Constructing International Politics
Author(s): Alexander Wendt
Source: International Security , Summer, 1995, Vol. 20, No. 1 (Summer, 1995), pp. 71-81
Published by: The MIT Press
Stable URL: https://www.jstor.org/stable/2539217
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
The MIT Press is collaborating with JSTOR to digitize, preserve and extend access to
International Security
This content downloaded from
134.117.10.200 on Sun, 20 Sep 2020 17:34:28 UTC
All use subject to https://about.jstor.org/terms
Constructing Alexander Wendt
biternational Politics
John J. Mearsheimer's
"The False Promise of International Institutions"1 is welcome particularly in
two respects. First, it is the most systematic attempt to date by a neorealist to
address critical international relations (IR) theory.2 Second, it reminds neoliberals
and critical theorists, normally locked in their own tug-of-war, that they
have a common, non-realist interest in the institutional bases of international
life.3 "False Promise" is likely, therefore, to spur productive discussions on all
sides.
Unfortunately, it will be hard for most critical theorists to take seriously a
discussion of their research program so full of conflations, half-truths, and
misunderstandings. However, to some extent misunderstanding is inevitable
when anthropologists from one culture first explore another. A dialogue between
these two cultures is overdue, and "False Promise" is a good beginning.
Critical IR "theory," however, is not a single theory It is a family of theories
that includes postmodernists (Ashley, Walker), constructivists (Adler,
Kratochwil, Ruggie, and now Katzenstein), neo-Marxists (Cox, Gill), feminists
(Peterson, Sylvester), and others. What unites them is a concern with how
world politics is "socially constructed,"4 which involves two basic claims: that
the fundamental structures of international politics are social rather than
strictly material (a claim that opposes materialism), and that these structures
Alexander Wendt is Associate Professor of Political Science at Yale University.
For their exceptionally detailed and helpful comments I am grateful to Mike Barnett, Mlada
Bukovansky, Bud Duvall, Peter Katzenstein, Mark Laffey, David Lumsdaine, Sylvia Maxfield, Nina
Tannenwald, Jutta Weldes, and the members of the Yale IR Reading Group.
1. John J. Mearsheimer, "The False Promise of International Institutions," International Security, Vol.
19, No. 3 (Winter 1994/95). Subsequent references appear in parentheses in the text.
2. Other efforts include Robert Gilpin, "The Richness of the Tradition of Political Realism,"
International Organization, Vol. 38, No. 2 (Spring 1984), pp. 287-304, and Markus Fischer, "Feudal
Europe, 800-1300," International Organization, Vol. 46, No. 2 (Spring 1992), pp. 427-466.
3. On neoliberalism and critical theory, see Robert Keohane, "International institutions: Two approaches,"
International Studies Quarterly, Vol. 32, No. 4 (December 1988), pp. 379-396, and Wendt,
"Collective Identity Formation and the International State," American Political Science Review, Vol.
88, No. 2 (June 1994), pp. 384-396. Mearsheimer treats collective security as a third form of
institutionalism, but this is unwarranted. Collective security is an approach to international order,
arguable on either neoliberal or critical grounds, not a form of institutional analysis.
4. This makes them all "constructivist" in a broad sense, but as the critical literature has evolved,
this term has become applied to one particular school.
International Security, Vol. 20, No. 1 (Summer 1995), pp. 71-81
( 1995 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.
71
This content downloaded from
134.117.10.200 on Sun, 20 Sep 2020 17:34:28 UTC
All use subject to https://about.jstor.org/terms
International Security 20:1 | 72
shape actors' identities and interests, rather than just their behavior (a claim
that opposes rationalism). However, having these two claims in common no
more makes critical theory a single theory than does the fact that neorealism
and neoliberalism both use game theory makes them a single theory. Some
critical theorists are statists and some are not; some believe in science and some
do not; some are optimists and some pessimists; some stress process and some
structure.5 Thus, in my reply I speak only for myself as a "constructivist,"
hoping that other critical theorists may agree with much of what I say I address
four issues: assumptions, objective knowledge, explaining war and peace, and
policymakers' responsibilities.
Assumptions
I share all five of Mearsheimer's "realist" assumptions (p. 10): that international
politics is anarchic, and that states have offensive capabilities, cannot be
100 percent certain about others' intentions, wish to survive, and are rational.
We even share two more: a commitment to states as units of analysis, and to
the importance of systemic or "third image" theorizing.
The last bears emphasis, for in juxtaposing "structure" to "discourse" and in
emphasizing the role of individuals in "critical theory" (p. 40), Mearsheimer
obscures the fact that constructivists are structuralists. Indeed, one of our main
objections to neorealism is that it is not structural enough: that adopting the
individualistic metaphors of micro-economics restricts the effects of structures
to state behavior, ignoring how they might also constitute state identities and
interests.6 Constructivists think that state interests are in important part con-
5. These are far more than differences of "emphasis," as suggested by Mearsheimer's disclaimer,
note 127.
6. "Constitute" is an important term in critical theory, with a special meaning that is not captured
by related terms like "comprise," "consist of," or "cause." To say that "X [for example, a social
structure] constitutes Y [for example, an agent]," is to say that the properties of those agents are
made possible by, and would not exist in the absence of, the structure by which they are "constituted."
A constitutive relationship establishes a conceptually necessary or logical connection between
X and Y, in contrast to the contingent connection between independently existing entities
that is established by causal relationships.
The identity-behavior distinction is partly captured by Robert Powell's distinction between
preferences over outcomes and preferences over strategies; Robert Powell, "Anarchy in International
Relations Theory," International Organization, Vol. 48, No. 2 (Spring 1994), pp. 313-344. The
main exception to the mainstream neglect of structural effects on state identity is Kenneth Waltz's
argument that anarchy produces "like units"; Kenneth Waltz, Theory of International Politics (Reading,
Mass.: Addison-Wesley, 1979), pp. 74-77. Constructivists think there are more possibilities than
this; see Alexander Wendt, "Anarchy is What States Make of It: The Social Construction of Power
Politics," International Organization, Vol. 46, No. 2 (Spring 1992), pp. 391-425.
This content downloaded from
134.117.10.200 on Sun, 20 Sep 2020 17:34:28 UTC
All use subject to https://about.jstor.org/terms
Constructing International Politics | 73
structed by systemic structures, not exogenous to them; this leads to a sociological
rather than micro-economic structuralism.
Where neorealist and constructivist structuralisms really differ, however, is
in their assumptions about what structure is made of. Neorealists think it is
made only of a distribution of material capabilities, whereas constructivists
think it is also made of social relationships. Social structures have three elements:
shared knowledge, material resources, and practices.7
First, social structures are defined, in part, by shared understandings, expectations,
or knowledge. These constitute the actors in a situation and the nature
of their relationships, whether cooperative or conflictual. A security dilemma, for
example, is a social structure composed of intersubjective understandings in
which states are so distrustful that they make worst-case assumptions about
each others' intentions, and as a result define their interests in self-help terms.
A security community is a different social structure, one composed of shared
knowledge in which states trust one another to resolve disputes without war.8
This dependence of social structure on ideas is the sense in which constructivism
has an idealist (or "idea-ist") view of structure. What makes these ideas
(and thus structure) "social," however, is their intersubjective quality In other
words, sociality (in contrast to "materiality," in the sense of brute physical
capabilities), is about shared knowledge.
Second, social structures include material resources like gold and tanks. In
contrast to neorealists' desocialized view of such capabilities, constructivists
argue that material resources only acquire meaning for human action through
the structure of shared knowledge in which they are embedded.9 For example,
500 British nuclear weapons are less threatening to the United States than 5
North Korean nuclear weapons, because the British are friends of the United
States and the North Koreans are not, and amity or enmity is a function of
shared understandings. As students of world politics, neorealists would probably
not disagree, but as theorists the example poses a big problem, since it
completely eludes their materialist definition of structure. Material capabilities
as such explain nothing; their effects presuppose structures of shared knowledge,
which vary and which are not reducible to capabilities. Constructivism
is therefore compatible with changes in material power affecting social relations
7. What follows could also serve as a rough definition of "discourse."
8. See Karl Deutsch, et al., Political Community and the North Atlantic Area (Princeton: Princeton
University Press, 1957).
9. For a good general discussion of this point, see Douglas Porpora, "Cultural Rules and Material
Relations," Sociological Theory, Vol. 11, No. 2 (July 1993), pp. 212-229.
This content downloaded from
134.117.10.200 on Sun, 20 Sep 2020 17:34:28 UTC
All use subject to https://about.jstor.org/terms
International Security 20:1 | 74
(cf. Mearsheimer, p. 43), as long as those effects can be shown to presuppose
still deeper social relations.
Third, social structures exist, not in actors' heads nor in material capabilities,
but in practices. Social structure exists only in process. The Cold War was a
structure of shared knowledge that governed great power relations for forty
years, but once they stopped acting on this basis, it was "over."
In sum, social structures are real and objective, not "just talk." But this
objectivity depends on shared knowledge, and in that sense social life is "ideas
all the way down" (until you get to biology and natural resources). Thus, to
ask "when do ideas, as opposed to power and interest, matter?" is to ask the
wrong question. Ideas always matter, since power and interest do not have
effects apart from the shared knowledge that constitutes them as such.10 The
real question, as Mearsheimer notes (p. 42), is why does one social structure
exist, like self-help (in which power and self-interest determine behavior),
rather than another, like collective security (in which they do not).
The explanatory as opposed to normative character of this question bears
emphasis. Constructivists have a normative interest in promoting social
change, but they pursue this by trying to explain how seemingly natural social
structures, like self-help or the Cold War, are effects of practice (this is the
"critical" side of critical theory). This makes me wonder about Mearsheimer's
repeated references (I count fourteen) to critical theorists' "goals," "aims," and
"hopes" to make peace and love prevail on Earth. Even if we all had such
hopes (which I doubt), and even if these were ethically wrong (though Mearsheimer
seems to endorse them; p. 40), they are beside the point in evaluating
critical theories of world politics. If critical theories fail, this will be because
they do not explain how the world works, not because of their values. Emphasizing
the latter recalls the old realist tactic of portraying opponents as utopians
more concerned with how the world ought to be than how it is. Critical
theorists have normative commitments, just as neorealists do, but we are also
simply trying to explain the world.
Objectivity
Mearsheimer suggests that critical theorists do not believe that there is an
objective world out there about which we can have knowledge (pp. 41ff). This
is not the case. There are two issues here, ontological and epistemological.
10. On the social content of interests, see Roy D'Andrade and Claudia Strauss, eds., Human Motives
and Cultural Models (Cambridge: Cambridge University Press, 1992).
This content downloaded from
f:ffff:ffff:ffff:ffff:ffff:ffff on Thu, 01 Jan 1976 12:34:56 UTC
All use subject to https://about.jstor.org/terms
Constructing International Politics | 75
The ontological issue is whether social structures have an objective existence,
which I addressed above. Social structures are collective phenomena that confront
individuals as externally existing social facts. The Cold War was just as
real for me as it was for Mearsheimer.
The epistemological issue is whether we can have objective knowledge of
these structures. Here Mearsheimer ignores a key distinction between modern
and postmodern critical theorists. The latter are indeed skeptical about the
possibility of objective knowledge, although in their empirical work even they
attend to evidence and inference. Constructivists, however, are modernists who
fully endorse the scientific project of falsifying theories against evidence. In an
article cited by Mearsheimer, I advocated a scientific-realist approach to social
inquiry, which takes a very pro-science line.1" And despite his claims, there is
now a substantial body of constructivist empirical work that embodies a
wholly conventional epistemology.12
Mearsheimer is right, however, that critical theorists do not think we can
make a clean distinction between subject and object. Then again, almost all
philosophers of science today reject such a naive epistemology All observation
is theory-laden in the sense that what we see is mediated by our existing
theories, and to that extent knowledge is inherently problematic. But this does
not mean that observation, let alone reality, is theory-determined. The world is
still out there constraining our beliefs, and may punish us for incorrect ones.
Montezuma had a theory that the Spanish were gods, but it was wrong, with
disastrous consequences. We do not have unmediated access to the world, but
this does not preclude understanding how it works.
Explaining War and Peace
Mearsheimer frames the debate between realists and critical theorists as one
between a theory of war and a theory of peace. This is a fundamental mistake.
11. See Alexander Wendt, "The Agent-Structure Problem in International Relations Theory," International
Organization, Vol. 41, No. 3 (Summer 1987), pp. 335-370; and, for fuller discussion, Ian
Shapiro and Alexander Wendt, "The Difference that Realism Makes," Politics and Society, Vol. 20,
No. 2 (June 1992), pp. 197-223.
12. See, among others, Michael Barnett, "Institutions, Roles, and Disorder," International Studies
Quarterly, Vol. 37, No. 3 (September 1993), pp. 271-296; David Lumsdaine, Moral Vision in International
Politics (Princeton: Princeton University Press, 1993); Samuel Barkin and Bruce Cronin, "The
State and the Nation," International Organization, Vol. 48, No. 1 (Winter 1994), pp. 107-130; Rey
Koslowski and Friedrich Kratochwilj, "Understanding Change in International Politics," Interna
tional Organization, Vol. 48, No. 2 (Spring 1994), pp. 215-248; Thomas Biersteker and Cynthia Weber,
eds., State Sovereignty as Social Construct (Cambridge: Cambridge University Press, forthcoming);
and Peter Katzenstein, ed., Constructing National Security (working title), forthcoming.
This content downloaded from
134.117.10.200 on Sun, 20 Sep 2020 17:34:28 UTC
All use subject to https://about.jstor.org/terms
International Security 20:1 | 76
Social construction talk is like game theory talk: analytically neutral between
conflict and cooperation."3 Critical theory does not predict peace.14 War no
more disproves critical theory than peace disproves realism. The confusion
stems from conflating description and explanation.
The descriptive issue is the extent to which states engage in practices of
realpolitik (warfare, balancing, relative-gains seeking) versus accepting the rule
of law and institutional constraints on their autonomy States sometimes do
engage in power politics, but this hardly describes all of the past 1300 years,
and even less today, when most states follow most international law most of
the time,15 and when war and security dilemmas are the exception rather tha
the rule, Great Powers no longer tend to conquer small ones, and free trade is
expanding rather than contracting.16 The relative frequency of realpolitik, however,
has nothing to do with "realism." Realism should be seen as an explanation
of realpolitik, not a description of it. Conflating the two makes it impossible
to tell how well the one explains the other, and leads to the tautology that war
makes realism true. Realism does not have a monopoly on the ugly and brutal
side of international life. Even if we agree on a realpolitik description, we can
reject a realist explanation.
The explanatory issue is why states engage in war or peace. Mearsheimer's
portrayal of constructivist "causal logic" on this issue is about 30 percent right.
The logic has two elements, structure and agency. On the one hand, constructivist
theorizing tries to show how the social structure of a system makes
actions possible by constituting actors with certain identities and interests, and
material capabilities with certain meanings. Missing from Mearsheimer's account
is the constructivist emphasis on how agency and interaction produce
and reproduce structures of shared knowledge over time. Since it is not possible
here to discuss the various dynamics through which this process takes
place,17 let me illustrate instead. And since Mearsheimer does not offer a
13. On the social basis of conflict, see Georg Simmel, Conflict and the Web of Group Affiliati
(Glencoe, Ill.: Free Press, 1955). This is also why I prefer to avoid the term "institutionalism," since
it associates sociality with peace and cooperation.
14. Fischer's suggestion that critical theory predicts cooperation in feudal Europe is based on a
failure to understand the full implications of this point; see Fischer, "Feudal Europe, 800-1300."
15. See Louis Henkin, How Nations Behave (New York: Council on Foreign Relations, 1979), p. 47.
16. On the inadequacy of "realist" descriptions of international politics, see Paul Schroeder, "Historical
Reality vs. Neo-realist Theory," International Security, Vol. 19, No. 1 (Summer 1994), pp.
108-148.
17. For a start, see Alexander Wendt, "Collective Identity Formation," and Emanuel Adler, "Cognitive
Evolution," in Emanuel Adler and Beverly Crawford, eds., Progress in Postwar International
Relations (New York: Columbia University Press, 1991), pp. 43-88. The best introduction to processes
of social construction remains Peter Berger and Thomas Luckmann, The Social Construction
of Reality (New York: Anchor Books, 1966).
This content downloaded from
134.117.10.200 on Sun, 20 Sep 2020 17:34:28 UTC
All use subject to https://about.jstor.org/terms
Constructing International Politics | 77
neorealist explanation for inter-state cooperation, conceding that terrain to
institutionalists, let me focus on the "hard case" of why states sometimes get
into security dilemmas and war, that is, why they sometimes engage in realpolitik
behavior.
In "Anarchy is What States Make of It" I argued that such behavior is a
self-fulfilling prophecy,18 and that this is due to both agency and social structure.
Thus, on the agency side, what states do to each other affects the social
structure in which they are embedded, by a logic of reciprocity If they militarize,
others will be threatened and arm themselves, creating security dilemmas
in terms of which they will define egoistic identities and interests. But if they
engage in policies of reassurance, as the Soviets did in the late 1980s, this will
have a different effect on the structure of shared knowledge, moving it toward
a security community. The depth of interdependence is a factor here, as is the
role of revisionist states, whose actions are likely to be especially threatening.
However, on the structural side, the ability of revisionist states to create a war
of all against all depends on the structure of shared knowledge into which they
enter. If past interactions have created a structure in which status quo states
are divided or naive, revisionists will prosper and the system will tend toward
a Hobbesian world in which power and self-interest rule. In contrast, if past
interactions have created a structure in which status quo states trust and
identify with each other, predators are more likely to face collective security
responses like the Gulf War.19 History matters. Security dilemmas are not act
of God: they are effects of practice. This does not mean that once created they
can necessarily be escaped (they are, after all, "dilemmas"), but it puts the
causal locus in the right place.
Contrast this explanation of power politics with the "poverty of neorealism."20
Mearsheimer thinks it significant that in anarchy, states cannot be 10
percent certain that others will not attack. Yet even in domestic society, I cannot
be certain that I will be safe walking to class. There are no guarantees in
life, domestic or international, but the fact that in anarchy war is possible does
not mean "it may at any moment occur.'"21 Indeed, it may be quite unlikely, as
it is in most interactions today Possibility is not probability Anarchy as such
18. A similar argument is developed in John Vasquez, The War Puzzle (Cambridge: Cambridge
University Press, 1993).
19. On the role of collective identity in facilitating collective security, see Wendt, "Collective
Identity Formation."
20. Richard Ashley, "The Poverty of Neorealism," International Organization, Vol. 38, No. 2 (Spring
1984), pp. 225-286.
21. Kenneth Waltz, Man, the State, and War (New York: Columbia University Press, 1959), p. 232.
This content downloaded from
134.117.10.200 on Sun, 20 Sep 2020 17:34:28 UTC
All use subject to https://about.jstor.org/terms
International Security 20:1 | 78
is not a structural cause of anything. What matters is its social structure, which
varies across anarchies. An anarchy of friends differs from one of enemies, one
of self-help from one of collective security, and these are all constituted by
structures of shared knowledge. Mearsheimer does not provide an argument
for why this is wrong; he simply asserts that it is.
Other realist explanations for power politics fare somewhat better. Although
neorealists want to eschew arguments from human nature, even they would
agree that to the extent human-beings-in-groups are prone to fear and competition,
it may predispose them to war.22 However, this factor faces countervailing
dynamics of interdependence and collective identity formation, which
sometimes overcome it. The distribution of material capabilities also matters,
especially if offense is dominant, and military build-ups will of course concern
other states. Again, however, the meaning of power depends on the underlying
structure of shared knowledge. A British build-up will be less threatening to
the United States than a North Korean one, and build-ups are less likely to
occur in a security community than in a security dilemma.
In order to get from anarchy and material forces to power politics and war,
therefore, neorealists have been forced to make additional, ad hoc assumptions
about the social structure of the international system. We see this in Mearsheimer's
interest in "hyper-nationalism," Stephen Walt's emphasis on ideology
in the "balance of threat," Randall Schweller's focus on the status
quo-revisionist distinction and, as I argued in my "Anarchy" piece, in Waltz's
assumption that anarchies are self-help systems.23 Incorporating these assumptions
generates more explanatory power, but how? In these cases the crucial
causal work is done by social, not material, factors. This is the core of a
constructivist view of structure, not a neorealist one.
The problem becomes even more acute when neorealists try to explain the
relative absence of inter-state war in today's world. If anarchy is so determining,
why are there not more Bosnias? Why are weak states not getting killed
off left and right? It stretches credulity to think that the peace between Norway
and Sweden, or the United States and Canada, or Nigeria and Benin are all
due to material balancing. Mearsheimer says cooperation is possible when core
interests are not threatened (p. 25), and that "some states are especially friendly
22. For a good argument to this effect, see Jonathan Mercer, "Anarchy and Identity," International
Organization, Vol. 49, No. 2 (Spring 1995).
23. John J. Mearsheimer, "Back to the Future," International Security, Vol. 15, No. 1 (Summer 1990),
pp. 5-56; Stephen Walt, The Origins of Alliances (Ithaca: Cornell University Press, 1987); Randall
Schweller, "Tripolarity and the Second World War," International Studies Quarterly, Vol. 37, No. 1
(March 1993), pp. 73-103; and Wendt, "Anarchy is What States Make of It."
This content downloaded from
134.117.10.200 on Sun, 20 Sep 2020 17:34:28 UTC
All use subject to https://about.jstor.org/terms
Constructing International Politics | 79
for historical or ideological reasons" (p. 31). But this totally begs the question
of why in an ostensibly "realist" world states do not find their interests
continually threatened by others, and the question of how they might become
friends. Perhaps Mearsheimer would say that most states today are status quo
and sovereign.24 But again this begs the question. What is sovereignty if not
an institution of mutual recognition and non-intervention? And is not being
"status quo" related to the internalization of this institution in state interests?
David Strang has argued that those states recognized as sovereign have better
survival prospects in anarchy than those that are not.25 Far from challenging
this argument, Mearsheimer presupposes it.
Neorealists' growing reliance on social factors to do their explanatory work
suggests that if ever there were a candidate for a degenerating research program
in IR theory, this is it.26 The progressive response (in the Lakatosian sense)
would be to return to realism's materialist roots by showing that the background
understandings that give capabilities meaning are caused by still
deeper material conditions, or that capabilities have intrinsic meaning that
cannot be ignored. To show that the material base determines international
superstructure, in other words, realists should be purging their theory of social
content, not adding it as they are doing.27 And anti-realists, in turn, should be
trying to show how the causal powers of material facts presuppose social
content, not trying to show that institutions explain additional variance beyond
that explained by the distribution of power and interest, as if the latter were a
privileged pre-social baseline.
Responsibility
An important virtue of "False Promise" is that it links neorealism and its rivals
to the ethical responsibilities of foreign policymakers. These responsibilities
24. Mearsheimer and Waltz both assume sovereignty, without acknowledging its institutional
character; see Mearsheimer, "False Promise," p. 11, and Waltz, Theory of International Politics,
pp. 95-96.
25. David Strang, "Anomaly and Commonplace in European Political Expansion," International
Organization, Vol. 45, No. 2 (Spring 1991), pp. 143-162.
26. "Degenerating" problem shifts are adjustments to a theory that are ad hoc, while "progressive"
shifts are those that have a principled basis in its hard core assumptions. See Imre Lakatos,
"Falsification and the Methodology of Scientific Research Programmes," in Lakatos and Alan
Musgrave, eds., Criticism and the Growth of Knowledge (Cambridge: Cambridge University Press,
1970), pp. 91-196.
27. The significance of Dan Deudney's work lies partly in his appreciation of this point; see Dan
Deudney, "Dividing Realism: Structural Realism versus Security Materialism on Nuclear Security
and Proliferation," Security Studies, Vol. 1, Nos. 2 and 3 (1993), pp. 7-37.
This content downloaded from
134.117.10.200 on Sun, 20 Sep 2020 17:34:28 UTC
All use subject to https://about.jstor.org/terms
International Security 20:1 | 80
depend in part on how much it is possible to change the structure of shared
knowledge within anarchy If such change is impossible, then Mearsheimer is
right that it would be irresponsible for those charged with national security to
pursue it. On the other hand, if it is possible, then it would be irresponsible to
pursue policies that perpetuate destructive old orders, especially if we care
about the well-being of future generations.
To say that structures are socially constructed is no guarantee that they can
be changed.28 Sometimes social structures so constrain action that transformative
strategies are impossible. This goes back to the collective nature of social
structures; structural change depends on changing a system of expectations
that may be mutually reinforcing. A key issue in determining policymakers'
responsibilities, therefore, is how much "slack" a social structure contains.
Neorealists think there is little slack in the system, and thus states that deviate
from power politics will get punished or killed by the "logic" of anarchy
Institutionalists think such dangers have been greatly reduced by institutions
such as sovereignty and the democratic peace, and that there is therefore more
possibility for peaceful change.
The example of Gorbachev is instructive in this respect, since the Cold War
was a highly conflictual social structure. I agree with Mearsheimer (p. 46) that
Soviet nuclear forces gave Gorbachev a margin of safety for his policies. Yet
someone else in his place might have found a more aggressive solution to a
decline in power. What is so important about the Gorbachev regime is that it
had the courage to see how the Soviets' own practices sustained the Cold War,
and to undertake a reassessment of Western intentions. This is exactly what a
constructivist would do, but not a neorealist, who would eschew attention to
such social factors as naive and as mere superstructure. Indeed, what is so
striking about neorealism is its total neglect of the explanatory role of state
practice.29 It does not seem to matter what states do: Brezhnev, Gorbachev,
Zhirinovsky, what difference does it make? The logic of anarchy will always
bring us back to square one. This is a disturbing attitude if realpolitik causes
the very conditions to which it is a response; to the extent that realism counsels
realpolitik, therefore, it is part of the problem. Mearsheimer says critical theorists
28. Hence, contra Mearsheimer, there is nothing problematic about the fact that critical theorists
do not make predictions about the future. What happens in the future depends on what actors do
with the structures they have made in the past.
29. This is not true of classical realists; for a sympathetic discussion of the latter from a critical
standpoint, see Richard Ashley, "Political Realism and Human Interests," International Studies
Quarterly, Vol. 25, No. 2 (June 1981), pp. 204-237.
This content downloaded from
134.117.10.200 on Sun, 20 Sep 2020 17:34:28 UTC
All use subject to https://about.jstor.org/terms
Constructing International Politics | 81
are "intolerant" of realists for this reason (p. 42). The ironies of this suggestion
aside, what matters is getting policymakers to accept responsibility for solving
conflicts rather than simply managing or exploiting them. If neorealism can
move us in that direction, then it should, but as I see it, neorealist ethics come
down to "sauve qui peut."
To analyze the social construction of international politics is to analyze how
processes of interaction produce and reproduce the social structures-cooperative
or conflictual-that shape actors' identities and interests and the significance
of their material contexts. It is opposed to two rivals: the materialist
view, of which neorealism is one expression, that material forces per se determine
international life, and the rational choice-theoretic view that interaction
does not change identities and interests. Mearsheimer's essay is an important
opening to the comparative evaluation of these hypotheses. But neorealists will
contribute nothing further to the debate so long as they think that constructivists
are subversive utopians who do not believe in a real world and who
expect peace in our time.
This content downloaded from
134.117.10.200 on Sun, 20 Sep 2020 17:34:28 UTC
All use subject to https://about.jstor.org/terms