Framing

I said that the first wave of interest in class after the Second World War appeared in the years around 1970, and it was associated with what was called Neo-Marxism (see Manza and McCarthy Reference Manza and McCarthy2011; an earlier assessment of which can be found in Bottomore Reference Bottomore, Bottomore and Nisbet1979, pp. 135-143). For many of those involved, this interest included a rejection of older, classic Marxism, which they saw as rigid, deterministic and economistic (but see Roseberry 1988). That older Marxism helps to define one pole of the dimension of class analysis that is of concern in this section. That is because it is concerned with processes and entities associated with whole social systems and even groups of social systems. In this sense, that older Marxist approach is structural, for it addresses the properties and operations of political-economic structures. However, this structuralism was not limited to that older Marxism, nor did it disappear with the emergence of Neo-Marxism.

For one thing, the other main theory of class that existed at the time was also concerned with structure. That was the theory associated with the work of Max Weber, laid out most concisely in the fragment “Class, Status, Party” (Weber Reference Weber, Gerth and Mills1946). The best known of Weber’s works at the time was his Protestant Ethic (Weber Reference Weber1958 [1904-1905]), which describes the Weltanschauung of ascetic Protestants and points to the historical contingency of the emergence of Western rational capitalism. This justified the common view that Weber is not as structural or deterministic as classic Marxism. However, the book also points to what looks very much like the inevitability of the Iron Cage and to the unintended and constraining consequences of the initial appearance of rational capitalism among ascetic Protestants. Moreover, in “Class, Status, Party” Weber presents classes in terms of a fairly abstract and general understanding of the structure of the market and of people’s location within it. That market may reflect people’s resources and activities, but it has an existence that people confront as ordered and external to their individual will.

Moreover, structuralism persisted in other important strands of Neo-Marxist analysis. This is most obvious in the work of the French structural Marxists, especially Louis Althusser (e.g. Althusser and Balibar Reference Althusser and Balibar1970), who was probably better known to sociologists than were many others of the school. For anthropologists, those people included a number of scholars who considered African societies, especially as they existed before and during early European incursion (overviews are contained in Robotham Reference Robotham and Carrier2012; Şaul Reference Şaul and Carrier2012). Some of these scholars were concerned with class per se, but even those who were not used Marxist models to describe systems in which different sets of people were identifiable as classes, for they stood in different relations to the means of production in their societies.

One of the more influential of these scholars was Claude Meillassoux, whose Maidens, Meal and Money (1981) illustrates this structuralism in an interesting way. The first part of that book is more conventionally anthropological, approaching the political economy of sub-Saharan societies in terms of production relations and class as they were manifest in kinship structures and practices. The second part is less conventional. It is concerned with those societies once there was substantial European capitalist enterprise in the region, especially enterprises that relied on migrant labour, particularly mining. Meillassoux argued that the relationship between rural areas and those enterprises, embodied in the migrant workers, strengthened the capitalist sphere through a hidden subsidy. If capitalist firms collectively are to survive in fully capitalised and monetised economies, they must pay enough to cover the cost of rearing the next generation of workers, and they do so through wages and taxes. However, Meillassoux says that in the African regions that concern him they need not do so, for the bulk of that rearing occurred within rural areas with their own mode of production. In making this argument about the articulation of modes of production, Meillassoux was identifying classes in terms larger than the single system that was the common concern of sociologists writing about class.

Meillassoux was not alone in this, and there is a body of work that was not, strictly speaking, Marxist, but that used the same structural approach. For instance, the analysis of the structural properties of systems spanning countries and regions of the world is a defining feature of work on underdevelopment (e.g. Baran Reference Baran1957; Frank Reference Frank1969) and, following Wallerstein (Reference Wallerstein1974), on the world-system (aspects are considered in Friedman Reference Friedman2000). Writers in these schools did not often address class directly, though some did. For instance, Sidney Mintz (Reference Mintz1985) linked class relations in sugar-growing areas of the Caribbean with the position of the expanding proletariat in England in the nineteenth century. However, even if the production at issue could range from the extraction of raw materials in one continent to their conversion into finished products and ultimate consumption in another, all of these people’s work revolved around the ways that different sets of people were related to the means of production, and hence to each other. That is the classic Marxist approach to class.

That structural approach persisted as well in work that is more purely sociological. For instance, some sociologists addressed the ways that the class structure was reproduced and legitimated over the course of time by the education system in France (e.g. Bourdieu and Passeron Reference Bourdieu1977) and the United States (Bowles and Gintis Reference Bowles and Gintis1976; and, in a different way, Milner Reference Milner1972). As well, some who wrote on consumption considered the ways that the structure of society is related to the structure of objects people consume (e.g. Baudrillard Reference Baudrillard1981) and how people’s consumption preferences reflect and reproduce their class location (esp. Bourdieu Reference Bourdieu1984). And, of course, this general approach continued in work on the occupational structure, whether treated in terms of social stratification (e.g. Goldthorpe and Jackson Reference Goldthorpe and Jackson2007) or class (e.g. Steinmetz and Wright Reference Steinmetz and Wright1989; Le Roux et al. Reference Le Roux, Rouanet, Savage and Warde2008).

While this structural orientation persisted well beyond the arrival of Neo-Marxism, there were Neo-Marxists who challenged it and who inspired a growing body of work that marks the other pole of the approach to class that concerns me in this section. Whereas structuralists are concerned with large-scale, systemic class processes and relationships, those at the other pole are concerned with class as it appears in fairly small-scale interactions and processes. Many factors encouraged this change in the framing of class, but one of the most important was E.P. Thompson’s The Making of the English Working Class (1968). Thompson explicitly rejected structuralist approaches, instead being concerned with the historical process of class formation.

Thompson’s topic was a systemic change in the political economy of England, but his concern with class formation was echoed in work that, while it may ultimately have been concerned with systemic change, investigated it in smaller, more local grounds. This shift is apparent, for instance, in work on a topic that became popular in the 1980s, the emergence of the middle class (which needs to be distinguished from the petit bourgeoisie; e.g. Bechhofer and Elliott Reference Bechhofer and Elliott1981). That emergence is a systemic event and so, on its face, could be approached in terms of large-scale structures and processes. This is the sort of approach that, for instance, Don Kalb (Reference Kalb1997) uses in his treatment of the changing nature of the working class in a part of the Netherlands, which he relates to the changing position of, and global pressures on, the dominant firm in that area. Unlike Kalb, with his concern for larger economic structures and processes, those describing the emergence of the middle class often attended to the pattern of life in fairly circumscribed localities, such as changes in employment patterns, church attendance and the like in a particular city or neighbourhood (Archer and Blau Reference Archer and Blau1993).

There are smaller and more circumscribed sites than a neighbourhood in Birmingham or a suburb of Cleveland, and works that attend to them mark the other end of the dimension that concerns me here. One such site is the factory shop-floor, which attracted significant sociological attention. The person who is perhaps best known in this regard is Michael Burawoy, and the technique that he used in the research that led to Manufacturing Consent (1982), the participant observation that is the conventional method of anthropology, illustrates my point that the boundary between it and sociology is fuzzy. Burawoy may be the best known of those investigating the processes, understandings and negotiations through which class and class relations appear, are challenged and reproduced on the shop-floor. He is, however, hardly the only researcher to be concerned less with the properties of large-scale structures and the relations they subsume, and more with the small-scale processes in which class, and especially the working class, exists (e.g. Nichols and Beynon Reference Nichols and Beynon1977; Willis Reference Willis1977; Halle Reference Halle1984). In this work, those concerned with such processes reveal the particularities and contingencies that shape class processes, which are often are hidden from view in more macroscopic studies.

Perspective

The second of the dimensions I mentioned is the perspective that is used in research bearing on class, and I said that I am concerned particularly with whether that perspective is more objectivist or subjectivist. I take an objectivist perspective as one in which the researcher starts with a definition of the entity at issue, however tentative that definition may be, and investigates the world in terms of it. On the other hand, with a subjectivist perspective the researcher effectively allows those being studied to define the entity through their ordinary usage, and then investigates the world in terms of that definition.

As was the case with the dimension described above, framing, structural Marxists lie at one end of this dimension, the objectivist. Such scholars approached African societies, for instance, with a concept of class in mind, and considered the extent to which the systems and relationships in those societies did or did not fit that conception. What the Africans they described actually thought about, say, their system of kinship or of production was a secondary consideration, perhaps something to be taken into account or explained in class terms, but not constitutive of class in their society. Weber’s approach to class, at least in the fragment “Class, Status, Party”, is similar. That is, he defined class in terms of the resources that people can deploy in the market, their ability to get what they want in their transactions. Their market power may be affected by how they and others think about those transactions, but market power, like class, is Weber’s concept, not theirs.

An objectivist approach has long been important in one of the strands of work on class that I mentioned at the outset, that concerned with social stratification. Such work commonly is concerned with objective measures of status, such as income or education, and their distribution in society (e.g., famously, Blau and Duncan Reference Blau and Duncan1967). The objectivist approach is found as well in work on very large-scale systems and their changes, such as that concerned with underdevelopment and the world-system, mentioned above. A clear example of an objectivist approach among Marxists is George Steinmetz and Erik Olin Wright’s (1989) investigation of the changing preponderance of the petty bourgeoisie in the class structure of the United States. In this, they define “petty bourgeoisie” as “self-employed”, use national statistics to identify how the number of self-employed in the US had changed over time and suggest reasons for those changes. The conceptual apparatus used to define class, then, arose from these two researchers, not from the people that they studied.

This objectivist approach is to be expected in the sort of survey and statistical work that I have just mentioned, but it occurs in more qualitative work as well (an early example is Sennett and Cobb Reference Sennett and Cobb1966). So, for instance, in his study of parts of the working class in Sheffield, Massimiliano Mollona (Reference Mollona2009) attends to how people understand themselves, their work and the world around them. However, Mollona relates these to their position in the structure of the social relations of production, not to their subjective sense of their social location. Loïc Wacquant’s (Reference Wacquant, Crompton, Devine, Savage and Scott2000) analysis of changing class polarisation in Western countries is similar. His concern is the different ways that a particular sort of urban lower class is expanding in those countries, the factors that have led to that expansion and the government policies that shape it. Orientations and attitudes are important in Wacquant’s argument, but only in the ways that they shape and reflect government policies and people’s political-economic situations. In other words, they help elucidate the processes and structures that Wacquant studies and describes, as they do for Mallona, but they do not define the object of scholarly attention.

An objectivist approach is also found in work on a topic that is frankly concerned with people’s perceptions of their worlds, consumption. One instance of this is Josiah Heyman’s consideration (Carrier and Heyman Reference Carrier and Heyman1997, pp. 364-367) of people’s consumption of housing and education as they seek to secure the reproduction and improvement of their households. In this, Heyman focuses on how people perceive aspects of their worlds, but does so in terms of categories and processes that those perceptions reflect and illustrate, but do not define. The best known objectivist approach to class and consumption is probably Pierre Bourdieu’s Distinction (1984), based on how people perceive aspects of their world, which is to say their tastes. As was the case with Heyman’s analysis, however, those perceptions do not define the dimensions of class that Bourdieu lays out. Rather, the patterns and regularities of those perceptions are the vehicle he uses to identify the structure and nature of class in French society. As one might expect, this objectivism is found among those who have been influenced by Bourdieu (see Sallaz and Zavisca Reference Sallaz and Zavisca2007). For instance, Brigitte Le Roux et al. (Reference Le Roux, Rouanet, Savage and Warde2008) effectively replicate aspects of Distinction using British data, and discuss the statistical technique Bourdieu used in that work.

Bourdieu’s objectivist approach to what people think in Distinction echoes his earlier argument about what he called “misrecognition”, a concept that itself echoes the older Marxist concept of false consciousness. For Bourdieu, misrecognition occurs when people see their social arrangements and the consequences of those arrangements as reflecting the natural order of things, rather than as being arbitrary social conventions. And for him, misrecognition is common: “Every established order tends to produce [...] the naturalization of its own arbitrariness” (1977, p. 164), a naturalisation that the analyst needs to explain in terms of those social conventions. Thus, Bourdieu and those like him use people’s perceptions in work on class, but approach those perceptions in terms of the analyst’s model of class.

I said that the other end of the dimension that concerns me is subjectivist, in which researchers use the perspectives of the people they study to identify and even define the conceptual entities at issue. The difference between these two perspectives can be cast in terms of the classic anthropological concern with The Native’s Point of View. Both objectivists and subjectivists can be concerned with that, but approach it in different ways and for different purposes. For those using an objectivist approach, like Bourdieu, that Point of View is a tool for understanding processes and regularities amongst the pertinent Natives, including those related to class. For those using a subjectivist approach, Clifford Geertz (Reference Geertz1973) being the defining instance in anthropology, that Point of View is the goal of research, and it exhausts the realm of investigation.

To a degree, the subjectivist approach to class emerged as a way of dealing with two problems that were difficult for Marxists to solve, though they presented no particular difficulty for Weberians. One problem was fairly specific: the nature of the middle class, which has long posed a problem not only for Marxists, but also more generally (that problem is raised, albeit indirectly, in Berle and Means Reference Berle and Means1932). One approach to this problem ended up echoing, perhaps unintentionally, the older observation that those who moved out of the lower classes changed their residence, consumption and religion (see Goldthorpe Reference Goldthorpe1987). It did so when it defined the middle class largely in terms of attitudes, orientations and styles of life, what Melanie Archer and Judith Blau (Reference Archer and Blau1993, p. 30) point to as a concern with the “‘structure of feeling’ that was both the expression and the legitimization of middle-class behavior and ideals”. The result, as I indicate below, was work that increasingly defined the middle class in terms of the subjective perceptions of the people being studied, so that “middle class” came increasingly to look like an identity, perhaps echoing the old notion of “respectable”, rather than a position in a social structure. The consequence is very close to what R.T. Smith (Reference Smith1984: 469) observed about a body of anthropological work on class: “Class [...] becomes just another word for culture-bearing group”.

The other problem was more general: people in different classes increasingly have seen themselves as middle class (e.g. Kingston Reference Kingston1994). Those interested in class could account for this by considering the effects of conventional social processes on people’s self-perceptions (e.g. Kelley and Evans Reference Kelley and Evans1995); equally, they could see it as indicating the need to reconsider the ways that researchers understand class consciousness (e.g. Fantasia Reference Fantasia1995). However, commonly the problem was resolved by abandoning the objectivist view of class and instead defining class in terms of identity (aspects of this are described in Fraser Reference Fraser1999). As Wendy Bottero (Reference Bottero2004, p. 988) observed, such a resolution puts “issues of cultural identity at the heart of class theory, but recognize[s] that such issues cannot be theorized from within traditional class analysis”. Identity is an amorphous concept at best, and to say that the issues that it raises “cannot be theorized from within traditional class analysis” understates, if anything, the profound differences that can separate subjectivist and objectivist approaches. Nancy Fraser’s (Reference Fraser2001) presentation of a theory of justice indicates the nature and breadth of those differences in the context of her efforts to deal with a topic of importance to many of those concerned with class.

The effect of the more subjectivist approach on the concept of class can be illustrated by two interesting analyses of the ways that certain sets of people in Britain appear to see other sets. These are by Stephanie Lawler (Reference Lawler2005a) and Bev Skeggs (Reference Skeggs2005), and both represent the interest in culture that has become more important in sociology and anthropology in the past few decades. The elements of culture that concern them are some of the ways that sections of what Lawler and Skeggs take to be the British middle class understand a different set of people, sections of what Lawler and Skeggs take to be the working class, as well as some of the ways that these sections of the working class are presented in the media. As one might anticipate, these sections of the working class are taken to be undesirable.

The substance of their analyses is an interesting, if somewhat depressing, presentation of how some people in Britain see aspects of the world. However, the identification of the undesirables in these two articles as lower class illustrates the subjectivist approach. That is because neither Lawler nor Skeggs starts out with an analytical definition of “lower class”; neither investigates how people who fit that definition are represented or comprehended. Rather, both adopt the categories and conceptions of middle-class people and the media reports that they analyse. The result is not a description of how one set of people see and think about an objective set of people, the lower class. Rather, it is a description of how those people think about a cultural category that they themselves in part define. That thinking amounts to their imagining of themselves and their differences from imagined others in a process of dialectical definition. Consequently, the arguments in these two articles resonate more with Said’s Orientalism (1978; see Carrier Reference Carrier1992) than they do with Marx or Weber.

These imaginings and their associated identities may be construed by the researcher as “class subjectivities” (Fraser Reference Fraser1999, p. 120), a concept that has a significant analytical background. However, as the subjectivist approach became more established and even taken for granted, the attention to the modifier, “class”, and to what it implies about the social and political-economic processes that shape people’s self-perceptions, has declined (they are considered in Friedman Reference Friedman2004). The result is that, in some cases, all that is left for the researcher to study is those subjectivities (e.g. Reay Reference Reay2008). As Lawler (Reference Lawler2005b, p. 799) put it, “the language of psychology has come to replace a grammar of exploitation”.

Approaches and what is approached

I have described two dimensions that can help us to make sense of the work on class in sociology and anthropology over the past few decades. One is the frame used in scholarly work, especially whether it is concerned more with macroscopic or with microscopic relations and processes. The other is the scholar’s perspective, especially whether it is more objectivist or more subjectivist. I said that these two dimensions are conceptually distinct, but appear to be related in practice. That is, since the boom in scholarly interest in class around 1970, work on the topic has tended to shift from the macroscopic and the objectivist, illustrated by the French structural Marxists and by classic Marxist work before them, to the microscopic and subjectivist, illustrated by work on class as identity.

This association of the microscopic and the subjectivist is not mysterious. Those who study smaller-scale events and processes, common fare for fieldworking anthropologists, are more likely to confront particularities and contingencies that are less visible to those using a more macroscopic approach, and hence are less likely to be able to account for what they see in terms of the general analytical models of class. One common response is to account for those events and processes in terms of the perspectives of those being studied, a tendency made stronger by the increasing interest in culture in both disciplines since the 1970s. It should not be surprising that this explanatory tactic can end up elevating the perspective of those being studied relative to that of those doing the studying. The consequence turned out to be the displacement of the earlier, more macroscopic and structural orientation, which saw people’s class as their positions in a system that was society-wide, springing ultimately from the organisation and operation of the economy (for Marx and Weber) or of the institutions through which people secured and improved their socioeconomic position (for stratification theorists). Increasingly this gave way to work that approached class as an aspect of people’s identity and perception of the world, which arise from their personal experience and individual psychology; in extreme form class became no more than that.

This change was not happenstance, nor does it seem to have arisen purely because of the intellectual inadequacies of the older approaches. Rather, it appears to have been motivated by at least three factors. One of these was internal to scholarly work on class, one was internal to the social sciences and humanities more generally, and one was pervasive in the Western societies that have been home to most of the people who work on class. I shall describe each of them in turn.

I said that my first factor is internal to scholarship on class. In fact it primarily affected Marxist scholars, and particularly those whom Raymond Firth (Reference Firth1972) called “gut” Marxists, concerned with the progressive potential of Marxism, people whom he distinguished from the more purely intellectual “cerebral” Marxists. That factor springs from the ways that important parts of the world have failed to behave as the classic Marxist model, at least in its more polemical form, predicts, so that the revolution is no nearer. The proletariat have not developed unified class consciousness; the class structure of capitalist societies has not become increasingly bifurcated into rich capital and starving labour. While important elements of classic Marxist thought say that these things should happen, and present reasons why they should, that thought has turned out to be of little help for those who want to understand these failures.

The result has been something like a division among Marxist scholars, which resembles the distinction I have made between more macroscopic and more microscopic frames for research. That is the division between what can be called political economists and those concerned with lived class. This division is by no means absolute, and interesting work has been undertaken that seeks to link the two sides of it (among others, Smith Reference Smith1999; Durrenberger and Erem Reference Durrenberger and Erem2005; Narotzky and Smith Reference Narotzky and Smith2006; Kalb and Halmai Reference Kalb and Halmai2011; Heyman Reference Heyman and Durrenberger2012), but the division is real even so. The political economists are concerned with the ways that Marxist models can help illuminate the nature and operation of important aspects of the world in the same way that, for instance, Aage Sørenson (Reference Sørenson2000) used models from neoclassical economics, and especially the concept of rent, to produce a different theory of class that is intended to help illuminate other important aspects of the world. So, for instance, I have mentioned scholars of underdevelopment and of world-system theory, whose work is built on, among other things, the existence of sets of people who stand in different relations to the means of production considered on a global scale. In being built on those sets, this work necessarily entails a concept of class, but political economists tend to be concerned with the nature of the larger political-economic system rather than with class. On the other side of this division are those who are concerned with the nature and significance of class in people’s lives: people’s experience of it, awareness of it, the ways that they seek to mobilise it and so forth. These scholars are, however, generally less interested in the political-economic system that generates and sustains those classes.

The next of my three factors is one that, I said, affected academics generally, and with it those who worked on class. That is the intellectual current called “postmodernism”, which I take to include poststructuralism. While postmodernism has been especially noticeable in anthropology (see Sahlins Reference Sahlins1999), it has affected sociology as well, perhaps most notably through the work of Michel Foucault. Although postmodernism flourished in the 1980s, precursors to it appeared in Neo-Marxist writing in the period around 1970. That was the era of structural Marxism, but also of work that criticised it, facilitated by the publication in English of Marx’s early writings (Bottomore Reference Bottomore1963). The most obvious critics were E.P. Thompson (Reference Thompson1968) and, somewhat later, Pierre Bourdieu (Reference Bourdieu1977). This criticism posited that it is necessary to relate the macroscopic structures of classic Marxist analysis, and by implication also of important aspects of Weber’s work, to people’s ordinary lives. This shift in orientation was widespread in sociology (e.g. Giddens Reference Giddens1984) and anthropology (e.g. Sahlins Reference Sahlins1985), and was intended to relate social and cultural structures, and the stability that they were taken to imply, to social process, and the autonomy and change that they were taken to allow.

However, and regardless of what those early critics of structural Marxism and of structuralism more generally may have intended, researchers increasingly appear to have abandoned interest in structure altogether, and instead attended to everyday lives and people’s perceptions of them. This took distinctive form in the postmodernist rejection of the idea of social or cultural systems and of the theories and models that sought to describe and account for them. These were the sorts of theories and models associated with classic Marxist analysis, with structural-functionalism in both disciplines, with older American anthropological concern with cultural structure and with Lévi-Straussian structuralism. In its more thoroughgoing forms, this rejection of structure led to the rejection of the very idea that there are things like social statuses or conditions, things like standing in a particular relationship to the means of production or having a particular amount of market power. Instead, and reflecting as well the cultural turn, increasing attention was paid to people’s views of themselves and their worlds. In a thoughtful introduction to an issue of Sociology devoted to class, David Byrne (Reference Byrne2005, pp. 807-808) pointed to this conjunction of the microscopic and the subjectivist: “The cultural turn in class [...] is often also a turn toward individual experience and personal response to that experience”. With this shift, society and culture came at times to be seen as little more than the processes and relationships in which people exist and that shape their views of themselves and their worlds; or, as Skeggs (Reference Skeggs2005, p. 976) puts it, “the process of evaluation, moral attribution and authorization in the production of subjectivities”.

In this rejection of the idea of social or cultural structure, and the concomitant shift of interest to individuals, their mentalities and perceptions, this growing body of postmodernist thought came to echo the third of the factors that, I said, motivated changes in the ways that sociologists and anthropologists approached class. That third factor is something that I have mentioned already: the growing general influence of neoliberalism in the closing decades of the twentieth century. The aspect of neoliberalism that concerns me here is the neoclassical economics with which it was associated and which became especially important in the 1970s.

Before that decade, the predominant economic orientation in Western countries was broadly Keynesian, which was reflected in government economic policies and public debate. That orientation was macroeconomic, and it sought to make sense of the nature and operation of the economic institutions and systems of countries. In 1974, however, a number of Western countries experienced “stagflation”, a combination of economic stagnation and inflation. According to neoclassical critics and their public supporters, that stagflation showed the failure of existing Keynesian approaches. Those critics could have used that failure to urge an improvement in Keynesian thought or the development of new macroeconomic orientations. Instead, however, they argued that Keynesian orientations were fatally flawed simply because they were macroeconomic, seeing economies as systems made up of institutions and relationships among them. Echoing the words of Friedrich Hayek (Reference Hayek1944, p. 204), they held that we must ignore “the craving for intelligibility” about how the economy works and why. Wielding Occam’s Razor with fatal effect, they said that the economy was only the arithmetic sum of the actions of individuals in their transactions in free markets, which were not only rational (Fox Reference Fox2009) but also were more efficient than governments at allocating resources.

The rise of neoliberalism was not, of course, due only to the arguments of neoclassical economists. The neoliberal rejection of systems and structures, echoed in postmodernism, was widespread in many Western countries, especially in the United States, with its long history of individualistic opposition to constraint. In the first part of the 1970s that country was undergoing the Civil Rights and Women’s Liberation movements, both of which sought to assert the rights of individuals against institutions and systems that infringed on them. Combined with this was the coming of age of the demographic bulge that was called the Baby Boom, the extraordinary number of people born in the years immediately following the Second World War. These people, who were in their twenties by the middle of the 1970s, were unhappy with what many of them saw as the constraints imposed by the existing social and political order. Doubtless, many of these people, like many of those in the Civil Rights and Women’s Liberation movements, had visions of replacing the old order with a better one. However, and doubtless for many different reasons, they turned out to resemble the neoclassical critics of Keynesianism, for they all ended up rejecting systems and structures, instead elevating the individual and individual lives, thoughts and values.

In retrospect, then, it appears that the changes that I described in orientations to class in sociology and anthropology were shaped by a range of broader changes both in academic thought and in society. Like that work on class, those broader changes amounted to a rejection of structure and of the ideas and bodies of thought related to it, and a celebration of individuals, their lives, subjectivity and agency. All the academic rejections were associated with problems generated by the older orientations that were discarded. What many saw as the classic Marxist predictions about the shape of society were apparently wrong. The older structuralist and structural-functionalist models in the social sciences did a poor job of accounting for change and of relating structures to the processes of daily life. The older Keynesian political economy did not prevent the stagflation of 1974. I said that these problems led some to try to revise the old orientations, such as Bourdieu attempted in Outline of a Theory of Practice. However, the result turned out to be not revision, but rejection, and those who maintained more structuralist, systemic and objectivist orientations increasingly were, at least until quite recently, reduced to the status of outliers in their fields.

In the preceding paragraphs I have described trends and forces that have affected the way that scholars have approached class. The points that I have made suggest that changes in that approach since the rise of Neo-Marxism have been no simple intellectual development within sociology and anthropology. In addition to that, perhaps even instead of that, they are disciplinary manifestations of broader social and political-economic forces. Particularly for these two disciplines, the boundary between academic field and broader society has always been porous, so the influence of those broader forces should come as no surprise.

The effects of that influence are complicated and revealing. I will illustrate this with one aspect of that influence, which relates to the early critics of structural, macroscopic models of class. As I have described, those critics said that we need to relate the abstract universe of those models more clearly to the actions that people take and that researchers study. For anthropology, Sherry Ortner (Reference Ortner1984 p. 148) stated that the important question is “the relationship(s) that obtain between human action, on the one hand, and some global entity which we may call ‘the system’, on the other”. For sociology, Jeffery Alexander (Reference Alexander1982, p. 98) put the same concern somewhat differently, asking “How [can] sociological theories which do accept the sui generis collective character of social arrangements [...] retain a conception of individual freedom and voluntarism?” Put differently, they said that we need to relate the structures contained in those models to the processes of social life. This injunction was reasonable on its face, but turned out to be difficult in practice.

On reflection, we should have foreseen that difficulty. Those more structural and macroscopic models of class are populated by abstract forces and pressures. There was no reason to think that relating them in a clear and straightforward way to social life in this neighbourhood or that shop-floor would be any easier than relating the abstract force of gravity in a clear and straightforward way to the fall of a leaf on a breezy day, or even on a still one (on the difficulties of this relating in anthropology, see Smith Reference Smith1984). Having such an expectation requires conflating two different sorts of answer to the question “Why?” One sort of answer is concerned with the visible chain of events that precedes, and can be taken to determine, the end state that concerns the researcher, such as an observed level of class consciousness in a given group of people or the final resting place of a leaf. The other sort of answer is that suggested by those general models, which are probabilistic rather than deterministic. They point to a tendency to certain sorts of states or outcomes, and they predict those states with no more certainty than we, with our knowledge of gravity, can predict the resting place of the leaf that is separated from its branch.

The result of this conflation is confusion, and perhaps the death of class in an objectivist sense. The critics who urged more attention to the relationship between structure and process were, as it turned out, taken to be calling for increasing, and increasingly microscopic, attention to the features of daily life. With the growing influence of postmodernism and the cultural turn, the generation following those critics had little taste for the comparison and generalisation that would allow them to aggregate the findings of these studies of daily life, interpret the results in terms of the general models that were the concern of those critics, and interpret these models in terms of those results.

The risks and costs that this confusion poses are illustrated especially in the more subjectivist work that sees class primarily in terms of people’s own cultural categories and the identities associated with them, the “subjectivities” to which Skeggs refers. From such a perspective, it is reasonable to say that, nowadays, “Forms of identity such as gender, generation, religion, or ethnicity [...] become as, if not more, salient to class formation, than does work” (Roberts Reference Roberts1990, 373-374). Such a concern with subjectivities and identities directs our attention to aspects of ordinary life that are shaped by so many factors that it would require extraordinary effort to convert them into terms that would allow us to see how well they accord with general and probabilistic models of class.

In addition, that concern, with its embrace of The Native’s Point of View, runs the risk of foreclosing certain sorts of questions, those that reflect things not visible from that Point of View. I illustrate this with reference to research by Michal Buchowski (Reference Buchowski, Schröder and Vonderau2008) on the middle class in a city in Poland. He found substantial difference between the set of people who meet a reasonable analytical definition of “middle class” and the set of people who meet the local, popular understanding of “middle class”: the latter was a relatively small and distinct fraction of the former. This difference is provocative, but only if we, like Buchowski, take The Native’s Point of View as something to be investigated and accounted for, perhaps in terms of changing practices and beliefs in a recently-socialist country. The provocation, the questions that this finding helps us to raise, disappear if we take people’s perceptions as defining the world of class.