|
|
|
|
The Logic of Anarchy: Neorealism to Structural Realism
Barry Buzan, Charles Jones and Richard Little
New York
1993
8. The Structure and Logic of Anarchy
Waltz's assessment of the logic of anarchy is very simple and yet absolutely central to Neorealism. He asserts that an anarchic system emerges as soon as independent political units begin to coact. Once this happens, the logic of anarchy requires that the agents of these units pursue actions that will ensure not only that the political units can survive and reproduce themselves in the anarchic system but also that the anarchic structure of the international system is simultaneously albeit unintentionally reproduced. Unintended actions, however, are not necessarily unanticipated actions. It is important to recognize that while agents may not intend to reproduce an anarchic arena, this does not mean that they are unaware that in consolidating the state they are simultaneously helping to reproduce an international anarchy.
This assessment of the logic of anarchy has been strongly criticized by poststructuralists on the grounds that it obscures some very important features of the contemporary structure of the international system. According to their line of argument, anarchy is a multifaceted phenomenon that can be explored from contradictory perspectives. This assessment strikes a common chord with our argument here: that anarchy itself is a differentiated phenomenon. In other words, we shall challenge Waltz's argument that the logic of anarchy is unaffected by the nature of the units that make up the anarchy. Because we have incorporated functional differentiation into the deep structure of the system, it becomes necessary to take account of the way states reproduce themselves. Because states reproduce themselves in different ways, it will be argued that this can affect the logic of anarchy. Despite the echoes of poststructuralism, the similarity is deceptive. The poststructuralists wish to argue that reality is inherently contradictory and they wish to expose that the international system embraces competing modes of anarchy. We wish to identify how different modes of anarchy are reproduced and transformed. After looking at the poststructural position, therefore, it is necessary to explore in more detail how the reproduction of political units interacts with the logic of the anarchic system.
Ashley's Deconstruction of Waltz
Using poststructuralist techniques, Ashley (1989) sets about "deconstructing" TIP to show how Waltz presents a one-dimensional and inadequate conception of anarchy, arguing that poststructuralists adhere to two fundamental but contradictory truths: that human beings are not free agents, because their lives are controlled by undetected, invisible structures; and that human beings are free agents and can organize their lives in whatever way they like. The purpose of "deconstructing" TIP is to reveal that in promoting the first truth Waltz has no alternative but to "silence" or "conceal" the second. Deconstruction requires the analyst to read the text for its overt meaning and then reread the text to show how the case being made is undermined by the text itself.
Ashley argues that Waltz is intending to suggest that statesmen are in the grip of structural forces over which they have no control. The structure of the international system requires them to take certain actions in order to hold at bay the international anarchy posing a constant threat to the domestic order it is their duty to protect. A poststructural reading of the text reveals that, in developing this argument, Waltz is closing off from the "passive" reader certain essential areas of discussion. In particular, Ashley argues that Waltz "silences" questions about the source of the islands of domestic order that exist in the international arena. The poststructural approach is said to provide a voice which can open up these silences. It reveals (1) that without the identification of a dangerous international anarchy, there could be no notion of a well-bounded domestic society; (2) that the expansion of the modern state's role in domestic society involves hardening the boundary distinguishing domestic and international society; (3) that in order to maintain the conception of domestic space it becomes necessary to constitute ever more encompassing international threats; and (4) that by casting the international society in this negative light, an ambiguous space is created where further problems can be projected.
It follows from this line of analysis that to free the analyst from the restrictions of structuralism it is necessary, first, to displace the state from its central position in discourse about international relations; second, to accept that the distinction between domestic and international politics can no longer provide the starting point for discussing international politics; and, third, to acknowledge that the distinction between domestic and international politics is indeterminate. When this is done, it becomes possible to see, as Ashley (1988) argues elsewhere, that there are two competing conceptions of international anarchy. One view of international anarchy, that of Waltz, identifies a world of competing states, another view, that of the early Keohane and Nye (1972), depicts international anarchy in terms of a transnational or pluralistic world of competing interest groups. Ashley argues that both of these views of international anarchy are valid. But because we have no alternative but to use language to describe reality, there is never any possibility of capturing this ambiguity, except by poststructural analysis, which self-consciously acknowledges that any description of reality necessarily closes off or silences alternative and equally valid descriptions. Ashley's line is unduly pessimistic, as we will try to demonstrate in Section III. Language is well able to express incompatible views and readers can get used to treating texts, like the world of their daily lives, as indeterminate.
The argument being made by Ashley is obviously very different from the one made earlier by Luttwak and Strange. From their perspective, there is a real world out there which is amenable to empirical analysis and description. It is, as a consequence, possible to argue not only that Waltz is representing just one possible view of international anarchy, but also that it is an outdated and unhelpful image, distorting what is actually happening. As Strange sees it, we are living in a period of dramatic change, and to understand what is happening it is necessary to appreciate that "like a chrysalis in the metamorphosis from caterpillar to butterfly, the American Empire today combines features of a national-exclusive past with features of a transnational-extensive future"(1989:11). Failure to appreciate this fact has led analysts, locked in an anachronistic framework, perpetrated by Waltzian type analysis, to argue about whether or not the American empire is losing its hegemonic position. As a consequence, they have overlooked fundamental questions about the nature of power and the basis on which hegemony is established.
Waltz has been remarkably unperturbed by the criticisms leveled at him about structural transformation. Although he has not responded to the most recent arguments developed by Ashley in his poststructural mood, it is likely that he would indicate that in so far as the argument is intelligible, Ashley is simply drawing attention to different levels of analysis. Waltz would insist that he has not "silenced" the analysis of transnational activity in the international system. He explicitly argues that such activity can be understood only at a different level of analysis.
In the same way, Waltz is perfectly happy to acknowledge that his theory cannot account for system change or transformation. But he does question whether any such transformation has ever taken place. For example, in commenting on Ruggie's "fine and rich account of the historical transition from the medieval to modern state" he remains convinced that the account tells us "nothing about the structure of international politics"(1986:328). What happened at the point of transition, according to Waltz, was that the nature of the units changed; the structure of the system was unaffected. Waltz can make this argument because as far as he is concerned, it is immaterial for his theory whether the units of the system are city states, medieval states, nation states, or empires. What his theory explains is why political units will be constrained to take the same form by the anarchic structure of the system. As a consequence, the theory predicts that once one state moved away from the highly decentralized medieval state toward the highly centralized and more powerful nation state, the other political units had quickly to follow suit.
While Waltz's position does not mean that he denies that the structure of the system can change (although he does find it difficult to envisage what form this structure would take) the argument that "the international system is not fully generative" is central to his position. In other words, he does not believe that a theory established at the level of the international structure can explain how changes in that structure can take place. He insists that, like Durkheim, he sees unit level processes "as a source both of changes in systems and of possible changes of systems, hard though it is to imagine the latter." He concludes, therefore, that "Neither structure nor units determine outcomes. Each affects the other"(1986:328). Waltz therefore clearly recognizes that there is an interaction between structure and agency. But he insists that it is necessary to separate out the two levels of analysis. When this is done, it can be seen that system maintenance can be accounted for at the level of structure, while system transformation can be understood only at the level of agency. This line of argument accords exactly with the position adopted by Archer and Taylor but differs from the position adopted by Wendt.
The argument developed so far suggests that it is invalid to criticize Waltz for failing to develop a theory of the state. It is worthwhile looking a little more closely at Wendt's related argument that the international system as well as the state must not be treated as an ontologically primitive concept. Again, Wendt fails to follow through the logic of Waltz's position when he suggests that Waltz has nothing to say about the origins and emergence of the international system. Waltz argues very clearly that the international system comes into existence only when states begin to coact. Before states come into contact with each other they operate in the absence of a political environment.
Wendt makes a further error when he suggests that Wallerstein also fails to examine this issue. Wallerstein, according to Wendt, treats the international system as a primitive or unproblematic concept and is consequently unable to account for the emergence of the international system. The international system is thereby reified and there is a failure to show that the existence of the system is historically contingent and a "problematic creation and recreation of state and class agents"(Wendt 1987:348). But Wendt's assessment fails to take into account that Wallerstein views the international system as the product of a fragmented empire. His theory of the international system is part of his broader cyclical theory of world systems. Built into Wallerstein's theory, therefore, is the notion of system transformation. By contrast Waltz, who assumes that the anarchic international system has been a permanent feature of world history, excludes this possibility.
Anarchy and Competing Modes of Reproduction
This ontological assessment of the international system leads Waltz to presuppose that the logic of anarchy is unaffected by the nature of the units operating in the anarchic arena. In other words, although his theory of the state acknowledges that the units themselves may be constantly evolving, for example, from the medieval state to the modern state, he presupposes that the way these different states seek to reproduce themselves will have no effect on the deep structure of the anarchic arena in which they operate. It is possible to challenge this argument by drawing attention to the well-known and long-standing criticisms of the comparison that has been made between the logic operating in Hobbes's state of nature and the logic operating in the international system. It is often asserted that the two systems share a similar logic. But critics of the comparison argue that the logic is different in the two cases because the nature of the units in the two systems is different. In the state of nature, the logic of the system is seen by Hobbes to generate a situation of absolute and unbearable insecurity. Each member of the system is seen to live in a constant fear of being killed by another member. The critical point about Hobbes's state of nature is that even limited cooperation is ruled out of court.
By contrast, Hobbes considers the condition of war to be tolerable. There exists what Warriner (1957) refers to as "relative security." In the world of states, as has often been pointed out, the potential for instant death does not exist. At least when Hobbes was writing, states could be subdued only over time and they could always inflict wounds on an adversary. In conditions where survival is possible, there is an incentive to cooperate. It becomes possible to conceive of forming alliances and thereby deterring potential aggressors. Hobbes also argues in favor of being heavily armed; to be lightly armed is to become very vulnerable to attack and begin to resemble individuals in a state of nature. The nature of the logic begins to change at that point. The distinction between the state of nature and the state of war suggests that the logic of anarchy is tied in a rather fundamental way to the nature of the units operating within the anarchic structure. As a consequence, to understand the character of anarchy it is necessary to take account of the nature of the constituent units. By incorporating the functional differentiation of political units into the deep structure of the international system, we have made provision for this line of argument, but its implications need to be further examined.
The idea of the state of nature is, of course, a fiction. But it draws attention to the possibility that the activity required to reproduce an anarchic arena is dependent upon the nature of the constituent units in the arena. It is possible to postulate that as the nature of the units change, so too do the methods used to reproduce them. This line of argument has rarely been explored because it has so widely been accepted that anarchy generates a competitive environment. Although this assumption would be challenged by theorists of political anarchy, there have been virtually no attempts by theorists in the field of International Relations, apart from essays by Falk (1978) and Alker (forthcoming), to examine the work of the theorists of political anarchy. This hiatus is not, in fact, surprising, because theorists of anarchy are interested in situations where the existence of the state has been dissolved, while international theorists have traditionally been concerned almost exclusively with the activities of the state, albeit in an anarchic arena. It has been left to theorists like Bull (1977) to emphasize that anarchy and society are not incompatible concepts.
It can be argued, however, that anarchists, like the analysts of Hobbes's state of nature, are dealing with hypothetical systems. And, according to most political theorists, they are unrealizable systems as well. But it is possible to reinforce the argument that there is a complex interaction between the deep structure of an anarchic arena and the units operating in the system by exploring an empirical example provided by anthropology. Generally speaking, anthropologists are drawn upon to support the idea that conflict between independent political units is ubiquitous. It is not possible, in other words, to identify primitive social systems that live in complete harmony with their neighbors. Indeed, many anthropologists have concluded that conflict is endemic between primitive tribes. This finding would surprise neither Hobbes nor Waltz, both of whom would expect conflict to be endemic in an anarchic arena made up of weak and vulnerable tribes.
The Kula and the Logic of Anarchy
There is an interesting exception to the pervasiveness of international conflict which poses a real problem for the Neorealist framework and raises important questions about the relationship between the structure of an anarchic system and the structure of the units within the system. The exception is provided by the tribes who lived on the Trobriand Islands in the Western Pacific. These tribes constituted "stateless" societies because they lacked any form of central government. Although it has become conventional to refer to such societies as "stateless," they do not lack political organization. They constitute independent political units. But because of the absence of a central government, political organization takes an anarchic form. As a consequence, the islanders operate within anarchic units set in an anarchic system. This is, of course, an extreme and unusual case. Although Waltz asserts that the internal constituents of a state will not affect its reaction to the structural constraints imposed by anarchy, he is assuming that the states themselves are hierarchical in structure. It is perhaps not surprising to find, therefore, that the external behavior of the Trobriand islanders does not fit Waltz's theory. What anthropologists have observed is that when these tribes interact, their activities serve to reproduce the essential characteristics of a stateless society and at the same time sustain the anarchic international system.
The tribes were observed by anthropologists to come into contact with each other on a regular basis in order to engage in trade. But the trade was also accompanied by an elaborate ceremony called the kula. It involves an exchange of valuables--Malinowski compared them to the crown jewels--which circulated continuously from one political community to another. There are two types of valuables: long necklaces made of red shell, which travel in a clockwise direction around the islands, and arm bracelets made of white shell, which move in an anticlockwise direction. Every other year, the members of a tribe who participate in the kula travel in a clockwise direction in their canoes to an adjacent island. There, each member of the expedition has a life-long partner from whom he receives a number of shells in the course of the kula. On a return expedition, the partner will receive some necklaces in exchange. There is a high degree of reciprocity in these exchanges. In return for a valuable necklace a donor will expect an equally valuable bracelet. Each political unit conducts the kula with two adjacent units and, because the islands form a ring, the valuables travel continuously in opposite directions.
The kula was first investigated by Malinowski (1922). But the evidence was later reassessed by Uberoi (1962) who has argued that the kula is not simply a peace ceremony but also represents a form of political organization that plays an important role in the power relationships within each society, as well as stabilizing relations between the societies. So, for example, in addition to the formally paired partners, ad hoc exchanges also take place and these give scope for cheating. For example, an individual with a very famous bracelet may use it as collateral on an expedition to a neighboring island, and receive in return several valuable necklaces from partners who each believe they are to receive the bracelet. When the sharp practice is revealed on the return visit, however, animosity is not directed against the member of the neighboring tribe who perpetrated the trick but against the member of the home tribe who is lucky enough to receive the famous bracelet. The acquisition of the bracelet enhances his prestige within the tribe and therefore becomes a source of jealousy. This complex interaction between internal and external transactions indicates that "conflicts within the smaller social group contribute to the cohesion of a wider society" (Uberoi 1962:74). The main source of prestige for all the tribes is, therefore, located in the international system and is obtainable only through the kula. As a consequence, transactions enacted through the kula, serve to reproduce not only the international system but also the decentralized system within each of the tribes. Although it can be argued that the kula is an exceptional mode of international transaction, it does provide a very clear illustration of the close interconnections between the domestic and the international system.
Other examples of isolated regional systems have displayed a distinctive logic. Before Rome unified the Mediterranean region, for example, the Carthaginian empire and the Greek city states were essentially autonomous systems of independent political units. Agents within these systems were, of course, aware that there were other political systems, but their aim was to preserve their independence by minimizing ties with them. Warmington (1960) describes how the Carthaginians endeavored to establish the Western Mediterranean as their special preserve and eliminate the influence of the Greek city states. He argues that a "deliberate attempt to cut a state off from the broad current of Mediterranean advance was not unique nor wholly impractical" (Warmington 1964:61). Because of the relative albeit temporary isolation of Carthage, and its distinctive mode of reproduction, it is possible to argue that relations with its trading partners in the West reflected a distinctive logic. It can then be shown that the logic underlying the relations amongst the Greek city states was quite different.
Carthage and the Logic of Anarchy
Carthage, originally established as a Phoenician colony in the eighth century B.C., was, unlike most of the other colonies, restricted to a narrow strip of land by a treaty with Libya. When the colony began to grow, it therefore became necessary to establish extensive trading links to provide staple products to support the population. This pattern persisted even after Carthage began to extend into its hinterland. Whittaker has argued that until the third century the extensive connections Carthage established in the Western part of the Mediterranean cannot be really identified as imperialism. Evidence of the structures usually associated with imperialism is almost completely absent and there is no evidence of tribute being paid to Carthage. It is true that Carthaginians emigrated, but there is no evidence that they did so to establish positions of dominance in their new community. Whittaker (1978) accepts that Carthaginian troops were posted abroad on the territory of allies but insists that they were there to ensure uninterrupted trade. It is stressed that recent research also reveals the polycentric character of trade in the Mediterranean at this time. There was no centralized trade empire directed by either Tyre or Carthage, although Whittaker does accept that Carthage may have had extraterritorial jurisdiction over its citizens within the boundaries of a foreign port. But Carthage was not interested in using this jurisdiction to promote domination. From its origins, therefore, Carthage was locked into a system where wealth and privilege was associated with trade rather than the possession of land. Its main aim was to promote conditions that would facilitate trade.
It can be suggested, therefore, that Carthage established an international system that involved interaction among what Fox has referred to as "commercial" societies engaged in "long-distance commerce in luxuries or the water-borne exchange of staples that had persisted in the Mediterranean for centuries" (1971:34). Water travel involves a considerable initial investment, but once made, it becomes a relatively inexpensive and easy way of moving goods in bulk. Fox goes on to argue that there is then established "a virtually unlimited potential for linear extension of water-borne communications" and with this ability came correspondingly "increased opportunities for agricultural specialisation" (1971:35). Rational actors recognizing the economic benefits that can accrue from specialization would take advantage of the system and thereby help to reproduce it. These rational actors will thereby find themselves committed to "the interdependence of a large linear (circular) economic system capable of indefinite rationalisation and expansion" (1971:37). Because of this interdependence, moreover, Fox argues that a "consultative approach to action proved more practical than constraint and compulsion." He also observes that because these commercial or linear societies did not have extensive areas to administer, they "normally had no need for government of the sort we usually take for granted in the state" (1971:38).
Herodotus provides an interesting description of Carthaginian trading practices that illustrates the informal and noncoercive basis on which trade was conducted.
The Carthaginians also tell us about a part of Libya and its inhabitants beyond the Straits of Gibraltar. When they reach this country, they unload their goods and arrange them on the beach; they then return to their ships and send up a smoke signal. When the natives see the smoke, they come down to the sea and place on the shore a quantity of gold in exchange for the goods and then retire. The Carthaginians then come ashore again and examine the gold that has been left; if they think it represents the value of the goods, they collect it and sail away, if not they go back to the ships and wait until the natives have added sufficient gold to satisfy them. Neither side tricks the other; the Carthaginians never touch the gold until it equals in value what they have brought for sale, and the natives do not touch the goods until the gold has been taken away. (Herodotus 1987 IV:196)
Although Rome and Carthage were eventually to engage in bitter conflict, for several centuries they coexisted without serious problems arising between them. There was some contact, but it was primarily to remove areas of friction. It is in this light that Whittaker assesses the treaty established between Carthage and Rome in 509 B.C. which gave Carthage control of the ports of trade in Eastern Libya and Sardinia and the treaty of 348 B.C. which prohibited Rome from trading or settling in these regions. Whittaker insists that these treaties arose "out of friendship" and were not "truces dictated from hostility." They are seen to be characteristic of agreements associated with Phoenician or Levantine trading history and were "designed to assist traders and strangers, not to keep them out" (Whittaker 1978:87-88). But the agreements do indicate that Carthage regarded this region as its sphere of influence. This meant that city states in the area were prepared to permit Carthage to define their foreign relations by guaranteeing to protect their shipping.
This relationship represented the defining characteristic of the Carthaginian empire. It can be suggested then that for a time the Roman and Carthaginian Empires coexisted but operated in international subsystems that were independent and subsystem dominant. Indeed, it can be suggested that the agreements established between the two empires were designed to institutionalize the separation, so that there was coexistence without coaction. This example of a regional system that managed to sustain its own distinctive subsystemic form of anarchy immediately raises questions about the factors that brought the subsystem within a broader system where the logic of anarchy described by Waltz dominated.
Whittaker notes that by the end of the fourth century the nature of Carthage's relations within its informal empire and regional subsystem began to undergo a shift. The growing power of Syracuse and Sicily made it more difficult to maintain a stable system, so the level of insecurity in the system increased; and the problem was exacerbated because of the development of central state institutions in both Syracuse and Carthage. Whittaker argues that there was a "shift in the balance in inter-state relations from the personal bonds of friendship between great houses to the more impersonal, collective interests in the ruling elite" (1978:9). Although the more informal bonds never disappeared completely, the shift created a need for formal definitions of alliances and intercity relations, as well as a clarification of spheres of influence and boundaries.
At the same time, Carthage began to move into the hinterland, with a concomitant growth in landed property ownership. Inevitably, this brought about greater inequality, and a greater potential for internal unrest in the event of any foreign invasion. As the internal structures changed, so also did the perception of the international system. Sicily, for example, was seen to be posing a security risk rather than as a source of trading opportunities because it was seen to be a potential base from which an attack on Carthage could take place. It was not long, moreover, before the Romans began to appear as the most likely perpetrator of such an attack. The increase first of Greek, and then of Roman power in the Western Mediterranean raised the interaction capacity of the wider international system, thereby threatening the integrity of the older subsystems. The distinctive logic that characterized the Carthaginian subsystem then quickly changed. The Carthaginian state became more centralized and the capacity to resist attack was enhanced.
The important role initially played by transnational forces in the Carthaginian system was possible because the agents of Carthage saw no need to develop the state in order to preserve their internal or external position. As the potential for internal and external threats increased, so did the desire to strengthen the state. The system became more centralized and in doing so it lost its regional or subsystemic character as the agents of Carthage began to respond to the anarchic structural force of the broader international system. This case can be contrasted with that of the Greek city states which coexisted with Carthage. The city states operated within a wider system, but can initially be characterized as an international subsystem. As will be seen, however, the structural constraints operating within the subsystem continued to influence the city states even after their behavior began to be constrained by the larger system.
Greece and the Logic of Anarchy
Subsystem dominance among the Greek city states began to be challenged when they were confronted by the threat of Persia. The Greek city states, in line with Waltz's balance of power theory, collaborated in order to maintain their independence. In the aftermath of the conflict, subsystemic forces continued to be profoundly affected by the wider international system. At the same time, changes in the internal structure of the city states began to affect the underlying logic of the subsystem. In the first place, agents of the Athenian state began to recognize that their commercial status in a highly competitive international system rendered the city state extremely vulnerable to external threats from the wider international system. When the war with Persia was over, therefore, the Athenians used the possibility of future oriental threats to consolidate their dominating position within the Greek alliance--the Delian League. Athens refused to allow the dissolution of the alliance and it was, as Anderson notes, "converted into a de facto Empire" with about 150 states--at its height--paying tribute to Athens (1974:41).
Athens was able to maintain support in its colonies because it established democratic institutions, which were favored by the poorer sections of the community. The wealthier class, on the other hand, which favored the displaced oligarchic system, looked to Sparta for support where that system continued to flourish.
The internal class division permeated every city state according to Thucydides and created the potential for ubiquitous intervention. Thucydides argues that during the Peloponnesian War between Athens and Sparta, when instability prevailed throughout the Hellenic world, there were
rival parties in every state--democratic leaders trying to bring in the Athenians and oligarchs trying to bring in the Spartans. In peace-time there would have been no excuse and no desire for calling them in but in time of war, when each party could always count upon on alliances which would do harm to its opponents and at the same time strengthen its own position, it became a natural thing for anyone who wanted a change of government to call in help from outside (Thucydides [1954]:208).
Because of these deep rifts in the Greek city states, Ferguson comes to the conclusion that "Dependence on Sparta or Athens was, in fact, regarded by none of their allies except as the lesser of two evils, the greater was dependence on domestic foes" (1913:24).
These powerful transnational ties obviously helped to sustain the Athenian informal empire. But practices that reproduced the democratic state in Athens could not be extended to its allies and it was impossible to weld them into a unified political system. In particular, as Anderson notes, "the direct residential democracy of the mass Assembly" was possible only "within a small geographical compass" (1974:42). In the same vein, he observes that the democratic nature of the system prevented the emergence of any separate or professional apparatus of the state. As a consequence, the Athenian democracy "precluded the creation of a bureaucratic machinery that could have held down an extended territorial empire by administrative coercion" (1974:43). The Athenian empire, therefore, was unsurprisingly short-lived. But the transnational links between states persisted, providing the anarchic system of city states in Greece with a very distinctive character that, as will become apparent later, substantially affected the expansion of Rome into the area.
The Carthaginian and Greek cases open up an obvious weakness in the Neorealist framework, which excludes the possibility of exploring the relationship between the structure of the state and the structure of the system. The two are irrevocably fused in Neorealist theory. Waltz, confident that the structural principle of anarchy will generate a competitive system of units that share a common internal political structure, fails to identify that weakness. It is important to recognize that Waltz's theory of the state does not deny the possibility of an anarchic unit operating within an anarchic system. His theory predicts, in fact, that before coaction takes place, political units can develop in innumerable ways. A state without a central government is certainly a logical possibility. But Waltz's theory also predicts that once coaction is initiated, then states structured in a way that renders them uncompetitive will either have to change their internal structure or accept external domination. Although Waltz does not explore the issue, it is likely that he would agree that states without a central government will be unable to compete successfully with states possessing centralized governments.
But this still leaves open the question of what happens when all the "states" in a system lack a central government. Waltz's framework can predict only that the relationship will be competitive; but the kula case study does not support this conclusion. It follows that there is a need to look more closely at the relationship between political units and the international system within which they interact. In Section I, where we accepted that the basic thrust of Waltz's argument about anarchy is correct, we give causal priority to the anarchic structure of the system, acknowledging Waltz's argument that "anarchy tends to generate like units." Functional differentiation is then characterized as a "deviant" structure in the international system because the existence of functional differentiation demonstrates that the generative effects of anarchy have been unsuccessful. This line of argument identifies a contradiction or tension in our analysis, but the contradiction has been resolved in this section. By defusing the structure of the state and the structure of the system, it has been shown that the external dimension of the balance of power can actually help to sustain functional differentiation.
The implications of this resolution have been extended here. In Waltz's theory of international politics, the logic of anarchy necessarily generates a competitive balance of power. But the cases we examined suggest that this assessment is oversimplified. In the Kula and Carthaginian examples, the logic of anarchy militates against the formation of a balance of power. In both cases, power remains highly decentralized within the political units as well as within the system. Because of the decentralization of power within the political units, neither the internal nor the external balance of power mechanisms come into play. Waltz, of course, can negate these examples by arguing that he is concerned only with the relations among political units where power is centralized. He can then argue that the examples on the whole tend to support his theory. The establishment of a more centralized state in Carthage as the power of Rome began to extend confirms the Waltzian argument that political units become undifferentiated over time. By the same token, the ability of the Greek city states to collaborate when confronted by Persia is also clearly in line with Waltz's balance of power theory. But the advantage of our Structural Realist approach is that it can accommodate for the possibility of differentiated anarchic systems.
The gap in Waltz's theory created by a "transitional" phase is particularly disturbing in the case of the Greek city states, where the political units were structurally differentiated--with power being centralized in some states (Athens and Sparta) and decentralized in others. As Thucydides reveals, in such a system, the logic of anarchy institutionalizes intervention. Since intervention is a good example of the kind of phenomenon which Waltz wishes to explain, the short-coming is significant.
But there are still gaps in our theoretical formulation and they exist in what we have identified as "transitional phases." The failure of the balance of power to protect Carthage, the Greek states, and the Diadochi Empires, which all succumbed to the Roman Empire, for example, is anomalous. We shall return to this gap in our theoretical framework, but first, it may be useful to take an empirical look at why the balance of power apparently failed to restrain Rome.
Rome and the Demise of the Hellenistic System
As already indicated, the Hellenistic system embraced the enormous Diadochi Empires, as well as the small and much more vulnerable Greek city states. At first sight, historical assessments of how the Hellenistic system operated tend to confirm Waltz's theory that the unintended outcome of states seeking to reproduce themselves is the formation of a balance of power that ensures the reproduction of the system. But a closer examination of the system reveals that the balance of power in this system is more complex than Waltz allows because the political units reproduce in a way that functionally differentiates the Greek city states from the Diadochi Empires.
Superficially, the application of Waltz's theory is unproblematic. It has been argued that although there is no evidence that the Diadochi Empires "formally or informally, recognized the principle of the balance of power, in fact such a balance had existed because no great state was in a position to destroy any of the other great states" (Walbank 1981:240). It has also been observed that the third century "saw the creation of an uneasy balance of power between the great kingdoms, with conflict confined to disputed areas: the Ptolemies and the Seleucids fought over Syria and Palestine, while the Greek cities of the Aegean area sought to manipulate the great powers in order to achieve independence" (Boardman, Griffen, and Murray 1986:845). But the independence of Greece was tenuous, because of the proximity of Macedonia. Nevertheless, the leaders of Macedon had to take account of the "equilibrium" as much as did the Ptolemies and Seleucids because to have "aimed at complete domination of Greece would certainly have been regarded by them as a threat to it" (Briscoe 1978:147).
According to this account, therefore, the behavior that generated the balance of power ensured the survival of small as well as large states. But a closer examination of the system reveals the need to incorporate functional differentiation within the deep structure of the system. As indicated earlier, when the internal divisions between the democrats and the oligarchs opened up in the Greek city states, a persistent potential for transnational links was created. Competing parties in the Greek city states displayed a willingness to call in partisan actors from other city states to resolve either internal or external conflicts in their favor. Once the Greek city states were coopted into the larger Hellenistic system, the dynamic that had precipitated this behavior in the Hellenic subsystem was extended to the Hellenistic subsystemic level. Inevitably the dynamic affected the balance of power, giving it a transnational twist. Macedon, the nearest great power, often received requests to support a particular party within one of the Greek city states. As Briscoe notes, if Macedon had turned down all the invitations, they would have been redirected to one of the other great powers, thereby posing Macedon itself with a security problem. But when Macedon did intervene, the hostility of the Greeks on the opposing side was inevitably aroused (Briscoe 1978:148). In the third century Macedonian intervention grew more dictatorial and the level of anti-Macedon sentiment within Greece increased. Then, when Macedonia decided to join forces with Carthage during the Second Punic War, the boundary separating the two subsystems centered on the Eastern and Western Mediterranean was effectively eliminated. With this development, the Greek opposition to the Macedonians began to be expressed in terms of regular calls to Rome for support.
It is important to recognize that the agents of the nascent Roman Empire, after taking control of the Italian peninsula, were still a long way from viewing the world in terms of Waltz's closed anarchic system. They did not see themselves as part of a system of independent states. To the West lay uncivilized, and disorganized, barbarian tribes. To the East and South, on the other hand, lay civilized and long-established communities believed by the agents of Rome to pose a serious and long-term threat to the survival of their state. The Romans were particularly concerned about the power of the Seleucid Empire. As a consequence, the Romans displayed no urgent desire to breach the boundary separating them from the Hellenistic system. Carthage, by contrast, posed a more immediate threat and the outcome of the conflict with Rome was not seen to be self-evident to contemporaries. During the second war between Rome and Carthage, for example, the Greek states were also engaged in a major conflict with each other. At a conference called at Naupactus to seek a resolution, one of the members argued that "it is self-evident even to those of us who have given but scanty attention to affairs of state that whether the Carthaginians beat the Romans or the Romans beat the Carthaginians, it is not the least likely that the victors will be content with the sovereignty of Italy and Sicily, but they are sure to come here and extend their ambitions and their forces beyond the bounds of justice" (Walbank 1981:231).
The association of the Eastern Mediterranean with the subsystem in the Western Mediterranean containing two dynamic and expanding states seemed to create an extraordinary degree of insecurity. The speech reveals not only that the outcome of the struggle was uncertain, but also a lack of faith in the balance of power. As far as this Greek is concerned, whatever the outcome of the struggle between Rome and Carthage, the security of the Greeks was likely to deteriorate. It could, however, also be argued that the Greek was merely making a rhetorical statement, designed to persuade the Greeks to make peace without outside intervention.
Taking a wider perspective, it appears that the response of the actors in the Hellenistic system accords very closely with the balance of power expectations in Waltz's theory. The alliances between Macedon and Carthage and between Rome and the Greeks can certainly be accounted for in terms of balance of power prescriptions. The success of the Roman Empire cannot be attributed to the failure of the external adjustments in the balance of power. The inability of the Hellenistic system to resist the Romans flowed in part from internal structural constraints that made it impossible for the political units in the Hellenistic system to copy the structural features of Rome, discussed earlier, which facilitated Rome's acquisition of power. But in addition to these balance of power considerations, Rome was able to take advantage of the fact that the Hellenistic system was rule governed. To understand the nature of this advantage, however, it is necessary to look in more detail at how the existence of rules can affect the logic of anarchy.
Rules and the Logic of Anarchy
Because Waltz establishes a conception of deep structure that militates against cooperation in the international system, his theoretical framework precludes the possibility of explaining the complex body of rules governing the relations between states (Ward 1991). As theorists have become increasingly interested in the process whereby cooperation takes place and rules get formed, they have become increasingly frustrated with the restrictive framework imposed by Waltz where rules and cooperation appear as anomalies (Kratochwil 1989; Onuf 1989). In Waltz's formulation, the anarchic system poses a structural obstacle to cooperation which has to be overcome. Cooperation and rules, as a result, are seen to be extremely tenuous and fragile, in constant danger of being overthrown because they exist in fundamental tension with the structure of the system. Yet this is a rather odd conclusion, because rules and cooperation are, like war and alliances, universal features of the international system. Why should they not be explained in terms of the structure of the international system? As Kratochwil and Ruggie have pointed out, one of the insights derived from game theory is that both "conflict and cooperation can be explained by a single logical apparatus" (1986:762).
Frustration with the self-imposed restrictions on analysis imposed by Waltz has led to the erroneous conclusion that he presupposes that states operate on the basis of a nonstrategic form of rationality (Snidal 1985). In other words, according to Snidal, Waltz works on the assumption that states do not take each other's preferences into account when they arrive at decisions. But this is quite wrong. Waltz's analysis can easily be accommodated within a game theoretic matrix. Waltz depicts states as rational actors seeking to survive in an anarchic arena. As Grieco (1988a, 1988b) has shown, Waltz's analysis suggests two structural factors that inhibit cooperation. Both require the state to take the preferences of other states into consideration. The first is the potential for cheating in an anarchic arena; all states are thereby encouraged to compete rather than cooperate. The second obstacle arises because the benefits of cooperation are rarely symmetrical, and in an anarchic arena which breeds competition, states would rather forgo the benefits of cooperation in preference to seeing a competitor improve its relative power position.
But despite the force of these arguments, the fact remains that states do cooperate and develop rules among themselves. Analysts working within the rational choice model have suggested that cooperation can take place in the anarchic arena when the potential for mutual benefits exists and when what has come to be called "the shadow of the future" prevails. (Oye 1985) Because states know that they are in a relationship, the threat of cheating is diminished because a potential cheat knows that other states will reciprocate and any benefit gained by cheating will be short-run. Under these circumstances, cooperation can enhance the chances of a state surviving in the anarchic arena. It is denied, in other words, that cooperation and rules exist in opposition to the deep structure of the international system.
The potential for cooperation and rule formation in the anarchic international system has also been reexamined from the perspective of literature which has endeavored to overcome the problem of structure and agency. This reexamination has led Dessler (1989), for example, to question the ontological assumptions underpinning Waltz's theory. In particular, he attacks Waltz's assumption that structure can emerge only as a byproduct of the interaction between states. (1989:450) Waltz, in other words, assumes that structures can be formed only as the result of the actions of rational agents who are not intending by their actions to reproduce the structure of the international system. Dessler rejects this ontological posture in favor of another, associated with analysts such as Giddens and Bhaskar, which, he argues, provides a richer and therefore more effective theory of social action. In essence, the new model not only embraces the interactions between states that unintentionally preserve the anarchic arena, but also makes room for intentional interactions that perform the same function.
This line of argument does not deny, of course, that actions taken to preserve the state have the unintentional consequence of reproducing the deep structure of an anarchic system. But such actions are not incompatible with the actions performed by states when they self-consciously set out to reach agreements with each other. In making such agreements, states explicitly confirm each other's sovereignty and therefore actively help to reproduce the deep structure of the system. Treaties and acts of cooperation all intentionally serve to reconfirm and reproduce the anarchic system of independent states.
So we are brought back to our starting point which asserted the importance of recursiveness in accounting for the relationship between structure and agent. Onuf has argued that "If recursiveness is the key conceptually, then rules are the key operationally" (1989:62). Waltz reveals the flaw in this line of argument, because he demonstrates that it is possible to conceive of an anarchic structure being reproduced unintentionally. But what the recent literature on anarchy and regimes has revealed is that the logic of anarchy does not preclude cooperation and that anarchy, therefore, is not incompatible with the formation of rules. Once these rules are formulated, they constitute in our terms a process formation, constraining the behavior of the member states. It then becomes possible to distinguish between anarchy operating in the absence of rules and anarchy operating in accordance with them. This corresponds to the distinction between an international system and an international society. As Bull and Watson put it:
By an international society we mean a group of states (or, more generally, a group of independent political communities) which not merely form a system, in the sense that the behavior of each is a necessary factor in the calculations of the others, but also have established by dialogue and consent common rules and institutions for the conduct of their relations, and recognize their common interest in maintaining these arrangements (1984:1).
Bull and Watson use this distinction to discuss the evolution of the modern international system, but it applies equally well to the Hellenistic subsystem into which Rome expanded (Wight 1977:chs. 2,3).
Rules and the Expansion of Rome
Having located the role of rules in an anarchic arena, it is possible to distinguish between the way Rome expanded to the West and to the East. In essence, in the West, the Romans used military force to expand the empire while in the East they were able to rely much more extensively on the rules associated with diplomacy to regulate or contain any potential threats. The Romans did send troops East, but never initially for the purpose of conquest.
To a very large extent, this divergence was dictated by the very different anarchic structures in the two sectors. In the West, expansion could be achieved only by military conquest. The conquest also proved to be a long drawn out affair, and the Romans were not anxious to see troops permanently tied down in the East before the West was secure. The Romans found, however, that in the areas influenced by Hellenic culture in the East they were able to establish diplomatic links similar to those which existed in Roman municipal life. These links established a system where the dominant party provided rewards in return for services rendered. In the context of the international system, Rome offered security guarantees in return for services rendered by the client states.
The Romans implemented this strategy when Macedon decided to join forces with Carthage during the Second War with Rome. Once this alliance was forged between Carthage and Macedon, Rome quickly moved to establish an alliance with one of the major power groupings in Greece, the Aeotolian League. The Greeks agreed to the link because of the disaffection with Macedon. As a result, the influence of Macedon was replaced by that of Rome. Instead of annexing Greece, Rome developed a stable system of alliances. These alliances were based on treaties which in turn were built on the bedrock of diplomatic rules and practices that represented an important feature of the Hellenic and later the Hellenistic subsystems. As a consequence, the alliances with client states in the Hellenistic system were always more effective than the client states which the Romans endeavored to establish in the barbaric kingdoms of continental Europe. Nevertheless, the Romans tried to promote the German tribes into clients by educating their rulers. By channeling money and favors to certain key chiefs, they helped them to increase control over their own subjects, thereby tightening the hold which the Romans had over the chiefs. (Luttwak 1976:36)
When Macedon tried to renew its influence in Greece in 174 B.C., however, the Romans chose to see the move as threat to the Greek allies, and in place of diplomacy they resorted to force. The Macedonians were decisively defeated at Pydia in 168 B.C. The Macedonian monarchy was brought to an end and the territory was divided up into four republics. Twenty years later, when this solution failed and Rome was once again confronted by resistance in Greece and Macedonia, both were annexed and Greece was placed under the control of the Governor of Macedonia.
As Crawford observes, whenever one solution failed, the Romans were always sufficiently flexible to try another (1978:94). The move East was slow and cautious and the Romans displayed a constant reluctance to annex territory, preferring always to rely on diplomatic methods that presupposed the existence of established rules and a mutual willingness to abide by agreements. Roman expansion to the East eventually came to a halt when the Parthian Empire was confronted. Once the Romans discovered that they were unable to use either force or diplomacy to dominate the Parthians they recognized that it would be beneficial to establish a diplomatic settlement with them. Roman security depended primarily on the agreements struck with the Parthians on the Eastern border of their empire.
The agreements between the Romans and the Parthians open up a number of theoretical questions about the nature of the relationship between the two empires. It is necessary, in particular, to pose the question as to whether the agreements enabled the two empires to coexist rather than coact. This distinction plays an important role in both Waltz's theory and in ours. In Waltz's framework the distinction is used to identify when a system comes into existence; in our framework, the distinction helps to identify the movement from subsystem dominance to system dominance. This issue needs to be located in the context of a wider discussion about the relationship between continuity and transformation in the international system.
Email this citation