CHAPTER 1
Realisms
Michael A. Jensen and Colin Elman
CONTENTS
z Introduction 18
z Classical realism 19
z Neorealism: Waltz’s Theory of International Politics 21
z Defensive structural realism 22
z Offensive structural realism 24
z Rise and fall realism 26
z Neoclassical realism 27
z Realisms and the rise of China 28
z Conclusion 31
ABSTRACT
In this chapter, students will learn about the various strands of the realist
research tradition and their different approaches to security studies: classical
realism, neorealism, rise and fall, neoclassical, offensive structural, and
defensive structural realism. Although sharing a pessimistic outlook about the
continuity of inter-group strife, each of these research programmes is rooted
in different assumptions and provides different explanations for the causes and
consequences of armed conﬂict. These differences are illustrated with reference
to what the contemporary strands of realism anticipate will happen in international
politics as China’s power continues to grow.
17
z Introduction
The realist tradition has exercised an enormous inﬂuence over the ﬁeld of security
studies (see Elman and Jensen 2014). Even its harshest critics would acknowledge
that realist theories, with their focus on power, fear and anarchy, have provided
centrally important explanations for armed conﬂict and war. This chapter discusses
several different realist approaches to security studies. Although there are signiﬁcant
differences among variants of realism, they largely share the view that the character
of relations among states has not altered. Where there is change, it tends to occur
in repetitive patterns. State behaviour is driven by leaders’ ﬂawed human nature or
by an anarchic international system. Selﬁsh human appetites for power, or the need
to accumulate the wherewithal to be secure in a self-help world, explain the seemingly
endless succession of wars and conquest. Accordingly, most realists take a pessimistic
and prudential view of international relations (Elman 2001).
In describing and appraising the realist tradition, it is customary to take a metatheoretic
approach that differentiates it from other approaches and that separates
realist theories into distinct subgroups (see Elman and Elman 2003). Accordingly,
accounts of twentieth-century realism typically distinguish political realist, liberal
and other traditions, as well as describe different iterations of realist theory. As noted
in Figure 1.1, this chapter distinguishes between six different variants of realism –
classical realism, neorealism and four ﬂavours of contemporary realism: rise and
fall, neoclassical, offensive structural and defensive structural realism. While this
ordering is not intended to suggest a strict temporal or intellectual succession,
classical realism is usually held to be the ﬁrst of the twentieth-century realist research
programmes.
18
M.A. JENSEN AND C. ELMAN
Defensive
Structural
Realism
Neoclassical
Realism
Classical
Realism
Neorealism
Offensive
Structural
Realism
Rise
and Fall
Realism
FIGURE 1.1 Six realist research programmes
REALISMS
19
Realists frequently argue that they draw on a long intellectual tradition and that
realist themes can be found in important antiquarian works from Greece, Rome,
India and China (e.g. Smith, M. 1986 and Garst 1989 for a contrasting view). They
also suggest that humankind has, in most times and in most places, lived down to
realism’s very low expectations.
z Classical realism
Classical realism is generally dated from 1939 and the publication of Edward Hallett
Carr’s The Twenty Year’s Crisis. Classical realists are usually characterized as
responding to then-dominant liberal approaches to international politics, although
scholars disagree on how widespread liberalism was during the interwar years. Key
works include those by Reinhold Niebuhr (1940), Martin Wight (1946), Hans
Morgenthau (1948), George F. Kennan (1951) and Herbert Butterﬁeld (1951, 1953).
It was, however, Hans Morgenthau’s Politics Among Nations: The Struggle for
Power and Peace that became the undisputed standard bearer for political realism,
going through six editions between 1948 and 1985.
According to classical realism, because the desire for more power is rooted in the
ﬂawed nature of humanity, states are continuously engaged in a struggle to increase
their capabilities. The absence of the international equivalent of a state’s government
is a permissive condition that gives human appetites free rein. In short, classical
realism explains armed conﬂict with reference to human failings. Wars are explained,
for example, by particular aggressive statesmen, or by domestic political systems
that give greedy parochial groups the opportunity to pursue self-serving expansionist
foreign policies. For classical realists international politics can be characterized as
evil: bad things happen because the people making foreign policy are sometimes bad
(Spirtas 1996: 387–400).
Although not employing the formal mathematical modelling found in contemporary
rational choice theory, classical realism nevertheless posits that state behaviour
can be understood as having rational microfoundations. As Morgenthau notes:
we put ourselves in the position of a statesman who must meet a certain
problem of foreign policy under certain circumstances and we ask ourselves
what the rational alternatives are from which a statesman may choose who
must meet this problem under these circumstances (presuming always that he
acts in a rational manner), and which of these rational alternatives this particular
statesman is likely to choose. It is the testing of this rational hypothesis
against the actual facts and their consequences that gives theoretical meaning
to the facts of international politics.
(2005: 5)
State strategies are understood as having been decided rationally, after taking costs
and beneﬁts of different possible courses of action into account. Classical realism’s
focus on the importance of individual leaders raises the intriguing question of
whether US President Donald Trump is a realist (see Box 1.1).
20
BOX 1.1 Is Donald Trump a realist?
On 27 April 2016, Donald J. Trump gave his ﬁrst major foreign policy speech as a
candidate for president of the United States. He declared that as president he
would usher in an era of ‘America First’ foreign policies to improve the US global
position by renegotiating unfavourable trade agreements, withdrawing from
binding international institutions, compelling allies to shoulder more of the
defence burden and taking steps to curb the growth of China. It is likely too soon
to offer more than some tentative observations, but, with that caveat, how might
some of the realist research programmes discussed in this chapter view Trump’s
foreign policy? In the context of different variants of realism (to be discussed),
Trump’s foreign policy appears at ﬁrst glance to align most closely with the views
of offensive structural realism. The United States has long been a regional
hegemon, and its primary interest should be preventing the rise of a peer competitor,
not trying to achieve an unattainable global hegemony. With that in mind,
the US should ﬁrst encourage states that are located near rising threats to deal
with them. The US should only act as a balancer of last resort. Trump’s pledge to
pursue US security interests at all costs, including withdrawing from or revising
long-standing alliances, could be interpreted as an attempt to refocus US foreign
policy on this kind of offshore balancing to preserve US regional hegemony.
However, to date, the administration’s actions have not aligned with its declarations.
By extending US involvement in Iraq and Syria and ramping up tensions with
North Korea, the Trump administration appears to be violating the principle of
sophisticated power maximization that drives offensive structural realism.
If Trump’s ‘America First’ approach leads to a reduction in overseas deployments,
defensive structural realism might view that as likely to reduce states’ perceptions
that the US is a threat. On the other hand, aspects of the administration’s
behaviour – including its step back from international institutions, its bellicose
rhetoric and its economic nationalism – might have the opposite effect.
Neoclassical realism’s focus on the domestic ‘transmission belt’ suggests that,
despite the Trump administration’s proclamations, some aspects of American
social, economic and political experience are likely to block the more ambitious
parts of ‘America First’. High levels of polarization, ﬁscal woes, legislative paralysis
and so on may prevent the United States from arriving at optimal foreign
policies. If the US cannot decide its goals, extract resources to pursue them or
select appropriate strategies, good results are unlikely to follow.
While it is too soon to forecast the fate of ‘America First’, it is prudent to
remember neorealism’s warnings about misguided foreign policies. Neorealism
suggests that systemic pressures will push states to behave in particular ways,
but that they are free to ignore those cues and behave as they wish. The theory
does not suggest that states will always behave wisely, only that they will pay
a steep price if they do not.
M.A. JENSEN AND C. ELMAN
z Neorealism
In 1979, Kenneth Waltz’s Theory of International Politics replaced Morgenthau’s
Politics Among Nations as the standard bearer for realists. Waltz argued that
systems are composed of a structure and their interacting units (1979: 77). Political
structures are best conceptualized as having three elements: an ordering principle
(anarchic or hierarchical), the character of the units (functionally alike or
differentiated) and the distribution of capabilities (Waltz 1979: 88–99). Two elements
of the structure of the international system are said to be constant: the lack of an
overarching authority means that its ordering principle is anarchy, and the principle
of self-help means that all of the units (states) remain functionally alike. Accordingly,
the only structural variable is the distribution of capabilities, with the main distinction
falling between multipolar and bipolar systems (see Chapter 10).
Contrary to classical realism, neorealism excludes the internal make-up of different
states when thinking about their foreign policy preferences. Unlike classical realists,
Waltz’s (1979: 91) theory is not based on leaders’ motivations and state characteristics
as causal variables for international outcomes, except for the minimal
assumption that states seek to survive.
In addition, whereas classical realism suggested that state strategies are selected
rationally, Waltz is agnostic about which of several microfoundations explain state
behaviour. State behaviour can be a product of competition among them, either
because they calculate how to act to their best advantage or because those that do
not exhibit such behaviour are selected out of the system. Alternatively, state
behaviour can be a product of socialization: states can decide to follow norms because
they calculate it is to their advantage, or because the norms become internalized.
Neorealism’s minimal account of preferences and microfoundations means it
makes only indeterminate behavioural predictions about foreign policy (Waltz 1996;
Elman 1996). Waltz nevertheless suggests that systemic processes will consistently
produce convergent international outcomes. Waltz notes that in international politics
the same depressingly familiar things happen over and over. This repetitiveness
endures despite considerable differences in internal domestic political arrangements,
both through time (contrast, for example, seventeenth- and nineteenth-century
England) and space (contrast, for example, the United States and Germany in the
1930s). Waltz’s purpose is to explain why similarly structured international systems
all seem to be characterized by similar outcomes, even though their units (i.e. states)
have different domestic political arrangements and particular parochial histories.
Waltz concludes that it must be something peculiar to, and pervasive in, international
politics that accounts for these commonalities. He therefore excludes as ‘reductionist’
most assumptions about the units that make up the system – suggesting they must,
at a minimum, seek their own survival.
By focusing only minor attention on unit-level variables, Waltz aims to separate
out the persistent effects of the international system. A system can be said to exist:
‘when (a) a set of units or elements is interconnected so that changes in some
elements or their relations produce changes in other parts of the system; and
21
REALISMS
M.A. JENSEN AND C. ELMAN
(b) the entire system exhibits properties and behaviors that are different from
those parts’
(Jervis 1997: 7)
Because systems are generative, the international political system is characterized
by complex nonlinear relationships and unintended consequences. Outcomes are inﬂuenced
by something more than simply the aggregation of individual states’ behaviour,
with a tendency towards unintended and ironic outcomes. Neorealists therefore see
international politics as tragic, rather than as being driven by the aggressive behaviour
of revisionist states (Spirtas 1996: 387–400). The international political outcomes that
Waltz predicts include that multipolar systems will be less stable than bipolar systems;
that interdependence will be lower in bipolarity than multipolarity; and that, regardless
of unit behaviour, hegemony by any single state is unlikely or even impossible.
Waltz’s Theory of International Politics proved to be a remarkably inﬂuential
volume, spinning-off new debates and giving new impetus to existing disagreements.
For example, the book began a debate over whether relative gains concerns impede
cooperation among states (e.g. Baldwin 1993), and added momentum to the extant
question of whether bipolar or multipolar international systems are more war-prone
(e.g. Hopf 1991; Mansﬁeld 1993).
Partly because of its popularity, and partly because of its rejection of competing
theories, Waltz’s Theory of International Politics became a prominent target. As time
went by, detractors chipped away at the book’s dominance (e.g. Keohane 1986).
Neorealism’s decline in the 1990s was ampliﬁed by international events that seemed
to provide strong support for alternative approaches. The Soviet Union’s voluntary
retrenchment and subsequent demise; the continuation of Western European integration
in the absence of American–Soviet competition; the wave of democratization and
economic liberalization throughout the former Soviet Union, Eastern Europe and the
developing world; and the improbability of war between the great powers all made
realism seem outdated (Jervis 2002). Not surprisingly, after the 9/11 attacks on the
United States, political realism enjoyed a resurgence. It is, however, ironic that this was
at least partly owed to transnational terrorist networks motivated by religious extremism,
actors and appetites that both lie well outside neorealism’s traditional domain.
Excluding neorealism, there are at least four contemporary strands of political
realism: rise and fall realism, neoclassical realism, defensive structural realism and
offensive structural realism. All four take the view that international relations are
characterized by an endless and inescapable succession of wars and conquest. The
four variants can be differentiated by the fundamental constitutive and heuristic
assumptions that they share. Brieﬂy, these four realisms differ on the sources of state
preferences – the mix of human desire for power and/or the need to accumulate the
power necessary to be secure in a self-help world – while agreeing that rational
calculation is the microfoundation that translates those preferences into behaviour.
z Defensive structural realism
Defensive structural realism developed out of neorealism but is distinct from it.
Defensive structural realism shares neorealism’s minimal assumptions about state
22
REALISMS
motivations, suggesting that states seek security in an anarchic international system –
the main threat to their well-being comes from other states (Glaser 2003; Walt
2002).
There are three main differences between neorealism and defensive structural
realism. First, whereas neorealism allows for multiple microfoundations to explain
state behaviour, defensive structural realism relies solely on rational choice. Second,
defensive structural realism adds the offence–defence balance as a variable. This is
a composite variable combining a variety of different factors that make conquest
harder or easier (see Lynn-Jones 1995, 2001). Defensive structural realists argue
that prevailing technologies or geographical circumstances often favour defence,
seized resources do not cumulate easily with those already possessed by the metropole,
dominoes do not fall, and power is difﬁcult to project at a distance. Accordingly,
in a world in which conquest is hard it may not take too much balancing to offset
revisionist behaviour. Third, combining rationality and an offence–defence balance
that favours defence, defensive structural realists predict that states should support
the status quo. Expansion is rarely structurally mandated, and balancing is the
appropriate response to threatening concentrations of power (e.g. Walt 1987, 1996).
Rationalism and an offence–defence balance that favours defence means that states
balance, and balances result.
Perhaps the best-known variant of defensive structural realism is Stephen Walt’s
‘balance of threat’ theory (e.g. Walt 1987, 2000). According to Walt, ‘in anarchy, states
form alliances to protect themselves. Their conduct is determined by the threats they
perceive and the power of others is merely one element in their calculations’ (1987:
x). Walt suggests that states estimate threats posed by other states by their relative
power, proximity and intentions and the offence–defence balance (2000: 200–201).
The resulting dyadic balancing explains the absence of hegemony in the system:
Together, these four factors explain why potential hegemons like Napoleonic
France, Wilhelmine Germany, and Nazi Germany eventually faced overwhelming
coalitions: each of these states was a great power lying in close
proximity to others, and each combined large offensive capabilities with
extremely aggressive aims.
(Walt 2000: 201)
Because balancing is pervasive, Walt concludes that revisionist and aggressive
behaviour is self-defeating, and ‘status quo states can take a relatively sanguine view
of threats. . . . In a balancing world, policies that convey restraint and benevolence
are best’ (1987: 27).
One difﬁcult problem for defensive structural realism is that the research programme
is better suited to investigating structurally constrained responses to
revisionism, rather than where that expansionist behaviour comes from. To explain
how armed conﬂict arises in the ﬁrst place, defensive structural realists must either
appeal to domestic-level factors (which are outside of their theory) or argue that
extreme security dilemma dynamics make states behave as if they were revisionists
(see Herz 1950; Chapter 9). Steps taken by states seeking to preserve the status quo
are ambiguous and are often indistinguishable from preparations to launch offensive
23
M.A. JENSEN AND C. ELMAN
strikes. ‘Threatened’ states respond, leading to a spiralling of mutual aggression that
all would have preferred to avoid. This is international relations as tragedy, not evil:
bad things happen because states are placed in difﬁcult situations.
Defensive structural realism has some difﬁculty in relying on security dilemma
dynamics to explain war, however. First, it is not easy to see how, in the absence
of pervasive domestic-level pathologies, revisionist behaviour can be innocently
initiated in a world characterized by status quo states, defence-dominance and
balancing (see Schweller 1996; Kydd 2005). Because increments in capabilities can
be easily countered, defensive structural realism suggests that a state’s attempt to
make itself more secure by increasing its power is ultimately futile.
Second, defensive structural realists contend that states routinely signal their
benign intentions to their peers, thus mitigating the uncertainty that drives the security
dilemma (see Rosato 2014/15; Glaser et al. 2015/16). States can reveal their peaceful
intentions to others by reducing their arsenals, by limiting the size of their armed
forces or by investing in arms that are good for defence and deterrence but provide
little offensive utility. Through these so-called ‘costly signals’, states can avoid the
action–reaction spirals of the security dilemma that produce arms races and war.
Hence, defensive structural realists suggest that the world is made up of states that
seek an ‘appropriate’ amount of power and signal to their peers that they intend no
harm. If states do seek hegemony, it must be due to domestically generated preferences;
seeking superior power is not a rational response to external systemic pressures.
z Offensive structural realism
Offensive structural realists, in contrast, argue that states face an uncertain international
environment in which any state might use its power to harm another
(Mearsheimer 2001, 2014a). Under such circumstances, relative capabilities are of
overriding importance, and security requires acquiring as much power compared to
other states as possible (see also Labs 1997). The stopping power of water means
that the most a state can hope for is to be a regional hegemon, and for there to be
no other regional hegemons elsewhere in the world.
John Mearsheimer’s (2014a: 30–31) theory of offensive structural realism makes
ﬁve assumptions: the international system is anarchic; great powers inherently
possess some offensive military capability, and accordingly can damage each other;
states can never be certain about other states’ intentions; survival is the primary
goal of great powers; and great powers are rational actors. From these assumptions,
Mearsheimer deduces that great powers fear each other, that they can rely only on
themselves for their security and that the best strategy for states to ensure their
survival is maximization of relative power (2014a: 32–6).
Mearsheimer argues that security requires the acquisition of as much power
relative to other states as possible and that increasing capabilities can improve a
state’s security without triggering a countervailing response (2014a: 427, n. 27).
Careful timing by revisionists, buckpassing by potential targets, and information
asymmetries all allow the would-be hegemon to succeed. Power maximization is
not necessarily self-defeating, and hence states can rationally aim for regional
hegemony.
24
REALISMS
Although states will take any increment of power that they can get away with,
Mearsheimer does not predict that states are ‘mindless aggressors so bent on gaining
power that they charge headlong into losing wars or pursue Pyrrhic victories’
(2014a: 37). States are sophisticated relative power maximizers that try ‘to ﬁgure
out when to raise and when to fold’ (2014a: 40). By expanding against weakness
or indecision and pulling back when faced by strength and determination, a
sophisticated power maximizer reaches regional hegemony by using a combination
of brains and brawn.
Mearsheimer (2014a: 140–55) argues that ultimate safety comes only from being
the most powerful state in the system. However, the ‘stopping power of water’ makes
such global hegemony all but impossible, except through attaining an implausible
nuclear superiority. The second best, and much more likely, objective is to achieve
regional hegemony, the dominance of the area in which the great power is located.
Finally, even in the absence of either type of hegemony, states try to maximize both
their wealth and their military capabilities for ﬁghting land battles. In order to gain
resources, states resort to war, blackmail, baiting states into making war on each
other while standing aside, and engaging competitors in long and costly conﬂicts.
When acting to forestall other states’ expansion, a great power can either try to
inveigle a third party into coping with the threat (i.e. buck-pass) or balance against
the threat themselves (2014a: 156–62). While buckpassing is often preferred as the
lower cost strategy, balancing becomes more likely (all things being equal) the more
proximate the menacing state and the greater its relative capabilities.
In addition to moving Mearsheimer’s focus to the regional level, the introduction
of the stopping power of water also leads to different predictions of state behaviour
depending on where it is located. While the theory applies to great powers in
general, Mearsheimer distinguishes between different kinds: continental and island
great powers, and regional hegemons. A continental great power will seek regional
hegemony but, when it is unable to achieve this dominance, such a state will still
maximize its relative power to the extent possible. An insular state, ‘the only great
power on a large body of land that is surrounded on all sides by water’ (2014a:
126), will balance against the rising states rather than try to be a regional hegemon
itself. Accordingly, states such as the United Kingdom act as offshore balancers,
intervening only when a continental power is near to achieving primacy (2014a:
126–8, 261–4). The third kind of great power in Mearsheimer’s theory is a regional
hegemon such as the United States. A regional hegemon is a status quo state that
will seek to defend the current favourable distribution of capabilities (2014a: 42).
Mearsheimer’s theory provides a structural explanation of great-power war,
suggesting that ‘the main causes . . . are located in the architecture of the international
system. What matters most is the number of great powers and how much power
each controls’ (2014a: 337). Great-power wars are least likely in bipolarity, where
the system only contains two great powers, because there are fewer potential conﬂict
dyads; imbalances of power are much less likely; and miscalculations leading to
failures of deterrence are less common. While multipolarity is, in general, more warprone
than bipolarity, some multipolar power conﬁgurations are more dangerous
than others. Great-power wars are most likely when multipolar systems are
unbalanced, that is, when there is a marked difference in capabilities between the
25
M.A. JENSEN AND C. ELMAN
ﬁrst and second states in the system, such that the most powerful possesses the means
to bid for hegemony. Mearsheimer hypothesizes that the three possible system
architectures range from unbalanced multipolarity’s war-proneness to bipolarity’s
peacefulness, with balanced multipolarity falling somewhere in between (2014a:
337–46).
z Rise and fall realism
Rise and fall realism emerged as a powerful alternative to the balance of power
theories that dominated International Relations scholarship during the 1950s. A.F.K.
Organski’s classic 1958 volume, World Politics, challenged the popular belief that
power parity is a virtue in international relations by insisting that throughout history
‘world peace has coincided with periods of unchallenged supremacy of power,
whereas the periods of approximate balance have been the periods of war’ (Organski
1968a: 364). Organski’s claim that hegemony is the foundation for peace, while
balance is often associated with war, has since become a central theme of rise and
fall realism.
In particular, this research programme emphasizes that war between major powers
is least likely when the international system is dominated by a single state and when
there is an absence of rising challengers vying for system leadership. Given its
privileged position, the dominant state is capable of shaping the rules and practices
of the international system in such a way as to satisfy its selﬁsh interests. Stability
is a product of this hegemonic order, as states that are dissatisﬁed with the status
quo lack the capabilities to change it. However, as power becomes more evenly
matched, war over system leadership is likely to occur. When two or more states
approach power parity, the declining hegemon may rationally calculate the need for
preventive war in order to preserve its status as the world’s top power (Gilpin 1981).
In the absence of a preventive attack, a dissatisﬁed rising challenger could initiate
a war in an attempt to capture the top spot and all of the beneﬁts that go along
with it (Organski 1968a, 1968b).
Rise and fall realism depicts the course of human history, or some signiﬁcant
portion of it, as the successive rise and fall of great powers. In order to explain this
historical trend, it pays particular attention to the mechanisms that cause states to
grow at different rates and at different times. In contrast to neorealism, rise and fall
realists contend that differential growth rates are mainly caused by processes internal
to states, including the timing of industrialization (Organski 1958, 1968b), social
formation and type of economic system (Gilpin 1981), bureaucratic politics and
productivity (Doran 1983), and military, economic and technological innovation
(Modelski 1978). Since these dynamics are not at work in all states at the same time
or to the same extent, rather than grow simultaneously, states tend to rise and fall
in relation to one another. Thus, internal developments and the timing of their onset
produce the periods of transition from one system leader to the next, which are
often marked by war.
The rise and fall research programme has spawned a number of theories to explain
differential growth patterns and the onset of major power war, including debates
over (1) whether it is the rising challenger or the declining hegemon that initiates
26
REALISMS
war; (2) what speciﬁc internal processes drive differential growth; and (3) whether
the theory is applicable across time and space, or limited to a period of history or
a particular region of the world (see Elman and Jensen 2014).
While continuing to emphasize the onset of major power war, rise and fall realists
have extended their scope to other important aspects of international relations. For
example, Douglas Lemke (1995, 1996) has applied power transition theory to dyads
other than those involving states directly contesting for system leadership and Dale
Copeland (2014) has shown how relative power shifts deter states from participating
in trade agreements. Moreover, adding alliances to the calculation of differential
growth (Kim 1991, 1992) and security concerns and polarity to theories of great
power competition (Copeland 2001), rise and fall realists have enriched explanations
of power trends and war.
z Neoclassical realism
Neoclassical realism suggests that what states do depends in large part on inﬂuences
located at the domestic level of analysis. Neoclassical realism employs a ‘transmission
belt’ (Rose 1998) approach to foreign policy, which illustrates how systemic pressures
are ﬁltered through variables at the unit level to produce speciﬁc foreign policy
decisions.
While neoclassical realists agree that the distribution of capabilities is a good
starting point for the analysis of state behaviour, they contend that the international
system rarely provides states with clear information on which to base foreign policy
decisions. Systemic threats and opportunities are not easily identiﬁable and there
are numerous policy options available to decision makers for meeting strategic goals.
Given these challenges, variables at the unit level often intervene between systemic
pressures and state behaviour to determine the precise nature and direction of a
state’s foreign economic and military policy.
Neoclassical realists have incorporated a large number of unit-level variables into
their theories of foreign policy decision-making (see Ripsman et al. 2016). Critics
contend that these variables are often chosen in an ad hoc manner to explain
particular historical cases, which limits the theoretical coherence and explanatory
power of the research programme (Walt 2002). In response, neoclassical realists
contend that how a state reacts to systemic imperatives is largely shaped by the
perceptions of its leaders, the culture of its military, bureaucracy and society, the
nature of its domestic political institutions and the ability of its state apparatus to
extract and mobilize domestic resources to achieve foreign policy goals (Ripsman
et al. 2016).
Randall Schweller’s (2006) theory of ‘under-balancing’ provides a good example
of how neoclassical realists incorporate unit-level variables into theories of foreign
policy. Schweller starts from the structural realist position that state behaviour is
primarily driven by the relative distribution of material power in the international
system. He notes, however, that exactly how states choose to react to threatening
accumulations of power depends on the degree to which they embody structural
realism’s unitary actor assumption. When systemic pressures are transmitted through
states that are uniﬁed at the elite and societal levels, decision makers ﬁnd it easy to
27
M.A. JENSEN AND C. ELMAN
recognize threats and carry out appropriate balancing strategies to counter them.
When states are fragmented, however, leaders often ﬁnd it difﬁcult to come to an
agreement on the nature of a threat and how best to deal with it. Furthermore,
divided states often lack the extractive capacity to mobilize the resources from society
that are necessary for restoring a balance of power. According to Schweller, both
France and Britain were fragmented states prior to the Second World War, and this
explains why they both under-reacted to the threat posed by a rising Germany.
z Realisms and the rise of China
As the preceding discussion illustrates, realism is a diverse research tradition. The
tradition’s theoretical diversity is nowhere more apparent than in the different, and
sometimes contending, predictions that realists make about how the growth of China
is likely to shape international relations. While offensive structural realists and rise
and fall realists generally share a pessimistic view of the consequences of China’s
rise for global security, defensive structural realists are more optimistic that the
relations between China, its neighbours and the United States will remain peaceful.
Neoclassical realists are open to a wide range of potential outcomes, ranging from
mutual accommodation to war. This section brieﬂy reviews how the four
contemporary varieties of realism – offensive structural, defensive structural, rise
and fall, and neoclassical – see the rise of Chinese power inﬂuencing the course of
international relations in the future.
Offensive structural realism
If China’s economy continues growing at a fast rate, offensive realism predicts a
future of intense security competition and global insecurity (e.g. Fravel 2010;
Kirshner 2012). Offensive realists argue that, given the required capabilities, states
will pursue regional hegemony as the best means of staying safe in a dangerous
world. Most stress that there is no reason to assume that China will behave any
differently.
In particular, offensive realism predicts that if China’s power continues to grow
it is likely to assert greater control in Asia. China will invest more of its resources
in military capabilities in order to become the predominant power in the region.
While China may not use those capabilities to conquer its regional neighbours, it
will use them to try to dictate how they behave. China will also look to force the
United States from the region through a mixture of military might and its own version
of the Monroe Doctrine (Mearsheimer 2006: 162, 2014a: 370–71).
China’s neighbours and the United States are not likely to sit idly by and watch
as China takes over. Fearing for their security, threatened countries will join the
United States in a balancing coalition to counter China’s rise (Mearsheimer 2006,
2014a: 383–92; 2014b). Intense security competition between China and the
American-led coalition is likely to result, with the United States pursuing aggressive
policies in an attempt to remain the world’s only regional hegemon. While this rivalry
will not guarantee armed conﬂict, it will make war signiﬁcantly more likely.
28
REALISMS
Defensive structural realism
Defensive realists are far more optimistic about China’s rise and the future of international
security. As they see it, the international system is relatively benign.
Aggressive behaviour and power maximization usually trigger self-defeating balancing
coalitions, technology and geography make offensive action difﬁcult, and states
can signal their peaceful intentions. Rational states, therefore, have little reason to
worry about each other based on considerations of power alone. Thus, while China’s
rise will not be welcomed by its neighbours or the United States, they need not
fear it.
As China continues to rise, defensive realists expect it will devote more of its
resources to military technology and capabilities. Although these developments
could spark a security dilemma leading to arms racing and war, defensive realists
stress that measures can be taken by China to signal to other states that these investments
are meant for security purposes alone (see Glaser 2010; Liff and Ikenberry
2014). For example, China could limit its military investments to technologies that
work well for defence but have little or no offensive use. When paired with the
belief that conquest is difﬁcult because of defensive advantage and the pervasiveness
of balancing, defensive realists expect that any security competition that occurs as
a result of China’s growth in power will be countered by a healthy dose of assurance
and rational restraint. This is not to say that defensive realists believe that war
between China and an American-led coalition is impossible. Rather, they would
suggest that if war were to occur it would not be because structure mandated it.
Some domestic-level pathology, such as log-rolling interest groups (Snyder 1991) or
gross misperception (Van Evera 1999), would likely be to blame.
Defensive structural realists have recently argued that the United States should
accommodate China’s rise in order to promote peaceful relations between the two
states. For example, Charles Glaser (2015) suggests that the United States can make
a number of policy concessions to China and maintain its current level of security,
including abandoning its commitment to defend Taiwan against Chinese aggression.
Such actions would elicit Chinese cooperation on a number of maritime and
territorial disputes, which would help sustain peaceful relations in East Asia for the
foreseeable future.
Rise and fall realism
Rise and fall realists are generally pessimistic about a rising China and the prospects
for cooperation and peace in international politics (e.g. Fravel 2010). Increased
security competition and major power war are most likely, they argue, when a rising
challenger and declining hegemon approach power parity. Thus, as China rises
relative to the United States, rise and fall realists expect to see relations between the
two countries become increasingly antagonistic, reaching crisis levels as they near
each other on measures of material power (Kugler and Lemke 2000). If this occurs,
the United States is unlikely to peacefully cede to China its position atop the international
system and the remarkable advantages that go along with it. Instead,
American ofﬁcials could deem that preventive actions, including war, are necessary
29
M.A. JENSEN AND C. ELMAN
to forestall China’s rise and preserve United States hegemony. Likewise, as China’s
power increases, Chinese leaders are likely to demand more inﬂuence in international
politics and a greater share of international spoils. If these expectations are not met,
China could try to dethrone the United States by launching a hegemonic war.
For rise and fall realists, whether China’s rise ultimately results in contained
security competition or a catastrophic war depends on a number of factors. First,
China must not simply rise but rise in relation to the United States. China’s growth,
while remarkable, is driven primarily by domestic processes that could break down
or end before it becomes powerful enough to challenge the United States for
international dominance. Moreover, even with unprecedented growth, China has
not made many gains in overcoming the massive technological and military
advantages that are enjoyed by the United States (e.g. Brooks and Wohlforth
2015/16; Wang et al. 2016).
Second, rise and fall realists stress that, even if Chinese growth continues at current
levels, much of how the two states behave towards each other in the future will
come down to American and Chinese evaluations of the status quo and whether
China can articulate a vision for a new international order that inspires collaboration
(Schweller and Pu 2011). American ofﬁcials are more likely to be open to taking
preventive action to stop the rise of China if they believe that the decline of the
United States is deep and inevitable (Copeland 2001). Similarly, Chinese leaders are
more likely to evaluate the status quo unfavourably if they deem that the policies
and actions of the United States signiﬁcantly limit their ability to achieve beneﬁts
in line with the country’s growing power (Kugler and Lemke 2000).
Finally, rise and fall realism suggests that a number of intervening factors could
moderate security competition between China and the United States and provide
for a peaceful transition. For example, Kugler and Lemke (2000) have argued that
power transitions are likely to be peaceful when both sides possess nuclear weapons.
Similarly, power transitions between liberal democracies produce more favourable
evaluations of the status quo and are less prone to destabilizing security competition
(Lemke and Reed 1996). Thus, Sino-American relations may remain peaceful if
China’s rise occurs alongside a parallel process of domestic political liberalization.
Neoclassical realism
Neoclassical realism does not make determinate predictions about China’s rise and
its implications for the future of Sino-American relations and international security.
Put simply, for neoclassical realists nearly anything is possible. That said, neoclassical
realists agree that the future of Sino-American relations are unlikely to be driven by
objective power trends alone. How China, its neighbours and the United States act
in the future will be determined in part by the domestic-level factors that shape how
states interpret and respond to systemic constraints.
In particular, neoclassical realism expects that the perceptions of key decision
makers and the ability of the state to mobilize resources for its foreign policy will
play a decisive role in determining how China behaves and how others respond.
Thomas Christensen’s (1996) study of Sino-American relations during the early part
of the Cold War showed that Chinese and American leaders frequently prolonged
30
REALISMS
short-term crises between the two countries, often risking war, in order to mobilize
resources from their domestic societies for the purpose of long-term grand strategy.
If this were repeated, neoclassical realism would expect future relations between
China and the United States to be more hostile than they are today as a result of
leaders in both countries trying to cope with weak state institutions and their
inability to extract resources from their citizens.
Neoclassical realism also expects that perceptions of China’s rise in the United
States and Asia will play a large part in determining whether China’s growth is met
with suspicion and fear or with reassurance and collaboration. If American and Asian
ofﬁcials perceive China’s growth to be threatening, they are considerably more likely
to adopt aggressive containment strategies than they are if they view China to have
benign intentions. Given that perceptions are based on a number of complex factors,
including past behaviours, shared expectations and the cognitive biases of individual
leaders, it is impossible to predict how China’s growth will be viewed years from
now by individuals who are tasked with making foreign policy decisions. Scholars
are studying factors that might impinge on these calculations. For example, Johnston
(2016/17) investigates whether Chinese nationalism, a major element in the narrative
that China is newly assertive, is in fact on the rise, and concludes that it is not.
z Conclusion
Realism is a multifaceted and durable tradition of inquiry in security studies, with
an extraordinary facility for adaptation. To conclude, we note how the development
of these six variants of the realist tradition has at least three signiﬁcant ramiﬁcations.
First, while the research programmes have some common characteristics, none
makes wholly overlapping arguments or predictions. Although it is possible to support
some general remarks about the realist tradition (for example, the observations
about realism’s continuity and pessimism), one should otherwise be wary of statements
that begin ‘Realism says . . .’ or ‘Realism predicts . . .’. Different realist theories
say and predict different things. They also have very different implications when
considered as the basis for prescriptive policy. For example, the best offensive
structural realism has to offer the world is an armed and watchful peace anchored
in mutual deterrence, punctuated by wars triggered by structurally driven revisionism
when a state calculates it can gain at another’s expense. The best defensive structural
realism has to offer is a community of status quo states that have successfully managed
to signal their peaceful intentions and/or refrained from obtaining ambiguously
offensive capabilities.
Second, realism’s capacity for change opens the tradition to some criticisms. For
example, realists have been scolded for making self-serving adjustments to their
theories to avoid contradiction by empirical anomalies. John Vasquez (1997) argues
that balance of power theory is degenerative when judged by Imre Lakatos’s (1970)
criteria. Vasquez suggests that balance of power theory is empirically inaccurate but
that succeeding versions of the theory have become progressively looser to allow it
to accommodate disconﬁrming evidence. A related critique was made by Legro and
Moravcsik (1999), who argued that realists often subsume arguments that are more
usually associated with competing liberal or constructivist approaches. The result,
31
they argue, is that realist theories have become less determinate, coherent and
distinctive. These critiques have provoked vigorous responses from realist scholars
(e.g. Feaver et al. 2000; Vasquez and Elman 2003).
Finally, despite its internal divisions and external critics, the realist tradition
continues to be a central contributor to security studies. Now fully recovered from
the excessive optimism of the immediate post-Cold War era, realisms are likely to
provide a substantial share of our explanations and understandings of the causes of
armed conﬂict and war.
Further reading
Robert Gilpin, War and Change in World Politics (Cambridge University Press, 1981).
A leading work in the ‘rise and fall’ realist tradition, which investigates the
consequences of unequal growth rates for great-power politics.
John J. Mearsheimer, The Tragedy of Great Power Politics (W.W. Norton, 2014). The
ﬂagship statement of offensive structural realism suggests that states pursue
sophisticated power maximization to achieve security. This edition revises the
original text by substituting a new closing chapter that addresses the question ‘Can
China Rise Peacefully?’
Hans Morgenthau, Politics Among Nations: The Struggle for Power and Peace, 7th
edition (McGraw-Hill, 2005). The most important classical realist text, and one of
the two most inﬂuential realist books written since the Second World War, this
volume suggests that unchanging human nature is the root cause of conﬂict.
Randall L. Schweller, Unanswered Threats: Political Constraints on the Balance of
Power (Princeton University Press, 2006). A groundbreaking neoclassical realist
volume that illustrates how unit-level factors inﬂuence foreign policy decisions.
Stephen M. Walt, The Origins of Alliances (Cornell University Press, 1987). An important
contribution to the defensive structural realist research programme, which argues
that threat (and not just aggregate capabilities) determines how states respond to
one another.
Kenneth N. Waltz, Theory of International Politics (Addison-Wesley, 1979). The seminal
neorealist work, and the second of the two most important realist volumes, this
book shifted the realist research tradition away from the individual and towards
the international system.
32
M.A. JENSEN AND C. ELMAN
CHAPTER 2
Liberalisms
Cornelia Navari
CONTENTS
z Introduction 34
z Traditional or Kantian liberalism 34
z Douce commerce 36
z The democratic peace thesis 40
z Neo-liberal institutionalism 42
z Conclusion 46
ABSTRACT
In this chapter, students will learn about the debates concerning security
within liberal thought. The ﬁrst section outlines traditional/Kantian liberalism.
The second section introduces liberal economic thought regarding peace and
war and the ideas of ‘douce commerce’. The third section describes the
democratic peace thesis, and reviews the major discussions on the idea that
liberal states do not ﬁght wars with other liberal states. The last section
outlines the major arguments in neo-liberal institutionalism. It concludes by
highlighting the main differences between realist approaches to security and
liberal approaches.
33
z Introduction
True internationalism and world peace will come through individual freedom,
the free market, and the peaceful and voluntary associations of civil society.
(Ebeling 2000)
The liberal tradition in thinking about security dates as far back as the philosopher
Immanuel Kant, who emphasized the importance of ‘republican’ constitutions in
producing peace. His pamphlet Perpetual Peace contains a peace plan and may fairly
be called the ﬁrst liberal tract on the subject. But liberal security has been elaborated
by different schools within a developing tradition of liberal thought. Andrew
Moravcsik (2001) has distinguished between ideational, commercial and republican
liberalism following Michael Doyle (1998), who distinguished international, commercial
and ideological liberalism, each with rather different implications for security
planning; and Zacher and Matthews (1995) have identiﬁed four different tendencies
in liberal security thought. Each reﬂects upon a family of loosely knit concepts,
containing in some cases rather opposed approaches. Kant believed that trade was
likely to engender conﬂict, while later, ‘commercial’ liberals saw in trade a beneﬁcial
and beneﬁcent development. Republican liberals argue that peace is rooted in the
liberalism of the liberal state – the internal approach – while neo-liberal institutionalists
emphasize the role of international institutions, which could ameliorate conﬂict
from without.
This chapter provides an overview of the major positions on security within liberal
thought. The ﬁrst section outlines traditional/Kantian liberalism. The second section
introduces liberal economic thought regarding peace and war and the ideas of
‘douce commerce’. The third section describes the democratic peace thesis and
reviews the major discussions on the idea that liberal states do not ﬁght wars with
other liberal states. The last section outlines the major arguments in neo-liberal
institutionalism. It concludes by highlighting the main differences between realist
approaches to security and liberal approaches.
z Traditional or Kantian liberalism
Immanuel Kant was an Enlightenment philosopher (some would say the greatest
Enlightenment philosopher), often noted for his approach to ethics. (Kant argued
that moral choices were guided by an inner sense of duty – when individuals behaved
according to duty, they were being moral.) But he was not only an ethicist; he philosophized
the ‘good state’ as well as its international relations. According to Kant, the
only justiﬁable form of government was republican government, a condition of
constitutional rule where even monarchs ruled according to the law. Moreover, the
test of good laws was their ‘universalizability’ – the test of universal applicability.
The only laws that deserved the name of ‘law’ were those one could wish everyone
(including oneself) obeyed. Such laws became ‘categorical imperatives’; they were
directly binding, and monarchs as well as ordinary citizens were subject to them.
Kant argued that republican states were ‘peace producers’; that is, they were more
inclined to peaceful behaviour than other sorts of states. He attributed this to habits
34
C. NAVARI
of consultation; a citizenry that had to be consulted before going to war would be
unlikely to endorse war easily. He also attributed it to the legal foundations of the
republican state because he believed a state built on law was less likely to endorse
lawless behaviour in international relations.
But being republican was not sufﬁcient to ensure world peace. According to Kant
– and it was the critical argument of Perpetual Peace – the situation of international
relations, its lawless condition, unstable power balances and especially the everpresent
possibility of war endangered the republican state and made it difﬁcult for
liberal political orders to maintain their republican or liberal condition. Hence, he
argued, it was the duty of the republican state to strive towards law regulated international
relations; they could not merely be liberal in themselves.
A critical part of Kant’s argument, which initiated the debate between liberals
and ‘realists’, was his critique of the concept of the ‘balance of power’: he refuted
the argument, becoming prevalent in his day, that the balance of power was a peacekeeper.
The idea of conscious balancing was fallacious, he argued, since ‘It is the
desire of every state, or of its ruler, to arrive at a condition of perpetual peace
by conquering the whole world, if that were possible’ (Kant 1991b [1793]), a view
shared by some leading realists (e.g. Mearsheimer 2001). As to the automatic
operations of such a balance, he held Rousseau’s view that such tendencies did indeed
exist. Rousseau (1917 [1754]) argued that states were naturally pushed into watching
one another and adjusting their power accordingly, usually through alliances. However,
this practice resulted merely in ‘ceaseless agitation’ and not in peace.
Kant’s peace programme consisted of two parts (Kant 1991a [1795]). There were
the ‘preliminary articles’ – the initial conditions that had to be established before
even republican states could make much contribution to a more peaceful international
environment. These included the abolition of standing armies, non-interference in
the affairs of other states, the outlawing of espionage, incitement to treason and
assassination as instruments of diplomacy, and an end to imperial ventures. These
had to be abolished by a majority of states, non-liberal as well as liberal, to end the
condition Hobbes had described as ‘the war of all against all’. There were then
the three deﬁnitive articles, which provided the actual foundations for peace:
1 The civil constitution of every state should be republican.
2 The law of nations shall be founded on a federation of free states.
3 The law of world citizenship shall be limited to conditions of universal hospitality.
Spreading republican constitutions meant, in effect, generalizing the striving for
peace, since, according to Kant, striving for peace was part of the natural orientation
of the republican state. The ‘federation of free states’ would provide, in
effect, a type of collective security system; and the provision of ‘universal hospitality’
would, in Michael Howard’s formulation, ‘gradually create a sense of cosmopolitan
community’ (2000: 31). Kant distinguished between the end of war and the
establishment of positive peace, and his plan made peace ‘more than a merely pious
aspiration’. Accordingly, he can properly be regarded as the ‘inventor of peace’
(Howard 2000: 31).
LIBERALISMS
35
C. NAVARI
During the nineteenth century, liberals tended to emphasize only Kant’s views
that liberalism inclined to peace. Through most of the nineteenth century, the liberal
approach to peace consisted of critiques of the ancient regime, and promised that
peace would automatically follow the overthrow of autocracy and the establishment
of constitutional regimes. According to Raymond Aron (1978), nineteenth-century
liberals had no peace plan. With the outbreak of the First World War, however, the
emphasis changed. Then, the dangers that Kant had foreseen for liberalism in a
dangerous international environment were rediscovered and liberal thinkers turned
from internal reform towards emphasizing arbitration, the development of international
law and an international court, to protect liberalism from without. When
the League of Nations failed, moreover, some would go so far as to recommend
either the abolition of, or severe restrictions upon, state sovereignty.
z Douce commerce
According to Moravcsik, ‘commercial liberalism’ focuses on ‘incentives created by
opportunities for trans-border economic transactions’ (2001: 14). This contemporary
formulation attempts to make speciﬁc the causal mechanisms behind the inclination
of economically liberal states to prefer peace to conﬂict. According to Moravcsik,
‘trade is generally a less costly means of accumulating wealth than war, sanctions
or other coercive means’ (2001: 50). But it is not the only theory – other commercial
liberals stress the structure of a liberal economy, not merely the preferences of
individual economic actors.
The origins of modern commercial liberalism lie in the liberal critique of mercantilism,
the aggressive economic policies recommended to, and to a degree practised
by, the autocrats of the ancient regime. Mercantilist doctrine advised doing all to
increase the amount of bullion held by a country, in an environment where bullion
was believed to be a ﬁxed quantum. The economic philosophes (called physiocrats)
such as François Quesnay and Victor de Mirabeau identiﬁed a structural proclivity
in mercantilism towards trade wars and territorial conquests. If your own nation
was to be wealthy, it could only be so by making others poorer. Tariff walls were
needed to protect the prosperity of domestic producers from the ‘attacks’ of foreign
competitors. Subsidies were required for export producers so that they could ‘seize’
the wealth of others in foreign markets. Resources in foreign lands had to be militarily
‘captured’ to keep them out of the hands of commercial rivals. The effect of
generalizing mercantilism was made explicit by Voltaire in 1764: ‘It is clear that a
country cannot gain unless another loses and it cannot prevail without making others
miserable.’
The physiocrats argued that giving up mercantilism and allowing freedom to trade
would civilize the citizens of a nation, facilitating a peaceful coexistence among fellow
citizens and guaranteeing the rule of law. Commerce and international trade would
exercise a restraining force on tyrannical and arbitrary leadership. Finally, international
trade would incline nations to peace. Montesquieu ﬁrst presented all three
arguments in De L’Esprit des Lois (Montesquieu 1989 [1748]: 338).
The explicit association between non-mercantilist, open trading orders and peace
was, however, not French but a British development. It ﬁrst appeared in Adam
36
LIBERALISMS
Smith’s Wealth of Nations in 1776, where he argued that ‘the hidden hand’, besides
increasing wealth, also promoted a lessening of economic hostilities. But even earlier,
Smith’s Scottish colleague and friend David Hume in his 1736 Treatise of Human
Nature had proposed that a division of labour and trade beneﬁted all participants,
and David Ricardo (1817) formulated the theory of comparative advantage.
According to Ricardo, wealth accrued in the degree to which states concentrated
production in areas where they had ‘comparative advantage’ and traded for other
products. Ricardo’s theory underpinned the notion of a benevolent division of
labour as well as the idea that trade was non-hostile competition.
The nineteenth-century ‘commercial liberals’ developed these ideas into doctrine.
Liberal trade doctrine, or the theory of douce commerce, held that trade among
states, like trade among individuals, was mutually beneﬁcial. All men would gain
through participation in a global division of labour – a way of life in which they
offered to each other the various products in the production of which they specialized.
Market competition was not conﬂict, they argued, but rather peaceful cooperation:
each producer helped to improve the quality of life for all through the production
and sale of superior and less expensive products than the ones offered by his market
rivals. The market was civil society and peace; economic policy in the hands of
governments was conﬂict and war (Silberner 1972).
Commercial liberalism also took on a sociological aspect. James Mill (1828: 5)
described the British Empire as ‘a vast system of outdoor relief for the upper classes’.
Joseph Schumpeter in the twentieth century agreed that conquest and imperialism
had economically favoured the old aristocratic elites, and argued that the social
changes accompanying capitalism made modern states inherently peaceful, since they
led to the decline of the aristocratic class.
The only formal non-liberal nineteenth-century riposte to the commercial liberals
(that is, the only argument one could credit with some respect) was the mid-nineteenth-century
idea of protecting infant industries. The German nationalist Jahn
argued that because of the time lag between developed and developing countries,
there was an argument for initial protection for ‘infant’ industry, but even then only
until it could compete in an open market. In the twentieth century, under the pressure
of the Great Depression, liberals would also argue that there was some justiﬁcation
for protecting economies from storms in the world economy, but again temporary
measures only. (The liberal tendency came to be to improve international regimes
so that storms could be avoided or ridden out without closures.) There also developed
a more reﬁned critique of the argument that everyone beneﬁted through trade. This
made it clear that the wealth accruing through opening economic exchange did not
automatically beneﬁt everyone in society; this depended on social policies that, among
other things, deliberately fostered the skills that would allow individuals to participate
in market economies.
During the twentieth century, the initial successes of Nazism, government-directed
labour programmes and the much-vaunted ‘Soviet model’ led the commercial liberals
to focus on government involvement in the economy and on protectionist ideologies.
Indeed, twentieth-century commercial liberals spoke less of economy than of
ideology, attacking ideas of economic closure and planning that derived from
‘scientiﬁc socialism’ and economic nationalism. The most famous of these is Friedrich
37
C. NAVARI
Hayek’s Road to Serfdom, but it had many echoes, especially in Central Europe.
During the 1930s, the German economist Wilhelm Röpke declared that the ‘genuinely
liberal principle’ required ‘the widest possible separation of the two spheres of
government and economy’. He recommended the largest possible ‘depoliticization’
of the economic sphere. In 1936, the Swiss economist and political scientist William
Rappard (the Rappard Chateau that houses the World Trade Organization in
Geneva is named after him) in a lecture entitled ‘The Common Menace of Economic
and Military Armaments’ identiﬁed ‘economic armaments’ with all of the legislative
and administrative devices governments use to politically inﬂuence imports and
exports as well as the allocation of commodities. Rappard argued that a new world
order of peace and prosperity would only come about when governments exited
from control of the economy. In similar fashion, in 1950, the free-market economist
Michael A. Heilperin presented ‘An Economist’s Views on International Organization’.
He told his audience:
It is an elementary, but often forgotten, knowledge that policies of national
governments have always been the principal obstacle to economic relations
between people living in various countries, and that whenever these relations
were free from government restrictions, equilibrium and balanced growth
would follow by virtue of the spontaneous and anonymous mechanism of the
market.
(Heilperin 1962: 181)
Attacks on ideology came to include the idea that peace could come through
abolitions of sovereignty, a favourite liberal idea of the late 1930s and 1940s. But
not one that was shared by all free-market liberals. According to the Austrian
economist Ludwig von Mises:
[Classical] liberalism did not and does not build its hopes upon abolition of
the sovereignty of the various national governments, a venture which would
result in endless wars. It aims at a general recognition of the idea of economic
freedom. If all peoples become liberal and conceive that economic freedom
best serves their own interests, national sovereignty will no longer engender
conﬂict and war. What is needed to make peace durable is neither international
treaties and covenants nor international tribunals and organizations
like the defunct League of Nations or its successor, the United Nations. If the
principle of the market economy is universally accepted, such makeshifts are
unnecessary; if it is not accepted, they are futile. Durable peace can only be
the outgrowth of a change in ideologies.
(1949: 686)
The notion that economic openness produces a more peaceful international posture
has become the subject of close empirical examination. In 1997, Oneal and Russett
(1997) declared that the ‘the classical liberals were right’ in their study of the record
in the post-war period. Similarly, Mansﬁeld and Pollins (2001) summarized a large
body of empirical work that, for the most part, supports the thesis. There are various
38
LIBERALISMS
exceptions and qualiﬁcations that are seen to limit the circumstances under which
economic interdependence results in conﬂict reduction. Stephen van Evera (1994)
has argued that the more diversiﬁed and complex the existing transnational commercial
ties and productions structures the less cost-effective coercion is likely to
be. By extension, the less diverse the production structure of a country and the more
it is characterized by monopolies, the more fragile will be the inclination to peace.
Moving beyond economic interdependence to the issue of economic freedom
within states, Erik Gartzke (2005) has found empirical evidence that economic
freedom (as measured by the Fraser Institute Economic Freedom Index) is about 50
times more effective than democracy in reducing violent conﬂict. Gartzke’s
conclusions are critical for the direction of liberal reforms, since they imply that it
is less important what sort of political regime a country has than its degree of
economic freedom.
The policy prescriptions enjoined by the commercial liberals – often called ‘economic
disarmament’ – focus on limiting the power of governments to impose trade
restraints, primarily through international regulation. Foreign exchanges were to be
open; tariffs were to be reduced to the minimum and quotas and other quantitative
restrictions positively forbidden. Governments were to pledge themselves to open
tariff borders, to abolish quotas and to allow currencies to move in line with market
forces. These policy prescriptions were immensely inﬂuential in the architecture of
the newly established international economic organizations, set up at the end of the
Second World War.
Recently, the literature on globalization has suggested that globalization, in its
aspect as unfettered free trade on a global scale, is a peace producer. Graham Allison
has opined that ‘global networks, particularly in economics, create demands by
powerful players for predictability in interactions and thus for rules of the game
that become, in effect, elements of international law’ (Allison 2000: 83). Thomas
Friedman’s The Lexus and the Olive Tree declares that
When a country reaches the level of economic development, when it has
a middleclass big enough to support a McDonald’s network, it becomes a
McDonald’s country. And people in McDonald’s countries don’t like to ﬁght
wars anymore, they prefer to wait in line for burgers.
(Friedman 2000: 14)
But it is not obvious that globalization has ﬁrmly entrenched economic liberalism.
Commenting on America’s foreign economic policy of the 1980s, Professor Richard
Ebeling, of the Future of Freedom Foundation, observed the emergence of traditional
mercantilist methods:
If some of America’s Asian trading partners ‘capture’ a large share of the
American consumer market, the government responds with a tariff-wall
‘defense.’ If American agriculture cannot earn the proﬁts it considers ‘fair,’
the U.S. government takes the ‘offensive’ by ‘attacking’ other lands through
export price-subsidies. If other nations will not comply with the wishes of the
Washington social engineers in some international dispute, the American
39
Box 2.1 Donald Trump’s neo-mercantilism
Donald Trump, elected to the US presidency in 2016, demonstrated clear neomercantilism
in his economic plan, announced during his campaign (Trump
2016), in which he promised to raise tariff walls and create domestic employment
by limiting imports. He described trade as a zero-sum game in which countries
lose by paying for imports.
The reputed peace effects of globalization are also countered in the literature by
some reputed war effects. These include increased vulnerability to threats from the
failure of the complex systems globalization relies upon, as well as from non-state
actors whose access to weapons and potential for disruption increase in a globalized
world. Advances in technology may also have made states more vulnerable to
coercive threats than would have been possible earlier (on some of the implications
of globalization for security, see Navari 2006). Liberal unease with globalization is
well-represented in a recent collection of essays (Held 2007), where Michael Doyle,
among other noted liberals, outlines the problem of democratic accountability in a
globalized political system.
z The democratic peace thesis
The ‘democratic peace’ thesis is the argument that liberal states do not ﬁght wars
against other liberal states. It was ﬁrst enunciated in a keynote article by Michael Doyle
in the journal Philosophy and Public Affairs (Doyle 1983). Doyle argued that there
was a difference in liberal practice towards other liberal societies and liberal practice
towards non-liberal societies. Among liberal societies, liberalism had produced a
cooperative foundation such that ‘constitutionally liberal states have yet to engage in
war with one another’ (Doyle 1983: 214). Doyle based his ﬁndings on David Singer’s
Correlates of War Project (COW) at Michigan University and the COW’s list of wars
since 1816 (see Small and Singer 1982). Using the list, Doyle observed that almost
no liberal states had fought wars against other liberal states, and that in the two
instances in which it seemed that liberal states had fought against other arguably liberal
states, liberalism had only recently been established. Doyle sourced the tendency in
Kant’s ‘three preconditions’; namely republican constitutions, collective security
arrangements and civic hospitality, in which Doyle included free trade.
The speciﬁc causes of the ‘liberal peace’ have become the subject of robust
research and discussion. The two major contending theories focus on liberal
institutions and liberal ideology, respectively. Liberal institutions include the broad
40
government inﬂuences and persuades them with government-to-government
ﬁnancial loans, grants and subsidized credits – all at American taxpayers’
expense, of course.
(Ebeling 1991)
C. NAVARI
41
franchise of liberal states and the need to ensure broad popular support; the division
of powers in democratic states, which produces checks and balances; and the
electoral cycle, which makes liberal leadership cautious and prone to avoid risk
(Russett 1996). But liberal institutions would tend to inhibit all wars, whereas
liberal states have fought robust wars against non-liberal states. The other contender,
which can explain the difference, is liberal ideology or ‘culture’. According to the
liberal culture argument, liberal states tend to trust other liberal states and to expect
to resolve conﬂict through discussion and compromise. But, equally, they distrust
non-liberal states. The major argument for liberal culture has been put forward by
John M. Owen, who suggests that, ‘Ideologically, liberals trust those states they
consider fellow liberal democracies and see no reason to ﬁght them. They view those
states they consider illiberal with suspicion, and sometimes believe that the national
interest requires war with them’ (1996: 153).
Since Doyle ﬁrst produced his ﬁndings, the theory has developed two variants:
one maintains that democracies are more peaceful than non-democracies, that is,
that they are more paciﬁc generally (see Russett 1993). This is sometimes referred
to as the monadic variant. The other maintains that liberal states are not necessarily
more peaceful than non-liberal states but that they eschew the use of force in
relation to other democracies; that is, the use of force depends on the recipient’s
form of government. In the later variant, sometimes called the dyadic variant, a few
have argued that democracies may be even more robust in the use of force than
non-democracies, owing partially to the ideological nature of democratic wars and
partially to the fact that liberal democracies are generally strong states with a large
wealth base (see Barkawi and Laffey 2001).
From the security point of view, the recommendations of democratic peace theory
are clear – in the ﬁnal analysis, security depends on encouraging liberal institutions
and a security policy must have as its long-term goal the spread of liberalism. In
the short term, it must protect liberalism, including liberal tendencies in non-liberal
states. Doyle himself argues that where liberalism has been deﬁcient ‘is in preserving
its [liberalism’s] basic preconditions under changing international circumstances’
(1983: 229). The route to peace is to encourage democratic systems, the universal
respect for human rights and the development of civil society.
But such a conclusion depends on an untroubled and robust correlation between
the democratic nature of a state and a peaceful inclination, at least towards other
liberal states, and it is not entirely clear that such a direct correlation exists. Chris
Brown (1992) has pointed out that liberal states have, during the period that many
states became liberal, faced determined enmity from non-liberal states. The fact that
liberal states have faced enemies of liberalism distorts the historical record; we do
not know how they may have diverged in the absence of such an enmity. It may
also be that in a world of diverse states in situations of conﬂict; that is, in an
anarchical society, liberal states make more reliable allies – they do not ﬁght one
another because they ally with one another. (This is called the liberal alliance thesis
and is compatible with realist approaches.) There is also the not insigniﬁcant fact
that the majority of liberal states are locked into economic integration, via the European
Union (a fact that may support the douce commerce variant). Finally, the
democratic transition phenomenon may be a statistical aberration. David Spiro
LIBERALISMS
C. NAVARI
(1996), for instance, has argued that historically there have not been many liberal
states, and that most states do not ﬁght wars against one another anyway. The fact
that liberal states have not fought wars against one another may not be statistically
signiﬁcant.
As to whether liberal states are more intrinsically peaceful than other states, this
is perhaps even more contentious. Kant, for his part, seemed to support the monadic
theory; he claimed not only that republics would be at peace with each other but
that republican government is more paciﬁc than other forms of government.
But the empirical work is indeterminate since it has so far concentrated on traditional
state-to-state wars and has ignored interventions – intervention could also be considered
an intrinsically hostile act involving the use of force outside of one’s borders.
Recent ‘liberal’ attempts to bring non-liberal states to liberal democracy (in Iraq,
for example) have raised fears that, far from being a recipe for peace, liberal foreign
policy may have its own tendencies towards war. This ‘dark side of liberalism’ has
occupied much of the resent research, which has turned to the conditions which
may lead liberal states to ﬁght wars (see Geis et al. 2006).
Despite some hesitations from the academy, the theory that democracies do not
ﬁght wars against other democracies has been immensely inﬂuential in public policy.
For example, it underpinned President Clinton’s A National Security Strategy for a
New Century (US 1998); it was also extensively used to support the neo-conservative
case for war in Iraq and has guided post-war reconstruction in insisting on a broadly
inclusive post-war government in Iraq and an early move to self-government with
elections. The democratic transition thesis has also come to dominate the peacebuilding
programme of the UN. Michael Barnett (2006) has been critical of what
he calls the ‘civil society’ model for post-war reconstruction, since it places emphasis
on mobilizing social forces in often unstable and divided societies, when more
attention should be placed on building state capacity and strengthening governmental
powers (see also Paris 1997).
The association between war, democracy and rights, prevalent in the immediate
aftermath of the Second World War, has also been revived. Founded upon the
principles of territorial integrity and state sovereignty, the UN has recently begun
to shift towards an emphasis on the rights of human beings as being at least as
important as the rights of states in the international realm. In a discussion on the
relevance of the Security Council, UN Secretary-General Koﬁ Annan clearly indicated
that ‘the last right of states cannot and must not be the right to enslave, persecute
or torture their own citizens’. In fact, rather than rally around sovereignty as its
sole governing idea, the Security Council should ‘unite behind the principle that
massive and systematic violations of human rights conducted against an entire
people cannot be allowed to stand’ (Annan 1999: 514).
z Neo-liberal institutionalism
Neo-liberal institutionalism concentrates on the role of international institutions in
mitigating conﬂict. Robert Keohane (1984) and Robert Axelrod (1984), who have
played a central role in deﬁning this ﬁeld, point to the ability of institutions such
as the UN to redeﬁne state roles and act as arbitrators in state disputes. Although
42
LIBERALISMS
institutions cannot absolve anarchy, they can change the character of the international
environment by inﬂuencing state preferences and state behaviour.
International institutions do this by a variety of methods that either create strong
incentives for cooperation such as favourable trade status, or powerful disincentives
such as trade sanctions.
‘Under what conditions will cooperation emerge in a world of egoists without
central authority?’ This question was posed by Robert Axelrod (1984) in his central
contribution to the theory, where he identiﬁed several critical factors. The ﬁrst was
the practice of tit-for-tat. He argued that when agents returned good for good this
initiated a potential spiral of cooperative behaviour. If this practice were repeated,
egoistic agents would gradually learn to trust one another, particularly when their
interests coincided. This situation was formally modelled as a reiterated prisoner’s
dilemma situation. It implied that if states repeatedly found themselves in a situation
in which they feared that their self-restraint would be taken advantage of, they
would not defect but would, instead, devise reinsurance devices that would allow
cooperation to ensue. Reinsurance devices produce institutions. He also theorized
the ‘shadow of the future’, arguing that once cooperation was institutionalized states
would hesitate to abandon it, for fear of what lay ahead. Axelrod went further by
advising participants and reformers to increase the likelihood of mutual cooperation
by enlarging the shadow of the future, by making interactions more durable and/or
more frequent – for example, by breaking issues under negotiation into smaller pieces
– and by changing the payoffs faced by the players.
Central to neo-liberal institutionalism is the notion of transaction costs. These
include ‘the costliness of information, the costs of measuring the valuable attributes
of what is being exchanged and the costs of protecting rights and policing and
enforcing agreements’ (North 1990: 27). Thus, institutions are desirable, despite the
constraints they impose on states, because they reduce transaction costs associated
with rule-making, negotiating, implementing, enforcing, information gathering and
conﬂict resolution. They are also durable. Existing regimes persist even after the
conditions that facilitated their creation have disappeared ‘because they are difﬁcult
to create or reconstruct’ (Keohane 1984: 12–14, 50). This is the logic that lies at
the core of neo-liberal institutionalism: cooperation in situations modelled by the
iterated prisoner’s dilemma can be achieved in highly institutionalized settings
because institutions can serve as means of providing information, reducing transaction
costs and altering the payoffs associated with cooperation. In consequence,
many neo-liberal institutionalists argue that international actors should promote
institutionalization as a means of promoting the collective interest in international
stability.
Constructivist institutionalism, on the other hand, conceptualizes institutions
as a collection of norms, rules and routines, rather than a formal structure (see
Chapter 3). In contrast to rational choice theories like Axelrod’s, institutions do not
simply change the preferences of actors but can also shape their identity (Barnett
and Finnemore 1999). Constructivism focuses on the central role of ideology, rules
and norms that institutions diffuse to constitute agents. Against a ‘logic of
instrumentality’ or ‘logic of consequences’ of rational choice institutionalism,
constructivism posits a ‘logic of appropriateness’, arguing that individuals’ actions
43
C. NAVARI
are guided by social expectations rather than utility maximization calculations.
Institutional routines are followed even when there is no obvious self-interest
involved (see March and Olsen 1989; Finnemore and Sikkink 1998).
There is, however, no single model of the most desirable sort of institution. On
the contrary, the notion of transaction costs points to very different sorts of institutions
for different cooperation problems. For example, the Strategic Arms
Limitation Talks (SALT II) during the 1970s between the US and the Soviet Union
required a set of specialists to determine what might be meant by a ‘technical
advance’, but avoiding the dangers of misreading technical information required a
telephone hotline between the major nuclear antagonists. Neo-liberal institutionalism
has spawned a voluminous institutional design literature that points to the variation
in international institutions and outlines the different institutional arrangements
necessary to address different types of cooperation problems (see e.g. Koremos et
al. 2004; Mitchell 2006).
In this approach, unlike other liberal approaches, states are central. They are the
agents who design institutions to advance their joint interests. Interests are ﬁrst
deﬁned outside the institutional context (in formal language, ‘individual preferences
are exogenous’; they are deﬁned outside of institutional contexts), and then
institutions are designed by state actors to facilitate the achievement of their joint
interests (Keohane 1989; Jupille and Caporaso 1999). Thus, institutions emerge and
survive because they serve to maximize the exogenously determined interests and
preferences of their members, especially those founding members who designed the
institution.
But state-centredness has also led to a central ambiguity in the approach: what
if the state is no longer able to cope with the pressures of interdependence? This
has led to a radical liberal school exempliﬁed by David Held (1995) and Seyom
Brown (1996). In this version, the state is no longer able to cope with international
crises such as the degradation of the environment, mass migration, starvation and
disease. In such a situation, Brown (1996) recommends that we substitute world
interests for the state interests envisioned by more conservative neo-liberals. These
world interests would include the survival of the human species, a reduction in world
violence, the provision of conditions for healthy subsistence to all people, the
preservation of cultural diversity and the preservation of the world’s ecology. But
the approach is rather vague on who should build these new ‘world interest’
organizations.
Neo-liberal institutionalism contrasts in several critical areas with realism. Both
agree that powerful states inﬂuence the formation and shape of international
institutions, but for different reasons. According to liberals, states create institutions
to maximize shared interests; for realists, however, it is to realize and maintain
domination. According to the leading American realist John Mearsheimer, ‘The most
powerful states in the system create and shape institutions so that they can maintain
their share of world power, or even increase it’ (1994/95: 13). Realism also focuses
on the extent to which powerful states dominate institutions; they argue that
latecomers or less powerful members will have less control over institutional decisions
and outcomes, beneﬁt less from their creation and will have less commitment to
maintaining the institution (Gruber 2000). This is quite apart from the general
44
LIBERALISMS
critique that realists make of institutional approaches. ‘Realists maintain that
institutions are basically a reﬂection of the distribution of power in the world. They
are based on the self-interested calculations of the great powers, and they have no
independent effect on state behavior’ (Mearsheimer 1994/95: 7). Neo-liberal
institutionalists argue, on the contrary, that the ‘shadow of the future’ – the possibility
to attain gains in the future – provides a strong incentive for all states to cooperate
and create institutions that beneﬁt all parties.
An equally harsh realist critique of neo-liberal institutionalism is Grieco (1993),
with his concept of relative gains. Grieco argues that relative gains – what a state
in a competitive situation might gain from cooperation relative to what his opponent
might gain, are more important than ‘absolute gains’ – the overall calculus of gains
versus losses. This is so, he argues, because power is a relational concept; power
can only be measured in relation to another’s power, that is, by comparison with
another power-seeker. It matters not if the other gains and I lose, but if the other
gains more than I do. He maintains that the calculus of relative gains often sabotages
hoped-for cooperative ventures if the cooperative venture threatens to change the
balance of power (for discussion of the relevance of the absolute vs. relative gains
argument to neo-liberal institutionalism, see also Snidal 1991; Powell 1991).
The question is, how suitable is neo-liberal insitutionalism with regard to security
issues? Jervis has observed that the realm of security has special characteristics that
at the same time make regime creation more difﬁcult and increase its need: ‘Security
regimes, with their call for mutual restraint and limitations on unilateral actions,
rarely seem attractive to decision-makers’ under the security dilemma (Jervis 1982:
360). Basic to the neo-liberal institutionalists is the idea of common interests that
states could achieve together. But what if antagonists do not share common interests?
According to Jervis (1999: 54), ‘states will establish an institution if and only if they
seek the goals that the institution will help them reach’. It does not seem, superﬁcially,
that institutions could do much to increase security.
The notion that security might lie outside the scope of neo-liberal cooperation
has led neo-liberal institutionalists to focus on cooperation in low politics such as
economy, society and environment and pay much less attention to military security
cooperation. But the persistence and expansion of NATO after the end of the Cold
War created a theoretical puzzle for realists and an opportunity for neo-liberal
institutionalism to move into high politics. Wallander and Keohane (1999), for
instance, explicitly regard NATO as a security institution and try to theorize the
conception of ‘security institution’. First, owing to transaction costs and uncertainty
it is easier to maintain than to create new institutions, which is a basic assumption
argued by Keohane (1984) in After Hegemony. Second, the duration of an institution
mainly relies on the function and extent of institutionalization and organization.
Third, most importantly, the conditions and objects for a security institution’s
persistence are not as narrow as those of alliances. An alliance is for dealing with
common threats, while an institution is for coping with risks, including regional
uncertainty. David A. Lake distinguishes hierarchic institutions from anarchic ones.
He argues that the former are effective in taking actions but can be evanescent, while
the latter, lacking dominant authorities, are less effective but more adaptable to a
changing environment and can last (Lake 2001: 136). In short, Lake, Keohane and
45
TABLE 2.1 Realist and liberal security systems
Theoretical base Realist (Alliance) Liberal
(Community of Law)
Structure of the Material; Static; Anarchic; Social; Dynamic;
international Self-help system Governance without
system government
Conceptions of Basic principles Accumulation of power Integration
security Strategies Military deterrence Democratization
Control of allies Conﬂict resolution
Rule of Law
Institutional Functional scope Military realm only Multiple issue areas
features Criterion for Strategic relevance Democratic system of rule
membership Reﬂects distribution of Symmetrical; high degree
Internal power power; most likely of interdependence
structure hegemonic
Decision-making Will of dominant powers Democratically legitimized
prevails
Relation of system Dissociated; perception Serves as an attractive
to its environment of threat opening for association
z Conclusion
In liberal International Relations theory, the state is not an actor but an institution
‘constantly subject to capture and recapture, even construction and reconstruction’
by coalitions of social actors (Moravcsik 2001: 5). The theory has distinct variants
that supply different motivations for action and that have different implications for
security theory. In ideational liberalism, the underlying motive is social identity, and
conﬂict will ensue if borders do not accord with social identity. Conﬂict will also
ensue across social identities. In commercial liberalism, the underlying motivation
is economic beneﬁt, which does not necessarily lead to cooperation but which
identiﬁes under what sorts of circumstances the economy can be a peace producer.
Wallender argue that NATO persisted because it was not a simple alliance; rather,
it was becoming a security institution (see Chapter 18).
The distinction drawn by Wallander and Keohane (1999) between an alliance
and a security institution has led to a signiﬁcant new typology. Dittgen and Peters
(2001) have contrasted two ideal-type security systems – the alliance-type system
and the community of law-type system – which provide models for the construction
of their respective security systems (see Table 2.1). One is rooted in a realist perspective;
the second in a liberal perspective. The key difference is the response to the
threat. In a liberal community of law, potential disturbances are not dealt with by
mobilizing superior power but rather diffused through integration, by reinsurance
and by conﬂict resolution. Threats are circumvented by common membership in a
security institution.
46
C. NAVARI
47
In republican liberalism, the critical factor is state form and states can be integrated
into long-term peace arrangements that at the same time encourage democratization
and internal state reform. The contribution of liberalism to security theory is dense,
speciﬁed and progressive.
Further reading
David Baldwin (ed.), Neorealism and Neoliberalism: The Contemporary Debate
(Columbia University Press, 1993). This collects the major articles in the debate
between realists and liberals, which still constitutes, arguably, the major axis of
theory in contemporary international relations.
Michael E. Brown, Sean Lynn-Jones and Steven Miller (eds), Debating the Democratic
Peace (MIT Press, 1996). This contains all the classic writings on the democratic
peace and the major criticisms.
Michael Howard, War and the Liberal Conscience (Cambridge University Press, 1978).
This presents what has become the classic account of real liberals in their encounters
with war.
Robert O. Keohane, After Hegemony: Cooperation and Discord in the World Political
Economy (Princeton University Press, 1984). This lays out the ﬁrst systematic
statement of neo-liberal institutionalism.
Andrew Moravcsik, Liberal International Relations Theory (Harvard University Press,
2001). This sets out what the various liberal theories explain, their limits, and how
to operationalize them.
LIBERALISMS