Making the State
The destructive capacity of war is self-evident. Less so is the manner in which war, or more accurately, the process of going to war, can be constructive. War is rejuvenating. The demands of war create opportunities for innovation and adaptation. Wars help build the institutional basis of modern states by requiring a degree of organization and efficiency that only new political structures could provide; they are the great stimulus for state building.1 States, in a sense, are by-products of rulers' efforts to acquire the means of war; war is inherently an organizing phenomenon from which the
state derives its administrative machinery. According to Hintze,
1. Huntington, Political Order in Changing Societies.
101
102    Blood and Debt
all state organization is principally military in nature. The shape and size of the state may even be seen as deriving from the managerial potential and limits of military technology.2 So, for example, the advance of bureaucratic forms may be in part a result of increasing demands for administrative efficiency generated by the needs of growing armed forces and the escalating costs of waging war.3
The notion that war supports the institutional development of the state is widely accepted in political sociology.4 This is not a recent discovery, but reflects the importance assigned to war by Weber and Hintze.5 Wars help build the institutional basis of the modern state by requiring a degree of organization and efficiency that only new political structures could provide. Charles Tilly has best summarized this process with his statement "States make wars and wars make states."
This at least is the scholarly consensus on the European experience. On that continent, wars served as a crucial causal mechanism behind the growth of the state. The rise of the modern European state may be traced to the military revolution of the sixteenth and seventeenth centuries.6 During this period, three critical organizational developments changed the nature of military struggle: control over the means of violence shifted from private to public control; the size of armies increased dramatically; and their composition became less varied and more based on a specific national identity.7
War made the territorial consolidation of a state more feasible and more imperative. Only those states that could wield great armies and guarantee control over their own territories could play the great game. Only those
2. Bean, "War and the Birth of the Nation State."
3. M. S. Anderson, War and Society in Europe.
4. Andreski, Military Organization and Society; Finer, "State and Nation Building in Europe"; Tilly, "Reflections"; Downing, The Military Revolution and Political Change; Porter, War and the Rise of the State; Ertman, Birth of the Leviathan.
5. Weber, General Economic History; Hintze, "Military Organization."
6. Outside Europe, the American Civil War both provided the major impetus for state expansion and allowed the industrial North to reshape the antebellum agenda (Bensel, Yankee Leviathan). Karsten describes the links between the rationalization of the armed services in the United States and similar organized efforts in other government sectors ("Militarization and Rationalization"). Bendix suggests that the effective rule of early Japanese shogunates may originate in the aristocracy's military experience (Kings or People).
7. Finer, "State and Nation Building in Europe"; Porter, War and the Rise of the State; Roberts, Essays in Swedish History; Parker, Military Revolution; Duffy, introduction to Duffy, Military Revolution and the State; Ralston, Importing the European Army; Kaiser, Politics and War.
Making the State 103
states able to impose that central control could survive the military revolution. Countries unable to do so—Poland being one example—disappeared. The decline in the number of European states after the fifteenth century (from fifteen hundred to twenty-five by 1900) is an obvious indicator of the centralization of power wrought by military conflict. Wars pushed power toward the center.8 War provided both the incentive and the means with which the central power was able to dominate. Peter Paret explains that "military force performed the essential task of defeating particularistic rivals to the crown, lending authority to the expanding process of government."9 Whether in the France of Louis XIII, seventeenth-century Prussia, or Restoration England, violence was used to impose the rule of the center. The means for this violence were provided by war.
The key to the relationship between war and state making in western Europe is what Finer calls the "extraction-coercion" cycle.10 We begin with the obvious fact that wars require capital: by the sixteenth century, combat became so expensive that the mobilization of an entire country was required. Professional armies clearly outperformed any rivals, but these needed "ample and continuous amounts of money."11 These changes causally linked military and political development. On the one hand, states penetrated their societies in increasingly complex forms in order to obtain resources. The organizational innovations that occurred during wartime did not disappear with peace, but often left an infrastructural residue that Ar-dant calls the "physiology" of the state.12 On the other hand, the new form of the post-Westphalian state was particularly well suited to the organizational task of managing this penetration and channeling the resources thus obtained into "productive" violence directed at some external enemy. Thus, wars both built and were an expression of political power.
Taxation is the best measure of effective political authority and institutional development, both representing and augmenting the strength of the state as measured by the capacity to enforce centralized rule on a territory and its population.13 Taxes partly determine the very size of states' institutions and shape relationships between these and society; they help mold the
8. Howard, Causes of War.
9. Understanding War, 41.
10. Finer, "State and Nation Building in Europe."
11. Howard, War in European History, 37.
12. Ardant, "Financial Policy and Economic Infrastructure."
13. Peacock and Wiseman, The Growth of Public Expenditure in the U.K.; Organski and Kugler, The War Ledger.
104    Blood and Debt
eventual form of the state.14 War is widely perceived as increasing the capacity of a state to tax its population.15 Combat simultaneously generates greater need for resources and temporary declines in the state's social constraints; it also provides a focus around which the state's organizational capacity may improve. Finally, armies raised for war might also serve as a means with which to collect resources.
The evidence for the positive link between war and the rise of taxes in early modern Europe is exhaustive.16 The pattern is also evident in the United States.17 In all these cases, not only does state revenue increase after war; the structure of taxation also changes. For example, wars led both the British state in the eighteenth century and the American in the nineteenth and twentieth to increase both the amount of revenue (which never returned to prebellic levels) and the relative importance of domestic and direct taxes (Fig. 3.1). Military conflicts allow—and force—the state to depend less on the administratively simple, but inelastic, custom taxes and to rely on the more politically challenging, but potentially more lucrative, domestic sources of revenue. The greater bureaucratic complexity required is at the heart of the institutional legacy of war.
Yet how automatic is the relationship between war and increased state strength? Appreciation of historical specificity and structural conditions is vital for the production of truly generalizable models of state development. Only some wars built states, only some states were built by wars. The European experience indicates that warfare in and of itself does not necessarily lead to state making. Until the sixteenth century, several centuries of prior warfare had not produced states in Europe. Rather, as Tilly has emphasized, particular circumstances found in parts of Europe between 1600 and 1800 promoted conflict-led state development.18
14. Tilly, "Reflections on the History of European State-Making"; and Ardant, "Financial Policy and Economic Infrastructure"; Schumpeter, "The Crisis of the Tax State"; Gallo, Taxes and State Power; von Stein, "On Taxation"; Levi, Of Rule and Revenue.
15. Peacock and Wiseman, Growth of Public Expenditure; Mann, Sources of Social Power, vols. 1 and 2; States, War, and Capitalism; Tilly, Coercion, Capital, and European States; Rasler and Thompson, War and Statemaking; J. Campbell, "The State and Fiscal Sociology."
16. Ames and Rapp, "The Birth and Death of Taxes"; Mathias and O'Brien, "Taxation in Britain and France"; Stone, An Imperial State at War; Brewer, The Sinews of Power; Aftalion, "Le financement des guerres."
17. Bensel, Yankee Leviathan; Skowronek, Building a New American State; David and Legler, "Government in the American Economy"; Hooks and McLauchlan, "The Institutional Foundation of Warmaking."
18. Tilly, Coercion, Capital, and European States.
Image not available
Fig. 3.1 United States: Change in revenue and expenditure
106    Blood and Debt
To an extent, this transformation remains a historical black box. We possess myriad references to the rise of the modern state and countless monographic descriptions of the specific historical sequences. Yet political sociology has generally failed to produce a coherent model of how violence was transformed into order. Sociological and comparative accounts of the relationship between war and state building have also not sufficiently emphasized historical order in their analysis. There is a causal ambiguity in Tilly's famous aphorism: Which came first, states or wars?19
I argue in this chapter that wars in and of themselves do not make anything. Rather, they merely provide a potential stimulus for state growth. Wars can only make states if they are preceded by at least a modicum of political organization. Without institutional cohesion, wars will make for chaos and defeat. Wars only provide an opportunity for those political organizations that are able to capitalize on them; they cannot create institutions out of thin air. The consolidation of central authority and the creation of a modicum of a bureaucracy appear to have preceded the state-making stage of war in England, France, and Prussia. The venality of the Spanish bureaucracy and the financial leakage of tax farming in a variety of other countries represented critical obstacles to state development.
It is vital to fully appreciate the social resistance that may be offered to state penetration. The combination of coercion and capital symbolized by the military draft and direct taxation, the defining characteristics of a war-made modern state according to Tilly,20 does not come about simply because a bureaucratic apparatus is in place and wishes it so. The capacity of a state to extract resources will be closely linked to the willingness of the population to accept these burdens. Reluctance on the part of either an economic oligarchy to part with its cash or a populace to move closer to penury may make expansion of taxes simply not worth the effort.21 Thus, state capacity is not an absolute phenomenon, but a relational one. It is not merely a question of strength, but also of the potential of the relevant societies to resist (or welcome) intrusion.
Wars only make states when there already exists some form of union between a politically or militarily dominant institution and a social class that sees it as the best means with which to defend and reproduce its privi-
19. Tallett, in War and Society in Early Modern Europe (citing I. A. A. Thompson's work on Spain), calls war less a stimulant than a test of state strength.
20. Tilly, European Revolutions, 32.
21. On the political costs of pushing too far, a wonderful recent addition is Markoff, The Abolition of Feudalism.
Making the State 107
lege. That is, following Perry Anderson, there has to be a prior agreement that the state will be responsible for collecting and disposing of social surplus.22 European cases demonstrate that the fragmentation of sovereignty, be it through the persistence of local autonomies (Spain), powerful but divided aristocracies (Poland), or direct external control (the Balkans), can and does prevent the solidification of states even when they are surrounded by conflict. In contrast, more successful war-making states established a coalition between central authority and potential aristocratic challengers either through alliance (England) or through coercion (France, Prussia).23
Following the methodological logic of the counterfactual, the failure of Latin American wars to generate similar state-building forces as seen in Europe after the seventeenth century can serve to improve our understanding of the relationship between conflict and institutional development. The findings from this exercise could then be used to explain both geographic and temporal variation outside of Latin America. Not only do these cases expand the relevant sample; they also allow us to discover whether the specific postcolonial conditions under which these states initially developed helped determine their further evolution. In this way at least, the Latin American experience may be much more relevant for contemporary analysis than western Europe in the early modern era.
In this chapter I analyze the contribution of war to the process of centralization and empowerment of the nineteenth-century Latin American state. I identify the critical elements that transform (or fail to transform) the anarchy of war into the imposition of order through monopolized violence and argue that war in Latin America was never able to break the disastrous equilibrium that existed between various powers and social interests. We must trace successful state developments not to war in itself, but to the presence of a united elite, willing—or forced to—accept the loss of individual prerogatives for a (still elite-defined) collective good, and leading a society not already torn asunder by ethnic or racial divisions. Europe has been exceptional not only in the immense amount of organized violence that has characterized continental geopolitics, but also in enjoying preconditions that allowed it to transform this bloodshed into modern political institutions.
This is not to imply that war is the only possible catalyst for state devel-
22. P. Anderson, Lineages of the Absolutist State.
23. Again, by this I do not mean to imply that such coalitions or alliances are either sufficient or necessary, but only that they increase the probability of being able to establish a successful central authority.
108    Blood and Debt
opment or that the Prussian model is the only one available from the European experience. The empirical question of whether war assisted the development of the state in Latin America allows us to isolate theoretically critical aspects of the continent's experience and to better highlight the crucial differences between the historical development of particular regions. In such an exercise we must engage in generalizations about both regions and about individual cases within them. Nevertheless, the general pattern of what I describe should be of interest even for those whose individual cases do not exactly fit the descriptions found in what follows.
War and the Latin American State
What was the effect of war on Latin American institutional political capacity? Measuring state strength, particularly without reliable comparative data, is a risky enterprise. One of the most critical aspects of a modern state is its ability to create and enforce what Frederic Lane and later Charles Tilly have called a "protection racket."24 From this vantage point, the state is often little more than the stereotypical Hollywood goon warning store owners of the potential disasters awaiting them should they fail to purchase his particular brand of insurance. For all its flags, anthems, and other symbolic paraphernalia, the state offers its citizens a simple proposal: in exchange for obedience to a set of laws, state institutions offer protection from both internal and external violence; the Weberian monopoly over legitimate use of violence.25 In the first section that follows, I discuss the creation of that monopoly, after which I turn to the political ability to extract rents.
Providing Protection
In order to maintain its racket, the state has to be able to defend preset frontiers and ensure obedience to its laws within those frontiers. It has to defend its right to exist and to demand internal recognition of its domination internally. The internal element involves two aspects (and they may be related, but it is important to keep them separate): (a) Only state officials
24. Tilly, "War Making and State Making as Organized Crime."
25. "[The state] is thus a compulsory organization with a territorial basis. . . . the use of force is regarded as legitimate only so far as it is either permitted by the state or prescribed by it" (M. Weber, Economy and Society, 56).
Making the State log
may have access to means of violence; and (b) the central state institutions (those claiming national coverage) have priority over any other regional or local competitors. The first is about controlling lawless violence, for example, banditry; as well as the elimination of rival claimants to the national territory, for example, Indians in Latin America or the North American West. The second is about the number of governments inside the territory who are claiming the right or duty to stamp out the violence. We will refer to the first as a process of pacification and to the second as centralization.
There is no precise measure that we may use to absolutely establish the date when centralization or pacification was achieved in Latin America. For the first, we may use the last date of significant regional revolts. For the second we might consider the effective end of banditry or the elimination of a viable Indian military threat. In some cases, defining the threshold is relatively easy. In Argentina, 1880 marks the last major regional rebellion within the political class and also the final conquest of the southern Indian tribes. Chile is effectively centralized by the early 1830s, and we see no explicitly regionalist rebellions after that date. Pacification is finally accomplished with the final assault on the Mapuches in the early 1880s. As the subsequent history of both of these countries makes clear, neither centralization nor pacification implies the end of social or even political struggle, but rather the establishment of central authority as a final arbiter or goal for a political project.
Paraguay represents something of an outlier in that despite the extreme degree of external violence it has suffered, the domestic architecture of political power was essentially settled in the early years of Dr. Jose Francia's rule. By contrast, Uruguay had to wait almost a century for this denouement, but has enjoyed relative stability since then. Other cases are more difficult to pinpoint. Major regional rebellions are a regular part of Brazilian history through the 1850s, but after that date the specific geographical aspect of uprisings becomes less clear. Regionalism still matters (witness Paulismo), but, again the contest is about who will control the center rather than attempts to redefine national borders. With the victory over the Canudos rebellion, the government effectively established its authority over most of the country.
We are clearly left with some judgment calls that can be debated. In Venezuela, Guzman Blanco established centralized control and the fifty years of Andinos established political peace. In Mexico, Porfirio Diaz both centralized authority and pacified the countryside (often meaning killing those in it). The revolution and its consolidation in the 1920s finished the process.
no    Blood and Debt
Yet recent events have indicated that even the apparently solid pax priiana may not have been as permanent as many would have guessed. Finally, one could argue that Peru, Bolivia, Ecuador, and Colombia have not yet accomplished the basic task of pacification. In two of these countries, guerrilla movements held the government at bay for years. In three, Indian communities oftentimes remain beyond the purview of central control and may represent a challenge to national unity. In all, the capacity of the government to guarantee safety and defend its authority is questionable. Note, moreover, that neither process of pacification nor centralization has been historically linear and each has involved often alternating trends.
Table 3.1 includes two possible proxies for state centralization/pacification. One is the date of the first national census. Such efforts require that government representatives not only have authority to ask sometimes difficult questions, but also can be protected from random violence while performing their jobs. An alternative measure is the development of a communications and transportation infrastructure here indicated by railroad mileage in 1900. If we use either measure as an indicator of infrastruc-tural development we note the gap that existed between even the most developed Latin American countries and the United States. Extensive railroad development does not begin on the continent until the 1880s, while few countries were able to thoroughly count their populations before the end of the century. Not coincidentally, it is also around this time that we note the disappearance of regional or local currencies and the establishment of a monopoly over legal tender by the central government. A final indicator of centralization would push the relevant date further into the twentieth century. Using the "governmental scope" measure in the POLITY II data set,26 we note that the social intrusion of the government is minimal until after World War I and only escalates during the Depression (Fig. 3.2).
Any of these time lines make it difficult to argue for the causal significance of war, as the most important conflicts had occurred at least a decade prior to the key dates. Even if we accept a historical lag in order to account for the possible influence of the Pacific and Triple Alliance Wars their immediate effect was limited. The relationship between even international wars and state building is largely spurious, as these wars did not play an important role in the centralization and pacification of these countries.
The Paraguayan War did provide Mitre and later Domingo Sarmiento with a much stronger instrument with which to crush continued regional
26. Gurr, "Polity II."
Table 3.1   Centralization and pacification
Date of First
Country	Centralization*	Pacification*	National Census**	RR km in 1900***	Relevant War
Argentina	1880	1881	1869	16,767	Triple Alliance
Bolivia	1900a	1952	1900 (1831?)	972	Confederation and Pacific
Brazil	1850	1890s	1872	15,316	Triple Alliance
Chile	1833	1881	1831/1835	4,354	Confederation and Pacific
Colombia	1880sa	1950s	1912 (1825?)	568	Thousand Days
Ecuador	1916	1950s	1950	92	?
Mexico	1880s	1920s	1895	13,585	French Invasion
Paraguay	1820s	1820	1899 (1847?)	240	Triple Alliance
Peru	1895	1940s	1876 (1836?)	1,800	Confederation and Pacific
Uruguay	1903	1900s	1908 (1852?)	1,730	Guerra Grande
Venezuela	1870s	1930s	1873	858	?
United States	1865	1800s	1790	311,160	Civil War
''"Estimates based on federalization of capital, end of major regional revolt, or effective end of banditry and Indian attacks of major centers. **First date is for recognized national census fulfilling basic criteria. Second date is for possible earlier effort with significant limitations. Source: Goyer and Domschke, Handbook of Censuses. * * * Source: Mitchell, International Historical Statistics.
aRegression in 1990s.
Image not available
Fig. 3.2 Governmental scope (source: Ted Robert Gurr, Polity II dataset)
Making the State 113
revolts in Argentina.27 The 1860s saw the last of the montoneras. But the war may have actually encouraged many of these regional revolts. During the five-year period of conflict there were eighty-five rebellions, twenty-seven mutinies, and forty-three military protests!28 The army that returns from Paraguay is much better equipped to suppress internal revolts. Both these presidents also used the army and local garrisons as a means with which to impose their authority on the provinces. Yet one could argue that it was not the war itself but the development of national institutions, of which the army was but one, under Sarmiento and his successors that consolidated rule from Buenos Aires. It was this success rather than the War of the Triple Alliance that killed regional autonomy. These developments were not born from military victories, but from a series of political contracts and defeats within elite circles. The war arguably had even less effect in Brazil. The army shrank and suffered from the very same political divisions and conflicts as the state; it certainly did not impose a solution until much later.
Paraguay's experience is perhaps closest to a typical "unifying" war, but the relevant struggle was not in the 1860s, when the country was left destroyed, but many decades earlier. By 1811, not only had the Paraguayans already defeated an army from Buenos Aires attempting to maintain the capital's control over the province, but even more important, the porteho threat and the possibility of a Brazilian intervention forced the local elite to unite behind a single program and junta. Having established a unitary government with all power centered on Asuncion, Francia was able to dominate the ruling junta and finally make himself dictator in 1816. It is interesting to note that Francia was able to use hatred of a white elite to unite an Indian and mestizo population in support of the central state.
It has often been said that war made Chile, and it is true that the war against the Peruvian-Bolivian Confederation was accompanied by efforts to create a more solid and effective central authority. The first Peruvian war "provided a basis for solidarity and legitimacy, as well as the leadership that spared Chile the political disorder and caudillismo."29 But one could argue that by the time Chile went to war in the 1830s, the Portalian Constitution had already established the mechanisms required to maintain central control. Moreover, it was the assassination of Portales that served to consoli-
27. Although Argentina's central authority really consolidated in 1880 when Roca defeated the Buenos Aires Guardia Nacional led by Carlos Tejedor.
28. Pomer, La guerra del Paraguay, 246.
29. Loveman, Chile, 141.
114     Blood and Debt
date support for the war and the regime.30 If by 1859 Manuel Montt could declare that "political parties have disappeared in Chile,"31 this had more to do with geography and elite interaction than with military developments. The "second" war of the Pacific increased government legitimacy and encouraged economic development. The large army created to defeat Peru was then available for the last conquest of the Indians to the south. Nevertheless, it would be difficult to argue that the 1879-83 war made the Chilean state that much more centralized or pacified.
Overall, infrastructural and political development in the late nineteenth century appears more closely related to the expansion of the primary export economy than to the logistical needs of war (Fig. 3.3). Argentina's railroads, for example, were not designed by a General Staff seeking to accelerate mobilization, but for and by the agricultural export economy. The growth in state capacity or centralization noted as beginning in the late 1880s matches exactly a continentwide expansion in the export of basic commodities. Of course, causal order is ambiguous here. To what extent did the increase in trade fuel state centralization and to what extent was it made possible by the prior political stabilization? The question is far too complex to settle here.32 For our purposes the critical issue is that state growth was linked more to the development of capital and trade than to military exploits and conflicts.33
Even such a cursory overview indicates that precisely in the period of its history when Latin America was most bellicose, the state was still struggling to establish its authority. During an extremely bloody century we can speak of stable regimes in Brazil after 1840, Chile between 1830 and 1891; and Venezuela, Colombia, and Ecuador until the 1840s. Up to the last decade of the nineteenth century, violence was endemic, power was fractured, and authority was fragile in much of the continent.34 Despite a great deal of
30. Collier and Sater, A History of Chile, 1808-1994, 66.
31. Collier and Sater, History of Chile, 118.
32. Isolated evidence would indicate that trade produced the state. In Bolivia, for example, the state was seen as a necessary police officer acting to defend investments and direct the needed infrastructural development (Klein, Bolivia, 148-52; Paz, Historia economica de Bolivia, 111).
33. Nearly simultaneously and across the entire continent, a variety of foreign missions led efforts to discipline national militaries and create more professional forces. These arguably produced more efficient, but much less internationally active, militaries (Nunn, Yesterday's Soldiers; Loveman, For la Patria). The link between these two trends is weak at best. Neither one was an expression of nascent states born out of war, but independent responses to Latin America's emerging role within a global economy and polity.
34. There were other exceptions. Castilla brought a semblance of order to chaotic Peru in the 1850s and Santa Cruz to Bolivia in the 1830s.
Image not available
Fig. 3.3 Latin American exports (source: B. R. Mitchell, International Historical Statistics)
116     Blood and Debt
military conflict, Latin America remained something of a continental frontier; a place where no one had an enduring monopoly on violence.35
The relevant model is much more that of Austria-Hungary than of Prussia.36 Wars seemed to only highlight the intrinsic weakness of the regime and the fragility of any sense of nationhood. Only with the beginning of the "long peace" do we observe the development of the forms of state capacity supposedly associated with military conflict. The apparent paradox is even more puzzling when we recognize that the forms of Latin American states with the most characteristics associated with "modern" political institutions (extensive bureaucracies, closer interaction with larger parts of the population) did not appear until the 1930s and 1940s under corporate and populist guises and were hardly influenced by the kind of geopolitical competition supposedly responsible for European development (see Fig. 3.2).
Collecting Rents
If the capacity to "protect" was weak and apparently not significantly influenced by war, it should not surprise us that Latin American states were not very successful at collecting the rents derived from the racket.
War did have some of the expected results in Latin America. The expenditure patterns of the states for which we have a modicum of information look a great deal like the classic bellicist state (Table 3.2).37 These were countries apparently devoted to war. Expenditures were concentrated on the military and paying the debt derived from war.38 (The overall decrease in military expenditures after the 1870s does reflect a general decline in interstate violence from the late nineteenth century on.) The exceptions to the pattern tend to prove the rule. The decline in Paraguayan expenditures after 1840, for example, actually reflects a change in the manner of account-
35. Borrowing from Lane's formulation of the frontier as cited in Duncan Baretta and Markoff, "Civilization and Barbarism," 590.
36. With thanks to Steve Topik.
37. I have avoided the use of formal statistical methods for two reasons. First, given the vagaries in the data, formal cross-national comparisons would be deceptive. Second, qualifying any individual year as peaceful or at war would be an extremely subjective process and involve distinctions beyond the purview of this book.
38. Conflict did not necessarily mean large organized armed forces (see Chapter 5). Colombia had a very violent nineteenth century, yet its army was down to eight hundred men in 1854 and 511 in 1858. The army was in fact almost nothing more than a palace guard well into the twentieth century (McGreevey, An Economic History of Colombia, 1845-1930, 87). In the 1830s, the entire Venezuelan army consisted of one thousand men (Bethell, Cambridge History, vol. 3, 520).
Making the State 117
ing adopted in order to disguise an increase in the attention paid by the state to military development.39 The low Uruguayan numbers come almost fifty years after formal independence; in those fifty years (1830-80) the state did little but fight internal and external enemies.
Among winners, such as Argentina and Brazil in the 1870s, war led to an increase in the size of government. Wars also provided the expected economic stimulus. Military procurement and the slightly higher war tariffs encouraged domestic industrial development in Brazil during the War of the Triple Alliance.40 The need to supply the Chilean expeditionary forces in Peru during the War of the Pacific increased the demand for domestic production of basic products such as textiles and foodstuffs. More factories were founded in Chile between 1880 and 1889 than had existed prior to the war.41 Recruitment and increased demand also lowered unemployment.42
But at least as measured by taxation, the Latin American states did not penetrate or extract from their societies in the expected manner. Comparisons of relative extractive capacities in the nineteenth century are difficult, given the lack of comparable national economic data and the questionable use of official exchange figures. The two countries for which we have the most reliable information that can be translated into an international currency are Chile and Brazil. During the entire period in question, neither Latin American state could extract even half the revenue per capita available to the British state, arguably the least rapacious European power at the time, despite the fact that these countries experienced considerable conflict.43 Moreover, both Latin American countries were much more dependent on customs revenues than were the United Kingdom or France during this time period. While such taxes accounted for roughly one-third of British revenues and were marginal for France, in Brazil and Chile they represented at least two-thirds and often more (Fig. 3-4).44 As discussed earlier, the upward trend in tax receipts of the Latin American countries reflects increased connection to the global economy, not a stronger state.
Even if the state grew, and this was not universally true,45 it did not de-
39. Thomas Whigham, private communication with the author.
40. Bethell, Cambridge History, vol. 3, 768.
41. Loveman, Chile, 169; Zeitlin, The Civil Wars in Chile, 77-78.
42. Cariola and Sunkel, Un siglo de historia economica de Chile.
43. Flora, State, Economy, and Society in Western Europe; Mitchell, International Historical Statistics; Direccion de Contabilidad (Chile), Resumen de la hacienda publica de Chile; Buescu, Historia administrativa do Brasil.
44. Cariola and Sunkel, Un siglo de historia economica de Chile.
45. No matter whether at war or peace, Colombia had one of the lowest levels of government expenditure per capita in Latin America (Tovar Pinzon, "La lenta ruptura con el pasado
118     Blood and Debt
velop the fiscal musculature associated with the warring state. The stimulus of war did not produce the dramatic increase in the institutional complexity of extraction associated with the theoretical model. Despite rises in expenditures, revenues lagged far behind. As in the European cases, war produced immediate deficits, but with one prominent exception, the Latin American states did not respond to these with increased extractions, at least not in the form of domestic taxes. Customs and royalties from the export of primary goods remained the mainstay of the most Latin American states (Table 3.3).
The internal and external wars experienced by the Brazilian Empire did not elicit the kind of fiscal intrusion seen in western Europe or even the United States.46 Overall, taxes on wealth and production contributed less than 4 percent of ordinary revenue even during the war years.47 In Chile, domestic taxes played an even smaller role. The most significant was a colonial tithe used to support the church, which represented less than 3 percent of all revenues by the 1860s. Early Argentinean governments attempted to impose a capital tax (contribution directa) with rates of 1-8 percent, but this never functioned and was not obeyed.48 Land rents never accounted for more than 3 percent of total receipts.49
Three important exceptions serve to clarify the pattern. Mexico's relatively low dependence on customs receipts reflects continual state dependence on domestic loans and the printing press. The central government was generally incapable of imposing domestic taxes prior to the Porfiriato. By contrast, Bolivia maintained the oppressive colonial Indian tribute and relied on these sources for a large part of its revenues through the mid-nineteenth century. Two aspects of this tax bear notice. First, it was not sensitive to the stimulus provided by external conflict, but rather reflected internal caste divisions. Second, such a tax did not rely on central government infra-structural development, but rather on the retention of special social privileges and localized fiefdoms.
colonial," 115). In 1871 Colombia gathered one-half of Mexico's revenues and one-fifth of those of Chile. During the same period one local authority estimated that the government received only 2 percent of the national product (Deas, "The Fiscal Problems of Nineteenth Century Colombia," 289, 310, 326).
46. During the War of the Triple Alliance a tax was instituted on buildings with a 3 percent surcharge on their value, but this excluded the most valuable rural properties. Moreover, an attempt to establish a tax on industry and the professions was never enforced and the rates remained minimal.
47. Buescu, Evolugäo econömica do Brasil, 89-91; Cavalcanti, "Finances," 326.
48. Alemann, Breve historia de la politica econömica argentina, 61.
49. Rock, Argentina, 1516-1987, 99.
Table 3.2   Combined military and financial expenditures as share of budget
	Argentina	Brazil	Chile	Ecuador	Mexico	Paraguay8	Peru	Uruguay Venezue
1820						0.81		
1821								
1822	0.92				0.96	0.80		
1823	0.93					0.85		
1824	0.85				0.81			
1825					0.95	0.84		
1826	0.98				0.66		0.84	
1827						0.91		
1828		0.90			0.92			
1829	0.90	0.88			0.94	0.89		
1830	0.85	0.85		0.74	0.93			
1831	1.00	0.82			0.92	0.86	0.70	0.83
1832	0.74	0.62		0.57	0.91	0.87		0.79
1833	0.78	0.88	0.57		0.81	0.85		0.84
1834	0.78	0.86	0.58		0.70	0.76		0.80
1835		0.86	0.60		0.90	0.85		0.83
1836		0.85	0.56		0.80			0.41
1837		0.86	0.51		0.93	0.92		0.65
1838		0.87	0.48			0.94		0.73
1839		0.88	0.47	0.58	0.86	0.89		0.68
1840	0.98	0.84	0.41		0.88	0.89		0.53
1841	0.96	0.86	0.34		0.85			0.64
1842	1.00	0.83	0.83		0.95			0.72
1843	1.00	0.82	0.83		0.94			0.57
1844	0.98	0.81	0.73		0.93	0.43		0.73
Table 3.2   Combined military and financial expenditures as share of budget {cont'd)
	Argentina	Brazil	Chile	Ecuador	Mexico	Paraguay3	Peru	Uruguay	Venezuela
1845	0.98	0.79	0.72		0.98				0.69
1846	0.93	0.78	0.70	0.77	0.93	0.81	0.51		0.63
1847	0.94	0.78	0.72				0.58		0.66
1848	0.95	0.79	0.71		0.89		0.66		0.84
1849	0.93	0.77	0.71		0.90		0.74		0.72
1850	0.95	0.78	0.70		0.40		0.75		0.76
1851		0.81	0.77		0.55		0.76		0.83
1852		0.77	0.73	0.74	0.86		0.82		0.83
1853		0.76	0.66	0.60	0.85	0.56	0.83		0.82
1854		0.74	0.60	0.62	0.73		0.85		0.35
1855		0.72	0.53	0.61	0.84		0.86		0.32
1856		0.74	0.55	0.62		0.27	0.87		
1857		0.74	0.57	0.60			0.88		
1858		0.70	0.44			0.40	0.84		0.83
1859		0.70	0.68				0.79		
1860		0.68	0.67		0.57	0.32	0.77		0.74
1861		0.71	0.66	0.80	0.75		0.74		
1862		0.72	0.64	0.62			0.80		
1863		0.72	0.67	0.69			0.73		
1864	0.80	0.73	0.70	0.62		0.72			
1865	0.88	0.84	0.71	0.57		0.71		0.94	
1866	0.73	0.83	0.79	0.66			0.69		0.79
1867	0.72	0.87	0.77	0.58	0.85		0.60		
1868	0.64	0.86	0.67		0.87		0.47		
1869	0.66	0.84	0.70	0.54	0.65		0.57		
1870	0.71	0.72	0.64	0.38	0.55		0.57		
1871	0.63	0.69	0.57		0.55		0.57		0.93
1872	0.88	0.69	0.58	0.51	0.51	0.58		
1873	0.82	0.67	0.57		0.51	0.53		0.60
1874	0.80	0.67	0.52		0.49	0.57		0.82
1875	0.54	0.65	0.54	0.61	0.70	0.74		0.67
1876	0.51	0.62	0.60		0.76	0.78		
1877	0.61	0.52	0.67		0.63	0.57		
1878	0.58	0.43	0.64	0.28	0.69	0.61		0.77
1879	0.54	0.57	0.78		0.82	0.53		
1880	0.57	0.62	0.68					0.69
1881	0.49	0.62	0.76					
1882	0.56	0.61	0.78					
1883		0.58	0.74					0.45
1884		0.57	0.68	0.43			0.35	0.29
1885		0.61	0.64	0.25			0.35	0.46
1886		0.59	0.72	0.41			0.32	
1887			0.68	0.50			0.31	
1888			0.44	0.47			0.32	
1889			0.41	0.35			0.31	
1890			0.37	0.40				
aFor Paraguay, the figures are military expenditures alone. Apparent drop after 1840 reflects changes in government accounting that appear to disguise continued military expenditure.
Source: Cortes Conde, Dinero deuda y crisis; Bürgin, The Economic Aspects of Argentine Federalism, 1820-1852; Halperin-Donghi, Guerra y finanzas en los origines del estado argentino; Mercea Buescu, Organizaqäo e administraqäo do Ministerio da Fazenda no Imperio; Direccion de Contabilidad, Resumen de la hacienda publica de Chile, 1833-1914; Rodriguez, The Search for Public Policy; Aguilar, Los prespuestos mexicanos; Arbulu, Politica econömico-financiera y la formaciön del estado; Vera Blinn Reber, unpublished tables; Repüblica de Uruguay, Anuario Estadistico; Tomas Enrique Carillo Batalla, Historia de las finanzas püblicas en Venezuela.
Image not available
Fig. 3.4. Reliance on customs (source: Direccion de Contabilidad, Buescu, Flora)
Table 3.3   Customs and royalties as share of ordinary income
Argentina    Bolivia    Brazil    Chile    Colombia    Ecuador8    Mexico    Paraguay    Peru    Uruguay Venezuela
1820 0.18
1821
1822	0.84						0.54	0.02	
1823								0.09	
1824	0.80						0.41		
1825							0.55		
1826							0.46		
1827		0.13					0.42	0.33	
1828			0.50				0.45		
1829	0.85		0.32				0.34		
1830	0.87		0.29			0.51	0.45		0.39
1831	0.83	0.15	0.29					0.01 0.52	0.63
1832	0.82	0.11	0.33		0.30	0.43	0.37	0.02	0.55
1833	0.87	0.12	0.57	0.51			0.42		0.52
1834	0.76		0.49	0.56		0.51	0.49		0.49
1835		0.09	0.59	0.56			0.20	0.01	0.52
1836		0.09	0.73	0.57	0.33	0.57	0.21		0.61
1837		0.10	0.78	0.57		0.57	0.25		0.50
1838			0.81	0.57		0.57	0.25	0.01	0.59
1839		0.07	0.84	0.57		0.57	0.19	0.03	0.65
1840	0.70		0.84	0.62	0.32		0.35	0.06	0.74
1841	0.93	0.07	0.80	0.59			0.25		0.88
1842	0.92	0.07	0.78	0.60			0.17		0.73
1843	0.90	0.06	0.66	0.55			0.22		0.62
1844	0.90	0.06	0.67	0.51			0.23		0.68
Table 3.3   Customs and royalties as share of ordinary income (cont'd)
	Argentina	Bolivia	Brazil	Chile	Colombia	Ecuador8	Mexico	Paraguay	Peru Uruguay	Venezuela
1845	0.89	0.09	0.67	0.53						0.67
1846	0.69		0.64	0.56	0.21	0.54		0.05	0.35	0.69
1847	0.84	0.07	0.65	0.53		0.60			0.40	0.57
1848	0.90	0.07	0.76	0.49		0.62	0.25		0.44	0.33
1849	0.93	0.09	0.77	0.51			0.35		0.37	0.45
1850	0.93		0.79	0.57	0.47		0.34	0.04	0.53	0.49
1851			0.84	0.57			0.55		0.57	0.57
1852		0.07	0.82	0.57		0.60	0.26		0.74	0.72
1853		0.11	0.80	0.43		0.59	0.45	0.06	0.82	0.66
1854		0.09	0.77	0.60	0.52	0.59	0.50		0.86	0.77
1855		0.10	0.79	0.56		0.59	0.43		0.91	0.74
1856		0.09	0.81	0.50		0.61		0.11	0.91	0.74
1857			0.79	0.58		0.61		0.22	0.95	0.74
1858			0.78	0.51		0.65		0.18	0.94	0.53
1859			0.76	0.41					0.86	0.63
1860			0.75	0.56				0.25	0.94	0.56
1861			0.76	0.49		0.70			0.95	0.62
1862		0.10	0.74	0.43		0.59			0.86	0.60
1863		0.15	0.73	0.47	0.32	0.52			0.56	
1864	0.81		0.78	0.36		0.50			0.62	0.57
1865	0.92		0.76	0.23	0.21	0.50			0.74	0.74
1866	1.00	0.05	0.75	0.18		0.64			0.87	0.75
1867	0.86	0.06	0.72	0.28	0.40	0.62	0.53		0.70	0.80
1868	0.95	0.04	0.74	0.47		0.77	0.53		0.72	0.89
1869	0.97	0.05	0.74	0.45		0.73	0.48		0.47	0.99
1870	0.94		0.70	0.34		0.73	0.53		0.75	
1871	1.00	0.12	0.73	0.45	0.47	0.64	0.45		0.96	
1872	0.83		0.72	0.53	0.45	0.53	0.45		0.62	0.76
1873	0.83	0.35	0.71	0.33	0.52	0.66	0.48	0.76		0.85
1874	0.80		0.70	0.50	0.59	0.61	0.47	0.94		0.79
1875	0.72		0.69	0.38	0.69	0.48	0.63	0.72		0.80
1876	0.89		0.70	0.40	0.69	0.67	0.57	0.73		0.74
1877	0.96		0.66	0.36	0.62	0.53	0.44	0.85		0.57
1878	0.82		0.67	0.37	0.62	0.64	0.36	0.90		
1879	0.72		0.69	0.25	0.62	0.49	0.36	0.85		0.75
1880	0.79	0.42	0.68	0.34	0.61	0.72				0.71
1881	0.89	0.31	0.70	0.58	0.61					0.70
1882	0.78	0.26	0.69	0.70	0.68	0.75				0.65
1883			0.70	0.65	0.63	0.70			0.53	0.72
1884			0.67	0.67	0.68	0.68			0.57	0.70
1885			0.67	0.62	0.60	0.57			0.50	0.63
1886			0.68	0.39	0.48	0.62			0.57	0.69
1887				0.44	0.47	0.74			0.60	0.73
1888				0.71		0.68			0.59	0.70
1889				0.66	0.52	0.61			0.61	0.74
1890				0.65		0.80			0.55	0.76
aEcuador figures are for indirect taxes largely consisting of customs.
Source: Cortes Conde, Dinero deuda y crisis; Bürgin, The Economic Aspects of Argentine Federalism, 1820-1852; Halperin-Donghi, Guerra y finanzas en los origines del estado argentino; Erwin P. Grieshaber, Survival of Indian Communities in Nineteenth-Century Bolivia; Buescu, Organi-zagäo e administragäo do Ministerio da Fazenda no Imperio; Direccion de Contabilidad, Resumen de la hacienda publica de Chile, 1833-1914; D. A.N.E., Boletin Mensual de Estadistica; Jose Antonio Ocampo and Santiago Montenegro, Crisis Mundial, Protecciön e Industrializaciön; Rodriguez, The Search for Public Policy; Carmagnani, "Finanzas y estado en Mexico, 1820-1880"; Arbulu, Politica economico-financiera y la formacion del estado; Vera Blinn Reber, unpublished tables; Repüblica de Uruguay, Anuario Estadistico; Raman Veloz, Economia y finanzas de Venezuela; Rafael Cartay, Historia econömica de Venezuela.
126     Blood and Debt
The third exception is the most interesting. Because of its location and the almost continual threats to its existence as well as the ideological proclivities of the dictator Francia, Paraguay could not rely on external financing for much of its early history. That is, unlike other Latin American countries, it could not count on either customs or loans to balance its books, but had to live on its own resources.50 The state financed itself through sales to soldiers and the populace as well as through confiscation of property. The structure of Paraguayan finances appeared to change immediately after Francia's death as the importance of customs increased to nearly half of state revenues. Unlike most Latin American countries, however, Paraguay ran a consistent trade surplus during this period.51 The state remained largely in charge of external trade (mostly of yerba mate) either through its own administration or through the patronage network of the Lopezes.52 Direct sales continued to play an important role as well. What is most significant is that Paraguay depended on foreign trade (directly or indirectly) for only 40 percent of its government revenue.53 Paraguay also did not rely on debt, that other great source of international financing; when the War of the Triple Alliance began, it did not have any external debt.54 Given the chaos found in Paraguay during the War of the Triple Alliance, there are no records of how Lopez financed it. It would appear, however, that he paid for the war through the complete mobilization of the country.55 How much of this was voluntary patriotism and how much it reflected the reach of the state is impossible to tell. The relevant point, however, remains the same: unlike other Latin American countries, Paraguay autofinanced its war.
Even in these cases, however, wars did not produce the expected institutional transformation in state civil relations or the structure of the political apparatus. The fiscal infrastructure was established independent of interstate conflict and remained relatively constant in peace and war. In general, Latin American states were not able to escape the "cycle of levies, bankruptcies and mutinies" that characterized warring states prior to the revenue revolution of the late seventeenth century;56 they were not "built" by war.
50. Pastore, "Trade Contraction and Economic Decline," 539.
51. Rivarola and Bautista, Historia monetaria de Paraguay, 104.
52. Whigham, The Politics of River Trade, 69-70.
53. Pastore, "State-Led Industrialization," 304.
54. Rivarola and Bautista, Historia monetaria de Paraguay, 119.
55. Williams, Rise and Fall, 217-21; Pastore, "State-Led Industrialization," 306, 318.
56. Kaiser, Politics and War, 35.
Making the State 127
Blood and Debt
What were the effects of the wars of nineteenth-century Latin America on the political capacity of the state? They produced blood and debt and not much more. If stability requires acts of force, authoritarian impositions, exercises in power justified by internal and external danger, why did the Latin American violence of the nineteenth century not produce a coherent state? What explains the limited significance of war on state development in Latin America? Certainly the eighteenth and nineteenth centuries saw enough organized violence that a concomitant evolution in state capacity could be expected. If the relationship between organized violence and institutional development were automatic, we might expect Latin America to follow the European pattern.
The easy availability of external financing allowed the state the luxury of not coming into conflict with those social sectors that possessed the required resources. In the 1820s and from the 1870s through the 1890s, loans were relatively easy to obtain. Increasingly throughout the nineteenth century, almost all the Latin American economies became integrated into a global economy through the export of a mineral or agricultural commodity. In any case, whenever the state did try to extract greater domestic resources, it was universally defeated. The European pattern includes a basic organizational capacity that was missing in Latin America. The wars occurred too soon after independence and were fought by countries not capable or responding in the pattern described by Tilly and others.
In the following sections I discuss the effect of the form and timing of the warfare that predominated on the continent. I then turn to focus more on the societies in which they were fought.
Wrong Kind of Wars?
It is important to first distinguish between the different kinds of wars and their respective state-building effects. For example, Latin America has not experienced one of the most significant forms of warfare producing centralized authority. In many of the European cases, war contributes to state formation not as an unintended by-product, but as a direct result of conquest.57
57. The sixteenth-century Conquest did produce an elaborate political institution and fostered separate identities (both between it and other imperial claimants and within it, between ruling whites and dominated Indians and blacks).
128     Blood and Debt
The two most obvious examples here are the Italian and German cases. These unifications essentially involved not a central authority systematically expanding control over provinces, but a regional contender for supremacy defeating all other claimants. Prussia and Piedmont essentially conquered Germany and Italy. The same could also be said for England, Ile-de-France, Castille, or Muscovy.58 One could say the same of the Northern victory in the U.S. Civil War.
The combination of colonial heritage and the particular military experience of South America produced a very different variant of state making than that seen in Europe, where states were built from the inside out; a region or province would carve its dominance over others. The state was built at the same time as the territory was acquired. In Latin America, the fight was about assuming control over what was left of a patrimonial state (the descendant of the crown) even if it did not exercise meaningful authority over large parts of the formally defined nation.59 Wars, therefore, were counterproductive for state building. They were either internal squabbles that left domestic territory pillaged or fights between political lightweights that did not produce the necessary stimulus for organizational and institutional evolution. Unlike in Europe, the military did not conquer territory in order to make a state, but had to impose order over a fractious set of local interests, each irrevocably married to the other. Wars were likely to remain within the family, with the particular destruction and divisiveness that civil struggles tend to leave behind.
Civil wars in Latin America were at times defined territorially, but more commonly involved competing claims to central power. These conflicts were rarely about delimiting internal or external frontiers, but instead about deciding who controlled the already defined national territory. Unlike in Europe, we have few examples—two being the creation of Uruguay and the War of the Peruvian-Bolivian Confederation—where international war combined with the process of regional conquest. The principle of uti possidetis (right of possession) consecrated colonial boundaries. This disallowed the kind of Darwinian geopolitics that arguably fueled state development in Europe. There are no Latin American equivalents to Sadowa or Sedan. The closest equivalent to the unification-through-conquest model would be Bue-
58. This is not to deny how non-Conquest wars may have contributed to the capacity of a region to establish itself as first among equals. Again, Prussia, Piedmont, and England were made the powerful regions they were partly through the experience of war. It is this process rather than explicit conquest that we need to analyze.
59. Morse, "Heritage of Latin America," 162.
Making the State 129
nos Aires and the half-century-long struggle for mastery over the confederation. But even here one could argue that the provinces conquered the capital as much as vice versa. Moreover, it is arguable to what extent the final union was a product of war. Urquiza was not vanquished at Pavon and the union of 1861-62 was really the result of elite pacts arrived at politically, not militarily.
Most commonly, wars resulted from attempts of states to consolidate their authority. The final set of conflicts between Liberals and Conservatives in Mexico, for example, is partly a product of attempts by the central government to finance itself through the appropriation of church property. The first war between Chile and Peru and Bolivia was in part a response to the efforts of Santa Cruz to consolidate his rule. A fear of the influence of Paraguay on still semiautonomous regions helps explain some of the unity of the Triple Alliance.
Interestingly, given the relative low level of technical capacity available to the combatants, the Latin American wars were also extremely destructive— arguably more so than most European wars between Westphalia and Sarajevo. The ruinous consequences of war were often not limited or far removed from population centers, but involved the destruction of one side. Losing a war was often disastrous for political order. Unlike in Europe, where the victor's rule would succeed that of the loser, the pattern in Latin America was closer to the creation of a vacuum. In the late 18 80s, for example, the Peruvian countryside was not accessible to the state.60 The Paraguayan case is even more extreme.
The historical record certainly supports analytical emphasis on war type. The two wars of the Pacific, the War of the Triple Alliance, and the Mexican-French War of 1862—in part an internal conflict that became a nationalist one—all appear to have played at least a limited role in the subsequent consolidation of the winning nation-states (but still smaller than might be expected). The much more common Conservative-Liberal conflicts of Mexico and Colombia, the regionalist rebellions of Brazil and Argentina, and the caudillo wars of Peru did little but destroy. Losing a war was universally a disaster.
Looking back at the critical contributions made by wars to state making in Europe, we find that Latin American conflicts did not generally embody the three critical characteristics of wars associated with "military revolution." First, they often did not accompany a shift from private to public
60. Mallon, Defense of Community, 102-3.
130    Blood and Debt
control of violence. In many cases, the chaos that followed them meant that powerful private actors remained in de facto control of their respective lands. Moreover, as wars often were about establishing claims to power, they were perceived as private battles over public goods. Second, armies generally remained small and logistics severely constrained. None of the Latin American conflicts, with the exception of the Paraguayan experience in the 1860s, were "total wars" requiring the militarization of every aspect of social and economic life. Third, the vast majority of the wars did not emphasize or support the development of a national identity, but consisted of struggles in which either such questions were irrelevant or the very definition of that identity was at stake.
Artificial Wealth
For the "coercion-extraction cycle" to begin, the relevant states must not have alternative sources of financing, while the domestic economy must be capable of sustaining the new fiscal and bureaucratic growth. Conflict-induced extraction will only occur if easier options are not available. Even then, the relevant societies might not be able to produce enough surplus to make the effort productive. Thus, for example, the availability of Latin American silver and the willingness of bankers to risk massive sums freed the Spanish Habsburgs from imposing greater fiscal control over their provinces as a means to pay for their wars. Conversely, the relative scarcity of such external supports drove the expansion of the early English state.
It is not clear how much even the most voracious state could have extracted from such extremely poor societies. The independence wars left little base on which to build a state and destroyed much of the economy on which it might have relied. The Mexican economy experienced a significant decline after independence.61 During the first forty years of the century, Bolivia saw the decapitalization of its mining industry—in 1840 there were ten thousand abandoned mines—and the depopulation of its cities. Venezuela was "left a wasteland" by a war that was "cruel, destructive, and total," with armies regularly destroying the property of enemies and paying their soldiers with
61. Per capita income declined from 35-40 to 25-30 pesos between 1810 and 1820 and did not attain colonial levels until the Porflriato of the last quarter century (Bethell, Cambridge History: Independence, 91). One indication of the fall is Mexico's changing position relative to the U.S. economy. By John Coatsworth's estimates ("Obstacles to Growth in Nineteenth Century Mexico," 82-83), Mexico had 44 percent of the U.S. per capita income in 1800, but only 13 percent by 1910, with most of the comparative decline coming prior to 1845.
Making the State 131
plunder.62 The Peruvian economy was crippled by the need to support loyal armies throughout the continent.63 The situation was made even worse by the collapse of the continental "customs union" maintained by Spanish mercantilism.64 Material goods were not the only thing destroyed by the wars of independence. An entire system of economic, juridical, and social relationships was also their victim.65 For most, the wars brought misery, hardship, and penury.
This did not change during most of the period in question. The Colombian economy, for example, was small and undeveloped, while difficult transport constrained the growth of taxable market exchange.66 In general, squeezing the rich did not yield very much because even this social group had relatively small amounts of capital available.67 In the late 1830s, Ecuador's exports totaled the equivalent of two hundred thousand pounds sterling.68 Even a thoroughly rapacious state would have produced per capita extractions much lower than those in Europe. Those states that did impose a direct tax on their populations could expect little. Even with an onerous Indian tribute, for example, Bolivia's government revenues per capita in the 1840s, following a war against Chile, were one-fortieth those of Britain.69
This poverty made it extremely difficult to use excise taxes as a form of income. Aside from relatively few commodities, large parts of the population did not consume very much that could be easily taxed. In any case, the hatred of the colonial alcabala would have made it practically impossible for the newly independent governments to impose such a tax. Equally important, there were few wage laborers whose income could be measured and taxed in any systematic manner. Even the landed oligarchy, while rich in land, in most cases did not possess large amounts of directly extractable resources.70 In several cases, the church did have considerable wealth, but even successful appropriations, as in Mexico, produced disappointing results.
62. Lynch, Spanish-American Revolutions, 202-18; Halperin-Donghi, Aftermath of Revolution, 8.
63. Lynch, Spanish-American Revolutions, 162.
64. Bulmer-Thomas, The Economic History of Latin America Since Independence, 29.
65. Bethell, Cambridge History, vol. 3, 307.
66. Bushnell, The Making of Modern Colombia, 76.
67. Deas, "Fiscal Problems," 314.
68. Bethell, Cambridge History, vol. 3, 511.
69. Dalence, Bosquejo estadistico de Bolivia, 316.
70. Nevertheless, it is possible to build a considerable state apparatus on the basis of a land tax. See Bird, "Land Taxation and Economic Development."
132     Blood and Debt
How then did the Latin American states pay for their wars? A common experience—and one not dissimilar to that of the European cases—was to print money. During its war with Brazil in the 1820s, Argentina resorted to currency emissions that produced a monetary cataclysm: the price of an ounce of gold on the Buenos Aires stock exchange went from 17 pesos in January 1826 to 112 pesos in December 1830.71 During the same war, the amount of available Brazilian currency doubled, and subsequently the reis lost half of its value.72 Even more dramatic was the endless printing of money in Brazil during the War of the Triple Alliance. In 1864 there were 29 million milreis in circulation; by 1870, there were 151 million.73 Between 1859 and 1901, the Uruguayan state issued 342 million pesos, of which 124 million were still outstanding. To give an idea of the degree of printing insanity, the customs house of Montevideo was producing an average of only 10 million pesos per year.74 Thus, the vast majority of Latin American governments resorted to a form of inflation tax in order to pay at least partially for their wars.
This profligate reliance on an inflation tax may represent a possible institutional and administrative legacy of war. Such a tax would require that the central government be able to establish a monopoly on the issue of currency. Thus war would provide an incentive to expand centralized authority as a way to tax through a printing press. It may be worthwhile analyzing the link between war and monetary development in Europe; the Latin American experience also would deserve further analysis. This form of taxation also serves as an indication of the relative power of social groups. Inflation taxes would favor rural over urban populations and exporters over importers. They were generally extremely regressive.75
States also borrowed from both domestic and international sources. In Mexico, the government increasingly relied on the agiotistas, who would provide funds during fiscal emergencies—an almost everyday occurrence as ministers often found empty tills upon assuming office. In exchange for the considerable risk, domestic lenders were often given very favorable terms, with rates in the range of 300-500 percent.76 Since borrowing was constant
71. Bürgin, The Economic Aspects of Argentine Federalism, 69.
72. Cardoso, ed., Mexico en el siglo XIX, 105; Barman, Brazz'/, 140; Nogueira, Ratzes de urna Nagäo, 313.
73. Castro Carreira, História financiera, 743.
74. Acevedo, Notas y apuntes, 457.
75. Rock, Argentina, 1516-1987, 107; Oszlak, La formación del estado argentino, 183; Halperín-Donghi, Guerra y finanzas, 161; Aldo Ferrer, The Argentine Economy, 61.
76. Bažant, História de la deuda exterior de Mexico, 44.
Making the State 133
and loans were continuously rolled over, the debt ballooned to 102 million pesos by 1840, 120 million in the 1850s, and 165 million by 1867.77 The Argentine government also borrowed: by 1840, the debt stood at 36 million pesos, while the total income for that year was 1,710,491 pesos.78 The response to later wars was no different. Between 1865 and 1876, Argentina acquired almost 19 million pounds of debt.79 By 1885, the figure was 26 million.80 Currency conversions make comparisons difficult, but this represented at least four times the revenue of the state during these years. The total public debt by 1888 was more than 60 million sterling, and indebtedness per capita tripled.81 The War of the Triple Alliance brought similar results in Brazil. Real net debt increased from 4.5 million pounds in 1863 to 9.3 million in 1871. Debt as a percentage of exports increased from 58 percent to 82 percent during the same decade.82 By the time an independent Uruguay appeared in 1830, the government had already accumulated a debt of 2 million pesos.83 It continued to spiral: by 1853, when the first attempt was made to organize and systematically manage public finances, it reached 40 million pesos. By 1854, the estimate was 60 million, and by 1858 it was 106 million. To put these numbers in context, the estimate for the annual budget of these years is 2 million pesos.84
In and of itself, however, relying on debt and the printing press does not explain why the Latin American countries did not impose domestic taxes after the wars. Many European countries initially used debt to pay for wars and later imposed taxes to meet their obligations. What distinguishes Latin America is that the fiscal reckoning never came. Moreover, government debt did not encourage the creation of a stable domestic financial market, a criti-
77. These are considerable sums, considering that the metal peso was officially set at practically dollar parity (López Cámarra, La estructura económica y social de Mexico, 171-72; López Gallo, Economía y política en la história de Mexico, 98).
78. Halperín-Donghi, Guerra y finanzas, 213.
79. Pomer, La guerra del Paraguay, 238.
80. Randall, A Comparative Economic History of Latin America, vol. 2, 215.
81. Oszlak, La formación del estado argentino, 217; Dirección de Contabilidad (Chile), Resumen de la hacienda púhlica de Chile, section VII.
82. Buescu, Evolucäo económica do Brasil, 119, 126; Nogueira, Raizes de urna Nagäo, 313; Castro Carreira, História financier a, 429.
83. Reyes Abadie and Vazquez Romero, Crónica general del Uruguay, 337.
84. Percentage of GDP or of export income would be much better measures of the relative size of these debts. Numbers on the size of the national economy even well into the twentieth century are notoriously unreliable. Export revenue and debts are often expressed in different currencies (e.g., paper vs. gold, or in different international currencies) making a more precise measure practically impossible. Reyes Abadie and Vazquez Romero, Crónica general del Uruguay, 339.
134     Blood and Debt
cal contribution of war in the cases of Britain and the Netherlands. Rather, government paper fueled unproductive cycles of speculation and ruin. Because of the risk involved, interest rates remained usurious, further hampering domestic development and increasing external dependence.
The availability of external resources freed the state from having to exploit the domestic economy. The relationship between the state and the global economy had three legs: foreign debt, the sale of commodities, and customs.
Much of the debt discussed earlier was to foreign banks.85 From the beginning, postindependence governments sought to supplement their inadequate domestic sources with foreign loans.86 Unlike the United States, for example, the new countries lacked allies and external aid, which meant that they had to pay hard cash for all the supplies that reached them. By 1820, the Gran Colombian government had already accumulated European debts of five hundred thousand pounds. Chile similarly contracted in 1822 for 1 million pounds to buy a navy. Peru had to obtain loans in order to pay back wages and a bonus to the victors of Ayacucho as well as some loans to the Gran Colombian government. Mexico borrowed 2.5 million pounds in 1824, 65 percent of which went to direct military expenses. Argentina borrowed in London beginning in 1824. Most of these proceeds went to the creation of a domestic financial system that was largely destroyed by the Argentine-Brazilian War of 1826-28.87 The first Brazilian loan of 1824 was used to pay for Portuguese loans from Britain. More money followed in 1825 and 1829. By 1830, Brazil had already borrowed 4.8 million pounds, or approximately four times its annual revenues.88 While Latin America was out of the international financial markets for nearly forty years following this early boom, it made a significant comeback after i860. For example, during the War of the Triple Alliance, Brazil borrowed 5 million pounds in 1865. In 1867, the state needed another loan from Rothschild for 71 million milreis.89
If they could not borrow on international markets (as was the case from roughly 1830 to 1870), Latin American states could sell access to a com-
85. Rippy, British Investments in Latin America; Marichal, A Century of Debt Crises.
86. These included some intracontinental debts, which usually represented payments for armies or military materials during the independence struggles. By the late 1820s, Bolivia owed Peru 725,000 pesos, Peru owed Colombia 6,000,000 pesos and Chile 3,000,000 pesos (Seck-inger, Brazilian Monarchy, 51).
87. Marichal, A Century of Debt Crises, Z7-36.
88. Cardoso, Mexico en el siglo XIX, 105.
89. Nogueira, Raizes de urna Nagäo, 378-380; Castro Carreira, História financiera, 429.
Making the State 135
modity. Guano allowed Peru to become what Shane Hunt has called a "rentier state."90 The availability of guano revenues retarded the development of the state by allowing it to exist without the remotest contact with the society on which it rested and without having to institute a more efficient administrative machine. Guano did allow the removal of the regressive contribution (in 1855), but it also permitted the state to avoid modernizing its fiscal structure while borrowing large amounts of money. A contemporary British observer noted that "a wise government would have treated this source of revenues as temporary and extraordinary. The Peruvians looked upon it as if it was permanent, abolishing other taxes, and recklessly increasing expenditure" (my emphasis).91 Much like the guano bonanza in the Peruvian case, the conquest of nitrate territories allowed the Chilean state to expand without having to "penetrate" its society and confront the rampant inequality.92 By 1900 nitrate and iodine accounted for 50 percent of Chilean revenues and 14 percent of GDP.93
Custom taxes represented an ideal solution to fiscal problems, given the organizational ease with which they could be collected. A few soldiers in the main ports could provide considerable income. More important, given their indirect nature, these taxes were the least likely to provoke popular protest.94 A revenue tariff was a characteristic feature of a society dominated by landed proprietors, who diverted taxation away from property toward the consumer.95 A reliance on customs also reflected the sectoral distribution of the continent's economies. For some countries, such as Peru, a large share of the national product was concentrated on the export of a commodity. Thus, government taxed that part of the economy that was most visible. Others having a less developed export market would target imports as well as exports. The distribution between these two often reflected the relative influence of importers over exporters. In the case of Brazil, for example, the heaviest taxes were on manufactured imports. This strategy, in turn, may have made this form of taxation even more regressive, depending on the distribution of goods within the typical import basket.
Argentina is an extreme example of this pattern. From the 1820s, the
90. Hunt, "Growth and Guano."
91. Markham, The War Between Peru and Chile, 1879-1882, 37.
92. Loveman, Chile, 169; Sater, Chile and the War of the Pacific, 227.
93. Mamalakis, The Growth and Structure of the Chilean Economy, 19-21; Sater, Chile and the War of the Pacific, 275.
94. Marichal, A Century of Debt Crises, 17.
95. Lynch, Spanish-American Revolutions, 150.
136     Blood and Debt
various incarnations of the Argentine state depended on customs duties for the vast majority of its income. The dictator Rosas continued the policy of allowing customs revenues to replace the more politically costly excise or land taxes.96 The fragility of such dependence was demonstrated by the European naval blockades of 1827, 1839, and 1846, which produced fiscal crises. Despite the massive changes in the Argentine economy during the last quarter of the century, including a fivefold increase in exports from 1862 to 1914, the fiscal system remained largely dependent on import taxes.97 Even by Latin American standards, this dependence is striking, as customs often accounted for more than 90 percent of ordinary revenues. Trade taxes were seen as the only way of maintaining some semblance of peace between the various politically relevant factions.98 All knew that this fiscal system was inadequate, but it was the only way of maintaining the social status quo.99 Tariffs were particularly attractive to the elite. They required no sacrifice, helped finance the expansion of the frontier from which the elite benefited disproportionately,100 and demanded few administrative resources.
The particular links between Latin America and the global markets had important domestic repercussions. First, they often linked the fiscal health of the state to the world economy and the price of a single commodity. In the well-known pattern, declines in trade or demand for a good could halve government receipts in a single year. Long-term planning and investment were impossible. Depending on the state apparatus as a political patron was also extremely risky. Precisely because the new governments were so fiscally strained and could not impose domestic taxes, they also could not risk losing foreign trade, from which they garnered such a huge part of their income. The fiscal use of trade thus contradicted any possibility of protectionist economic policy.101
96. Halperin-Donghi, Guerra y finanzas, 242-243.
97. Oszlak, La formacion, 173.
98. Oszlak, La formacion, 186.
99. Gorostegui de Torres, Historia Argentina, 120-21.
100. Oszlak, La formacion, 189.
101. This, combined with the already significant ideological bent toward laissez-faire, devastated what little domestic industry existed and worsened the already considerable tensions between urban centers and the interior. For example, the dependence on the guano receipts (either through loans or through customs) allowed the Peruvian government to follow an open-market policy, since there was no fiscal requirement for tariffs. This "scorched earth free trade" further devastated what was left of a domestic producer class (Gootenberg, Between Silver and Guano, 134). In Peru, unlike in Europe, economic liberalism was not used to help the victory of an industrial bourgeoisie over a rural oligarchy, but was employed by foreign capital to establish its dominance (Yepes del Castillo, Peru, 1820-1920, 38). The low tariffs enforced by the British on Brazil until 1844 had the expected devastating impact on domestic industry. A
Making the State 137
Simultaneously, the availability of foreign capital prevented the government from challenging elite groups and forging with them a national alliance. In fact, rather than war leading to greater central control, the absence of such sovereignty may have led to conflict. The War of the Pacific may best demonstrate the consequences of the external orientation of these states and the lack of domestic domination. It was "at heart a bald struggle over exports among jealous Chile, Bolivia, and Peru."102 "All three countries were hard up, and run by oligarchies which disliked paying taxes and looked to revenue from these fertilizers [nitrate] as a substitute."103 Each country was competing with the others for those resources that would allow it to maintain its "rentier" status and not challenge the domestic status quo.104 War came because the states were too weak to fight their respective elites. For example, because the elites of the Altiplano were too powerful to tax, the Bolivian state saw the littoral and the nascent nitrate industry as the best source of fiscal support.105 This brought it into conflict with Chile. But, precisely because it did not have adequate support from its home base, Bolivia could not hope to win.106
Wrong Timing
Rather than focusing on the importance of the "wrong kind" of wars or resources, it might be more accurate to say that the "right kind" of war came too soon. The Spanish Conquest in the sixteenth century had already subjugated the most powerful enemy of the criollos, the Indians. Arguably, where that fight continued well into the nineteenth century, it provided an important impetus for state making. Frontiers were areas of continuous warfare.107 On the northern Mexican frontier, for example, the fight against the Indians did generate a social consensus on the need to develop an extensive military force and the subsequent need for the ability to pay for it.108 The state of Nuevo Leon established relationships with its merchant and
similar decimation of the colonial artisanal and "manufacturing" class occurred in most other countries. Thus, Latin America delayed developing a national bourgeoisie around which a modern state apparatus could evolve.
102. Gootenberg, Imagining Development, 182.
103. Kiernan, "Foreign Interests in the War of the Pacific," 14.
104. Bonilla, "La crisis de 1872," 179.
105. Morner, The Andean Past, 140-43.
106. Klein, Bolivia, 144-46.
107. Duncan Baretta and Markoff, "Civilization and Barbarism."
108. Cerrutti, Economia de guerra y poder, 29-30.
138     Blood and Debt
landowning classes much more along the lines of the European model than what we see in the central government. The final Argentinean war against the Indians, also known as the "Conquest of the Desert," in 1879-80 perhaps made a more significant economic contribution than the War of the Triple Alliance, as it freed the frontier of the often expensive Indian raids while allowing vast expansion of agribusiness and the promotion of greater immigration. The war also helped to solidify the legitimacy of the state and made President Julio Roca's political career.109 The Conquest of the Desert was partly financed by land sales, which further established the power of the landowning elite.110 This territorial expansion, combined with developments in the international economy, helped consolidate the oligarchic control that was to allow the Argentinean state some measure of coherence during the following five decades. In Chile, the army that was to defeat its northern neighbors during two critical wars was largely born and trained on the frontier. Given the important role of the "frontier" in the development of the European state111 and the United States, as well as the experience of Argentina, Chile, and Mexico, the presence of an easily defined ethnic enemy outside the borders of a state could also have played an important role in Latin American political development.112
No matter the form of war, the Latin American states were not structurally, politically, or ideologically ready to exploit the political opportunities provided. The birth of the Latin American state, despite being announced by gunfire, did not produce the expected political apparatus. Liberation from Spain produced a much weaker institution (at least as measured by fiscal structure) and one much more dependent on the international economy. The independence wars wrecked the economy and indebted the countries, making the rise of the structural equivalent of a national bourgeoisie much more difficult. Thus Latin America was deprived of both a political and a social anchor on which to base institutional development. The Latin American state was never able to impose the internal unity required for the extraction process, even in the face of military threats. As strange as it might appear, given the oppression so endemic to the continent, the Latin Ameri-
109. Puigbo, Historia social y economica Argentina, 125-27; Gasio and San Roman, La conquista del progreso, 115-25; Alemann, Breve historia de la politica economica argentina, 101.
110. Rock, Argentina, 1516-1987, 154; Ferns, The Argentine Republic, 1516-1971, 70.
111. Robert Bartlett, The Making of Europe.
112. I owe much of this point to conversations with Michael Jimenez, Stephen Aron, and Jeremy Adelman.
Making the State 139
can state may have suffered from an incomplete process of internal domination. In the European cases, representatives of the monarchy, the landed oligarchy, or the newly developing bourgeoisie were either willing to bear part of the burden in order to protect themselves, or were able to impose that obligation on recalcitrant social sectors. In Latin America, control of the state remained in contention.
It may be that wars can only make states within an ideological framework of enlightened despotism.113 If so, independent Latin America provided an inauspicious setting. The postindependence period was not ideologically predisposed to state growth, in part a reaction to the expansion of the Spanish colonial state during the eighteenth century.114 This met significant resistance; after the 1770s, rebellion against colonial authority became increasingly commonplace. Thus it was particularly difficult for postindependence governments to impose new tax measures, as these were associated with the absolutism that had just been defeated. Old taxes were abolished before new ones could be instituted.115
The dominance of liberal economic thought throughout the continent also went against the idea of a powerful and intrusive state.116 The acceptance of classic liberalism set the ideological stage for the challenges to follow. Perhaps most important, none of the successful independence rebellions involved radical social reforms and most represented leading economic sectors. Those movements that did call for changes in the distribution of wealth, such as Hidalgo and Morelos in Mexico, may have actually contributed to the conservative bias of postindependence government by raising the specter of race war.
Latin America was not alone in this attitude toward taxation and government penetration, and certainly the United States had a very different institutional development despite similar ideological constraints. But Latin America faced obstacles not found further north. Even had the fiscal spirit been willing, it would have been difficult for the body to follow. Taxes do not collect themselves, but require a considerable administrative apparatus.
113. Kaiser, Politics and War, 206.
114. Similar to the experience of absolutism in western Europe, this generated a remarkable increase in resources available to the colonial governments. Mexican revenues, for example, increased fourfold during the eighteenth century (Bethell, Cambridge History: Independence, 10).
115. Burkholder and Johnson, Colonial Latin America, 330.
116. Collier, Ideas and Politics of Chilean Independence; Hale, Mexican Liberalism in the Age of Mora, 1821-1853; Burns, Poverty of Progress; Peloso and Tenenbaum, Liberals, Politics, and Power; Love and Jacobsen, Guiding the Invisible Hand.
140    Blood and Debt
The states facing war simply did not have the administrative capacity to respond with increased extraction. There was not enough "there" there to follow the coercion-extraction cycle.117 For example, the administrative backwardness of the Ecuadorian bureaucracy was such that double-entry bookkeeping was not successfully imposed even after the 1850s and 1860s.118 In 1851 and 1852, the Brazilian government attempted a census, to considerable opposition from members of almost all social sectors, who saw it as an effort to establish a list for either new taxes or conscription. As a result, the idea was abandoned.119 In addition, despite the obvious benefits of a land tax, the sheer task of a cadastral survey would have been beyond the capacity of the Brazilian state.120
In the following section I discuss the reasons why the state was so unable to make use of the opportunities presented by war. It was not only the number or type of wars that distinguished Latin America, but also the social context in which these were fought.121 Understanding the impact—or lack thereof—of war on the continent, requires analysis of both the conflicts themselves and the societies that fought them. While to some extent the institutional product of European war was technologically determined by the greater expenses involved in post-seventeenth-century warfare, the important break had more to do with the social and economic contexts in which the wars took place than with what happened on the battlefield.122 In Latin America, the institutional and political stimulus of the drums of war fell on deaf ears. The structures or authority required to make use of these opportunities were not consolidated. Thus, even if the subsequent history had looked much more like that of western Europe, the starting conditions were too different for the same outcome to result.
117. Lofstrom, "From Colony to Republic."
118. Rodriguez, The Search for Public Policy.
119. Barman, Brazil, 236.
120. For an interesting contrast with the Japanese case, see Bird, "Land Taxation and Economic Development." The key comparative issue here would be that the military revolution in Europe was also accompanied by a bureaucratic revolution that dramatically increased the administrative capacity of the state.
121. It should be clear that I am addressing questions of state building from above. That war may have contributed to nationalism from below or that it fostered different types of communities is now fairly well accepted (Mallon, Peasant and Nation). The question still stands of why it did not seem to affect the construction of authoritative and inclusive political institutions at the level of the nation-state.
122. Kaiser, Politics and War; Tallett, War and Society in Early Modern Europe; Waller-stein, Modern World System.
Making the State hi
Shadow States and Divided Societies
Wars provide opportunities for institutional development, but centralization requires a preexisting elaborated political logic. This can come from an already united elite that sees the growth of the state as in its interest, or alternatively, from a nascent class seeking to augment the territory under which it may function. Whether nationalist aristocracy or expanding bourgeoisie, these groups use war to defeat rivals or competing claimants. Latin America did not posses either group during its century of wars. There were too many divisions and claims to power. Unlike in Europe, wars did not provide opportunities for a single elite faction or family to impose its will on others, but rather functioned to perpetually maintain the possibility of rebellion.
We might wish to recall the distinction between a dominant and a ruling class.123 Latin America possessed the first, but arguably not the second. In general, Latin American countries lacked a single class able to impose its will and organize the capacities of the state toward war. At best, military caudillos, urban merchants, and large landowners made temporary and unstable alliances. The independence wars failed to produce the hegemony required for the conjunction of military action and internal extraction. No faction of the dominant class was able to establish a strong enough hegemony to prioritize national collective interests (even if still defined in class terms). Because of the absence of this dominion, the state apparatus was not truly fiscally sovereign.
The key to understanding the Latin American failure to "benefit" from war lies in the myriad divisions that characterized these societies, detectable even prior to independence, when Latin America underwent an attempt at centralization of authority partly fueled by military concerns: the Bourbon reforms. The Seven Years War, which ended in 1763, and the Spanish involvement in the campaigns against France beginning in 1793 challenged the status quo relationship that had developed between Iberia and America.124 The reforms initiated by Charles III, involving attempts to re-centralize authority and increase revenue, could be interpreted as an effort to transform the unwieldy Bourbon Empire into something resembling a
123. I owe this point to John Womack.
124. Bethell, Cambridge History: Independence; Lynch, Spanish Colonial Administration; Bourbon Spain, iyoo-1808; Fisher, Government and Society in Colonial Peru; Brading, Miners and Merchants in Bourbon Mexico; Haciendas and Ranchos. See also Stein, Silver, Trade, and War.
142     Blood and Debt
cross-Atlantic state. The Bourbon army was essentially created to deal with these external challenges.125 The reforms were considerably successful, although met with opposition, including violent protest.126 Could these reforms have produced a state closer to the European model? Following Rodriguez one could imagine the development of such a political entity, or more likely, the consolidation producing three or four larger states than arose out of the independence wars.127 In any case, the Bourbon reforms indicate that the role of war as a stimulus or incentive for institutional development and political consolidation was relevant to the continent prior to the nineteenth century.
This experience provides a hint of the future role of war in state building.128 Thanks to the increased need for armed force, the military was granted greater institutional autonomy through the fueros. The result was that the military played a more important role in the region, but remained outside society and above the state, establishing a pattern that would continue for years to come. Most important, while the greater military and administrative capacity may have developed as a response to international threats, armed force was increasingly directed inward. The colonial state came to be oriented not to protecting the society from an external menace, but to repressing internal threats.129
It is these real and perceived threats that best explain the particular relationship between state and war making on the continent. The Bourbon reforms produced a more efficient political apparatus, but they also brought to a head the internal divisions that would haunt Latin America over the following one hundred years. Military tensions and concerns created administrative and fiscal crises that encouraged the state to impose its authority, but the last quarter of the eighteenth century already saw the conflicts that would plague attempts to create more solid political structures. The society on which the state rested was not sufficiently united to provide an adequate arena for institutional consolidation. Institutional developments fueled by military conflict were frustrated by the strength of geographical, social, and racial divisions. These appeared in slightly different forms across the conti-
125. Rodriguez, introduction to Rank and Privilege, x.
126. Phelan, The People and the King; Stern, Resistance, Rebellion, and Consciousness in the Andean Peasant World, Eighteenth to Twentieth Centuries.
127. Rodriguez O., Independence in Spanish America.
128. Archer, Army in Bourbon Mexico; L. G. Campbell, "The Army of Peru and the Tupac Amaru Revolt," i; McAlister, The "Fuero Militar" in New Spain, 1764-1800.
129. It was largely unsuccessful in the face of external menace. For example, the British were thrown out of Buenos Aires in 1806 by forces organized and led by nonoffkial actors.
Making the State 143
nent (Table 3.4), but the result was generally the same. War led not to increased order and unity, but to chaos and division. In the following pages I discuss each of these divisions in turn.
Regionalism
Despite the efforts of Charles III, Latin America entered the nineteenth century arguably more divided than ever. The Americas as a whole resented the imposition of an order dictated in Madrid. The various subunits of the Bourbon domains wished to protect and expand their autonomy vis-a-vis the central power. The viceregal seats in turn resisted efforts to create small, more autonomous governments from their individual parts (for example, the captaincy-generals and intendencias). Granting greater local autonomy to one group, as in the creation of the Viceroyalty of Rio de Plata, often meant diminishing that of others, such as Paraguay, or forcing others to shift their administrative allegiances, as in Charcas. Thus, efforts to create a greater whole—whatever the benefits to be derived from this—were met with resistance from those who felt that their individual part would suffer. Autonomist rebellions consistently brought to head salient social divisions within the various regions.
The notion of a fragmented sovereignty had a long history on the continent. The colonial regime had already recognized considerable regional diversity and autonomy and even had floated a plan of dividing the continent into three kingdoms—Mexico, Peru, and Nueva Granada.130 The conflicts over sovereignty were not simply between province and capital, but also within the provinces themselves, between regional and municipal governments.131 In any case, the colonial state barely controlled large parts of the empire. Most of northern Mexico was beyond its control, as was the southern area of the continent. In part because of geographical and logistical challenges, and in part because of Spain's fears of transatlantic autonomy, the Bourbon reforms established at least a semicentralized state, but not a unitary one.132 Each part of the empire was connected to the center, but the separate regions were not linked with one another. The resulting political entity that had to face the challenge of the Napoleonic invasion in 1808 thus lacked a solid territorial cohesion. The crown was the real source of
130. Jaramillo Uribe, "Nación y región,"
131. Annino, "Soberanías en lucha," 250.
132. Kossok, "Revolución, estado y nación," 163.
Table 3.4   Internal divisions in Latin America
	Regionalism	Racism/Class Conflict	Elite Division
Argentina	Buenos Aires vs. provinces	Indian conflict; gauchos	Unitarios vs. Federalistas; different
			caudillos
Bolivia	La Paz vs. Sucre	Minuscule elite with	No dominant group
		disenfranchised Indian masses	
Brazil	Rampant through 19th century	Concern with postslavery inclusion	Declines after regency
Chile	Santiago soon dominant	Becomes more relevant with mining	Later 19th-century ideological
		development	divisions
Colombia	Rampant through 19th century	Centered on class divisions	Liberals vs. conservatives
Ecuador	Coast vs. Sierra	Indian vs. white elite	Liberals vs. conservatives
Mexico	Provinces at margin	Very strong in postindependence	Liberals vs. conservatives; caudillo
			politics
Paraguay	Asuncion dominates	State co-opts Indian identity	None left after Francia
Peru	Lima vs. Sierra vs. Coast	Fear of Indian uprising	Caudillo and follows geography
Uruguay	Montevideo vs.	Urban class conflict	Blancos/Colorados
	rural		
Venezuela	Llanos vs. coast	Very strong in postindependence	Caudillos
Making the State us
sovereignty, and when the political order came under strain and the crown lost much of its inherent legitimacy, there were relatively weak links keeping the various parts together.
The early experiences with the wars of independence exacerbated these conflicts of sovereignty. In large part because of both sides' need to acquire resources, rebel and loyal administrations each sought to centralize power over areas under their control. Such efforts continued after the 1820s because of fear of postbellic chaos. Since both the royal governments and their successors were generally too weak to enforce their constitutional claims, attempts to at least formally increase their authority merely exacerbated local fears without resolving the conflict in their favor. Instead of a true federalism with enough assurances to ensure provincial loyalty, or an autocratic centralism resolving regional differences, efforts at centralization merely produced resentment and rebellion. The result was the worst of all possible worlds: the threat of central authority kept the provinces and local powers restive, while the limits on the state prevented a final resolution.
The wars of independence witnessed the dispersion and dissipation of political authority. In New Granada, the authority of the autonomist government in Bogota did not extend much past that city as Cundinamarca and Santa Marta, among others, claimed different allegiances, and patrias bobas survived repeated attempts to centralize authority. In Ecuador, the longstanding rivalry between coastal Guayaquil and highlands Quito moved to new levels as the latter became politically dominated by Venezuelans from the invading liberators. In Bolivia, Sucre provided the Charcas elite with the military protection it needed to separate itself from Peru and establish its own independent nation. The very same logic that would allow La Plata to break from the rest of the empire seemed to permit each of the provinces to divorce itself from Buenos Aires. Each new subunit produced further claims to autonomy. The state of Tucuman faced secession from Santiago de Estero and La Rioja broke off from Cordoba. The failure of Buenos Aires to stop this process is most critical to understanding the following fifty years. Forces in both what would become Bolivia and Paraguay defeated porteho attempts to keep them in the old viceroyalty. These failures served as an inspiration to other regions. Moreover, the military and political effort expended by the various expeditions made it difficult to gather the resources needed to keep other provinces in line.
The discrepancy in resources available to the provinces actually abetted regionalism, since it often made the richer regions, often associated with the capital, reluctant to enter into political contracts that necessitated sharing
146     Blood and Debt
their wealth. The Argentine situation was emblematic of a pattern found across the continent. The income of Tucumän, for example, was one-thousandth of that available to Buenos Aires, while that of Jujuy was almost nonexistent. In 1827, the province of Cordoba had an income of seventy thousand pesos whereas Buenos Aires had 2.5 million.133 If unwilling to share its bounty, Buenos Aires might have yet conquered the rest of the country. But poverty did not mean military incompetence, and the poorer regions were able to hold the richer ones at bay while never being able to conquer them.
In Mexico, the pattern continued after independence, when the country essentially dissolved into a "series of satrapies dominated by caudillos."134 Throughout the nineteenth century, the central Mexican state could not even eliminate interstate barriers to national commerce. The Texas War, which began in 1835 and arguably continued until 1847, certainly did not aid the institutional development of the Mexican state. The many civil wars actually impeded efforts to impose central authority, as individual provinces could usually find a general or pretender to challenge any dictates from Mexico City or attempt to change the government. Ironically, Mexico City was accused of attempting to establish a new Tenochtitlän even as it could not control the road to Veracruz.135 In fact, no single authority possessed the capacity to impose its project on the whole country.136 Political regionalism arguably survived the Diaz era and was only stamped out after 1910.137
The independence project also suffered from a critical contradiction at its core.138 On the one hand, the independence of each region was always subordinated to the independence of the continent. No province was allowed to concentrate on its purely protonational interests. The struggle was to free America. On the other hand, however, the sovereignty of each region was never sacrificed to a central entity. Conquest was not allowed. San Martin could not conquer Chile for La Plata. But Chile did not completely control the armies within its borders. This resulted in a disastrous combination of a supranational military with regional political authority, in turn, producing a disjunction between military might and territorial limits that fatally weakened the centralizing effect of war.
133. Scobie, Argentina, 94; Bürgin, Economic Aspects of Argentine Federalism, 126.
134. Coatsworth, "Obstacles to Growth in Nineteenth Century Mexico," 95.
135. This was not a unique situation. In the 1820s and 1830s, the road from Lima to Callao was often unusable because of bandits (Dobyns and Doughty, Peru, 171).
136. Cardoso, Mexico en el siglo XIX, 60.
137. G. Thompson, "Federalism and Vantonalism in Mexico."
138. Kossok, "Revolucion, estado y nacion."
Making the State 147
As Halperin-Donghi has often noted, the independence wars established a particularly vicious connection between regional identities and divisions and military force. Thanks to the duration and savagery of the independence struggle, by the 1820s the continent had literally dozens of armed groups contending for control. Perhaps more important, military power was essentially autonomous. With few exceptions, armies did not evolve past a provincial militia mode, with a few professionals leading local masses.139 The military was not the armed representative of the state, but fought for either an individual leader or some vague notion of "liberty." Partly because of relatively low technical requirements, partly because of the absence of competition, leaders of irregular armed forces could easily become independent of the governments that organized them. Armies were not regulated by any central authority, but were often under the control of precisely those who wished to be left alone by such powers. Armies, in this case, were not a way of establishing political authority, but keeping it at bay. Military mobilization was not accompanied, much less preceded, by political mobilization.
Even countries that were spared ruinous independence wars found regionalism an almost insurmountable obstacle. The first thirty years of the Brazilian Empire witnessed yearly efforts to bring a recalcitrant region or social group under the control of Rio. While the army did serve as a unifier during much of this period, particularly during the 1830s and 1840s, it also suffered from regional divisions. Military camps and barracks, for example, would often be divided according to the geographical origins of the troops.140 The empire never established a strong central authority. Instead it could be interpreted as a patrimonial state operating a network of favors, guaranteeing some legitimacy, and serving as the police of last resort for elite squabbles.141
Rather than mitigating regionalism, war often made geographical divisions even more detrimental to centralized authority. To survive in often difficult logistical situations or to obtain needed resources, armies (whether claimants to power or representatives of official authority) often had to negotiate deals with local powers. As recruitment was often geographically concentrated, significant parts of armies also reflected their provincial origins and replicated their loyalties to their towns through their chiefs.142
139. Annino, "Soberanias en lucha," 252.
140. Nunn, Yesterday's Soldiers, 58-59.
141. Graham, Patronage and Politics in Nineteenth-Century Brazil.
142. Annino, "Soberanias en lucha," 252.
148     Blood and Debt
Racism
One could argue that some early military conflicts helped create a nascent Latin American identity. The militias and even formal armies established by the Bourbon reforms were largely staffed by colonials.143 The military was arguably the first truly American institution and one that helped develop and consolidate a separate identity of criolloism. Yet a more salient response to the Bourbon reforms, and again one that would haunt Latin America into contemporary times, was increasing conflict between races, classes, and castes. The Tupac Amaru and Quito revolts grew from foundations much older and more complex than the Bourbon reforms. In the case of the Peruvian rebellion, we should take into account the frustrations of the old Inca aristocracy, the heavy weight of the mita forced labor, regionalist pressures to create a separate intendancy for Cuzco, and most important, racial strife. What is most relevant for our purposes, however, is that the specter of race war hampered the progress of the rebellion and helped shape attitudes toward the loyalist victory in 1783.
Much more than in Europe and or even the United States, Latin American elites lived in constant fear of the enemy below. More than any competing elite across a border, nonwhite subalterns represented the most significant threat to the social status quo. Such fears need to be understood in light of the demographics facing the white elite at the beginning to middle of the century. In Brazil, for example, slaves made up one-third of the population.144 New Spain at independence consisted of 20 percent whites, 20 percent mestizos, 40 percent Indians, and 20 percent castas.145 New Granada at independence was 33 percent whites, 43 percent mestizos, 17 percent Indians, and 6.5 percent slaves while Venezuela had sixty thousand slaves.146 Quito was 90 percent Indian.147
Racial conflicts were particularly critical to the historical impact of the war of independence in Mexico. The Hidalgo and Morelos revolts of the first part of the decade convinced a significant part of even autonomist criollo opinion that the dissolution of the political status quo would produce a race war in which they would suffer the fate of their equivalents in Guanajuato or, the true specter, Haiti. The criollo elite united in supporting the authori-
143. Loveman, For la Patria, 15.
144. Bushnell and Macaulay, Emergence of Latin America, 150.
145. Torre Villar, "El origen del estado mexicano," 128.
146. Jaramillo Uribe, "Nacion y region," 342-47.
147. Ocampo Lopez, "La separacion de la Gran Colombia," 370.
Making the State 149
ties, and by 1814, loyalist control was assured. Rather than an assertion of liberal sentiment, the eventual independence of Mexico came as a reaction to the Cadiz Mutiny of 1820 and the threat it represented to vested interests and ideologies. The almost bloodless Iturbide revolt that produced an independent Mexico in 1821 was a classic rebellion from above meant to slow social change through political restructuring.148
Racism and fear of armed nonwhites also hampered Bolivar's early campaigns. The royal government successfully exploited these tensions. By 1812, a royalist army allied with blacks, Indians, and pardos was able to impose its authority over Venezuela. Bolivar's next attempt did include an alliance with the pardos, and more important, with the llanero cowboys under Jose Antonio Paez. Later campaigns in Peru, in turn, appeared to have been hampered by perceptions of the independence army as Indian dominated.
Throughout the following century the prospect of armed Indians alarmed white elites (much as the prospect of armed ex-slaves and freedmen alarmed both Northern and Southern whites in the American Civil War). This was not only because of the immediate access to violence that guns provided, but also because of the perhaps more dangerous and insidious notion that participation in battle bestowed equality on Indians. Peasants who fought did begin to believe in their own equality as soldiers and demanded to be treated accordingly.149 Thus, the very fact that the military was at least perceived as a ladder for social and ethnic mobility made its role as a national unifier problematic. The "tool" was tainted by the very problems it was meant to solve.
This contradiction was complicated by the army's central mission in most of Latin America: protection of "civilization" either from revolts by internal subalterns or through defense of frontiers from "savages." Yet large numbers of those in the army were ethnically (as well as socioeconomically and geographically) related to the "enemy." The military was composed of the very same threat from which it was to protect the nation.
Racism also often limited the potential authority of promising leaders, even when faced by war. Andres Santa Cruz attempted to unite Peru and Bolivia and faced war with Chile. Yet he received little support from the Lima elite, who despised him for his race and class origins. Members of the
148. Sometimes, however, racial fears could actually assist central authority. Creole Yuca-tecos' willingness to obey Mexico City certainly increased when they realized that they needed the central government's help to put down local Indian revolts in the 1840s.
149. Mallon, Defense of Community, 88.
150    Blood and Debt
Peruvian elite (along with future presidents Agustin Gamarra and Ramon Castilla) fought on the side of the Chileans against Santa Cruz.150 Rather than consolidating a sense of nation and Lima's sovereignty, the war maintained the isolated fiefdoms defined by geographical regions. Gamarra and Castilla subsequently faced the same discrimination against their mixed origins when they assumed power.
Racial divisions enjoyed some institutional sanctions. Well into the nineteenth century, for example, there existed republicas indias where the national government did not rule, as well as pockets of territorial loyalties or patrias cbicas often associated with indigenous groups. Liberal critiques of these separate nations were not completely off the mark in suggesting that no nation could arise as long as these communities existed. This is not to deny the disastrous consequences for the Indian populations when liberals later in the century dismantled these protections. Yet, while they did serve to protect segments of the population against the commercial onslaught that was often to leave them landless, their existence made the consolidation of a single nation very difficult. The idea of two—if not more—nations haunted the nineteenth century.
While all the major European nations created their nationalities as they were developing their territories, they were not as internally divided as Latin America. If making Italians, Germans, Britons, or the French involved the forceful imposition of the culture of a single region or creation of a compromise national language, it did not require the mending of centuries-long racial gulfs closely correlated with the distribution of political and economic power. The very composition of "the nation" was fraught with conflict. Under these circumstances, wars were not occasions for institutional unity, but represented opportunities for groups to opt out of the national project. In the end, the official military was not organized to protect or even co-opt the people, but to coerce them.151
Elite Divisions
Despite, or perhaps because of, their extremely privileged position, elites in most Latin American countries have been internally divided.152 This took a
150. Dobyns and Doughty, Peru, 158.
151. I emphasize the form of military, since we need to distinguish it from more popular armed organizations such as resisted the Chileans or fought against the French (Mallon, Peasant and Nation).
152. My argument here of course follows the work of Halperin-Donghi.
Making the State 151
variety of forms during the Bourbon period. One was the struggle between state and church. The Bourbon reforms sought to increase the power of the former at the cost of the latter. In some places, most prominently in the case of the Jesuits of Paraguay, this involved removing church influence altogether. In others it simply meant a shift in control of resources. A second form of intra-elite struggle exacerbated by the Bourbon reforms was that between American-born criollos and peninsulares, born in Spain. The latter had the most to benefit from attempts to delocalize administration and associated efforts to increase immigration from Spain. Although often dressed in the Enlightenment guise of liberty and rights, American opposition to the reforms frequently had more to do with the protection of sinecures and social positions. The imposition of mercantilist policies, which again might have benefited the Iberian-American Empire as a whole, met with a similar response.
There were also divisions within the elite regarding the benefits of remaining within the imperial system. The Spanish crown offered some American elites a series of important advantages, protection from external threats and internal security being perhaps the most important. While large portions of the American population might chafe under the control of Madrid—such as it was—significant groups saw it as preferable to a situation in which they would not be able to maintain their social, political, and economic control. It is not surprising that some elites (and nonelites) were loyal to the crown and that the independence struggle took nearly two decades to complete.
Elite divisions made it difficult to unify coercive capacity and use it constructively. At least theoretically, Argentina and Chile should have enjoyed the best of all possible worlds in their experience with the wars of independence. While both (but especially Argentina) provided important logistical support for the liberation of other regions, neither suffered significant destruction. Yet in both cases, the wars of independence did not produce the consolidated state we might have predicted and in one, such an institution would not arise until fifty years later. How do we explain this pattern?
In the case of Argentina, the first decade following May 1810 was characterized by elite divisions not simply along peninsular/crioMo lines, but also between groups wishing different levels of separation from the government in Spain. By the Congress of Tucuman, Argentina had seen "two juntas, two triumvirates, one assembly, one directorate with four office holders, and a constituent congress."153 In part this may be explained by the very absence
153. Navarro de Garcia, "El orden tradicional," 156.
152     Blood and Debt
of a believable threat to Buenos Aires, but it also reflected real divisions in the Argentinean elite along both instrumental and ideological lines. A fateful decision in this regard was to allow Buenos Aires's best general and considerable troops and resources to be devoted to the defeat of Loyalist forces in Chile and Peru. A San Martin in Entre Rios or Santa Fe might have given a Buenos Aires-dominated union a better chance. In this case, it would appear that external war (if so we can count the Chilean and Peruvian campaigns) sucked strength out of the nascent state. Had San Martin been needed to fight a believable Spanish threat in Montevideo after 1814, he might have been able to impose a unitary order on the other provinces.
The initial Argentinean involvement in Uruguay and fears of Brazilian reaction in the late 1820s produced greater support for a stronger president. Bernardino Rivadavia certainly benefited from some of the victories against the Brazilians. He even envisioned using the army that had fought Brazil in Uruguay for local consolidation: "haremos la unidad a palos."154 But his inability or unwillingness to exploit the peace created tension and actually led to the subsequent political dissolution.
On the other side of the river, the Cisplatine war was a disaster for Pedro I. The defeat—and Uruguayan independence was perceived as such— weakened his authority and complicated the political balance between centrists and federalists. In both cases, divisions inside the ruling elites led to the waste of the opportunity presented by international conflict.155 In general, however, Brazil after 1840 may represent an important exception to this pattern of elite division, as it was able to develop something resembling a governing class whose professional and political interests were linked to the preservation and even expansion of state authority.156 In this instance a war may have assisted in the imposition of a governmental directive. The War of the Triple Alliance provided the government with a better opportunity than it had ever enjoyed to challenge the slave owners' power and impose at least gradual manumission.
Elite divisions also helped shape the independence struggle on the northern coast of the continent. In the early stages of the struggle, some of the
154. Rock, Argentina, 1516-1987, 102.
155. Wars in the Platine region did provide resources for central authorities to purchase political loyalty. Rosas's campaign against the Indians in the early 1830s gave him a great deal more land with which to reward allies. Similarly, Brazilian success in Uruguay following Rosas's defeat improved relations between the emperor and the elites of the southern states.
156. Merquior, "Patterns of State-Building"; Graham, Patronage and Politics; Carvalho, "Political Elites and State-Building."
Making the State 153
elite feared that the collapse of the royal government in Spain would provide an opportunity for the nonwhite majority to threaten the social status quo, but a more radical segment wished to move toward full independence. The resulting power struggle quashed both hopes. Criollos killed criollos under a variety of flags. Not even Bolivar's considerable military successes could stop the infighting. By 1819, Bolivar once again controlled most of what would become Colombia and Venezuela. But the wars had left the region devastated; various groups (most important, Päez's army) remained largely beyond the control of the central government; and regional antagonisms remained. When Bolivar moved his army south, his representative Francisco Santander was not able to hold the various parts together.
Postindependence Peru continued to suffer from elite divisions, often correlated with regions. Members of the Lima elite were divided in their attitudes to Bolivar and San Martin and their successors. In turn, Lima was split from the merchants and miners of the highlands, who did not see eye to eye with sugar planters on the coast.157 With the possible exception of Ramon Castilla, no single political entrepreneur could establish a monopoly over national power. From 1826 to 1865, thirty-four different men served as Peru's chief executive.158 These divisions played a major role in the eventual defeat of Peru in the War of the Pacific. In the words of Florencia Mal-lon, "[N]o measure of heroic exploits or symbols could compensate for the lack of unity and national purpose of the Peruvian elite."159 In Bolivia, the independence army was soon torn apart in a struggle between those who sought union with southern or western neighbors and those who wanted full independence.160
Further, the Mexican war of independence failed to provide the opportunity to consolidate authority in a postcolonial setting. On the one hand, elites were united enough to resist a popular insurrection that might have created a more socially revolutionary national government. On the other, once the threat of race war was removed, no individual segment was strong enough to impose its will on the others. Even more than in Argentina and Chile, independence destroyed what political authority had existed without leaving the framework for domination by a central government. The construction of a new nation-state was begun without the existence of a hege-
157. Halperin-Donghi, Contemporary History, 99.
158. Dobyns and Doughty, Peru, 158.
159. Mallon, Defense of Community, 82.
160. Buisson, Kahle, König, and Pietschmann, Problemas de la formacion del estado y de la naciön en Hispanoamerica, 502.
154     Blood and Debt
monic power bloc; instead of durable alliances, Mexico had hard battles.161 Santa Anna's disastrous, decades-long rebellions began in 1822—merely a year after his helping Iturbide come to power. Perhaps worse than Santa Anna's meddling was his failure (or lack of interest?) in establishing a permanent domination over the Mexican state. Neither domineering autocrat nor obedient subject, Santa Anna made Mexican political stability impossible.
Latin America was torn in a half-century struggle between what would be called, despite localized differences, Liberal and Conservative views of the role of the state.162 Liberals—arguing for greater political inclusion and intellectual and commercial freedom—had dominated the independence struggles. But in the immediate aftermath of victory they faced a dilemma: how to protect individual rights while also constructing a new political order.163 In opposition, Conservatives sought to protect what they saw as worthwhile inheritances from the colonial period: protection of property, the church, and some economic sectors. The two strands were associated (in varying ways) with federalist versus centralist projects. Divisions also persisted during much of this period over notions of citizenship, sovereignty, and the relation between state and society. For much of the nineteenth century, Latin America was stuck between a liberalism that did not guarantee order and a form of nationalism that would inherently exclude a large part of the population. Supporters of opposing visions "prolonged the military phase of the independence movements and virtually guaranteed that chaos would be an inescapable legacy of newly formed states."164 Much as in fifteenth-century Italy, the prevalence of factions and their myopic preoccupation with local battles retarded the creation of a political union capable of acting on the international stage and expanding its domestic authority.165
The nature or outcome of disputes was not as important as the fact that for many years neither tendency was able to completely dominate political
161. San Juan Victoria and Velazquez Ramirez, "La formacion del estado," 67.
162. There were also cases where the struggles were more openly between elite factions not encumbered with much ideological baggage, as was the case with Blancos and Colorados in Uruguay. In this instance elite divisions appeared to have been at least partly the product of international squabbles and subsequent conflicts. In this sense, wars were very much responsible for the underdevelopment of the Uruguayan state (Lopez-Alves, "Wars and the Formation of Political Parties").
163. Adelman and Centeno, "Law and the Failure of Liberalism in Latin America"; Bo-tana, La tradicion republicana.
164. Knight, "State of Sovereignty and the Sovereignty of States," 18.
165. Morse, "Theory of Spanish American Government," 79.
Making the State 155
life. The "German" road to nationhood was not open, given the ethnic divisions, but the state was not strong enough to enforce a "French" road.166 Nor could any one camp construct systems acceptable to the others. Without either consensus or hegemony, many Latin American states could not consolidate their rule. One war did help resolve these struggles and arguably played a role similar to that of conflicts in Europe. The Mexican war against the French (in many ways a continuation of the Reforma) destroyed the classic conservative elite. It also necessitated the creation of a truly national army garrisoned throughout the country. This army and the elimination of some elite divisions provided the basis for the Porfiriato.
Because these divisions played themselves out militarily all over the continent, armies, for all intents and purposes, were the state, and they certainly consumed the largest part of its resources.167 But unlike in Europe, the military did not serve a single master. First, it was willing to be bought by whichever actor promised the best reward for its services. Equally important, because of the lack of technical sophistication, it could not impose a monopoly on the means of violence. The costs of entry into the military-political competitive market were generally low. Throughout the continent, provinces and local caudillos raised and maintained militias that protected their interests. Militias served to defend property, not governments.168 The militarization of Latin America during this period represented the worst of all possible worlds: armies fought without being able to dominate and they coerced without extracting. Although draining large amounts of money, the military did not provide a means with which to pay for itself. Here we come to the crux of the Latin American puzzle that may explain not only the state's relationship with war, but also that of the military with civilian authority. Through a series of historical events beginning with the independence war, the military assumed a political autonomy separate from the state as such. Much has been written about how this encouraged the military to set itself up as the ultimate judge of national virtue. We have appreciated less the fact that this divorce between state and military robbed the former of an assured means with which to impose its will.
The absence of an institutional consensus and the difficulties facing the establishment of order often forced the most successful rulers to ignore constitutional principles. That is, order was too often based on the disregard of
166. Brubaker, Citizenship and Nationhood.
167. Halperin-Donghi, Aftermath of Revolution, 74-75.
168. Halperin-Donghi, Aftermath of Revolution, 9.
156     Blood and Debt
law that made the long-term consolidation of a political system and the creation of an elite consensus still more difficult. Vincente Rocafuerte best expressed this contradiction when he declared himself "a true lover of enlightenment and civilization," adding, "I consent to pass for a tyrant."169
But it was not tyranny as such that represented a problem for the growth of the state, but on what forms of authority that tyranny rested. Here it is important to analyze the role of political entrepreneurs in Latin America. For much of the first seventy-five years of independence, the region's critical political actor was not the institutionalized authority of the state, but the much more personalized rule of the caudillo.170 Arising from the destruction of colonial institutions, the emergence of local power centers, and the need for some form of order, the caudillos sought to alternatively appropriate the power of the central state, as in the case of Paez or Santa Anna; or challenge it, as in the classic Argentine regional caudillos such as Estanislao Lopez or Facundo Quiroga.171 If the "worst" were full of passionate intensity and political prowess, the "best" were unable to defend their authority. Santa Anna's antagonist, Lucas Alaman, for example, never served a government strong enough to impose a permanent centralized republic.
The more interesting cases are the caudillos who could have become much more. Paez gave Venezuela two decades of peace by successfully constructing an alliance between the military and pre-independence elites and Rosas temporarily managed an equilibrium of various regional leaders' interests. Such gifted caudillos sometimes were able to construct the semblance of states, but these were hardly institutional orders and rarely, if ever, survived their founders. Perhaps the most mysterious of cases is Rosas. There is no question that he pacified the province of Buenos Aires or that he was at least first among equals in the confederation. Once he had established his position, however, he seemed uninterested in expanding the region under his direct control. Rosas's nearly constant warfare, especially his conflicts with Britain and France, helped to consolidate his popular legitimacy, but he never used this to do anything but reinforce his control over Buenos Aires Province. The forty years of civil war following independence brought little more than superficial change to an Argentina still dominated by caudillos.
Neither Argentina nor any of the other nations produced by indepen-
169. Bethell, Cambridge History, vol. 3, 369.
170. Lynch, Caudillos in Spanish America.
171. Lynch, Caudillos in Spanish America.
Making the State 157
dence experienced a post-Thermidor, which would have institutionalized the changes brought about through political revolution and re-created the state in a new bureaucratic form. The result was a disastrous combination of local autocracy with little weak central domination; a continent of repressive islands with few links between them.
Nor was the appearance of the Latin American state accompanied by the rise of a hegemonic class willing and able to ride it to social and political dominance. The wars of independence were produced by the collapse of the Spanish crown's legitimacy, not by internal changes in the colonial societies. The wars disrupted the old order, but they did not establish an alternative system of domination. When the colonial apparat disappeared, no social group had an interest in replacing it with one equally strong. What the criollos wanted was as little interference as possible in their immediate profit making. The availability of international moneys allowed elites an exit, thereby inhibiting the development of class loyalty to the state.
In an almost complete reversal of the European pattern, the appearance of the modern state strengthened the political power of the landowning class. It was a Fronde in reverse. Those who possessed resources were completely successful in protecting their wealth. The only exception came, perhaps, in the immediate period of the independence wars. When San Martin finally arrived in Peru in 1821, for example, he gave a great deal of Spanish-held property to those who fought in his army. It is important to note, however, that the criollo landowning class was largely spared these sacrifices, as they were borne by the Spanish peninsulares and, in some cases, the church. When the criollo elite was asked to pay for its independence, it almost always refused, a pattern that was to be repeated for the following hundred years, if not longer. Given that the criollos were unwilling to pay even for the elimination of the old masters, we should not be surprised that they would be reluctant to pay for a new one.
The response to subsequent wars was similar. While the Chilean armies were marching on Lima, Peruvian finance minister Quimper suggested a small tax on capital to pay the troops in the field. These measures were defeated.172 The government also asked for an internal loan of 10 million soles. This request generated 1 million, largely from the "popular classes," as the rich did not want to risk their money.173 During the same war, the
172. Ugarte, Historia economica del Peru; United States Department of State, Foreign Relations, 165-68.
173. Bonilla, "War of the Pacific," 99. The British ambassador noted with surprise that "Peru appears struck with paralysis; the people themselves seem as indifferent to the future as
158     Blood and Debt
Chilean legislature was repeatedly unable to impose a wealth or an income tax.174 When Mexican finance minister Lorenzo Zavala attempted to impose a direct tax to finance a defense against the possible Spanish invasion of 1829, he was defeated and his government overthrown by an elite-sponsored coup.175 Similar efforts during the so-called Pastry War with France produced identical results. Even as the U.S. army marched toward Mexico City in 1847, the government frantically negotiated with the church and domestic lenders for funds.176 In Brazil the Chamber of Deputies consistently refused to give any funds to Pedro I to fight in Uruguay.177
Taxes and the avoidance thereof made it very clear where the line marking off the dominant from the dominated was to be drawn.178 In Brazil, the fazenderio was systematically avoided as an object of taxation. Discussions regarding land or income taxes had no result; the landowning elite was considered fiscally untouchable.179 Bolivia's Sucre attempted to impose a direct tax on wealth in the 1820s; within a year, this tax had been abolished. Resistance was both economically and racially based. Along with the rich resisting the new imposition, whites resented being placed under a contribution and being placed on the same level as Indians.180 Argentinean attempts to expand the tax base faltered because of the successful opposition of powerful social interests already well represented in the legislature.181 The contribution directa was a farce, as the legislature would not allow the creation of an independent system of assessment.182 In the 1830s, a ranch with nineteen thousand head of cattle paid a total of 540 depreciated pesos.183
While the avoidance of taxes is perhaps one of the few truly universal traits, the absolute regressivity of the Latin American cases compares unfavorably with some European cases. The resistance of the French is famous, but the British and German propertied classes were made to pay in one form
the governing classes, who are thinking more of their personal ambition than the welfare of their country" (quoted in Bonilla, "War of the Pacific," 98).
174. Sater, Chile and the War of the Pacific, 131-54.
175. Tenenbaum, The Politics of Penury, 34-35.
176. Tenenbaum, Politics of Penury, 79.
177. Haring, Empire in Brazil, 35 n. 17.
178. Gomes, The Roots of State Intervention in the Brazilian Economy, 93-94.
179. Buescu, Evolugao economica do Brasil, 142; Leff, Underdevelopment and Development in Brazil.
180. Lofstrom, "Attempted Economic Reform," 282-86; Paz, Historia economica de Bolivia, 52.
181. Halperin-Donghi, Guerra y finanzas, 155.
182. Friedman, The State and Underdevelopment in Spanish America, 185.
183. Burgin, Economic Aspects of Argentine Federalism, 189.
Making the State 159
or another.184 In Latin America, what little was paid appears to have come from those on the bottom, caste taxes being perhaps the most glaring example. While there are differences depending on the import basket and the specific rates, the general view is that customs taxes were also extremely regressive.185
An important factor here is that the relevant elites did not see the wars as threatening their social positions and thus did not have the incentive to permit greater political penetration. That is, the relevant elite did not appear to care which state ruled them as long as it was not markedly stronger than its predecessor. No state was alien to the elites' immediate interests.186 A transfer of political allegiance did not imply a change in property. Certainly in most cases, their concern appears to have been with protection from internal enemies, either ideological or, more commonly, class and racial ones. The maintenance of such internal control did not require an expansive and expensive state. In this, as in perhaps many factors, the Latin American elites were much closer to their Italian and Polish counterparts than to the English gentry or the Dutch bourgeoisie. For both these last two, fear of external threats, be it the Spaniards or "Popery," drove the elites to support high levels of taxation.
Interestingly, losing wars appeared to have created the base not for a more powerful state, but at least for a closer union between political goals and the interests of the dominant elite.187 Following the defeat by Chile, Bolivian elites appear to have been more open to paying for a state that could protect them as well as build the infrastructure needed for the exploitation of natural resources. Mexico's defeat in 1848 and the subsequent Treaty of Guadalupe produced a split among the agiotistas regarding the need for stronger government. Some members of the elite began to recognize the advantage of a better-integrated national economy and the need for a government to nurture it.188 For the first time, the state was perceived as something other than a massive feeding trough. The Liberals who took
184. Stone, An Imperial State at War.
185. Further research needs to be done on the composition of imports during the nineteenth century to determine the class distribution of payment of customs taxes.
186. I owe this point to Michael Mann.
187. The consequences of war could be disastrous. From 1870 to 1894, Peru went from having 18 millionaires, to none; from 11,587 classified as rich to 1,725; from 22,148 classified as well off to 2,000. Yet despite this looming disaster, the Peruvian elite seemed more concerned with resistance by peasants than invasion by Chileans (Mallon, Defense of Community, chap. 2).
188. Tenenbaum, Politics of Penury, 83-85, 116-17.
160    Blood and Debt
power in 1855 had the support of some of the wealthy who had begun to understand the potential benefits of a stronger state and they looked to the considerable wealth of the church for funds. The most interesting aspect of the ecclesiastical reforms is that in its battle with the church, the government enjoyed the support of a faction of the agiotistas who sought a securer basis for their loans. Thus, for the first time, the government had social allies supporting its encroachment on a part of civil society.189 In this way, at least, wars constituted the foundational first steps toward a state.
Toward the end of the century, several key figures who in previous times might have remained personalistic caudillos began to build the institutional basis for a state (Diaz in Mexico, Guzman Blanco in Venezuela, or even Roca in Argentina). I would argue that this was the result of a different institutional context of caudilloism. If during the first part of the century caudillos had secured their material and political base by controlling regions in conflict with central authority, toward the end of the century the road to power and wealth lay in expanding the capital's domain. The causes of this shift were rooted not in war or military competition, but in the requirements of capital and export production. As long as a hacienda economy writ large dominated the continent, it made political sense for the state to be weaker than its most powerful subjects.190 When that changed so did the goal of the armed political actors.
Differences at the Margin
Within the Latin American pattern there are, of course, relative exceptions, and these serve to prove the rule. The violence of the wars between i860 and 1880 clearly resulted in a much more powerful Argentine state. The key difference is that unlike in the 1810s and 1820s, the later Argentina did possess a semblance of a central government that could and did use the war both against Paraguay and the Indians to stamp out provincial opposition and impose uniform control over the entire country. More important, by the second half of the century, the central state had found its social ally whose interests it could serve: the export of meat and wheat to European markets required much more political and institutional infrastructure than
189. Tenenbaum, Politics of Penury, i6z-66.
190. I am borrowing the language from Bethell, Cambridge History, vol. 3, 663, but the original only refers to Uruguay.
Making the State I6i
the sale of salted beef to the slave owners of Brazil. Although the rural oligarchs of Argentina remained unwilling to pay for the new state, they were also unwilling to accept challenges to its authority. With this narrow support, Mitre and his successors were able to establish their domination.
The key to Brazil's relative unity would seem to lie in its avoidance of the struggles for independence. Neither the Brazilian economy nor its polity were destroyed by years of civil war, nor did the empire have to maintain an absurdly large military in order to establish its authority. Conflicts during the reign of Dom Pedro I helped resolve the intra-elite struggle between the "native" aristocracy and the Portuguese courtiers brought by Pedro's father and thereby consolidated the creation of a Brazilian political class. While there is considerable debate regarding the autonomy of this sector,191 there is no doubt about the existence of an "imperial" class that gave Brazil a particular coherence. The secessionist wars in the 1830s and 1840s helped consolidate this group. By the time of the War of the Triple Alliance, Brazil possessed enough institutional coherence to survive, if not necessarily prosper.
Although Chile experienced considerable political dislocation during the independence wars, its economy was not crippled by them, and may have even grown.192 More important, even before the rule of Diego Portales and certainly afterward, the Chilean elite displayed a remarkable cohesion.193 To what extent this was the result of the small size of the country, the concentration in a single city, the pervasiveness of dense interfamilial networks, or just sheer luck is the subject of debate. For our purposes, what is most important is that the Chilean state preceded war and thus was able to extract some benefits from it. Yet it is important to note that even the Chilean "exception" still fits the general Latin American pattern discussed earlier. Even as the state expanded, it did so without extracting from the domestic economy. Overall, the wars helped make Chile, not by a combination of blood and iron, but by allowing a fiscal improvisation fueled by duties on exports of commodities.194
Paraguay represents perhaps the most interesting exception to the Latin American pattern. Following the requisite period of instability following independence in 1814, the country was ruled by three dictators: Jose Francia until 1840, followed by Carlos Antonio Lopez, and then his son
191. Graham, "State and Society in Brazil."
192. Cariola and Sunkel, Un siglo de historia economica, 25.
193. Collier, Ideas and Politics of Chilean Independence.
194. Bethell, Cambridge History, vol. 3, 610.
162     Blood and Debt
Francisco Solano Lopez until Francisco's death in 1870. Francia created an all-encompassing state that dominated every aspect of public life and was completely controlled by him. The state owned all the land and largely managed all external trade. During the rule of the first Lopez, the state was involved in economic development, building some infrastructure and attempting to achieve self-sufficiency through the production of several industrial goods. Lopez fils encouraged military development to the point that the small country had arguably the strongest army in South America.195
The early Paraguayan state enjoyed a rare degree of autonomy. Unlike in the other Latin American countries, there existed an agent within the state that drove it to impose itself on the society.196 Francia served as the structural equivalent of an absolutist monarch, which helped ensure the continuance of Paraguayan autonomy.197 Francia's centurions allowed him to funnel all social resources toward his political apparatus.198 If we follow White, who claims that the rise of the military was a direct response to external threat, it would then appear that early Paraguay was perhaps the only example of the classic European variant of war-led state development. Paraguay could maintain this independence in part because the revenue that it could gather covered the state's needs. Unlike its neighbors, the Paraguayan state ran a consistent surplus during the entire postindependence period prior to the War of the Triple Alliance. This reflected the limited demands placed on it, but also the monopoly that the state enjoyed over almost all economic activity.
It is no longer possible to speak of Paraguay as the best example of anti-dependency and of a successful state-led development.199 Nevertheless, it is
195. Williamson, The Penguin History of Latin America, 2.73.
196. Pastore, "Trade Contraction and Economic Decline," 587.
197. White, Paraguay's Autonomous Revolution, 101-2.
198. The one consistent demand on the Paraguayan state during Francia's rule was the military budget. Despite the fact that the army never included more than two thousand men, the military absorbed an average of at least 64 percent of government expenditures during this period (White, Paraguay's Autonomous Revolution, 104). Much of the cost was associated with maintaining the military industries that supplied the armed forces. This relative self-sufficiency also helped protect Paraguay's international autonomy. While there are obviously different interpretations of the role of the Paraguayan army (107; Pastore, "Trade Contraction and Economic Decline," 591-92; Williams, Rise and Pall, 60-61), it is clear that it served to protect the state (or perhaps better said, Francia) from both external and internal enemies. Note that the apparent decline in military expenditure after 1840 in Table 3.2 may simply reflect a change in accounting practices (Thomas Whigham, private communication).
199. Whigham, Politics of River Trade, 83-84; Pastore, "State-Led Industrialization," 321-24.
Making the State 163
clear that the Paraguayan state was a very different institutional animal from its continental counterparts. The Paraguayan experience in building a much more powerful state apparatus than those of its neighbors even in the absence of war prior to the 1860s indicates again that while conflict does provide a stimulus and an opportunity, what matters is the organizational and political base of the state and its sources of support.
Conclusions: How Context Matters
In the end, wars did not make states in Latin America. The best that states could do was to survive wars or gain enough of their neighbors' territory to finance expansion. Nowhere did military action generate the kind of societal penetration seen in Europe. Latin America was caught in an inertial equilibrium: no class was powerful enough to impose its domination and no state was strong enough to enforce its control. The path to the modern state required one or the other.
Having argued that Latin America fought different wars and suffered from social, racial, and geographical barriers that precluded state development, the question then becomes why these were particularly acute on the continent. Europe also had divided elites, regional identities, and ethnic and class divisions. Why did these represent a more daunting obstacle in Latin America?
As discussed earlier, the forms of warfare were drastically different in the two regions. I would add, however, that differences in societal contexts were more important and that the analysis of these can make a more important contribution to our understanding of state making on the continent than any adjustment to the bellicist theory of political development.
First, regionalism in Latin America was encouraged by an important natural ally. The physical geography of the continent presented logistical and administrative obstacles only replicated in selected parts of Europe.200 Communications from the capital were uncertain and military support was irregular. Because of these problems, efforts to impose central authority in much of Latin America might be better compared with those of empires rather than of nation-states. The United States also faced geographical challenges,
200. One could even find a rough correlation between the success of state making in those regions and suitability of terrain. Certainly the plains of France made it easier to impose central authority than the mountains of southeastern Europe.
164     Blood and Debt
but in that country territory expanded much more in line with the capacity of the central state to administer. Imagine a United States having to fight an independence war across the Appalachians or trying to procure revenue from California to pay for the War of 1812. In Latin America, political institutions suited for a city-state were given empires to rule. We should not be surprised that they failed to do so.
The apparent (if illusory) cultural homogeneity of the continent also supported regionalism albeit in a perverse fashion. Given the absence of clear, strong distinctions across regions, the natural centrifugal attraction of the nation-state appeared less obvious. Thus, Latin American countries faced more significant natural obstacles while not enjoying the intrinsic attraction of differentiated pockets of cultural cohesion.
Ethnic divisions were also much more significant in Latin America, as they were not only accompanied by visible racial characteristics, but also supported by a legal and social system that institutionalized the minutest differences. Paris had to absorb Bretons and Provencal speakers, but it was much less successful with Basques. One can only imagine Spain's history with a sizable Morisco population, even with no ethnic differences involved. Once again, the relevant European model is Austria-Hungary, where internal ethnic/national divisions overpowered most notions of shared legacy or destiny. The presence of significant ethnic divisions, and their legal recognition, is perhaps the characteristic that most distinguishes the experience of Latin America from that of Europe.201 Again, we might best understand the birthing pains of independent states in Latin America by imagining them not as nations, but empires.
Finally, few elites could be as unruly as the European aristocracy that constructed states after the sixteenth century. Could the Argentinean or Peruvian elites really claim to be more fractious than their French or English counterparts? One major difference was the long European association of elite status with military prowess (broken in Latin America soon after the conquest). This relationship established a close link between martial competition and the viability of any elite group, which never existed in Latin America. The control of violence was an intrinsic part of elite functions in Europe; considering the state irrelevant was never an option. In Latin America, by contrast, political power was often secondary to economic con-
201. The experience of the United States is clearly relevant here. The counterfactual comparison might be an independent Confederacy having to construct a democratic state in the aftermath of slavery.
Making the State 165
trol, and this made the necessity to construct a state less urgent. European protostates were also helped by the institution of monarchy, which gave at least one family and its political network a very strong stake in the development of political capacity. Not even the Brazilian monarchy developed the congruence between individual and collective interests that may be so crucial at early stages of political growth.
Violence pervaded Latin American and European life during the development of these regions' respective states. There was violence between elites, between classes, between races, and between regions. Yet this did not generate the institutional development one might have expected from the European experience. The various regions of Europe competed with one another for supremacy and sovereignty, but they did so while re-creating a political map and not attempting to conform to a colonial geography. Ethnicities clashed, but they rarely were as hard to disentangle through territorial division. Elites might fight, but political entrepreneurs with monarchical legitimacy could impose institutional orders. With limited exceptions, Latin America did not possess the institutional or social kernel from which nation-states might have arisen, and wars did little to encourage their development. Where local conditions more closely approximated the European cases, war did provide the necessary institutional cement to secure the development of more powerful and stable states. In general, however, the military road to political development was not available on the continent.
The combination of weak central power and external economic direction is the defining characteristic of postcolonial states. The delegitimation of political authority as associated with the colonial power, the fragility of elite coalitions and lack of national cohesion or even identity, and the orientation toward a metropole and away from the interior and regional neighbors—all have characterized, in one form or another, the experiences of independent countries in Latin America, Africa, and Asia. Many have also experienced considerable violence without the benefits of the organizational development seen in western Europe. This pattern should make us wonder about the advisability of using such an idiosyncratic experience as the early modern western European one for the construction of universalistic paradigms. At the very least, the experience of Latin America should make us more curious about the particular circumstances that allowed states to flourish following the "Military Revolution" of the sixteenth and seventeenth centuries.
As discussed earlier, several special conditions allow wars to make states. The first of these is pressure on the state to respond to the financial challenge
166     Blood and Debt
of war through increased domestic extraction. There is no reason to expect states to undertake the political and organizational challenge of penetrating their societies if resources can be found more easily. Second, enough of an administrative core must already be in place that the state can use as a base on which to develop its strength. The chaos and violence of war do not provide the appropriate incubation for underdeveloped polities. Third, no political body can amass enough authority to coerce and extract without social allies. Domestic threats to sovereignty have to be resolved prior to "productive" conflict. Further research might test the relative significance of these three factors through their application to a variety of geographical and historical cases. Certainly these might help explain why it took nearly a millennium of violence for war to produce states in Europe. The case of Poland and the Balkans, suffering from both war and relatively weak states, would also merit attention.
The central lesson to be drawn from the Latin American experience is that we cannot assume that a state exists simply because the symbols of independence are there. States are not actors in and of themselves. They are shells—potentially powerful shells—but nevertheless hollow at the core. The machine of the state needs a "driver" able to use the stimulus provided by war to expand its reach and power. Without such a driver, whether it be state personnel, a dominant class, or even a charismatic individual, the political and military shell of the state has no direction. Without this direction, wars do not present opportunities for growth, but are mere challenges to survival. A fiscal system requires constitutional powers as well as a bureaucratic capacity to enforce them. This will not appear without an alliance between a political institution and a significant social sector. Without such an identification of interests, it is practically impossible for the state to grow, no matter the stimulus of violence.