COURSE MATERIAL REVIEW 2

SELECTED READINGS

Liberalism and World Politics MICHAEL W. DOYLE

. the "democratic peace" by re-examining . article the author makes a major contribution to the literat:r:;;for individual liberty exercis~ restr~int

::,:diti,,,;.I Ubeml ,/aim th"'''.""'"~" Z:7!'. :: ;.;:pdi,ti"" th,Offlical "'.ditiom ;/t';':= and peaceful intentions in_ their foreign po~- h. velli and Kant. Despite the cont:adictton_s o i ~a l'b - attributable to three theonsts: Schumpeter, afic ; .;h Kant and other democratic republicans, t at i r

and liberal imperialism, Professor D?yle n . s, w'. states are different and indeed peaceful, 7et t ey !:. doe, lew, a "hmnt legaq, '" fa'."'" •ffa,"-:;,~;,I wo,k ,,., the U,eorim h, dt«, "'?la".'" "'.'t are also prone at times to make war. Budtldtn~ ~p~~y te~ed some of the arguments presented tn this artic e, t ll developed an empmca years have concep ua_ y . th "democratic peace" literature. which is an excellent mtroductton to e

I look at three distinct theoretical traditions oflib~~- . m attributable to three theorists: Schumpetehr, _a rll.- 1s ' 'b al cifism . Mac iave i, liant explicator of ... li er pa .... ' . erialism a classical republican whose glory is an imp

we often ~racthtice; a~d a~:~ns ofliberal pacifism and Despite . ~ con r with Kant and other liberal

liberal !111pen::~if b:1!sm does leave a c?herent republicans, . ffi . Liberal states are different. l on foreign a airs. egacy ful yet they are also prone to

They are indeef ib::~ st~tes have created a separ~te make war. · · · uld and have also d1s- peace, as Kant argued they w;essi~n as he feared they covered liberal reasons for ag~ th ~ the differences might. I conclude by '.'1"gu11:g a_ali and Kant's

liberal acifism, liberal impen sm, . ~ong_ ~ alism are not arbitrary but rooted m liberal mternation . . d the state differing conceptions of the citizen an. .

Political participation, private property, andhequali] 'ty that most liberal states s are, -

of opportum :- erfected them all. Joseph Sc~um- though none has ? h' this family when he considers peter clearly fits wit m . . and democracy. the international effects of capitalism . 1· ,,

, "Sociology of Impena isms, Schumpeter s d t . ed bl. bed in 1919 made a coherent an sus am f

pu 1s '. 'fy' (in the sense o argument c~ncernmg the p~~~r:~nstitutions and' nonaggress1ve) effects of l D yle 1986

h 1955· see a so o , ' princi~le;9~S~~:~:::e of the earlier liberal theo- p_pt. l!h: fo~used on a single feature such as trade or ns s . . II h uments they were failed to examine cntica y t e a~g . f pi-

the mteract1on o ca advancing, Schumpeter sa; foundation of liberal t lism and democracy as e . l a 'fism and he tested his arguments in a socio ogy pac1 , .

1 •

of historical imp_enah~ml ~- " n objectless dispo- ,. He defines imperia ism as a . . d i: 'ble :.

f state to unhm1te iorc1 ' sition on the part o a 1955 6) Excluding Ll 'beral Pacifism expansion" (Schumpeter, " t' t' ord~" and those, :

. erialisms that were mere ca c w . I' ) , . . cal description of liberalism. What ~p "objectful" ( e.g., defensive impena 1~m , .:,.

Th,,'.,'.~';;; ;:~:;:,ml resembles a fumily pomait ?f h::,:~:'ii,, ,oots of ob/ectless hnpcrialism ~~:.'. ; we d . tit tions recognizable by certain sources, each an atavism. Modern impe ,-principl~ ~ ms u 'i indi,idu,l freedom, ... charactenst1cs-for examp e, i

. . Vi I 80 (December 1986),;, . l fi American Political Science Review, o . i I' . " b Michael W. Doy e rom . . "Liberalism and World Po 1t1cs y . . f Cambridge University Press. . . pp. 1151-1169. Reprinted with the perm1ss1on o

MI c HAE L w. Dov LE / Liberalism and World Politics 151 according to Schumpeter, resulted from the com- bined impact of a "war machine," warlike instincts, and export monopolism.

Once necessary, the war machine later devel- oped a life of its own and took control of a state's foreign policy: "Created by the wars that required it, the machine now created the wars it required" (Schumpeter, 1995, p. 25). Thus, Schum peter tells us that the army of ancient Egypt, created to drive the Hyksos out of Egypt, took over the state and pur- sued militaristic imperialism. Like the later armies of the courts of absolutist Europe, it fought wars for the sake of glory and booty, for the sake of warriors and monarchs-wars gratia warriors.

A warlike disposition, elsewhere called "instinc- tual elements of bloody primitivism:' is the natural ideology of a war machine. It also exists indepen-

. dently; the Persians, says Schumpeter ( 1955, pp. 25-32 ), were a warrior nation from the outset.

Under modern capitalism, export monopolists, :J the third source of modern imperialism, push for &f imperialist expansion as a way to expand their closed l\,markets. The absolute monarchies were the last ·.,).·.clear-cut imperialisms. Nineteenth-century imperi- }' 'alisms merely represent the vestiges of the imperi- :t;-1;uisms created by Louis XIV and Catherine the ;·'Great. Thus, the export monopolists are an atavism {of the absolute monarchies, for they depend com- :;pletely on the tariffs imposed by the monarchs and /their militaristic successors for revenue (Schum peter, \#955, pp. 82-83 ). Without tariffs, monopolies would \]be eliminated by foreign competition. }t Modern (nineteenth-century) imperialism, i;.therefore, rests on a atavistic war machine, militaris-

market train people in "economic rationalism"; the instability of industrial life necessitates calcula- tion. Capitalism also "individualizes"; "subjective opportunities" replace the "immutable factors" of traditional, hierarchical society. Rational individu- als demand democratic governance.

Democratic capitalism leads to peace. As evi- dence, Schumpeter claims that throughout the cap- italist world an opposition has arisen to "war, expansion, cabinet diplomacy"; that contemporary capitalism is associated with peace parties; and that the industrial worker of capitalism is "vigorously anti-imperialist." In addition, he points out that the capitalist world has developed means of preventing war, such as the Hague Court and that the least feu- dal, most capitalist society-the United States-has demonstrated the least imperialistic tendencies (Schumpeter, 1955, pp. 95-96). An example of the lack of imperialistic tendencies in the U.S., Schumpeter thought, was our leaving over half of Mexico unconquered in the war of 1846-48.

Schumpeter's explanation for liberal pacifism is quite simple: Only war profiteers and military aris- tocrats gain from wars. No democracy would pursue a minority interest and tolerate the high costs of imperialism. When free trade prevails, "no class" gains from forcible expansion because

raw materials and food· stuffs are as accessible to each nation as though they were in its own territory. Where the cultural backwardness of a region makes normal economic intercourse dependent on colonization it does not matter, assuming free trade, which of the "civilized" nations undertake, the task of colonization. (Schumpeter, 1955, pp. 75-76},'

;~t, ~ttitudes left over from the days of monarchical rars, and export monopolism, which is nothing i.ti6te than the economic residue of monarchical ~·, ance. In the modern era, imperialists gratify their ~tivate · interests. From the national perspective, fpei~ imperialistic wars are objectless. f>\;Schumpeter's theme now emerges. Capitalism gd 4emocracy are forces for peace. Indeed, they 1'fantithetical to imperialism. For Schumpeter, the

·~er development of capitalism and democracy p.s that imperialism will inevitably disappear. . · · aintains that capitalism produces an unwar- 0'. disposition; its populace is "democratized, vidualized, rationalized" (Schumpeter, 1955,

Schumpeter': arguments are difficult to evalu- ate. In partial tests of quasi-Schumpeterian propo- sitions, Michael Haas (197 4, pp. 464-65) discovered a cluster that associates democracy, development, and sustained modernization with peaceful condi- tions. However, M. Small and J. D. Singer (1976) have discovered that there is no clearly negative cor- relation between democracy and war in the period 1816-1965-the period that would be central to Schumpeter's argument.

Later in his career, in Capitalism, Socialism, and Democracy, Schumpeter (1950, pp. 127-28) acknowl- edged that "almost purely bourgeois commonwealths were often aggressive when it seemed to pay-like the

);,The people's energies are daily absorbed in "' tiction. The disciplines of industry and the Ll:i

152 M I c H A E L w. D o Y L E / Liberalism and World Politics

Athenian or the Venetian commonwealths:' Yet he stuck to his pacifistic guns, restating the view that cap- italist democracy "steadily tells ... against the use of military force and for peaceful arrangements, even when the balance of pecuniary advantage is clearly on the side of war which, under modern circumstances, is not in general very likely" (Schumpeter, 1950, p. 128). A ... study by R. J. Rummel (1983) of"liber- tarianism" and international violence is the closest test Schumpeterian pacifism has received. "Free" states ( those enjoying political and economic free- dom) were shown to have considerably less conflict at or above the level of economic sanctions than "non- free" states. The free states, the partly free states (in- cluding the democratic socialist countries such as Sweden), and the nopfree states accounted for 24%, 26%, and 61 %, respectively, of the international vio- lence during the period examined.

These effects are impressive but not conclusive for the Schumpeterian thesis. The data are limited, in this test, to the period 1976 to 1980. It includes, for ex- ample, the Russo-Afghan War, the Vietnamese inva- sion of Cambodia, China's invasion of Vietnam, and Tanzania's invasion of Uganda but just misses the U.S., quasi-covert intervention in Angola (1975) and our not so covert war against Nicaragua. More im- portantly, it excludes the cold war period, with its numerous interventions, and the long history of colo- nial wars ( the Boer War, the Spanish-American War, the Mexican Intervention, etc.) that marked the his- tory of liberal, including democratic capitalist, states.

The discrepancy between the warlike history of liberal states and Schumpeter's pacifistic expecta- tions highlights three extreme assumptions. First, his "materialistic monism" leaves little room for noneconomic objectives, whether espoused by states or individuals. Neither glory, nor prestige, nor ideo- logical justification, nor the pure power of ruling shapes policy. These nonmaterial goals leave little room for positive-sum gains, such as the compara- tive advantages of trade. Second, and relatedly, the same is true for his states. The political life of indi- viduals seems to have been homogenized at the same time as the individuals were "rationalized, individu- alized, and democratized:' Citizens-capitalists and workers, rural and urban-seek material welfare. Schumpeter seems to presume that ruling makes no difference. He also presumes that no one is prepared to take those measures (such as stirring up foreign quarrels to preserve a domestic ruling coalition)

that enhance one's political power, despite detri- mental effects on mass welfare. Third, like domestic politics, world politics are homogenized. Materially monistic and democratically capitalist, all states evolve toward free trade and liberty together. Coun- tries differently constituted seem to disappear from Schumpeter's analysis. "Civilized" nations govern "culturally backward" regions. These assumptions are not shared by Machiavelli's theory of liberalism.

Liberal Imperialism Machiavelli argues, not only that republics are not pacifistic, but that they are the best form of state for imperial expansion. Establishing a republic fit for imperial expansion is, moreover, the best way to guarantee the survival of a state. · Machiavelli's republic is a classical mixed repub- lic. It is not a democracy-which he thought would quickly degenerate into a tyranny-but is character- ized by social equality, popular liberty, and political participation (Machiavelli, 1950, bk. 1, chap. 2, p. 112). The consuls serve as "kings;' the senate as an aristocracy managing the state, and the people in the assembly as the source of strength.

Liberty results from "disunion"-the competi- tion and necessity for compromise required by the di- vision of powers among senate, consuls, and tribunes (the last representing the common people). Liberty also results from the popular veto. The powerful few threaten the rest with tyranny Machiavelli says be- cause they seek to dominate. The mass demands not to be dominated, and their veto thus preserves the lib- erties of the state (Machiavelli, 1950, bk. 1, chap. 5, p.122). However, since the people and the rulers have different social characters, the people need to be "managed" by the few to avoid having their reckless- ness overturn or their fecklessness undermine the ability of the state to expand (Machiavelli, 1950, bk. I, chap. 53, pp. 249-50). Thus the senate and the consuls plan expansion, consult oracles, and employ religion to manage the resources that the energy of the people

supplies. Strength, and then imperial expansion, results

from the way liberty encourages increased popula- tion and property, which grow when the citizens know their lives and goods are secure from arbitrary seizure. Free citizens equip large armies and provide soldiers who fight for public glory and the common good because these are, in fact, their own (Machiavelli,

M I C H A E L W. D O Y L E / Liberalism and World Politics 153

1950, bk. 2, chap. 2, pp. 287-90). If you seek the honor of having your state expand, Machiavelli ad- vises, ~ou. should organize it as a free and popular republic like Rome, rather than as an aristocratic republic like Sparta or Venice. Expansion thus calls for a free republic.

"Necessity"-political survival-calls for ex- pan~ion. If a ~ta~,le aristocratic republic is forced by foreign conflict to extend her territory, in such a case w~ shall see her foundations give way and her- self qmckly brought to ruin"; if, on the other hand ~omestic security prevails, "the continued tranquil~ 1ty would enervate her, or provoke internal dissen- si?ns, which together, or either of them separately, will apt to prove her ruin" (Machiavelli, 1950, bk. 1, chap. 6, p. 129). Machiavelli therefore believes it is necessary to take the constitution of Rome, rather than that of Sparta or Venice, as our model.

Hence, this belief leads to liberal imperi~lism. We are lovers of glory, Machiavelli announces. We s~ek to rule or, at least, to avoid being oppressed. In either case, we want more for ourselves and our states than just material welfare (materialistic monism). Because other states with similar aims thereby threaten us, we prepare ourselves for expansion. Because our fellow citizens threaten us if we do not allow them ei- ther to s~tisfy their ambition or to release their politi- cal energies ~roug~ imperial expansion, we expand. . There 1s considerable historical evidence for

hberal imperialism. Machiavelli's (Polybius~) Rome and _Th~cydides' Athens both were imperial re- publics m the Mac~iavellian sense (Thucydides, 1954, bk. 6). The historical record of numerous U.S. interventions in the postwar period supports ~achiavell!'s argument, but the current record of l~beral_pac1fism, weak as it is, calls some of his in- sights mto question. To the extent that the modern popul_ace actually controls (and thus unbalances) ~e mixe~ r~public, its diffidence may outweigh elite ( senatorial ) aggressiveness.

We can conclude either that (1) liberal paci- fism has at least taken over with the further devel- opment of capitalist democracy, as Schumpeter predicted it would or that (2) the mixed record of liberalism-pacifism and imperialism-indicates tha~ some _liberal states are Schumpeterian democ- ~acies while others are Machiavellian republics. . efore we accept either conclusion, however, we

.;,\must conside~ ~ third apparent regularity of mod- / ern world politics. i" ·,,

Liberal Internationalism

Modern liberalism carries with it two legacies. They do not affect liberal states separately, according to -:hether they are pacifistic or imperialistic, but simultaneously.

. The firs~ of these legacies is the pacification of f~re1gn relat10ns among liberal states. I During the m~et_eenth century, the United States and Great Britam engaged in nearly continual strife; however, after t~e Reform Act of 1832 defined actual repre- sentat~~n as th~ formal source of the sovereignty of the British parliament, Britain and the United States negotiated thei~ ?ispu~es. They negotiated despite, for ~xample, British grievances during the Civil War agamst the North's blockade of the South with which Britain had close economic ties. Despite severe Anglo-French colonial rivalry, liberal France and liberal Britain formed an entente against illib- eral Germany before World War I. And from 1914 to 1~15, Italy, the liberal member of the Triple Alliance with Germany and Austria, chose not to fulfill its obligations under that treaty to support its allies. Instead, Italy joined in an alliance with Britain and ~ranee, which prevented it from having to fight other hbera~ states a~d then declared war on Germany and Aus~ria. Despite generations of Anglo-American tens10n and Britain's wartime restrictions on American trade with Germany, the United States leaned toward Britain and France from 1914 to 1917 before entering World War I on their side.

~egin_ning in the eighteenth century and slowly growmg smce then, a zone of peace, which Kant called the "pacific federation" or "pacific union;' has begun to be established among liberal societies. M~re than 40 li~eral states currently make up the umon. Most are n1 Europe and North America, but they can be found on every continent.

. Here the predictions of liberal pacifists ( and President Reagan) are borne out: liberal states do exercise peaceful restraint, and a separate peace exists among them. This separate peace provides a solid foundation for the United States' crucial alliances with the liberal powers, e.g., the North Atlantic Treaty Organization and our Japanese alliance. This fo_undation _appears to be impervious to the quarrels w1th_o~r all~es that bedeviled the Carter and Reagan a_dm_1mstrat10ns. It also offers the promise of a con- tmumg peace among liberal states, and as the num- ber of · liberal states increases, it announces the

154 M I c H A E L W. D o Y L E / Liberalism and World Politics

possibility of global peace this side of the grave or world conquest.

Of course, the probability of the outbreak of war in any given year between any two giveri states is low. The occurrence of a war between any two adjacent states, considered over a long period of time, would be more probable. The apparent absence of war between liberal states, whether adjacent or not, for almost 200 years thus may have significance. Similar claims cannot be made for feudal, fascist, communist, authoritarian, or totalitarian forms of rule (Doyle, 1983a, pp. 222), nor for pluralistic or merely similar societies. More significant perhaps is that when states are forced to decide oh which side of an impending world war they will fight, liberal states all wind up on the same side despite the complexity of the paths that take them there. These characteristics do not prove that the peace among liberals is statistically signifi- cant nor that liberalism is the sole valid explanation for the peace.2 They do suggest that we consider the possibility that liberals have indeed established a sep- arate peace-but only among themselves.

Liberalism also carries with it a second legacy: in- ternational "imprudence" (Hume, 1963, pp. 346-47). Peaceful restraint only seems to work in liberals' rela- tions with other liberals. Liberal states have fought numerous wars with non-liberal states.

Many of these wars have been defensive and thus prudent by necessity. Liberal states have been attacked and threatened by nonliberal states that do not exercise any special restraint in their dealings with the liberal states. Authoritarian rulers both

· stimulate and respond to an international political environment in which conflicts of prestige, interest, and pure fear of what other states might do all lead states toward war. War and conquest have thus characterized the careers of many authoritarian rulers and ruling parties, from Louis XIV and Napoleon to Mussolini's fascists, Hitler's Nazis, and Stalin's communists.

Yet we cannot simply blame warfare on the authoritarians or totalitarians, as many of our more enthusiastic politicians would have us do. Most wars arise out of calculations and miscalculations of in- terest, misunderstandings, and mutual suspicions, such as those that characterized the origins of World War I. However, aggression by the liberal state has also characterized a large number of wars. Both France and Britain fought expansionist colonial wars throughout the nineteenth century. The

United States fought a similar war with Mexico from 1846 to 1848, waged a war of annihilation against the American Indians, and intervened militarily against sovereigq states many times before and after World War II. Liberal states invade weak nonliberal states and display striking distrust in dealings with powerful nonliberal states (Doyle, 1983b).

Neither realist (statist) nor Marxist theory accounts well for these two legacies. While they can account for aspects of certain periods of interna- tional stability, neither the logic of the balance of power nor the logic of international hegemony explains the separate peace maintained for more than 150 years among states sharing one particular form of governance-liberal principles and institu- tions. Balance-of-power theory expects-indeed is premised upon-flexible arrangements of geostrate- gic rivalry that include preventive war. Hegemonies wax and wane, but the liberal peace holds. Marxist "ultra-imperialists" expect a form of peaceful rivalry among capitalists, but only liberal capitalists main- tain peace. Leninists expect liberal capitalists to be aggressive toward nonliberal states, but they also (and especially) expect them to be imperialistic toward fellow liberal capitalists.

Kant's theory of liberal internationalism helps us understand these two legacies. The importance of Immanuel Kant as a theorist of international ethics has been well appreciated, but Kant also has an important analytical theory of international poli- tics. Perpetual Peace, written in 1795 (Kant, 1970, pp. 93-130), helps us understand the interactive na- ture of international relations. Kant tries to teach us methodologically that we can study neither the systemic relations of states nor the varieties of state behavior in isolation from each other. Substantively, he anticipates for use the ever-widening pacification of a liberal pacific union, explains this pacification, and at the same time suggests why liberal states are not pacific in their relations with nonliberal states. Kant argues that perpetual peace will be guaranteed by the ever-widening acceptance of three "definitive articles" of peace. When all nations have accepted the definitive articles in a metaphorical "treaty" of perpetual peace he asks them to sign, perpetual peace will have been established.

The First Definitive Article requires the civil constitution of the state to be republican. By republican Kant means a political society that has solved the problem of combining moral autonomy,

M I C H A E L W · D O Y L E / Liberalism and World Politics 155 individualism, ~nd social order. A private property and ~arket-o~rented economy partially addressed tha! ~rlemma m the private sphere. The public, or political, sphere was more troubling. His answer was a republic that preserved juridical freedom-the le- gal equality of citizens as subjects-on the basis of a representati_v~ government with a separation of powers. Jundrcal freedom is preserved because the morally_autonomous_ individual is by means of rep- resenta~r?n a self-legrs~ator ~aking laws that apply to all c1t1zens equally, mcludmg himself or herself Tyranny is avoided because the individual is subjec~ to laws he or she does not also administer (Kant PP. pp. 99-102),3 ' '

Liberal republics will . progressively establish pea~e among t~emselves by means of the pacific fed- eratron, or umon (foedus pacificum), described in K~nt's Sec?nd Definitive Article. The pacific union will establish peace within a federation of free states and sec~rely maintain the rights of each state. The world will ~ot have achieved the "perpetual peace" that provides the ultimate guarantor of republican freedom until "a late stage and after many unsuc- cessful attempts" (Kant, UH, p. 47). At that time all nations ~ill have le~rned the lessons of p:ace t~ro~gh nght conceptrons of the appropriate con- stitution, gr~at _an~ ~ad experience, and good will. ~nly then will md1v1duals enjoy perfect republican nghts or the full guarantee of a global and just ~eace. In the meantime, the "pacific federation" of liber~l republics:-"an enduring and gradually ex- p~n~m? federatron likely to prevent war" -brings ~rthm rt more and m~r~ republics-despite repub- lican collapses, backslidmg, and disastrous wars- creating an ever-expanding separate peace (Kant, PP, p. 105),4 Kant emphasizes that

it can be shown that this idea of federalism extending gradually to encompass all states and thus leading to perpetual peace, is practicable and has objective reality. For ifby good fortune one powerful and enlightened natibn can form a republic (which is by nature inclined to seek peace_), ~his will provide a focal point for federal assoc:ratron among other states. These will join up with the first one, thus securing the freedom of each state in accordance with the idea of international right, and the whole will gradu- ally spread further and further by a series of alliances of this kind. (Kant, PP, p. 104)

. The pacific union is not a single peace treaty endmg one war, a world state, nor a state of nations. Kant finds the first insufficient. The second and third are i~possible or potentially tyrannical. National sovere!gnty precludes reliable subservience to a state of nat~ons; a world state destroys the civic freedom on which the development of human capacities rests (Kant, UH, p. 50). Although Kant obliquely refers to various classical interstate confederations and mod- ern di~Ioi:iatic congresses, he develops no systematic orgamzatronal embodiment of this treaty and pre- sumably does not find institutionalization necessary. He appears to have in mind a mutual non-aggression pact, perhaps_ a collective security agreement, and the cosmopolitan law set forth in the Third Defini- tive Article. s

The Third Definitive Article establishes a cos- mopolitan law to operate in conjunction with the pacific union. The cosmopolitan law"shall be limited to conditions of universal hospitality." In this Kant calls for the recognition of the "right of a foreigner not to be treated with hostility when he arrives on someone else's territory." This "does not extend beyond those conditions which make it possible for them [foreigners] to attempt to enter into relations [commerce] ~ith_ the native inhabitants" (Kant, PP, P· 1?6). Ho~prtality does not require extending to foreigners either the right to citizenship or the right ~o s~ttlement, unless the foreign visitors would per- ish rf they were expelled. Foreign conquest and plun~er ~!so find no justification under this right. Hospitality ~oe~ appear to include the right of access and t_h~ obligatron of maintaining the opportunity ~or crt!zens to exchange goods and ideas without rmposmg the o~ligation to !rade (a v9luntary act in all cases under liberpl constitutions).

~~rpetual pe~ce, for Kant, is an epistemology, a cond1t10n f~r ethrcal action, and, most importantly, an expla~~tron ~f_h?w the "mechanical process of nat~re vISibly exh1b1ts the purposive plan of pro- ducr?g concord among men, even against their will and mdeed by means of their very discord" (Kant, PP,~- 108; Uf!, pp. 44--:'15). Understanding history reqmres an ep1stemolog1cal foundation, for without a teleology, ~uch as _the promise of perpetual peace, the complexity of history would overwhelm human understanding (Kant, UH, pp. 51-53). Perpetual pe~ce, however, is not merely a heuristic device with which to interpret history. It is guaranteed Kant explains in the "First Addition" to Perpetual Peace

156 M I c H A E L w. D o v L E / Liberalism and World Politics

("On the Guarantee of Perpetual Peace"), to result from men fulfilling their ethical duty or, failing that, from a hidden plan. 6 Peace is an ethical duty because it is only under conditions of peace that all men can treat each other as ends, rather than means to an end (Kant, UH, p. 50). In order for this duty to be p~a~- tical, Kant needs, of course, to show that peace is m fact possible. The widespread sentiment of approba- tion that he saw aroused by the early success of the French revolutionaries showed him that we can indeed be moved by ethical sentiments with a cos- mopolitan reach (Kant, CF, pp. 181-82). This does not mean, however, that perpetual peace is certain ("prophesiable"). Even the scientifically regular course of the planets could be changed by a wayward comet striking them out of orbit. Human freedo?1 requires that we allow for muc~ greater r~~ersals m the course of history. We must, m fact, anticipate the possibility of backsliding and destruct_ive wars- though these will serve to educate nations to the importance of peace (Kant, UH, pp. 47-48).

In the end, however, our guarantee of perpet- ual peace does not rest on ethical conduct. As Kant emphasizes.

we now come to the essential question regarding the prospect of perpetual peace. What does nature do in relation to the end which man's own reason prescribes to him as a duty, i.e., how does nature help to promote his moral purpose? And how does nature guarantee that what man ought to do by the laws of his freedom (but does not do) will in fact be done through nature's compulsion, without prejudice to the free agency of man? ... This does not mean that nature imposes on us a duty to do it, for duties can only be imposed b! practical reason. On the contrary, nature does it herself, whether we are willing or not: facta volen- tem ducunt, nolentem tradunt. (PP, p. 112)

The guarantee thus rests, Kant argues, not on the probable behavior of moral angels, but o~ t~,at of "devils, so long as they possess understandmg (PP, p.112). In explaining the sources of each of the three definitive articles of the perpetual peace, Kant then tells us how we ( as free and intelligent devils) could be motivated by fear, force, and calculated advantage to undertake a course of action whose outcome we could reasonably anticipate to be perpetual peace. Yet while it is possible to conceive ~f the Kantian road to peace in these terms, Kant himself

recognizes and argues that social e:olution also makes the conditions of moral behav10r less oner- ous and hence more likely ( CF, pp. 187-89). In trac- ing the effects: of both political and 1?oral development, he builds an account of why liberal states do maintain peace among themselves and of how it will (by implication, has) come about that the pacific union will expand. H~ also exp:ains how these republics would engage m wars with nonre- publics and therefore suffer the "sad expe~ience" of wars that an ethical policy might have av01ded.

The first source of the three definitive articles derives from a political evolution-from a constitu- tional law. Nature (providence) has seen to it that human beings can live .in all the regions where they have been driven to settle by wars. (Kant, who once taught geography, reports on the Lapps, the Samoyeds, the Pescheras.) ''Asocial soci~bility" draws men together to fulfill needs for secunty and mate- rial welfare as it drives them into conflicts over the distribution and control of social products (Kant, UH, pp. 44--45; PP, pp. 110-11). This violent natural evolution tends towards the liberal peace because "asocial sociability" inevitably leads toward republi- can governments, and republican governments are a source of the liberal peace.

Republican representation and separation of powers are produced because they are the means by which the state is "organized well" to prepare for and meet foreign threats (by unity) and to tame the am- bitions of selfish and aggressive individuals (by au- thority derived from representation, by general laws, and by nondespotic administration) (Kant, PP, pp. 112-13). States that are not organized in this fashion fail. Monarchs thus encourage commerce and private property in order to incre~se nation~l wealth. They cede rights of representation to theu subjects in order to strengthen their political sup- port or to obtain willing grants of tax revenue.

, Kant shows how republics, once established, lead to peaceful relations. He argues that once the aggressive interests of absolutist _m~n~rchie~ are tamed and the habit of respect for md1V1dual nghts engrained by republican government, wars would appear as the disaster to the people's welfare that he and the other liberals thought them to be. The fun- damental reason is this:

If, as is inevitably the case under this constitu- tion, the consent of the citizens is required to

M I C H A E L W. D O Y L E / Liberalism and World Politics 157

decide whether or not war should be declared, it is very natural that they will have a great hesita- tion in embarking on so dangerous an enter- prise. For this would mean calling down on themselves all the miseries of war, such as doing the fighting themselves, supplying the costs of the war from their own resources, painfully making good the ensuing devastation, and, as the crowning evil, having to take upon them- selves a burden of debts which will embitter peace itself and which can never be paid off on account of the constant threat of new wars. But under a constitution where the subject is not a citizen, and which is therefore not republican, it is the simplest thing in the world to go to war. For the head of state is not a fellow citizen, but the owner of the state, and war will not force him to make the slightest sacrifice so far as his banquets, hunts, pleasure palaces and court festivals are concerned. He can thus decide on war, without any significant reason, as a kind of amusement, and unconcernedly leave it to the diplomatic corps ( who are always ready for such purposes) to justify the war for the sake of propriety. (Kant, PP, p. 100)

Yet these domestic republican restraints do not end war. If they did, liberal states would not be warlike, which is far from the case. They do introduce repub- lican caution-Kant's "hesitation"-in place of monarchical caprice: Liberal wars are only fought for popular, liberal purposes. The historical liberal legacy is laden with popular wars fought to promote freedom, to protect private property, or to support liberal allies against nonliberal enemies. Kant's posi- tion is ambiguous. He regards these wars as unjust 11nd warns liberals of their susceptibility to them (Kant, PP, p.106).At the same time, Kant argues that each nation "can and ought to" demand that its neighboring nations enter into the pacific union of l,iberal states (PP, p. 102). Thus to see how the pacific µnion removes the occasion of wars among liberal states and not wars between liberal and nonliberal states, we need to shift our attention from constitu- tional law to international law, Kant's second source.

Complementing the constitutional guarantee of caution, international law adds a second source for the definitive articles: a guarantee of respect. The sep- aration of nations that a social sociability encourages is reinforced by the development of separate languages

and religions. These further guarantee a world of sep- arate states-an essential condition needed to avoid a "global, soul-less despotism:' Yet, at the same time, they also morally integrate liberal states: "as culture grows and men gradually move towards greater agreement over their principles, they lead to mutual understanding and peace" (Kant, PP, p. 114). As re- publics emerge ( the first source) and as culture pro- gresses, an understanding of the legitimate rights of all citizens and of all republics comes into play; and this, now that caution characterizes policy, sets up the moral foundations for the liberal peace. Correspond- ingly, international law highlights the importance of Kantian publicity. Domestically, publicity helps en- sure that the officials of republics act according to the principles they profess to hold just and according to the interests of the electors they claim to represent. Internationally, free speech and the effective commu- nication of accurate conceptions of the political life of foreign peoples is essential to establishing and pre- serving the understanding on which the guarantee of respect depends. Domestically just republics, which rest on consent, then presume foreign republics also to be consensual, just, and therefore deserving of accommodation. The experience of cooperation helps engender further cooperative behavior when the consequences of state policy are unclear but (potentially) mutually beneficial. At the same time, liberal states assume that nonliberal states, which do not rest on free consent, are not just. Because nonlib- eral governments are in a state of aggression with their own people, their foreign relations become for liberal governments deeply suspect. In short, fellow liberals benefit from a presumption of amity; nonlib- erals suffer from a presumption of ~pmity. Both pre- sumptions may be accurate; each, however, may also be self-confirming.

Lastly, cosmopolitan law adds material incen- tives to moral commitments. The cosmopolitan right to hospitality permits the "spirit of commerce" sooner or later to take hold of every nation, thus impelling states to promote peace and to try to avert war. Liberal economic theory holds that these cos- mopolitan ties derive from a cooperative interna- tional division of labor and free trade according to comparative advantage. Each economy is said to be better off than it would have been under autarky; each thus acquires an incentive to avoid policies that would lead the other to break these economic ties. Because keeping open markets rests upon the

158 M I c H A E L w. D o y L E / Liberalism and World Politics

assumption that the next set of transactio?s will also be determined by prices rather t_han coe_rcion, a_ sense

f t al security is vital to avoid security-motivated o mu u 'd' searches for economic autarky. Thus, avm mg a challenge to another liberal state's security or _even enhancing each others' security by means of alliance naturally follows economic interdepend~nce.

A further cosmopolitan source of libera~ peace is the international market's removal of difficult decisions of production and distribution from the direct sphere of state policy. A foreign state thus does not appear directly responsible for these outcomes, and states can stand aside from, and to some degree above . these contentious market rivalries and be ready~o step in to resolve crises. The interdependence of commerce and the international contac:s of st~te officials help create crosscutting transnational _ties that serve as lobbies for mutual acc~mmoda!ion. According to modern liberal scholars, mternat10nal financiers and transnational and transgovernmental organizations create interests in favor of accommo- dation. Moreover, their variety has ensured that _no single conflict sours an entire rela~io?ship by se~tm~ off a spiral of reciprocated retaliation (Brzezinski and Huntington, 1963, chap. 9; Keohane and Nye, 1977, chap. 7; Neustadt, 1970; Polanyi, 1944, chaps. 1-2). Conversely, a sense of susp~cion, such as t~at characterizing relations between liberal and nonlib- eral governments, can lead t? ~estriction~ on ~he range of contacts between societies, a°:d thi~ can in- crease the prospect that a single conflict will deter- mine an entire relationship.

No single constitutional, int~rnational, or cos- mopolitan source is alone su~cient, but together ( and only together) they plausibly conn~ct th_e char- acteristics ofliberal polities and economies with sus- tained liberal peace. Alliances founded. on mutual strategic interest among liberal and nonliberal ~tates have been broken; economic ties between liberal and nonliberal states have proven fragile; but the political bonds of liberal rights and interests have proven a remarkably firm foundat~on for mut~al nonaggression. A separate peace exists among lib- eral states.

In their relations with nonliberal states, how- ever liberal states have not escaped from the insecu- rity 'caused by anarchy in the world political sys:em considered as a whole. Moreover, the very constitu- tional restraint, international respect for individ~al rights, and shared commercial interests that establish

grounds for peace among !ib~ral st~tes establish grounds for additional conflict m relations between liberal and nonlibera, societies.

Conclusion Kant's liberal internationalism, M~ch~avelli's li~- eral imperialism, and Schumpeter s liberal paci- fism· rest on fundamentally different view~ of the nature of the human being, the state, and mte~na- tional relations.7 Schumpeter's humans are ration- alized, individualized, and democratiz~d. ~hey are also homogenized, pursuing maten~l mteres~s "monistically?' Because their material mte~ests he · peaceful trade they and the democratic state m ' . 'fi' that these fellow citizens control are pa~i isti~. Machiavelli's citizens are splendidly diverse m their goals but fundament~lly u~equal i~ them as well, seeking to rule or feanng bemg dominated: ~xtend- ing the rule of the dominant elite or ~t:di~g the political collapse of their state, each ca s or impe-

rial expansion. . . . , Kant's citizens, too, are diverse m their goals

and individualized and rationalized, but most im- portantly, they are capable of apprecia!ing the ~or~l equality of all individuals and of treating other m?i- viduals as ends rather than as means. The Kantian state thus is governed publicly according to law, as a republic. Kant's is the state that solves the problem of governing individualized equals, whether they are the "rational devils" he says we often find ourselves to be or the ethical agents we can and should become. Republics tell us that

in order to organize a group of rational bein~s who together require universal laws for their survival but of whom each separate individual is secre~ly inclined to exempt hims~lf from them, the constitution must be so designed so that, although the citizens are opposed to one another in their private attitudes, these oppos- ing views may inhibit one anothe~ i? such ~ way that the public conduct of the citizens ~ill ~e the same as if they did not have such evil atti- tudes. (Kant, PP, p. 113)

Unlike Machiavelli's republics, Kant's republics are capable of achieving peace ~mon~ themselves be~ cause they exercise democratic cautio? and are ca~a ble of appreciating the international nghts of forei?n republics .. These international rights of republics

MI c HAE L w. Do Y LE / Liberalism and World Politics 159

derive from the representation of foreign individuals who are our moral equals. Unlike Schumpeter's cap- italist democracies, Kant's republics-including our own-remain in a state of war with nonrepublics. Liberal republics see themselves as threatened by aggression from nonrepublics that are not con- strained by representation. Even though wars often cost more than the economic return they generate, liberal republics also are prepared to protect and promote-sometimes forcibly-democracy, private property, and the rights of individuals overseas against nonrepublics, which, because they do not authentically represent the rights of individuals, have no rights to noninterference. These wars may liber- ate oppressed individuals overseas; they also can gen- erate enormous suffering.

Preserving the legacy of the liberal peace with- out succumbing to the legacy of liberal imprudence .is both a moral and a strategic challenge. The bipo- lar stability of the international system, and the near

· certainty of mutual devastation resulting from a nuclear war between the superpowers, have created a "crystal ball effect" that has helped to constrain the tendency toward miscalculation present at the out- break of so many wars in the past (Carnesale, Doty, Hoffmann, Huntington, Nye, and Sagan, 1983, p. 44; Waltz, 1964). However, this "nuclear peace" appears to be limited to the super-powers. It has not curbed military interventions in the Third World. Moreover, itis subject to a desperate technological race designed

· to overcome its constraints and to crises that have . pushed even the super-powers to the brink of war. ' We must still reckon with the war fevers and moods

of appeasement that have almost alternately swept liberal democracies.

Yet restraining liberal imprudence, whether ~ggressive or passive, may not be possible without threatening liberal pacification. Improving the

i strategic acumen of our foreign policy calls for •. hHroducing steadier strategic calculations of the >national interest in the long run and more flexible 'responses to changes in the international political , ~nvironment. Constraining the indiscriminate med- dling of our foreign interventions calls for a deeper :~~predation of the "particularism of history, cul- .~ure, and membership" (Walzer, 1983, p. 5), but poth the improvement in strategy and the constraint on intervention seem, in turn, to require an execu- 'ye freed from the restraints of a representative leg- fature in the management of foreign policy and a

political culture indifferent to the universal rights of individuals. These conditions, in their turn, could break the chain of constitutional guarantees, the re- spect for representative government, and the web of transnational contact that have sustained the pacific union of liberal states.

Perpetual peace, Kant says, is the end point of the hard journey his republics will take. The prom- ise of perpetual peace, the violent lessons of war, and the experience of a partial peace are proof of the need for and the possibility of world peace. They are also the grounds for moral citizens and statesmen to assume the duty of striving for peace.

Notes 1. Clarence Streit (1938, pp. 88, 90-92) seems to have

been the first to point out (in contemporary foreign relations) the empirical tendency of democracies to maintain peace among themselves, and he made this the foundation of his proposal for a (non-Kantian) federal union of the 15 leading democracies of the 1930s .... I use the term liberal in a .... Kantian sense .... (Doyle, 1983a). In that essay, I survey the period from 1790 to the present and find no war among liberal states.

2. Babst (1972) did make a preliminary test of the sig- nificance of the distribution of alliance partners in World War I. He found that the possibility that the actual· distribution of alliance partners could have occurred by chance was less than 1 % (Babst, 1972, p. 56). However, this assumes that there was an equal possibility that any two nations could have gone to war with each other, and this is a strong assumption. Rummel ( 1983) has a further discussion of the issue of statistical significance as it applies to his libertarian thesis.

3. All citations from Kant are from Kant's Political Writings (Kant, 1970), the H.B. Nisbet translation ed- ited by Hans Reiss. Th~ works discussed and the abbre- viations by which they are identified in the text are as follows:

PP Perpetual Peace (1795) UH The Idea for a Universal History with a

Cosmopolitan Purpose ( 1784) CF The Contest of Faculties ( 1798) MM The Metaphysics of Morals ( 1797)

4. I think Kant meant that the peace would be estab- lished among liberal regimes and would expand by ordi- nary political and legal means as new liberal regimes appeared. By a process of gradual extension the peace would become global and then perpetual; the occasion

160 R o B E RT o . K E o H A N E / From Interdependence and Institutions

for wars with nonliberals would disappear as nonliberal regimes disappeared.

5. Kant's foedus pacificum is thus neither a pactum pacis (a single peace treaty) nor a civitas gentium (a world state). He appears to have anticipated something like a less formally institutionalized League of Nations or United Nations. One could argue that in practice, these two institutions worked for liberal states and only for lib- eral states, but no specifically liberal "pacific union" was institutionalized. Instead, liberal states have behaved for the past 180 years as if such a Kantian pacific union and treaty of perpetual peace had been signed.

6. In the Metaphysics of Morals (the Rechtslehre) Kant seems to write as if perpetual peace is only an epistemo- logical device and, while an ethical duty, is empirically merely a "pious hope" (MM, pp. 164-75)-though even here he finds that the pacific union is not "impractica- ble" (MM, p. 171). In the Universal History (UH), Kant writes as if the brute force of physical nature drives men toward inevitable peace ....

7. For a comparative discussion of the political foun- dations of Kant's ideas, see Shklar (1984, pp. 232-38).

References Babst, Dean V. 1972. A Force f6r Peace. Industrial

Research. 14 (April): 55-58. Bizezinski, Zbigniew, and Samuel Huntington. 1963.

Political Power: USA/USSR. New York: Viking Press. Carnesale, Albert, Paul Doty, Stanley Hoffmann, Samuel

Huntington, Joseph Nye, and Scott Sagan. 1983. Living With Nuclear Weapons. New York: Bantam.

Doyle, Michael W. 1983a. Kant, Liberal Legacies, and Foreign Affairs: Part 1. Philosophy and Public Affairs 12: 205-35.

Doyle, Michael W. 1983b. Kant, Liberal Legac~es, a~d Foreign Affairs: Part 2. Philosophy and Public Affairs 12: 323-53.

Hume, David. 1963. Of the Balance of Power. Essays: Moral, Political, and Literary. Oxford: Oxford Uni- versity Press.

Kant, Immanuel. 1970. Kant's Political Writings. Hans Reiss, ed. H. B. Nisbet, trans. Cambridge: Cam- bridge University Press.

Keohane, Robert, and Joseph Nye. 1977. Power and Inter- dependence. Boston: Little Brown.

Neustadt, Richard. 1970. Alliance Politics. New York: Columbia University Press.

Polanyi, Karl. 1944. The Great Transformation. Boston: Beacon Press.

Rummel, Rudolph J. 1983. Libertarianism and Interna- tional Violence. Journal of Conflict Resolution, 27: 27-71.

Schumpeter, Joseph. 1950. Capitalism, Socialism, and Democracy. New York: Harper Torchbooks.

Schumpeter, Joseph. 1955. The Sociology oflmperialism. In Imperialism and Social Classes. Cleveland: World Publishing Co. (Essay originally published in 1919 .)

Shklar, Judith. 1984. Ordinary Vices. Cambridge, MA: Harvard University Press.

Streit, Clarence. 1938. Union Now: A Proposal for a Fed- eral Union of the Leading Democracies. New York: Harpers.

Thucydides. 1954. The Peloponnesian War. Rex Warner, ed. and trans. Baltimore: Penguin.

Waltz, Kenneth. 1964. The Stability of a Bipolar World. Daedalus, 93: 881-909.

Walzer, Michael. 1983. Spheres of Justice. New York: Basic Books.

From Interdependence and Institutions to Globalization and Governance

ROBERT 0. KEOHANE

The author provides his understanding of how the world works that is "both individua(ist ~nd _instituti~nal- ist." He summarizes much of his earlier work on interdependence and subsequently on institutions and ideas rooted in liberalism. With increasing globalization (albeit in a still only partially globalized world), he sees greater institutionalization as the world becomes more like a polity with governance esse~tial to tra~e, ~na~ce, environment, security, and other matters of global import. Devising better, more effective global institutions to serve the needs of humankind is an imperative.

From Power and Governance in a Partially Globalized World by Robert 0. Keohane. Copyright © 2002, Routledge. Reprinted by permission of Taylor & Francis Books UK.

R o B E RT O . K E o H A N E / From Interdependence and Institutions 161

The ... conception ... of how the world works is both individualist and institutionalist, regarding institutions both as created by human action and as structuring that action. 1 The principal motor of action in this view is self-interest, guided by rational- ity, which translates structural and institutional condi- tions into payoffs and probabilities, and therefore incentives. But my conceptions of self-interest and rationality are broad ones. Self-interest is not simply material; on the contrary, it encompasses one's in- terest in being thought well of, and in thinking well of oneself. One's self-interest is not divorced from one's principled ideas or identity but closely con- nected with them. Furthermore, not all action is necessarily self-interested: actions such as those of firemen rushing into the burning World Trade Center on September 11, 2001, reflect commitment and courage rather than interest.

The resulting conception of how the world works is complex, seeking to take into account subjectivity as well as objectivity, primal urges for power as well as institutional constraints, princi- pled beliefs and worldviews that cannot be vali- dated as well as rational calculation. It therefore lacks parsimony. The core of my contribution to this view of the world has been to explore how in- ternational institutions operate, in the context of interdependence. But my exploration of institu- tions and interdependence has taken place in the context of an awareness of how they are affected by other, broader factors. Hence, I do not assume that institutions and interdependence are the most important aspects of contemporary world politi<:s, that they somehow contain the unique key to his- tory. Indeed, they only make sense if they are fit into the larger puzzle ....

From Interdependence to Institutional Theory

Over thirty years ago, astute observers of the world political economy began to comment on striking in- creases in economic connections among societies and the growing role of multinational corporations. Meanwhile, the literature on the European Commu- nity, pioneered by Ernst B. Haas, focused on how

· economic interdependence affected arrangements for governance. Nye and I picked up on these themes, beginning with our edited special issue of International Organization on transnational relations

(1972), a term that we did not invent but that we did insert into the literature on world politics.

At that time the buzzword for these changes was "interdependence." In the 1970s, Nye and! built a theory elucidating the notion of"complex interde- pendence:' an ideal type for analyzing situations of multiple transnational issues and contacts in which force is not a useful instrument of policy. We defined interdependence itself more broadly, to encompass strategic issues involving force as well as economic ones. In our analysis, interdependence is frequently asymmetrical and highly political: indeed, asymme- tries in interdependence generate power resources for states, as well as for non-state actors. Power and Interdependence, published first in 1977, elaborated this theory and applied it to fifty years of history (1920-1970) in two issue-areas (oceans and money) and two country relationships (US-Australia and US-Canada). There were a number of gaps in our analysis, some of which we acknowledged a decade later,2 but the analysis of the relationship between asymmetrical interdependence and power contin- ues to be useful. ...

Power and Interdependence contained an incipi- ent theory of institutions, in the form of what Nye and I called an international organization model of regime change (1977). But this theory was not well- developed. What preoccupied me for seven years after the publication of Power and Interdependence was the puzzle of why states establish international regimes- rule-oriented institutions that limit their Members' legal freedom of action. In After Hegemony: Coopera- tion and Discord in the World Political Economy (1984), I presented a theory of international institu- tions based on rationalist theory, in particular eco- nomic theories of the firm and of im,perfect markets. I argued that institutions perform important tasks for states, enabling them to cooperate. In particular, insti- tutions reduce the costs of making, monitoring, and enforcing rule&.-transaction costs-provide informa- tion, and facilitate the making of credible commit- ments. In this theory, the principal guarantors of compliance with commitments are reciprocity (in- cluding both threats of retaliation and promises of re- ciprocal cooperation) and reputation ....

My formulation of institutional theory has of- ten been referred to as "liberal institutionalism" or "neo-liberal institutionalism." These labels do not appeal to me, not just bec;use they are awkward. My theory does have its roots in liberalism ... , but the