Anya Palkowski

Progressive Theories and the Future of the Realist Paradigm in International Relations

Theory of World Politics | Professor Erica Chenoweth | Wesleyan University | 2010

Realist thought dates back to Thucydides and Thomas Hobbes, substantiated by Hans Morgenthau in 1948 with Politics in Nations and now a dominant school of thought in the field of international relations.  Designating self-interest and security concerns as the apex of state considerations and declaring that states are the main actors in an absence of universal authority, realism’s base of assumptions has produced a wide field of theoretical work, and like any good paradigm, its legitimacy has seen its share of challengers.

John Vasquez is one such critic, who in a 1997 article questions realism’s persistence as a paradigm of political thought.  He adapts Imre Lakatos’s methods for judging scientific theories and concludes that Kenneth’ Waltz’s neorealism, a theory refined from the core concepts of realism, displays degeneration.  Although there are problems with the accuracy with which Vasquez’s criteria – most importantly the tool of falsification – claim to judge the worth of theories, we will see that accusations against neorealism are ultimately legitimate.  However, the theoretical soundness of the realism on which Waltz’s theory is based, as well as its ability to produce advancing and progressive “theoryshifts,” gives it a valuable edge against Vasquez’s indicators of degeneration.

Degeneration, as defined by Vasquez, is one of two paths of theoretical evolution; in identifying degenerative theories, he gives two criteria.  First, they are characterized by a lack of significant forward movement in either predictive or explanatory power, and rely on using “auxiliary hypotheses” to “explain away discrepant evidence”[i] in ways that are unhelpful in reaching new conclusions within the same theoretical structure.  Rather than devise logically consistent additions to a flawed theory in order to make it more theoretically robust, a theory’s proponents use what Vasquez labels “ad hoc” extra premises to rescue the theory from an empirical anomaly.  This is inconsistent with the ideal of rational consistency that good theories should maintain, as “to add to a theory something that one believes has been omitted requires showing how it can take its place as one element of a coherent and effective theory.”[ii]  Second, the shallow, impromptu defenses that patch up the holes of a theory often give rise to contradicting predictions, with which the theory will consistently test positive regardless of the outcome or behavior against which it is tested.  Vasquez redoubles this point numerous times, saying of neorealist balance of power theory that “permitting the paradigm to be supported by instances of either ‘balancing’ or ‘not balancing’ reduces greatly the probability of finding any discrepant evidence.”[iii]  Numerous hypotheses based on the same configuration of circumstances break the logical flow of the theory; the fact that nothing can defy empirical tests weakens the power of corroborative evidence and increases the chances that the theory can be considered degenerative.

Progression is the desirable alternate model for Vasquez’s progression of theories. Through the process of criticism and revision, a fruitful research program develops until it is discredited with a preponderance of evidence to the contrary or replaced by a better theory; ideally, along the course of this evolution it remains meaningfully salvageable even after strong, seemingly damning criticism has been delivered.  Its amendments address the criticism and anomalous evidence, but when revised actually provide explanatory power additional to the scope of the original theory.  A refinement of this sort is useful in meaningfully fixing a theory’s errors and admitting that that it did not pass important empirical tests, with the end result that the old model is discarded and the new adopted.

Progressive “theoryshift,” as implied by its criteria, also saves the theory in question.  It is the range of new successes in empirical testing that partly determines Vasquez’s first distinction between degeneration and progression; while a degenerative theory uses reparations of narrow scope to superficially guard the theory’s weaknesses against the objections raised, a progressive theory will be revised with new hypotheses that enable the theory to explain more examples and predict more outcomes than its previous iteration could.  This theory, creating new ways of looking at the world alongside new corroborative examples, is considered progressive.  Using these criteria, we will observe that realism meets Vasquez’s criteria for progressive theories, both by distancing itself significantly from the theories built from it and thus avoiding unfalsifiability by contradictions, and by meeting criticisms with meaningful theoretical and conceptual shifts that “tell scholars something about the world other than what was uncovered by… discrepant evidence.”[iv]

Vasquez raises objections foremost to Kenneth Waltz’s neorealism as a vehicle through which he believes the realist paradigm has undeservedly survived.  A number of Waltz’s own steps for testing a theory display, and even encourage, the essence of Vasquez’s critique: “If a test is not passed,” directs Waltz, “ask whether the theory flunks completely, needs repair and restatement, or requires a narrowing of the scope of its explanatory claims.”[v]  Vasquez holds that neorealism, by way of the devastating falsifying evidence of its opponents, has flunked, and yet continues to gather repair and restatement; we will see that though his arguments do not hold for the whole of realism, they ring true for the neorealist research program.

A theory attempts to explain trends in behavior by testing two variables against each other with all others held constant, thus making it important to know “what a theory claims to explain before attempting to test it.”[vi]  A theory falsified on the basis of an empirical anomaly is not necessarily in need of theoretical revisions if variable allowances (which Vasquez might be tempted to call ad-hoc hypothetical additions) can be used to complicate and explain the theory’s application.  This is the mechanism by which Waltz claims his theory of neorealism has been treated by authors like Stephen Walt and Rober Schweller.  Falsification by opponents led Waltz’s colleagues to recast his statements within the same research program, showing us how “forcing more empirical content into a theory”[vii] can dilute its previous logical consistency.

Contrary to Waltz’s defenses, the branches of and “improvements” on neorealism are not dissimilar enough, or stated to be so, to be considered by the study of international relations as entirely new theories that must be evaluated on their own terms.  Unfortunately, his saying that Walt’s theoretical revision is only an explanation based on extra historical information adduced into the theory does not make it so.  Nor does Walt’s semantic revision fall under Waltz’s argument that “changing the concepts of a theory… makes an old theory into a new one that has to be evaluated in its own right.”[viii]  What Waltz believes Walt did was to offer a variable to explain the empirical anomalies presented by balance-of-power theory; he differentiates this from devising a theoretical addition much as he would distinguish a theoretical article like Robert Gilpin’s 1981 “War and Change in International Politics” from an empirically-based explanatory article like Robert Pape’s 2005 “Soft Balancing against the United States.”  However, we see in the language of Walt himself that his work “should be viewed as a refinement of traditional balance of power theory.”[ix]  What Walt really did was create a branch of neorealism that, enveloped into the theory as part of its defining literature, imbued it with two contradictory predictive paths that marred its original logical consistency.  Despite Waltz’s claims that revisionists like Randall Schweller intend to “show that the central theory of neorealism is wrong,”[x] the corrections they made have become part of the same testable body of neorealist literature. Those who wrote after Waltz and semantically revised his theses ultimately reinforced the doctrine of neorealism, rendering it unfairly omni-predictive based on contradictory propositions.  Here we see why unfalsifiability can be damning to a research program: though the ultimate goal of theorizing is to devise a theory that passes all empirical tests, it should also be possible to conceive of a situation in which the theory would fail, thus giving it logical coherency.   As Vasquez notes of the overflexible nature of neorealist predictive power, “opposite behaviors are both logically compatible with the assumptions of anarchy.”[xi]  This underscores the unhelpful nature of neorealist predictive power in predicting the outcomes of state interactions, and thus its limited usefulness to the scientific community.  Combined with the tendency of neorealist revisions to lack empirical power beyond that provided by the original theory, this phenomenon suggests that neorealism is a degenerative research program.

With the model of a degenerative theory in hand, we now move to realism as a paradigm and examining the differences between it and the theories it produces that make it less susceptible to the label of “degenerative.”  It is important to note that despite the criticisms that discredit neorealism, the disproving of a hypotheses spun from the tenets of a certain theorem do not necessarily invalidate the theorem; similarly, realism should not be considered degenerative on the sole basis that neorealism has been shown to be.

Vasquez betrays a deep opposition to the logical inconsistency he believes defines the paradigm of “realism.”  I argue that this inconsistency is illusory, and that as a school of thought, realism has been more fruitful than degenerative.  While neorealism is made degenerative by the sheer number of branching paths of possibility that have been raised to defend its shortcomings, realism itself is a paradigm whose many theoretical offspring represent abundances of “excess empirical content compared to the original theory,”[xii] something Vasquez considers vital in the judgment of theory progressiveness.  For the paradigm of realism to be named degenerative, all the theories that spring from it must pass Vasquez’s tests: they must be continually repaired and semantically restated in order to close a theoretical gap presented by the original realism, they must be cursory and provide the theory with no additional explanatory power, and over time they must develop contradictory hypotheses that preclude the theory from falsification.

Realism is indeed constantly “revised” into a diverse field of theories, all relying on “the use of fundamental concepts to develop theories covering ever more phenomena”[xiii] and all of which rightfully fall under the wide umbrella of realist origin.  However, Vasquez oversimplifies the relationship between the paradigm and its theories by accusing them of “claiming a family resemblance” as they “predict several competing outcomes as providing support for the paradigm.”[xiv]  In doing so he fails to realize that sharing assumptions at realism’s core does not mass a theory automatically into the semantic label of “realism.”

The aspects of classical realism generally receive different treatments from the theories that adopt them.  For example, the way realist theories handle the pursuit of power may not simply embody “the primacy of the international struggle for power and the unitary rational nature of the state,”[xv] and to treat the concept as a static application that permeates all realism-based theories would be misrepresentative.  To say that a balance of power will arise may have different theoretical reasons ranging from the intentions of statesmen to the inevitable result of interactions between states.  To say that states desire power is a fact; to form a hypothesis on the reasons states want power – as an end in itself, as a means to an end – Is a matter handled differently by separate theories.  Knowing this makes it illogical to consider branching theories mere “revisions” of the original realism; the competing outcomes of offensive, defensive and neorealism exist in separate spheres and separate research programs, and we find that they are meaningfully differentiated from the paradigm from which they originate.

Perhaps intending to throw classical realism under the wheels of falsification while defending his own treatment of realist ideas, Waltz cites Morgenthau’s belief that “balances are intended and must be sought by the statesmen who produce them,”[xvi] an idea already fruitfully challenged with arguments that statesmen are seldom motivated to balance power.  Later he expounds his own claim that balance arises “willy-nilly,” whether or not statesmen seek them actively, and that his theory “does not direct the policies of states [but] it does describe their expected consequences.”[xvii]  This follows on his original theorizing that “states will engage in balancing behavior, whether or not balanced power is the end of their acts.”[xviii]  This exposes but one aspect of the theoretical divergence of these two theories, both labeled “realist” by Vasquez but in fact fundamentally dissimilar in the way they handle the balance of power in the international system.

Because of this variety in concepts, the competing predictions of the many realist-based research programs in existence cannot constitute contradictions, as do the many semantic refinements under neorealism.  When John Mearshimer defines the parameters of his model of offensive realism by proposing two branching possibilities, given rise by two separate sets of circumstances, these represent the conditional outcomes of a single theory.  “A threatened great power operating in a bipolar system must balance against its rival because there is no other great power to catch the buck,” he says of the first possibility, and of the second, “buck-passing tends to be widespread in multipolarity when there is no potential hegemon to contend with, and when the threatened states do not share a common border with the aggressor.”[xix] These are two distinct predictions for equally distinct situations, a far cry from saying that a threatened power in bipolarity will either pass the buck or balance, the type of contradiction likely to arise in degenerative theories.  Mearshimer’s theory itself is falsifiable – in the shallowest sense of the word – and can be brought against a preponderance of evidence discrediting a state’s ever-increasing buildup of raw power (as observed by the decisions of countries like Ukraine signing nuclear nonproliferation treaties in return for security guarantees[xx]), but these tests apply only for Mearshimer’s specific interpretation of the notion of “power.”  In short, we evaluate offensive realism as not simply an addendum to realism that decreases its chances of falsification, but as its own intellectually viable theory.

The remaining criterion of degenerativeness is the capacity of realism-based theories for ever-increasing explanatory power.  As Vasquez states himself, “a paradigm can only be appraised indirectly by examining the ability of the theories it generates to satisfy criteria of adequacy.”[xxi] Should the realist tree bear fruitful theories, then it should be called progressive.

Though neorealism itself lacks in progression and explanatory advancement, its first refinement from classical realism embodied the ideal progressive shift from initial theory to second incarnation with new predictive paths and expansively improved explanatory power.  As one example, Waltz’s switch from the classical view of individual motivations to systemic theory accounted for the similar nature of state decisions despite the human biases central to Morgenthau’s realism by explaining that “the situation in which [internal forces] act and interact constrains them from some actions, disposes them towards others, and affects the outcomes of their interactions.”[xxii]

Similarly, hegemonic stability theory built over its realist progenitors to a clear predictive and explanatory advantage.  Hegemonic stability takes Morgenthau’s realist assumptions and Morton Kaplan’s work on systemic change and marries them into a logically consistent theory predicting a number of things out of the scope of either original argument: that conflict and system change occurs when equilibrium cannot be restored, that unipolarity presents the most stable arrangement of power possible in international relations, and that war is most likely when a challenger rises to meet the dominant state, among others.[xxiii]  “As long as unipolarity obtains,” says William Wohlforth, “there is little uncertainty regarding alliance choice or the calculation of power.”[xxiv]  Using this claim, the theory predicts a difficulty of stability that arises from unstable perceptions of power hierarchy, using 1856 Britain and its failure to deter conflict as an example of a weak unipolarity that “could not perform the conflict-dampening role that a unipolar state can play.”[xxv]  Substantiated by modern examples of a lack of balancing against the United States despite its “dominant position in all the key ‘leading sectors’ that are most likely to dominate the world economy into the twenty-first century,”[xxvi] hegemonic stability theory significantly alters the focus and range preceding theories, and produces a different sphere of predicted successes.

Because its individual theories have stood the test of empirical and theoretical progression, the realist paradigm remains a logical base of assumptions on which to construct more specific theories.  Though Vasquez might take this as a clear sign that degeneration has “saved” the paradigm from falsification, the fact remains that each separate theory under realism requires its own research program, and that individual theories neither prove nor disprove realism as a whole.  Either one theory based on realism is explanatorily sound, or another is, or they both fail; one should not evaluate them all together as Vasquez does.  Thus we conclude that realism is neither homogenous nor stagnant enough to merit being called degenerative.

Safe from degeneration, realism has begun to secure its rightful place in the field of international relations.  However one of Vasquez’s most salient points arises when he attempts to address the possible shortcomings of realism itself.  His statement that states act with domestic interests at heart “because they are not the unitary rational actors the realism paradigm thinks they are”[xxvii] embodies his misgivings that some shared aspects of the realist school of thought are themselves untenable, and thus should be de-emphasized as the dominant school of thought in international relations.  These objections may affect realism’s prospects for the future of international studies, making it necessary to elucidate the shortcomings that his methods of falsification may entail.

Using a specific historical anomaly to discredit a theory is an extremely clinical use of falsification; though contradiction-based immunity from falsification is a valuable judge of a theory’s inadequacy, the legitimacy of falsification on an individual basis is limited and should not constitute sole grounds for theoretical revision.  Even a strong theory explains only what it sets out to about the actions of states; it predicts general directions in the progression of international politics by examining “expected behaviors and outcomes… repeatedly found where the conditions contemplated by the theory obtain.”[xxviii]  Theories will naturally experience individual anomalies displayed by tests that contain what Waltz calls “perturbing variables not included in the theory under test.”[xxix] The mutable nature of reality, and a constant possibility of irrational behavior, precludes the exact predictive and explanatory power of social science; this accurately frames the role of the theorist as a kind of political climatologist rather than an all-seeing fortuneteller.

Because theories claim only to test certain variables, the addition of historical information and empirical complications can often provide a more complicated or “imaginative application of a theory.”[xxx]  Any given theory does not claim to explain all situations and all configurations of the world system, but if its concepts are viable, then informational application may suffice to apply them differently to an anomalous situation.  It is only when these applications become fused with the theory itself, as they have for neorealism, that the theory begins to lose explanatory credibility.  Countless theories may be saved from the semantic traps of degeneration by recognizing where a theory needs explanation and where it truly requires revision of core concepts; realism has thusfar avoided legitimate condemnation by producing the dynamic and fruitful theories that now define the realist school of thought.

The purpose of a paradigm is explained clearly by John Vasquez himself.  It “promises scholars that if they view the world in a particular way, they will successfully understand the subject they are studying.”[xxxi]  The realist paradigm is just so: it presents a particular way of viewing the world that has guided its scholars into prolific theorizing.  From the original principles of the realist paradigm, the ideas of anarchy, balance and self-help have given life to new theories – defensive and offensive realism, neorealism, liberal realism – that address larger spheres of phenomenon and predict useful things far outside the scope of the original realist base.  Through the mechanics of progressive theories, we learn that just as realism has endured since the days of Thucydides, the strength and sensibility of its basic tenets will allow it to survive further still.  The realist paradigm, encompassing self-help, the role of nation-states and the importance of power, retains validity as a model of state-level interactions, and continues to explain the behavior of states with value and meaning.

Endnotes