Class and Class Conflict

We have the framework (4.1.1) and its machinery (4.1.2). Now the thing that moves inside it.

The Communist Manifesto opens, after the famous line about a spectre haunting Europe, with a claim so large it is easy to skim past: "The history of all hitherto existing society is the history of class struggles."

This lesson asks what a class actually is in Marx's sense — which is not what most people mean by the word — and what that sentence is really claiming.

Two households in the same English county town, 1860.

At number 14 lives Mr Harrow, chief clerk to a shipping agent. He earns £180 a year, which is comfortable. He wears a good coat, keeps a servant, sends his son to a grammar school, and would be extremely offended to be described as working class. He reads the newspaper, votes, and is on speaking terms with the vicar.

At number 52 lives Mrs Kelly, who takes in laundry. She earns perhaps £25 a year, in coins, when the work comes. Her hands are permanently damaged. Her children have worked since they were nine.

By income these two are in different worlds, and both of them know it. Mr Harrow earns seven times what Mrs Kelly does and would not dream of mixing with her.

Now a third household. On the hill lives Mr Ashby, who owns the mill. He also earns money — a great deal of it — but here is the difference Marx wants you to see:

If Mr Ashby stays in bed for a year, money still arrives. The mill runs; the looms turn; the goods sell. His income comes from something he owns.

If Mr Harrow stays in bed for a year, he is destitute by spring. His £180 comes from something he does, and he does it for someone else, on that person's premises, with that person's equipment, and only for as long as that person requires it.

If Mrs Kelly stays in bed for a week, her children do not eat.

By income: three tiers. By the question where does your living come from — from property, or from selling your capacity to work — Mr Harrow and Mrs Kelly are on the same side of a line, and Mr Ashby is on the other.

That line is what Marx means by class.

Why insist on a definition that so obviously cuts against how people experience their own lives? Mr Harrow would reject it flatly, and he is not stupid or deluded about his circumstances.

Marx's answer has two parts.

First, the definition is doing analytic work, not describing feelings. It picks out a structural relationship that has consequences whether or not anyone notices it: what happens to you in a downturn, whether your income survives your illness, what you can pass to your children, what you must do on Monday morning.

Second — and this is the part people miss — Marx knew perfectly well that Mr Harrow does not feel any solidarity with Mrs Kelly. He built a whole concept for the gap between structural position and lived identity: class-in-itself versus class-for-itself (4.1.8). He was not claiming people automatically know their class. He was claiming there is something there to know.

The test of the definition is not whether it matches self-description. It is whether it predicts anything. Hold that question open through the lesson.

The idea, in plain words

In capitalism, Marx's two principal classes:

The bourgeoisie — owners of the means of production, who buy labour power and live from the surplus it generates.

The proletariat — those who own no means of production and must sell their labour power to live.

Note the definition of the proletariat carefully: it is not "poor people," and it is not "manual workers." It is people who have no way of living except by selling their capacity to work. A software engineer on a large salary is, on this definition, a proletarian. This sounds absurd until you ask what happens to that engineer when the company restructures, and notice that the answer is categorically different from what happens to the person who owns the company.

This is the single most common misunderstanding, so let us kill it properly.

Marx's class is not a ladder. Sociology has two quite different traditions here. Gradational schemes rank people along a continuous scale — income deciles, socio-economic status, "upper / middle / lower." Relational schemes define classes by their relationships to each other. Marx is emphatically relational.

The difference is not academic. On a gradational view, classes exist independently — you could remove the top decile and the rest would be unchanged. On a relational view, the classes constitute each other: there is no capitalist without a worker to employ, and no proletarian without an owner to sell to. The relation is prior to the categories.

Which is why "class conflict" is built into the definition rather than added to it. The relation contains an opposition of interest: wages are a cost to one party and a livelihood to the other, and the same pound cannot be both.

Marx's other classes

The two-class model is Marx's analytical skeleton, not his description of any actual society. His historical writing is populated.

The petty bourgeoisie — small proprietors who own means of production but work them themselves: shopkeepers, artisans, small farmers, the self-employed. Marx expected them to be crushed between the two main classes and largely proletarianised. This has partly happened and partly not; the self-employed sector in most rich countries is stubbornly persistent, and gig platforms have created a large category of workers who are formally petty bourgeois — owning their vehicle, bearing their own costs, legally independent — while being economically dependent in the manner of employees. Marx would have found this arrangement extremely interesting.

The landed aristocracy — owners of land drawing rent, a residual class from the previous mode of production, allied and in conflict with industrial capital by turns. Much of nineteenth-century British politics, including the Corn Law fight, is legible as a quarrel between landed and industrial property.

The lumpenproletariat — Marx's contemptuous term for those outside regular production: vagrants, criminals, the chronically destitute. He regarded them as politically unreliable and available for hire by reaction. This is the least defensible part of his class analysis — moralistic, empirically thin, and the ancestor of a great deal of unpleasant talk about an "underclass." Modern sociology of poverty (8.6) has demolished the notion of a distinct population permanently outside the labour force; movement in and out of insecure work is the norm.

The peasantry — a class by structural position, but one Marx thought incapable of acting as a class. His image in The Eighteenth Brumaire is famous: French smallholding peasants are like potatoes in a sack, "formed by the simple addition of homologous magnitudes" — similar to one another but not connected to one another, and therefore "incapable of enforcing their class interest in their own name." Twentieth-century history in Russia, China, Vietnam and Mexico was rather rough on this judgement.

What class struggle actually claims

Back to the Manifesto's opening. Engels added a footnote in 1888 restricting it to "all written history," in deference to the anthropology of the time — an early sign that the claim was felt to need qualification.

Read carefully, the claim has three parts, of quite different strength.

One: every society since the emergence of a surplus has been divided into classes with opposed interests over the disposal of that surplus. This is strong and largely defensible. Freeman and slave, patrician and plebeian, lord and serf, guild-master and journeyman — Marx's own list.

Two: the conflict between these classes is the principal motor of large-scale historical change. This is contestable, and contested. Marx's own historical writing gives real causal weight to states, wars, religion and contingency. Weber's counter-case is that status groups, parties, bureaucracies and religious ideas are independent forces (4.3.6, 4.3.7). The honest position is that class conflict is one major motor, demonstrably so, and that claiming it is the motor requires ignoring or reducing a great deal.

Three: the conflict is structural, not personal. This one is essential and frequently missed. Marx is not saying capitalists are bad people. He says so explicitly in the preface to Capital: he deals with individuals "only in so far as they are the personifications of economic categories, embodiments of particular class-relations and class-interests," and "my standpoint... can less than any other make the individual responsible for relations whose creature he socially remains."

A kind employer who pays above the market rate, provides good conditions and knows his workers' names is still occupying a position defined by the extraction of surplus — and is under competitive pressure from employers who do none of those things. The analysis is about positions and pressures, not about virtue. This is precisely what makes it sociology rather than moralising (1.1.1).

A firm makes components. Margins are thin because three competitors are bidding for the same contracts.

The owner would prefer to pay well. She likes her staff. But if her labour costs run 8% above her competitors', she loses the contract, and then everyone loses their job — so she holds wages down, and she is not being greedy; she is responding to a structural pressure she did not create and cannot exit.

The workers would prefer higher wages. They are not being greedy either; their rent has risen 12% and their wages have not.

Each side is behaving reasonably. Neither is villainous. And their interests are still directly opposed, because the same money cannot be both wage and margin.

This is what Marx means by class conflict as a structural feature. Remove every individual and replace them with better people, and the opposition remains, because it is a property of the relation and not of the persons occupying it.

Now notice what the conflict becomes in practice: not a battle but a permanent negotiation, conducted through wage bargaining, union recognition, employment law, minimum wage legislation, working-time rules and safety inspection. Most class conflict in developed capitalist societies is institutionalised and boring. Dahrendorf (5.2) made this the centre of his analysis: conflict does not vanish, it gets regulated.

Go deeper

Did Marx get the trajectory wrong?

He made several predictions about class structure, and it is worth grading them honestly rather than defending or dismissing wholesale.

Polarisation into two classes. Partly wrong. The twentieth century produced an enormous stratum of professionals, managers, technicians and public employees who fit neither pole comfortably. Erik Olin Wright built a career on this, developing the concept of contradictory class locations: managers, for instance, are employees who nonetheless exercise authority over other employees and act for capital. That is not a fudge; it is an attempt to keep a relational framework while acknowledging what actually exists.

Immiseration — the claim that workers' conditions would worsen absolutely. Wrong for the rich countries over the long run. Real wages in the industrialised world rose substantially from the late nineteenth century onwards. Whether they rose because of capitalism or because of organised labour, suffrage and state regulation wrung concessions from it is a live argument — but the absolute claim failed. A weaker version, relative immiseration (the workers' share of a growing product falling), has more support: the labour share of national income has declined in most rich countries since around 1980, and top income shares have risen towards early-twentieth-century levels.

Disappearance of the petty bourgeoisie. Partly wrong, as noted.

Proletarian revolution in the most advanced capitalist countries. Wrong. Revolutions claiming his name occurred in Russia and China — agrarian societies where his own theory said they should not. This is the largest single failure of prediction in the framework, and it deserves to be stated plainly.

Concentration of capital. Right, strikingly. Firms grew, merged and dominated exactly as he expected. The scale of contemporary corporate concentration would not have surprised him at all.

The globalisation of capitalist relations. Right, and early. The Manifesto's passage on the bourgeoisie chasing over the whole surface of the globe, nestling everywhere, settling everywhere, establishing connections everywhere — written in 1848 — is quoted approvingly by people who agree with nothing else he wrote.

What class does and does not predict now

The honest state of the evidence, which is neither "class is dead" nor "class explains everything."

Class position, however measured, remains among the strongest predictors of: life expectancy and healthy life expectancy; child mortality; educational attainment; exposure to occupational injury; job security in downturns; wealth (far more unequal than income); and — importantly — the wealth and class position of one's children.

Class has weakened as a predictor of: party vote in many countries, where education, age, region and cultural identity now do more work; self-identification, which in surveys tends to bunch enormously in the middle; and immediate consumption patterns, which are cheaper to imitate than they were.

And Marx's specific definition competes with others. Weber (4.3.6) splits the field into class (market situation), status (social honour) and party (organised power) — three dimensions rather than one, and better at explaining why Mr Harrow despises Mrs Kelly. Bourdieu (5.10) adds cultural and social capital, showing how taste and comportment transmit advantage. Erik Olin Wright attempted a synthesis, defining classes by ownership, authority and scarce skills together. Goldthorpe's schema, used across European official statistics, classifies by employment relations — the terms of the employment contract, not ownership.

Note that every one of these is arguing with Marx, and every one retains the core relational insight: class is about how you stand in relation to other people in the productive order, not about a number.

And the largest omission

Marx's class analysis has a systematic blind spot that his successors had to repair.

Gender. The wage labourer's ability to appear at work depends on someone having fed him, washed his clothes, raised his replacements and cared for him when sick. That labour was and largely remains unpaid, and it was and largely remains women's. Marx knew domestic labour existed and had almost nothing analytic to say about it. The domestic labour debate of the 1970s and the social reproduction theory that grew from it (8.5) are the repair.

Race. Slavery, indenture and colonial forced labour were not archaic survivals awaiting absorption into free wage labour; they were constitutive of the capitalism Marx analysed, and he substantially underweighted them. Marx did see part of this — his line that "labour in the white skin can never emancipate itself where it is branded in the black skin" is a genuine recognition — and he supported the Union in the American Civil War and the abolition of slavery unequivocally. But the systematic analysis came later: Du Bois (4.5), Oliver Cox, Cedric Robinson's racial capitalism, and the split labour market literature (8.4).

"Class means income bracket." It means relationship to the means of production. The two correlate; they are not the same, and confusing them makes the theory unfalsifiable and boring.

"Marx said capitalists are evil." He said they are personifications of economic categories, and denied that individuals are responsible for relations whose creatures they remain.

"Everyone is middle class now." Self-identification has bunched in the middle. Ownership of productive assets has not: wealth remains far more concentrated than income in every rich country measured.

"Class conflict means violence." Most of it is wage bargaining, employment law, union recognition disputes and strikes — institutionalised, procedural and dull. Dahrendorf's point.

"Marx invented class analysis." Adam Smith, Ferguson, Millar (3.1.6) and the French historians Thierry and Guizot all wrote about class before him. Marx said so in a letter of 1852, claiming as his own contribution only that classes are tied to particular historical phases of production.

"Class is obsolete." It has weakened as a predictor of voting and identity. It has not weakened as a predictor of how long you live, which is a fairly consequential outcome.

Why it matters

It gives inequality a mechanism instead of a description. Stratification research can tell you the distribution; class analysis proposes why it takes that shape and what reproduces it. You need not accept the answer to see the value of demanding one.

It explains why conflict persists through personnel changes. Structural opposition survives the replacement of individuals. This is the same lesson as 1.1.1, applied to the economy: stop looking at character, look at position.

It underlies the largest arguments in the discipline. Weber's three dimensions, Bourdieu's capitals, Wright's contradictory locations, Goldthorpe's employment relations, intersectionality's insistence on class with race and gender (8.7) — all of it is a conversation begun here.

And it names the mechanism the rest of Topic 4.1 will specify. We have said the owner takes a surplus. We have not yet said how, given that the worker is paid and free and the transaction looks fair. That is 4.1.5, and getting there requires first understanding what the arrangement does to the person inside it — which is 4.1.4.

For Marx, a class is defined by its common relationship to the means of production — do you own productive property, or must you sell your labour power to those who do? Not income, not status, not lifestyle.

Bourgeoisie: owners of the means of production. Proletariat: those who must sell labour power to live. Plus the petty bourgeoisie, the landed aristocracy, the peasantry ("potatoes in a sack") and the lumpenproletariat — the last being the weakest and least defensible category.

Class in Marx is relational, not gradational: the classes constitute each other, and the opposition of interest is built into the relation rather than added to it.

"The history of all hitherto existing society is the history of class struggles" unpacks into three claims of unequal strength: that surplus-producing societies have opposed classes (strong); that class conflict is the principal motor of change (contested); and that the conflict is structural rather than personal (essential, and constantly misread).

Scorecard: concentration of capital and globalisation — right. Polarisation into two classes and the disappearance of the petty bourgeoisie — partly wrong. Absolute immiseration in rich countries — wrong; relative decline in labour share — supported. Revolution in the most advanced economies — wrong.

Class still strongly predicts life expectancy, health, education, job security and wealth; it predicts voting and self-identification far less well than it did. And it required major repair on gender (social reproduction) and race (racial capitalism), which Marx underweighted.

Class (Marxian) — a group sharing a common relationship to the means of production.

Means of production — what is needed to produce: land, plant, machinery, raw materials, and now patents, platforms and data. Distinct from means of consumption.

Bourgeoisie — owners of the means of production who buy labour power.

Proletariat — those without means of production who must sell labour power to live.

Petty bourgeoisie — small proprietors who own and work their own means of production.

Lumpenproletariat — Marx's dismissive category for those outside regular production; largely discredited.

Relational vs gradational class — defined by relationships to others versus position on a continuous scale.

Contradictory class locations — Erik Olin Wright's term for positions with features of more than one class, such as managers.

Immiseration thesis — the claim that workers' conditions worsen; absolute version disconfirmed in rich countries, relative version better supported.

Racial capitalism — Cedric Robinson's term for capitalism's constitutive dependence on racial differentiation.

One — run the Ashby test. For three people you know in different jobs, ask: if they stopped working for a year, where would money come from? The answers sort people by relation to the means of production far more sharply than salary does.

Two — locate yourself. Do you own means of production? Do you sell labour power? Do you exercise authority over others who sell theirs? Do you hold scarce credentials? Wright's three dimensions will probably put you somewhere uncomfortable and interesting.

Three — find the institutionalised conflict. Look for a current wage negotiation, strike or employment-law dispute in the news. Identify each side's interest without moralising about either. Then note the procedure through which it is being conducted — that procedure is class conflict, domesticated.

Four — test a prediction. Find published figures for life expectancy by occupational class, or by neighbourhood deprivation, in your country. The gaps are usually larger than people expect. Ask what mechanism could produce them.

Five — the gig question. Take a platform worker who owns their vehicle, bears their own costs and is legally self-employed. Are they petty bourgeois or proletarian? Argue both. The difficulty is the point, and courts in several countries have found it difficult too.