Ideas of class1

Ideas de clase

David S. Parker*

Queen’s University (Canada)


Middle class

Abstract: This text was originally published by David S. Parker in 1998 as an introduction to the book The Idea of the Middle Class: White-Collar Workers and Peruvian Society, 1900-1950, published by Penn State University Press. In the book, the author reflects on the union organization of private employees in Lima (Peru) and the origin of the 1924 law that established a legal distinction between empleado (white-collar employee) and obrero (blue-collar worker). Both the “constructivist” school and the linguistic turn argue that social classes are abstractions, inventions of the collective imaginary, that is, ideas that compete in an ideological market. Among the infinite ways of conceptualizing society, only a few images and discourses become common sense, influencing the formation of identities and inspiring laws and public policies. This text affirms and characterizes that the formation of the middle class concept in Peru is due to ideological, discursive and political processes.

Palabras clave

Clase media

Resumen: Este texto fue originalmente publicado por David S. Parker en el año 1998 como introducción al libro The Idea of the Middle Class: White-Collar Workers and Peruvian Society, 1900-1950, publicado por la editorial Penn State University. En él, el autor reflexiona sobre la sindicación de los empleados de Lima (Perú) y sobre el origen de la ley de 1924 que estableció una distinción legal entre empleado y obrero. Tanto la escuela «constructivista» como el giro lingüístico plantean que las clases sociales son abstracciones, inventos del imaginario colectivo, es decir, ideas que compiten en un mercado ideológico. De entre las infinitas maneras de conceptualizar la sociedad solo unas cuantas imágenes y discursos llegan a convertirse en sentido común, influyendo en la formación de identidades e inspirando leyes y políticas públicas. El presente texto afirma y caracteriza que la formación del concepto clase media en el Perú se debe a procesos ideológicos, discursivos y políticos.


* Correspondence author / Correspondencia a: David S. Parker. Queen’s University. Department of History. 212 Watson Hall, 49 Bader Lane, Kingston (Ontario, Canadá K7L 3N6) – –

How to cite / Cómo citar: Parker, David S. (2023). «Ideas of class». Papeles del CEIC, vol. 2023/1, papel 274, 1-16. (

ISSN 1695-6494 / © 2023 UPV/EHU

logo%20CC%20atrib%204_0%20int.jpg This work is licensed under a
Creative Commons Atribución 4.0 Internacional


On a Saturday morning in late September of 1919, shoppers entering Lima’s central retail district found stores closed and shuttered. Few could have been surprised: labor unrest had long ceased to be a novelty in that strike-filled year. In January, anarchist-led workers had paralyzed the city in a general strike that won them the eight-hour workday. In May, demonstrations for lower food prices had erupted into a wave of looting and mob violence, throwing the Peruvian capital into virtual chaos for several days. In July, President-elect Augusto B. Leguía had ousted his predecessor in a military coup rather than trust Congress to ratify his earlier victory at the polls. At each outbreak of violence, store owners habitually closed their doors, prudently hoping to avoid broken windows, looting, or worse.

On that particular morning, however, tranquility reigned. Orderly, courteous, well-dressed young men moved unthreateningly through the streets, stopping to discuss the merits of their cause with employers and passers-by. The government had not condoned their strike, but neither had it sent mounted police to break things up, often the response in such circumstances. Indeed, the National Assembly later approved a unanimous motion in support of the strikers’ demands (El Tiempo, 1919). There was no violence, not even the slightest hint of a threat. Still, shop owners viewed events with alarm, for one good reason. While Lima’s previous strikes had involved factory operatives, stevedores, bakers, or other manual laborers, this time the merchants’ own employees were the ones on the picket lines. For the first time in modern Peruvian history, a strike had been called specifically by empleados, white-collar workers in Lima’s import-export firms, banks, insurance companies, and retail stores.

While the strike by empleados sparked a fair amount of interest at the time, it has since become little more than a footnote (Basadre & Ferrero, 1963; Basadre, 1983; Temoche Benites, 1987). The neglect is unfortunate, because this small, nonviolent action arguably marked a turning point in Peruvian social history. September 1919 was the country’s first white-collar strike ever, and more importantly, it was the first time that any group of demonstrators had taken to the streets in the name of “the middle class”, a concept virtually absent from the discourse of the previous century. Employees made a point to emphasize that theirs was not the cause of the proletariat, but of the forgotten, long-suffering clase media. Underlying their rhetoric was an increasingly clear picture of what it meant to belong to that middle class, an assertion of what separated them from the workers below and from the aristocrats above.

It is important to understand who these strikers were, where they came from, what they wanted, and how they fought to achieve their goals, both in 1919 and over subsequent decades. Embedded in that simple task, however, is the greater challenge of reconstructing the historical emergence of a middle class, not just as a group of people but as an idea. Why did Lima’s white-collar employees paint their specific labor disputes as part of a larger struggle of the middle class? How did employees describe that middle class? What were its supposed qualities, values, and needs? Why did the idea of the middle class enter public debate at that time, rather than two or three decades earlier or later? A set of related questions follow from these: To what extent did other Peruvians accept the employees’ claim to represent a middle class with certain alleged attributes? How did the government respond to demands couched in the language of middle class needs and rights? Finally, how did this emerging idea of the middle class contribute to larger social, political, and cultural changes in twentieth-century Peru? The answers to all of these questions are more complicated than one might first imagine.

The middle class is hardly a new topic for Latin Americanists. As early as 1958, John J. Johnson’s Political Change in Latin America exploded the myth that the nations of Latin America had no middle classes to speak of. Against the popular stereotype of a continent peopled only by landed oligarchs and impoverished masses, Johnson described a region in which large and growing “middle sectors” increasingly dominated public life. Among their political achievements were the promotion of economic nationalism and state-sponsored development, the rapid expansion of public education, the introduction of social welfare policies, and perhaps most importantly, the impulse, albeit halting and imperfect, toward democratization and the broadening of the electorate (Johnson, 1958). This image of progressive, reformist middle sectors had already appeared in earlier writings by Latin Americans, but Johnson’s succinct formulation made a profound impression in both Americas and came to symbolize the vision of mesocratic reformism as one guiding theme of twentieth-century Latin American history (Salazar, 1986).

As a result, the Johnson thesis also served as a lighting rod for critics who rejected his faith in the middle sectors as a force for progressive change. From both the right and the left emerged a common, more pessimistic portrait. Bert F. Hoselitz noted in 1960 that the Latin American countries with the largest middle sectors were also those in which economic stagnation had become most intractible. He reasoned that, unlike the bourgeoisies of Europe or the United States, Latin America’s middle sectors lacked entrepreneurial spirit, promoting consumption and redistribution instead of the creation of new wealth (Hoselitz, 1965). In a frequently-cited 1963 article, Fredrick Pike painted a similar picture of Chile’s early twentieth-century middle class, underscoring its dependence on elite patronage, its lack of cultural autonomy, and its arrogant disdain for the masses (Pike, 1963). By the end of the 1960s, the vision of a dependent middle class had all but completely supplanted Johnson’s optimistic scenario. The new left in particular, inspired by dependency theory and the example of the Cuban Revolution, rallied behind an analysis that denied the existence of that old chestnut of the popular front era: the so-called “national bourgeoisie” (Petras, 1970; Sunkel, 1973). From there it was a short step to the assertion that meaningful social change in Latin America could only come by means of a popular revolution from below.

The dependent and progressive visions marked the poles around which subsequent debate would revolve, as writers in the late 1960s and 1970s attempted to overcome the conceptual impasse. Luis Ratinoff (1967) and José Nun (1968), among others, hoped to solve the contradiction by periodizing it, arguing that the middle sectors had at one time indeed been the progressive, reformist force that Johnson had described, but once they had sucessfully assaulted the fortress of oligarchic control (between about 1910 and 1945, depending on the country), these middle sectors abandoned their insurgent mentality and took up the task of consolidating their victory. The former champions of “the people” against “the oligarchy” now found reason to disavow and even suppress those workers and peasants who sought to follow in their footsteps: hence their support for military coups in countries like Brazil, Argentina, and Chile. Luis Costa Pinto (1964) and Jorge Graciarena (1967) sought to overcome the progressive-dependent conflict not by periodization but by classification. Their central conceit was a distinction between “residual” and “emergent” middle sectors. The residual middle sectors (read dependent) included public employees, liberal professionals, small landowners and urban rentiers; they were largely recruited from families of old lineage, enjoyed patronage ties to that traditional elite, and were economically and socially stagnant if not downwardly mobile. The emergent middle sectors (read progressive) included small businessmen, teachers, and salaried employees in the modern sector of the economy; they were largely recruited from the common people, lacked connections to the elite, and were on balance upwardly mobile.

In the 1960s and 1970s, however, major strikes by bank and other white-collar workers seriously challenged both the progressive and the dependent stereotypes, once again recasting the debate in new terms. A few writers looked at the turn by ostensibly middle-class employees to radical forms of trade unionism and Communist, Trotskyist or even Maoist ideologies, and saw a textbook illustration of class polarization. While the higher middle strata had fused with the bourgeoisie, they argued, the lower strata had suffered acute proletarianization, effectively joining the ranks of the working class (Germaná & Westwell, 1983). Other analysts, unwilling to accept the idea that the middle class had disappeared in Latin America, focused instead on the imbalance between expanding public education and stagnant employment opportunities. In this formulation, huge numbers of new high school and university graduates, many of modest social origins, had believed that their diplomas would guarantee them upward mobility and ensure their access to non-manual jobs and middle-class lifestyles. Finding instead that the massification of public education had devalued their credentials and thwarted their aspirations, they turned to ideas of revolution with a fervor born of betrayal (Angell, 1982; Filgueira & Geneletti, 1981).

If scholars of the 1960s and 1970s found in Latin America’s middle class everything from modernizing democratic reformers to status-obsessed clients of the aristocracy to frustrated revolutionaries, what are we to do today, as the very concept of class has come under increasing attack? Two major difficulties immediately come to mind. First, of course, is the problem of definition. What is the middle class? How is it possible to describe the class characteristics of groups as diverse as shopkeepers, lawyers, teachers, stockbrokers, medium-sized farmers, bureaucrats and white-collar employees? Johnson deliberately used the term “sectors” rather than “class” in order to underline this heterogeneity, but his choice merely begged the question: If the middle sectors were too heterogeneous to be a class, could anything at all useful be said about them? The second, more basic problem arises from the temptation to conceptualize classes as concrete, conscious historical actors. Was there, to cite a well-known analogy, such a group as “the bourgeoisie” in eighteenth-century France and did it bring about the French Revolution? A generation of revisionist scholarship has answered in the negative. In the process, as William Reddy pointed out, “the question that is raised is whether it is possible to continue to speak of socially distinct sets of individuals, united by some identifiable trait or traits, as having shared intentions” (1987: 8). At very least we must reject the idea that social classes speak with a single voice, act as a single individual, or play a single role in the shaping of historical destiny. Classes are, at the end of the day, abstractions that we reify at our peril. Splitting classes into “residual” and “emergent”, “bourgeois” and “proletarianized”, “old” and “new” cannot solve this fundamental epistemological problem.

In the case of Latin America, precious little seems to unite the millions of lives we tend to lump together statistically as middle-class. The closer one looks, the greater the differences appear: immigrant versus native-born, white versus mestizo, salaried versus independent, public sector versus private sector, university-trained versus self-taught, urban versus rural. These distinctions create an enormous spectrum of living standards, consumption patterns, cultural tastes, social circles, economic interests, and political affiliations. If our social history is a study of daily life, then there can be no such thing as a single social history of the middle class, for Peru or anywhere else. We would need a hundred different histories before a sufficiently complete picture emerged.

Is anything gained, therefore, by using the concept of class, specifically “the middle class”, to tell the social history of clerks and bureaucrats in twentieth-century Lima? In my view, the answer remains an emphatic yes. For three-quarters of a century, Peruvians have routinely employed the concept of “the middle class” with a broadly-accepted meaning and clear, vivid connotations. Had a sample of ordinary Peruvians been interviewed in 1920, 1950, or 1980 —at least in urban areas— they almost certainly could have identified quintessentially middle-class occupations, neighborhoods, schools, restaurants, theaters. Some people saw the term as a badge of honor, others did not, but nowhere did “the middle class” want for comprehension. For all its objective heterogeneity, the middle class was and is a palpable cultural reality.

From whence did that palpable cultural reality, that widely-shared understanding of what the middle class meant, arise? I would argue that Peruvians’ idea of the middle class was basically an invention of the early twentieth century. At that time, white-collar workers and other similar occupational groups chose to identify themselves explicitly as members of the middle class, in part to gain a sense of identity, but also to lend legitimacy to their fight for social legislation. In so doing, they created and diffused a new vision of what it meant to be middle-class in Peru. Why the early twentieth century? Because these were the years when the so-called “social question” first appeared on the radar screen of Peru’s public debate and private consciousness. These were the years when literate urban Peruvians for the first time started talking and thinking about their society as divided along socioeconomic lines. These were the years when Peruvian governments took their first steps toward formulating social policy and intervening in social conflict. In other words, these were the years when people first began to think in class terms and to legislate as if classes existed.

From Europe Peruvians imported new ideas and a new vocabulary, which they adapted to local circumstances and superimposed onto prior beliefs. The novelty of the social question gave those ideas a fluid, unfixed quality that would only last a short time. Employees benefitted from that novelty, finding that their claims, their myths, and their vocabulary drifted easily into the public debate. Their ideas joined a chorus of competing voices, but they also won popular acceptance to a degree that few could have anticipated. As people began to think and talk regularly about the middle class, and as those ideas increasingly influenced government policy, the Peruvian middle class was, in a sense, “made”: tentatively and impermanently, but made nonetheless.

1. Ideas of Class

Telling the story of the Peruvian middle class in this way betrays my attempt to strike a balance between the material forces that give rise to social conflict and the role that ideas play in making that conflict intelligible and real to ordinary people. As social historians we need to examine concrete battles over wages, benefits, working conditions, and standards of living, but we must also, equally, explore the language and concepts people employed as they prosecuted those battles. This relationship between structure and discourse has become a growing source of controversy, as poststructuralism and the “linguistic turn” of social history have simultaneously found favor and sparked opposition. Let me briefly stake out some of the positions that inform my approach.

First of all, most scholars would now agree that while individual and collective battles over the appropriation of wealth, the distribution of power, and the terms of labor are a permanent facet of daily life, overtly class-conscious collective action is the historical exception. From the ordinary worker’s perspective, co-employees may be comrades in struggle, but they can also be competitors for promotion. Bosses may be antagonists, but they can also be fatherly guardians. Supervisors may be agents of control, but they can also be godparents, friends, or leaders. Race, nationality, gender, seniority, job category, and a host of other differences complicate identification of an “us” versus a “them”. Historians have therefore rightly jettisoned the idea that class consciousness is a normal state of affairs, and that the lack of class consciousness is an aberration that needs special explanation.

Social history has benefitted from this realization. It has allowed proponents of the idea of “class formation” to emphasize the unevenness, historical contingency, and local specificity of the process by which artisans and ragpickers evolved into “workers” (Katznelson & Zolberg, 1986). Reveling in the richness of lived experience, students of class formation have rejected the narrow idea of class consciousness for the much broader notion of a class culture, and have freed that notion from rigid economic determinism. The crisis of the concept of class consciousness has also sparked the growth of subaltern studies, a field dedicated to the ambiguities of power and subordination. With their emphasis on inchoate ideas of opposition and everyday forms of resistance, subaltern studies have shown that something very much resembling class struggle can flourish without trade unions, without collective action, without class ideologies, even without clearly-defined classes (Mallon, 1994; Scott, 1985, 1990). Most importantly, the decline of this mechanical perception of class consciousness has nurtured a long, heated, but fruitful controversy over the power of ideology to influence social relations and to change the outcome of material conflicts. In the sense that linguistic approaches to social history are also primarily about the power of ideology, they are the latest round in a debate that traces back to Althusser, Gramsci, and Marx (Abercrombie, Hill & Turner, 1980).

It is essential to pay close attention to the role of ideology in shaping how people interpret the world, and consequently, how they live their lives within it. Ideas do not merely reflect power, they are power. This recognition depends, however, upon three prior suppositions. First, “ideology” needs to be defined in the broadest possible manner. Far beyond explicitly political ideas, ideology extends to the entire universe of thoughts, perceptions, prejudices, stereotypes, habits, and language that form the essence of subjectivity. Ideology provides the organizing logic by which people comprehend an otherwise incomprehensible “reality”; it gives meaning to their surroundings, forms their sense of identity, and profoundly influences their actions (Bourdieu, 1985, 1990). Second, ideology must be understood not as a hegemonic imposition from above, but as a collective enterprise of construction, a dialogue in which a significant amount of conflict is taking place. Third, while admitting that ideology shapes history, we should never go so far as to think that it is the only force that shapes history.

A coherent study of “ideology” so broadly defined and so largely writ borders on the impossible. I therefore limit my scope to class ideologies, starting from one simple premise: classes, like other abstractions, are products of the mind (Furbank, 1985; Wahrman, 1995). Because society is so extraordinarily complex, it can be described in an almost infinite number of ways; therefore, when people identify themselves as members of a particular social class, they are making profound ideological choices, rooted in a string of assumptions about the nature of society and their place in it. A class identity, no matter how deeply it seems to be rooted in material experience, requires a vision of what classes are and what one’s own class looks like. What are a class’s defining characteristics? Who belongs to a particular class and who does not? Who are its allies and who are its enemies? What does the class need, what does it deserve? None of these are simple questions, and there are no inevitably “right” answers. These ideas of class are invented constructs that serve ideological ends: they place people in an imaginary hierarchy, exalting some and stigmatizing others, and they negotiate the rules by which some people deem themselves better than the rest.

Ideas of class must compete for acceptance; they must appeal to those whom they would unite by explaining reality in a convincing way. Like all ideas, ideas of class have their producers and consumers. They may be created by intellectuals, by opinion-makers in the media, or by potential leaders hoping to build a base of support. They are often propagated by “interest entrepreneurs” who invest their time and effort in organizing a specific group, a group to which they themselves may or may not personally belong (Salisbury, 1969). And while it might be possible in certain rare historical moments for ideas of class to spring up spontaneously —as when the violent repression of a strike engenders the camaraderie of the barricades— more typically they emerge and compete with little fanfare, in newspapers, in public debate, in conversation, and in the private thoughts of ordinary men and women.

In order to win acceptance, ideas of class must successfully appear to reflect everyday experience. They must be consistent with collective identities, the sense of “us” versus “them”, that are being forged on a daily basis among people who carry out the same tasks, earn the same pay, and suffer the same indignities. Ideas of class must have resonance and explanatory power; they cannot simply be decreed from above. Beyond that, ideas are unlike any other product: the consumer of ideas transforms them in his or her own mind, constantly shaping and reshaping them until they mesh with pre-existing attitudes and experiences. In other words, the generation of ideas is a uniquely collective enterprise, where consumers are at the same time producers. To that extent, but to that extent only, class identification arises from below as a reflection of shared experience (Joyce, 1991).

Ideas of class face tremendous competition in the marketplace of ideas. There are opposing collective identities grounded in religion, gender, ethnicity, and so on, just as there are visions of society that deny altogether the existence of classes or class conflict (Bourdieu, 1985). Even workers in evident struggle against oppression have often eschewed the rhetoric of class, choosing instead to define themselves simply as “the people” (Joyce, 1991). Ideas of class also compete amongst themselves. Centuries of social theory have generated a wide range of concepts and definitions, of which the Marxist division of society into bourgeoisie and proletariat, or the Aristotelian three-class model (upper, middle, and lower), are but two of the more influential. Ideas of class may draw upon previous ideological traditions of rank and hierarchy, as with seventeenth-century British conceptions of “better sorts”, “middling sorts” and “common sorts”, or they may be tied to the special legal privileges of a society of estates, as in pre-Revolutionary France or colonial Latin America (Wrightson, 1991). Beyond that, even apparently similar images of class can mask very different operating rules: people may perceive class membership as a function of income, occupation, race, education, family name, consumption, residence, or a combination of attributes.

Being ideological creations, ideas of class gain or lose currency as a result of ideological struggles. When some people deny that class barriers exist, or others contend that social inequalities reflect intrinsic differences in ability or character, their arguments serve an agenda. The same can be said for those who assert that class distinctions are purely arbitrary impositions, or that they are uniquely determined by the division of labor. Which ideas win out has much to do with who has the power to impose their vision over the competing voices of others. It is probably safe to say, moreover, that in most modern societies, ideas propagated by the rich and powerful are more likely to gain the spotlight than the ideas of the poor and weak: the former are able to exercise influence through a variety of channels.

Yet the construction of ideology is by no means a unilateral process of imposition and acceptance, hegemony and socialization. Ideological battles are a dialogue —a dialogue between unequals, to be sure, but a dialogue nonetheless—. Not only do subordinate groups frequently contest and reject those ideas with which they disagree, they also appropriate elements of the dominant discourse and use them in new ways to challenge accepted truths. Alternatively, the powerful regularly co-opt the arguments of the opposition, recasting those ideas in less dangerous terms, but at the same time increasing their legitimacy. This dialogue is almost always disorderly, inconclusive, and only partially understood by those involved. Ideological change comes piecemeal, rather than in leaps and bounds (Gardiner, 1992; Steinberg, 1994). To further complicate matters, the dialogue is not two-sided, but multisided. Innumerable voices are clamoring to be heard; some succeed, others do not. The point is that ideology matters, and the conflict of ideas has a great deal of autonomy from the many other forms of conflict.

This dialogue of ideological competition is mediated by language. The words and concepts that exist to describe the existing order are essential building-blocks of ideology: to manufacture consent, the powerful seek to create a world in which their vocabulary and their definitions are the only ones used (Fairclough, 1989). Because of this, literary critics have in recent years engaged in an active debate over whether language is really a transparent tool of expression, or if it is instead an opaque, immutable structure that is in itself an ideology, controlling or at least distorting the very ideas expressed. I side with the contention, typically attributed to Bakhtin, that language is neither of these: language is not a fully open system, in which people can create any discourse they like, but neither is it a closed system that pre-empts people’s abilities to say what they want to say (Holquist, 1990).

On the one hand, words come with a great deal of historical baggage. That baggage, made up of past meanings and usages, accepted definitions, and others’ use of the same words, cannot easily be jettisoned or subverted. This is particularly true for emotive political words such as “freedom”, “democracy”, “equality”, and “rights”, but it is no less the case for the vocabulary employed to describe the social order. People are handed an already-constituted language with which to work; the culturally-understood meanings of existing signs place a limit on people’s ability to signify new things. In other words, some ideas may be impossible to formulate because the words and concepts with which to express them do not exist.

But on the other hand, language is not a closed and immutable system (Steinberg, 1994). There is enormous diversity within the shared community of language, and even greater room for movement over time. Different people employ the same words with entirely distinct meanings and connotations, and those meanings can and do compete with one another. Old terms are appropriated and used in new ways. New meanings for old words and even entirely new words are invented on a regular basis. Concepts, vocabulary, and definitions also cross boundaries from one language to another: scholars of Latin America in particular have long understood the importance of ideological and linguistic borrowing, particularly from Europe and North America. All of these mechanisms come into play in the case of ideas of class.

Finally, these symbolic struggles over language and ideology can and do concretely affect the course of social conflicts, their outcomes, even the winners and losers. Specifically, ideas and vocabularies of class contribute in absolutely concrete ways to the constitution of real historical actors. This assertion, while by no means original, is still controversial (Joyce, 1991; Palmer, 1990)2; the point is best made with examples. First of all, ideas of class delineate the fault lines of collective organization. Is the working class limited to manual workers, or does it include all who sell their labor for a wage? Does it include the unskilled occasional worker? The poorly paid clerical worker? The unemployed? Women? The character and scope of labor unions may vary enormously depending on how their members answer those questions. Second, ideas of class have an impact on the kinds of demands that groups make, and the language in which they express those demands. A vision of the working class that includes factory foremen will be difficult to reconcile with demands to abolish foremen’s special privileges; by the same token, an idea of the working class that includes office personnel will be incompatible with a rhetoric emphasizing workers’ role as the producers of goods. Gendered ideas of work as inherently “manly” can lead unions to ignore or marginalize the demands of women, sometimes even in spite of their leaders’ conscious intent (Wolfe, 1993).

Third, and perhaps most important, ideas of class are claims on society. The invented image of what a class looks like contains explicit arguments and implicit assumptions about what that class should look like, the rights it should enjoy, the duties it should fulfil to others, and the benefits it rightly deserves. Such claims are a crucial element of class struggles, for if they achieve legitimacy, they can inspire new laws or other equally tangible changes. To cite a classic example, debates over minimum wage legislation typically invoke competing definitions of an “acceptable” minimum living standard. Scientific pretensions aside, this definition ultimately hinges upon an idea of class par excellence: the socially and ideologically constructed image of what a worker should and should not be expected to consume.

2. White-Collar Workers and the Middle Class

Critics of the linguistic turn in social history have argued that we err in overstressing the importance of language and ideology as deciding factors in class formation. They point out that the harsh material reality of working-class life shaped workers’ experience and culture, forming them as a class more surely than any discourse ever could. Whether or not workers talked about themselves as a class, the argument goes, this common experience and culture, born out of deprivation and struggle, surely made them one (Palmer, 1987). I concede that this may quite possibly be the case, not because linguistic approaches to the study of class formation are inherently wrongheaded, but because manual workers are so typically subordinated on many fronts, each of which reinforces the others. Manual workers generally earn low incomes, lack autonomous control over the labor process, often suffer degrading working conditions, have few opportunities for education or advancement, frequently live in the poorest and most unhealthy circumstances, and usually enjoy an inferior social status, sometimes compounded by racial or ethnic prejudice. One can reasonably argue that it does not matter in the slightest whether manual workers think of themselves as a class or not. If they do define themselves as a class, it may not matter whether they define that class in terms of income, occupation, power, lifestyle, or prestige, nor may it make a big difference exactly where they draw class boundaries. Whichever way they conceptualize the social order, workers remain near the bottom, looking up. For perhaps a majority of workers the ideological dimension of class consciousness remains relatively unimportant, simply because the material facts of poverty and subordination are so self-evident and omnipresent.

It is difficult to say the same thing about white-collar employees. First there is the question of heterogeneity. While manual workers as a group are enormously diverse —a fact that most labor historians now recognize— this is only more the case for white-collar workers, who can be found everywhere from the typing pool to the top executive office. The complexity of management hierarchies and the fluidity of career patterns underscore the absence of anything even remotely resembling a common class experience rooted in material circumstances. Compounding this situation is the fact that for each individual white-collar employee, the fit between power and influence, quality of work environment, income, and social status is hardly ever neat and tidy (Lockwood, 1958). Compared with manual workers, white-collar employees are much more likely to rank high on some counts but low on others, further fragmenting the group and impeding the identification of a shared class culture.

Where the material bases of class experience are so diffuse and ambiguous, the ideological component of class formation takes on greater importance. As the rules of class structuration are up for grabs, symbolic conflict over the representation of social reality becomes a struggle with high stakes (Bourdieu, 1990). Moreso than manual workers, white-collar workers need ideas of class to make sense of their circumstances, and find that those ideas fundamentally shape their organizational practices, their demands, their strategies, and their rhetoric. For this reason above all others it is important to balance any study of employees’ material lives with an equally intense examination of how white-collar workers themselves debated, adopted, and promoted their own class identity.

If any idea of class has historically appealed to non-manual, white-collar workers, it is the idea that they form part of the middle class. The identification is by no means inevitable, but it enjoys a certain logic: the bulk of employees find themselves somewhere between the highest and lowest ranges of company hierarchies, their income and living standards often place them closer to the societal average than to either extreme, and they are typically better off than some but worse than others in terms of prestige and social status. Beyond that, the idea of the middle class appeals to a common sense of being neither fish nor fowl, neither sybaritic aristocrat nor part of the poverty-stricken underclass.

However, the feeling of “middleness”, no matter how deep and sincere, is not enough to provide a coherent idea of class on which to construct a collective identity. Even more than in the case of working-class consciousness, ideas of the middle class must come to terms with the complexity of the social structure. Ideas of the middle class must answer hard questions: Where are the boundaries of that class? By what criteria do those boundaries exist? A middle class may define itself by income, occupation, education, family name, lifestyle, race, or something else. The conceptual line between the middle class and the upper class can be drawn almost anywhere: the CEO of a Fortune 500 corporation is obviously not of the middle class, but how about the vice president in charge of operations? The assistant head of personnel? If income is the standard, where is the line? At 250,000 dollars per year? $100,000? $60,000?3 The difficulties are obvious and have real effects. At what level should tax cuts aimed at “the middle class” be capped? Which executives should be barred from joining employee unions? Moreover, all the same problems are replicated at the other end, at the dividing line between middle class and working class or lower class. It is just this kind of indeterminacy that led a full 80 percent of Americans to identify themselves as middle-class in a 1940 survey. Such a survey does not say much about the U.S. social structure: it speaks loudly, however, about American class ideology (Rodgers & Wilenz, 1991).

To sum up, ideas of the middle class are by no means determined simply by the objective structures of occupation, income, or status. Not only do distinct images of class compete within any given society, but the outcomes of those struggles are likely to differ greatly from one place to the next. Each nation has its own unique history of economic development, racial and ethnic tension, social conflict, political competition, and ideological legitimation. We should hardly expect countries with such diverse experiences to arrive at identical or even similar visions of what the middle class is, where its boundaries lie, or what qualities it embodies. Thus, models of the middle class developed in Europe or North America are quite likely to be inapplicable in Latin America, a region with very different economic, demographic, political, social, legal, moral, religious, and cultural traditions. In order to understand both the material and the ideological forces that shaped Peru’s emerging middle class, it is necessary to shine a spotlight on how Latin America’s unique cultural heritage, combined with its distinct economic and social structures, led to a process of class formation that diverged widely from the experience of the industrial, Anglo-Saxon West.

3. The Case of the Peruvian Empleado

Throughout the early twentieth century, Peruvians both explicitly and implicitly debated the nature of the middle class, its boundaries, its interests, and its rights and duties. From the very beginning, the central protagonists in this debate were empleados de comercio, employees in the import-export houses and retail stores of Lima and its nearby port of Callao. They were joined by white-collar workers in banks, insurance companies, and some mining, agricultural, and industrial concerns. The Sociedad Empleados de Comercio (SEC), formed in 1903 as a mutual-aid society, became an important defender of employee interests, and in so doing, played a central role in defining Peruvians’ idea of the middle class. Employees were increasingly recognized as the leading sector of the middle class, and by the 1920s they had earned acceptance as natural spokesmen for the class as a whole.

Lima’s commercial employees were ideally suited to carry the banner of the middle class. On the one hand, most Peruvians agreed that empleados were inherently different from manual workers, or obreros4. The distinction was recognized by obreros themselves —even a Communist union activist could recall the nervousness and distance he felt when asked to speak before an audience of empleados in 1918: “I found myself for the first time in such a situation. The streetcar drivers were empleados. They were always well dressed; they had even elected a Congressman” (Portocarrero, 1987: 52)—. On the other hand, few empleados could consider themselves part of the elite, though many were distant relatives of the “best families”. Differences in income, education, background, and lifestyle separated Peru’s employees from the ranks of the truly privileged. Other groups with roughly the same social status as commercial employees might have led a movement in the name of the middle class but for a variety of reasons did not. Public employees, for example, enjoyed notoriously precarious job tenure, which undermined their capacity for independent action, at least in the years prior to around 1940. Most shopkeepers and small businessmen came from one of several distinct immigrant communities, making them enormously difficult to organize. Even when they did unite to voice their concerns, they rarely spoke of themselves as a class.

So partly by design, partly by default, commercial employees in Lima and Callao became the first to organize on behalf of the middle class and were the first to propagate a specific vision of middle-class qualities and interests. To understand just how significant a group are we talking about, it is important to underscore the relative weakness of Peru’s industrial base at the beginning of the twentieth century. When economist Alexander Garland wrote in 1906 that “Peru is not a manufacturing country” (1907: 211), he only simplified to a degree. The tenuous industrial growth that Peru had experienced in the 1890s was short-lived, and gave way in the 1900s and 1910s to a major commercial boom, fueled by high world prices for Peru’s exports and an ever more abundant supply of reasonably priced imports (Bardella, 1964; Yepes del Castillo, 1972). The Peruvian elite, moreover, demonstrated a clear preference for imported goods, so much so that a newspaper in 1926 suggested placing counterfeit foreign labels on locally-made products in order to increase their marketability (Herbold, 1973). What industrialization did take place centered around products that were difficult to import because they were perishable or heavy: processed foods and beverages, construction materials, furniture. Even textile manufacturing, the notable exception to this rule, was largely owned and operated by the large merchant houses, the same companies that exported Peruvian cotton and imported Manchester cloth (Thorp & Bertram, 1978)5.

But if Peru was not an industrial nation, neither was it simply backward and undeveloped. The lucrative import-export business underwrote the modernization of Lima, and merchant capital generated some immense private fortunes. While many of the largest commercial houses first arrived in Peru during the guano export boom from the 1840s to the 1870s, the twentieth century sparked a new wave of prosperity and expansion (Chavarría, 1979; García Calderón, 1907; Miró Quesada Sosa, 1946; Romero, 1951). By 1920, Lima was visibly changed; by 1930, it was hardly recognizable. Commercial houses formerly operating out of tiny storefronts now built massive, sumptuous offices. In 1917, for example, a German shopkeeper named A.F. Oechsle opened an elegant four-story department store, the first of its kind in Peru, on the Plaza de Armas looking out toward the National Cathedral. Boasting elevators, all the modern conveniences, and the last word in European styles, this shrine to consumption provided work for over sixty empleados (Ilustración obrera, 1917). Import-export firms like Gildemeister and W.R. Grace built equally grand offices, the largest covering half a city block and employing hundreds (Partido Democrático Reformista, 1935). Banks and insurance companies grew even more rapidly, cementing Lima’s transformation from a sleepy colonial capital to a bustling commercial center.

Banks and commercial firms were among the largest employers in Lima, and white-collar employees made up a significant part of the city’s working population. Census data for 1908, 1920, and 1931 placed the number of commercial employees —depending upon how the group was defined— between 15 and 25 percent of the adult male population (Parker, 1990). The commercial white-collar workforce becomes even more significant when compared to the numbers of industrial workers. As late as 1931, small workshops and artisanal forms of organization dominated in every sector of production except textiles and food processing. Construction and agriculture remained the two largest manual occupations, illustrating the extent to which Peru’s blue-collar workers remained peones, toiling in unstable and frequently unskilled jobs with little chance of advancement or collective organization. This is not to say that Peru’s obreros did not influence the society in which they lived —quite the contrary— but rather that the primarily commercial orientation of the Peruvian economy gave workers in commerce and banking a greater voice and impact than most scholars have realized.

When those empleados first walked off the job in 1919, they could not have imagined the impact they would have on Peruvian society, yet their impact was arguably immense. All they wanted was an eight-hour day, higher salaries, and a longer lunch break, but in the pursuit of those goals they ended up achieving a great deal more: they invented a new conception of their social order, in which the clase media figured prominently. Their vision employed the established stereotypes of a hierarchical society, ordered by family name, prestige, and “respectability”, but recast those concepts into a modern explanation of why empleados, as born members of the middle class, needed and deserved certain exclusive benefits. Little by little the empleados’ arguments, championed by representative organizations like the SEC, filtered into Peru’s political discourse and captured public opinion to a degree that would have been hard to imagine even a decade earlier.

In the mid-1920s, this idea of the middle class succeeded in inspiring precedent-setting social legislation, which gave new legal force to the traditional distinction between manual workers (obreros) and non-manual empleados. This obrero-empleado division, a central and enduring feature of Peruvian labor law, cannot be understood except as an outgrowth of the empleados’ vision of a middle class defined by non-manual work and intrinsically, almost biologically, distinct from the working class. This castelike idea of the middle class survived surprisingly intact through the 1930s and 1940s, and continued to inform Peruvian government policies on everything from unemployment relief to summer vacations. At the same time, the middle class became a significant ingredient of Peru’s public debate, as emerging leaders and political movements, most notably the APRA Party of Víctor Raúl Haya de la Torre, adopted elements of the employees’ rhetoric in their official platforms. Distinct and even contrary ideas of class never completely disappeared: APRA leaders sometimes talked about the middle class but at other times talked about “manual and intellectual workers”, while Peruvian Communists promoted the belief that employees were destined to join the proletariat. Yet to a surprising degree the conception of the middle class that white-collar employees had created continued to set the agenda for subsequent discussion of the nature of Peruvian society. Ideas that were new and dangerous in the 1910s became the hackneyed clichés of the 1940s and 1950s, clichés that often still echo through the present.

It is for this reason that we should give the white-collar strikers of 1919 their rightful place in Peru’s social and cultural history. Because they invented a new idea of the middle class, empleados changed the way that ordinary Peruvians, rich and poor, imagined and described their social order. Because their idea of the middle class influenced a generation of lawmakers, empleados set into motion a chain of events that permanently altered their nation’s juridical history. Because the middle class played a central part in Peru’s early partisan discourse, empleados contributed to the emergence and development of twentieth-century Peruvian politics. The invention of the middle class was part and parcel of the making of modern Peru.

4. References

Abercrombie, N., Hill, S., & Turner, B. S. (1980). The Dominant Ideology Thesis. London: Allen & Unwin.

Angell, A. (1972). Politics and the Labour Movement in Chile. London: Oxford University Press.

Angell, A. (1982). Classroom Maoists: The Politics of Peruvian Schoolteachers Under Military Government. Bulletin of Latin American Research, 1(5), 13-15.

Bardella, G. (1964). Setenta y cinco años de vida económica del Perú: 1889-1964. Lima: Banco de Crédito del Perú.

Basadre, J. (Ed.) (1983). Historia de la República del Perú, 1822-1933. Lima: Editorial Universitaria.

Basadre, J., & Ferrero, R. (1963). Historia de la Cámara de Comercio de Lima. Lima: Editorial Santiago Velarde.

Bourdieu, P. (1985). The Social Space and the Genesis of Groups. Theory and Society, 14(6), 723-744.

Bourdieu, P. (1990). In Other Words: Essays Toward a Reflexive Sociology. Stanford: Stanford University Press.

Chavarría, J. (1979). José Carlos Mariátegui and the Rise of Modern Peru, 1890-1930. Albuquerque: University of New Mexico Press.

Costa Pinto, L. (1964). Estructura de clase y cambio social. Buenos Aires: Paidós.

Davis, S. M. (1972). Empleados and Obreros. In S. M. Davis & L. W. Goodman (Eds.), Workers and Managers in Latin America (pp. 31-34). Lexington, Mass: D.C. Heath.

El Tiempo (1919). El Tiempo. Lima, 19 December.

Fairclough, N. (1989). Language and Power. London: Longman.

Filgueira, C., & Geneletti, C. (1981). Estratificación y movilidad ocupacional en América Latina. Santiago: Cuadernos de CEPAL.

Furbank, P. N. (1985). Unholy Pleasure, or the Idea of Social Class. Oxford: Oxford University Press.

García Calderón, F. (1907). Le Pérou contemporain. Paris: Dujarric et Cie.

Gardiner, M. (1992). The Bakhtin Circle and the theory of ideology. In The Dialogics of Critique: M. M. Bakhtin and the Theory of Ideology (pp. 59-98). London: Routledge.

Garland, A. (1907). Peru in 1906. With a Brief Historical and Geographical Sketch. Lima: La Industria.

Germaná, C., & Westwell, P. (1983). The Middle Strata and the Problem of Class Alliances. Latin American Perspectives, 10(2-3), 171-184.

Graciarena, J. (1967). Poder y clases sociales en el desarrollo de América Latina. Buenos Aires: Paidós.

Herbold, C. F. (1973). Developments in the Peruvian Administrative System, 1919-1930: Modern and Traditional Qualities of Government Under Authoritarian Regimes [Unpublished doctoral dissertation (PhD)]. Yale, Yale University Press.

Holquist, M. (1990). Language as dialogue. In Dialogism: Bakhtin and his World (pp. 39-64). London: Routledge.

Hoselitz, B. F. (1965). El desarrollo económico en América Latina. Desarrollo Económico, 2(4), 49-65.

Ilustración Obrera (1917). Ilustración Obrera, 77, 13-21. Lima, 25 August.

Johnson, J. J. (1958). Political Change in Latin America: The Emergence of the Middle Sectors. Stanford: Stanford University Press.

Joyce, P. (1991). Visions of the People: Industrial England and the Question of Class, 1848-1914. Cambridge: Cambridge University Press.

Katznelson, I., & Zolberg, A. R. (Eds.). (1986). Working-Class Formation: Nineteenth-Century Patterns in Western Europe and the United States. Princeton: Princeton University Press.

Lockwood, D. (1958). The Blackcoated Worker: A Study in Class Consciousness. London: Allen & Unwin.

Mallon, F. E. (1994). The Promise and Dilemma of Subaltern Studies: Perspectives from Latin American History. American Historical Review, 99(12), 1491-1515.

Miró Quesada Sosa, A. (1946). Lima: ciudad de los reyes. Buenos Aires: Emecé.

Nun, J. (1968). A Latin American Phenomenon: The Middle-Class Military Coup. In J. Petras & M. Zeitlin (Eds.), Latin America: Reform or Revolution? A Reader (pp. 145-185). Greenwich, Conn: Fawcett.

Palmer, B. D. (1987). Response to Joan Scott. International Labor and Working-Class History, 31(1), 14-23.

Palmer, B. D. (1990). Descent into Discourse: The Reification of Language and the Writing of Social History. Philadelphia: Temple University Press.

Parker, D. S. (1990). The Rise of the Peruvian Middle Class: A Social and Political History of White-Collar Workers in Peru, 1900-1950 [Unpublished doctoral dissertation (PhD)]. Stanford, Stanford University.

Partido Democrático Reformista (1935). Lima 1919-1930. Lima: Partido Democrático Reformista.

Petras, J. (1970). Politics and Social Structure in Latin America. New York: Monthly Review Press.

Pike, F. B. (1963). Aspects of Class Relations in Chile, 1850-1960. Hispanic American Historical Review, 43(2), 14-33.

Portocarrero, J. (1987). Sindicalismo peruano: primera etapa 1911-1930. Lima: Labor.

Ratinoff, L. (1967). The New Urban Groups: The Middle Classes. In S. M. Lipset & A. Solari (Eds.), Elites in Latin America (pp. 61-93). New York: Oxford University Press.

Reddy, W. M. (1987). Money and Liberty in Modern Europe: A Critique of Historical Understanding. Cambridge: Cambridge University Press.

Rodgers, D. T., & Wilenz, S. (1991). Languages of Power in the United States. In P. J. Corfield (Ed.), Language, Class, and History (pp. 240-263). Oxford: Basil Blackwell.

Romero, F. (1951). Evolución industrial y educación técnica. Lima: Editorial Hora del Hombre.

Salazar, G. (1986). Documento de Trabajo, nº 60. Para una historia de la clase media en Chile. Santiago: LOM.

Salisbury, R. H. (1969). An Exchange Theory of Interest Groups. Midwest Journal of Political Science, 13(2), 11-12.

Scott, J. C. (1985). Weapons of the Weak: Everyday Forms of Peasant Resistance. New Haven: Yale University Press.

Scott, J. C. (1990). Domination and the Arts of Resistance: Hidden Transcripts. New Haven: Yale University Press.

Stedman Jones, G. (1983). Rethinking Chartism. In Languages of Class: Studies in English Working Class History, 1832-1982 (pp. 90-178). Cambridge: Cambridge University Press.

Steinberg, M. W. (1994). The Dialogue of Struggle: The Contest over Ideological Boundaries in the Case of London Silk Weavers in the Early Nineteenth Century. Social Science History, 18(4), 508-515.

Sunkel, O. (1973). Transnational Capitalism and National Disintegration in Latin America. Social and Economic Studies, 22(3), 132-176.

Temoche Benites, R. (1987). Cofradías, gremios, mutuales, y sindicatos en el Perú. Lima: Escuela Nueva.

Thorp, R., & Bertram, G. (1978). Peru 1890-1977: Growth and Policy in an Open Economy. New York: Columbia University Press.

Wahrman, D. (1995). Imagining the Middle Class: The Political Representation of Class in Britain, c. 1780-1840. Cambridge: Cambridge University Press.

Wolfe, J. (1993). Working Women, Working Men: São Paulo and the Rise of Brazil’s Industrial Working Class. Durham: Duke University Press.

Wrightson, K. (1991). Estates, Degrees, and Sorts: Changing Perceptions of Society in Tudor and Stuart England. In P. J. Corfield (Ed.), Language, Class, and History (pp. 30-52). Oxford: Basil Blackwell.

Yepes del Castillo, E. (1972). Perú 1820-1920: un siglo de desarrollo capitalista. Lima: Instituto de Estudios Peruanos.

1 This text has been slightly modified so that it can be read as an independent work. The reference to the unmodified version published as a book chapter is the following: Parker, D. S. (1998). Introduction. In The Idea of the Middle Class: White-Collar Workers and Peruvian Society, 1900-1950 (pp. 1-21). University Park: Penn State University Press. We would like to thank both the author and Penn State University Press for the permission to reproduce that chapter in this monographic issue of Papeles del CEIC.

2 Much of the criticism of Gareth Stedman Jones’ (1983) work on the language of Chartism revolves around this point.

3 Values since the 1998 edition of the book have been updated to reflect inflation to date.

4 For an introduction to the obrero-empleado distinction, see: Davis, 1972; and Angell, 1972.

5 See preferably chapters 3, 6 and 7 of that book.