously. We pass that test, too, given that a simple extension of our original argument allowed Christensen (1997), in a follow-up article, to explain a variety of nineteenth-century alliance choices. Unlike degenerative Ptolemaic astronomy, we explain a lot with a little.

A final criterion is whether modifications to the original theory introduce assumptions that contradict the theory's core logic. Our article took great pains to show that Waltz's basic assumptions overlap extensively with the core axioms of Robert Jervis's (1978) theory of the security dilemma. This theory has always been understood as having dove-tailed structural and perceptual elements. In anarchy, actors may misperceive structural constraints, but insofar as their perceptual antennae remain focused on figuring out what those structural constraints are, we contend that perceptual arguments are neither contradictory to structural ones nor especially degenerative. Moreover, as Waltz argues, alliance dynamics are more complex and uncertain in multipolarity. Therefore, under these structural conditions, miscalculation and misperception should be more likely.


Aside from his theoretical arguments, Vasquez calls into question some of our historical interpretations. He criticizes our account of alliance behavior before the two world wars on two counts. Before World War I, he argues, Britain did not behave in a way consistent with our theory. It did not chain itself tightly to allies and, according to Vasquez, did less than it did in the early phases of World War II. He also suggests that in the 1930s there was really no balancing against the rise of German power. Instead there was appeasement, something he believes is fundamentally at odds with balancing.

Once again, Vasquez misrepresents our article. We stated clearly that Britain did not chain-gang in 1914 to the same degree that France and Russia did, in part because Britain's off-shore strategic position gave it a defensive advantage that the continental states lacked. Thus, British behavior differed for reasons that our hypothesis readily explains (Christensen and Snyder 1990, 155-6).

Consistent with our argument, Britain responded rather slowly to the security challenges on the continent before both world wars. It is inaccurate to state that it was slower to respond in 1914 than in 1939. British faith in defensive advantage on land was much stronger in 1939 than in 1914, and Britain's commitment to Europe was therefore much smaller and slower to arrive than its commitment to France in summer 1914. London managed to assemble no more than ten divisions for the British Expeditionary Force by the time Germany invaded the Low Countries and France in May 1940 (eight months after the Anglo-French declaration of war). That force was only about half the size sent to Europe by the first half of 1915, and it constituted only 10% of joint Anglo-French forces on the continent (Christensen 1997). Finally, regarding Vasquez's comparison between August 1914 and March 1939, our theory readily explains why the Serbian question sparked quick escalation in 1914, whereas the strategically much more important issue of German control of Czechoslovakia did not do the same in 1938-39.

On the issue of "appeasement versus balancing" Vasquez asks (p. 907): "Were Britain, France, and the USSR passing the buck in the late 1930s, or were they just slow to balance? Or were Britain and France pursuing an entirely different strategy, appeasement?" In fact, Britain, France, and Russia were slow to balance and also were buck-passing. The reasons for both postures were similar: British and Russian leaders in particular underestimated the threats posed by German offensives against France and the Low Countries. Therefore, in both their internal balancing efforts (arming) and their external balancing efforts (alliances) they responded anemically in the short term in order to preserve fighting power for the longer war that they anticipated would follow Germany's initial attempts to subdue either Britain or France.

British, French, and Russian leaders were still concerned about the growth of German power and aggressiveness, however. It is wrong to suggest that they were "appeasing" instead of "balancing." Once again, they did both. It is strange that Vasquez takes an "either/or" view of realist balancing and diplomatic appeasement. One of the first major realist works (published 40 years before Waltz's Theory of International Politics) was E. H. Carr's classic Twenty Years' Crisis. The 1939 edition of this book called on England to appease Hitler in the short term and to build up military power to balance against German expansionism in the longer term (Carr 1939). Appeasement is a diplomatic strategy that can either accompany or preclude balancing strategies, in the same way that "talking tough" and leveling coercive threats can accompany or preclude taking concrete measures to improve one's power position in the world.

Moreover, Britain's misguided and dangerous appeasement policy was largely abandoned after March 1939. By then, British strategists were thinking almost solely in terms of balancing against German power. Britain and France both increased their defense spending in 1939-40. Moreover, they formed an alliance in early 1939. Clearly, some balancing was occurring. Yet, Britain continued to adopt rather anemic responses to the threat posed to France and the Low Countries, even after the formation of a formal Anglo-French alliance. The pace and direction of the British rearmament effort as well as the deployment of available forces all demonstrated Britain's false faith in the hardiness of defensive positions in northern and eastern France (Christensen 1997).

As for the Soviet Union, Stalin's appeasement of and collusion with Hitler were accompanied by a serious arms build-up. As we reported in the original article, Stalin was shocked that France fell so quickly in spring 1940 (Christensen and Snyder 1990, 157). He not only underestimated the effectiveness of German offensives but also grossly overestimated the comprehensive power of Britain and France in comparison to Germany. Given his confidence about French defenses and the power of the Anglo-French condominium, Stalin was much less interested in alliance formation in the months leading up to Guderian's blitz than he likely otherwise would have been (Christensen 1997).

Stalin's misjudgements and the ways that they affected Soviet strategy demonstrate particularly well the real shortcomings in straightforward realist foreign policy analysis. Even when state leaders think of their security in terms consistent with realist tenets - weighing the general balance of power and the ways in which military power can be used to threaten national security-leaders still will often analyze their security environment so badly that they will adopt policies that should appear puzzling to realists.

By underreacting to the threat posed by Hitler, the British, French, and Soviet leaders did not maximize the security of the future alliance. But to suggest that they did not balance at all is simply wrong. The most obvious fact in this regard is that Germany was defeated. If no one balanced, then how did this happen? Soviet, French, and British leaders built up their militaries beginning in 1938, and Britain and France formed an alliance several months before the German invasion of Poland. Britain sent ten divisions to its ally before the German invasion of France; British leaders considered sending more immediately thereafter, until they received the shocking news of the total collapse of allied forces; finally, after the fall of Paris, London guaranteed support for the remaining French forces, thus assuring that Britain would be Hitler's next target. Though Britain should have done more to help France and should have done so earlier, it hardly "distanced" itself from the crisis in France (Christensen 1997).

When designing their security policies before the Cold War, European leaders did think in terms of the balance of power (both the number of great powers and the distribution of power among them) and the relative efficacy of offensive versus defensive strategies. Although they worried about their national security in ways familiar to realists, they often misread their strategic environments on both counts, and thus behaved in ways that should confound realists. This is true not only for the early twentieth century but also for the late nineteenth century (Christensen 1997).

This does not mean that all regions in all times have leaders who are obsessed first and foremost with Realpolitik. Rather than offering a theory of all foreign policy, we were offering a perceptual approach to explain why behaviors vary greatly even in structurally similar worlds in which leaders are generally thinking in terms of power and military doctrine. So, in this limited sense, our article may be seen as truly critical of realism. It uncovers the problems of straightforward realism in explaining the cases in which realism should prove most effective, and it explains the anomalies using perceptual variables. But we never meant to suggest that even our "emendation" of realism would prove decisive in analyzing all international settings. Just as it would be very dangerous to try to ignore power-related factors such as multipolarity, leaders' calculations of the balance of power, and beliefs about the offense-defense balance in analyzing security relations in pre-Cold War Europe, it would be equally dangerous to assume that power politics is always primary in the thinking of all leaders in relations with all outsiders. Both dogmatic realism and dogmatic antirealism will prove crippling in understanding many aspects of international relations. Ideology, institutions, domestic politics, and interdependence may indeed combine in various ways that make balance-of-power considerations irrelevant to various regions or bilateral relationships (Goldgeier and McFaul 1992, Jervis 1991/92). Yet, there have been, there still are, and there likely will be again times and places in which Realpolitik thinking prevails. Our 1990 article and Christensen's 1997 article merely emphasize that, to understand such worlds, we will have to know more about the region in question than the balance of power or the relative efficacy of offensive and defensive doctrines. We will have to know what leaders think about those factors in order to explain and predict the likely patterns of alliance formation and potential causes of conflict escalation.


Carr, E. H. 1939. The Twenty Years' Crisis: 1919-1939. London: Macmillan.

Christensen, Thomas J. 1996. Useful Adversaries: Grand Strategy, Domestic Mobilization, and Sino-American Conflict, 1947-1958. Princeton, NJ: Princeton University Press.

Christensen, Thomas J. 1997. "Perceptions and Alliances in Europe, 1865-1940." International Organization 51(Winter):65-98.

Christensen, Thomas J., and Jack Snyder. 1990. "Chain Gangs and Passed Bucks: Predicting Alliance Patterns in Multipolarity." International Organization 44(Spring):137-68.

Goldgeier, James, and Michael McFaul. 1992. "A Tale of Two Worlds: Core and Periphery in the Post-Cold War Era." International Organization 46(Spring):467-92.

Jervis, Robert. 1978. "Cooperation under the Security Dilemma." World Politics 30(January):167-74.

Jervis, Robert. 1991/92. "The Future of World Politics: Will It Resemble the Past?" International Security 16(Winter):39-73.

Lakatos, Imre. 1970. "Falsification and The Methodology of Scientific Research Programmes." In Criticism and the Growth of Knowledge, ed. Imre Lakatos and Alan Musgrave. Cambridge: Cambridge University Press.

Snyder, Jack. 1991. Myths of Empire. Ithaca, NY: Cornell University Press.

Vasquez, John A. 1997. "The Realist Paradigm and Degenerative versus Progressive Research Programs: An Appraisal of Neotraditional Research on Waltz's Balancing Proposition." American Political Science Review 91(December):899-912.

Waltz, Kenneth. 1979. Theory of International Politics. Reading, MA: Addison-Wesley.

Waltz, Kenneth. 1996. "International Politics Is Not Foreign Policy." Security Studies 6(Autumn):54-7.
COPYRIGHT 1997 Cambridge University Press
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1997 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:response to the article of John Vasquez,in this issue, p. 899
Author:Christensen, Thomas J.; Snyder, Jack
Publication:American Political Science Review
Date:Dec 1, 1997
Previous Article:Evaluating theories.
Next Article:Lakatos and neorealism: a reply to Vasquez.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters