Abstract
This article assesses the relative merits of the “reversal of fortune” thesis, according to which the most politically and economically advanced polities of the precolonial era were subject to institutional reversal by European colonial powers, and the “persistence of fortune” view, according to which early advantages in state formation persisted throughout and beyond the colonial era. Discussing the respective arguments, the article offers a synthesis: the effect of early state formation on development trajectories was subject to a threshold condition. Non-European states at the highest levels of precolonial political centralization were able to resist European encroachment and engage in defensive modernization, whereas states closest to, yet just below, this threshold were the most attractive targets for colonial exploitation. Since the onset of decolonization, however, such polities have been among the first to regain independence and world patterns of state capacity are increasingly reverting to those of the precolonial era.
Keywords
Among the best-known theses concerning the effect of colonial rule on patterns of political and economic development is that of the “reversal of fortune,” according to which the most urbanized and centralized non-European polities at the start of the modern era, such as the Inca and Aztec Empires, Bengal, or the Malacca Sultanate, subsequently became the most underdeveloped as a result of Western colonialism. Building on the arguments of world-systems and dependency theorists, for example, Acemoglu, Johnson, and Robinson argue that coercive labor institutions and natural resources made centralized polities attractive to European powers: they could subsequently be turned into extractive colonies for the purpose of securing rents from mining, slaving, and tax farming. 1 By contrast, the more sparsely populated territories of Australasia, Northern America, and the African Cape offered fewer possibilities for labor-extractive practices and were instead made into centers of settlement, inheriting European political and economic institutions. In this way the most powerful empires outside Europe on the eve of the modern age became peripheral zones of the world economic system, whereas formerly marginal territories in the Americas and Antipodes became linked to its core and sites of productive growth.
However, in recent years a contrasting argument, which we might call the “persistence of fortune” thesis, has maintained that the most centralized non-European states of the early modern era were the most likely to resist European encroachment, undergo defensive modernization, and experience “catch-up” growth in the late twentieth century, citing Japan, China, and Turkey as examples. 2 Historical studies including Michael Mann’s Sources of Social Power, Victor Lieberman’s Strange Parallels, and Francis Fukuyama’s Origins of Political Order argue that non-European powers such as China and the Ottoman Empire were the first to implement measures such as military conscription or merit-based examination, whereas Japan after the Meiji Restoration offered a path to defensive modernization subsequently taken up by other Asiatic powers, linking with an earlier literature on the developmental state. 3 Advocates of the “persistence” view point out that the countries experiencing the fastest rates of economic growth in the era following decolonization were almost universally non-European states with long state histories, capable of implementing policies of state-directed development, many of whom, including Thailand, China, or Iran, had already resisted attempts at European colonization. The persistence of fortune argument thus contends that non-European powers that entered the modern age strongest also left it the strongest.
These “reversal” and “persistence” arguments would appear to generate flatly contradictory expectations regarding the relationship between political development in the early modern era and the strength of political institutions today. According to the reversal view, areas of the world that hosted centralized empires at the start of the modern era should now contain the poorest and most fragile states; according to the persistence view, such areas should today contain politically integrated and high capacity regimes. That both persistence and reversal arguments are supported by rich historical, sociological, and economic literatures only makes the contradiction more acute.
This article attempts a reconciliation by arguing that each is correct—subject to a threshold condition. If a non-European polity had sufficient state capacity at the start of the modern era effectively to resist colonial rule, defensive modernization occurred; but if state capacity was sufficient only to extract labor and taxes, the result was colonial exploitation and institutional reversal. In the first set of cases persistent states formed as precolonial regimes were able to mobilize armies and defend their borders. However, in those cases where institutions existed only to facilitate tax collection or coerced labor, “arrested states” emerged as European colonists targeted these polities for control in the sixteenth and seventeenth centuries, interrupting indigenous processes of state formation. Further, since decolonization, because of the role of early states in creating indigenous elites capable of mobilizing for independent rule and engaging in effective postcolonial state building, the advantages of early state formation have reemerged. Postcolonial regimes with a precolonial state inheritance have since experienced the most rapid increases in postcolonial state capacity and the greatest success in state-directed development, with patterns of state formation around the world increasingly reverting to their precolonial pattern.
The rest of the article proceeds as follows. First, I consider the theoretical basis for the reversal and persistence hypotheses, and the evidence of existing studies. Second, I outline the threshold condition hypothesis, and suggest a tripartite taxonomy of non-European states between “persistent,” “arrested,” and “new” polities, depending on how the exposure to European colonialism affected indigenous processes of state development. I then show how the strategies of European colonial powers led highly centralized regimes to become colonized in the first phase of colonial expansion, whereas less centralized regimes experienced colonial regime only during the final phase of colonial rule from the 1880s to 1960s. Finally, I argue that in the postcolonial era, patterns of state formation have steadily begun to revert to the precolonial pattern, as polities with more centralized precolonial states were among the first to reassert their independence, and experienced more rapid improvements in state capacity and growth through state-directed development.
The Reversal of Fortune Thesis
A large number of empirical studies have examined the negative consequences of colonialism, and the reversal of fortune thesis in particular. Acemoglu, Johnson, and Robinson utilize data on city size and population density in 1500, from McEvedy and Jones’s Atlas of World Population History, to show that the most densely settled areas and largest urban centers outside of Europe in the period before 1500 were to be found in the indigenous polities of the Americas, and in South and Southeast Asia and that these became the most underdeveloped parts of the world by the end of the colonial era (Figure 1). 4 This hypothesis is supported by a range of studies from Latin America, including the findings by Melissa Dell regarding the mita in Bolivia and Peru; 5 by Engerman and Sokoloff with respect to the labor-rich areas of the Andes and Mesoamerica; 6 by Marcus Kurtz and Hillel Soifer regarding the relative failure of postcolonial state building in feudal Peru versus its success in remote Chile; 7 and by Acemoglu, García-Jimeno, and Robinson’s comparison of gold mining and nonmining areas of Colombia. 8
The Reversal of Fortune Thesis: Population Density in 1500 and GDP per Capita in 1950.
At the core of the reversal argument is that in the areas that were targeted first, a combination of labor-extractive practices and Western colonial rule undermined the indigenous development of state capacity. 9 These varying arguments for the weakness of the postcolonial state can be summarized into three main categories: the limitation of scope, the displacement of indigenous state capacity, and the creation of unviable states.
The limitation of scope. First, colonial regimes sought only to extract resources and tax, rather than provide public goods, fight wars, or implement a wide-ranging platform of societal transformation and behavioral regulation, in the fashion of postrevolutionary regimes. 10 The colonial state was therefore a “thin” state by design; a “gatekeeper” state based on the coast, with interest in ruling and extracting natural resources rather than institution building, and did not establish the state’s presence outside the administrative capital or those areas that contained valuable natural resources. Nor did colonial regimes generally invest in building tax infrastructure, such as registration of income or cadasters of land, opting instead for “easier” sources of revenue such as tariffs on trade or the use of commodity marketing boards. As a result, postcolonial regimes were left without ready sources of revenue, and without any means of accomplishing the ambitious transformative schemes of postcolonial leaders other than by resorting to the same distortionary apparatus and the recourse of international debt markets. 11 Further, though the labor and resource-extractive nature of colonial regimes may have entailed establishing a minimal layer of state capacity around the commercial operations of the colonial elite, it also meant that legal institutions were not extended beyond this group, and the failure of colonial institutions to establish broader legitimacy resulted in pervasive noncompliance, resistance, and insurgency outside of the metropole. 12
The displacement of indigenous capacity. Second, colonial regimes relied for their functioning on a cadre of administrative officers provided by the colonial core state, such that on independence many postcolonial regimes found themselves facing a severe shortage of capable bureaucrats. This lack of administrative capacity has been reinforced by the tendency for postcolonial states to rely on their former colonial rulers and international donors and multilateral agencies for technical support, military defense, finances and soft loans. The ready availability of external financial support has hindered the responsibility to develop domestic fiscal capacity, while the availability of external technical assistance has inhibited the development of comparable bureaucratic capacities domestically. 13
The creation of “unviable” states. Finally, whereas European states formed endogenously to the conditions of state viability, that is, geographic defensibility, ethnic cohesion, and access to resources, colonial regimes were formed arbitrarily by European powers with limited respect for existing ethnic boundaries or natural geographic conditions of governability. As a result, colonial regimes unintentionally left behind unviable states, either because ethnic fractionalization led to weak collective identities and clientelism, 14 or because geographic conditions were unfavorable to territorial consolidation. 15
The Persistence of Fortune Thesis
The reversal of fortune thesis provides a strong basis for understanding why persistent patterns of underdevelopment have occurred in areas of the world that, until the modern era, seemed most advanced on indicators of social and political development. 16 Yet in recent years a contrasting “persistence of fortune” argument has been developed among both historical studies examining the relationship between long-run processes of state formation and recent developmental states, and econometric studies that measure early state inheritance and its consequences for contemporary state capacity and economic growth. In the former camp are historical works, such as Victor Lieberman’s Strange Parallels, which compares processes of state formation across East and Southeast Asia, Europe, and Russia, and studies of state formation such as Francis Fukuyama’s The Origins of Political Order, which links the study of historical state formation to a more established literature on developmental states by authors such as Peter Evans and Meredith Woo-Cumings. 17 If studies of the reversal thesis largely center on Latin America and India, it is perhaps no coincidence that scholarship of persistence has focused more on Africa and East Asia: regions where precolonial and postcolonial continuities are easier to identify. In the African context, Gennaioli and Rainer have found that countries with a greater proportion of centralized ethnic groups have more paved roads; a greater percentage of infants immunized against diphtheria, pertussis, and tetanus; lower infant mortality; a higher adult literacy rate; and greater attainment of schooling. Michalopoulos and Papaioannou have investigated the role of precolonial institutions in shaping comparative regional development within African countries, using nighttime satellite density maps to measure electricity output around precolonial ethnic boundaries. 18 At the global cross-country level, Bockstette, Chanda, and Putterman find similar patterns, having developed an index of “state antiquity” that measures the stock of accumulated time over which a supratribal polity has existed within the present-day boundaries of each of 149 countries; they find that an early and durable history of political organization is a powerful predictor of modern rates of economic growth (Figure 2). 19
The Persistence of Fortune Thesis: GDP per Capita Growth (1950–2006) and State Antiquity in 1500.
Beyond calling doubt on the negative treatment effect of colonialism, the persistence argument offers at least three potential mechanisms explaining why postcolonial outcomes would reflect the strength of precolonial states:
The persistence of “mental models” of governance. First, one argument is that postcolonial governing practices based on neopatrimonialism or sultanism are a legacy of precolonial political orders and the familistic norms that governed pre-state or early-state society. Scholars of African politics such as Michael Bratton and Nicolas van der Walle have argued that on independence, postcolonial leaders reverted to these patterns because the brevity of formal colonial rule had failed to establish alternative “legal-rational” modes of legitimation. 21 By contrast, where colonial regimes built on existing models of bureaucratic authority, as in Vietnam, Taiwan, or Korea, postcolonial regimes were able to maintain strong, insulated states with limited opportunities for patron-client relations. Similarly, where colonial regimes preserved longstanding traditional modes of legitimation, as in the kingdom of Morocco or Malay Kedah dynasty, those modes provided institutional stability in the postcolonial era, such that postcolonial state trajectories have continued processes of state formation from the precolonial era.
Endogenous colonialism. Second, scholars have called into doubt the ability to identify a clear colonial “treatment effect,” because whether countries were colonized at all was endogenous to prevailing levels of state capacity; much of the variance in state capacity attributed to a colonial legacy or to a mode of colonial administration (e.g., direct versus indirect rule) may simply reflect differences in the ability of the precolonial state to resist foreign incursion. 22 In other words, the treatment effect of colonial rule may well have been negative, but it was only administered in territories that were already occupied by weak states, not in areas subject to stronger and more centralized precolonial regimes. This is obviously the case for states such as China or Japan that resisted colonial rule actively and by and large successfully, and more ambiguously in areas sufficiently resistant to merit a strategy of indirect of governance. Empirical studies of colonial legacies have often neglected the need for a robust identification strategy to isolate the exogenous component of colonial rule. 23
The preservation of abolition of tribal institutions. Third, along with colonial regimes, precolonial states played an important role in preserving or abolishing local tribal chieftaincies, which when preserved led to negative consequences for postcolonial states due to widespread clientelism and the balkanization of legitimate authority. In Africa, for example, studies suggest that positive developmental consequences are associated with indirect rule in the case of the larger and stronger precolonial states such as Buganda or the Ashanti Empire, but not where indirect rule meant devolution of government to a fragmented network of local chieftaincies. 24
In making these arguments, the persistence view makes a number of objections to the thesis of colonial underdevelopment maintained by the reversal of fortune literature. First, theories that root the low state capacity of non-Western countries in the legacy of colonialism lack a “counterfactual conditional” 25 —a hypothetical alternative scenario in which the treatment, in this case colonial rule, had not been applied. 26 Second, the critique of colonialism thesis “reads history sideways,” by comparing the current state capacity gap between former colonial powers and postcolonial states in order to conclude that the latter were at some point deinstitutionalized; yet, as the reversal thesis asserts, at the time of early colonization in the sixteenth and seventeenth centuries, many colonial powers were themselves relatively weak states, and augmented their capacity in the eighteenth and nineteenth centuries for reasons that were largely endogenous to Europe: interstate competition, socioeconomic modernization, and technological innovation. 27 Third, the persistence literature asserts that we cannot attribute the weakness of the postcolonial state to the colonial past alone, because this colonial state was not simply a product of European design but heavily endogenous to preexisting precolonial institutions. European powers did not discover the world as a tabula rasa but as a series of small and large states, equally struggling with challenges of resource mobilization, military modernization, and the consolidation of their administrative reach. 28 Therefore, both the strategies of foreign rule and the resultant levels of state capacity were determined by preexisting patterns of institutional development. 29 Whether a territory fell to European rule at all depended on the capacity of local rulers to resist. 30 Last, the “treatment” of colonial rule was not constant over time and space: The conquests of Mesoamerica or the Andean region by the conquistadors in the 1500s were clearly different processes (e.g., in respect to the degree of bureaucratization, the relationship to the indigenous population, or nature and goals of the governing actors) from the administration of Hong Kong under the British Empire, say, or the administration of Korea by Japan, as Atul Kohli has argued. 31 Earlier forays by European powers were commercially motivated and extractive, and may have had especially negative institutional consequences as the growth of the slave trade during the seventeenth and eighteenth centuries, 32 yet later colonial endeavors were motivated more by the logic of interstate competition, and could involve significant investments in building bureaucratic infrastructure. 33 European powers, moreover, were not the polities engaged in colonial acquisition at this time, and the effects of non-European imperialism also appear ambiguous. 34
Toward Synthesis
How are scholars of colonialism to deal with the apparent contradiction between these two competing theories? In this article, I argue that both the persistence and reversal arguments can be seen as correct and, moreover, consistent with the observed data, once we introduce a discontinuity threshold. This threshold is whether a state had achieved sufficient institutional development before the arrival of Europeans to allow it successfully to resist colonization (in which case the encounter with colonial powers gave rise to defensive modernization), or not, in which case a territory was conquered and subjected to institutional reversal. The path of defensive modernization was that taken by ancient polities such as China, Persia, and Turkey, which responded to the European challenge by reforming their fiscal and military infrastructures, securing their territory, and mobilizing against foreign rule. However, states that fell just below this threshold suffered institutional reversal; and the degree of reversal was sharpest among those that were first to be targeted by colonial powers during the initial phase of colonialism, led by the mercantile and extractive impulses of the Spanish conquistadores and the English, French, and Dutch East India companies. That was the path taken by the Aztecs, the Inca, or the sultanates of Bengal or Malacca, as well as the few early African trading posts, such as Senegambia or the kingdom of Kongo, which became early bases for slave raiding into the African interior. Finally, a separate category must be created for non-European regions that were not suitable for European settlement, in which much of the population remained in pre-state societies—such as much of the remainder of sub-Saharan Africa, upland southeast Asia, and Central Asia. Those were the last to be fought over by imperial powers, largely in the late nineteenth century, because of the difficulty of conquest and the limited opportunities they afforded for labor exploitation. Here again, there was institutional persistence, if a persistence of absence: non-state societies were left by colonial powers with the nominal authority of a central postcolonial bureaucracy, yet, the institutions of tribal government still largely in place.
This pattern can be seen in Figure 3, which shows the century of conquest by European powers and the state antiquity index assessed in 1500. As suggested by the persistence of fortune argument, the most established states on the eve of the modern era, such as China, Japan, and Turkey, were those most likely to resist European colonialism and retain their independence through defensive modernization. However, in accordance with the reversal of fortune argument, it can be shown that the first territories to be targeted by European colonial powers were indeed established polities offering possibilities for extractive surplus, and that conquest of pre-state societies was left until late in the colonial era.
Century of Conquest by European Powers and Average Level of State History in 1500 (Non-European Colonies Only).
As a result of this interaction, the encounter with European colonialism can be seen as having created three kinds of state in the modern world: persistent, arrested, and new colonial states. Which of these outcomes arose was determined by the level of state formation achieved before colonial influence. States at the highest level of political development at the point of contact with European powers responded to the colonial challenge through defensive modernization: these states “persisted,” or remained independent, and continue to exhibit the highest levels of state capacity today, as measured by such conventional metrics as fiscal depth, the rule of law, or the quality of the bureaucracy. Yet a second group of polities were sufficiently developed by the early modern era to offer opportunities for labor-extractive practices such as taxation and mining, but not sufficiently developed in their fiscal or military infrastructures to resist foreign rule. These were the first to be targeted by European colonialists in the sixteenth and seventeenth centuries, resulting in an interruption of their indigenous institutional development. Finally, a third group of polities were created entirely under colonial rule, in areas in which there was little or no indigenous state capacity beforehand. These territories were targeted only in the final phase of European imperialism, and combine a postcolonial central state with almost complete lack of state capacity outside of the capital territory.
To this list we may add a fourth regime category of regime—“settler states”—in which climatic and disease environments were suitable for colonial settlement, and where European political and economic institutions could transplanted along with European populations. Such settler states include the cases of North America, Australasia, southern America, and the African Cape; however, because the purpose of this article is to explain the effect of colonialism on non-European polities, these may be considered a category apart, even though such polities had a clear and negative effect on the indigenous institutions they found in place via forcible and brutal displacement of indigenous populations and institutional reversal of indigenous political structures. 35 A summary of the three-state typology is provided in Table 1.
Persistent states are polities that survived the era of European colonialism as independent powers, and indeed were more likely to have been strengthened by their interaction with Europeans as a result of engaging in defensive modernization. Persistent states were already long-standing powers at the start of the modern era (circa 1500). Examples of persistent states with long state histories were China, Japan, the Ottomans (Turkey), Morocco, Persia (Iran), and Siam (Thailand).
Arrested states were already centralized polities at the point of contact with European colonialism, but were insufficiently advanced technologically or organized bureaucratically to resist foreign rule. Their state history scores were high at the outset of the modern era, but not as high as the persistent states, and were likely to have been covariate with weaker bureaucratic penetration, fiscal capacity, and military readiness. Arrested states were the first to be targeted by European trading companies and conquistadors in the sixteenth and seventeenth centuries. Such reversal states include India and Pakistan (under the Mughal Empire), Indonesia and Malaysia (Malacca sultanates), Mexico (the Aztecs), Peru and Bolivia (the Incas), Cambodia (the Khmers), and Myanmar (Burma).
New states are found where colonial powers encountered pre-state societies in areas unsuitable for settlement by colonists, and in which there was limited indigenous state capacity, as a consequence of which colonial rulers developed entirely new political institutions. These include many of the states of sub-Saharan Africa and Central Asia.
Persistent, Arrested, and New States.
Case Studies in State Trajectories under Colonial Influence
The optimal method of case comparison is that of the natural experiment, in which assignment to a treatment or control group occurs via a process that resembles random assignment. 36 This section will therefore illustrate the difference between persistent and arrested trajectories by considering two cases in which colonial rule and deinstitutionalization were driven by geographic and temporal circumstances: Burma and Siam (Thailand). These are supplemented by an “ideal-typical” illustration of a purely new colonial state from within the region, in the form of Papua New Guinea, a case that is ideal-typical according to the defining features of such a polity, namely the absence of prior state formation and the resultant construction of state attributes entirely under the impetus of colonial administration.
Persistent States—the Case of Siam
Thailand, until the modern era the kingdom of Siam, was governed by a succession of kingdoms from the thirteenth to the eighteenth centuries. Although Siam began as a relatively weak and patrimonial state, repeated warfare between Siam and its neighbors—above all Burma, against which the kingdom fought no less than twenty wars in the years between the mid-sixteenth and the mid-nineteenth centuries—forced a greater degree of centralization. By the nineteenth century, France had succeeded Burma as Siam’s main challenger on the peninsula thanks to its expanding empire in Indochina, and the state engaged in extensive defensive modernization in order to prepare militarily for this threat. Under King Rama V (who reigned 1853–1910) a Royal Military Academy was founded along western lines (1887), the executive branch was restructured into ministerial government (1888; including a new ministry of war), a Royal Naval Academy was founded (1898), and universal conscription introduced (1905). The Siamese kingdom also sought to increase the reach of its legal and judicial system over its population and rationalize its method of taxation: to this end the Thai Survey School was established in 1882, and a first land cadaster conducted several years later in 1896. In 1932, a “breakthrough” coup by the army established a constitutional monarchy, and during the Second World War Thailand declared war on France in order to regain control over Laos.
Critically for the purpose of identification, Siam’s survival as an independent state was contingent on historical circumstance. It is unlikely the Siamese state would have survived a lengthy anticolonial war such as endured by Burma (at war with the British Empire on three occasions from 1824 to 1886) and Vietnam (at war with France from 1858 to 1886). By virtue of geography, however, Siam was not located next to the base of the East India Company in Bengal, as was Burma, nor to the French base of operations for its conquest of southeast Asia, which began in northern Vietnam following the Sino-French war and expanded from there to southern Vietnam, Cambodia and Laos. Having persisted through to the late nineteenth century, however, the kingdom was able to rationalize its administration and modernize its army in order effectively to ward off further attempts at colonial encroachment.
In common with most other persistent states—notably Japan, Nepal, Bhutan, and Morocco—Thailand has survived as a constitutional monarchy, and was among the only regimes of southeast Asia to successfully suppress revolution during the Cold War era. Persistent states represent many of the surviving monarchies of the world today, largely because of the greater legitimacy accumulated by historical sovereigns; monarchs artificially imposed by colonial powers, such as the postcolonial kings of Egypt, Iraq, or Libya, experienced a short duration in office after the retreat of colonial rule. Persistent states are also responsible for the “autocratic legacy” of early statehood, as the stock of legitimacy and compliance built up by absolute monarchs has allowed for the construction of more highly centralized and noncompetitive political orders in the contemporary era. 37
Arrested States—the Case of Burma
An example of an arrested state would be Burma, since 1989, the Union of Myanmar. During the early modern era, Burma was one of the paramount powers of Indochina, fighting repeated wars against neighboring Siam, and repulsing four invasions by Qing Dynasty China in late eighteenth century. In the nineteenth century, the polity’s strongest regional rival became the British Empire, then consolidating control over neighboring India. After losing a first war against the East India Company in 1826, Burmese elites made attempts to modernize their polity, including the introduction of a salary system for the bureaucracy, fixed judicial fees, penal laws, increases in direct taxation, and comprehensive military reform. However, geographic proximity to the capital of the British Raj made Burma the primary target for further wars, such that they were unable to stave off successive British attempts to wrest control over Burmese territory, and the remainder of Burma was incorporated into the British Indian Empire in 1886 despite some of the most costly wars of the British Raj.
Under British colonial rule, Burma’s indigenous monarchical institutions, as important a linchpin of political legitimacy as in neighboring Siam, were not preserved. The king and his family were instead sent into exile in India, and a colonial administration was formed in the new capital of Rangoon, in place of the palace capital at Mandalay. In most arrested states, the colonial era saw the construction of a parallel administrative apparatus that marginalized the existing, precolonial structures, and in many cases marginalized its former administrators and elites. This was the case in Burma, where Anglo-Burmese dominated the civil service. The result was a divided society, a strong anticolonial movement, and postindependence politics marked by conflict between radical factions demanding revolutionary change and military officers seeking to restore political order. With the existing sources of political order displaced, the postcolonial history of many arrested states has often been unstable. In Myanmar, a 1962 military coup established a junta that ruled the country subsequently, punctuated by widespread though unsuccessful civic uprisings in 1974, 1988, and 2007.
New States—the Case of Papua New Guinea
An ideal-typical example of a new state, lacking any precolonial political centralization, would be Papua New Guinea: Formed under colonial administration by the merger of Papua and New Guinea in 1945, the country is one of the most heterogeneous in the world, with hundreds of ethnic groups and over eight hundred languages; and the country’s entire process of state formation occurred under colonial tutelage, until the country’s late independence in 1975. Everything that defines Papuan statehood, from the capital (founded in 1873 and named after a British captain), to the language (a form of patois English), to the legal system (imported from English common law), was constructed under the aegis of its colonial rulers, and in this respect it is an ideal-typical new colonial state. Papua may represent an extreme case; however, it typifies the dynamics of many postcolonial states in sub-Saharan Africa, Oceania, and Central Asia in which French, British, and Russian administrators attempted to craft a metropolitan state, in the image of the colonial power, grafted atop a tribal, precolonial society. Because of the postcolonial state’s weak or absent penetration into local communities, it is typically weak and detached, with limited territorial control outside the capital, and clientelism toward competing tribal groups and affiliations.
Understanding the Nonmonotonic Relationship: Mercantile and Imperial Phase Colonialism
Why is there a nonmonotonic relationship between indigenous state formation and the effect of European colonialism? In order to answer this question, it is important to understand not only the interaction of indigenous state formation with colonial rule, but also the temporal sequencing of the colonial enterprise. The decision of European powers to conquer or not to conquer a territory can be seen as a function of two variables: the cost of conquest, determined by the degree of indigenous state capacity and the ability of its elites to withstand military attack; and the payoff to the imperial ruler of victory, determined by the ability to extract rents through such practices as mining and tax farming. The cost of conquest rose linearly with the degree of indigenous state building, as a high degree of state organization, comparable to that of European powers, was required to achieve the most effective military defense through practices such as conscription and a sufficient fiscal extraction to invest in gunpowder weaponry. However, taxation and corvée labor exist even in very early states, providing opportunities for labor-extractive practices in polities with weak ability to resist external aggression. These were the polities most attractive to European colonialism, as they offered possibilities for extraction without also possessing the military institutions required to ward off foreign invasion.
The Mercantile Phase of Colonialism
Figure 4 illustrates the nonmonotonic relationship regarding the date of conquest (or nonconquest) by European colonial powers. In the first or “mercantile” phase of European colonialism, approximately from the fifteenth to the nineteenth centuries, colonial ventures were led by private enterprise. This enterprise was either formal, as with the Dutch, English, Portuguese, Swedish, Danish, and French East India Companies, or informal, in the case of the Spanish conquistadors. Under this system, the relevant decision to engage in the colonial enterprise was a logic of cost and benefit; whether the expected return on investment of conquering a territory would exceed the expense of manpower and resources required to engage in such an acquisition. (The directors of the English East India Company initially used the term “imperialism” in contempt, as engaging in expensive territorial ventures without regard to economic benefit.) European colonizers were therefore selective in their targets: smaller powers such as the Portuguese, Swedes, and Danes would rather avoid military expeditions altogether, preferring to establish trading posts along the main routes for sugar, slaves, and spices, though even the colonial companies of the larger powers preferred to adopt peaceful means where possible, eschewing explicit conquest in favor of establishing strategic trading forts and treaties with local rulers. To the extent that trading companies engaged in war, it was more often with one another; and not with the local potentates whose cooperation they required.
Two Phases of Colonialism.
Where early colonizers did engage in territorial acquisition, priority was given to those holdings that allowed for opportunities for profit through mining, tax farming, or slaving. This greatly restricted the range of potential ventures that might be reasonably entertained, as the armies of the Ottoman, Mughal, and Chinese Empires were vastly greater in manpower and resources than the largest of Europe’s domestic armies, to say nothing of the resources that could be mobilized at great distance by a collection of privateers, and limited the range of feasible ventures to a restricted number of sites. Even the largest European powers during this phase of world history were minor players in the global system, economically, diplomatically, and militarily. European colonialists needed to be selective in their choice of targets, opting for acquisitions in areas that offered the weakest resistance relative to the potential benefits available. In practice, this necessity meant engaging in wars against established polities in which institutions of slavery and vassalage had already been established by local elites, but which did not yet have sufficient knowledge of Eurasian military technology or bureaucratic organization in order to mount effective resistance. Such polities were found in the Americas and in South and Southeast Asia: in the Americas because of the absence of familiarity with cavalry and gunpowder weaponry, as well as the devastating impact of European disease; and in Southeast Asia because of a relative lack of political centralization. 38 Thus the first colonial acquisitions were in the territories of the Inca and Aztec empires, and in the weaker sultanates of the South China Seas. Yet despite these acquisitions, European colonial companies inserted themselves into a world economy that was still dominated by Asiatic powers. Fifty percent of all silver mined from the Americas by the Spanish made its way to China, and to the extent that the Spanish acquired posts in Formosa and the Philippines, it was to be closer to their main client. 39 Similarly, much of the trade along the spice route operated by the Dutch and Portuguese from Indonesia to Europe via Malabar and the Arab Gulf occurred internally; only a small amount in fact originated or terminated in Lisbon or Amsterdam. Even the African slave trade remained dominated by Asiatic and internal African demand until the eighteenth century. 40 During the mercantile phase of colonialism, European companies operated in a world of Asiatic powers and required their cooperation: a Dutch colony on Taiwan was destroyed in 1662 when 50,000 Chinese soldiers turned up under the generalship of Koxinga, and forts operated by the Danish, French, and Dutch East India Companies were seized by indigenous rulers when they displeased their local hosts. 41
During this mercantile phase of European colonialism, the established Eurasian polities such as Persia, China, and Vietnam were left undisturbed; likewise untouched were those areas still populated by pre-state societies, whether in the Amazonian interior, upland southeast Asia, or much of Africa. The latter were difficult to subjugate and offered little obvious benefit; no sooner had one tribe been pacified, than another might erupt in rebellion. Thus the simplest policy in dealing with pre-state societies was a logic of displacement and extermination, as was carried out ruthlessly in the “neo-Europes” of the Americas and Australasia. 42 Yet even in these cases of European settlement, the land had to be comparatively sparsely populated by indigenous peoples; the Americas had been decimated by the impact of European disease, and the African Cape not yet reached by the great Bantu migration.
The Imperial Phase of Colonialism
A second phase of European expansion abroad began from the mid-nineteenth century, which we may refer to as the “imperial” phase. During this era, colonial ventures were no longer being led by private actors, but rather by governments that sought to maximize territorial control in their competition with other states. The English East India Company was brought under parliamentary authority in 1773 and finally nationalized in 1858, the French East India Company in 1769, and the Dutch in 1798, and decisions over statecraft transferred to national administrators. As a result, military expeditions were frequently conducted despite a lack of obvious economic benefit, and in areas of limited economic value but nonetheless perceived geostrategic importance. This is illustrated by the shift in utility curve shown in Figure 4. Overseas colonies became an extension of national statecraft, and were administered not with the goal of profit but with the aim of enhancing the state’s competitive position against other states. In the nineteenth century, European powers began waging wars against strong polities that they had previously left undisturbed; the British fought a new round of wars in South Asia, primarily against Burma, Afghanistan, and Nepal; and the French fought new wars in Southeast Asia, claiming the ancient states of Vietnam and Cambodia. These were difficult and costly conflicts, which a fully commercial logic would have avoided. It took three Anglo-Burmese wars to subjugate Myanmar, the first of these making the longest and most expensive war in British Indian history, while Afghanistan was never successfully conquered. It took France over thirty years to gain territorial control over Vietnam, from the first expedition by Rigault de Genouilly in 1858 to the conclusion of the Tonkin campaign in 1886. On the other hand, not only a number of “strong” states became targets of imperialism, but also the vast remaining areas inhabited by pre-state societies that had previously remained untouched. For the first time European powers began to acquire territory in Africa for administrative ends, rather than simply engage trading settlements on the coasts for the end of securing slaves; in this way Germany declared its possession of Togoland, the Cameroons and South West Africa in 1884, and King Leopold of Belgium claimed Congo as his personal possession the following year. The conquest of pre-state societies also came at high military cost, due to repeated rebellions such as the Dervish, Herero, and Namaqua uprisings. A similar logic of territorial consolidation dictated that Russia, having completed its administrative hold over the tundras of Siberia, begin subjugating the nomadic peoples of Central Asia by acquiring the territories that were to become Kazakhstan, Kyrgyzstan, and Uzbekistan; both Russia and Britain eyed Afghanistan. The “great game” in Central Asia and the “scramble for Africa” were the defining events of the last phase of imperialism: a competition among colonial powers for territorial control with little regard to economic benefit and often at substantial net economic cost.
An implication is that it is difficult to specify a single threshold of state capacity at which state persistence was likely, that threshold being a relational one, relative to the state capacity of European colonial powers. Polities such as Ottoman Turkey and Japan, which made contact with European influence early and initiated attempts at defensive modernization during the mercantile phase, were able not only to persist as independent states but to become colonial powers in their own right. Later modernizers had a higher threshold to attain, and initial success at repulsing European colonial projects could be met with defeat later: Ethiopia could resist Italian encroachment at the battle of Adwa in 1896, but not during the invasion of 1937. The stronger regimes of eighteenth-century India, such as Mysore, Travancore, or the Marathas, won their initial battles against European powers but could not defeat the East India Company of the nineteenth century. An important difference, however, lies in the phase of conquest: early colonial conquest, as implemented by mercantile companies, led to extractive institutions, whereas later colonial rule, during the imperial phase, was driven by the logic of interstate competition.
Conclusion
In recent years two apparently inconsistent literatures have developed regarding the long-run legacies of colonialism, which can be termed the reversal and persistence of fortune theses. At first glance, these two literatures seem to generate contradictory expectations regarding how colonial rule interacted with trajectories of early state development in order to produce the variation that can be observed in the world today. This article has assessed the relative merits of the reversal thesis, which holds that the most developed states of the precolonial era were subject to institutional reversal by European powers, and the persistence view, which holds that the states most established in the early modern era, such as China, Japan, or Ottoman Turkey, were the most likely to resist European encroachment, undergo defensive modernization, and experience “catch-up” growth in the late twentieth century. The reversal and persistence perspectives can be reconciled by introducing a threshold condition, according to which the effect of early state formation on a country’s development trajectory depends on whether it allowed for effective resistance to colonial rule. Non-European states at the highest levels of precolonial political centralization were able to resist European encroachment and engage in defensive modernization, whereas states closest to, yet just below, the threshold proved the most attractive targets for colonial rule.
Colonial history, however, is not destiny. Although the effects of colonial rule may have been especially disruptive among those polities that were relatively centralized at their point of contact with European colonizers, and in particular for those polities subject to institutional reversal during the mercantile and extractive phase of European colonial rule, the postcolonial era has seen a gradual reassertion of precolonial legacies. Since the onset of decolonization, among polities that fell under colonial influence, an important difference remains between arrested states that inherited a precolonial state legacy and new states that did not: there has been a “reversal of the reversal” by which colonies with a precolonial state inheritance fostered indigenous elites capable of mobilizing for earlier political independence and experienced more rapid increases in state capacity in the period since decolonization. Countries that acquired their independence first during decolonization, such as India, Ethiopia, or Malaysia, were those in which indigenous elites were able to mobilize against colonial powers, whereas independence came much later to colonies in which challenger elites did not exist as a result of historical states (Figure 5).
State History in 1500 and Date of Independence during Decolonization.
Thus while countries with relatively higher state capacity may have proved more attractive initial target for colonialism, they ultimately proved more difficult to retain. Colonies with higher levels of precolonial state capacity have had more success in building state capacity during the postcolonial era than those with weak or absent precolonial state institutions. As a result, with the passage of time the effect of colonialism has begun to reverse, and patterns of global political development have begun to resemble once more those of the precolonial period.
Footnotes
Acknowledgements
The author would like to thank James Robinson, Louis Putterman, Steven Wilkinson, and Daniel Ziblatt for useful feedback and comments on this article, as well as attendees of the workshop on legacies of colonialism at The Ohio State University.
Declaration of Conflicting Interests
The author declares no potential conflicts of interest with respect to research, authorship, and/or publication of this article.
Funding
The author received no financial support for the research, authorship, and/or publication of this article.