Most available explanations fail because they ignore the fact that many different kinds of states were viable at different stages of European history, because they locate explanations of state-to-state variation in individual characteristics of states rather than in relations among them.1
THIS CHAPTER presents my model of change. It begins with a discussion of two theories of change which have informed my nonlinear view of institutional evolution. I argue that a change in the constitutive units of the system is only likely to occur after a broad exogenous change, or an environmental shock, if you will. Such an exogenous change will lead to political and social realignments.2 The divergent composition of these political alliances accounts for the variation in new institutional types. In the long run, however, some institutions will disappear given the competitive nature of the international system. In short, following a broad, contingent change in milieu, actors choose to create institutions that meet their material interests and ideological perspectives. But the subsequent viability of institutions is constrained by their relative competitiveness.
This book is influenced particularly by the images of change proposed by two theorists, Fernand Braudel and Stephen Jay Gould. Both classify change by periodization and pace. Fernand Braudel, the French historian, suggested that the pace of change operates at three levels.3 This parallels the different types of systems change in international relations. Gould’s view of Darwinian selection, specifically the punctuated equilibrium model, serves as a useful metaphor in classifying institutional change.
Braudel argued that the pace of history can be captured by a threefold classification. It is fastest at the level of everyday occurrences, a heterogeneous tapestry textured by individuals’ decisions and events in life. This is l’histoire evenementielle, the history describing continually changing microlevel actions. The tempo of such change is rapid and is paced by individual time. At the intermediate level, history is less transitory. Economic changes and political shifts are more structural in nature. This is social history that aggregates individuals and human structures. Change, particularly in economic affairs, fluctuates in cycles of ten, twenty, or perhaps even fifty years. Using not only the metaphor of economic cycles but even its terminology, Braudel called this the history of conjunctures. Finally, there is the level of history that seems to move at a virtually imperceptible pace. Looking at his own rural roots, Braudel saw little change in village life between the Middle Ages and the early twentieth century because the objective conditions of life had not altered all that much. Demography, geographic conditions, and the technology of communication all conspired to make life appear unchanged to the individual. Such forces, however, were critically important in explaining the long-term fortunes of geographic areas such as the Mediterranean basin, or in demonstrating the relative similarity in conditions facing the French farmer of the Middle Ages and of the nineteenth century. The objective conditions that created such structural stasis made for the history of the longue durée.
In the Braudelian scheme, the tempo of change thus varies between the instant, the cyclical, and the longue durée. The task of the historian is to demonstrate the interplay between these three. Seen from another perspective, l’histoire evenementielle and longue durée occupy opposite ends of the spectrum of agency and structure. Through microlevel choices, individual agents create a history of events. Conversely, factors such as weather and mortality rates often structurally determine the course of history over a long time frame.
Braudel’s taxonomy of pace resonates across disciplines.4 Scholars of international relations have been similarly confronted with the problem of how to categorize change.5 One way of doing so strongly resembles Braudel’s categorization of tempo. Change in international relations might be categorized as interaction change, rank order change, or change in the constitutive units.6 Interaction change, the change of diplomatic practices, is the most susceptible to individual decision making. Such practices are influenced by the presence of particular decision makers and by specific strategic choices. By contrast, shifts in the distribution of capabilities occur less frequently.7 Changes in relative power, and the subsequent challenges to the existing rank order by ascending powers, occur, by some accounts, every century or century and a half. Such changes might correspond with periodic cycles in the economy. Finally, unit change, for example, the change from city-states to empires, or from empires to feudal organization, occurs the least often. When a particular type of unit comes to dominate the international system, it transforms the deep structure of the system. The more frequent changes in interactions and rank order occur without affecting the particular character of this deep structure. For example, diplomatic practices and the rank order of states have changed in the past decades, but all this has happened without affecting the existence of a system of sovereign, territorial states. These historical periods in which particular types of units tend to dominate constitute the longue durée of international politics.
But using a Braudelian scheme does not explain why such transformations occur infrequently or how such changes might look when they do occur. Here the punctuated equilibrium model can be of some use.8 This second image of change comes—somewhat ironically—from biology.9 Stephen Jay Gould rearticulates the Darwinian perspective and challenges common interpretations of Darwin’s view of evolution as a linear and progressive process.10 Both the “ladder” and the “cone” views of evolution are, according to Gould, mistaken. In the ladder interpretation, evolution is defined as progression along steps of a linear process. Each step follows from and is superior to the preceding step. In the cone view, evolution is perceived as a process of continual change wherein life becomes progressively more diversified and more complex, and hence more advanced. Both of these views are wrong. “Evolution, to professionals, is adaptation to changing environments, not progress.”11 One explains evolution not by suggesting that the existing life form is superior to the previous—Gould beautifully illustrates that early life forms were highly complex—but by demonstrating how it is well suited to exist in the new environment.
Gould differs from Darwin, however, on the pace of evolution. In Gould’s view, change can be dramatic and very quick. It takes the form of “punctuated equilibrium.” Stages of relative tranquility are interrupted by sudden and dramatic changes. Such broad exogenous change—punctuation—will lead to a flurry of radically new forms.12 In the long run, some of these forms may die out and a period of relative tranquility will ensue—a period of relative equilibrium. Whatever forms survive are not explained by reference to the types preceding the exogenous shock but by reference to the new environment and the now simultaneously existing forms which emerged after the shock.13 Gould’s discussion of this two-staged nature of evolution is worth emphasizing in detail.
Darwinism, on the other hand, is a two-step process, with different forces responsible for variation and direction. Darwinians speak of genetic variation, the first step, as “random.” This is an unfortunate term because we do not mean in the mathematical sense of equally likely in all directions. We simply mean that variation occurs with no preferred orientation in adaptive direction. . . . Selection, the second step, works upon unoriented variation and changes a population by conferring greater reproductive success upon advantageous variants.14
Gould’s view of evolution serves as an alternative to the unilinear imagery that I criticized in Chapter 1. But is it appropriate for social science, and international relations in particular? Social and biological evolution are different, and the critical difference between them is intentionality in the first, variation, phase of evolution. Although biological organisms mutate at random, social groups may intentionally form political coalitions to deal with particular environmental constraints. Nonetheless, even here Gould’s theory proves to be a useful metaphor.15 First, the institutional outcomes of such bargains need not be efficient responses to external challenges. Furthermore, these outcomes are not a priori predictable because they involve elements of political strategy (more on this below). Moreover, the actors, when forming these coalitions, are incognizant of the long-term ability of their preferred institutions to compete with rivals. Created institutions are thus unoriented toward their long-run survivability, and the causes behind their emergence are different from the causes of their demise.
Second, empirically we can distinguish periods in which particular units (feudal, imperial, or territorial states) have become dominant following a dramatic change in military organization, economy, or culture. That is, changes in political organization seem to be preceded by broad shifts in constraints and opportunities imposed on social actors by the external milieu. Punctuations such as defeat in war, revolution, or emergent capitalism lead to a flurry of institutional innovations. Such phenomena are the sociopolitical equivalent of speciation events in biology.
Third, there are good reasons why actors do not redesign institutions unless conditions force them to do so. Transaction costs, set belief systems, and standard operating procedures mitigate against frequent overhaul. Moreover, given the fact that institutions reflect a particular distribution of power, such changes are unlikely to occur without fundamental shifts in that distribution.16 Once one form has established itself as dominant, relative stability in institutional types should follow.17 There is a certain path dependency in institutional design.
In short, the uneven nature of political change is well captured by the punctuated equilibrium model. Given that existing institutions cater to particular interests and reflect specific distributions of power, institutions will be “sticky.” Only when dominant coalitions change, or interests and perceptions shift, will there be an opportunity for institutional transformation.
Most importantly, this metaphor warns us that history does not work by optimal design. The species or institution that proves to be successful over time is only successful vis-à-vis its synchronic alternatives. Following a punctuation event, type A might prove to be better than type B or C. But type A need not be an optimal response. Indeed, if type A did not emerge as a type in the variation stage, type B might become dominant in the subsequent selective process with C. For example, changes in the economic and military milieu of China did not lead to territorial states there. Yet it is clear that the Tang and Sung empires were well ahead of European development. But neither expanding trade nor gunpowder caused territorial states to emerge.18 As in the first step of Darwinian evolution, the emergence of new types is not predictable by the change in milieu alone. However, once some new types have established themselves, the second step of the Darwinian process—selection—will reward only some or even only one of these.
I essentially argue that broad-based external change has a variety of internal repercussions. In response to such an external change, new political coalitions will form, based on material interests and shared conceptual frameworks. The expansion of trade was the critical external change that set this process in motion during the High Middle Ages. The variety of novel institutional arrangements—city-leagues, city-states, and sovereign, territorial states—can be explained by the particular nature of new political coalitions in response to the changing environment. The three cases I examine—the Hanseatic League, the Italian city-states, and the French territorial state—are all explained in this fashion. The growth of trade and corresponding increase in urban centers gave birth to new sets of arrangements between kings, aristocracy, burghers, and church.
Many theorists have argued that the internal developments of a country cannot be understood without taking into account that country’s position within its external environment. This view lies at the heart of Hintze’s explanation of how a state’s external vulnerability, because of geographic location, might lead to authoritarianism.19 The argument that changes in the economic market led to political realignment lies at the basis of Barrington Moore’s explanation of authoritarianism and democracy.20 Theda Skocpol’s explanation of revolutions and Peter Gourevitch’s account of foreign economic policies also follow this logic.21
Similarly, social realignments explain the emergence of new types of political organization, for example, during the High Middle Ages.22 Actors have particular interests and perspectives and corresponding preferred forms of organization. An external change, a change in the overall milieu in which that society is placed, will lead to a shift in the relative power of social and political actors. Individuals will seek to capitalize on their improved relative position and change the existing political institutions. I thus take a methodological individualist perspective. Individuals have reasoned preferences to support one institution over another. The emergence of new institutions must thus be traced back to the ability of actors to pursue those preferences.
The outcome of such realignments, however, is not a foregone conclusion.23 Realignments are essentially permutations and combinations of bargains based on material interests and shared belief systems.24 Some actors choose in a satisficing way. They might lack information or might be ignorant about the probability of achieving their most preferred outcome. Moreover, because powerful actors may have widely divergent preferences, second-best solutions and compromises abound. Furthermore, actors work with divergent discount rates. The long-term future of the city-state clearly was not in anybody’s mind in the period leading up to the Italian Renaissance. Political entrepreneurs and burghers were more concerned with immediate success in war or profit. To make matters worse, outcomes might even be serendipitous or accidental to what actors set out to achieve. Finally, institutions might be caught in low-level equilibrium traps.25
In short, we cannot simply deduce institutional outcomes from preferences or impute preferences from observed outcomes. The pure rational choice variant of methodological individualism, in assuming outcomes from efficient calculations, runs the risk of committing the post hoc ergo propter hoc fallacy Institutions are interpreted “as having arisen because of the functions they must have served, when they in fact appeared for purely adventitious reasons.”26 As Keohane notes, a sound functional analysis should show a causal connection between the functions of an institution and its existence. Similarly, James Caporaso criticizes rational choice for not examining the actual choice process.27
For these reasons, my research tries to inductively ascertain what preferences individuals actually had and what choices they made. I further examine how preferences were aggregated and played out in political bargaining. In doing so, the role of beliefs and norms must be taken into account, in conjunction with the material interests of the individuals in question. Beliefs and norms inform one’s preferences. Hence, in order to ascertain the actual preferences of an actor, that individual’s belief system must be examined. To label particular types of behavior, for example, a noble’s pursuit of honor, as irrational by holding it to a strict cost-benefit logic misses the point. The choice of that individual should be understood by the conceptual context through which that individual’s preferences are formed.28 Weber’s notion of elective affinity is one way of reconciling material interests and belief systems, without claiming the priority of the first and reducing the latter to epiphenomenal status. By using his insight, I try to demonstrate the preferences I ascribe to individuals.29
The focus on individuals’ ideas is important, moreover, because it addresses a problem that is neglected in some accounts of rational bargaining but that should be clearly central to it. Such accounts, although they correlate interests to outcomes, fail to clarify how second-order collective action problems are overcome. Even when new institutions might benefit the members of a potential political coalition, that coalition might not form because of freeriding. The existence of shared beliefs will thus make the link between material preferences and institutional outcomes more plausible.
I will analyze three cases: the rise of the sovereign, territorial state in France, the emergence of the Hanseatic city-league, and the rise of the Italian citystates.30 My analysis of the three cases starts at roughly the end of the tenth century In France, the Capetian Dynasty ascended to the throne. In three centuries it would extend its control from a small royal domain around Paris to an area roughly similar to contemporary France. In those same three centuries, German imperial pretensions came to naught. Both Germany and Italy fragmented into a variety of urban leagues and city-states. It is in this period, then, that feudal lordships, universalist claims by the church, and German imperial designs started to give ground. Instead, sovereign statehood, urban leagues, and city-states made their appearances.
My discussion ends at about the time of the Peace of Westphalia (1648), which formally acknowledged a system of sovereign states.31 This is not to say that the process of eliminating alternatives to states had been completed by then. But it did indicate that the variety in the types of units that existed in the Late Middle Ages was gradually being reduced, until later only a system of states remained.
I chose these cases for several reasons. First, I argue that all three were affected by the rise of trade, my independent variable. During the eleventh, twelfth, and thirteenth centuries, Europe went through a remarkable period of economic growth. This exogenous variable set a process of realignments in motion. Second, the variation in types of units allows me to demonstrate that the success of the state was not a foregone conclusion, as many theories on the rise of the state suggest. Third, the differences in capacities—not necessarily in favor of the state—clarify that the rise of the state was not simply a matter of efficiency of scale, that is, of size. As shown in this study, all these cases were viable contenders with states. Fourth, this variation dispels some alternative explanatory accounts. Emergent capitalism (neo-Marxist theory), dynamic density (neo-Durkheimian arguments), military technology (competitive ability perspectives), and conceptual shifts (neo-Weberian theory) do not suffice as ultimate explanatory variables. Each of the three cases were susceptible to emergent capitalism and increased transactions. All of these also had access to the new military technologies and sets of ideas. Yet they demonstrate different institutional solutions.
In short, by acknowledging this variation in contending alternatives, I can more accurately identify the causes of unit change. Furthermore, this variation allows me to suggest reasons why the other forms were inferior, without simply affirming the consequent.32
Critical to my analysis is the separation of the generative causes of such new institutional arrangements from the causes that led to the demise of some. It is clear from the contemporary state system that states have displaced other forms of organization. The second part of the explanation of unit transformation therefore requires an account of the selective process that favored states over these other types. The full account of this selective process is presented in Part III of this book.
I will argue that selection occurs both by systemic pressure on the less competitive types and by societal choices of actors who switch their allegiances. Systemically, from a top-down perspective, there are three main reasons why states survived and displaced other forms of organization. First, the internal logic of organization of the sovereign state had less deficiencies than its rivals. Sovereign, territorial states were better at rationalizing their economies and mobilizing the resources of their societies.33 Second, state sovereignty proved to be an effective and efficient means of organizing external, interunit behavior. Sovereign states could more easily make credible commitments than their nonsovereign counterparts. Third, sovereign states selected out and delegitimized actors who did not fit a system of territorially demarcated and internally hierarchical authorities. The organizational principles of territorial states and city-leagues were mutually incompatible, exactly because the latter had no specific borders.
Selection, however, also occurs by social choice, from the bottom up. Since sovereignty is an institutional mechanism which creates a focal point for coordinating behavior, members of sovereign societies can achieve gains from long-run, iterative cooperation. If the benefits of long-term cooperation outweigh the benefits of short-run defection, social actors have an incentive to defect from institutions that do not provide for these benefits.34
Institutional mimicry constitutes a second form of social choice.30 For example, once the benefits of sovereign statehood became apparent, the German feudal lords started to behave like territorial rulers. They started to call themselves princes and to perform regalian tasks such as regulating weights and measures, standardizing coinage, and rationalizing justice. Although technically still part of the empire (which would last in name until 1806), these principalities gained all the trappings of sovereign statehood. The Peace of Westphalia formally acknowledged their status and granted them all the rights of state actors, such as signing treaties, exchanging diplomats, and waging war (albeit not against the empire itself).
My key point, therefore, is that institutional structures have particular domestic and international consequences. This is not a comprehensive theory of late medieval and early modern Europe. Readers searching for a detailed account of the Hundred Years War, the defeat of Burgundy, or a clarification of the implications of the new Atlantic trade routes will search in vain. Instead my focus is on the advantages of fixed territorial borders and the benefits and costs of weakly established hierarchy. I do not deny the specific importance of individual traits of kings, weather patterns, or fortunes of war to clarify the rise and fall of specific states. But this is not an argument about specific states. In a sense, this is not even the history of France, Germany, or Italy. Instead it is an argument about the relative merits of particular logics of organization. The historical cases should be seen as examples of a larger set of diverse organizations—France was not the only territorial state, the Hansa was not the only city-league. My account suggests why some actors preferred one form of organization over others and why in the long run only sovereign territoriality survived as a logic of organization.
Charles Tilly suggests, as I do, that any account of change in the type of units must deal with the nature of institutional variation that existed in medieval Europe and the subsequent consolidation of the state. He explicitly recognizes the omission of this in some of his earlier work.36 He also gives one of the few accounts of change that takes internal and external variables into account.
Tilly’s recent model of explanation is extremely powerful.37 The explanation he presents explicitly acknowledges the variety of institutional types that differed from the national state. Indeed, in its broad outline, Tilly’s formula parallels my own. Shifts in the milieu lead to internal crises. Domestic factors produce a certain outcome. Then international systemic factors eventually lead to the dominance of one of the organizational forms. It is precisely the similarity in general perspectives that makes Tilly’s model a valuable benchmark for evaluating my analysis. It will also clarify how my argument differs from a commonly held view that the rise of the sovereign state must be primarily attributed to warfare.
Tilly argues that the creation of new types of units in Europe basically proceeded along two dimensions. The first dimension was coercive intensive and the second was capital intensive. “In the coercion intensive mode, rulers squeezed the means of war from their own populations and others they conquered, building massive structures of extraction in the process.”38 This method of resource mobilization occurred in regions that lacked strongly developed towns, such as Brandenburg and Russia. Areas that had well-developed towns, such as Italy and the Netherlands, had to make bargains with capitalists—the capital-intensive mode of resource mobilization.
National states such as England and France combined both coercive- and capital-intensive modes. “In the intermediate capitalized-coercion mode, rulers did some of each, but spent more of their effort than did their capital-intensive neighbors on incorporating capitalists and sources of capital directly into the structures of their states.”39
Gradually, competitive pressure selected out those organizations that used only one of the two methods of resource mobilization. The national state, which utilized both of these methods, proved to be better at waging war.40 The series of wars that permeated Europe from 1500 on thus selected out the city-states and city-leagues. These unit types proved unable to muster the resources that national states could bring to bear.
My account differs from Tilly’s on several issues. First, Tilly explains the variation between types of organization by differential responses to the functional demand of waging war. My account is based on the impact of economic change and subsequent politics of coalitional bargaining. I see the economic transformation of medieval Europe as the primary independent variable which made new political coalitions possible. These coalitions embarked on different institutional paths prior to the military revolution. Second, Tilly explains selection among rival modes of organization by their ability to wage war. From my perspective, the ability to wage war is itself determined by the efficacy of particular institutional arrangements. For example, the ability of a particular mode of organization to raise revenue and prevent freeriding will affect its war-making capacity. That is, the ability to wage war is an intervening variable, itself determined by institutional makeup. Third, Tilly’s focus is different from mine. Tilly is primarily concerned with the state’s ability to raise revenue from its society for war. Given this focus, there is no doubt that the military revolution led to an exponential increase in state demands on society following the Middle Ages. By contrast, I am interested in the question of how sovereign territoriality became one of the logics of organization that emerged in the Middle Ages. Thus, although I am also concerned with the capacity of government to extract resources, I focus more specifically on the emergence of sovereign territoriality in contrast to nonterritorial and nonsovereign types of organization.
Given Tilly’s focus and presuppositions, it is logical that he sees the variety of coercive-intensive or capital-intensive paths emerge particularly after 1400. This corresponds with major breakthroughs in military technology—artillery, modern fortifications, massed infantry—after 1400 or even later by some accounts. This environmental change forced governments to seek more revenue from their populations in order to guarantee the security of their respective polities.
However, some historians, such as Joseph Strayer, place the origins of the state at an earlier date. Tilly suggests that if Strayer is right, then his own argument is flawed.41 It is not merely a question of the exact dating of the origins of the state. As I discuss in Chapter 5—and this is the critical point if Strayer’s chronology is correct—centralization cannot be explained by war making alone. First, weak kings were not the logical providers of protection. Second, given that there was no increase in competitive pressure, since the military revolution had not yet occurred, central governments must have provided other inducements to powerful groups.
The French Capetian kings started from a very weak material basis. Powerful lords, such as the duke of Normandy, wielded more military might. If war making were the primary explanatory factor, then such lords should have been more successful than the king. That is, if state making were simply a bargain for protection, there is no reason to assume that the king would be the logical provider of protection, or that a centralized kingdom should be the result. The question thus becomes, how could weak kings bargain their way to greater strength?
This suggests that a focus on social alliances will be more fruitful. To argue that national states centralized and accumulated coercive force by combining elements of the coercive- and capital-intensive path still leaves unanswered how this occurred. A focus on social alliances will, for example, clarify why rulers started to protect property rights42 and will thus suggest why centralized rule was more attractive than feudal rule.43 This is of course Douglass North’s point. Although Tilly views North’s theory as a direct counterargument to his own,44 I believe that a closer analysis of microlevel incentives and political bargains, such as North’s or my own, actually fills some gaps in Tilly’s argument.
I agree that the relative strength of towns is the first step in explaining the variation between states (city-states, national states, etc.). However, the relative power of the towns alone cannot explain everything. One needs an account of how towns bargained or allied with rulers. For example, one could not predict that many German towns would become independent solely by taking their relative power into account. The course of German urban independence was equally determined by Frederick II’s choice to ally with the lords against the towns. The specific strategy pursued by political coalitions remains, therefore, important.
Furthermore, if towns are critical, as Tilly suggests—and I agree with him— then their presence needs to be explained. I argue that the emergence of many of the new towns in northern Europe, and the growth of the Italian towns, can be precisely attributed to the economic dynamics of the Late Middle Ages. Hence, this reinforces my point that the important change in milieu is the change in the economic system, which preceded changing methods of warfare.
In short, Tilly explains the variation in unit type by different responses to the functional necessity of waging war. These different responses can in turn be explained by regional variation in the strength of towns. I argue, by contrast, that variation can be explained by different responses to a changing economic environment. The particular different responses can in turn be explained by the nature of internal political coalitions.
The second difference with Tilly’s model is his account of international selection. For Tilly, selection is rooted in the unit’s security performance. From my perspective, the critical variable is institutional structure. This in turn explains the ability to wage war. Whether or not a particular type of organization will survive depends on its ability to prevent freeriding, its credibility to commit in international treaties, compatibility with other types of organization, and the benefits that it provides to its subjects (in order to prevent defection).
Sometimes Tilly’s account appears to explain the success of national states simply by the amount of resources under control, not by the particular mix of coercive-intensive or capital-intensive means. The national state centralized and accumulated more coercive means than its rivals.45 The physical size of the state makes the difference. Tilly suggests, for example, that “large states” replaced other types.46 The Italian city-states fell to the French and the Netherlands to England, not because they were capital intensive but because they were not large enough to muster sufficient resources.
I locate competitive ability in the nature of the unit’s organizational structure.47 One cannot argue that capital-intensive states lost out to national states because the latter could use coercion and capital-intensive means to mobilize resources. One could very well argue that the more capital-intensive units were more suited to purchase resources for war. In fact, states that rely more on coercive means to extract resources might be at a disadvantage.48
My alternative explanation of variation and selection has several benefits over Tilly’s perspective and similar accounts that focus predominantly on warfare. If, as Tilly proposes, war making is the crucial selecting mechanism, then it is difficult to explain why so many small states survived. Why, for example, were the minuscule German states recognized as treaty signees equal to France or England, even in the middle of the nineteenth century? As Tilly himself notes, he regards some small entities such as Monaco or San Marino today as states, because other states treat them as such.49 In other words, it is the empowerment by other states of such entities that allows them to continue to operate in world affairs rather than their ability to wield force. Furthermore, if the size of the national state is actually the explanation, how can smaller states replace a larger? Why did the Netherlands replace Spain? In my account this can be explained by looking at possible institutionally superior arrangements.50
In summary, I basically disagree with Tilly on two dimensions. First, I explain institutional variation by the specific nature of social coalitions following economic change rather than by different responses to the functional prerequisite of war making. Second, I argue that whereas war is an important factor, the ability to wage war itself needs to be explained by institutional analysis. I would argue that on these dimensions, however, my theory complements Tilly’s analysis by specifying the particular bargains underlying the state-making process.