From Conquest to Centralization

Huexocingo Codex

By Francisco Garfias and Emily Sellars

This year marks the 500th anniversary of the siege of Tenochtitlan, the final battle of the Aztec-Spanish War. It ended in the defeat of the Triple Alliance/Aztec Empire and the eventual establishment of Spanish colonial rule in Mexico. Few events occupy as important a place in Mexican or even world history.  Following the military victory of the Spanish and their Indigenous allies, Tenochtitlan (now Mexico City) became the center of Spanish colonial power in the Americas, serving as a base for the subsequent conquest and subjugation of much of the Western Hemisphere under Spanish rule.

Like Columbus’s voyage in 1492, the fall of Tenochtitlan has come to symbolize the Spanish conquest of the Americas and everything that came with it: the destruction and abuse of Indigenous societies, the political and economic transformations of colonial rule, and, as written on a plaque in the Plaza of the Three Cultures in Tlatelolco, “the painful birth of the mestizo [mixed-race] Mexico of today[1].” In Mexico, the ambiguous anniversary is being commemorated through numerous cultural events, the renaming of a central street in Mexico City, a symbolic reverse voyage by a group of Zapatista activists to Spain, and even a dubious reinterpretation of Tenochtitlan’s founding date for equally symbolic reasons. Beyond Mexico, real and imagined stories about the Spanish invasion and defeat of the Triple Alliance—Cortés’s decision to sink the ships that had brought his party from Cuba, the legend that the local population mistook the Spanish invaders for gods, the image of Cortés weeping under a tree after being driven out of Tenochtitlan only to return victorious a year later, the torture of Tenochtitlan’s last emperor, Cuauhtémoc, in the hunt for Moctezuma’s secret treasure—play an outsized role in how most of us think about the Conquest.[2]

This post is about what happened after the Aztec defeat. As historian Ross Hassig writes, the real conquest of Mexico “came later, after the battles” as the Spanish Crown gradually extended and consolidated political control over a widening territory.[3] This process was contentious, uneven, and arguably still incomplete at the end of colonial rule 300 years later.

How did the Spanish Crown extend its authority over this faraway territory? How did colonial institutions evolve over the first century of Spanish rule? What general lessons can we learn from this case?

The siege of Tenochtitlan, Florentine Codex

The Spanish encomienda as indirect rule

After the victory at Tenochtitlan, the Spanish Crown was faced with the difficult task of extending political authority over a vast and distant territory of which they had little information and little control. To do this, the Crown effectively outsourced the task of conquering territory to freelance “conquistadors” and relied on an institution of indirect rule, the encomienda, to administer these new holdings. Through the encomienda, local elites (encomenderos) were granted the right to extract labor and tribute from the population in exchange for providing for local defense, collecting taxes from Indigenous caciques, and promoting Christian conversion. This feudal-like institution was first employed during the Reconquista of the Iberian peninsula in the prior century. In central Mexico, the Crown adapted the encomienda to local Indigenous institutions, which were already organized to provide tribute and taxes to the Triple Alliance. As historian Bernardo García Martínez has argued, the organization of the early Mexican colonial state under the encomienda exhibits the key features of the indirect type of rule that the British would eventually use in Africa and South Asia.

This arrangement not only proved successful in attracting settlers and extending territorial control. It also provided a practical way for the Crown to extract revenue and assert political control using only a bare-bones administration in the decades following the fall of Tenochtitlan. The Crown relied on encomenderos’ superior information about their holdings to collect taxes and more effectively deploy repression to pacify the local population.

Tax roll and encomienda contract for Huitzilopochco (1554)

The move to centralization

Though the encomienda provided a low-cost way of bringing populations under Spanish rule, it had considerable economic and political costs for the Crown. Because of their feudal-like privileges and autonomy, encomenderos were able to retain a sizable share of local revenue and came to be seen as a political threat to royal authority. As political control grew less tenuous, the Crown sought to centralize control by converting encomiendas to corregimientos, public offices through which royal officials directly collected taxes for the Crown. Unlike encomenderos, corregidores were paid directly by the royal government and answered to higher-level royal officials. They became, in historian Alan Knight’s words, “useful agents of centralization.”[4]

Centralizing power was much easier said than done, however. Encomenderos controlled militias and held independent coercive power. Earlier efforts to rein in their authority—notably through the “New Laws” of 1542—had been met with armed resistance elsewhere in the Empire.[5] In addition to the potential for direct resistance by encomenderos, centralizing power also risked increasing the threat of rebellion from below. Encomenderos acted as a critical line of defense during times of political unrest given their informational advantages and physical presence in the territory. The famous chronicler Bernal Díaz del Castillo, for example, describes his participation as an encomendero in a number campaigns to quell tax revolts in southeast Mexico. In addition to the expense of setting up an administrative apparatus, a move to more direct forms of rule would force the Crown to take on the costly and difficult task of policing the territory.

As long as the risk of popular rebellion and elite defection remained high, the Crown found it preferable to rely on indirect rule through the encomienda despite the loss of revenue and control that this entailed. What changed to make the transition to direct rule more attractive? In a forthcoming article (ungated), we examine how a shock that lowered the risk of popular rebellion and elite resistance provided an opening for political centralization.

Left: Cocoliztli, Codex Osuna. Right: Rodents, Florentine Codex

Institutions and Mexico’s demographic collapse

Many factors determine why authorities might or might not choose to centralize authority, including the role of external threat, the geography or institutional structure of agrarian production, and demographic characteristics of a territory, among many others. We examine the effect of the catastrophic collapse of Mexico’s Indigenous population during the first century of colonial rule. Though figures vary, it is estimated that the Indigenous population of Central Mexico declined by up to 90% during this period, primarily due to disease. The severity of the collapse varied considerably across space and shaped the Crown’s ability and willingness to centralize power in different regions.

The sharp decline of the population had numerous consequences for the political economy of Mexico over the long and short term.  We argue that it facilitated the transition to direct rule in two ways. First, it greatly depressed the threat of Indigenous rebellion and thus reduced the benefits of relying on encomenderos for local defense. In the wake of epidemics, Indigenous social institutions that facilitated collective action collapsed, population pressures on land decreased, and survivors were left, in the words of historian Friedrich Katz, “demoralized and disorganized.”[6] Second, encomenderos’ opportunities to extract local tax revenue and labor evaporated due to the population collapse, reducing both their incentives and ability to contest the loss of political authority. The demographic shock thus provided a window for the Crown to centralize power.

The regional pattern of direct rule adoption across Mexico is suggestive of a relationship between the population collapse and political centralization. As shown in Figure 1, political districts in New Spain and Nueva Galicia (central and north-central Mexico) experienced both a marked decline in population (left panel) and an acceleration in the adoption of direct rule, as measured by the proportion of encomiendas that were transformed into corregimientos (right panel). By contrast, in the Yucatan, where Indigenous populations remained relatively dense, the encomienda thrived well into the 18th century.

Figure 1: The graphs plot the mean logged population (left) and direct rule (right) over time and across districts in each region, with 95% confidence intervals.

Using a difference-in-differences approach to exploit within-district variation in population and the form of rule over time, we find that areas that experienced a more rapid decline in population saw a faster transition to direct rule under the corregimiento. The negative association between population and direct rule adoption is illustrated on the left panel of Figure 2.

One might be cautious to interpret the difference-in-differences estimates as causal given the potential for measurement error, omitted variables bias, or reverse causality.[7] To address these concerns, we also employ an instrumental-variables empirical strategy based on the characteristics of a series of severe epidemics in the late 16th and early 17th centuries.

While some outbreaks—such as the well-known smallpox of 1519–21—were caused by European diseases, others have been traced to a virulent disease known as “cocoliztli,” which is believed to have been a rodent-transmitted pathogen similar to human hantavirus or possibly salmonella. Like other rodent-transmitted diseases, cocoliztli outbreaks were climate-related, typically emerging during years of above-average rainfall immediately following severe droughts. During periods of drought, disease-carrying rodents concentrate around limited water and food resources, allowing the pathogen to spread among the rodent population. When climatic conditions improve, the rodent population rebounds, and rodents spread into agricultural fields and homes, infecting people when they ingest food or breathe air contaminated with the virus from rodent droppings.

Figure 2: The right panel plots a Nadaraya-Watson regression of direct rule on logged population (both in terms of deviations from the within-district mean) with a 95% confidence interval. The left panel plots the average level of direct rule (with a 95% confidence interval) in each year, for districts that experienced (red/dashed) and did not experience (black/solid) drought-rain shocks around the time of cocoliztli outbreaks.

Based on this climactic sequence associated with cocoliztli outbreaks, we construct time-varying instruments using tree-ring chronologies that measure yearly soil moisture. In line with prior work, these climate sequences are highly predictive of the decline in population in the first stage and are unlikely to be directly related with trends in centralization once other factors (like average rainfall) are accounted for. Using this IV approach, we again find evidence that the population collapse facilitated the transition to direct rule (the reduced-form relationship is presented on the left panel of Figure 2).

We further show that the effects of the population collapse were greater in areas where encomenderos had profitable outside options (and thus would have fewer incentives to contest the loss of power to the center)  and where the underlying risk of Indigenous rebellion would have been larger (where a decline in the risk of rebellion would have been most important). Taken together, the empirical evidence is strongly supportive of the role that the demographic collapse played in facilitating the transition to direct rule by mitigating the risk of domestic conflict.

The population collapse was not the only factor that determined where the Crown chose to centralize power. Other scholars have highlighted the incentive to target especially powerful elites or the need to continue to rely on the encomienda in frontier areas where political control was tenuous, among other factors. One thing that all of these works illustrate is that the choice to centralize authority in colonial Latin America differed dramatically across time and space. In some areas, the Crown gained firm and direct control over political institutions within a decade or two of the Conquest. In others, control remained fragmented and fragile for the entire 300 years of colonial rule.  The standard narrative about the Conquest skips the long, messy, and contentious process through which these institutions were established and evolved over time.

Lessons and legacies

What broader lessons can be learned from this example? One clear lesson is that the establishment of Spanish rule over Mexico was far from a clear and linear process that started with the defeat of the Aztec Empire in 1521 and ended with Hidalgo’s Cry of Dolores in 1810. Colonial institutions across the territory and over time differed in the extent of control awarded to central authorities, local elites, the Catholic Church, and Indigenous groups. Royal authority expanded and contracted based on factors like the evolving threat of internal and external conflict, the fiscal demands of warfare in Europe, the 17th century “decline” of Spanish power, or the Bourbon transformation of the 18th century. Its symbolic importance notwithstanding, Cortés’s military victory in 1521 was barely the beginning of the process through which colonial rule was established and evolved.

Beyond Mexico and colonial Latin America, the history of the encomienda in Central Mexico provides some general lessons about territorial conquest and centralization. Central rulers in many parts of the world continue to rely on indirect forms of rule to extend authority over areas that would be otherwise difficult to govern.[8] Centralizing power may be attractive, but local powerholders are often better positioned to monitor and control local populations, particularly when the threat of resistance from below is high. Mexico’s demographic collapse may be unique in its scope and severity, but it calls attention to how a shock that reduces the threat of resistance can open up opportunities to centralize and consolidate state authority (for better or worse, as several Broadstreet posts have highlighted).

One final lesson is that the longevity and structure of Spanish colonial rule was not inevitable. As Hassig notes, when the Tlaxcaltecs and other Indigenous allies joined Cortés in the fight against the Triple Alliance for their own reasons, it was not at all clear that the result of this military victory would be the establishment of Spanish rule across what is now Mexico and far beyond for the next three centuries. It is difficult to say what would have happened had Indigenous populations survived and the threat of rebellion remained high. Our work—and that of many, many others—suggests that the trajectory of colonial rule may have looked very different.


[1] “Fue el doloroso nacimiento del pueblo mestizo que es el México de hoy.” The full text of the plaque in Spanish can be found here.

[2] For a fascinating account of the social and political construction of the enduring myths surrounding Cortés’s expedition, see Matthew Restall’s When Montezuma Met Cortés. Relatedly, an ongoing historical outreach project to mark the anniversary reconstructs the events surrounding the Conquest and engages with some of the prevalent narratives.

[3] Hassig, Mexico and the Spanish Conquest, Second Edition. p. 183

[4] Knight, Mexico: The Colonial Era, p. 54.

[5] The encomendero revolt in Peru following the introduction of the New Laws severely threatened Crown control. See, for example, Lockhart’s classic Spanish Peru, 1532-1560: A Social History, Chapters I and II.

[6] Katz, Riot, Rebellion, and Revolution: Rural Social Conflict in Mexico, p. 80.

[7] For example, the encomienda itself may have exacerbated mortality. Charles Gibson, along with many others, describes the encomienda’s coercive labor arrangements, restrictions on internal movement, and overextraction of tribute, The Aztecs Under Spanish Rule, p. 77-80.

[8] For detailed descriptions of these institutional forms and historical and contemporary examples, see, for example, Gerring et al. (2011) and Naseemullah and Staniland (2016).


  • Emily Sellars

    Sellars is an Assistant Professor in Political Science at Yale University, where she is affiliated with the MacMillan Center for International and Area Studies, the Leitner Program in Political Economy, and the Institution for Social and Policy Studies. Her research interests are at the intersection of comparative political economy, development economics, and economic history. She is particularly interested in the politics of emigration and in the historical political economy in Mexico and Central America. Her work has been published or is forthcoming in the Journal of Politics, the Journal of Development Economics, the Journal of Urban Economics, and the Journal of Political Institutions and Political Economy. Sellars received her Ph.D. jointly in Political Science and Agricultural and Applied Economics from University of Wisconsin-Madison.

  • Francisco Garfias

    Francisco Garfias is an Assistant Professor of Political Science at the University of California San Diego’s School of Global Policy and Strategy. He studies the political economy of development, with a primary focus on state capacity, taxation, and social conflict. His work is published or forthcoming in leading political science journals such as the American Political Science Review, the American Journal of Political Science, and the Journal of Politics. His research on state capacity has received several recognitions, including the Michael Wallerstein Award for best published article in Political Economy from the American Political Science Association. He received his Ph.D. in political science and M.A. in Economics from Stanford University.

3 thoughts on “From Conquest to Centralization

Leave a Reply