The Promise and Peril of the Popular: Interpretations of Nineteenth-Century Popular Liberalism in Mexico

Abstract

This article examines the literature on popular liberalism in nineteenth-century Mexico and the shortcomings of two interpretations: popular liberalism as an alternative to elite liberalism, and popular liberalism as a strategy to ultimately pursue non-liberal ends. It argues that both interpretations tend to overstate the distance between the liberal elite and its popular supporters because of an unexamined, dichotomous conception of liberalism and the people (generally Indigenous and non-Indigenous peasants) as opposites. It draws its examples from studies of local politics and sides with the interpretation of ‘liberalism tout court’ as the best available option to avoid reifications of liberalism and the popular.

Este artículo examina la literatura sobre el liberalismo popular en el siglo XIX en México y las deficiencias de dos interpretaciones: el liberalismo popular como una alternativa al liberalismo de élite, y el liberalismo popular como una estrategia para perseguir fines no liberales en última instancia. Aquí se argumenta que ambas interpretaciones tienden a exagerar la distancia entre la élite liberal y sus seguidores populares debido a una concepción no analizada y dicotómica del liberalismo y el pueblo (generalmente campesinos indígenas y no indígenas) como opuestos. El artículo toma sus ejemplos de estudios de política local y se alinea con la interpretación de un ‘liberalismo tout court’ como la mejor opción disponible para evitar reificaciones del liberalismo y lo popular.

Este artigo examina a literatura sobre o liberalismo popular no México do século XIX e as deficiências de duas interpretações: o liberalismo popular como alternativa ao liberalismo de elite, e o liberalismo popular como estratégia para, em última instância, perseguir fins não liberais. Aqui se argumenta que ambas as interpretações tendem a exagerar a distância entre a elite liberal e os seus apoiantes populares devido a uma concepção não examinada e dicotómica do liberalismo e do povo (geralmente camponeses indígenas e não indígenas) como opostos. O artigo retira seus exemplos de estudos sobre política local e apoia a interpretação do ‘liberalismo tout court’ como a melhor opção disponível para evitar reificações do liberalismo e do popular.

In the last three decades the historiography of nineteenth-century Mexico moved from studying a confrontation between peasants and Indigenous peoples, on the one hand, and liberal state-building, on the other, to analysing the negotiation between them.¹ When a new literature on ‘popular liberalism’ emerged in the 1990s, it looked as if the idea that liberalism necessarily excluded Indigenous peoples and peasants could be put to rest.² Or could it? To address this question, I make explicit historians’ assumptions about ‘liberalism’ and the ‘popular’. I argue that the study of popular liberalism from the 1990s onwards brought an empirical and a conceptual challenge to the historiography and that we have only partially risen to them. Empirically, it maintained that the gulf between elites and the people had not been as wide as historians had thought before.³ Conceptually, it questioned common preconceptions about peasant and Indigenous politics and rejected idealised models of liberalism. This article focuses on the conceptual challenge as a starting point for a broader empirical and conceptual debate.

Historians of popular liberalism argued that previous accounts of confrontation had overstated the rural population's autonomy from elite politics. The crux of their argument rested in an empirical demonstration of greater convergence between elite and popular practices than had been hitherto acknowledged, at least for some regions. I call their interpretation ‘liberalism tout court’ and consider that it has mostly, and rightly, stayed away from fixed and normative assumptions of liberalism and the popular.⁴ I also argue that this perspective is compatible with the new intellectual and political history of liberalism in Hispanic America (of which more below).

As research on popular liberalism expanded, its premises and conclusions varied. I identify two further interpretations: ‘popular liberalism as an alternative’ and ‘popular liberalism as a strategy’. The former considers popular liberalism was an ‘alternative’ to elite liberalism. It has an optimistic view of the strength of popular agency and its capacity to appropriate and transform elite liberalism into a different practice, which should nonetheless be called liberal. By contrast, the interpretation of ‘strategic liberalism’ emphasises the superficiality of popular liberalism, concluding that it ultimately pursued very different objectives to the elites’, and they were not liberal. While we need to account for cases in which liberal policies were effectively rejected, resisted, paid lip service or surreptitiously subverted, I contend that the interpretations of alternative and strategic liberalism do not always offer a convincing case. They are often hampered by a problematic conceptual premise: the unacknowledged reification of liberalism and popular politics. By reification I mean a process whereby specific characteristics and social relations at a certain time and place are taken as if they were ‘inherent and invariant’.⁵ Additionally, these assumptions are part of a dichotomous frame of mind (the liberal, modern, Western, or elite versus the popular), where ‘liberalism’ and the ‘popular’ are understood as fixed and mutually exclusive ‘ideal types’.⁶

My distinction between ‘alternative’ and ‘strategic’ liberalism, and my use of ‘liberalism tout court’ as a desirable option in some specific cases, is informed by two methodological discussions in Latin American history and Mexican Indigenous studies, which can help us prevent the reification of liberalism and the popular. The first discussion has taken place in the recent intellectual and political history of liberalism in Latin America, which has joined the linguistic turn and criticism of the ‘history of ideas’ to contribute to an Ibero-American conceptual history, and to what Elías J. Palti has called a ‘history of political languages’.⁷ These perspectives argue that we cannot define even a minimal core of liberalism that will stay true to itself across time and space. They seek to avoid normative assessments of whether liberalism has been rightly or sufficiently applied.⁸ A similar methodological position emerged in Mexico in the 2010s for the history and anthropology of Indigenous peoples.⁹ It argues that there is no minimum set of characteristics that define indigeneity, but a complex history of varying criteria for the attribution of indigeneity by actors who self-describe, or describe others, as Indigenous.¹⁰ Recognition of the variability of definitions of the ‘Indigenous’ may be applied to definitions of ‘peasants’ and the ‘people’, or the ‘popular’, for the purposes of this article. The point is to renounce fixed definitions of liberalism and the popular so that actors who supported politics which were considered liberal at the time are taken seriously. Only then may popular liberalism stop being interpreted as a paradox, to become a possibility amongst others.

To discuss popular liberalism, I use examples of the introduction of popular sovereignty and representative government including a wide male franchise, with a focus on the history of liberal municipal government in the form of ayuntamientos (town councils). Attention to local politics is appropriate because it is local culture that is often considered the most different, resistant or opposed to national liberal policies. While I do not include economic liberalism in my discussion, my arguments are compatible with Emilio Kourí's analysis of land disentailment in that we both contend that some of the individualising logics of liberal policies were accepted and reproduced more frequently than has been acknowledged.¹¹

In the next section I summarise the broader historiographical debate on liberalism developing from the early twentieth century. The following three sections examine a selection of arguments representative of the interpretations of liberalism tout court, alternative liberalism and strategic liberalism developing from the 1990s onwards. In the last section, I propose shifting our questions from asking whether popular actors were liberal to how inclusive the practices and effects of everyday governance were.

Official, Revisionist and Post-Revisionist Interpretations

Interpretations of nineteenth-century Mexican liberalism started with an official version first penned by its protagonists and from 1917 onwards by authors close to post-revolutionary governments. It was an upbeat, linear account of the progressive advance of liberalism, including popular participation. With elite leaders as protagonists, and no empirical research on popular actors, they produced a simplified view of a single pueblo identified with its leaders.¹² In a Mexican version of the Whig theory of history, liberal ideas and popular sovereignty produced a ‘social liberalism’, which resulted in an ongoing democratisation from the Independence War in the 1810s through the Liberal Reform (starting in 1855) to the 1910 Mexican Revolution.¹³ As is well known, this teleological narrative benefitted the post-revolutionary party in power and was discredited after 1968 when the government's legitimacy deteriorated. There followed a revisionist interpretation of the 1910 Revolution with a less optimistic account of nineteenth-century liberalism.¹⁴ An unfortunate result of post-1968 revisionism was that together with the dirty bathwater of teleology and official history, the baby of nineteenth and twentieth-century popular participation and unreified liberalism was thrown away too. Where Jesús Reyes Heroles saw Mexico's ‘social liberalism’ as encompassing popular interests, and appropriate for the country, his critics saw the opposite: a self-interested elite in the form of strongmen (caudillos and caciques) had betrayed their mass following.¹⁵ This revisionism dovetailed with a positivist and nationalist interpretation of liberalism in Mexico (and the Hispanic world) as a failed attempt to impose foreign ideas on a culture that was not ready for them. In this view, liberal reforms sought to dissolve peasant and Indigenous communities, and promoted individualism while the masses clung doggedly to corporativism and communitarianism. Popular support for liberalism was therefore unthinkable unless it was conceived as exception. Historians saw liberal disentailment of land as the principal attack on former colonial Indian pueblos (pueblos de indios), who invariably resisted.¹⁶ Liberals and popular actors were antagonists.

From the 1980s, post-revisionist studies of the 1910 Revolution argued for the impact of popular actors in nation-state building.¹⁷ In the 1990s, a similar perspective emerged in monographs on the nineteenth century, broadening the scope of popular actions studied to include local government, and regional and national political alliances. Varying definitions of political liberalism have been used in this context; for the purposes of this article, I use ‘liberalism’ in two senses, which pertain to different analytical levels. One is a broad meaning which associates liberalism with popular sovereignty and representative government, and conflates it with ‘Western’ and ‘modern’ politics.¹⁸ Once a ‘liberal’ form of government is taken for granted, a second meaning of liberalism refers to specific policies considered liberal at the time; here the core distinction in Mexico was between the Liberal and the Conservative parties. Studies have accounted for popular liberalism, understood in both senses, for various regions and periods. These include the 1808 crisis of the Spanish empire that led to the 1812 Cádiz Constitution (in turn leading to the first use of ‘liberal’ in its modern political sense); the Independence wars and the following decades of experiments in liberal federalist and conservative centralist governments (1812–50); the mid-century Liberal Party struggles against conservatives and foreign intervention (1850–67), as well as the restoration of the Republic (1867–76); and, finally, the period during Porfirio Díaz's presidency and influence, known as the Porfiriato (1876–1911). Research on the ‘first liberalism’ (circa 1812–55) has documented popular appropriation of electoral rule and egalitarian legal guarantees.¹⁹ Amongst the most solid cases for popular liberalism was the radical, martial version of the Puebla Sierra. During the 1850–80 period its population participated in the National Guard battalions fighting the mid-century wars on the Liberal Party's side, supported representative government and liberal taxation, and enjoyed a significant measure of local autonomy as well as an effective defence of civil and political rights.²⁰

Historians agree that after 1880, the revolutionary spirit of popular sovereignty and citizens in arms gave way to a more authoritarian regime inspired by the positivist motto of ‘order and progress’. National government shifted its policies from a more radical liberalism, sensitive towards municipal autonomy and in greater tune with the people (whose support had been crucial in times of war), to a ‘developmental liberalism’ ready to sacrifice local autonomy, as well as civil and political rights, to political centralisation and economic growth.²¹ Popular pressure to defend previously enjoyed autonomy and rights, including the right to vote, persisted in some cases, and the gap between the elite and the people widened. What is significant here is that such distance was the result of popular support for liberal guarantees and policies, rather than lack thereof, together with the government elite's renunciation of some of those liberal principles.

Liberalism Tout Court: The Old Adapts to the New

I have described liberalism tout court as the interpretation which best avoids the reification of liberalism and the popular.²² I consider its advantages through two strategies. Firstly, in this section, I show how the history of liberalism ‘from below’ may be compatible with the underlying premises of the intellectual and political history which has questioned longstanding interpretations of Hispanic and Latin American liberalism as failure, an idea ‘out of place’, or too ‘advanced’ for ‘backward’ societies.²³ When these two corpora of literature take historical actors’ self-definition as liberals seriously, they allow for a recognition of the breadth and depth of nineteenth-century political transformations, arguing that even if old practices persisted, they had to adapt to the new. Secondly, in the sections on alternative and strategic liberalism, I will use examples from liberalism tout court as contrasting explanations.

In recent intellectual and political history of the Hispanic or Ibero-Atlantic world, liberalism is no longer reduced to an authentic, original or ideal type, as it had previously been following a British, US or French model. Hispanic liberalism is now seen as one of the Atlantic liberalisms.²⁴ Given that all liberalisms ‘select – deliberately or unconsciously – certain items from an accumulated and crowded liberal repertoire and leave others out’, there is room for those that are not completely individualistic.²⁵ Hence, there is no reason to consider defective, alternative or hybrid a liberalism like that of the Cádiz Constitution (1812), which retained Catholicism, built on a legal culture used to ‘acting in corporative and jurisdictional terms rather than on an individualist basis’, and put ‘more emphasis on the nation than on individuals, whose rights are confirmed precisely as a result of being members of the national community’.²⁶ This is no more contradictory, hybrid or inauthentic than the varieties of Anglo-liberalisms which supported both ‘the imperial civilising mission and its passionate denunciation’.²⁷ In London or Mexico, elite or popular, liberalism on the ground was messy.

In this framework, the persistence of corporativism amongst the elite or the people is not necessarily a sign of alternative, instrumental or fake liberalism. The question is, what kind of corporativism are we talking about? Antonio Annino has asserted that during its installation, early nineteenth-century ayuntamientos appropriated popular sovereignty and liberal citizenship, resulting in a shift of power from the cities to the rural world. For him, the ayuntamientos’ autonomy increased and it was not a mere preservation of colonial corporativism but rather the result of the deep rupture brought about by the French invasion of Spain, the ensuing crisis of the Spanish monarchy in 1808, and the 1812 Cádiz Constitution's recognition of popular sovereignty and representative government. In this new context, some colonial corporative practices persisted, but they were no longer supported by the same old system. Instead, they had to adapt to the new, so that for Annino they are better described as neocorporativism.²⁸

In this respect, Annino's view is compatible with Palti's argument that there was a fundamental level at which the effects of certain changes, such as the introduction of popular sovereignty, could not be avoided. Palti sees this transformation as part of a new political language, which he differentiates from ideology in that it does not refer to ideas or specific content but to something taking place at a deeper level: the modes of production and articulation of ideas, and the conditions of possibility of performative discourses. The transition from 1808 onwards transformed the playing field and changed the rules. Anything that happened afterwards, including a wide array of uses of the old vocabulary, rituals, hierarchies and practices, would never be the same; ‘even very old ideas gained a completely different meaning’ as they were articulated under new logics.²⁹

Annino's assertion that ayuntamientos appropriated liberal citizenship, resulting in a deep rupture with the old political order, has been challenged by subsequent studies. However, if we agree with Palti, objections to Annino's thesis may be seen in a new light. Annino's argument that local autonomy was gained after Cádiz and that this was a radical rupture has been qualified.³⁰ Available evidence suggests that Annino's model applies better in some regions than others.³¹ Still, acknowledging the existence of limits to municipal autonomy and regional variations should not stop us from considering the effect of the wars, Cádiz and subsequent constitutions at a deeper level.

The replacement of the king's sovereignty with the principle of popular sovereignty changed the underlying logic of government at all levels. The new, radical practice of national elections was introduced. Even if some corporative forms of representation persisted (for instance, in the Insurgents’ 1813 elections to the Chilpancingo Congress where corporations had the right to vote), these practices were put to the service of new provincial and national political networks, ‘to build the government from the bottom up’, as Michael Ducey has recently argued.³² In the following decades, at the local level, popular sovereignty bolstered the power of commoners against the nobility by giving it a legal and institutional basis, upsetting some of the hierarchies of age and status, even in districts with no strong regional or national war or party alliances, as shown by Peter Guardino.³³ It could also stop commoners from trying to obtain hereditary privileges, attempts that had taken place in the late-colonial period.³⁴ The new logic set the rules for the creation of a more level playing field by denying anyone any privilege.

When historians question Annino's thesis of a deep rupture to instead emphasise continuities,³⁵ they perhaps lose sight of the broader picture, thus underestimating the effects of the introduction of national elections, the expansion of the electorate, or the significance of everyday defence of civil and political rights throughout the century.³⁶ Details are many and the jury is still out on the character and degree of local autonomy, but there is no denying that new paths were opening up to fight for a more inclusive and egalitarian regime. Hierarchies did not disappear but were reconfigured under new patterns.³⁷ Key to this interpretation is the fact that it avoids explaining inequality exclusively as a result of the persistence of pre-1812 practices. Instead, it allows for an analysis of new forms of inequality spurred by the introduction of liberal policies and their contradictions.³⁸

As will be seen in the next two sections, the interpretation of liberalism tout court has been challenged by those of popular liberalism as an alternative and as a strategy. In the Spanish-language historiography, strategic liberalism is probably the most frequent interpretation of the three, so I will address it in greater length. Alternative liberalism, while less common in either English- or Spanish-language historiography, has been influential, so it also has its own shorter section below.

Popular Liberalism as an Alternative

The interpretation of popular liberalism as an alternative was originally and boldly argued by Florencia Mallon's 1995 book Peasant and Nation: The Making of Postcolonial Mexico and Peru (University of California Press). Analysing peasant contributions to nation-state building in the second half of the nineteenth century, Mallon persuasively criticised peasant and subaltern studies for attributing an exaggerated autonomy to their subjects, making it ‘theoretically impossible for subalterns to do more than negate state power’.³⁹ Her work also opposed Eric Van Young's view that popular participation in the Independence War was motivated exclusively by local concerns (religious preoccupations and class- and ethnic-based enmities), and not by any of the national projects.⁴⁰ Thus Mallon brought ‘peasant’ and ‘nation’ much closer together. Taking a ‘new cultural history’ perspective and encouraging broad theoretical dialogues, her analysis of how Mexico achieved a more inclusive hegemony than Peru still stands as a crucial contribution to the study of Latin American popular politics. The prominence of Mallon's argument in the wider political history debates is probably one key reason why popular liberalism is often understood by those researching national reforms as an alternative form of liberalism.⁴¹ By contrast, histories ‘from below’ written after Mallon's book have generally considered her depiction of a strong popular agency too optimistic. Indeed, her assertion that popular liberalism was a counter-hegemonic alternative to elite liberalism has not been taken up by subsequent studies of Mexico. I argue that some of Mallon's assumptions were flawed, leading her to overstate popular autonomy.

Examining the mid-century wars in Morelos and Puebla, Mallon paid close attention to peasants as petitioners, soldiers, taxpayers and land tenants and owners who supported the Liberal Party. She understood popular liberalism as a ‘counter-hegemonic’ alternative to elite liberalism and conflated the popular with the ‘communal’ or ‘communitarian’. She argued that ‘communal’ life provided the basis for an ‘alternative liberalism’, a more ‘democratic’ and inclusive vision of the nation, which she defined as ‘communitarian liberalism’ or ‘popular liberalism’.⁴² She focused on collective forms of decision-making that she called communal politics; as well as collective control of land privatisation defined as an alternative discourse on land property. Her argument rested on a clear differentiation between elite and communitarian/popular liberalism.

Mallon described the interaction between customary politics (which she called ‘communal politics’), municipal politics and the National Guard battalions organised by the Liberal Party. I contrast her interpretation with Guy P. C. Thomson's, which serves as an example of ‘liberalism tout court’. By foregrounding the mediation between local communities and national Liberal Party leaders, Thomson's account allows me to show how Mallon overestimated the force of ‘communal politics’.

In her book, Mallon replaced the term ‘community’ with ‘communal hegemony’ to abandon idealised views of continuity and consensus in peasant communities and engage with change and conflict, as well as inequalities within.⁴³ Yet she stopped short of a full dismantling of assumptions about community. Her communal politics (also referred to as communal political culture, communal democracy and democratic patriarchy) were comprised of three elements: civil-religious hierarchies or cargo systems (the informal juxtaposition or intermingling of legally recognised municipal posts and customary posts that served civil administration and religious ritual, and were arranged hierarchically); councils of pasados (elders holding the highest authority in local customary organisation); and ‘communal assemblies’.⁴⁴

Mallon described communal assemblies as ‘arenas of communal discourse in which interest groups, factions, or individuals sought collective approval for their projects or status’,⁴⁵ and argued that National Guard battalions (of which more below) held them ‘to consult village leaders and to legitimize their decisions to enter combat’.⁴⁶ Her discussion was based on archival sources and twentieth-century ethnography. Those of us who have consulted minutes of meetings in Puebla Sierra archives have found that they sometimes listed some male participants other than those with municipal posts, such as pasados; occasionally other male residents of voting age (vecinos) were present too.⁴⁷ Difficult decisions such as supporting a war probably involved them, but without details of village politics, we cannot know what percentage of the population participated, nor their age, gender or status. There is no archival evidence to state, as Mallon does, that ‘women and younger men’ sometimes participated as a ‘communal chorus’ expressing approval or disapproval. Indeed, Mallon based her description of this and other practices on ethnography carried out in Hidalgo in the 1980s.⁴⁸ Using ethnographies of different time periods, and sometimes of different regions too, to give fuller accounts than the available historical sources allow for a given case, is called ethnographic upstreaming, and it became the distinctive method in ethnohistory. As I will further argue in the section on ‘popular liberalism as a strategy’, this method's temporal premise is no longer sustainable. That Mallon used it is contradictory. If we reject the idea that communities were static, why retain a methodology of ethnographic upstreaming based on the assumption of continuity across decades and regions?⁴⁹ Moreover, the late twentieth-century evidence she grafts on to the mid-nineteenth century suggests greater inclusion than her archival evidence shows.⁵⁰

Mallon found documents mentioning pasados and their negotiations with civil and National Guard authorities, but further archival evidence of an actual council formed by pasados, or of a stable hierarchy of religious and civil posts is lacking for the Puebla Sierra. Yet she asserts that ‘by the mid-nineteenth century’ cargo systems, such as the ones observed by anthropologists in the twentieth century, had been formed and exercised ‘communal supervision’ over municipal politics.⁵¹ On the basis of ethnographic upstreaming, she could have assumed that there was a more continuous, systematic and self-contained cargo system than actually existed.⁵² Clearly separating customary and municipal politics allowed for her definition of the communal as relatively autonomous from the legally recognised municipal politics, but on the ground the distinction between them might not have been as clear-cut as she suggested.

Pasados had representation and government roles and some negotiations involved more people than the legally sanctioned authorities. However, it is not clear that the participation of elders (selected for their life-long service but frequently also because of their wealth and status), or cargo systems as a whole, if they had already taken shape as such, made communities more politically egalitarian. In fact, Mallon's strongest evidence of increased inclusion was the ‘loosening of ethnic and age hierarchies’, which came from the empowerment of young soldiers who spoke Indigenous languages and formed the National Guard battalions, thus gaining voice vis-à-vis pasados and commercial elites.⁵³ So, in fact, it was at least partly against one element of communal politics (the pasados), and by joining the National Guard organised by the regional Liberal Party, that inclusion of young Indigenous men took place and they became nationally important. It is then necessary to pay attention to such regional and national Liberal Party links.

The National Guard was created as an auxiliary force during the 1847–8 invasion by the United States; it was later reactivated by the Liberal Party and became crucial for defeating the conservatives and foreign powers during the War of the Reform (1858–61) and the European Intervention (1862–7). As argued by Thomson, the National Guard's decentralised organisation was key to attracting the population of the Puebla Sierra to its ranks. It differed from the traditional army in that it was administered by state governments from 1855, adult men could choose to pay a tax instead of serving in a battalion, and soldiers voted to elect their commanders; they never fought too far from their hometowns and were sometimes rewarded with land and musical instruments for their service. The Liberal Party's recruitment strategy succeeded because it ‘involved a clearly defined contractual relationship with the state, replacing the arbitrariness which had hitherto characterised military administration’.⁵⁴

Mallon saw the National Guard's pueblo-friendly conditions as the result of communitarian pressure. Indeed, local communities were forces to be reckoned with, but their power lay more in the elite's need for military support than the community's initiative per se. Additionally, mediation between communities and national leaders was crucial. Yet Mallon moved her focus away from mediators such as the Nahua leader of the regional Liberal Party, Juan Francisco Lucas, and subsumed the National Guard into communal politics. She thereby found the ‘relatively autonomous’ communities she was looking for, but at the cost of underestimating the connections with regional and national elites that empowered them.⁵⁵ Mallon's interpretation at many points was nuanced yet her bottom line rested on two complementary preconceptions: that elite liberalism was necessarily exclusionary, and that popular or communitarian practice was inherently more inclusive, egalitarian and democratic. One could argue that there were also moments when elite liberalism could be inclusive and the popular exclusionary or hierarchical.⁵⁶

Popular Liberalism as a Strategy

In the previous section I showed how Mallon considered popular liberalism as an alternative to elite liberalism, even if she had initially argued against exaggerating subaltern autonomy. I also introduced some of Thomson's arguments as an example of the interpretation of ‘liberalism tout court’, which showed a relatively close relationship between elite and popular liberals. Mallon, by contrast, depicted a more distant relationship. Yet, for a growing number of historians, the differences of interpretation between Mallon and Thomson are less important than the fact that both over-emphasised the depth and spread of liberalism. Those sceptical of popular liberalism perceive it as a strategy rather than an end in itself.

Strategic or instrumental popular liberalism considers that people used liberal policies only as a means to pursue local interests or local autonomy, which are construed as non-liberal ends. It includes the cases when historians suggest that popular support took place ‘more out of necessity than conviction’.⁵⁷ This interpretation has been frequent, whether as a casual comment, a description of a specific case or a general conclusion. It is used to express doubts about the significance of popular support for different liberal policies in different regions. Compliance could be grudging rather than enthusiastic, of course, but I would argue that this did not necessarily preclude politically effective support.⁵⁸

Many studies of popular politics are dense microhistories covering short periods, with conclusions often limited to a region, or a single municipality or village. Thus sub-national historiography has been enriched, but it remains disconnected from wider debates on elite liberalism and national reforms.⁵⁹ In this fragmented field, I picked some of the few analyses that connect their findings with those of other regions and periods. I focus on two discussions which are representative of common assumptions about popular politics, and keep further examples in the footnotes. Firstly, I show how 1990s research by Mallon and Thomson on mid-century popular liberalism was criticised by historians working on different topics or periods but interested in peasant politics. Secondly, I examine representative politics at the municipal level by focusing on Leticia Reina's recent history of elections (1876–1910), which, while studying a later period, is in instructive dialogue with the work by Guardino for Oaxaca (1750–1850) and Annino for the impact of the Cádiz Constitution up to the 1820s. I show how Reina's ethnicised history tends to emphasise instrumental liberalism and I question some of her assumptions.

Peasants: Practice vs Ideology

The application of James C. Scott's perspective on peasants has given us a fine tradition of social history taking popular agency seriously and providing nuanced accounts of social mobilisation. Scott's understanding of moral economy and resistance usefully rejected assumptions of false consciousness and apolitical actors.⁶⁰ Yet one unfortunate tendency in some of the uses of Scott's work has been to separate ‘ideology’ and ‘practice’. Such distinction often underlies interpretations of strategic liberalism. This was John Tutino's comment on Mallon's book:

[Peasant leaders] in Morelos used liberal discourse when they allied with the liberals, and a deeply traditional discourse when they addressed Emperor Maximilian. I suspect that, ultimately, they were not ‘communitarian liberals’, nor cultural traditionalists, but they simply pursued local interests and developed their own culture in interaction with local elites and peasant families […] Were they ever really ‘liberal’ or ‘nationalist’, did they even have any specific peasant or popular ideology? I doubt it.⁶¹

A similar view was espoused by Keith Brewster, who studied Gabriel Barrios, the military commander and cacique of the Puebla Sierra in the 1920s. Brewster objected to Mallon's and Thomson's interpretations:

While I agree that participation in widespread conflicts indicated community engagement in crucial political issues, I believe that the underlying motives were more pragmatic than ideological. In this respect, I question whether evidence of ‘popular liberalism’ should more accurately be viewed as the deployment of what James C. Scott refers to as ‘weapons of the weak’. Mobilization in favour of the liberal cause, whether popular or not, may have indicated nothing more than a community's realization that such a posture represented the best hope of achieving their aims.⁶²

For Tutino, peasants had their own ‘interests’ and ‘culture’, but no ‘ideology’; similarly, Brewster opposed pragmatic to ideological motives and saw liberalism as a ‘weapon of the weak’: an instrument rather than an end. They shared an understanding of ideology in which a sharp line was drawn between discourse and practice.⁶³ Additionally, they suggested that peasant actions cannot be taken at face value. While in some cases there might have been a ‘hidden transcript’ that was opposed to what was said in public, assuming that this was generally the case denies the possibility of a real interest in at least some aspects of liberalism, and the practical consequences of adhering to a discourse and a political faction, even when the result was the defeat of conservatives and foreign intervention, and the development of a number of liberal institutions.⁶⁴

Strategic liberalism is defined against an ideological or authentic attachment.⁶⁵ It presupposes that only a strong belief and commitment counts, while strategic adherence is lesser or even deceptive: a façade. The idea of liberalism as a mask to hide non-liberal objectives is a frequent trope in Mexican historiography, as when historians propose to strip away ‘political rhetoric and postures’ to unveil the real objective: ‘local autonomy’.⁶⁶ It is often emphasised that it was not ordinary people but savvy mediators who properly understood the liberal programme, and it was they or their secretaries who penned liberal declarations and petitions, implying that only they were true supporters.⁶⁷ This is an idealist view of politics and ideologies, focusing on deep knowledge or conviction to the detriment of more pragmatic considerations. Historians are quicker to question the masses’ attachment to party politics and ideology than the leaders’, and leave no room for the possibility that superficial or accidental attachments might have eventually become effective or long-lasting.⁶⁸ They ultimately follow a dichotomous framework where the people are actors in materialist interpretations, and only elites can be properly ideological. We can leave this double standard behind by discarding axiomatic oppositions between idealist and materialist frameworks, and between the elite and the people.

A frequent argument is that ‘regardless of the deployment of cosmopolitan political discourse, the village remained the focus of peasant priorities’.⁶⁹ Here the local is opposed to the national or global in a zero-sum game. Historians who speak of popular federalism and liberalism as a mere strategy consider that the properly peasant or Indian demands referred to ‘protection from unrestrained external interference, whether this be forced conscription, crippling taxes, or threats to local autonomy’.⁷⁰ While we have abundant evidence that such protection was sought, it is not clear that these local demands always diverged from the national form of liberalism. In the Puebla Sierra, it was precisely the Liberal Party's negotiation of a locally acceptable implementation of national projects − such as the National Guard, the end of compulsory payments and services to the Church, and the collection of universal, direct taxes (replacing former ‘crippling taxes’) − that successfully bridged the gap between people and elite, local and national, for the 1855–80 period and, to a lesser extent, between 1880 and 1910.⁷¹

Indians: Binary Logics and Temporal Alterity

After historians’ initial enthusiasm for the impact of the Cádiz Constitution, the Spanish-language historiography has grown increasingly unconvinced by the idea that there was a meaningful popular liberalism in the long nineteenth century. The strongest form of scepticism has highlighted the prevalence of colonial continuities and/or non-liberal practices in the last instance, even for the second half of the century. To explore this view, I focus on Reina's book on municipal elections, including liberal elections prescribed by law and customary elections (‘sistemas de usos y costumbres’). This is a crucial topic because one of the ways in which local autonomy has been considered incompatible with national liberalism pertains to the persistence of local, customary forms of representation and government. Reina studies the states of Oaxaca, Puebla, Veracruz, Estado de México and Michoacán, all of which had the highest number of electoral conflicts in the country between 1876 and 1910. Reina's systematisation of abundant documentary sources from different states is a key step to understand political representation at the local level. However, I differ from her interpretation. I argue that her strategic liberalism rests on a contradiction rendered legitimate by binary logics and temporal assumptions which come from ethnohistory and anthropology, and have been persuasively criticised within those disciplines.

The study of nineteenth-century elections has been developing since the 1990s. For the 1820–70 period Annino and Hilda Sabato argued that the instability brought about by electoral conflict was not due to the weakness of republicanism or liberalism in Latin America, but the opposite: it was the result of extensive support for popular sovereignty and the difficulty of building stable institutions to channel it peacefully.⁷² Regarding Mexico after 1870, Reina's study finds widespread and deep engagement with municipal elections, which may be read as an extension of Annino's and Sabato's argument to a later period.⁷³ Ordinary people continued to participate in liberal politics, even if their chances of being heard were decreasing.⁷⁴ On the other hand, studying customary elections has at least two effects leading Reina to conclusions different to those of Annino and Sabato. Firstly, it curbs the potential overestimation of the legitimacy of liberal municipal elections. Secondly, Reina argues that wherever ‘Indigenous’ electoral practice existed, the liberal, secret vote was ‘used à la carte’, suggesting a conditional and contingent adherence.⁷⁵ Would Reina then conclude that this was an instance of instrumental liberalism?

Customary elections derived from the practices of colonial Indian pueblos. Once liberal elections were introduced, the old forms of local elections lost legal backing in some states, but they did not disappear and, in others, they were officially allowed to co-exist with liberal elections. In the old forms the decision-makers were Indigenous notables or elders (pasados, principales or consejo de ancianos), or assemblies that reached a consensus, sometimes rotating different members from each of the barrios (hamlets).⁷⁶ Reina states that the persistence of colonial forms did not bring about a political syncretism, as argued by Annino for the first decades of the century, because in her view colonial corporative logics and liberal individualist logics did not mix.⁷⁷ Neither does she see an ‘encounter between local custom and liberal law in which Indigenous communities were fundamentally changed’, thus departing from Guardino's conclusions for Oaxaca up to 1850. Reina argues that the two forms of election co-existed. Former Indian pueblos reinforced ‘traditional and consensual forms of election’, at the same time that they increasingly adopted liberal elections and, under certain circumstances, defended them fiercely.⁷⁸

Were customary and liberal elections completely separate histories? The idea of co-existence neatly separates them and, although Reina considers interaction at some points,⁷⁹ her final resolution is to emphasise the particularity of the Indigenous as expressed in customary elections. This is done following two operations. One is to keep a binary narrative opposing ‘liberal’ and ‘Indigenous’ throughout, leading to the conclusion that Indigenous liberalism was necessarily instrumental and used, as Tutino had suggested, for non-liberal ends. A second operation discards the idea of homogeneous, static and exclusively traditional communities by emphasising their capacity for change yet leaves some of its underlying temporal conceptions unchanged.

Let us examine the binary oppositions in Reina's argument. For her, communities switched between customary and liberal elections, and even if the two logics changed in time, they remained separate. She emphasises that liberal elections were not hegemonic but with her own evidence neither is it possible to conclude that customary forms predominated. It is clear that customary elections were not used systematically but ad hoc, even in states like Estado de México or Oaxaca, where they were officially recognised until 1870 and 1900 respectively.⁸⁰ She states that further research is needed but puts forward the hypothesis that customary elections were preferred ‘when there was a relative harmony’ or ‘agreement’ amongst political forces; in other words, when traditional forms of election were deemed legitimate.⁸¹ By contrast, in crisis, they resorted to municipal elections as prescribed by law (for example, in cases where customary authorities were no longer believed to represent community interests fairly, or in conflicts between a locality and a jefe político or other higher authorities seeking to impose a municipal president of their choice).⁸² Reina concludes that ‘ethnic groups’ adopted liberal elections and thereby articulated themselves with the national system, at the same time that they ‘reformulated their internal social organisation without losing their own particular culture, different to the national’, in order to ‘perpetuate themselves as a group’.⁸³ She sees the electoral practices of liberal municipal cities as an ‘alien institution’ for Indigenous pueblos and considers their ‘appropriation’ as a strategy for ethnic group survival, an ‘adoption’ that included the ‘refunctionalisation’ of liberal practices to fit their needs.⁸⁴ Thereby Reina aligns with strategic liberalism as a means to an end that was not sought by liberal policy: ‘ethnic group survival’. The underlying incompatibility between Indigenous populations and liberalism, which 1990s research had appeared to upend, is reinstated at a certain level: liberalism is not anathema anymore; it was used but only instrumentally and was kept separate from the ‘Indigenous’.

And yet both customary and liberal elections were used ad hoc. Based on Reina's description, we may fairly conclude that the legitimacy of customary elections was as conditional and contingent as that of the official elections. Whatever colonial elements persisted, what Reina's vital findings suggest (even if she does not state as much) is that customary forms in the post-1812 world lacked what we normally understand as custom-based or traditional legitimacy: following a certain practice simply because this is the way it has always been done. Instead, given that there was now the option of the individual ballot, custom became subject to performance: when there was agreement on the method and the elected authorities, they were followed; when there was not, the liberal ballot was used and its results were defended by legal and sometimes extra-legal means. Neither was the secret, individual ballot hegemonic. Yet if customary elections were subjected to the logic and legitimacy of performance, then they existed under very different rules to the times in which they were the only norm. And this took place in an era of popular sovereignty; dispute over the meaning of the latter, and how to operationalise it, does not suggest rejection but rather engagement. Further research could explore what forms of popular sovereignty took place on the ground and how inclusive existing practices were. Did customary agreements rest on the choices of pasados/principales or on assembly discussions by a minority or a majority? What constituted a majority for them? Had the logic of adding a certain number of individual male voters, as in the computations of liberal elections, seeped into customary assemblies making decisions? Is it really possible to completely separate ‘customary’ from ‘liberal’ logics in political practice?

Let us now consider the temporal assumptions in Reina's argument. She asserts that the colonial electoral practices that ‘persisted’ were also ‘reformulated’ under the new circumstances, including the fact that liberal legislation enabled the contestation of authorities’ hereditary privileges. However, neither this key transformation, nor others, become a central concern. Instead, she moves from examining electoral practice to describing cargo systems by combining archival sources referring to customary elections during the period of study with twentieth-century ethnographies.⁸⁵ In a topic for which it is difficult to find any traces, Reina expertly compiled a wealth of primary sources. It is then unfortunate that she does not stick exclusively to them and, like Mallon before, her description of cargo systems relies partly on ethnographic upstreaming.

Mallon and Reina refer to two 1980s articles on cargo systems which rejected ethnographic upstreaming. These articles used only documents produced during their periods of study and found that cargo systems had not developed in the pre-Hispanic or early colonial period but later, from the late eighteenth century in Villa Alta and the Valley of Oaxaca, from 1770 to 1850 in Jalisco, and in the late nineteenth and early twentieth centuries for Zinacantán and Chamula in Chiapas.⁸⁶ Yet neither Mallon nor Reina followed the methodology in the aforementioned articles by restricting their sources to evidence produced at the time in their specific regions. Instead, Mallon assumed that the dating of cargo systems in one of these articles would apply to her region of study, and both used twentieth-century ethnographies to flesh out their descriptions, thereby assuming continuity for about 100 years.⁸⁷

In Reina's narrative, as in Mallon's before, Indigenous peoples’ ability for change is constantly emphasised but a shift to seeing them as fully participating in the novelties of their time is truncated by reliance on a methodology based on the temporal alterity of its subject of research. The problem rests in the origins of anthropology and ethnohistory, which defined their object of study as ‘primitive’, ‘traditional’ or ‘underdeveloped’ peoples, thereby assuming that they belonged to another time.⁸⁸ Moreover, the anachronism of ethnographic upstreaming was enshrined as the specific methodology for ethnohistory, freezing its subjects in time. As Alejandro Araujo has demonstrated, the temporal assumptions embedded in the origins of ethnohistory, while persistent, are no longer pertinent.⁸⁹ In the revised model, Indigenous people are no longer stuck in time; indeed, they are adeptly changing but not so much as to become confused with the non-Indigenous, or with liberal practice. Why? In the past, the objective was to not lose the specificity of the national in the ideology of mestizaje; today it is perhaps the highly valued diversity of multicultural nationalism that wants to be preserved, just like Mallon did not want to give up the promise of the popular as one of greater inclusion and egalitarianism. Additionally, the criteria to define who and what is Indigenous or popular remain firmly in the hands of the historian, anthropologist or ethnohistorian and their assumptions, particularly the equation between Indigenous peoples and an overly consistent image of customary politics, both in opposition to liberalism.

Finally, there is a significant number of studies showing a weaker scepticism and inconclusive results. They cover earlier periods and have appropriated to a greater extent the new intellectual history premise of not taking for granted a specific model of liberalism, such as an Anglo-American individualist version. They emphasise that there are many liberalisms and analyse a variety of responses in municipalities nationwide. However, they stop short of classifying them, and when a conclusion is due, perhaps overwhelmed by diversity and detail, they clutch onto an ethnic or culturalist explanation as a last resort. For example, they state that liberal principles were ‘adapted’ to the needs of the pueblos and they used liberal ‘language as a strategy to defend their rights to communal property and to demand justice and respect for their “usos y costumbres”’.⁹⁰ The problem with such conclusions is that they reinstate the reification of liberalism and the popular that their own acknowledgement of the existence of diverse liberalisms was supposed to supersede. They also flatten out the complexity of their own findings by reducing them to changes and adaptations that are ultimately there to defend non-liberal or illiberal local or traditional practices; thereby underestimating the breadth and depth of change. A similar procedure has been questioned in anthropological research. José Luis Escalona demonstrates how a great variety of interactions, conflict and tensions are reduced to a ‘linear ethno-narrative’.⁹¹ As Paula López Caballero compellingly shows, anthropologists have sorted out intricately interwoven practices and produced many simplified, binary taxonomies, such as Indigenous versus non-Indigenous/Mestizo.⁹² In the historiography ultimately subscribing to alternative or strategic liberalism, the underlying binary is Indigenous, peasant or popular versus liberal. This assumption has biased historians against recognising the cases in which popular liberalism is not necessarily alternative, nor merely strategic.

Coda

Examining local governance, Karen Caplan argues that it is not always possible to distinguish pristine customary practices isolated from liberal logics. For instance, in 1848 the barrios of El Mixteco and El Zapoteco in Santa Anna Zegache, Ejutla, Oaxaca, supported different forms of municipal representation. The first wanted to retain the colonial practice of electing one alcalde per barrio; the second, which was numerically bigger, favoured a liberal vote, which made their preferred candidate win since the population of El Zapoteco was greater. Was El Zapoteco's choice self-interested rather than ‘liberal’? Perhaps, but what effects did it have? The Ejutla authorities, concerned with keeping the existing ‘liberal order’, noted that liberal elections in the past had resulted in conflict, riots and even homicides, while allowing one alcalde per barrio mostly kept the peace. El Mixteco probably did not present their case in Tocquevellian language (and nor does Caplan) but, in effect, allowing one alcalde per barrio contained the potential tyranny of El Zapoteco's numeric majority, thus suggesting de facto a dialogue and mutual influence between a territorial (perhaps also ‘ethnic’ in this particular case) form of representation and the numerical argument in favour of a majority vote. Another case identified by Caplan in Oaxaca further suggests the imbrication of the customary and the liberal. In Cacahuatepec in 1849 there existed an ‘Indigenous regidor’ which effectively retained the functions of colonial Indigenous alcaldes, and was backed by the ‘Indigenous class’. The latter, however, also defended their constitutional right not to be obliged to pay money or give personal services to the parish priest. These were hardly staunch defenders of a monolithic old regime. Whatever combinations of ethnic, civil and religious rule occurred, they were taking place in a context where obligations previously imposed by the Church were becoming voluntary; the old regime was no more.

Caplan rightly presents these cases as empirical problems on the ground, arguing that ‘for Oaxacan officials disentangling the liberal from the customary was a daunting and probably impossible task’.⁹³ My point is that they are also conceptual problems and correspond to Palti's argument that what we call ‘the traditional and the modern would imbricate in each other’ to the point of becoming ‘indistinguishable’ in many cases.⁹⁴ In the confrontation and mutual influence of so-called old and new, the dilemmas in both came up. Strictly, none of the involved supported ‘tradition’ per se, not wholesale, instead they all made a political choice to respond to contingent situations with the tools at hand. Practices like these are often considered as examples of the co-existence or hybridity of Indian and liberal, old and new. I propose that instead of focusing on the ambiguity of co-existence or hybridity, which tends to reinforce binary oppositions, we concentrate on questions such as the dilemmas inherent in the problems of representation, legitimacy and order, and the available options to manage them. Both the old and the new had paradoxes to face, but at a basic level certain new logics ultimately had the upper hand.⁹⁵ If we describe them with greater precision, we may find paths out of our current impasse.

The terms ‘co-existence’ or ‘hybridity’ often gloss over important questions. Sometimes hybridity is understood as ostensibly modern principles embedded in old power dynamics; indeed, it could be read as one more instance of instrumental liberalism, positing a prevalence of ‘tradition’ which allegedly explains the ultimate failure of liberalism.⁹⁶ Other uses of juxtaposition and hybridity are ambivalent about which of the components, if any, predominated, and about the extent to which there was a clear rupture with the past or not.⁹⁷ Annino has been criticised mostly for an overemphasis on rupture, but his uses of ‘cultural mestizaje’, ‘syncretism’ and the more recent ‘assimilation’ sometimes remain ambiguous as to whether the old or the new had greater weight.⁹⁸ Other historians, while mostly ambivalent, imply in their conclusions that the ‘inertia of the past’ was the strongest trend.⁹⁹ In fact, liberalisms everywhere produced hybrids, the difference is whether they are acknowledged or not. Latin American hybrids have been either lamented or proudly paraded as our distinguished contribution to world diversity, while the contradictions and hybridities of the Global North for a long time rested unacknowledged as if to preserve the region's status as the original locus of liberalism.¹⁰⁰ I suggest that hybridity is ultimately a red herring.

Together with our balance of the weight of the past, and the force of the nineteenth-century revolutions, there remains the question of when and where can we speak of a convergence of elite and the people. I believe that the empirical case for greater convergence than generally admitted has already been persuasively argued by the literature of liberalism tout court cited throughout this article. It recognises popular defence of local autonomy but contends that such defence was not always divorced from the liberalism espoused by national and regional elites and mediators. Perhaps Annino's strong hypothesis of ‘municipal sovereignty’ spurred by the Cádiz Constitution was a stretch. But it is clear that ‘the new constitutional precepts’ were intertwined with a ‘certain bedrock of consensus to accommodate the pueblos’.¹⁰¹ Additionally, a growing number of studies have shown how those ‘who fought for a federalist system of government did so not because they were parochial or hostile to a liberal political culture but because […] only local, popular governments were able to guarantee the rights and freedoms that independence had promised’.¹⁰² Late in the century, when the balance tipped to centralisation and economic development, ‘national liberalism was incapable, in the end, of reflecting all the things that liberalism had come to mean locally’.¹⁰³

I have sought to make explicit the underlying logics in our discussions of popular liberalism. All three existing interpretations ask how ‘liberal’ popular actors were, thereby running the risk of leaving us trapped in a competition for liberal authenticity. Perhaps a way out of this cul-de-sac is to focus instead on how inclusion and exclusion took place in everyday governance.¹⁰⁴ My hope is that we can find broader interpretations which are free of reifications, binary logics and assumptions of temporal alterity. The promise of the popular does not lie in an egalitarian or Indigenous essence but in the open question of how inclusive our politics are.