The Division of Labour and the Two Solidarities

Durkheim's doctoral thesis, The Division of Labour in Society (1893), is where he first works out the question that occupied him for the rest of his life. It is also the exact mirror-image of Marx's project — same historical transformation, opposite question.

Marx looked at industrial society and asked what was tearing it apart. Durkheim looked at the same society and asked why it had not already flown to pieces.

Two places, and a question.

A village in the Auvergne, 1830. Four hundred people. Almost everyone farms. They do roughly the same work with roughly the same tools, follow the same calendar, attend the same church, know the same songs, and are related to one another in ways that would take an afternoon to map. A man from the next valley is recognisably a foreigner.

Ask why they hold together and the answer is easy: they are the same. Same beliefs, same work, same lives. Their unity is the unity of a thing made of identical parts.

Now break one of the rules here — steal, blaspheme, marry the wrong person — and the reaction is immediate, communal and severe. Everyone is offended, because the rule violated is everyone's rule.

Paris, 1893. Two and a half million people. A man spends his life setting type for newspapers he does not read. Beside him lives a clerk in an insurance office, upstairs a seamstress who works for a shop she has never entered, next door a chemist. They come from six different provinces and speak French in four accents. Most hold no strong religion in common. Many hold none.

They agree on almost nothing. They do entirely different work. Most are strangers.

And this arrangement is not fragile. It is spectacularly productive and remarkably stable.

By every intuition available in 1830, it should not work. Durkheim's question is: what is holding it together, given that similarity plainly is not?

State the difficulty precisely, because it is not obvious.

Almost every social thinker before Durkheim assumed that social cohesion comes from shared belief. Common religion, common custom, common blood. On that assumption, modernity — which dissolves all three — should produce a war of all against all, or at best an unstable aggregation held down by force.

This was the conservative diagnosis after 1789 (3.2.3): the Revolution destroyed the bonds, and what remained was atomised individuals and chaos.

It was also, in a different key, the liberal-economic assumption: society is a contract between self-interested individuals, held together by mutual advantage.

Durkheim rejects both, and his reason is one of the sharpest arguments in the book.

Against the conservatives: modern society does hold together, demonstrably, without shared belief. So shared belief cannot be the only mechanism.

Against the contract theorists: contracts cannot be the foundation, because contracts presuppose a society that enforces them. His formulation is famous — "not everything in the contract is contractual." A contract works only because there is already a shared understanding that promises bind, that certain terms are unconscionable, that courts will enforce, that fraud is wrong. None of that is in the contract; all of it must exist before any contract is possible.

So: what is the modern bond?

His answer is startling. It is the division of labour itself — the very thing that makes people different — that binds them together. Difference, organised, produces dependence, and dependence is a stronger bond than similarity.

The idea, in plain words

The evidence: law as a visible index

Here Durkheim does something methodologically clever, and it is the part that makes the book a work of sociology rather than philosophy.

Solidarity is invisible. You cannot observe it, measure it or survey it. So — following his own rule about external indicators (4.2.1) — he looks for something visible that varies with it.

His choice: law. Every society writes down its rules and its reactions to their violation. Law is solidarity made visible.

The reasoning behind repressive law is worth spelling out, because it is one of his best arguments. Why punish at all? Not, he says, primarily to deter or reform — those are rationalisations offered after the fact. Punishment exists to repair the damage done to collective sentiment. A crime offends the shared moral consciousness; the public ritual of punishment reasserts that the rule still holds and that everyone still shares it. This is why punishment is public, why it is expressive and often disproportionate, and why people who will never meet the victim feel personally affronted by a notorious crime.

The evidence, honestly. Durkheim's historical claim has fared badly. Legal historians pointed out early that ancient legal systems contain vast amounts of restitutive law — Roman law is largely a body of civil and commercial regulation, and Babylonian and Hindu codes contain elaborate provisions on contract, property and compensation. Meanwhile modern states have expanded criminal law enormously — regulatory offences, drug law, traffic law, immigration law — and imprison far more people than nineteenth-century states did. The neat historical curve is not there.

What survives is the distinction between the two functions of law, and the insight that punishment is expressive rather than merely instrumental. That insight founded the sociology of punishment: Durkheim's line runs directly to David Garland's Punishment and Modern Society (1990) and to every argument about why penal severity tracks public anxiety rather than crime rates.

The cause: dynamic density

If the division of labour produces organic solidarity, what produces the division of labour?

Durkheim rejects the obvious answer — that people specialised because it made them happier or richer. That, he says, explains why specialisation persists, not why it began; and it is exactly the confusion of function with cause he warned against (4.2.1).

His causal claim is morphological, and it is a social fact explaining a social fact:

Material density — more people in a given space, as population grows and settlements concentrate.

Moral or dynamic density — the frequency and intensity of interaction between them, which rises with material density and is multiplied by roads, canals, railways, post and telegraph.

Rising dynamic density intensifies competition for the same resources and roles. And here Durkheim borrows explicitly from Darwin: when organisms compete for the same niche, one resolution is elimination — but another is differentiation. Species that diverge into different niches can coexist.

So specialisation is the peaceful solution to competition. Two shoemakers in a village fight for the same customers; in a city, one becomes a bootmaker and the other a maker of ladies' shoes, and both survive. Multiply this across every trade and you get the modern division of labour.

Note the shape of the explanation. No one intended it. No one designed it. It emerges from the aggregate of individual adaptations under competitive pressure — and it produces a new form of social bond as a by-product. This is a genuinely sophisticated piece of theorising, and it is why the book still gets read.

Consider what it took for you to eat this morning.

Someone grew the grain, in a country you may not know. Someone bred the seed; someone made the fertiliser; someone drilled for the gas the fertiliser required. Someone drove the harvester, which someone else designed and a factory of a thousand people built. Someone milled, someone shipped, someone baked, someone drove, someone stocked the shelf. Someone maintained the road. Someone generated the electricity. Someone wrote the software routing the delivery lorry.

You could not survive a month if these people stopped. Not because you lack willingness, but because you lack the knowledge. You cannot make bread from scratch — you cannot make a plough, and you could not smelt the iron for one.

Durkheim's point: this dependence is a bond of extraordinary strength, and it does not require you to share a single belief with any of these people. They may hold opposite politics, worship differently or despise your way of life. It makes no difference to the bond. The tie is functional, not moral, and it is unbreakable in a way that shared belief never was — you can stop believing together, but you cannot stop needing each other.

And the fragility is instructive too. When such a chain breaks — a fuel blockade, a port closure, a pandemic emptying shelves — people discover in a week how thoroughly they depend on strangers. Those moments are organic solidarity becoming visible, which is exactly what Durkheim said breakdowns do (4.2.1's pathological cases).

When it goes wrong: the abnormal forms

Durkheim does not think the division of labour automatically produces solidarity. Book III of the thesis is on the abnormal forms — the conditions under which it fails — and it is the most Marx-adjacent part of his work.

The anomic division of labour. When specialisation outruns the moral rules regulating it, functions are no longer properly coordinated and people do not know what they owe one another. Durkheim's own examples are industrial crises, the antagonism of labour and capital, and the fragmentation of the sciences into mutually unintelligible specialisms. Rules have not kept pace with structure. This is the seed of the concept that becomes 4.2.3.

The forced division of labour. When the distribution of roles does not correspond to the distribution of talents — when position is inherited, imposed or bought rather than earned. Durkheim is explicit and, on this point, radical: solidarity requires that "external inequality" be abolished, that competition be genuinely fair, and that inherited wealth be substantially curtailed. He argues for abolishing inheritance. His argument is not egalitarian in outcome — he accepts unequal rewards for unequal talents — but he insists that where roles are forced, the division of labour produces conflict rather than solidarity, and the resulting arrangement is a constrained one that people rightly resent.

Poor coordination. Where specialisation exists but activity is insufficiently linked, producing waste, idleness and lack of purpose — the bureaucratic pathology.

Go deeper

The critique of Spencer, and why it matters

A large portion of the book is an argument with Herbert Spencer (3.3.2), and the argument is the one that separates sociology from economics.

Spencer's position: modern society is held together by contract. Individuals pursue their interests, exchange freely, and the aggregate is a functioning order requiring minimal state or shared morality.

Durkheim's reply, in three moves.

Contracts presuppose non-contractual conditions. Someone must have established that promises bind, which terms are void, what counts as fraud, who adjudicates. That framework precedes any particular contract and is not itself contracted.

Contracts presuppose trust. Every exchange rests on the expectation that the other party will not simply take the money and vanish. Where that expectation fails, exchange stops — as it does in societies with weak legal institutions, at enormous economic cost.

And self-interest is a weak glue. "There is nothing less constant than interest. Today it unites me to you; tomorrow it will make me your enemy." A society whose only bond is mutual advantage dissolves the moment advantage shifts.

This argument is now foundational to economic sociology (1.4.2). Every demonstration that markets require trust, norms and institutions to function — Granovetter on embeddedness, the entire literature on institutions and development — descends from these pages.

The relation to Marx, stated fairly

The two are usually opposed, and the opposition is real but narrower than the textbooks suggest.

Same object. Both are explaining the transition to industrial society and both regard the division of labour as central.

Different diagnosis of the pain. Marx says the suffering of the industrial worker follows from property relations — from who owns the means of production, which produces alienation (4.1.4) and exploitation (4.1.5). Durkheim says it follows from insufficient moral regulation — the division of labour has grown faster than the norms coordinating it.

Different remedy. Marx: change ownership. Durkheim: build the regulative institutions — occupational groups, professional ethics, a state that guarantees fair access.

And an underrated overlap. Durkheim's "forced division of labour" is a critique of inherited advantage as sharp as anything in the reformist tradition, and his account of anomic industrial conflict identifies the same phenomena Marx did. Where they differ is on whether the cause is ownership or regulation — which is, in the end, an empirical question that has been argued ever since.

And a real blind spot in Durkheim. He treats the division of labour as broadly the product of impersonal competition, which underplays the extent to which it was imposed: by enclosure, by colonial conscription, by the deliberate deskilling Braverman described (4.1.4). Who decides who specialises in what, and who gets the specialism that pays — these are questions his framework raises and does not answer.

What the evidence says now

On interdependence as a bond: strongly supported, and stronger than Durkheim could have known. Global supply chains have made functional interdependence planetary. The pandemic and subsequent shipping disruptions gave everyone a practical demonstration.

On the decline of shared belief: partly. Religious homogeneity has declined in most rich societies (3.2.6), but the "cult of the individual" Durkheim predicted — a shared moral commitment to human dignity, rights and personal autonomy — is arguably the strongest common creed modern societies possess, and he called it (4.2.5). Human rights discourse is his prediction coming true.

On whether interdependence is enough: the open question, and the one his critics press. Contemporary work on polarisation, social trust and civic disengagement suggests that functional interdependence coexists comfortably with mutual hostility. People who depend absolutely on each other's labour can also loathe each other's politics, live in separate information environments, and refuse to share a table. Durkheim would say that is precisely anomie — structure without adequate moral regulation — which makes his framework a diagnosis of the present rather than a description of a solved problem.

"Mechanical solidarity is modern." The reverse. Mechanical = solidarity of the similar, characteristic of small undifferentiated societies. Organic = solidarity of the different.

"Durkheim thought modern society has no shared morality." He thought the collective conscience becomes thinner, more abstract and more general — centring on the dignity of the individual — not that it disappears.

"The repressive-to-restitutive law shift is established." It is Durkheim's hypothesis and the historical evidence largely contradicts it. The distinction survives; the trend does not.

"The division of labour is caused by its benefits." Durkheim explicitly rejects this as confusing function with cause. His cause is rising dynamic density and competitive differentiation.

"Durkheim was a conservative defender of the status quo." He argued for abolishing inheritance, for equality of opportunity as a condition of legitimate solidarity, and for occupational associations to regulate industry. Reformist, certainly; complacent, no.

"Interdependence guarantees cohesion." Durkheim's own abnormal forms say it does not. Without regulation, specialisation produces anomie and conflict — which is the next lesson.

Why it matters

It answers the question the whole discipline started from. The problem of order (1.2.1) gets its most influential modern answer here: modern societies cohere through organised difference, not shared belief.

It shows how to make an invisible thing researchable. Solidarity cannot be observed, so Durkheim found a visible index that varies with it. The hypothesis failed; the move is what students of method should take away, and it is the same move as measuring integration through suicide rates (4.2.4).

It founds economic sociology. The demolition of contract-as-foundation — "not everything in the contract is contractual" — is why economics and sociology are different subjects, and why questions of trust, norms and institutions are now central to understanding markets.

And it introduces the concept that made him famous. The abnormal forms — specialisation outrunning its rules — name a condition that Durkheim would spend the next four years developing into the idea most people know him for, and which he would use to explain something nobody thought sociology could explain at all.

Durkheim's question is the mirror of Marx's: not what tears modern society apart, but what holds it together when its members share neither work, belief nor acquaintance.

Mechanical solidarity = cohesion through likeness (small, undifferentiated societies, strong collective conscience). Organic solidarity = cohesion through difference and interdependence (complex societies, thinner collective conscience). The names are counterintuitive: mechanical = same parts; organic = unlike organs that need each other.

Law as the visible index: repressive law punishes and expresses shared sentiment; restitutive law restores and expresses interdependence. Punishment exists to repair damaged collective feeling, not primarily to deter — an insight that founded the sociology of punishment. The predicted historical shift, however, is not supported by the evidence.

The cause of the division of labour is not its usefulness (that would confuse function with cause) but rising material and dynamic density producing competition, resolved through differentiation rather than elimination.

Against Spencer: contract cannot be the foundation, because "not everything in the contract is contractual" — contracts presuppose a framework of rules and trust they cannot themselves create. This is the founding argument of economic sociology.

Abnormal forms: the anomic division of labour (specialisation outrunning its regulation) and the forced division of labour (positions distributed by inheritance and constraint rather than talent — for which Durkheim proposed abolishing inheritance).

Versus Marx: same phenomenon, different diagnosis — property relations versus insufficient moral regulation — and correspondingly different remedies.

Social solidarity — the bonds holding a society together.

Mechanical solidarity — cohesion based on likeness, characteristic of undifferentiated societies.

Organic solidarity — cohesion based on specialisation and interdependence.

Segmental society — Durkheim's term for a society composed of similar, self-sufficient units (clans, villages).

Repressive law — law that punishes, expressing the strength of shared sentiment.

Restitutive law — law that restores, expressing the complexity of interdependence.

Material density — the concentration of population in space.

Dynamic (moral) density — the frequency and intensity of interaction, amplified by transport and communication.

"Not everything in the contract is contractual" — Durkheim's argument that contracts rest on prior non-contractual foundations.

Anomic division of labour — specialisation insufficiently regulated by shared norms.

Forced division of labour — the distribution of roles by inheritance or constraint rather than by aptitude.

Cult of the individual — the shared modern moral commitment to human dignity and personal autonomy.

One — trace a day's dependence. List everything you used before noon and estimate how many people's specialised work it required. Then ask how many of them you would recognise, and how many share your beliefs. The gap is Durkheim's point.

Two — test the two solidarities. Think of one group you belong to that coheres through similarity and one that coheres through complementary roles. Ask which survives disagreement better. The answers are usually surprising.

Three — find the non-contractual foundations. Take any everyday transaction — buying a coffee, renting a flat — and list everything both parties assume that is nowhere written down. That list is Durkheim's reply to Spencer.

Four — check the law hypothesis. Find out whether the number of criminal offences in your country's statute book has grown or shrunk over fifty years. Durkheim predicted repressive law would recede. Report honestly what you find.

Five — diagnose an abnormal form. Identify one occupation where the rules have not kept up with how the work is now done — platform work, care work, content moderation, AI-assisted professions. That is the anomic division of labour, and it is what 4.2.3 is about.