To Build or Not to Build?


State Building in Latin America 
By Hillel David Soifer
Cambridge University Press, 2015, 307 pps

Hillel Soifer’s powerful new book proposes a solid and original theory of state-building in Latin America. In recent years the study of how states formed and develop has become a burgeoning subfield, and Soifer aims to tell the Weberian—if not Hobbesian—story of Latin American states; one that is told from within the state. 

Soifer tells us, “... state-building was a state project rather than a sectoral or class project” (emphasis in the original, p. 9). In this account, Latin American state leaders in the nineteenth century enjoyed a considerable amount of “autonomy”; that is, a strong or weak state did not result primarily from its encounters with the society it aimed to govern, but rather: “I argue that the institutional choices made by leaders in populating the bureaucracy shape the fate of state-building efforts” (my emphasis, p. 233). 

Soifer’s book is masterfully anchored to substantive theoretical issues that he illuminates with analytical clarity and impressive empirical work. The book’s departure point is that neither pre-independence institutions nor the aftermath of independence in Latin America can explain the divergent capacity of states in contemporary days. Instead, “the liberal era,” that is, more or less, the period between 1850 and 1900, established the resilient foundations of either successful or failed state projects. Three decisions and trajectories were taken in that “critical juncture”: 1) elites decided not to build a state, hence an effective state never emerged; 2) elites decided to build a state but failed; 3) elites choose to build a capable state and they succeeded. 

According to Soifer, Colombian elites never did actually attempt to build an effective state, so that state has been ineffective throughout history. Peruvian elites, instead, had a state project but failed to implement it. Although these arguments, I am sure, will trigger reactions among historians of both countries, they make a main analytical contribution: causes of state inefficiency diverge and institution builders must be aware of them.

Why would some national elites decide not to build a national state? Soifer’s answer is that a country with several important urban centers, like Colombia, will not undertake the state-building effort, whereas a country with the undeniable primacy of one urban center would attempt do it. The reason is that this kind of geography/demography produces different sets of ideas in elites leading them to either attempt to build a state or not. Backed by quantities of data in three crucial dimensions of state activity (education, coercive capacity and tax extraction) Soifer argues that Colombian elites —embedded in a laissez-faire liberalism derived from the abundance of urban centers—chose not to engage in the costly effort of building a state. In Chile, Mexico and Peru, conversely, liberalism “was more statist”—an ideology derived from the undeniable prominence of Santiago, Mexico City and Lima—and that is what led central elites to seek a national state. 

However, why did countries choosing to build a central state, like Peru, fail, while others, such as Chile and Mexico, succeeded? Soifer dismisses explanations built on war effects on the state, and those that give too much importance to society: “...I attribute more autonomy to Latin American state leaders than do either of my interlocutors...”(p. 8). According to the author, the crucial step to set a path of state capacity was an administrative decision: either deciding to rule the country with deployed bureaucrats, or in alliance with local/regional elites. In Soifer’s account, this mid-19th-century decision is the key and critical choice for state-building in Latin America. Where central leaders decided to rely on local/regional elites to rule the periphery, the state undertook a path of weakness; where they decided to circumvent local powers and deploy a troop of bureaucrats, they put the state on the right track. This is because deployed bureaucrats’ salary depends on central institutions and because the absence of local allegiances helps them to rule in a more independent and effective way, thus constructing more solid institutions: Peruvian leaders relied on regional powers, which prevented the emergence of a strong state; Chilean leaders, instead, aggressively deployed bureaucrats and kept local people out of administrative state positions, which placed their state in the successful path of state building (Mexico followed a path closer to the Chilean one, though relying a bit on regional powers). This is key to the whole argument:  “The result of this difference in administration is nothing less than the success and failure of the state-building efforts...” (p. 86).

Soifer argues that the chosen strategy depended on the elites’ perception of their periphery. If they perceived subaltern classes as rebellious, they preferred to ally with local powers; if they perceived them as non-threatening, they ruled directly. Secondly, if central elites perceived local powers as backward, they would try to defeat them with deployed bureaucrats. Therefore, since Chilean elites perceived a non-threatening periphery, they deployed bureaucrats; because Peruvian elites perceived a tumultuous periphery, they relied on local notables to manage it. Hence, this perception constitutes Soifer’s Archimedean supporting point for the whole proposed theory (and for the institutional future of Latin American countries). 

To conclude, this is a clean analytical and persuasive book; it will inform the debate on the Latin American state for long time. Yet, I will raise three sources of skepticism in this account of Latin American state-building. 


Soifer argues that both an urban and ideological variable play a key role in why some elites do not engage in state-building. Yet I don’t see why the author emphasizes the decisive contribution of ideas when his argument makes it clear that in a country with several strong regions, regional elites have interest in building regional institutions rather than alien central institutions. And, actually, Soifer’s cases almost perfectly correlate: Chile, Peru and Mexico had a single urban center and developed state-building efforts; Colombia had several urban centers and the effort did not emerge. The role of ideas would be much clearer if we had an outcome that “defeats” the urban variable, that is, a case where despite the existence of multiple urban centers, ideas led leaders to undertake a state-building effort; or one country that had a sole main urban center, but the elites’ ideas pushed them not to develop a central state. But the argument lacks such dealignment, which dilutes the weight of the ideological variable. Although Soifer spends two pages (pp. 56-58) warning us to not make this critique, the reader feels that this caveat confirms that the author is aware that ideas have a shaky role in the explanation. Indeed, this feeling seems justified when in the book’s conclusions the whole argument is summarized in a table (p. 260) that combines the geographical variable with state efforts and almost perfectly places/explains every country, with no need of the allegedly crucial ideological dimensions.

In the same realm of “ideas” the reader misses a more detailed reconstruction of the crucial ideological debates and the intellectual context in which they must have happened. The author succinctly establishes the emergence of a laissez-faire liberalism in Colombia and a more statist liberalism in Chile, Mexico and Peru. But how did such homogenous ideological elites’ consensus occur? We miss a more careful intellectual history of such ideas and debates. 

Likewise the reader misses a more in-depth analysis of the “central elites perceptions” of their peripheries. It is precisely because I do not doubt those perceptions were crucial—and because their consequence, according to the book, was the most important one in state-building—that I miss a careful reconstruction of those cognitive structures that, in other century, people used to call les mentalités.  


Soifer’s most ambitious point is to rule out interpretations where the state is the result of its relation with the society it aims to control. Instead, the state (the 19th-century Latin American state) is quite autonomous from society. I agree with Soifer that state leaders’ decisions may have a lasting effect on the state itself, and that they may enjoy some levels of autonomy or isolation from societal forces. However, is this tantamount to proclaiming that Latin American states can be explained on their own, putting society aside? I doubt it, for two reasons. The state leaders that Soifer conceptualizes/sees as insulated state functionaries actually belonged to parties, classes, indigenous groups, clienteles, loggias, clubs, etc... The Latin American state was not kidnapped by a specific class as a basic Marxist view would tell us, but neither was it an autonomous artifact. The state had porous borders with society. Alan Knight, a historian free of subaltern perspectives, has shown that well into the 20th century the successful Mexican state was constantly involved in a give and take with different sectors of society. My own research on Bolivia and Peru during the 20th century gives me a similar impression.

My second point of skepticism about this “administrative” explanation of state capacity in Latin America is that Soifer’s book is actually full of examples of how the state ends up being shaped not only by administrative procedures and political choices, but by the society it aims to govern. At several times Soifer’s main actors, state leaders and bureaucrats alike, face rebellion, resistance, anti-fiscal revolts, etc... that shaped their policies and decisions. Where Soifer proclaims state-building as a political choice, I see a much more embedded choice. 

This last observation brings me to make a final, epistemological comment: to what extent can we explore such a vast, long and complex process as the development of states from a framework that seeks to discover the single independent cause that would allow us to build a general model of state-building? I think Soifer does a superb job in showing that deploying bureaucrats instead of relying in regional strongmen had a true positive effect on state-building in Latin America, but I doubt this can be the ultimate theory that rules out the other ones.  Instead, I tend to lean more towards another author who also masterfully inquired about state-building in Latin America: “I hope to offer a challenge to the implicit ‘claims for essential, invariant universals’ that Charles Tilly asserts have become too predominant in the field. [...] I hope to demonstrate that contingency, contextuality and relationality play too an important role in historical developments to allow for all-encompassing general laws [...].” (Miguel Ángel Centeno, Blood and Debt. War and the Nation-State in Latin America, Pennsylvania State University Press, 2002, p.18).

Alberto Vergara is a Banting post-doctoral fellow at the Weatherhead Center for International Affairs and lecturer on Latin American politics at Harvard ’s Government Department. He recently published La Danza Hostil: Poderes Subnacionales y Estado Central en Bolivia y Perú (1952-2012) (Lima, Instituto de Estudios Peruanos, 2015) 

See also: Book Talk, Democracy