When I began writing about class in the mid-1970s, I viewed Marxist and positivist social science as foundationally distinct and incommensurable warring paradigms. I argued that Marxism had distinctive epistemological premises and methodological approaches which were fundamentally opposed to those of mainstream social science. In the intervening period I have rethought the underlying logic of my approach to class analysis a number of times.footnote1 While I continue to work within the Marxist tradition, I no longer conceive of Marxism as a comprehensive paradigm that is inherently incompatible with ‘bourgeois’ sociology.footnote2

Having previously argued for the general superiority of Marxist class analysis over its main sociological rivals—especially Weberian approaches and those adopted within mainstream stratification research—I now take the view that these different ways of analysing class can all potentially contribute to a fuller understanding by identifying different causal processes at work in shaping the micro- and macro- aspects of inequality in capitalist societies. The Marxist tradition is a valuable body of ideas because it successfully identifies real mechanisms that matter for a wide range of important problems, but this does not mean it has a monopoly on the capacity to identify such mechanisms. In practice, then, sociological research by Marxists should combine the distinctive Marxist-identified mechanisms with whatever other causal processes seem pertinent to the explanatory task at hand.footnote3 What might be called a ‘pragmatist realism’ has replaced the ‘grand battle of paradigms’.

For the sake of simplicity, in what follows I will focus on three clusters of causal processes relevant to class analysis, each associated with a different strand of sociological theory. The first identifies classes with the attributes and material life conditions of individuals. The second focuses on the ways in which social positions afford some people control over economic resources while excluding others—defining classes relative to processes of ‘opportunity hoarding’. The third approach conceives of classes as being structured by mechanisms of domination and exploitation, in which economic positions accord some people power over the lives and activities of others. The first is the approach taken in stratification research, the second is the Weberian perspective, and the third is associated with the Marxist tradition.

Both among sociologists and among the lay public, class is principally conceived in terms of individual attributes and life conditions. Attributes such as sex, age, race, religion, intelligence, education, geographical location, and so on, are held to be consequential for a number of things we might want to explain, from health to voting behaviour to childrearing practices. Some of these attributes are acquired at birth, others later in life; some are stable, others quite dependent upon a person’s specific social situation, and may accordingly change over time. In the stratification approach, people can also be categorized by the material conditions in which they live: squalid apartments, pleasant suburban houses or mansions in gated communities; dire poverty, adequate income or extravagant wealth, and so on. ‘Class’, then, identifies those economically important attributes that shape people’s opportunities and choices in a market economy, and thus their material conditions. Class should neither be identified simply with people’s individual attributes nor with their material conditions of life; rather, it is a way of talking about the interconnections between these two.

Within this approach, the key individual attribute in economically developed societies is education, but some sociologists also include more elusive attributes such as cultural resources, social connections and even individual motivations.footnote4 When these different attributes and life conditions broadly cluster together, then these clusters are called ‘classes’. The ‘middle class’ here denotes people who have enough education and money to participate fully in some vaguely defined ‘mainstream’ way of life (which might include particular consumption patterns, for example). The ‘upper class’ designates people whose wealth, high income and social connections enable them to live their lives apart from ‘ordinary’ people, while the ‘lower class’ refers to those who lack the necessary educational and cultural resources to live securely above the poverty line. Finally, the ‘underclass’ are those who live in extreme poverty, marginalized from the mainstream of society by a lack of basic education and skills needed for stable employment.

In the individual-attributes approach to class, the central concern of sociologists has been to understand how people acquire the characteristics that place them in one class or another. Given that for most people in the countries where sociologists live, economic status and rewards are mainly acquired through employment in paid jobs, the central focus of research in this tradition has been the process through which people obtain the cultural, motivational and educational resources that affect their occupations in the labour market. Because the conditions of life in childhood are clearly of considerable importance in these processes, this approach devotes a great deal of attention to what is sometimes called ‘class background’—the family settings in which these key attributes are acquired. In a stripped down form, the causal logic of these kinds of class processes is illustrated in Figure 1 (below).

Skills, education and motivations are, of course, very important determinants of an individual’s economic prospects. What is missing in this approach to class, however, is any serious consideration of the inequalities in the positions people occupy, or of the relational nature of those positions. Education shapes the kinds of jobs people get, but why are some jobs ‘better’ than others? Why do some jobs confer a great deal of power while others do not? And is there any relation between the power and wealth enjoyed by some and the lack of it experienced by others? Rather than focusing exclusively on the process through which individuals are sorted into positions, the other two approaches to class analysis begin by examining the nature of the positions themselves.

Article figure NLR60wrightpicture1