Most explanations of the mode of production give you a list. Primitive communal, ancient, feudal, capitalist, and sometimes an Asiatic entry and a socialist one, each with a paragraph describing what people did for a living. A reader finishes the list able to recite it and no better placed to answer the question that actually arrives in a seminar or an examination, which is what mode of production a given society was in and how you would know.

The list is not the concept. The concept is a specification: a mode of production is identified by the way surplus labour is pumped out of the direct producers, and everything else follows from that. Once you have the criterion, the list becomes a set of worked examples rather than a thing to memorise, and you can apply the analysis to a society nobody has classified for you. That is the difference between a category and a method, and it is the difference this guide exists to install.

Call the criterion the surplus-extraction test: a mode of production is identified not by its technology and not by its politics but by how surplus labour is transferred from those who perform it to those who do not, and any classification that has not answered that question has not identified a mode at all. The test is what disqualifies the common answers. It is not the level of technology, because societies with comparable techniques have organised extraction in opposite ways. It is not the presence of markets, because markets long predate capitalism and were compatible with several modes. It is not the political system, because the same relations of production have persisted under republics, monarchies, and colonial administrations. What distinguishes a mode is the mechanism, and the mechanism is what the analysis has to name.

What a mode of production is, stated precisely

A mode of production is the unity of the productive forces of a society and the relations of production through which those forces are set in motion, considered as a system with its own conditions of reproduction and its own characteristic contradictions.

Three components are packed into that sentence and each needs unpacking.

The productive forces are the means by which a society produces: instruments and machinery, raw materials and land under cultivation, technical knowledge, and the labour power of the producers themselves, including their skills and their capacity for cooperation. The last of these is regularly dropped from summaries and should not be, because Marx treats the organisation of labour as a productive force in its own right, arguing in the first volume of Capital that the productive power of combined labour exceeds the sum of the individual powers combined. A reorganisation of work counts as a development of the forces even where no new machine appears.

The relations of production are the social relations through which production is carried on, and specifically the relations determining effective control over the productive forces and over the labour that uses them. In a class society these are relations of appropriation. They specify who works, who commands the work, and how the product above the subsistence of the producers is transferred to someone else. A wage relation is a relation of production. A relation between a lord and a peasant owing labour services is a relation of production. A relation between a slaveholder and an enslaved worker is a relation of production. What makes each of them a relation of production rather than a merely legal or personal relation is that surplus labour flows through it.

The unity of the two is what makes a mode. Neither component alone identifies anything. A steam engine is compatible with wage labour, with slave labour, and with state-directed labour, all of which have historically existed. A relation of dependency between a landholder and a cultivator is compatible with an enormous range of techniques. The mode is the combination, considered as something that reproduces itself: a set of arrangements that regenerates its own conditions period after period, and that contains within itself the tensions capable of destroying it.

That last clause matters more than it looks. A mode of production is not a snapshot. It is a system with conditions of reproduction, which means an analysis must be able to say what has to keep happening for the arrangement to persist into the next season, the next generation, the next century. If the answer is that the producers must be kept separated from the means of production, that is a fact about capitalism with enormous consequences. If the answer is that a coercive apparatus must be maintained to prevent flight, that is a fact about serfdom with different consequences. Naming the conditions of reproduction is half of what the concept is for.

How would you define the concept in one sentence?

It is the whole way a society produces its material life: the technical means it uses, and the social arrangements determining who works for whom and how the surplus is taken. What identifies one mode against another is the mechanism of that extraction rather than the level of technology or the presence of markets.

Forces, relations, and the terms people confuse them with

Three distinctions cause more trouble on this topic than anything else, and each is worth drawing sharply before the substantive material begins.

The first is between the mode of production and the means of production. This is the single commonest error in student writing and in general explainers. The means of production are things: land, tools, machinery, buildings, raw materials, the physical and infrastructural resources with which production is carried on. The mode of production is a whole way of organising production, of which the means are one element. Ownership of the means of production is a relation, and it is part of what specifies the mode; the means themselves are objects. The sentence that a society changed its means of production describes a technical change. The sentence that a society changed its mode of production describes a transformation of who works for whom. Confusing the two produces the technological reading of the theory that its most careful defenders reject.

The second is between the forces and the relations. The forces are what a society can do; the relations are how it is organised to do it. The distinction is analytical rather than physical, which is why some elements sit awkwardly: a machine designed to deskill a workforce is a productive force whose design encodes a relation, and the labour process is simultaneously a technical arrangement and a structure of control. That awkwardness is not a defect in the distinction; it is what the whole fetters argument is about, since the theory holds that the two components can come into conflict, which requires that they be distinguishable in the first place. The complete guide to historical materialism sets out the full range of positions on which of the two carries explanatory weight, and this article does not adjudicate that dispute.

The third is between the mode of production and the base. The base, in the base and superstructure model, is the economic structure constituted by the relations of production, considered as the foundation on which a legal and political order rests. The mode of production is the larger unit including the forces. The two terms answer different questions: the base is a term in a model of explanation, while the mode is a term in a classification of societies. Substituting one for the other misplaces technology, which sits inside the mode without straightforwardly sitting inside the base, and the treatment of the base and superstructure pair works through the boundary questions that generates.

Why do these two terms get confused so often?

The means of production are things: land, tools, machinery, raw materials. The mode of production is the entire way a society organises production, including who owns those means, who works them, and how surplus labour is extracted. One is an inventory of objects; the other is a description of a social system.

Where the concept comes from in the texts

The concept is assembled from four bodies of material with different statuses, and knowing which is which is what separates a citation that survives scrutiny from one that does not.

The manuscripts written with Engels between 1845 and 1846 supply the earliest treatment, organised around forms of ownership succeeding one another as the division of labour develops: tribal ownership, ancient communal and state ownership, feudal or estate ownership. The vocabulary there is ownership rather than mode, and the sequence is offered as a description of European development rather than as a universal law. The text was unpublished in either author’s lifetime and its famous first part survives as manuscript bundles whose ordering is an editorial decision, so a writer citing it should say what they are citing.

The preface to A Contribution to the Critique of Political Economy of 1859 supplies the compressed canonical statement and the list of epochs. In a single sentence at the end of a few pages of theoretical compression, Marx names the Asiatic, ancient, feudal, and modern bourgeois modes of production as epochs marking progress in the economic formation of society. Four terms, one sentence, no claim that every society passes through all of them. Everything the textbooks have made of this list is extrapolation, and the article on historical materialism in this series documents when and by whom the extrapolation was performed.

The notebooks written in 1857 and 1858, published long afterwards as the Grundrisse, contain the fullest treatment of pre-capitalist arrangements anywhere in Marx. The section usually printed separately under a title concerning pre-capitalist economic formations examines the different forms in which a community’s relation to land is organised, distinguishing arrangements in which the individual holds land as a member of a community from those in which a state or its representative stands as the ultimate proprietor. It is exploratory writing, not a finished position, and citing it as though it carried the authority of a published work is a sourcing error a specialist will notice. It is also indispensable, because nothing else in the corpus does the same work.

The first volume of Capital, published in 1867, supplies the analysis of the capitalist mode itself and, in its final part on so-called primitive accumulation, the account of how the mode came into being. The argument there is that capitalist relations require a prior historical process producing a population separated from the means of subsistence and therefore compelled to sell labour power, and that this separation was accomplished through expropriation rather than through thrift. The mechanism has its own article in this series, and the treatment of primitive accumulation covers it in full.

The third volume, assembled by Engels from manuscripts and published in 1894, contains the passage that makes surplus extraction the defining criterion of a mode. Marx argues there that the specific form in which unpaid surplus labour is pumped out of the direct producers determines the relationship of rulers and ruled, and that this relation is the innermost secret and hidden basis of the entire social structure. That is the sentence the surplus-extraction test comes from, and it is worth noting its status: it appears in a volume Marx did not prepare for publication, in a discussion of ground rent, and it is nonetheless the clearest statement of the criterion anywhere in his work.

Which text should a researcher cite for the concept?

For the definition and the epochs, the preface to A Contribution to the Critique of Political Economy. For the criterion of extraction, the discussion of ground rent in the third volume of Capital, cited with the acknowledgement that the volume is a posthumous editorial construction. For pre-capitalist forms, the relevant section of the Grundrisse, cited as notebook material rather than as a finished position.

The surplus-extraction test

Every society above the barest subsistence produces more than its direct producers consume. The question that identifies a mode of production is how that surplus gets from the people who produce it to the people who live on it, and the answer takes a different form in each mode. Running the test has four steps and they can be applied to any society, historical or contemporary, that a reader is asked about.

Ask first who the direct producers are. This is not always obvious and the answer is frequently plural. In a society with slave plantations, peasant smallholdings, and urban artisans coexisting, there are three distinct groups of direct producers standing in three distinct relations, and any classification that names only one has already lost most of the information.

Ask second what they must do to gain access to the means of production. This question separates the modes more sharply than any other. Where producers hold their own means of subsistence and access is secured by membership of a community, no market compulsion exists and the surplus must be extracted by some other force. Where producers hold land but owe dues to a superior, access is conditional on the dues. Where producers hold nothing and must sell their capacity to work, access is mediated by a market and no direct coercion is required at the point of production.

Ask third by what mechanism the surplus is taken. This is the heart of the test, and the crucial distinction is between extraction that requires extra-economic compulsion and extraction that does not. Under serfdom, a peasant working their own holding produces their own subsistence, so the surplus can only be obtained by compelling additional labour or additional payment, which requires a coercive apparatus: courts, armed retainers, legal disabilities on movement. Under capitalism, the worker’s entire working day is bought and the surplus is contained within it, which means no direct coercion is needed at the point of extraction and the transaction appears as a voluntary exchange of equivalents. That difference in visibility is not a detail. It is why capitalism required a new science to analyse it and why earlier forms of exploitation were obvious to everyone including their beneficiaries.

Ask fourth what has to keep happening for the arrangement to persist. Every mode has reproduction conditions and naming them is what turns a classification into an analysis. Serfdom requires that peasants not flee, which requires a legal and coercive apparatus and explains why the enforcement of restrictions on movement is a recurring theme in the historiography. Capitalism requires that the separation of producers from the means of subsistence be maintained, which explains why the enclosure of common resources and the treatment of independent subsistence are political questions rather than incidental ones. A slave economy requires continual replacement of the enslaved population, whether through reproduction or through capture and purchase, which makes its relation to external supply a structural rather than a contingent matter.

A classification that has answered all four questions has identified a mode. A classification that has answered none of them has identified a period, which is a different and much less useful thing, since periods are named after rulers, centuries, and wars rather than after the way surplus labour moves.

How do you tell which mode of production a society is in?

Ask who the direct producers are, what they must do to gain access to the means of production, by what mechanism their surplus labour is transferred to others, and what has to keep happening for the arrangement to reproduce. The answers identify the mode. Technology, market activity, and political form do not.

The mode-of-production specification sheet

The table below applies the test to each of the modes that appear in the literature. It is not a list of stages and should not be read as one; the ordering is roughly chronological in European experience and carries no implication that any society must pass through each. The final row is included because it is discussed constantly, and it is marked as disputed because whether it names an achieved mode of production is one of the tradition’s genuinely open questions.

Mode	Dominant productive forces	Characteristic relations	Mechanism of surplus extraction	Form of the ruling stratum	Principal internal contradiction
Primitive communal	Simple tools, foraging or early cultivation, cooperative labour within kinship groups	Communal access to means of subsistence, no permanent separation of producers from resources	No systematic extraction, since little durable surplus exists and no group is separated from the means of subsistence	None, in the developed sense; leadership without permanent appropriation	Growth of surplus and of settled production creating the conditions for permanent inequality
Ancient or slave	Agriculture and craft using slave labour, substantial urban concentration, military organisation	Ownership of the producer as property; the enslaved worker’s whole person is the commodity	Direct appropriation of the entire product beyond the maintenance of the producer, secured by ownership and force	Landholding citizens, military and civic elites	Dependence on external supply of enslaved people and the political costs of maintaining a large unfree population
Asiatic, as Marx sketches it	Irrigated or extensive agriculture organised at a scale exceeding the individual community	Village communities holding land collectively with a state or its representative as ultimate proprietor	Tax and tribute taken by a state apparatus, with tax and rent effectively merged	A state bureaucracy rather than a private landowning class	Contested; Marx’s own account emphasises the stability and reproduction of the village unit rather than a developmental dynamic
Feudal	Agriculture on peasant holdings with limited technical change, local craft production	Peasants in possession of their means of subsistence but subject to a lord’s claims and to restrictions on movement	Rent in labour, in kind, or in money, requiring extra-economic compulsion because producers hold their own subsistence	Landholding nobility with judicial and military powers	The struggle over the rate of rent and over the peasants’ freedom of movement and disposal
Capitalist	Machinery, generalised commodity production, a technically transformed labour process	Producers separated from the means of production and compelled to sell labour power; means of production held as capital	Surplus value produced within the working day and appropriated through the wage relation, requiring no direct coercion at the point of extraction	Owners and controllers of capital, increasingly separated from personal ownership	Accumulation dependent on labour while continually displacing it; the social character of production against private appropriation
Socialist, disputed	Industrial forces inherited from the previous mode	Formal social or state ownership of the means of production, with the character of effective control disputed	Contested. The question is whether a surplus is extracted from producers by a controlling stratum, and if so through what mechanism	Contested. Candidates in the literature include a bureaucratic stratum, a state bourgeoisie, and no ruling class at all	Contested, and the dispute is the substance of the argument about whether the row describes a mode at all

Three cautions belong with the table.

There is no agreed count of modes and no reader should assert one. Marx names four epochs in the 1859 preface, uses different vocabulary elsewhere, and the tradition has proposed additional candidates including a Germanic form, a Slavonic form, a colonial mode, and a petty commodity mode. A source that tells you there are exactly five modes is reporting a codification rather than a finding, and the codification in question has a datable political history rather than a scholarly one.

The absence of an agreed count is not a weakness in the concept, though it is often presented as one. The criterion generates categories rather than the categories generating the criterion, which means new mechanisms of extraction discovered in the historical record can produce new categories legitimately, provided each is specified by the same test and each carries its own reproduction conditions. What would be a weakness is a fixed list defended against the evidence, and that is precisely what the five-item version became wherever it was taught as doctrine.

The rows are ideal types rather than descriptions of any actual society. No historical society has ever been a pure instance of any row, which is not a defect in the analysis but a consequence of what the rows are for. The section below on social formations explains why.

The last row is not filled in because the tradition has not filled it in. Presenting a confident specification for a socialist mode of production would misrepresent the state of the argument, and the honest treatment states the candidate positions and their evidence rather than choosing one silently.

The modes examined

Primitive communal production

The category names arrangements in which no permanent separation exists between producers and the means of subsistence, access is secured through membership of a kin or territorial group, and no stratum lives systematically on the labour of others. It is the limiting case of the analysis: a form in which the surplus-extraction test returns no answer because there is no systematic extraction to identify.

The category’s status in the tradition owes more to Engels than to Marx. Engels developed it at length in his work of 1884 on the family, private property, and the state, drawing heavily on the anthropological research of Lewis Henry Morgan, and much of that research has been superseded. What survives of the argument is the general proposition that class divisions are historically produced rather than natural, and the specific proposition that the emergence of a durable surplus is the condition of permanent stratification. What has not survived is the specific evolutionary sequence Morgan proposed and the confident reconstruction of early kinship organisation built on it. A writer citing this material should distinguish the general claim, which the archaeological record broadly supports, from the particular reconstruction, which it does not.

The ancient or slave mode

The defining relation is ownership of the producer. Where a worker is themselves property, the entire product beyond their maintenance is appropriated directly, and the maintenance itself is a cost borne by the owner in the same way as the upkeep of any other asset. This makes the extraction mechanism the most visible of any mode and the surplus the most complete.

Three features follow from the specification and are worth carrying. Because the worker is capital rather than a seller of labour power, the owner bears the cost of subsistence in slack periods, which creates pressure towards work regimes that use the asset intensively and towards markets in which surplus workers can be sold. Because the population of producers cannot reproduce itself at the required rate under many conditions, the mode has a structural dependence on external supply, whether through conquest, purchase, or organised reproduction, which links it to military and commercial arrangements beyond the productive unit. And because the producer has no interest whatever in the product, the mode has a persistent difficulty with any technique requiring initiative or care, which is a genuine constraint on its productive development and one that contemporaries observed.

The category’s application beyond the classical Mediterranean is contested. Plantation economies in the Atlantic world used slave labour on a large scale within a world market organised on capitalist lines, which produces the question of whether they constituted an ancient mode surviving into a later period, a distinct plantation mode, or a form of capitalist production using unfree labour. The question is live, the answers are argued from evidence, and the surplus-extraction test does not settle it by itself, because the extraction at the point of production and the appropriation at the level of the world market have different characters.

The feudal mode

The defining feature is that the direct producer is in possession of their means of subsistence. A peasant holding land can feed their household without the lord’s involvement, which means the lord’s claim on the surplus cannot be secured through the production process itself. It must be secured through some other force: legal obligation, judicial power, restrictions on movement, the threat and use of violence. This is what the literature means by extra-economic compulsion, and it is the single most important structural fact about the mode.

Everything characteristic of feudal societies follows from that fact rather than from castles or chivalry. The fusion of political and economic power in the same hands, so that the extractor is also the judge, is a requirement of the extraction mechanism rather than an accident of the period. The legal disabilities on peasant movement are the reproduction condition. The fragmentation of sovereignty is the political form that a decentralised extraction apparatus takes. And the persistent conflict over the rate and form of dues, which is what the historiography of peasant resistance is about, is the mode’s principal internal contradiction operating in plain view.

The shift from labour rent to rent in kind and then to money rent, which Marx discusses in the third volume, is a change in form that leaves the mode intact while altering its dynamics considerably. Money rent requires the peasant to sell part of the product, which draws the household into market relations without changing who holds the land or who owes what to whom. That is a useful case for a reader learning the test, because it demonstrates that market involvement is not the criterion.

The capitalist mode

The defining features are three and they must be held together, because each alone is compatible with other arrangements.

The first is that the direct producers are separated from the means of production. They hold nothing they can live on, which means they must obtain access to the means of subsistence by selling the only thing they have, which is their capacity to work. The second is that this capacity is a commodity bought and sold on a market, with a price determined like other commodity prices by the cost of its reproduction. The third is that production is generalised commodity production: goods are produced for sale rather than for use by the producer or for direct transfer to a superior, and the coordination of the economy occurs through exchange rather than through custom or command.

The extraction mechanism is what distinguishes the mode most sharply. The employer purchases labour power for a period and obtains its use for that period. The value the capacity creates during the period exceeds the value of the capacity itself, and the difference is surplus value. Nothing in the transaction is coerced at the point of extraction, no legal disability is required, and the exchange is one of equivalents by the standards of the system: the seller receives the value of what they sold. This is why exploitation in the technical sense is compatible with a scrupulously fair wage bargain, and it is the point on which almost every popular account of the theory goes wrong.

Two structural consequences follow. Competition compels each capital to reduce costs, which drives technical change continually rather than episodically, making the capitalist mode the first with a built-in dynamic of transformation of its own productive forces. And accumulation depends on living labour while continually displacing it through that same technical change, which is the mode’s principal internal contradiction and the source of its crisis tendencies.

The disputed socialist mode

Whether a socialist mode of production has ever existed is one of the tradition’s most consequential unresolved questions, and the honest treatment sets out the positions rather than choosing.

The question is answered by running the same test. If the producers in a state socialist economy performed surplus labour that was appropriated by a stratum they did not control, then a relation of exploitation existed and the formation contained a class division whatever its legal form of ownership. If the surplus was appropriated by an apparatus accountable to the producers, then it was not extraction in the relevant sense. Everything turns on what effective control looked like in practice, which is a historical question rather than a definitional one.

The main positions can be stated briefly. The degenerated workers’ state analysis, developed by Trotsky in the 1930s, holds that the property forms were post-capitalist while the political apparatus had been captured by a bureaucratic stratum that consumed part of the surplus without constituting a class, since it had no property relation to transmit. The state capitalist analysis, developed from the 1940s by Tony Cliff and by others working independently, holds that the apparatus functioned as a collective capital under the pressure of international competition, so the relation between planners and workers was a capital relation in a different institutional form. The bureaucratic collectivist analysis holds that a genuinely new class formation existed, neither capitalist nor socialist, defined by control rather than by ownership. And the official position of the states themselves held that a socialist mode of production had been achieved and that the remaining divisions were non-antagonistic.

Assessing these is beyond a concept article and the regional and historical articles in this series carry it. What the concept contributes is the discipline: the question is not what the constitution said about ownership but who controlled the surplus and by what mechanism, and any answer that does not address that has not used the concept.

The Asiatic mode and why it is contested

No part of this topic is more politically charged or more frequently misrepresented, and the responsible treatment separates four things: what Marx wrote, what the category was made to do, why it was suppressed, and why it was revived.

What Marx wrote is thin. The Asiatic mode appears in the list of epochs in the 1859 preface and is developed, exploratorily, in the notebooks of the late 1850s. The core idea is a form in which village communities hold land collectively, in which the individual is a proprietor only as a member of the community, and in which a state or its representative stands as the ultimate proprietor above the villages, so that tax and rent are effectively merged into a single payment to a central apparatus. Marx associated this with large-scale hydraulic works requiring coordination beyond the capacity of individual communities, an association he took partly from the political economy and travel literature available to him rather than from independent research. His journalism on British rule in India works with a version of this picture, and those articles contain the passages describing an unchanging village order broken up by colonial commerce that later readers have found most objectionable.

What the category was made to do is where the trouble starts. A mode defined by a state apparatus extracting surplus from village communities, with no private landowning class and no internal dynamic of transformation, was available for two very different uses. It could function as an analysis of a genuinely distinct social formation, which is a legitimate research programme. Or it could function as an explanation of why certain societies did not develop as Europe did, which slides towards a doctrine of stagnation with an evident colonial application. The second use is what makes the category dangerous, and it is present in the sources rather than being an invention of later critics.

Why it was suppressed is a matter of Soviet intellectual history. Theoretical debates in the Soviet Union at the start of the 1930s effectively removed the category from official doctrine, and the five-stage schema that emerged from that period has no place for it. The reasons were not purely scholarly. A category describing a formation in which a state bureaucracy extracts surplus from a peasantry, with no private property in land, had applications uncomfortably close to the Soviet situation itself, and it also cut against the political requirement that colonial societies be understood as feudal and therefore as candidates for a bourgeois-democratic stage. A concept that had been in the founding text was removed by decision, which is a fact about the doctrine’s formation that every reader should know.

Why it was revived is a matter of the politics of the mid-twentieth century. Karl Wittfogel’s study published in 1957 developed a theory of hydraulic society and oriental despotism, arguing that societies organised around large-scale water management generate total bureaucratic power, and he explicitly extended the analysis to the Soviet Union and to the Chinese state. Wittfogel had been a communist and the book is a work of anti-communist argument as well as of comparative history, which is relevant to how it should be read. Its scholarly reception has been severe: the causal claim that hydraulic works generate despotic states has been extensively challenged by historians and archaeologists who find that centralised states frequently precede large-scale irrigation rather than arising from it, and the geographic and chronological range of the argument has been criticised as far exceeding its evidence.

Where the debate stands now is that most historians of the relevant regions do not use the category, that the empirical claims associated with its most influential defender have not held up, and that its underlying question remains real. The question is whether there are class societies in which surplus is extracted by a state apparatus rather than by a private propertied class, and that question survives the collapse of the specific answer. Formations in which taxation is the primary extraction mechanism are not well described by either the feudal or the capitalist specification, and the analytical gap is genuine. The concept’s afterlife in the analysis of Chinese history has been particularly consequential, and the complete guide to Marxism in China treats the political stakes of that argument, which were considerable.

Why is this category so politically charged?

It names a form in which village communities hold land collectively and a state stands as ultimate proprietor, extracting surplus through tax that functions as rent. It is controversial because it slides easily into a doctrine of oriental stagnation with colonial applications, because it was removed from Soviet orthodoxy by decision rather than by evidence, and because its most influential revival was an explicitly political work whose empirical claims have not survived scrutiny.

What makes capitalism capitalist

The most useful single application of the test is to the question of what capitalism actually is, because the popular answer is wrong in a way that has consequences for everything downstream.

The popular answer is markets, trade, profit-seeking, and private wealth. On that definition capitalism is very old, since long-distance trade, commercial fortunes, banking, and the pursuit of gain are attested across millennia and on every continent. Merchant houses in medieval Italian and north German cities accumulated enormous wealth, financed states, ran credit networks across a continent, and behaved in ways any observer would call commercial. If commerce and profit made a society capitalist, those cities were capitalist, and the concept would have nothing to explain about the transformation of the following centuries.

Marx’s definition produces a different answer. Capitalism is generalised commodity production resting on wage labour, which requires that the direct producers have been separated from the means of production and are therefore compelled to sell their capacity to work. Merchant wealth does not meet this specification. A merchant who buys cheap in one place and sells dear in another is appropriating a share of a surplus produced elsewhere, under whatever relations of production obtain where the goods were made. The merchant profits from the difference between prices, not from the difference between the value of labour power and the value it creates. Where the artisans producing the goods own their tools and workshops, sell their own products, and are not separated from their means of production, no capitalist relation exists no matter how much money changes hands.

This is why merchant wealth in medieval cities did not make those cities capitalist, and it is the most efficient correction available to the popular definition. It also explains a fact the popular definition cannot: that enormous commercial wealth existed for centuries in numerous societies without generating anything resembling industrial capitalism, and that the transformation, when it came, came out of changes in agrarian relations rather than out of the growth of commerce.

Two further consequences follow and both are examinable. The first is that state ownership does not by itself abolish the capitalist relation, since what matters is whether producers are separated from the means of production and must sell labour power, not who holds the title. This is what makes the state capitalist analysis of the socialist states a coherent position rather than a rhetorical insult. The second is that the presence of markets in a society tells you very little on its own. Markets in products existed under feudal relations, under slave relations, and in state-organised economies. What distinguishes capitalism is a market in labour power, which is a much rarer historical arrangement and a much more demanding one, since it requires a population with no other means of subsistence.

Is capitalism a mode of production?

Yes, and it is the one Marx analysed in detail. It is defined by generalised commodity production resting on wage labour, which requires that producers be separated from the means of production and compelled to sell their labour power. Markets, trade, and private profit are not sufficient, since all three long predate it.

Modes and social formations: why real societies are hybrids

A reader who has absorbed the specification sheet will immediately hit a problem: no actual society matches any row cleanly. Rural England in the period of agrarian transformation contained capitalist tenant farming, surviving customary tenures, wage labour, and household production simultaneously. Colonial economies contained plantation slavery, peasant subsistence production, wage labour in mines and ports, and metropolitan capital, all at once. Even fully industrialised economies contain household production, self-employment, informal work, and state provision alongside the wage relation.

The concept that handles this is the social formation. A mode of production is an abstraction identifying a specific way of organising production and extraction. A social formation is an actual society, which typically contains more than one mode, with one of them dominant in the sense that it subordinates the others and shapes the terms on which they persist. The distinction is essential and its absence is what makes many summaries useless: they present the modes as descriptions of societies, which they are not, and then the reader finds that no society fits.

The theoretical development that made this usable is the concept of articulation, worked out principally in French structuralist Marxism from the 1960s and applied extensively by anthropologists working on colonial and postcolonial economies. The argument is that modes do not merely coexist; they are combined in specific ways, with the dominant mode reproducing itself partly through the subordinate ones. A colonial economy in which mining wages are set below the cost of reproducing a worker’s household, on the assumption that subsistence agriculture in the rural areas makes up the difference, is a case where the capitalist sector depends on the persistence of a non-capitalist sector rather than simply displacing it. The persistence of unpaid domestic labour reproducing labour power for the wage sector is the same structure operating inside industrialised economies. In both cases the subordinate arrangement is not a survival awaiting extinction; it is a functioning part of how the dominant mode reproduces itself cheaply.

Articulation solved a real problem and it created another one, which is discussed under the strongest objection below. Used with discipline, it generates specific and testable claims about how sectors depend on each other. Used without discipline, it becomes a device for absorbing any awkward case, and the accusation that Marxist classification is unfalsifiable rests substantially on that abuse.

Why does no real society match a single row?

A mode of production is an abstraction naming one way of organising production and extraction. A social formation is a real society, which almost always contains several modes with one dominant. Expecting an actual country to match a single mode cleanly is a category error, and the concept of articulation exists to describe the combinations.

How many modes can coexist at once?

Yes, and virtually every historical society has contained several. What the analysis identifies is which is dominant, meaning which subordinates the others and sets the terms on which they continue. The interesting question is usually not which modes are present but how the dominant one uses the subordinate ones to reproduce itself more cheaply than it otherwise could.

How one mode changes into another

The transition question is where the concept does its hardest work and where the tradition has produced its most substantial historical scholarship. Two debates carry the argument and a reader who knows both can follow almost any discussion of the topic.

The first debate: internal contradiction or external commerce

The exchange between Maurice Dobb and Paul Sweezy in the 1950s posed the question of how feudalism dissolved in Europe, and it matters because the two answers are both materialist and are incompatible, which makes the disagreement a genuine test rather than a dispute about vocabulary.

Dobb’s position, set out in his study of the development of capitalism, is that the decisive dynamic was internal to the feudal relation: the conflict between lords and peasants over the rate of rent and over the terms of tenure, intensified by the lords’ rising need for revenue, produced a crisis of the extraction mechanism itself. On this account the growth of trade did not dissolve feudalism from outside; it was one of the conditions within which an internal conflict worked itself out, and its effects varied depending on the balance of forces in each region.

Sweezy’s position is that feudalism was a system of production for use with no internal tendency to transform itself, and that the solvent came from outside, in the growth of long-distance trade and of towns organised on different principles, which drew the manorial economy into production for exchange and undermined the relations that had sustained it. On this account the dynamic element is commerce and the feudal relation is essentially static.

The exchange drew in a number of historians and never resolved cleanly, but it established the terms. Dobb’s position has generally been judged the stronger by subsequent scholarship, for a reason that matters beyond this case: a theory in which the dynamic comes from outside the mode has difficulty explaining why the same commercial expansion produced dissolution in some regions and intensified servile obligations in others. That observation is what the second debate built on.

The second debate: class relations or productive forces

Robert Brenner’s intervention, published in 1976 and answered by a series of historians over the following years before the exchange was collected in book form, is the most consequential piece of Marxist historical argument of the past half-century, and its structure is worth understanding whatever one concludes about its verdict.

Brenner’s method is comparative. He observes that the great demographic contraction of the fourteenth century and the commercial expansion of the sixteenth affected large regions of Europe in broadly similar ways, and that the social outcomes diverged sharply. Serfdom declined in the west and was intensified in the east. In England, landlords and large tenant farmers emerged with agricultural production organised on capitalist lines, while in France the peasantry secured possession of the land and smallholding persisted. If demographic pressure or commercial expansion were the drivers, similar inputs should have produced similar outputs. That they did not is evidence that something else was doing the work, and Brenner’s candidate is the balance of class forces on the land, meaning the relative capacity of lords and peasants to secure their claims through organisation, custom, and political power.

The argument has a strong theoretical consequence. If outcomes turn on the contingent balance of class power, then the emergence of capitalism was not the working out of a developmental tendency but the unintended result of a particular struggle whose outcome could have been different. That places Brenner squarely against any version of the theory that gives the productive forces explanatory primacy, and the resulting current, sometimes called political Marxism, has been at odds with the primacy thesis ever since.

The critics of the argument raised substantial points that should be recorded rather than passed over. Several disputed Brenner’s characterisation of French and English property relations and the chronology on which the comparison rests. Several argued that he understated the role of the state, of demographic factors, and of the international context. Several argued that the account of why the balance of class forces differed is itself underspecified, which risks moving the explanatory burden to a factor that is asserted rather than explained. The exchange remains unresolved in its particulars, and any writer summarising it should say so.

What the two debates established jointly is a methodological standard that outlives their specific verdicts. A materialist explanation of a transition must be comparative, must specify a mechanism running through identifiable agents, and must be able to say what different conditions would have produced. Explanations that fail those tests were the casualties of the exchange, and their disappearance improved the historiography regardless of who won.

The mechanism the tradition supplies

Underneath both debates sits the account Marx gives in the final part of the first volume of Capital. His argument is that the capitalist mode requires a prior historical process producing a class with no means of subsistence, and that this class was produced by expropriation: the enclosure of common lands, the dissolution of household retainers, the clearing of estates, and the legal machinery that criminalised the resulting population’s alternatives. The point of the argument is polemical as well as analytical, since it is directed against an account in which capital originates in the thrift of the industrious, and it insists that the origin was coercive and political rather than economic.

The mechanism is important for the concept because it identifies what a transition actually requires. It is not enough for a new technique to appear or for markets to expand. A new mode requires that the conditions of the old extraction mechanism be destroyed and that the conditions of the new one be created, which in this case means the destruction of the peasant’s access to means of subsistence. Transitions are therefore violent as a matter of structure rather than of accident, and the treatment of primitive accumulation develops that argument and the evidence for it.

What does a transition actually require?

Through the destruction of the conditions that sustained the old extraction mechanism and the creation of conditions for a new one. In the case Marx analyses, that meant separating producers from their means of subsistence so they would have to sell labour power. Technical change and market growth are not sufficient on their own.

The misreadings, named and corrected

Five errors dominate general-audience writing on this concept, and each has a specific correction a reader can apply on sight.

The first is that a mode of production is a synonym for an economic system, in the sense in which that phrase is used in journalism and in introductory economics. It is not. An economic system in that usage is characterised by its coordination mechanism, meaning market, plan, or tradition, and by its distribution of ownership. A mode of production is characterised by its extraction mechanism. The two classifications cut across each other: two economies coordinated by markets can have different modes of production if one rests on wage labour and the other on slavery, and two economies with the same extraction relation can be coordinated very differently.

The second is that a mode of production is defined by its technology. This is the technological reading and it fails on the evidence, since the same techniques have been operated under slave, servile, wage, and state-directed relations. The steam engine is compatible with all four. It also fails on the texts, since the criterion Marx supplies concerns the form in which surplus labour is pumped out of the producers, not the instruments used.

The third is that each society has exactly one mode. Almost no society ever has. The correct formulation involves a social formation containing several modes with one dominant, and any classification that assigns a country to a single row without saying what else is present has thrown away most of the information.

The fourth is that the sequence of modes is a compulsory ladder every society climbs. Marx names four epochs once, does not present them as universal, and explicitly refused in his last decade to have his account of Western European development converted into a general theory of a path every people must tread. The rigid ladder is a later codification and the complete guide to historical materialism documents its origin.

The fifth is that the Asiatic mode is a settled component of the theory. It is not. It is a sketch in exploratory writing, it was removed from official doctrine by decision, its most influential revival was an explicitly political work whose empirical claims have been heavily criticised, and most historians of the relevant regions do not use it. Presenting it as a standard fifth category alongside the others misrepresents both its status and its history.

A sixth error is worth adding because it does specific damage in student writing: the definition of capitalism by markets and trade, corrected above. If a reader takes only one correction from this article, that should be it, because it is the one that produces wrong answers in every subsequent application.

The strongest objection to the concept

Stated at full strength, in the terms its own proponents use, the objection is this.

The concept of a mode of production appears to generate determinate classifications, but in practice no awkward case is ever allowed to falsify a classification, because three escape routes are always available. If a society does not fit a row, it is a transitional formation. If two arrangements coexist, it is an articulation of modes. If the fit is poor in some respect, it is a hybrid social formation with a dominant mode. Between them these three devices absorb every anomaly, which means no observation could ever count against a classification, which means the classifications are not empirical findings at all. The concept survives not because it is well supported but because it cannot fail.

The objection has a second edge that is sharper still. The number of modes is not fixed and the tradition has generated new ones whenever the existing ones fitted badly: a colonial mode, a petty commodity mode, a tributary mode, a plantation mode. A classificatory scheme that adds a category whenever a case resists is doing the work of description under the appearance of theory, and the addition of categories to preserve a scheme is the classic signature of a degenerating research programme.

Three things can be said in reply, and the first is a concession.

Undisciplined use of articulation does exactly what the objection describes, and the accusation is not unfair to a substantial body of writing. Where a case is labelled an articulation without any specification of how the modes interact, what each supplies to the other, and what would count as evidence for the alleged relation, the label is doing no work and the objection lands cleanly. Defenders of the concept should say so, because the practice has been common enough to discredit the concept by association.

The reply that matters is the surplus-extraction test itself, used as a discipline rather than as a definition. Run properly, the test produces determinate and contestable answers. Asked whether a medieval Italian merchant city was capitalist, it returns no, on the specific ground that the producers were not separated from their means of production and that merchant profit derives from price differences rather than from the purchase of labour power. Asked whether a colonial mining economy paying below-reproduction wages is a simple capitalist formation, it returns no, on the specific ground that the reproduction of the workforce depends on a non-capitalist agrarian sector, and that claim is checkable against wage levels, household budgets, and migration patterns. Asked whether a state socialist economy contained a class relation, it returns a question about effective control over the surplus that historical research can address. Those are not unfalsifiable results. They are claims about who did what to whom, and they can be wrong.

The second reply concerns the proliferation of categories. Adding a category is legitimate where the new category is specified by the same criterion as the others and generates claims of its own, and illegitimate where it is a label attached to a residue. A tributary mode specified by extraction through taxation is a legitimate addition on that standard, since it names a mechanism and predicts specific reproduction conditions. A colonial mode specified by nothing except that the society in question was a colony is not, since it names a political status rather than an extraction relation. That standard is available and it does real filtering work, which is what an objection of this kind requires as an answer.

What survives on the critical side is a warning rather than a refutation. The concept is a good analytical instrument and a bad taxonomic one. Used to ask how surplus moves in a given society, it is productive. Used to file societies into boxes, it produces exactly the endless qualification the objection describes, and the tradition has often used it the second way.

Is the concept of a mode of production falsifiable?

The classifications it produces are, provided the surplus-extraction test is actually run. The claim that a society rests on wage labour, or that one sector’s wages depend on subsistence production elsewhere, can be checked against evidence. What is not falsifiable is a classification supported only by labelling an awkward case as transitional or hybrid.

Applying the concept: three worked cases

The claim of this article is that the concept is something you do rather than something you memorise, and a claim of that kind has to be demonstrated.

A commercial city with great merchant wealth

Take a medieval trading city with substantial banking houses, long-distance commerce, and fortunes exceeding those of many territorial rulers. Step one: identify the direct producers. They are artisans in workshops, journeymen and apprentices under masters, and agricultural producers in the surrounding countryside. Step two: how do they gain access to the means of production? The masters own their workshops and tools; the journeymen work under a master with an expectation, however attenuated, of eventually holding a workshop themselves; the rural producers hold land under customary or servile tenure. Step three: how is surplus extracted? From the rural producers by rent and dues, requiring extra-economic compulsion. From the artisans, largely not at all in the relevant sense, since they own their means of production and sell their own products. The merchant’s gain comes from buying and selling, which is appropriation of a share of surpluses produced elsewhere under other relations. Step four: what sustains the arrangement? Guild regulation, customary tenure, and the political and legal apparatus securing both.

Conclusion: the dominant relations are feudal in the countryside and petty commodity production in the city, with a powerful merchant stratum whose wealth is commercial rather than capitalist. The city is not capitalist despite the money, and the analysis says precisely why.

A colonial mining economy

Take an economy in which a mining sector employs migrant workers at wages below the cost of reproducing a household, while those workers’ families continue subsistence cultivation in rural districts. Step one: the direct producers are the miners and the rural cultivators, who are frequently the same households at different points in a life cycle. Step two: access to the means of production is split, with the miners separated from any means of subsistence during their employment and the rural households retaining land. Step three: surplus is extracted from the miners through the wage relation, and the wage is set below reproduction cost because part of the cost of reproducing the workforce is borne by the rural sector. Step four: the arrangement requires that the rural sector persist and that movement between the two be regulated, which explains why the legal control of migration and the maintenance of rural land tenure are political questions in such economies rather than incidental ones.

Conclusion: a capitalist mode dominant, articulated with a subordinate agrarian sector on which it depends for the reproduction of its labour force. Note what the analysis produces: a specific and checkable claim about wage levels relative to household budgets, and an explanation of why particular legal arrangements exist. That is the concept doing work.

A contemporary economy with extensive self-employment

Take an economy in which a substantial share of workers are formally independent contractors owning their principal instrument of work while depending entirely on intermediaries for access to customers and for the terms on which they work. Step one: the direct producers are the contractors. Step two: they own an instrument of production, which on the classical criterion means they are not separated from the means of production. Step three: the surplus question is where the analysis becomes interesting, because the intermediary’s revenue derives from a share of each transaction rather than from the purchase of labour power for a period, which is formally closer to merchant appropriation than to the wage relation. Step four: the arrangement requires that the contractors lack independent access to customers, which is a form of dependence not captured by the classical criterion of separation from the means of production.

Conclusion: a genuinely hard case that the classical specification does not resolve cleanly, which is the honest answer. What the analysis produces is not a label but a specification of what is unusual: dependence secured through control of market access rather than through ownership of instruments. A reader who reaches that conclusion has used the concept correctly, and a reader who confidently assigns the case to a row has not.

Where the concept is used in scholarship now

The vocabulary is used less than the analysis is, and a reader tracking the influence needs to know where to look.

In economic and social history the concept remains a working tool, particularly in agrarian history, where the questions of tenure, rent, and the terms of peasant obligation are exactly the questions the specification asks. The historiography of European agrarian change, of plantation economies in the Atlantic world, and of agrarian relations in South Asia are all fields where the arguments are conducted in these terms, frequently by historians who would not describe their politics as Marxist but who find the questions the sharpest available.

In anthropology the concept had its largest impact through the articulation literature, applied to colonial and postcolonial economies from the 1960s onwards. That literature has receded, partly because its structuralist vocabulary aged badly and partly because the specific claims about how sectors depended on each other proved difficult to test. What survived is the general insight that non-capitalist arrangements within a capitalist economy are often functional to it rather than residual, which is now a standard assumption in the study of informal economies and of unpaid care work.

In development economics and economic history the concept operates under other names. The analysis of agrarian structures and their effects on growth, of tenancy arrangements and their incentive properties, and of the relationship between labour coercion and economic development, all ask questions descended from this framework, and some of the resulting literature engages with it explicitly.

Where the concept has lost ground is as a scheme of universal historical stages. Almost nobody now defends a fixed sequence of modes through which societies pass, and the disappearance of that claim is the largest change in the concept’s standing. What remains is the analytical unit and the test, which is a substantial thing to remain but is less than the tradition originally proposed.

What mode of production are we in now?

The dominant mode across most of the world economy is capitalist, identified by generalised commodity production resting on wage labour rather than by the presence of markets or of private wealth. Every actual society also contains subordinate arrangements, including household production, self-employment, and unpaid care work, on which the dominant mode depends.

For researchers and historians

This concept is the working tool of Marxist historiography, and four sourcing and methodological cautions prevent most of the errors that appear in professional writing.

Get the status of the texts right. The material on pre-capitalist forms is notebook writing from the late 1850s that Marx did not prepare for publication, and citing it as a settled position overstates its authority while dismissing it as marginalia understates its importance. The passage that supplies the extraction criterion sits in the third volume, which Engels assembled from manuscripts, and a writer relying on it should say so. The earliest treatment of forms of ownership is in manuscripts published long after both authors died, with an editorial history that affects what can be cited as a unit. None of this makes the material unusable; all of it should be stated.

Do not assert a count of modes. Marx names four epochs in one place, uses different vocabulary elsewhere, and the tradition has proposed additional candidates on varying grounds. A sentence stating that there are five modes of production reproduces a codification with a specific political history rather than reporting a finding, and readers who know the material will treat it as a signal about the source.

Treat the Asiatic mode with the care its history requires. The category’s removal from Soviet doctrine happened through theoretical debates at the start of the 1930s and was a decision rather than a result, the most influential revival was a work with an explicit contemporary political argument, and the empirical claims attached to that revival have been extensively criticised. A writer covering this should distinguish Marx’s sketch, the category’s suppression, its polemical revival, and the underlying analytical question about state-based extraction, which survives independently of all three.

Check the transition literature before summarising it. The exchange of the 1950s and the debate opened in 1976 are both frequently reported in simplified forms that misstate the positions, and the second in particular is often described as though its verdict were settled when the responses raised substantial objections about chronology, about the characterisation of property relations, and about the role of the state. Describing the debate as unresolved in its particulars while noting what it established methodologically is both accurate and more useful than picking a winner. Researchers assembling material across a body of contested literature of this kind can save these guides and build a citation-linked reading list free on VaultBook, which keeps each claim attached to the source and the edition it came from rather than to a note written six months earlier.

A fifth caution applies to comparative work. The specification sheet is a set of ideal types constructed largely from European material, and applying it outside that context without asking whether the categories fit is how the concept acquired its worst reputation. The productive practice has been the opposite: taking a case that fits badly and asking what the misfit reveals, which is how the articulation literature, the debates on agrarian relations in South Asia, and the analysis of plantation economies all began.

For students and exam candidates

The specification sheet is the answer skeleton for almost any question on this topic. Asked to explain the concept, define it as the unity of forces and relations of production, state the surplus-extraction criterion, and then use one or two modes as worked illustrations rather than reciting the whole list. Asked to compare modes, run the same four columns across each: the characteristic relation, the extraction mechanism, the ruling stratum, and the principal contradiction. Answers structured that way read as analysis, while answers that describe each mode in narrative paragraphs read as summary and are marked accordingly.

The distinction that earns marks is between the mode of production and the means of production, because a large proportion of candidates confuse them and a marker will notice immediately. Writing one sentence that names the difference, that the means are the material resources while the mode is the whole system of production and appropriation, buys credit cheaply and signals precision.

The evaluative point that lifts an answer is the social formation distinction. Observing that no actual society is a pure instance of a single mode, that real societies contain several with one dominant, and that the concept of articulation was developed to describe the combinations, demonstrates that the candidate knows the concept as an analytical tool rather than as a list. Adding that the same device attracts the charge of unfalsifiability, and that the surplus-extraction test is what disciplines it, is a genuine evaluation rather than a recited criticism.

The traps are three. Defining capitalism by markets and trade, which is wrong and which a good marker will treat as a conceptual error rather than a slip. Asserting five compulsory stages, which reproduces a codification the primary text does not support. And describing the Asiatic mode as a settled category, which misrepresents a contested and politically loaded piece of the tradition. Avoiding all three is straightforward once a candidate knows they exist, and very few do.

For teachers and seminar leaders

The misconception that dominates a class on this topic is that capitalism means commerce, and it is unusually resistant because it matches everything students hear outside the classroom.

The exercise that fixes it takes about fifteen minutes. Describe a medieval trading city with major banking houses, credit networks spanning a continent, merchant fortunes financing wars, and workshops producing luxury goods for export. Ask the class whether it is capitalist. Most will say yes, and their reasons will be commerce, profit, banking, and wealth. Then ask a single follow-up: who owns the workshop the weaver works in, and what happens to the cloth they make? When the answer comes back that the weaver owns the workshop and sells the cloth, the class can see for themselves that no wage relation exists and that the merchant’s profit comes from trade rather than from employing anyone. The correction is theirs rather than the teacher’s, which is the only form in which it holds.

The follow-up that consolidates it is to run the same question on a case where the answer is capitalist and the commerce is invisible: a farm where the land is held by a tenant employing wage labourers and selling into a market, which most students will not identify as capitalist because it lacks factories and cities. The pair of cases teaches the criterion better than any definition, and it also introduces the agrarian origins question that the transition debates turn on.

For assessment, the productive task is to hand students an unfamiliar case and ask them to run the four questions of the test rather than to identify the mode. Marking the reasoning rather than the label rewards exactly the skill the concept exists to build, and it makes the answers far more interesting to read. For a politically mixed class, the concept has an advantage worth using: it is a piece of analytical machinery rather than a political position, and a session spent applying it to historical cases produces argument about evidence rather than about allegiance.

Feudalism as a concept, and the historians’ quarrel with it

No row in the specification sheet has attracted more scholarly hostility than the feudal one, and the reasons are worth knowing because they bear directly on how the concept should be used.

The difficulty is that the word feudalism carries two quite different meanings. In the tradition of legal and constitutional history it refers to a set of arrangements among a military elite: the grant of land in return for service, the relations of vassalage and homage, the fief as a legal institution. On that definition feudalism concerns the relations among lords rather than the relations between lords and peasants, and it describes a narrow band of European practice over a limited period. In Marxist usage it refers to a mode of production defined by the extraction of rent from peasants in possession of their means of subsistence, which is a much broader category applicable wherever that relation obtains.

Medieval historians have argued at length that the first term is itself a construct of later jurists and historians rather than a description of how contemporaries organised their affairs, and that its use flattens enormous regional and chronological variation. That critique is directed at the legal-historical usage rather than at the Marxist one, but it has been read as discrediting both, and a writer using the word should say which meaning they intend.

The Marxist usage has its own difficulties. Applied outside Europe, it has often functioned as a placeholder for any pre-capitalist agrarian society with a landed elite, which drains it of the specificity the extraction criterion is supposed to supply. Debates in South Asian historiography over whether the term describes Mughal or pre-colonial agrarian relations turned precisely on this: whether the extraction mechanism was the localised rent of the European specification or a centralised revenue system, which is a different mechanism requiring a different classification. Those debates were productive because they took the criterion seriously; where the term is applied without running the test, the criticism that it is a Eurocentric residual category has force.

There is also a popular usage that should be kept out of analytical writing entirely. The claim that a society is returning to feudalism, made about contemporary arrangements involving landlordism, rentier income, or platform intermediation, uses the word to mean something like extraction without production or dependence without mobility. Whatever merit those observations have, they are not claims about a mode of production in the technical sense, since none of the arrangements described involves peasants in possession of their means of subsistence subject to extra-economic compulsion. A writer who wants to make the substantive point should make it in the vocabulary of rent, of monopoly, and of dependence, which is precise, rather than in the vocabulary of feudalism, which is not.

Is feudalism coming back?

Not in the technical sense. The feudal mode is defined by extraction of rent from producers who hold their own means of subsistence, secured through extra-economic compulsion. Contemporary arrangements involving landlordism, rentier income, or platform intermediation do not meet that specification, whatever else can be said about them.

The concept outside Marxism

A reader assessing how much weight to give this framework should know that several of its components have counterparts in research produced by people with no relation to the tradition, arrived at independently.

The distinction between coerced and market-mediated labour, which is the heart of the extraction criterion, is a standard analytical category in the economic history of labour institutions, where the questions of why coercion is used in some settings and not others, and what conditions make it profitable, have generated a substantial formal and empirical literature. The findings there, concerning the relation between land abundance, labour scarcity, and the incidence of unfree labour, bear directly on the specification sheet and are usable evidence for or against particular classifications.

The analysis of tenancy arrangements in agrarian economics addresses the same object as the discussion of rent forms in the third volume: the different incentive properties of fixed rent, share tenancy, and wage employment, and the conditions under which each is chosen. The vocabulary is entirely different and the questions overlap almost completely.

The study of household production and unpaid care work as a condition of the wage economy, which arrived at conclusions the articulation literature had reached from another direction, is now a substantial field with its own measurement traditions. And the analysis of institutions that structure who captures the returns from production, which is a central question in comparative economic history, asks the concept’s question in different terms while resisting its answer about explanatory priority.

The point of noting this is not to claim the tradition’s priority, which would be both petty and hard to establish. It is that a reader who finds the vocabulary alienating can reach much of the analytical content by another route, and that a writer who wants to defend the concept has evidence available from sources that cannot be accused of partisanship.

Rent, profit, and the forms of surplus

The extraction criterion becomes considerably sharper once the forms the surplus takes are distinguished, and this is where the concept connects to the economic analysis that occupies most of Marx’s mature work.

Under servile relations the surplus takes the form of rent, and the third volume distinguishes three sub-forms with different properties. Labour rent, in which the producer works part of the week on the lord’s land, makes the division between necessary and surplus labour visible in time and space: the days are counted and the fields are separate. Rent in kind, in which a share of the produce is handed over, obscures the division slightly while leaving the producer in control of their own labour process. Money rent, which requires the producer to sell part of the output, draws the household into market relations and creates differentiation among peasants, since some will do better at it than others. The sequence is a change in form within a mode rather than a change of mode, which is exactly the sort of distinction the concept is for, and it is also the mechanism through which one mode prepared the conditions of another in the European case.

Under capitalist relations the surplus takes the form of surplus value and is distributed among industrial profit, interest, and ground rent, which is what the third volume is largely about. The analytical point that matters for the concept is that all three derive from the same source. Rent under capitalism is not extraction from peasants; it is a claim on surplus value produced elsewhere, secured by monopoly of a natural resource. That is a different relation from feudal rent despite the shared word, and confusing the two produces the mistaken belief that landlordism as such constitutes a feudal remnant.

Under tributary arrangements the surplus takes the form of tax, and the defining feature is that the extraction is performed by a state apparatus rather than by a private proprietor, so that tax and rent are not distinguishable. This is the mechanism the Asiatic category was constructed to name, and it is the reason the underlying question survives that category’s difficulties, since formations where taxation is the primary extraction mechanism are genuinely not well described by either the feudal or the capitalist specification.

Distinguishing these forms is what allows an analysis to say something determinate about a mixed case. An economy in which some producers pay rent to landlords, some pay tax to a state, and some sell labour power for a wage is not confusing once the forms are separated; it is a social formation with three extraction mechanisms operating, and the analytical question is which is dominant and how the others are subordinated to it.

The vocabulary: mode, formation, system, epoch

Four words are used interchangeably in careless writing and mean different things, and a reader who holds the distinctions will read the literature more accurately.

A mode of production is an abstraction: a specific combination of productive forces and relations of production, with its own extraction mechanism and conditions of reproduction. It is a type, not a place or a period.

A social formation is a concrete society at a given time, containing one or more modes, with one dominant. It is the unit that has a name on a map and a date.

An economic system, in the ordinary usage borrowed from mainstream economics, is characterised by its coordination mechanism and its ownership pattern. It is a classification along a different axis and it cross-cuts the modes, which is why market socialism and planned capitalism are not contradictions in terms even though they are unusual.

An epoch is a period of history in which a given mode was dominant across a region. The 1859 preface uses this vocabulary and it is the loosest of the four, since the boundaries of an epoch depend on the region chosen and on the threshold for dominance, neither of which the concept specifies.

The practical consequence is that several familiar disputes dissolve once the vocabulary is sorted. Whether a particular country was feudal in a particular century is a question about a social formation and admits of a mixed answer. Whether feudalism is a coherent mode of production is a question about an abstraction and admits of a yes or no. Whether the feudal epoch had ended by a given date is a question about a region and a threshold. Writers who move between the three without noticing produce arguments that cannot be resolved because the participants are answering different questions.

What the concept cannot do

An honest account of an analytical tool includes its limits, and this one has three that matter.

It does not supply a theory of the state. The specification sheet identifies who extracts and by what mechanism, and it says nothing systematic about the form of political organisation, which is why the same mode has been compatible with monarchies, republics, colonial administrations, and dictatorships. Attempts to derive the political form from the mode have been among the tradition’s least successful projects, and the complete guide to the Marxist theory of class and the state articles in this series take up the question the concept leaves open.

It does not explain the pace or the direction of technical change on its own. The concept identifies the relations within which technical change occurs and can explain why particular techniques are adopted or resisted, which is a real contribution supported by evidence. It does not predict what will be invented or when, and accounts that treat productive development as an autonomous process which the relations merely accommodate are doing something the concept does not license.

It does not adjudicate questions about justice. The specification of an extraction mechanism is a description of how a surplus moves, and whether that movement is unjust requires a normative argument the concept does not contain. Marx’s own position on whether he was making a moral claim is genuinely difficult, and treating the identification of exploitation as though it settled an ethical question confuses a technical measure with a verdict.

The reproduction conditions of each mode

The fourth step of the test asks what has to keep happening for an arrangement to persist, and it repays being worked out for each mode, because reproduction conditions are where the concept produces its most checkable predictions.

A slave economy must replace its producers. Whether replacement occurs through organised reproduction within the enslaved population or through continual purchase and capture is a structural question with enormous consequences, since the second creates a permanent dependence on external supply and ties the productive unit to military and commercial arrangements far beyond it. It also makes the closing of a supply route an existential rather than a commercial matter, which is why the politics of such closures is so bitter wherever it occurs. And because the producer has no interest in the product, the mode requires a supervisory apparatus whose cost rises with the intensity of the work regime, which sets a ceiling on how far the extraction can be pushed.

A servile agrarian mode must prevent flight. A peasant in possession of their own means of subsistence has an obvious alternative to paying rent, which is to leave, and every feature of the legal apparatus restricting movement is a response to that fact. It must also maintain the fusion of judicial and economic authority, since a lord who cannot enforce is not a lord, which is why the fragmentation of jurisdiction is structural rather than incidental. And it must reproduce the peasant household as a viable unit, since a peasantry ground below subsistence produces no rent at all, which places a floor under the extraction that the extractors themselves have an interest in respecting.

The capitalist mode must maintain the separation of producers from the means of subsistence. This is the reproduction condition that explains the most and is noticed the least. It makes the availability of independent subsistence a political question rather than a private matter, which is why the enclosure of common resources, the terms of access to land, and the treatment of alternatives to wage employment are contested wherever the mode is establishing itself. It must also reproduce labour power itself, generationally and daily, which is work performed largely outside the wage relation and largely without payment, and the recognition that a mode depends on activity it does not pay for is one of the more consequential developments in the analysis of this row. And it must find outlets for accumulated surplus, which is what connects the mode’s reproduction to its expansionary character.

A tributary arrangement must maintain the state apparatus that performs the extraction, which means the revenue must cover the cost of collecting it plus the cost of the coercion that makes collection possible. That constraint explains a great deal about the political history of such formations, including the recurring pattern in which intermediaries interposed between the centre and the villages gradually convert a revenue function into a proprietary claim, which changes the mode from within without any change in formal doctrine.

Setting the conditions out this way makes the concept predictive in a limited but real sense. Each mode implies a set of political questions that will be contested wherever it operates, and the presence or absence of those contests is evidence about the classification. A society in which the legal control of movement is a central political issue is telling you something about how its surplus moves.

Unfree labour and the boundary of the capitalist mode

One boundary question recurs often enough to deserve separate treatment, because it is where the classical specification is most often tested and where careless application does the most damage.

The specification says that capitalism rests on wage labour, meaning producers who are legally free to sell their capacity to work and compelled by circumstance rather than by law to do so. Yet capitalist economies have made extensive use of labour that is not free in that sense: enslaved workers on plantations producing for world markets, indentured workers under penal contracts, workers bound by debt, and workers whose legal status restricts their ability to leave an employer. The question is whether such arrangements fall inside or outside the mode.

Three positions are defended in the literature. The first holds that unfree labour is by definition non-capitalist, so plantation slavery constituted a distinct mode articulated with capitalist commerce rather than a part of it. The second holds that what matters is the circuit of capital rather than the legal status of the producer, so that production organised by capital for a market, with surplus realised as profit, is capitalist regardless of whether the worker is bought or hired. The third holds that unfree labour is a variant form within the mode, adopted where the conditions producing a compelled free workforce are absent, particularly where land is abundant relative to labour and workers would otherwise take up independent cultivation.

The third position has the advantage of connecting to an independent and substantial body of evidence, since the relation between land abundance, labour scarcity, and the incidence of coerced labour has been studied extensively by economic historians with no interest in this framework. Their finding, that coercion tends to appear where free workers would have attractive alternatives, is precisely what the reproduction analysis above would predict, and it is usable evidence for the classification.

What the case demonstrates for a reader learning the concept is that the boundary questions are the productive ones. A framework whose categories fit every case cleanly would be describing rather than analysing. The value of the specification is that it tells you exactly which feature of an awkward case is awkward, which is where the research begins.

There is a related lesson about how the concept should be taught and written about. The rows of the specification sheet are worth knowing, but a reader who can only place clear cases has learned the easy half. The skill worth having is the ability to say, of a case that resists, exactly which of the four questions returns an unusual answer and what would have to be established to settle it. That is a transferable analytical habit and it is the reason this concept has outlasted most of the theoretical apparatus built around it.

Petty commodity production, the arrangement nobody classifies

One arrangement appears in almost every social formation, is essential to several of the arguments above, and has no settled place in the specification sheet. It deserves a section because a reader will meet it constantly and will find most summaries silent on it.

Petty commodity production names the case in which a producer owns their means of production, works them personally or with family labour, and sells the product on a market. The medieval urban artisan is the classic instance. The independent smallholder selling a cash crop is another. The self-employed tradesperson is a third.

The difficulty is that the surplus-extraction test returns no answer, because there is no extraction. The producer owns the means of production and keeps the product, so no surplus labour is transferred to anyone. On the criterion that identifies a mode, this is not a mode of production in the same sense as the others: it has no class relation at its core and therefore no internal contradiction of the kind the specification sheet records. Some writers accordingly treat it as a form rather than a mode, present in many social formations without ever being dominant.

It nonetheless does essential analytical work in three places. It is what makes the medieval trading city non-capitalist despite its commerce, since the artisans producing the exported goods stood in this relation rather than in a wage relation. It is the arrangement that the establishment of capitalist relations must dissolve, since a population that owns its means of production has no reason to sell labour power, which is why the account of primitive accumulation is an account of dispossession. And it is where a great deal of subordinated production sits in developed economies, in the form of self-employment, family businesses, and contracting, frequently under conditions where formal independence coexists with real dependence on a single purchaser.

That last case is where the concept is currently under most strain, and the honest position is that the classical specification does not resolve it. A producer who owns their principal instrument of work but depends entirely on an intermediary for access to customers, and who has no influence over the terms of that access, is neither separated from the means of production in the classical sense nor independent in any meaningful one. The dependence is real and it is secured through control of market access rather than through ownership of instruments, which is a mechanism the specification sheet does not have a column for.

Two responses are available and both are being pursued. One is to argue that the formal ownership is a legal appearance and that the substance is a wage relation, which requires showing that the intermediary effectively directs the labour process. The other is to argue that a distinct form of subordination exists here, requiring its own specification, which requires naming the mechanism and its reproduction conditions rather than simply noting that the case is difficult. Which response succeeds is an open question, and the general lesson is the one this article has argued throughout: the concept is worth having because it tells you precisely what is unresolved, which a description could not do.

How the concept generates a class structure

The relation between this concept and the theory of class is direct and is worth stating explicitly, because it explains why the mode is the primary unit and class the derived one rather than the reverse.

Classes, in this framework, are the positions that a mode of production creates. Specify the extraction mechanism and the classes follow: an arrangement in which rent is taken from producers holding their own subsistence generates a landholding stratum with judicial powers and a peasantry with a claim on the land it works, while an arrangement resting on the purchase of labour power generates owners of capital and sellers of labour power. The classes are not groups that happen to occupy the mode; they are the mode’s structure described from the standpoint of the people in it.

Two consequences follow that matter for anyone using either concept. Because a social formation typically contains several modes, it typically contains more class positions than a two-class model allows, and the messy class maps of Marx’s own historical writing are a consequence of the analysis rather than a departure from it. And because the classes are generated by the extraction relation, a change in that relation changes the class structure whether or not anyone’s income or status changes, which is why the framework cuts across stratification schemes built on occupation, education, or consumption.

The concept also sets a limit on what class analysis can do. It identifies positions, not the political behaviour of the people occupying them, and nothing in the specification of a mode determines whether the occupants of a position will recognise their situation, organise around it, or act. That gap is where the theory of class consciousness lives, and the complete guide to class in Marxist theory takes up the structure this concept supplies and works out what can and cannot be inferred from it.

What is settled and what remains disputed

Several things about this concept are not in serious dispute among people who have read the primary material, and stating them is useful because general-audience writing suggests otherwise.

It is settled that the mode of production is not the same as the means of production; that the criterion identifying a mode is the form in which surplus labour is extracted from the direct producers; that capitalism is defined by generalised commodity production resting on wage labour rather than by markets, trade, or private wealth; that merchant wealth does not by itself constitute capitalist relations; that no actual society is a pure instance of a single mode and that the concept of the social formation exists to handle this; that Marx names four epochs in the canonical passage and that the compulsory five-stage sequence is a later codification; that the Asiatic mode was removed from Soviet doctrine by decision rather than by evidence; and that there is no agreed count of modes in the tradition. Each of these is contradicted routinely in summaries, and a reader who holds all eight can identify an unreliable source in a paragraph.

It remains disputed whether the productive forces or the class relations carry explanatory weight in transitions, which is the axis the transition debates turn on; whether the Asiatic category names anything useful once its political history is stripped away, and whether a tributary category should replace it; whether Atlantic plantation slavery constituted an ancient mode surviving into a later period, a distinct mode, or capitalist production using unfree labour; whether the state socialist economies contained a class relation and, if so, what the extraction mechanism was; whether articulation is a genuine analytical device or a way of absorbing anomalies; and whether the concept can be applied outside the European material from which its ideal types were constructed without doing violence to the cases. These are disputes conducted inside the tradition, in its own literature, and the sharpest positions on each side are frequently held by people who share a general framework and disagree sharply about how to use it.

A reader who has reached this point can take an unfamiliar society, identify its direct producers, say what they must do to gain access to the means of production, name the mechanism by which their surplus labour is transferred, state what has to keep happening for the arrangement to persist, and say which of these arrangements is dominant where several are present. Those five moves are the concept, and they are what this guide was built to install. The natural next step is the treatment of primitive accumulation, which supplies the mechanism by which the capitalist mode came into being, followed by the complete guide to class in Marxist theory, which takes the relations identified here and works out the class structures they generate.

Frequently Asked Questions

Q: What is a mode of production in simple terms?

It is the whole way a society organises the production of its material life, combining the technical means it uses with the social arrangements determining who works for whom and how the surplus is transferred. What distinguishes one mode from another is that last element: the mechanism by which surplus labour is extracted from the people who perform it. Under serfdom it is rent backed by legal and physical compulsion. Under slavery it is direct appropriation secured by ownership of the producer. Under capitalism it is surplus value contained within a working day that has been bought on a market. The level of technology and the presence of markets do not identify a mode, because societies with comparable techniques and extensive trade have organised extraction in entirely different ways.

Q: What are the forces of production and the relations of production?

The productive forces are the means by which a society produces: instruments and machinery, raw materials and land in use, technical knowledge, and the labour power of the producers including their skills and their capacity for cooperative work. That last element is often dropped from summaries and should not be, since Marx treats the organisation of labour as a productive force in its own right. The relations of production are the social relations governing effective control over those forces and over the labour that sets them in motion, which in class societies means the relations determining who performs surplus labour and who lives on it. The two together constitute a mode of production, and the claim that they can come into conflict is the engine of the theory of historical change.

Q: How many modes of production did Marx identify?

There is no agreed count and nobody should assert one. In the canonical passage of the preface to A Contribution to the Critique of Political Economy he names four epochs: Asiatic, ancient, feudal, and modern bourgeois. Elsewhere he uses different vocabulary, discusses communal forms that do not appear in that list, and treats several arrangements without classifying them. The later tradition has proposed additional candidates including Germanic and Slavonic forms, a tributary mode, a colonial mode, and a petty commodity mode, on varying grounds. A source stating confidently that there are five modes is reporting the codification that hardened in Soviet doctrine rather than a finding in the primary material, and the five-item version does not match the four-item list Marx actually wrote.

Q: What is the Asiatic mode of production and why is it controversial?

It names a form in which village communities hold land collectively, the individual is a proprietor only as a member of the community, and a state or its representative stands as ultimate proprietor above the villages, so that tax and rent merge into a single payment to a central apparatus. It appears in the preface of 1859 and is developed exploratorily in the notebooks of the late 1850s. It is controversial for three reasons. It slides easily into a doctrine of unchanging societies with an obvious colonial application, and Marx’s journalism on India contains passages of that kind. It was removed from Soviet orthodoxy after theoretical debates at the start of the 1930s, by decision rather than by evidence, partly because it described the Soviet situation uncomfortably well. And its most influential revival, published in 1957, was an explicitly anti-communist work whose empirical claims about hydraulic society have been heavily criticised.

Q: What is the difference between a mode of production and a social formation?

A mode of production is an abstraction: a type of arrangement defined by its extraction mechanism and its conditions of reproduction. A social formation is a real society at a particular time, which almost always contains several modes with one of them dominant. The distinction matters because readers who expect a country to match a single mode cleanly will conclude that the concept fails, when what has failed is the expectation. Agrarian England during its transformation contained capitalist tenant farming, surviving customary tenures, wage labour, and household production simultaneously. Colonial economies contained plantation slavery, peasant subsistence, and metropolitan capital at once. The analytical question is which mode is dominant and how the others are subordinated to it, not which single row a country belongs in.

Q: Can two modes of production exist in the same society?

Yes, and nearly every historical society has contained several. The theoretical apparatus for this is articulation, developed principally in French structuralist Marxism from the 1960s and applied extensively to colonial and postcolonial economies. The argument is that modes do not merely coexist but are combined, with the dominant one reproducing itself partly through the subordinate ones. A mining economy paying wages below household reproduction cost, on the assumption that rural subsistence agriculture makes up the difference, is the standard case: the capitalist sector depends on the persistence of a non-capitalist sector rather than simply displacing it. The same structure operates in industrialised economies through unpaid domestic labour reproducing the workforce. Used with discipline this generates checkable claims; used loosely it becomes a way of absorbing any awkward case.

Q: How does one mode of production change into another?

Through the destruction of the conditions that sustained the old extraction mechanism and the creation of the conditions for a new one. Marx’s account of the emergence of the capitalist mode, in the final part of the first volume of Capital, argues that it required a prior process producing a population with no means of subsistence, accomplished through the enclosure of common lands, the dissolution of retinues, and the legal machinery that criminalised the alternatives available to the dispossessed. The point is that technical change and market growth are not sufficient. A transition requires that producers lose whatever secured their old position, which is why transitions are coercive as a matter of structure rather than of accident, and why the historiography of these periods is full of legislation about vagrancy, movement, and common rights.

Q: What mode of production came before capitalism?

In the European case, feudal relations, defined by the extraction of rent from peasants who possessed their own means of subsistence and were therefore subject to extra-economic compulsion. That answer carries three qualifications. It describes one regional trajectory and Marx explicitly refused to convert his account of Western European development into a universal sequence. It describes a dominant mode within social formations that also contained petty commodity production, urban artisan work, and in places slave labour. And the transition itself is the subject of a long dispute, running from the exchange between Dobb and Sweezy in the 1950s to the debate opened in 1976, over whether feudalism dissolved from its own internal conflict over rent or under the solvent action of expanding commerce.

Q: Is there a socialist mode of production?

The tradition has not settled this and the honest answer sets out the positions. The question is answered by running the same test: did the producers perform surplus labour appropriated by a stratum they did not control, and if so through what mechanism? The degenerated workers’ state analysis holds that property forms were post-capitalist while a bureaucratic stratum consumed part of the surplus without constituting a class. The state capitalist analysis holds that the planning apparatus functioned as a collective capital under international competitive pressure. The bureaucratic collectivist analysis holds that a genuinely new class formation existed, defined by control rather than ownership. The official position of the states themselves held that a socialist mode had been achieved. What the concept contributes is the discipline of asking who controlled the surplus rather than what the constitution said about ownership.

Q: What is the difference between mode of production and means of production?

The means of production are things: land, tools, machinery, buildings, raw materials, the physical resources with which production is carried on. The mode of production is the entire way a society organises production, including who controls those means, who works them, on what terms, and how the surplus above the producers’ subsistence is transferred. One is an inventory; the other is a description of a social system. The confusion is the commonest single error in student writing on this topic and it does real damage, because it produces the technological reading in which a change of machinery counts as a change of mode. A change of means is a technical change. A change of mode is a transformation of who works for whom.

Q: Does the concept apply outside Europe?

It was constructed largely from European material and applying it elsewhere without asking whether the categories fit is how it acquired its worst reputation. The productive practice has been the opposite: take a case that fits badly and ask what the misfit reveals. Debates in South Asian historiography over whether pre-colonial agrarian relations were feudal turned on exactly this, since a centralised revenue system is a different extraction mechanism from localised rent and requires a different classification. The articulation literature emerged from the same impulse applied to colonial economies. The general point is that the test travels even where the ideal types do not: asking who the direct producers are, how they gain access to the means of production, and how their surplus is taken is a question that can be put to any society.

Q: Is the concept still used by historians?

Yes, particularly in agrarian history, where the questions of tenure, rent, and the terms of peasant obligation are precisely what the specification asks about. The historiography of European agrarian change, of Atlantic plantation economies, and of agrarian relations in South Asia is conducted substantially in these terms, often by historians who would not describe their politics as Marxist but who find the questions the sharpest available. What has largely disappeared is the use of the concept as a scheme of universal stages through which societies pass. What remains is the analytical unit and the extraction test, which is a substantial survival though it is less than the tradition originally proposed.

Q: Was the Soviet Union a distinct mode of production?

That is exactly the disputed question and it is worth seeing why it is hard. Formal ownership of the means of production was held by the state, which means the classical criterion of private ownership does not identify a capitalist relation. But the concept’s own discipline says that formal ownership is not the criterion; effective control over the surplus is. If a planning apparatus decided the rate of investment, the level of consumption, and the allocation of labour without accountability to the producers, then a surplus was being appropriated by a stratum, and the question becomes what kind of relation that was. Answering it requires historical evidence about how decisions were made and who benefited, which is why the argument has been conducted through economic history rather than through definitions.

Q: How should a student structure an answer on this topic?

Define the concept as the unity of productive forces and relations of production, state the extraction criterion immediately, and then use one or two modes as worked illustrations rather than reciting the whole list. Run the same columns across each mode you discuss: the characteristic relation, the extraction mechanism, the ruling stratum, and the principal contradiction. Include one sentence distinguishing the mode of production from the means of production, since a large proportion of candidates confuse them. Lift the answer with the social formation point, that no real society is a pure instance and that articulation was developed to describe the combinations, and add that the same device attracts the charge of unfalsifiability, which the extraction test is what disciplines.

Q: What should a teacher watch out for when covering this?

The belief that capitalism means commerce, which every class arrives with because it is what the surrounding culture supplies. The exercise that fixes it is a medieval trading city with banking houses and merchant fortunes, followed by the question of who owns the workshop the weaver works in and what happens to the cloth. When the class establishes for themselves that the weaver owns the workshop and sells the product, they can see that no wage relation exists. The consolidating case is the reverse: a tenant farm employing wage labourers and selling into a market, which students will not identify as capitalist because it lacks factories. The pair teaches the criterion better than any definition and introduces the agrarian origins question at the same time.

Q: Which edition or text should a researcher rely on?

For the definition and the epochs, the preface to A Contribution to the Critique of Political Economy. For the extraction criterion, the discussion of rent in the third volume of Capital, cited with an explicit note that the volume was assembled posthumously by Engels from manuscripts. For pre-capitalist forms, the relevant section of the notebooks of the late 1850s, cited as exploratory material rather than as a settled position. For the transition mechanism, the final part of the first volume, which is the only one of these Marx himself saw through the press. Stating the status of each text is not pedantry in this field; it is the difference between a citation that survives a specialist reader and one that does not.

Q: Does the concept explain why some societies industrialised and others did not?

It reframes the question rather than answering it, which is a smaller claim than the tradition once made and a more defensible one. The reframing is to ask what the prevailing extraction mechanism permitted and obstructed: whether it created pressure to raise productivity, whether it left producers with an interest in improvement, whether it concentrated resources in hands that could invest them, and whether it generated a population compelled to sell labour power. Those questions have been productive in comparative economic history. What the concept does not supply is a general law connecting a mode to a growth trajectory, and versions of the theory that claimed one have not survived the comparative evidence.

Q: Is the concept just a way of relabelling ordinary economic history?

The objection has force where the concept is used taxonomically and much less where it is used analytically. Filing societies into rows adds nothing that a good description would not supply. Asking who the direct producers are, what compels them to work for others, and what has to keep happening for the arrangement to reproduce, generates questions that a great deal of economic history does not ask systematically, particularly the last one. The most useful test of whether the concept is doing work in a given piece is whether it produces a claim that could be wrong. If the analysis yields only a label, the objection is correct in that instance. If it yields a statement about wage levels, migration controls, or the dependence of one sector on another, it is not.