The Logic of Anarchy: Neorealism to Structural Realism

The Logic of Anarchy: Neorealism to Structural Realism

Barry Buzan, Charles Jones and Richard Little

New York

1993

11. Analogy and Metaphor

At this point the argument moves from an agenda set by the traditional epistemological mission of much modern British and American philosophy to an agenda explored until the last twenty years or so almost exclusively by literary critics and by philosophers in the Continental tradition. It is time to firm up our portrayal of rhetoric, or "the study of how people persuade" (McCloskey 1986:29).

Rhetoric and Metaphor

Now that the attempt to locate the analogy between balance of power theory and microeconomics within a positivist logic of testing and confirmation of theory-as-knowledge has been exposed as incoherent, an alternative path deserves to be explored, in which the relation between economics and international politics is characterized as metaphor. It makes much better sense of Waltz's view of theorizing and explanation as creativity or intuition. It reformulates the difficult question of how, for the pragmatist we take him to be, international theory, microeconomic theory, or any other social theory relates to reality. It fits his views about language as outlined in the previous chapter. It also offers solutions to two abiding weaknesses of political Realism identified in the opening paragraphs of this section: the problem of relativism, identified by E. H. Carr building upon the work of Karl Mannheim in the 1930s, and the all-pervading question of the relationship between science and policy, analysis and prescription.

Here is a theory of metaphor to which we do not subscribe. Normally, people use literal language. Sometimes they embellish in order to gain attention, as a form of shorthand, to focus on some special aspect of our subject, or simply to dazzle. We may speak of "the Eagle" and "the Bear" instead of the United States and Russia. But such figures of speech are optional. We can do without them, and it is always possible, though it may be cumbersome, to spell out literally what is conveyed for ease and pleasure through metaphor or by some other figurative use of language.

A corollary of this view is that when we wish to speak with precision, as in scientific inquiry or where great issues are at stake, we are best advised to eschew metaphor. Thomas Hobbes was very much of this view. He is best known, especially to students of international relations, for his claim that social order could not be maintained except by a central authority backed by force. But he also offered, as a further reason why men and women could not live in ordered self-regulated anarchy, as bees did, that mankind had the ability, through language, to deceive intentionally. The way to suppress deceit in human affairs, he argued, was to avoid metaphor. Counsellors, for example, in assisting the deliberations of princes, were to avoid "all metaphoricall Speeches, tending to the stirring up of Passion, (because such reasoning, and such expressions, are useful onely to deceive)" (Hobbes [1909]:199). Again, in his attack on the temporal pretensions of the Church, Hobbes criticized clerics' deceitful use of terms referring to "incorporeal substances" to mystify the faith. "Satan," he claimed, was simply a word meaning "the enemy." The personification of demons, one of his favorite targets, arose from leaving abstract terms untranslated (Hobbes 1909:354).

The irony of later metaphorical extensions of Hobbes's treatment of individual persons in a state of nature to apply to relations between states in an anarchical system will be apparent. But the immediate point of referring to Hobbes has been to make clear the historical, and hence relative, character of the emphasis placed in modern Europe on the literal. Earlier writers had used metaphor extensively. But a reaction had set in during the second half of the sixteenth century. The task undertaken by Hobbes had already been begun by Peter Ramus, who had produced a revised taxonomy of traditional rhetoric conducive to a separation of content from form, argument from embellishment, the literal from the metaphorical (Hawkes 1972; McCloskey 1986:29-30). He may be seen as redeeming rhetoric for the time being by advocating the combination of logical analysis and rhetorical expression, argument, and embellishment. But in making the distinction between the two so clear he has been seen as paving the way for the rationalist correspondence theories of language of the eighteenth century.

This move, Hawkes argues, coincided with a privileging, facilitated by the new technology of printing, of the written over the spoken word. It coincided also with the parallel tendency, facilitated by the development of printing, for codified law more and more to replace the deliberative processes of king in council and customary or common law and for printed notices to take the place of spoken announcements or judgments (Jones 1987). Writing began to be regarded as more precise, more permanent, and less swayed by the passions than spoken language. Writing took language half way towards Euclidian geometry, which provided an ideal of precision and certainty for the seventeenth century. Hobbes finally codified these new approaches to language so persuasively that for most of the modern period people of European descent have been sharply distinguished from their Muslim contemporaries by their willingness, whether believers or not, to read the Christian scriptures as metaphorical works, variously looking for richnesses of meaning in them through their metaphors, dismissing them as meaningless because metaphorical, or else, rather feebly, claiming that they possessed a different, and more or less privileged, kind of meaning than literal texts, but always accepting a duality which would be regarded as deeply offensive by Muslims if applied to the Qur'an.

This consensus, and the beginning of its downfall, are both evident in a remark made by Immanuel Kant in the preface to the first edition of the Critique of Pure Reason, published in 1781. Kant, and this particular work of Kant's, is referred to here because it provides at one and the same time the most powerful summation of the epistemological mission of Western philosophy and the seeds of a dissolution of this project to be achieved by German philosophy over the following hundred and fifty years. To epistemologists it still provides the most complex statement available at the very end of the European Enlightenment of the tension between empiricism and rationalism. It represents in short the apogee of the philosophical tradition in which Waltz places himself. Yet it can also be interpreted as a Romantic work offering, as does political Realism, the possibility of a radical break with the epistemological project. This makes Kant's remarks on rhetoric and politics of more than passing interest.

In the first critique Kant admits the heuristic value of examples and illustrations, but feels that "such assistance is not required by genuine students of the science, and, though always pleasing, might very well in this case have been self-defeating in its effects." Too often, he continues, "the bright coloring of the illustrative material intervenes to cover and conceal the articulation and organisation of the system, which . . . chiefly concerns us" (Kant [1933]:13). It seems a clear enough opting for the literal. But why on earth should Kant choose to make the point metaphorically, seeming almost to deny his commitment to the truth of the proposition by the way he expresses it? The long locked door between content and expression begins to open. Why else, in the work that follows, do metaphors abound, and abound most freely in the treatment of error in the Dialectic? If metaphor is a source of confusion and error, should it not be eschewed most of all in a treatment of the errors to which human reason is subject? Why, above all, is the Dialectic, explicitly concerned with human reason and knowledge, shot through with political metaphors? And why, finally, do those political metaphors, on examination, turn out to refer to a multiplicity of mutually inconsistent constitutions and plots? It is hard to resist the conclusion that no text can achieve its rhetorical or political objective without at some point compromising the truths it purports to convey. There is always more than one audience, and in an anarchic world one at least of the audiences must be lied to if truth is to prevail? Realism proceeds by the taking off and putting on of masks.

A Romantic Account of Metaphor

This is not the time to spell out detailed answers to these questions. The short answer is that Kant's linguistic practice is indicative of a moment in which the supremacy of the literal was beginning to be challenged. A strain of thought about language running from Vico, through Herder and the English Romantic poets was giving rise to a new attitude towards metaphor, summarized neatly by Terence Hawkes. For the Romantics, he suggests, "metaphor cannot be thought of as simply a cloak for a pre-existing thought. A metaphor is a thought in its own right." As a corollary, "there is . . . no way in which language can be 'cleared' of metaphor' " (Hawkes 1972:55). An extension of this revival of rhetoric has been the widespread substitution for language-as-representation of language-as-metaphor in characterizing the relation between language and the world. This removes the stress from a whole family of concepts including theory, measurement, accuracy, approximation, and truth and places it instead on creativity, richness, elegance, exploration, inquiry, suggestion, interaction, conversation, and success in the general characterization of what scientists are about when they do science. It will readily be seen that there is a family resemblance here between literalist views of language and empiricist characterizations of knowledge on the one hand, romantic theories of language and pragmatist epistemology on the other.

For many contemporary philosophers, even those writing in the Anglo-American analytical tradition, things have now come full circle. Willard Van Orman Quine captured the reversal of view very neatly, affirming that "it is a mistake . . . to think of linguistic usage as literalistic in its main body and metaphorical in its trimming. Metaphor or something like it," he continued, "governs both the growth of language and our acquisition of it. Cognitive discourse at its most dryly literal is largely a refinement rather, characteristic of the neatly worked inner stretches of science. It is an open space in the tropical jungle, created by clearing tropes away" (Quine 1960:188). ("Trope," meaning "turn," is the term used by rhetoricians to describe figurative uses of language.).

To get a sense of the far-reaching implications of the Romantic approach to metaphor the student of International Relations need only consider that most familiar of metaphors in the field, the personification of the state. It is customary to discuss the rights and obligations of states, to speak of international society, and also to point out the limits of the metaphor: states are not so easily killed as individual persons; they have no natural point of physical maturity; they are much more unequal in (physical) power than are individuals; etc. (Bull 1977). There can be few better examples of a metaphor which, far from expressing felicitously an existing state of affairs, creates a reality and a set of obligations that would not otherwise exist. As Carr once put it: "There is a world community for the reason (and for no other) that people talk, and within certain limits behave, as if there were a world community" (Carr 1939:206).

A literalist approach to metaphor might concede this act of linguistic and social creation, but would surely wish to maintain that it was rooted in a fundamental use of the language of personality, obligation, and rights relating to individual persons. The Romantic approach takes a further step, revealing the literal as already metaphorical.

How is this done? In the present example the path leads to the treatment of personality provided by Marcel Mauss. By means of comparative anthropology Mauss set out to persuade his reader that the relationship between the self and the person was not everywhere and always as close as in modern European culture. Mauss observed the manner in which a proper name together with social roles and obligations pertaining to it would be assigned to an individual within some indigenous American cultures for a lifetime, passing on the death of that individual to another carrier. Personality of this kind had a continuity extending beyond any single life and was clearly distinct from one's more primitive sense of being-just-here, of being oneself and not some-body else. He went on to claim that in the West it was only during the classical period that the concept of personality as something superimposed upon individuals began to be extended to "the individual, with his nature laid bare and every mask torn away." Here, surely, was an act of linguistic construction of reality every bit as audacious as the later personification of the state or the partial substitution of political argument for physical battle (Carrithers 1985; Mannheim 1936:34).

And if anyone recalls that Mauss himself preserved the literal even while exposing the conventionality of notions of personality when he noted how a distinction still remained, after this superimposition of personality on the individual, between "the sense of what is the innermost nature of this 'person' (personne), and the sense of what is the 'role-player' (personnage)," it can be conceded that what was true of Mauss was true also of other neo-Kantian theorists of language, metaphor, and consciousness for whom the transcendental deduction, with the burden which it placed on selfhood, remained the most alluring yet problematic passage in the Critique of Pure Reason (M;auuller 1887, 1889). All that can be said here is that any claims for universality of a human selfhood beneath the socially constructed personality is itself vulnerable to attack through social-constructivist approaches to the emotions and the most basic facets of rationality, or, more comprehensively, through philosophical strategies such as that of Nietzsche, which portray the self as created, not discovered (Harr;aae 1986; Murray 1978; Rorty 1989).

It was not until the historical transition described by Mauss, in which personality changed from an incidental to a seemingly essential--even a natural--attribute of each human individual, had taken place: that is to say, it was not until our current "literal" sense of "person" had taken firm root that its metaphorical use to describe corporations, states, universities, and other "moral persons" could develop. And indeed, Mauss dates the emergence of this most significant usage precisely to the early Christian period (Carrithers 1985:19). Further developments take place, which can hardly be traced here. Yet it must be noted, because it is of such great consequence for the original formulation of the current social sciences, that Hobbes, with his stress on the force of human passions rooted in the material, corporeal individual, participated in a materialistic and individualistic revolution in the concept of personality every bit as fundamental as, and parallel to, his assertions of the primacy within language of the literal and the written over the metaphorical, or of positive over natural law. The image that best captures the whole process is of the gradual development of a coral reef in which successive generations of dead metaphors provide the habitat for new tropes.

To sum up, it is not the case that personality, as a quality of human beings, is a clear and uncomplicated foundation on which to build a second-order, metaphorical, more shadowy or conditional notion of group personality. Personality attaching to individual human beings is itself a social institution that may post-date human society and has certainly undergone change over time. Yet this very lack of certainty at the foundation of the metaphor may make it more effective. If we have no trouble handling the dead metaphor of ourselves-as-persons, may not the trouble we encounter with the metaphor of states-as-persons stem from nothing more than its novelty and lack of familiarity?

The Reflexive Nature of Realism

One common reason given by for resisting adoption of the sort of attitude to language and tropes advanced here is that it is held to be subversive of rationality itself. If the literal base (microeconomics). of which much of Waltz's discussion of the balance of power is a metaphorical extension is not in some way more solid than that extension, if it has no superior authority, then what precisely is the force of the metaphor? The argument is first cousin to that employed by empiricists against the sociology of knowledge and glanced at briefly in the opening subsection of the previous chapter. Since this approach to knowledge entered the study of international relations early on, as we have seen, through Carr's reading of Mannheim, it will be best to consider the two arguments in tandem.

In brief, two approaches to a sociological explanation of knowledge may be discerned, roughly parallel to the literalist and Romantic accounts of metaphor. The first holds that error requires causal explanation while truth is guaranteed by the reasoning that leads to it. Error is a deviation from truth just as metaphor is a deviation from the literal. The so-called "strong" position, by contrast, holds that all knowledge is caused, whether it is true or mistaken, including those rational procedures such as mathematics and logic so often considered universal and foundational (Bloor 1976). Here truth is put in question in a manner precisely analogous to that in which the literal is exposed as metaphorical in romantic theories of metaphor. Our contention, recalling the closing paragraph of the previous chapter, will be that awareness--a public awareness--of this problem of reflexivity is a raft offering as much support as the romantic Realist can hope for. The raft is to be preferred when the supposedly solid earth beneath is quaking.

Carr deploys a sociological approach to knowledge in the early chapters of The Twenty Years' Crisis. His critique of idealism or utopianism depends substantially upon this argument. "The outstanding achievement of modern Realism--he writes--has been to reveal . . . the relative and pragmatic character of thought itself" (Carr 1939:87). In brief he notes that thought is generally conditioned not only by the concerns and position of the thinker but also by intention. He then moves swiftly to consider the reflexive character of this criticism, noting "our promptness to detect the conditioning or purposeful nature of other people's thought, while assuming that our own is wholly objective" (Carr 1939:92). Wholly consistent Realism, Carr suggests, would require Realists to accept that their own emphasis on continuity, the power of structural causes, or the inevitability of determined historical change is an index of their own conservatism and cannot therefore place them in a superior position to those idealists whom they criticize (Carr 1939:113). Indeed, Realists generally break their own rules, as they must, when acting in the public sphere, by pragmatically "assuming an ultimate reality outside the historical process" in which to ground their preferred policies (Carr 1939:114).

This argument is neat. If you expose the principles advanced by others as the merest cloaks for interest you expose yourself to the same charge. If you try to claim that all language is metaphorical, you cease to be able to make sense of metaphor, which can only be defined by reference to the literal. If you regard truth as a matter of coherence rather than of the accuracy of knowledge claims you destroy any basis for the defense of your own beliefs. A pragmatist approach to politics has plainly to address reflexive arguments of this form if it is not to inhibit effective action by those who adopt it.

It is helpful to realize, first of all, just what Carr is up to when he turns on Realism after using it to demolish idealism. His book is based entirely around a set of dichotomies: between power and morality, determinism and freedom, practice and theory. The structure of the argument is dialectical: thesis (utopianism), antithesis (Realism), synthesis (principled acquiescence in history). The dialectical structure was calculated to appeal at one and the same time to the Left, who would have recognized it as a distinctively Marxist strategy, and to the British Conservative policymakers it was chiefly intended to influence, who, by and large, would have taken it simply as a somewhat high-falutin' exercise in reasonableness and compromise. To the first audience Carr is saying "The Soviet Union, too, is a revisionist power." To the second: "Ought we not to behave more reasonably to the Germans, who are, after all, our cousins?." Carr would have known well and might have chuckled over the preface to Lenin's Imperialism: "The careful reader will easily substitute Russia for Japan, etc." (Lenin 1939:2).

In short, the reflexive argument purporting to confound thoroughgoing Realism is deployed by Carr as part of a really Realist strategy aimed at effective persuasion of policymakers. For Carr, as for any Realist, persuasion comes higher than truth, rhetoric than methodology, action than theory, and the point is never more effectively proved than through this attempt to dissociate himself from Realism. The apologist of Realism, like the prince, "should appear to be . . . faithful to his word . . . And indeed he should be so. But his disposition should be such that, if he needs to be the opposite, he knows how" (Machiavelli [1961]:100-101). Once again, the taking off and putting on of masks.

Carr, the Realist, surely knew how. He had doubtless seen in Karl Mannheim's work, to which he draws special attention in his preface, a possible escape from the paradox with which Carr, the man of sound common sense and compromise, subdues Realism. The argument goes like this. Understanding how someone came to hold a particular view does not demolish that view. Nor does the fact that the sociological approach may be applied to one's own knowledge claims (or Mannheim's) demolish the sociology of knowledge. On the contrary, understanding the determination of one's own views enhances one's rationality. "What seems to be so unbearable in life itself, namely to continue to live with the unconscious uncovered, is the historical prerequisite of scientific critical self-awareness," writes Mannheim. He continues, closely following the trail already blazed by Dilthey: "Man attains objectivity and acquires a self with reference to his conception of his world not by giving up his will to action and holding his evaluation in abeyance but in confronting and examining himself." He draws attention to the underlying paradox, that "the opportunity for relative emancipation from social determination increases proportionately with insight into this determination" (Mannheim 1936:42-43; Rickman 1961:ch.6).

Reflexive arguments do not defeat pragmatist Romantics; they simply defer "the literal," "truth," and "certainty" indefinitely, allowing the unresolved tensions of the dialectics of metaphor and literal, fiction and truth, subjective and objective, theory and action, to be sustained. We are not, in the end, all liars, trying to convince each other of our particular fantasies. Why not? Because these dialectics, opened up as they are by pragmatism, provide evident means to discriminate between those "lies" and "fantasies" which are able to command sustained assent and those which are not. Hackneyed it may be, but you really can't fool all the people all the time.

Some Implications of Metaphor

If all of this were accepted and the analogy between the balance of power and microeconomics seen as metaphorical in character, to do with creation and political action rather than the testing of knowledge claims, what difference if any would it make? Adopting a neo-Romantic approach, the big difference turns out to be that both the tenor (literal) and the vehicle (embellishment) are transformed by metaphor; it is not a matter of learning from a better developed science (economics) but of seeing what happens to both international politics and microeconomics when they are compared one with another. What was billed as a lecture by professorial microeconomics to neophyte international politics turns out instead to be an interactive seminar.

The resultant conversation is also much more noisy and unruly than that reported by Waltz because there are other participants. Scanning the wave-bands, the well-rehearsed antiphonal exchange between balance of power theory and oligopoly theory suddenly gives way to what sounds like the rehearsals of an amateur orchestra. Four or five principals are playing a recognizable piece. Different groups can be discerned: the play of authority systems against self-regulating anarchic systems, the political against the economic. But some instruments plainly don't quite know quite which group to play with. States and firms are now with one group, now with another. It is not clear whether there is a conductor or not. From time to time there is a snatch from some familiar melody: the theory of the firm, public goods theory, imperfect competition, hegemony, or game theory. Can that be Waltz, on timpani, playing from a score? The others seem to be improvising, but it is hard to tell because the audience are talking a good deal as they move in and out of the auditorium, plainly excited by undisclosed events taking place in the city outside. While the concert has been going on Stalinism has collapsed, Latin America has been humbled, and the disaster of United States arms in Southeast Asia redeemed, so it might appear, in the Gulf.

The methodological conformity with which Waltz needlessly burdens himself restricts international political economy to the application of established wisdom deriving from economics, a rather narrow diet of well-established theory of imperfect competition. The revision proposed here makes it much more reasonable to bring in insights that are perhaps more peripheral or less universally agreed within economics and to try them for size against the structure of the international anarchy. Conversely, it suggests that economists might carry home some fresh insights were they seriously to apply their techniques of analysis to competition between states.

It does not matter that there is no space to do more than scratch the surface of these opportunities here, because to a large extent what is being proposed is simply an overarching framework or methodological rationalization for a great variety of work of the past two decades that has already advanced quite far and has come to be known as international political economy. But to illustrate the nature and breadth of the conversation it may help to look briefly at recent approaches that challenge the quantity theory of competition employed by Waltz, and at the confluence of game theory and public goods theory as applied to international relations.

The standard view of competition which Waltz shares with neoclassical economists holds that "the number of firms and their size distribution [Waltzean distribution of capabilities] is crucial in determining the extent of competition in an industry" (Jenkins 1987:44). Although determination is incomplete, and there will be abberant cases for game-theoretic reasons, it will generally be the case that as the number of firms approaches unity, or as the number of dominant states approaches unity in the bipolar world favored by Waltz, so competition reduces and socialization through imitation becomes more important.

This image of competition is consistent with the idea that concentration of capital through the emergence of dominant multinational corporations in, say, the world motor industry, represents a reduction in the level of competition. But recent attempts to revive the classical concept of competition have challenged this. "Mobility of resources rather than the number of firms in a particular market becomes the crucial determinant of the degree of competition" (Jenkins 1987:45). Competition may be heightened for any given size distribution to the extent that effective concentration of authority over resources within the competing units enables anticipation to substitute for retrospective response to market change, and cost-reduction for the raising of barriers to entry as a competitive strategy. Translated into Structural Realist terminology, differentiation of function may produce very different outcomes at two successive periods in the history of the international anarchy even when the distribution of capabilities appears, arithmetically, to be the same.

In short, some economists--a minority to be sure, but a persistent minority--have been challenging from their side of the fence the simple idea, already challenged in earlier Sections of this book, that competition, even in a necessarily anarchic system, entails units of like function. And by doing so they have, by implication, cast doubt on the notion that a given structural disposition of the international political system will ever determine outcomes, even in the limited sense of shoving in a predictable direction.

In the world of firms it is with the giant--often transnational--corporation that mobility of resources is perfected, and this has led James Clifton to argue that the world economy has become more, not less, competitive with the concentration of capital in giant corporations and the development of oligopolistic world markets. "Capital is always searching out its highest reward at all stages of capitalist development," Clifton concludes. "The fact that it is typically the modern corporation rather than the independent capitalist that pursues this search today does not at all imply a lessening of competition in the capitalist economy" (Clifton 1977:150).

Jenkins, too, stresses the dynamic character of this approach to competition, emphasizing the systematic alternation over time of competition and monopoly. "TNCs," he writes, "give rise to increased mobility of capital between industries and countries. But an important part of competition between TNCs is the search for surplus profits which frequently involves the creation of quasi-monopolistic positions and attempts to protect such positions which are under threat from other TNCs" (Jenkins 1987:58).

What this talk of the mobility of capital achieved by large well-administered firms means is that large firms increasingly face an array of markets (sometimes both as buyers and sellers). including the markets for factors of production, for intermediate goods, for new technology, and for each of their main products. They are able to scan this array searching for opportunities to secure rents which may, in turn, enable them to enhance their competitiveness in less secluded markets. In addition, firms constantly face the issue of internalization, first identified by Coase. What is the limit of advantage to be had from further extension of the administrative grip of the firm? Where is the frontier beyond which the market will provide cheaper solutions? Furthermore, it is apparent that in reaching decisions firms have to anticipate the decisions that their competitors will make, and that subsequent decisions of each competitor will be influenced by past decisions of the others, any series of which is rapidly interpreted as a strategy. Whatever may be the attractions of elegance and simplicity in theory-building, there is always the counter-attraction of a model that retains formality even while incorporating further complexities. Here, the appropriate formal image seems to be of a number of kinds of player simultaneously engaging in a series of multiplayer reiterated variable-sum games.

Translating this into the world of international competition, it may readily be seen that there are indeed several multiplayer games going on between overlapping, though far from identical, groups of participants. States have their games, many of which, like arms racing and reduction, may be modeled in the formal terms of game theory; national economies (represented chiefly by constituency representatives in the legislature, trade union leaders, and employers' organizations) have theirs; truly transnational firms have theirs; publicists have theirs. Players from any game may intervene, either occasionally or habitually, in a game primarily played out between one of the other groups. The winning strategy may not be the same in each group. Deciding how much attention to give to each of the various games is itself a game in which all players participate. Some get this wrong either in the short term (ITT in Allende's Chile) or in the long term (the East India Company). Others get it right (Virginia, perhaps, in its translation from company to state?). The internalization game may also be quite readily transferred from the world of firms to the world of states. Just how great a range of functions can a state perform with economy and across how wide a territory? The answer has varied greatly across time with changes in interaction capacity stemming from new techniques of transport, communications, and warfare. It has varied also, for individual states, according to the relative speed and success with which they and their rivals have adopted new techniques and with the extent to which they were able to give attention to this, as well as to the many other games states play.

The relationship between the British state and British-based capital will provide an illustration of what may variously be called the internalization game, Coase's game, or, when played by states, "Public and Private." In brief, the weak and impecunious state under the Tudors and Stuarts gladly delegates political authority to groups of British merchants operating abroad, allowing them to set up courts, raise troops, engage in diplomatic relations with other sovereigns, and behave as quasi-sovereign entities. At home, the same state attempts centralization and places ever more emphasis on statute law but remains perpetually dependent on local Justices of the Peace and on the customary framework of common law for the maintenance of social order. During the eighteenth century a drift towards more effective central control takes place. Better and cheaper communications and transport progressively remove the barriers to entry which had formerly protected local centers of British power, both at home and abroad. Toward the end of the nineteenth century, during a period of temporary strategic overextension the British once again delegated imperial power, this time to land-based trading organizations in Southern and East Africa.

At the same time that it was being played out territorially, the internalization game was also being played out by states within their boundaries as popular pressure and the need to be able to operate an effective war economy together brought an expansion of state intervention in the economy, the family, local government, and, indeed, every facet of civil society.

These developments, at home and abroad, were clothed in a victorious ideology of welfare or compensatory liberalism. A succession of writers were able to redefine the public realm to embrace many matters of local government or commercial behavior formerly regarded as private. Here, a number of games played out by publicists, businessmen, and politicians came into synchronization in a manner nicely captured in Gramsci's notion of "hegemony" (Gill and Law 1988:64).

It has been argued that one of the first fruits of the replacement of positivist talk of analogy as a form of testing or confirmation by a Romantic view of analogy as metaphor might be to open up the possibility of a fruitful conversation between the two elements of the metaphor--economics and international politics--to which each may contribute and by which each may be changed. But preliminary discussion of the gains to be had from this methodological transformation is dangerously exposed so long as the issue of the scope of state power remains unresolved. It therefore plainly becomes critically important to return to a question anticipated in chapter 10, when the issues of contingent versus necessary anarchy and of the implications of the analogy between microeconomics and the balance of power were first raised. Resolution hinged on the extent to which varieties of social power were ultimately dominated by or dependent upon state power. The question now recurs. Must the kind of elaborate multi-game model sketched just now form part of a viable international theory or does it represent a needless sacrifice of theoretical elegance? Is social power essentially dispersed or is Waltz literally correct in his assertion that "the economic, military, and other capabilities of nations cannot be sectored and separately weighed" (Waltz 1979:131).