Few words in the Marxist vocabulary earn their keep as unevenly as contradiction. In one paragraph it names the engine of historical change, the deep structure that explains why a stable looking order carries the seeds of its own transformation. In the next it names something far more modest, a difficulty, a friction, a policy that pulls against itself, a trend that cannot continue indefinitely. A reader meeting the word for the first time is usually handed a gloss that covers both uses and distinguishes neither: contradictions, the gloss runs, are opposing forces whose struggle drives development. That sentence is not false, exactly, but it is empty in a specific and damaging way. Opposing forces exist everywhere, in weather systems and football matches and committee meetings, and a definition that fits everything a word might mean tells the reader nothing about what the word excludes. Worse, the gloss invites the sharpest objection the concept faces. If two tendencies push in opposite directions, why call that a contradiction at all? A contradiction, in the only rigorous sense most readers know, is a statement that cannot be true, the assertion of a proposition together with its negation. Two tendencies in a labour market assert nothing. They simply pull. So the suspicion forms early and never quite leaves: that contradiction in Marxist writing is a metaphor wearing the uniform of a logical term, borrowing an air of rigour and profundity that its content has not earned.

This article takes that suspicion seriously, because the suspicion is partly right, and working out exactly where it is right is the fastest way to learn what the term genuinely means. The method throughout is the glossary method of this series: a precise definition with its boundary drawn, the wrong definitions named and corrected one by one, the near neighbours separated with the exact difference stated in each case, the textual sources checked, and the strongest hostile argument heard at full length before any reply is attempted. The reader who stays to the end should be able to do four things that the standard gloss leaves impossible. The reader should be able to distinguish a dialectical contradiction from a logical contradiction without hesitation or embarrassment. The reader should be able to name the four distinct things Marx and the tradition after him call contradictions, and to say which of the four deserves the name under a stated test. The reader should be able to apply that test, the constitution test, to a fresh case and reach a defensible verdict. And the reader should be able to state the deflationary objection, that everything the word does could be done by the word conflict, in a form its proponents would recognise as fair, together with the honest concession the objection extracts.

What contradiction means in Marxist method

Begin with the term as Marx actually uses it, before the schools systematised him. The German word is Widerspruch, and its ordinary meaning already contains a hint that the English loses: Widerspruch is a speaking against, a counter statement, but in German philosophical usage from Kant onward it also covers opposition in things and not only in sentences. Marx inherited the word from Hegel, for whom contradiction was not a defect of careless thinking but the structure of anything that develops: a thing is what it is only through its relation to what it is not, and the tension between a determination and its opposite is what moves a category, and in Hegel’s larger claim a reality, forward into a richer form. Marx took the word out of the philosophy of spirit and put it to work on factories, commodities, rents and wages, and that relocation changed its centre of gravity. Where Hegel’s contradictions unfold primarily among concepts, even when the concepts are meant to be the concepts of real institutions, Marx’s mature contradictions are located inside social relations and material processes: inside the commodity, inside the production process, inside the relation between those who own productive property and those who must sell their capacity to work.

Stated as carefully as the material allows, the definition this article will defend and test is the following. A contradiction, in the strong Marxist sense, is a relation between two poles within a single process or structure, where each pole exists only in and through its relation to the other, where the poles press against one another in opposed directions, and where the working out of that pressure is a source of change in the structure itself. Every clause of that definition does work, and each clause excludes something the loose usage admits. The poles must belong to one unity, which excludes collisions between independent things. Each pole must depend on the other for its very existence as the pole it is, which excludes pairs that merely happen to compete. The pressure must be internal to the structure’s own operation, which excludes accidents arriving from outside. And the pressure must matter for how the structure changes, which excludes decorative tensions that leave everything as it was. Hold that definition in reserve. Much of what follows consists of watching Marx’s own usage fall short of it in identifiable places, and of asking whether the shortfall condemns the concept or only condemns careless uses of it.

Is a dialectical contradiction a logical contradiction?

No. A logical contradiction asserts a proposition and its negation and cannot be true. A dialectical contradiction describes opposed tendencies or mutually constituting poles within one system, such as capital’s simultaneous need to raise productivity and to employ labour. Nothing is asserted and denied in the same breath.

That short answer is the single most valuable sentence in this article, and it is worth pausing over why confusion on the point is so persistent. Part of the reason is the word itself, which in English does its logical work first and its dialectical work second, so that a reader meeting the dialectical sense hears a logical claim being smuggled past. Part of the reason is translation history: Widerspruch can be rendered contradiction or opposition, and translators have not been consistent, so the same German sentence appears in one edition as a claim about contradiction and in another as a claim about opposition, with very different effects on an English reader’s expectations. And part of the reason, candidly, is that some dialectical writing has enjoyed the confusion. A writer who can hint that his opponents are bound by a merely formal logic, while his own method grasps a reality in which opposites interpenetrate, gains a rhetorical altitude that precise statement would cost him. The precision this article insists on removes that altitude deliberately. Whatever dialectical contradiction turns out to be, it is not a licence to assert impossibilities, and any version of the doctrine that requires such a licence has already lost.

The parent doctrine within which the term lives, its three laws and its claim to describe nature as well as society, is treated in the series guide to dialectical materialism complete guide, and the Hegelian source of the vocabulary, together with what Marx kept and what he discarded, is treated in the comparison of hegel dialectic vs marx dialectic. This article has the narrower job: one word, taken apart until its legitimate and illegitimate uses can be told apart at sight. Readers building a working vocabulary for the whole cluster will find the surrounding terms gathered in the marxism glossary key terms, and the discipline practised here, definition before verdict and test before slogan, is one worth carrying to every entry in it. A useful way to keep organised notes on distinctions of this kind, with each term’s test and worked example kept beside the passages that prompted them, is to save your notes and build a citation-linked reading list free on VaultBook, since contradiction is a word a reader will meet again in nearly every subsequent article of the series and will want the fourfold distinction ready to hand rather than reconstructed from memory.

The precise definition and where its boundary falls

A definition earns its keep at its boundary, in the cases it refuses as much as in the cases it accepts, so the boundary deserves a section of its own. Take the four clauses of the definition in turn and press each one against a case designed to strain it. The unity clause first. Two firms competing for the same customers press against one another, sometimes to the death of one of them, but each firm could in principle exist without the other, in another market or another century, and their collision is therefore an external opposition rather than an internal contradiction in the strong sense. Marxists do sometimes call competition a contradiction, and later sections will examine what justifies or fails to justify that extension; under the strict definition it fails the first clause, because the unity to which both firms belong, the market as a whole, is a looser thing than the unity of two poles inside one relation. Contrast the relation between a capitalist and a wage labourer. A capitalist in Marx’s technical sense, the owner of money who purchases labour power and sets it to work on owned means of production, cannot be specified at all without the labourer who sells that labour power, and the wage labourer, the person free in the famous double sense of being legally unbound and separated from any independent access to production, cannot be specified without the class that buys. Delete either pole and the other does not become lonely; it becomes unintelligible, a term without a referent. That is the unity clause satisfied at full strength.

The constitution clause deepens the same point, and it is the clause on which the article’s namable claim will be built. Many pairs are found together constantly without constituting one another. Rain and umbrellas are found together; neither brings the other into being as what it is, since rain fell before umbrellas and umbrellas hang in shops through droughts. Master and slave in Hegel’s celebrated analysis stand differently: the master’s status as master is nothing but the slave’s recognition and labour, and the slave’s selfhood is formed inside a relation the master’s dependence secretly structures, so that each term carries the other inside its own definition. Marx’s economic pairs repeatedly exhibit this structure. Value and use value in the commodity, concrete and abstract labour in the labour that produces it, capital and wage labour in the production relation: in each case the analysis proceeds by showing that the first term cannot be stated without smuggling in the second. The constitution clause is what separates contradiction from the vast family of mere correlations, rivalries and frictions, and it is a demanding clause. Most pairs casually baptised as contradictions in journalism and in a good deal of academic writing fail it within a minute of honest examination.

The internal pressure clause excludes a different family of impostors. A factory whose roof is struck by lightning faces a problem, even a catastrophe, but not a contradiction, because the pressure arrives from outside the relation that defines the factory’s operation. A factory that must cut its wage bill to survive competition while depending on the morale, skill and cooperation of the very workers whose wages it cuts faces pressure generated by its own normal functioning. The distinction matters because the Marxist claim for contradiction is a claim about endogenous change, about structures that transform themselves through tensions their own operation continually reproduces. A meteor can destroy a mode of production in a story; it cannot be one of its contradictions in the analysis. Finally, the change clause excludes static oppositions, pairs locked in a balance that nothing internal will ever disturb. A tug of war team matched perfectly against another, pulling forever with no movement, exhibits opposition without development, and while a physicist may study it with profit, the dialectical vocabulary gains nothing by annexing it. Marx’s contradictions are interesting because they are unstable in principle: the same relation that holds the poles together also drives at least one pole to act against the conditions of the relation itself. Capital must both employ labour and expel it, must both socialise production across vast cooperative networks and appropriate its product privately, and in each case the pole’s own logic, pursued consistently, strains the unity that contains it.

Does a dialectical claim assert a proposition and its negation at once?

It does not, and any formulation that sounds as though it does has confused levels of description. The claim concerns tendencies within a process, each tendency real and each opposed to the other, and tendencies are not propositions. A river can tend toward the sea and toward silting its own channel without any sentence being both affirmed and denied.

With the boundary drawn, the logical point can be made once and then left alone, which is where it belongs. Formal logic’s law of non contradiction states that a proposition and its negation cannot both be true in the same respect at the same time, and nothing in Marx’s economic work requires violating it. When Capital says that the commodity is both a use value and a value, no proposition has been asserted and denied: two predicates have been attached to one object, and predicates are cheap. When the analysis says that labour is both concrete and abstract, it distinguishes two aspects under which one activity is considered, the tailor’s specific motions and the expenditure of human effort in general, and aspects are not negations of one another. The appearance of logical scandal arises only when a summariser compresses an aspect distinction into a slogan of the form A is not A, and then either marvels at the profundity or condemns the nonsense. Both responses mistake the summary for the analysis. The serious philosophical dispute about contradiction in Marxism, which the Colletti section below takes up, is not about slogans at all. It is about whether the relations Marx describes are oppositions of the kind Kant called real, which nobody disputes can exist, or something stronger and stranger that Kant’s framework cannot house. That dispute is genuine, and it survives the removal of every slogan.

One more boundary marker belongs here, because it protects the definition from its friends. A contradiction, on the account being built, is a property of relations and processes, not a synonym for a mistake, a hypocrisy, or an embarrassment. A government that promises lower taxes and higher spending may be deceiving its voters or deceiving itself, and a newspaper may call the pair of promises contradictory in the loose journalistic sense, but nothing in the Marxist analysis is engaged until someone shows that the two commitments are poles of a single relation that constitutes them and presses them apart from within. Likewise a thinker caught believing incompatible things has committed a logical error, not discovered a dialectical contradiction in his own head, and the tradition’s occasional habit of dignifying its own inconsistencies as dialectical subtleties deserves the ridicule it attracts. The definition’s strictness is not pedantry. It is the only thing standing between a usable analytical term and a universal solvent that dissolves every distinction it touches.

The wrong definitions that circulate

Glossary work proceeds as much by correction as by construction, and contradiction attracts wrong definitions the way few terms do. The first and commonest is the dictionary substitution: contradiction as any opposition whatsoever, two forces, two interests, two ideas pulling in different directions. Under this definition the word adds nothing to opposition, tension, rivalry or conflict, and the deflationary critic discussed later in this article wins without a fight, since a synonym with grander manners is still only a synonym. The second wrong definition reaches for drama: contradiction as a flaw in a system, the crack that shows the system is failing, so that to identify a contradiction is already to pronounce a death sentence. Marx’s own usage resists this reading directly. Some of his contradictions persist for centuries as the normal working condition of a structure rather than as its terminal illness, and the wage relation, contradictory on any account from the moment it appears, has proved the most durable social form of the modern world rather than a brief convulsion on the way to somewhere else. A contradiction in the strict sense diagnoses a structure of tension, and structures of tension can be managed, displaced, subsidised, exported and endured, as well as resolved.

The third wrong definition is the moralising one: contradiction as hypocrisy, the gap between what an order professes and what it practises, between a society’s declared values of freedom and equality and its delivered realities of dependence and hierarchy. Marx certainly exploited such gaps rhetorically, and the young Marx of the 1844 manuscripts and the journalism can be quoted doing so with relish. But a gap between profession and practice is a feature of nearly every social order that has ever announced ideals, including orders Marx analysed as historically progressive in their time, and treating the gap itself as the contradiction confuses an ideological symptom with the relational structure that produces it. The mature analysis digs under the gap: the question is not why a society fails to honour its values but why its normal, honest, law abiding operation generates outcomes its values condemn, which is a question about poles and relations rather than about sincerity. The fourth wrong definition is the temporal one, contradiction as any situation that cannot last. Plenty of situations cannot last, for reasons of arithmetic, geology or fashion, without any poles constituting one another inside a unity. A candle burning down cannot last. Nothing in the candle’s structure resembles the capital and wage labour relation, and a vocabulary that calls both contradictions has traded its analytical edge for a mood.

The fifth wrong definition is the most scholarly and therefore the most seductive: contradiction as Hegel’s logical machinery transplanted intact, a stage in the self movement of a concept, so that to speak of the contradiction of the commodity is to speak of the category of value negating itself and passing over into money by conceptual necessity. There is genuine Hegelian architecture in the opening chapters of Capital, and the section on form and content below will give it its due. But Marx’s contradictions are not, in the main, movements among categories in a logic textbook. They are claims about what identifiable people, placed in identifiable relations, are structurally pressed to do, and about what those pressures do to the relations over time. Reading them as pure logic dematerialises them, and reading them as pure sociology loses the constitutive structure that makes them more than conflicts. The definition offered in this article is an attempt to hold both truths at once: the poles are real social positions and material tendencies, and their relation is internal in the strong sense that each is what it is only through the other.

Against all five wrong definitions stands a simple diagnostic habit worth acquiring early. Whenever the word contradiction appears, in Marx, in a textbook, or in a pamphlet, ask four questions of the passage. What are the two poles, named exactly? What single unity contains them? In what respect does each pole depend on the other for its existence or its character? And what change does their internal pressure tend to produce? A passage that answers all four has earned the word. A passage that answers two has probably found a conflict or a tension and dressed it in borrowed robes. A passage that answers none has used the word as an incantation, and the reader is entitled to translate it, mentally, into the word problem, and to notice how much of the argument’s apparent depth evaporates in the translation. This diagnostic is the constitution test in its working clothes, and the test receives its full statement after the four uses have been separated, since a test can only judge what has first been sorted.

Four uses hiding inside one word

The brief for this article, and the scholarship behind it, distinguish four things that Marx and his successors call contradictions, and the separation is the article’s central act of precision. They are not four species of one genus, neatly coordinate. They differ in logical type, in textual location, in the kind of evidence that supports them, and in their vulnerability to the critics. The first use names opposed tendencies within a single unity: one process generating two pressures that work against each other, as when capital must raise the productivity of labour while depending on living labour as the source of new value, so that its own success in cheapening labour power also threatens the basis of its expansion. The second use names conflict of interest between classes occupying complementary positions: capital and labour again, but now viewed as groups whose gains and losses are structurally linked, so that a rise in the share going to wages is, other things equal, a fall in the share going to profit. The third use names the gap between a form and its content, or between how a relation appears and what it is: the wage appears as payment for labour performed and conceals, on Marx’s account, the division of the working day into paid and unpaid portions, so that the form contradicts the content it presents. The fourth use names real opposition in Kant’s sense, forces acting in opposite directions, as when two tendencies in the falling rate of profit debate, the rising composition of capital and the cheapening of its elements, press against one another like weights on a balance, a configuration that involves no logical or constitutive puzzle at all.

Only the first use, and the second use where the positions are genuinely complementary and mutually constituting, satisfy the strict definition in full. The third use satisfies it in a qualified way that needs its own discussion, since forms and their contents are internally related in Marx’s analysis even though their opposition is one of presentation rather than of tendency. The fourth use fails the definition outright, and its presence in the canon is the factual basis of the most serious philosophical attack the concept has sustained. Yet all four travel under one word, in Marx’s own pages as well as in the literature after him, and the failure to mark which use is in play at a given moment is responsible for a large share of the confusion, and of the counterfeit profundity, surrounding the whole topic. A writer can establish a fourth use opposition, two forces in counterpoise, and then draw conclusions that only a first use contradiction would support, about necessary transformation and the self undoing of a structure, and the slide passes unnoticed because the word never changed. Precision here is not a scholastic luxury. It is the difference between an argument and a conjuring trick performed with a single noun.

How many distinct senses does the word carry in Marx’s writing?

Four, on the analysis defended in this article, and they are not interchangeable. Opposed tendencies inside one process, opposed interests between complementary positions, a form at odds with the content it presents, and forces in simple counterpoise exhaust the main uses, and each carries different implications for evidence, for resolution and for criticism.

The fourfold sorting also explains a puzzle that troubles careful readers of Marx: the unevenness of his rigour. In the great analyses, the contradiction of the commodity, the contradiction between socialised production and private appropriation, the contradiction inside the working day between capital’s drive to extend it and labour’s resistance grounded in the physical and moral limits of the worker’s life, the term is doing precise constitutive work, and the passages repay the four diagnostic questions with exact answers. In the journalism and the polemics, and in the more prophetic stretches of the programmatic writings, the term relaxes into emphasis, a way of saying that a situation is unstable, unjust, or pregnant with change, without the relational anatomy being supplied. Marx was a writer of genius working at speed across fifty years and several genres, not a terminology committee, and the honest scholarly position is that his usage is a field to be mapped rather than a doctrine to be harmonised. The tradition after him made the harmonising error systematic, elevating every occurrence into doctrine and thereby exposing the whole vocabulary to refutations aimed at its weakest instances. A glossary article serves the tradition better by grading its master’s uses than by gilding them, which is also, not incidentally, the treatment Marx himself applied to Ricardo and to Hegel.

The table that follows sets the four uses side by side in the form this series reserves for its findable artifacts, with a worked example, a logical verdict, a resolution profile and the specific objection that attaches to each use. It is designed to be used, not merely read: a reader confronting an unfamiliar claim about contradiction can locate the claim’s row, check whether the conclusions being drawn belong to that row’s logic, and detect the slide between rows that is the commonest defect in the literature. The four contradictions table is referred back to throughout the later sections, and the constitution test proposed in this article can be understood as a rule for deciding which rows of the table deserve the title term and which should be renamed without loss.

Use	Definition	Worked example from Marx	A contradiction in any logical sense?	What would count as its resolution	Objection specific to this use
Opposed tendencies within one unity	One process generates two pressures that undermine each other	Capital raises productivity to cheapen commodities while depending on living labour as the source of surplus value	No; tendencies are not propositions	Transformation of the process generating both pressures, or a durable counterweight that holds them in managed balance	The tendencies may be compatible in practice for long periods, so necessity of breakdown cannot be read off the structure
Conflict between complementary positions	Groups constituted by one relation pursue structurally opposed interests	Capital and wage labour over the length of the working day and the division of the product	No; opposed interests are not assertions	Abolition or fundamental renegotiation of the relation constituting both positions	Interest conflict is the common property of many theories and needs no dialectical vocabulary
Form against content	A social form presents a relation in a guise its own content belies	The wage form presents payment for all labour while the working day divides into paid and unpaid parts	No; appearance and essence can both be described without asserting a negation	A form adequate to its content, which for Marx means transforming the relation the form misrepresents	The appearance may be a functional necessity of the relation rather than a deception, which complicates the language of contradiction
Real opposition	Forces or magnitudes acting in opposite directions, cancelling or outweighing	The rising organic composition of capital against the cheapening of the elements of constant capital	No; and not a contradiction in the constitutive sense either	No resolution is required; the stronger force prevails or a resultant emerges	This is Kantian Realrepugnanz, ordinary counterpoise, and calling it contradiction conflates two distinct categories

Read down the fourth column of that table and the article’s recurring point stands in a row: not one of the four uses asserts a proposition together with its negation. Read down the last column and the article’s critical point stands beside it: each use has its own proper objection, and answering the objection proper to one use with resources drawn from another is the standard evasion the sorting is meant to prevent. The tendency use is vulnerable on necessity, the interest use is vulnerable on originality, the form use is vulnerable on the status of appearance, and the real opposition use is vulnerable on category. A defence of contradiction that does not specify which vulnerability it is answering has not yet begun.

Use one in slow motion: productivity against labour

The first use deserves a worked example followed step by step, because the difference between a genuinely analysed contradiction and an asserted one is visible only at walking pace. The example is the one embedded in the snippet answer and in the table: capital’s drive to raise the productivity of labour, set against its dependence on living labour as the source of surplus value. Step one is to fix the poles and the unity exactly. The unity is a single capitalist production process, and more generally the capitalist mode of production considered as a self reproducing system. The poles are not two groups of people. They are two tendencies generated by that unity’s own rules: the tendency to economise on labour, replacing workers with machinery and organisation wherever doing so cheapens the commodity, and the tendency to expand the mass of labour set in motion, since on Marx’s theory only living labour adds new value, and only the unpaid portion of its working day constitutes surplus value. Already the naming of the poles does analytical work that the word conflict cannot do, because the opposition has been located inside a process rather than between agents who might in principle walk away from one another.

Step two is to exhibit the pressure as internal. No external shock is required to generate it. Competition among capitals enforces the first tendency on pain of extinction: the firm that fails to cheapen its product loses its market to the firm that succeeds, a coercive law Marx compares to the force of gravity when a house falls about a person’s ears. The search for surplus value enforces the second tendency just as impersonally: a capital that employed no labour would produce no new value at all, however splendid its machinery, since on the theory the machines transfer their own value to the product piecemeal but add nothing beyond it. Each pole is thus not a preference of capitalists, who might be enlightened or greedy, far sighted or foolish, without the structure changing. Each is a compulsion generated by the unity’s ordinary operation, felt by every participant whether understood or not. Step three is to show the poles pressing on one another. As productivity rises across the economy, the labour time socially necessary to produce a given mass of commodities falls. The value created per worker per day does not rise merely because the worker produces more objects; the objects are cheaper. What rises, when productivity rises in the industries producing wage goods, is the rate of surplus value, because the portion of the day needed to reproduce the worker’s own wage shrinks. So far the tendencies cooperate. The strain appears at the level of the total mass: a system that continually economises on labour per unit of output must continually expand output, markets and the scale of production merely to keep the employed mass of labour, and with it the mass of surplus value, from contracting relative to the accumulated capital demanding valorisation. Expansion is therefore not a taste for growth. It is a structural requirement imposed by the contradiction itself, and the drive toward new markets, new products and new needs, which observers often attribute to the psychology of greed, is derived here from a relation.

Step four is to identify what the pressure does over time, without smuggling in a predetermined ending. The contradiction does not dictate a date or a drama. It predicts a pattern: recurrent pressure on profitability as the labour basis narrows relative to accumulated capital, counteracted by the cheapening of machinery and raw materials, by the expansion of the labour force drawn into wage work across the globe, by the lengthening or intensification of labour where resistance permits, by the growth of credit that anticipates future surplus value, and by periodic destruction of capital values in crises that resets the proportions violently. Each counteracting factor belongs to the same analysis as the tendency it counteracts, which is why the series treatment of the marxist crisis theory guide must be consulted for the economic sequel; the present article’s concern is the form of the claim. And the form is exactly what the first use at its best provides: two poles, one unity, mutual dependence, internal pressure, and a family of possible developments whose common root in the structure can be exhibited step by step. Compare a typical asserted version, in which a writer announces that capitalism is riddled with contradictions and moves on within the same sentence to unemployment or to war. The asserted version borrows the authority of the analysed version. The glossary discipline consists in refusing the loan.

One further feature of the worked example should be marked, because it recurs in every strong use of the term. The contradiction is invisible from inside either pole taken alone. Viewed purely as engineering, raising productivity is an unmixed good, more wealth from less toil. Viewed purely as employment, expanding the workforce is an unmixed necessity for those who must sell their labour power to live. The tension exists only at the level of the unity, where the product must be both cheapened and valorised, both produced with less labour and made the vehicle of unpaid labour. This level relativity is why contradiction claims are so often talked past in debate: the critic examines one pole, finds it perfectly coherent, and declares the alleged contradiction a phantom, while the defender examines the unity and finds the pressure real. Both can be describing the same process accurately at different levels, and a precise vocabulary would mark the level at which the claim operates as carefully as it names the poles. The constitution test, when it arrives, will build this requirement in: a contradiction claim that cannot state the unity within which its poles are constituted has not yet located its own subject matter.

Use two: complementary positions and opposed interests

The second use shifts the camera from tendencies to people, or more exactly to positions that people occupy. Capital and wage labour, landlord and tenant farmer in the agrarian analyses, creditor and debtor in the discussions of finance: pairs of positions defined by a single relation, each unintelligible without the other, whose occupants find their material interests set against one another by the relation itself rather than by temperament or misunderstanding. The classical theatre of this use in Capital is the chapter on the working day. Capital, as the purchaser of labour power, claims the right to use its purchase to the fullest, and presses toward extending the working day toward its physical limits and beyond them, into the worker’s sleep, childhood and lifespan. Labour, as the seller, claims the right to a working day of normal duration, on the ground that what was sold was labour power for a determinate period and not the worker’s life outright. Marx’s celebrated verdict on the collision is that between equal rights, force decides, meaning that the boundary is settled by struggle, organisation and law rather than by any deduction from the concept of exchange itself. Both claims arise from the same contract, honestly read from opposite ends. That is the signature of a second use contradiction: one relation, two valid derivations, opposed conclusions, and a history that records the fight.

Apply the four diagnostic questions and the second use scores well. The poles are exact: the class that buys labour power and the class that sells it. The unity is exact: the wage relation itself, without which neither class exists as a class. The dependence is constitutive in the strictest sense, stronger here than anywhere else in the system, since the classes are defined by the relation and by nothing else, not by income level, culture, or self identification. And the pressure demonstrably generates change: factory legislation, the normal working day, the shifting frontier between paid and unpaid labour, the whole institutional sediment of industrial politics. Why, then, does the table attach a reservation to this use, the objection that interest conflict is the common property of many theories and needs no dialectical vocabulary? Because the reservation is just. Pluralists, Weberians, public choice theorists and ordinary trade unionists all recognise opposed interests between buyers and sellers of labour, and a Marxist who presents the recognition itself as a discovery of dialectics is claiming patent rights over common land. What the Marxist analysis adds, and what alone entitles the dialectical name, is the constitutive depth beneath the conflict: the claim that the opposed parties are not independent agents who happen to meet in a market, as buyer and seller of apples meet, but positions created by one relation, so that the conflict is internal to the relation’s reproduction and cannot be finally settled while the relation stands. A wage bargain can be renegotiated indefinitely; the opposition that generates bargaining cannot be bargained away by the parties it constitutes, since abolishing it would abolish them in their current form. That is a stronger and more interesting claim than the claim that interests conflict, and it is the claim the word contradiction is needed to carry, where it is needed at all.

Can opposed classes stand in contradiction without either side being logically mistaken?

Yes, and the working day dispute shows how. Each side derives its claim from the same exchange relation read from its own end: the buyer claims full use of what was purchased, the seller claims a limit set by what was actually sold. The opposition is practical and structural, and involves no error of reasoning on either part.

The distinction between a contradiction of positions and a quarrel between persons deserves emphasis because political rhetoric constantly blurs it, in both directions. Supporters of the tradition sometimes write as though employing workers were a personal vice, and its opponents sometimes reply as though showing an employer to be generous refuted the analysis. Both moves miss the level at which the second use operates. The mill owner who shortens hours from conviction and the mill owner who lengthens them from greed occupy the same position in the relation, and the pressure of competition visits the consequences of the first owner’s virtue on his costs with the same indifference it visits the consequences of the second owner’s vice on his workers. Marx’s own preface to Capital disclaims in advance any portrait of individuals as personally responsible for relations whose creature they are, a disclaimer that is not politeness but method. Conversely, the analysis gains nothing by treating every industrial dispute as an emanation of the deep structure. A strike over a foreman’s insult is a strike over a foreman’s insult, and the analyst who reaches for the capital and labour contradiction to explain its timing will explain less than the historian who reads the local record. The second use, like the first, is a tool with a specified range, and precision consists in knowing where the range ends.

A second theatre for the second use, less famous than the working day and in some ways more instructive, is the relation between industrial capital and landed property in the third volume’s treatment of ground rent. Landowner and capitalist are both, on Marx’s account, appropriators of surplus value, and their interests coincide against the labourer whose unpaid work supplies the fund they divide. Yet within that alliance their interests collide structurally over the division itself: what the landowner captures as rent, on the basis of a monopoly over a non reproducible condition of production, the capitalist cannot retain as profit, and the political history of the nineteenth century, corn laws and their repeal at the head of it, is unintelligible without the collision. The example refines the concept in a useful direction. Complementary positions need not be the two great classes of the simplified schema; any relation that constitutes its poles and sets their claims against one another exhibits the form. It also displays the form’s explanatory modesty. The contradiction between land and capital did not abolish either party. It was managed politically, priced into leases, and eventually dwarfed by the declining weight of agriculture, a reminder that resolution by transformation is one outcome among several, and that the second use predicts struggle over terms rather than any particular victor.

Use three: the form that contradicts its content

The third use is the most distinctively Marxian of the four and the hardest to state without distortion, because it concerns not tendencies or interests but presentation: the way a social relation appears in the forms through which participants experience and transact it, and the claim that some of those forms systematically present the relation as something other than what the analysis shows it to be. The wage form is the exemplary case. A worker is hired for a week and paid a sum at its end. The transaction presents itself, to both parties and to the law that enforces it, as payment for labour performed, a fair exchange of a quantity of work for a quantity of money. Marx’s analysis divides the working day into two portions that the form renders invisible: necessary labour, in which the worker produces the equivalent of the wage, and surplus labour, in which the worker produces value appropriated without equivalent. The wage form, he argues, extinguishes every trace of the division, making all labour appear as paid labour, and in this precise sense the form contradicts its content. Nothing has been asserted and denied. What has happened is stranger and more interesting: a real relation generates, by its own mechanics, an appearance that misdescribes it, and the misdescription is not a trick played by anyone but a property of the form itself.

The commodity form supplies the second exemplary case, in the analysis of fetishism in the first chapter of Capital. Producers work independently, and their labours are socially validated only when their products exchange. The social character of their labour therefore confronts them as a property of things, the value of the coat or the linen, and relations between people assume the fantastic form of relations between objects. Again the structure is the same: a content, the mutual dependence of private labours in a divided economy, appears in a form, the exchange ratios of things, that both expresses the content and conceals its nature. Marx reaches here for religious analogies, misty realms and wooden tables turned spiritual, and the theatricality of the prose should not obscure the analytical claim, which is sober and even austere. Where social relations are mediated by the exchange of products, the mediation hides the relations it performs, and no amount of individual clear sightedness removes the hiding, since a producer who saw through the form entirely would still have to sell in the market it governs.

Why call this a contradiction rather than a deception or an illusion? The glossary answer must be exact. Not because anyone lies: Marx is emphatic that the wage form deceives without a deceiver, and the capitalist who pays the going wage breaks no rule of the exchange that constitutes his world. The justification for the word lies in the internal relation between form and content. The form is not an external costume draped over the relation; it is the relation’s own necessary mode of appearance, generated by the same structure whose content it misstates. Content without this form could not operate, since labour power is in fact bought and sold under the wage form and products do in fact exchange under the value form, and form without this content would be an empty shell. Each side requires the other, which is the constitution clause knocking at the door again, and yet the two sides stand in genuine opposition, because the form’s account of the transaction, all labour paid, is the direct inverse of the content’s structure, part of the labour unpaid. Form against content thus passes the constitution test in a distinctive register, constitutive but presentational, and the article’s verdict will be that the third use earns the name contradiction on condition that its register is marked. Unmarked, it slides: a writer establishes that the wage form misrepresents the working day and then writes as though a first use contradiction, with its dynamics of pressure and transformation, had been established in the same breath. The slide is common and it is illegitimate. A misrepresenting form can be extraordinarily stable, as the wage form’s long career attests, and nothing about the third use alone predicts that its object will transform rather than endure.

The young Marx’s version of this structure should be separated from the mature one, because the tradition’s blending of them causes needless confusion. In the 1844 manuscripts the governing gap is between human essence and human existence: labour, the activity in which the species realises itself, confronts the worker as an alien power, so that what people are contradicts what their conditions make of them. The structure is form against content wearing anthropological dress, essence as content, estranged labour as the form that belies it. The mature version replaces essence with mechanism. The wage form analysis needs no doctrine of what labour essentially is; it needs only the division of the working day and the form’s erasure of the division, both of which can be exhibited to a reader who rejects the early anthropology entire. Keeping the two apart matters for criticism, because objections that land on the early version, its speculative humanism and its unfalsifiable essence, are routinely treated as having landed on the later one, which stands on different and narrower ground. Precision about which Marx is speaking, at which date, in which manuscript tradition, is a recurring demand of this series, and nowhere is it more rewarded than here.

Use four: real opposition, or counterpoise without constitution

The fourth use takes the analysis out of Marx’s texts for a moment and into Kant’s, because the category at stake was defined there with a clarity nobody has improved. In his 1763 essay on negative magnitudes, Kant distinguished two kinds of opposition that careless language merges. Logical opposition, contradiction proper, occurs when a predicate is affirmed and denied of the same subject, and its result is nothing at all, an impossibility that cannot be thought as existing. Real opposition, which Kant called Realrepugnanz, occurs when two real grounds, each positive in itself, cancel one another’s effects in whole or in part, and its result is something entirely thinkable: rest, equilibrium, a diminished magnitude, a debt balanced against a credit. Two equal forces pressing a body from opposite sides produce real opposition and a stationary body. A pleasure and a displeasure of equal weight produce, in Kant’s analysis, not a logical scandal but a state of indifference. The category is indispensable to physics, to accounting and to any economics that models offsetting tendencies, and it has one decisive property for the present inquiry: real opposition requires no internal relation between its terms. The opposing forces do not constitute one another. Each would be exactly what it is if the other vanished tomorrow.

Marx uses opposition in this Kantian register constantly, and legitimately. The third volume of Capital, in the chapters on the tendency of the rate of profit to fall, sets out a rising organic composition of capital pressing profit downward and then enumerates counteracting influences pressing it upward: more intense exploitation, depression of wages, the cheapening of the elements of constant capital, relative overpopulation, foreign trade, the growth of share capital. The architecture of that argument is a balance of forces. Each influence is specified, its direction given, its strength left to empirical circumstance, and the net outcome is a tendency hedged by counter tendencies, a resultant rather than a destiny. Nothing in the passage requires, or is illuminated by, the constitutive machinery of the first and second uses. The cheapening of machinery does not owe its existence to the rising composition of capital in the sense in which wage labour owes its existence to capital; the two are independent causal streams that happen to intersect in the profit rate’s denominator and numerator. To call their intersection a contradiction in the strong sense is to borrow dignity the passage neither needs nor claims, and Marx’s own language in the passage, which speaks more often of counteracting influences and of the law as tendency, is more careful than the language of many of his summarists.

Why did Kant insist that real opposition is not contradiction?

Because the two oppositions have different results and different logics. Contradiction between a predicate and its negation yields nothing thinkable at all, while real opposition between two positive grounds yields a determinate outcome, rest or a resultant, that can be measured. Forces that cancel are real, opposed and entirely free of logical offence.

The reason an entire use must be quarantined in this way is not antiquarian. It is that the fourth use is the door through which the strongest philosophical objection to dialectical contradiction enters, the objection associated with Lucio Colletti, and the objection cannot be fairly stated until Kant’s distinction is in hand. If all of Marx’s substantive oppositions are real oppositions, then the Hegelian claim that contradiction, in a sense stronger than counterpoise, inhabits reality itself collapses into a metaphor, and the dialectic becomes a dramatic way of describing balances of forces that a sober vocabulary describes better. If some of Marx’s oppositions are more than real, if the constitutive structure of the first three uses marks a genuine category that Kant missed, then the metaphor charge fails for those uses, and Colletti’s argument must be answered where it is strongest rather than where it is weakest. The next sections build the instrument for deciding the question case by case, and then give the objection its full hearing.

The constitution test

Everything so far has been preparation for a single discriminating instrument, and the instrument can now be stated in one sentence. The constitution test holds that a genuine dialectical contradiction obtains where the two poles constitute each other, so that neither pole could exist, in the form under analysis, without the other, and it directs the analyst to sort every alleged contradiction by asking whether that mutual constitution holds, holds partially, or fails. The test is this article’s namable claim, offered in the spirit of the series: a portable rule, sharp enough to be wrong in particular applications, that a reader can carry to texts this article never mentions and apply without permission. Its power lies in its asymmetry. Confirming a contradiction under the test requires exhibiting the constitutive dependence in detail, pole by pole. Refuting one requires only a single clear demonstration that a pole survives, conceptually and materially, the removal of its alleged opposite. The test therefore reverses the usual burden in this literature, under which the sceptic is asked to disprove a mist. Under the constitution test the mist must condense into a mechanism or evaporate.

Applying the test is a three step discipline, and each step has its characteristic failure mode. The first step is specification: state each pole in terms that make no reference to the disputed conclusion, and state the unity that contains them. The failure mode is circularity, defining the poles so that their mutual dependence is true by stipulation, as when a writer defines capital as that which exploits labour and labour as that which capital exploits, and then announces their internal relation as a discovery. Honest specification gives independent content to each term: capital as accumulated value advanced to purchase labour power and means of production for the sake of expansion, wage labour as the condition of persons separated from means of production who must sell their capacity to work. Only when the terms stand on their own does their interdependence become a finding rather than a definition. The second step is the removal experiment, conducted in thought with historical discipline: subtract one pole and ask what becomes of the other, not in a slogan but in institutional detail. Remove wage labour from the capitalist relation and the owner of money and machinery does not become a capitalist without partners; he becomes a proprietor who must work his own assets, hire no one, or lend at interest to those who do, each a different social figure with a different logic. Remove capital and the wage labourer does not become a free producer in the same gesture, as the history of dismissed workers confronting locked factories demonstrates; he becomes unemployed, which is a condition defined precisely by the absence of the pole that constitutes his normal existence. The experiment’s asymmetry in that second case is instructive rather than embarrassing: constitution can be mutual without being symmetrical in every respect, and the test asks for dependence, not for mirror images.

The third step is the pressure audit: given the constituted poles, exhibit the pressure each exerts on the relation’s own conditions of existence, arising from the poles’ normal operation rather than from accident or vice. This step is where first use contradictions separate from second use ones in practice, since interest opposition alone, while sufficient for the second use, does not yet show either pole acting against the unity itself. Capital pressing to cheapen labour power acts against the condition that labour power be reproduced in health, skill and willingness; labour organised to shorten its day and raise its share acts against the condition that surplus value be produced at the expected rate; and both pressures are generated by fidelity to each pole’s own logic, not by departures from it. Where the pressure audit finds nothing, where two constituted poles coexist in a stable symbiosis that neither strains, the honest verdict is an internal relation without contradiction, a possibility the vocabulary needs and rarely admits. Not every internal relation is contradictory, just as not every opposition is internal, and a test that cannot return a negative answer to a friendly case is not a test at all.

Fresh cases calibrate the instrument better than canonical ones, so consider three in ascending order of difficulty. First, buyer and seller in a simple market for a finished good. The roles are reciprocal, and a sale requires both, which tempts a quick verdict of constitution. The removal experiment dissolves the temptation: the apple grower who fails to sell remains a grower with apples, the shopper who fails to buy remains a person with money, and each role is a temporary posture adopted by agents whose existence and character are constituted elsewhere, in production, in household need, in law. Buyer and seller are externally related phases of an act, not poles of a structure, and their occasional conflicts of interest over price are real oppositions with a bargaining range, not contradictions. The test returns a clean negative, and in doing so explains why Marx locates contradiction in production relations rather than in the sphere of circulation, however noisy circulation’s conflicts may be. Second, the relation between a platform and the workers whose labour it dispatches, which this series examines elsewhere as a test of class categories, also tests the present instrument instructively. The platform as a social figure, owner of the matching infrastructure and the customer relation, is constituted by the availability of a workforce it does not employ in law; the dispatched worker, considered in the role the relation assigns, is constituted by access to that infrastructure, without which the worker’s vehicle and willingness are unsaleable. Constitution holds at the level of the relation, in a historically novel form, and the pressure audit finds the familiar strain, the platform pressing to cheapen and intensify the labour it depends on while disclaiming the obligations of the dependence. The test travels to new material and bites there, which is what an instrument is for.

Third, the hardest canonical case: the contradiction between the forces and relations of production, the formula of the 1859 preface, in which the material forces of production come into conflict with the existing relations of production, and the relations turn from forms of development into fetters. Run the test. The poles can be specified independently: forces as the labour process’s technical and cooperative powers, relations as the property and control structure within which production proceeds. The removal experiment gives a mixed verdict, since forces and relations are not related as capital and wage labour are; a given technology can persist across very different property regimes, as the history of the steam engine under slavery, feudal obligation and wage labour attests, and a given property regime can house successive technologies. Constitution in the strict sense is therefore weak in this celebrated formula, and the relation is better described as a developing incompatibility, an internal tension of the second or fourth kind, than as a first use contradiction. That verdict will discomfort readers for whom the 1859 formula is the dialectic’s headquarters. It should instead reassure them about the instrument. A test that grades even the tradition’s most quoted formula, finds it misclassified, and specifies the correct classification, is a test with edges. The alternative, a defence that certifies everything Marx called a contradiction as a contradiction of the strongest type, is available in many textbooks and convinces nobody who was not convinced before opening them.

Internal relations: the philosophy underneath

The constitution test is not an invention floating free of the literature. It codifies the doctrine of internal relations, the philosophical load bearing wall of the dialectical tradition, whose most systematic modern statement is associated with Bertell Ollman’s work on alienation and on dialectical method. The doctrine holds that some relations are internal to their terms: the relation enters into what each term is, so that abstracting the term from the relation does not leave the term intact but produces a different and thinner object. External relations, by contrast, join terms that remain what they are whether the relation holds or not, as a book remains a book whether it sits on this shelf or that one. Ollman’s exposition ties the doctrine to a method of inquiry: begin from the whole whose internal relations structure the parts, move by abstraction to the parts as moments of that whole, and resist the empiricist habit of treating factors as independently given variables that subsequently interact. Interaction is the key foil. Variables that interact are presumed to exist fully formed before their encounter, and their encounter modifies their behaviour without touching their identity. Internally related poles have no such pre history. Capital is not first capital and then, additionally, related to labour; the relation is its substance.

Stating the doctrine exposes both its strength and the precise point at which critics should push. The strength is that internal relations are demonstrably real in social life, and not only in Marx’s examples. A husband in a legal order recognising marriage, a debtor under enforceable contract law, a goalkeeper within the rules of football: each is constituted by relations to other terms, and removing the relation removes the position, not merely the position’s circumstances. Social ontology cannot proceed without such relations, and a philosophy that treats every connection as external will misdescribe institutions as systematically as a biology that treated organs as independent machines would misdescribe bodies. The pushing point is scope. From the truth that some relations are internal, nothing follows about which ones are, and the dialectical tradition has at times slid from the modest truth to a maximal holism in which everything is internally related to everything else, a doctrine that explains nothing because it excludes nothing and that no working analysis, Marx’s included, actually observes. Capital itself proceeds by abstractions that provisionally isolate: value from use value, production from circulation, the individual capital from the total social capital, each isolation an analytic externalisation performed in full knowledge that the separated moments belong to one process. The method is therefore not holism against analysis. It is analysis conducted with a memory, abstractions held as moments rather than mistaken for independent things, and the constitution test is best understood as the operational form of that memory: a standing question, put to every pair of terms an analysis separates, whether the separation is a convenience or a falsification.

Hegel’s contribution and Marx’s alteration of it can be located exactly at this point, which is why the comparison article in this series, hegel dialectic vs marx dialectic, repays reading alongside this one. For Hegel, internal relations hold among determinations of thought that are also determinations of being, and contradiction is the motor by which a determination, fully thought through, shows its dependence on its opposite and passes into a unity that preserves both at a higher level. Marx keeps the structure of mutual determination and discards, or claims to discard, the idealist motor: the poles of his contradictions are not categories embarrassed by their one sidedness but classes, forms and tendencies embarrassed by their material conditions, and the passage beyond a contradiction is not a conceptual reconciliation but a historical transformation that somebody wins and somebody loses, or that nobody wins because the transformation is a shared ruin. The famous image of standing Hegel on his head is too theatrical to be precise, but the precision it gestures at is real: the constitution of poles is retained, the guarantee of reconciliation is withdrawn. A Hegelian contradiction carries its resolution within its logic. A Marxian contradiction carries only its pressure, and pressure is compatible with resolution, with stalemate, with displacement onto other populations or other continents, and with catastrophe. That withdrawal of guarantee is the mature tradition’s most underrated act of intellectual honesty, and it is the feature the next section’s objection will test most severely.

Colletti’s challenge: real opposition against dialectical contradiction

The objection can now be stated in its strongest scholarly form, the form given to it by the Italian philosopher Lucio Colletti in his 1975 New Left Review essay Marxism and the Dialectic, building on his earlier work on Hegel and Marxism. Colletti was no external enemy of the tradition. He wrote as a Marxist philosopher of formidable learning who had concluded that the tradition’s central methodological inheritance was rotten, and his argument deserves the compliment of exact reconstruction. Its first move is the Kantian taxonomy already established in this article: science and common sense alike know logical contradiction, a relation among propositions, and real opposition, a relation among things and forces, and the two must never be merged. Its second move is a claim about Hegel: that Hegelian dialectic requires precisely that merger, transposing contradiction from the sphere of propositions into the sphere of being, so that reality itself, and not merely our accounts of it, is said to contradict itself. Its third move is the application to Marx: that Marx’s materialism, insofar as it is a materialism, deals in real oppositions, forces, interests, tendencies in counterpoise, which are scientifically respectable and need no dialectical dressing, while insofar as Marx imports the Hegelian contradiction into matter he inherits an idealist mystification incompatible with his own science. The conclusion is surgical. Keep Marx the analyst of real oppositions and discard Marx the dialectician, or keep the dialectic and forfeit the science. The conflation of the two oppositions, Colletti argued, is not a depth but an error, and a fatal one for the claim that dialectical contradiction names anything in the world.

Feel the argument’s strength before reaching for replies, because its strength is considerable. It explains, as no defence does, the textual unevenness documented earlier in this article: the places where Marx’s contradiction talk reduces without remainder to offsetting tendencies are exactly the places where a materialist analysis was doing its work, and the places where the talk swells into the self movement of opposites are exactly the places where the Hegelian inheritance was writing cheques the materialism could not cash. It also explains the tradition’s defensive habits. Faced with the demand to exhibit one contradiction that is not a real opposition, defenders characteristically produce an example, capital and labour say, describe its real oppositions vividly, the struggle over the working day, the contest over the product, and then declare the constitutive depth to have been exhibited, when what has been exhibited is the conflict and the depth has been asserted in the commentary. Colletti’s taxonomy gives the sceptic a standing question that cuts through this manoeuvre: name the property the relation has, beyond the opposition of its terms and the interdependence of the positions, that earns the logical word. If no property can be named, the word is ornament.

The replies available to the tradition are three, and they are not equally good. The weakest reply is jurisdictional: that Colletti assumes a Kantian framework whose authority the dialectician need not grant, and that dismissing dialectical contradiction for failing Kant’s taxonomy begs the question against Hegel’s Logic, where the taxonomy itself is put in question. There is a real philosophical dispute underneath this reply, but as stated it functions as an exemption claim, a refusal to sit the examination on the ground that the examiners are positivists, and it has persuaded few who did not already hold the doctrine. The second reply is textual and historical: that Marx’s considered view, especially in the Grundrisse and in the method of presentation in Capital, treats the unity of opposites as a feature of the object as presented by a science that reconstructs the concrete in thought, so that contradiction names the structure of the presentation of a totality whose moments mutually presuppose one another, and the charge of conflating logic with physics rests on reading a presentational doctrine as a cosmological one. This reply is stronger, and it correctly relocates much of the dispute to the theory of science rather than to metaphysics. The third reply is the one this article has been constructing, and it concedes the taxonomy while disputing the inventory. Grant that logical contradiction and real opposition are distinct and that neither, alone, is what the strong Marxist usage means. The strong usage names a third configuration that Kant’s scheme omits, not because Kant was careless but because his examples were drawn from mechanics and mathematics rather than from institutional life: poles internally related, constituting one another within a unity, whose internal pressure drives the unity’s transformation. A goalkeeper and the offside rule, a debtor and a creditor under a shared legal order, capital and the wage labour it simultaneously requires and expels: these are not propositions, so no logical contradiction arises, and they are not independent forces, so real opposition underdescribes them, since deleting one term deletes the other term’s identity and not merely its sparring partner.

Whether the third configuration deserves the contested word is, at the end of the Colletti exchange, partly a legislative question about vocabulary, and the honest article should say so. Call the configuration constitutive opposition and Colletti’s scientific scruples are satisfied while the analysis proceeds unchanged, which suggests that part of the quarrel is over custody of a noun. But custody matters more than the deflationary mood admits, for a reason the next section develops: the word contradiction, whatever its liabilities, carries a standing reminder that the poles are internally related and that their conflict is therefore not amenable to the remedies appropriate to external collisions, neither separation nor compromise in fixed proportions nor the victory of the stronger force leaving the field’s structure intact. Colletti’s challenge survives the exchange as a permanent discipline rather than as a refutation. It convicts the fourth use, it convicts every slide among uses, and it obliges every surviving use to exhibit its constitutive structure rather than intone it. A tradition that accepted that discipline would write differently, more slowly, with fewer contradictions per page and more anatomy per contradiction. Whether the tradition’s flagship term also survives the second great objection, the deflationary one, is a separate trial, and it is conducted next.

The deflationary objection at full strength

The second great objection is less scholarly than Colletti’s and more damaging, because it requires no Kant and no Hegel to state and because every reader can test it at the desk. It is pressed in various forms by the analytical critics of dialectics, and the series article on analytical marxism and dialectics gives the school its full hearing; here the objection is stated as a single challenge with three supporting counts. The challenge is a substitution test. Take any passage of Marxist writing that leans on the word contradiction and replace the word, mechanically, with conflicting tendencies, opposed interests, or structural tension, whichever fits the grammar. If nothing is lost in the replacement, if every inference the passage drew can still be drawn and every fact it cited still cited, then the word was decoration, and a literature that leans on decoration at its conceptual centre has a problem that no amount of exegesis can dissolve. The burden, the objection insists, sits squarely on the dialectician: produce one claim, just one, that cannot be stated without the term, or concede that the term is a prestige synonym whose function is sociological rather than analytical, a badge of belonging for those who use it and a haze for those who do not.

The first supporting count is semantic. Contradiction already has a precise job in the language, the logical job, and borrowing it for tendencies and interests creates a standing equivocation that dialectical writing exploits whether or not individual writers intend the exploitation. A critic who objects that tendencies cannot contradict one another is answered with the constitutive depth of the first use; a critic who probes the constitutive depth is answered with examples of the fourth use, forces in counterpoise that nobody denies; and the word shuttles between its senses, gathering the plausibility of each in turn and surrendering none. The second count is methodological. A term that can denote tendency, interest, appearance and force can be stretched to cover any unwelcome fact, and a theory whose central diagnostic term stretches on demand has insured itself against refutation, since every failure of a predicted transformation can be redescribed as a contradiction displaced, deepened or temporarily resolved, with the redescription costing the theory nothing. The third count is historical and the least polite. The vocabulary flourished institutionally in parties and academies where the appearance of possessing a method superior to formal logic and to empirical social science conferred authority, and vocabularies that confer authority are retained for that reason long after their analytical yield has been questioned, as the histories of other prestige terminologies in other orthodoxies illustrate. The objection therefore ends not with a request but with a diagnosis: contradiction persists, on this view, because it is useful to its users in ways that have little to do with its truth.

What analytical work does the word perform that conflict cannot?

It compresses three claims into one term: that the opposed poles belong to a single unity, that each pole is constituted by its relation to the other, and that their pressure arises from the unity’s own operation. Conflict asserts opposition and leaves all three claims unstated, which is why the substitution test fails exactly where the constitution test passes.

That compressed answer is the strongest reply available, and the sections that follow from here are an extended audit of how much of the literature it actually covers. Notice first what the reply does not claim. It does not claim that the three compressed claims are true of every pair the tradition has called contradictory; the table and the test have already surrendered that ground. It does not claim that the three claims could not be stated in other words; of course they could, at greater length, since any content that can be thought can be paraphrased. It claims something narrower and more defensible: that where the constitution test passes, the word contradiction functions as an abbreviation for a specific and checkable structure, and that abbreviations for checkable structures are legitimate instruments of analysis rather than ornaments. Science is full of such compressions, terms that pack a mechanism into a noun and save the labour of unpacking it at every mention, and their legitimacy is always conditional on the unpacking being available on demand. The dialectician who cannot unpack, who meets the request for poles, unity and constitution with a further rotation of the vocabulary, stands convicted by the objection in his own person, whatever the doctrine’s merits in better hands.

The reply audited, and the concession recorded

An honest audit must now run the substitution test across the literature’s actual range rather than across its best cases, and the result of that audit is the concession this article was commissioned to record without flinching. A great deal of published Marxist writing would survive the substitution unchanged. In the journalism of movements, in the commemorative essay, in the textbook chapter written to a syllabus, and in a sobering proportion of academic production, contradiction functions exactly as the deflationary objection alleges: as a solemn synonym for tension, problem, or conflict, lending an air of method to observations that any competent reporter could have made in plain words. The sentence announcing that a government’s housing policy is contradictory, that a corporation’s green commitments contradict its lobbying, that a celebrity’s image contradicts a celebrity’s conduct, loses nothing when the key word is replaced, because nothing was present to lose. Even in more serious registers, the fourth use analysis shows the substitution succeeding wherever the content is a balance of tendencies, and the form against content analysis shows it half succeeding wherever the point is simply that appearances mislead, a point the vocabularies of ideology critique and of accounting fraud both make without dialectical assistance. The concession, stated as a quantity, is that the word is necessary in a minority of its occurrences and ornamental in the majority, and that the majority’s ornamentation draws its prestige parasitically from the minority’s substance. A term in that condition is not refuted. It is in debt, and the debt is callable by any reader armed with the substitution test.

What survives the audit, and survives it in good order, is the constitutive core. Where the constitution test passes, the substitution demonstrably fails, and the platform case and the capital and labour case show the failure concretely. Replace contradiction with conflicting tendencies in the analysis of the working day and the inference structure changes under the replacement: conflicting tendencies can be weighed, balanced, or traded, and a reader so equipped will expect the dispute to be settleable by a sufficiently clever compromise inscribed in law, whereas the analysis’s whole point is that the dispute regenerates from the relation itself after every settlement, because each settlement leaves the constituting relation, and therefore the poles and their opposed derivations, exactly where they were. The word contradiction, disciplined by the constitution test, is the marker of that regenerative property. It says: this opposition will not stay settled, not because the parties are stubborn or the negotiators unskilled, but because the unity that constitutes the parties reproduces the opposition as a condition of its own continuation. That is a falsifiable, substantive, and distinctive claim. It can be wrong about a given case, as it would be wrong applied to a trade dispute between independent states whose economies barely interlock, and its wrongness in such a case is discoverable by exactly the removal experiment the test prescribes. A vocabulary item that marks a falsifiable structural claim has earned its place, and the deflationary objection, which began as a demand for one such claim, ends as an instrument for finding the places where the claim is genuinely made. The objection therefore succeeds against the vocabulary’s necessity as a whole while failing against its coherence at the core, which is the most instructive outcome an objection can have: it leaves the concept smaller, cleaner and usable.

The audit also clarifies what defenders should stop doing, and a glossary article is the right place for plain speaking. Defenders should stop answering the substitution challenge with volume, piling example on example of conflict and calling the pile depth. They should stop treating the request for a mechanism as a failure of dialectical sensibility on the requester’s part. They should stop citing Hegel’s authority as though the Logic were a certificate exempting its holders from the removal experiment. And they should stop, above all, using the word in the fourth sense while spending the authority of the first, the slide this article has flagged from its opening pages as the literature’s characteristic vice. The corresponding duties for critics are milder but real. A critic who has run the substitution test only across journalism has refuted journalism. A critic who denies that internal relations exist owes an account of debtors, goalkeepers and spouses. And a critic who treats Colletti’s taxonomy as the end of thought, rather than as the beginning of an inventory, will miss the one configuration, constitutive opposition under internal pressure, that the taxonomy’s own author may have underestimated because his examples came from mechanics. Between the disciplined defenders and the disciplined critics, the word’s fate is not a verdict but a workload: case by case, test by test, which is where this article leaves it and where the persona layers below take it up for different kinds of reader.

Overdetermination: when no contradiction stands alone

One complication must be absorbed before the vocabulary can be used responsibly on real history, and it arrives under the name overdetermination. The term is borrowed from Freud’s account of dreams and symptoms, in which a single element is determined by several converging chains of causes, and it enters Marxist method through Althusser’s essays of the nineteen sixties, where it serves to correct a picture the simpler contradiction talk encourages: the picture of a society organised around one contradiction, capital against labour, whose development proceeds by its own logic and whose resolution will transform the whole in a single intelligible movement. Althusser’s counter claim is that every contradiction in a social formation exists in a specific conjuncture of other contradictions, unevenly developed, displaced and condensed by one another’s action, so that the moment of rupture, when it comes, arrives not as the maturation of the principal contradiction in isolation but as a fusion of many, each at its own tempo, in a configuration nobody’s schema predicted in detail. His standing example is the Russian revolution, located not in the most advanced capitalist formation, where the principal contradiction should by the simple picture have ripened first, but at the weakest link, where national, agrarian, military and dynastic contradictions had accumulated into a single inflammable heap.

Overdetermination complicates the constitution test without dissolving it, and the complication should be stated precisely. The test is applied to pairs, and real formations present networks. A pair that passes the test in abstraction, capital and wage labour say, may in a given conjuncture be displaced in its effects by a national contradiction, condensed with a religious one, or held in suspension by a state form whose own contradictions intersect both poles asymmetrically. The analyst who has certified the pair and stops there has a true structural claim and a false historical expectation, since the structure’s pressure is real but its timetable and its theatre are set by the network. The correction overdetermination supplies is therefore about inference, not about inventory: from the existence of a contradiction, nothing follows about its dominance, its visibility, or its imminence, and every step from structure to event must be argued through the conjuncture rather than read off the pair. This is also the point at which to register the honourable function of the concept of the principal contradiction, before the boundary discipline of the next section assigns its political development elsewhere. Within a network, analysis must still rank: some contradictions condition the others more deeply, set the limits within which the others move, and would transform the whole field if they shifted, while others are local, derivative, or episodic. To call one contradiction principal is to make that ranking claim, and like every claim in this article it is checkable in principle and contestable in fact. What the ranking is not, on any disciplined use, is a permission slip to ignore the subordinate contradictions, since the lesson of overdetermination is that the subordinate ones decide when and how the principal one becomes historical.

Readers should also be warned against the mirror image error. Overdetermination is sometimes used, by writers weary of the old certainties, as a solvent in its own right: everything is overdetermined becomes a reason to analyse nothing in particular, a sophisticated shrug dressed as a method. The concept has no such licence in it. In Freud, from whom it was borrowed, overdetermination multiplied the analyst’s work, requiring each converging chain to be traced separately before their convergence could be understood. In social analysis it does the same. Each contradiction in the network must pass its own constitution test, each displacement must be shown rather than announced, and the fusion at a rupture must be reconstructed chain by chain. Properly used, overdetermination raises the standard of proof for contradiction talk across the board. Abused, it becomes the most prestigious way yet devised of avoiding proof altogether, and the difference between the uses is visible, as usual, in whether the writer can name poles and unities or only atmospherics.

Principal and antagonistic contradictions, at the boundary

Two terms from the political development of the doctrine belong in a glossary of contradiction even though their full treatment belongs to another article, because readers meet them constantly and because their misuse feeds back into the parent concept. The principal contradiction, in the usage systematised by Mao in the 1937 essay On Contradiction, is the contradiction within a process whose development conditions the development of the others, and its identification is presented as the first task of practical analysis, since strategy consists above all in knowing which opposition, among the many a situation contains, is the one whose resolution would transform the rest. The distinction between antagonistic and non antagonistic contradictions follows in the same development: antagonistic contradictions are those between poles whose conflict cannot be resolved within the existing unity and therefore tends toward the unity’s rupture and the poles’ transformation into something else, while non antagonistic contradictions, the famous contradictions among the people, occur between poles sharing a fundamental interest and can be resolved by criticism, adjustment and education inside a unity that persists. Mao’s political application of the whole apparatus, including its refinement into the doctrine of the principal aspect, is the subject of the series article on mao on contradiction and practice, and it is deliberately not re-argued here, in line with the anti cannibalisation discipline of this series.

What falls within the present article’s remit is the pair’s effect on the parent term’s precision, which is mixed and must be honestly weighed. On the credit side, the antagonistic distinction builds into the vocabulary the lesson this article has laboured to establish from the other direction: that contradiction does not mean impending rupture, since a whole class of contradictions is defined by resolvability within the unity, and that the analyst’s task includes grading the contradictions found rather than merely collecting them. On the debit side, the political career of the terms shows both being used as instruments of classification by authority, in which the labelling of a dispute as antagonistic or otherwise preceded and pre-empted its analysis, and in which the principal contradiction was whatever the directing centre announced it to be. A glossary can neither celebrate nor bury that career in a paragraph; it can only fix the terms’ analytical meanings tightly enough that their political use becomes visible as a use, and answerable to the meanings. Under the constitution test, both terms operate downstream of a prior verdict: a pair must first be a contradiction, constitutive and internally pressured, before the questions of its rank among contradictions and of its resolvability within its unity can so much as arise, and much confusion in the secondary literature comes from running the downstream questions on pairs that never passed the entrance examination.

Resolution, transformation, and the refusal of guaranteed endings

The last conceptual piece is the one on which popular understanding is most reliably wrong: what happens to a contradiction. The disambiguation guard for this article assigns the economics of crisis to its own guide, and the question here is the general grammar of outcomes. Popular usage, friendly and hostile alike, assumes a two term grammar: a contradiction either deepens until the system collapses, or the system somehow muddles on and the contradiction talk was overwrought. The tradition’s own serious literature recognises at least five outcomes, and distinguishing them converts many sterile arguments into answerable questions. Resolution proper occurs when the unity itself is transformed so that the poles, as constituted, cease to exist: the wage relation abolished, in the limiting case the tradition imagines, would resolve the capital and labour contradiction by removing both relata in their capitalist form, and nothing less counts as resolution in the strict sense. Transformation without resolution occurs when the poles persist but their relation is reconstituted on new terms, as the factory acts reconstituted the working day contradiction inside a legal frontier that both sides thereafter treated as terrain rather than as the question itself. Displacement occurs when the pressure is exported beyond the unity’s visible boundary, onto colonial populations in the classic analyses, onto households and unpaid labour in the feminist extensions of them, onto the natural environment in the ecological ones, so that the home contradiction quietens precisely because its costs have been relocated to poles the original analysis did not count. Adaptation occurs when institutions grow up whose function is to metabolise the pressure continuously, collective bargaining machinery, welfare provision, countercyclical management, so that the contradiction becomes a governed process rather than an accumulating charge. And catastrophe, the shared ruin of the contending classes that the Communist Manifesto itself lists as an outcome of class struggle alongside revolutionary reconstitution, remains on the list as the tradition’s own admission that pressure guarantees nothing except that the status quo is not self stabilising by definition.

Each outcome corresponds to a failure mode in criticism that a precise vocabulary prevents. The hostile critic who points to a century of adaptation and declares the contradiction imaginary has confused the absence of resolution with the absence of pressure, as though a river dammed and channelled had been shown to have no current. The friendly writer who greets every adaptation as a contradiction deepening in disguise has confused the presence of pressure with the imminence of resolution, as though a river’s current guaranteed the dam’s collapse by a date the writer declines to specify. Both errors are nourished by the two term grammar, and both are starved by the five term one, since adaptation and displacement are exactly the outcomes the grammar’s extra terms make speakable. There is also a sober implication for the constitution test’s users. Because outcomes are plural, the test cannot be validated by outcomes: a certified contradiction may adapt for generations, and a merely external conflict may explode tomorrow for reasons of accident and personality. The test certifies a structure, and structures constrain and pressure histories without scripting them. Readers who require of a method that it script history are asking for prophecy, and the tradition’s prophetic reputation, earned in its pamphleteering and forfeited in its scholarship, is the one inheritance this glossary article is content to leave unclaimed.

A final clarification belongs to the grammar of outcomes, concerning the word resolution itself, which misleads by suggesting a puzzle solved and filed away. In the Hegelian ancestry, resolution carries the sense of Aufhebung, a cancelling that preserves, and whatever one thinks of that machinery as logic, its historical translation is sound: transformations that resolve a contradiction preserve much of what the poles had built, their skills, their institutions, their accumulated productive powers, under new relations, which is why the tradition’s own images of resolution are of inheritance under new management rather than of a blank field. The opposite error to the collapse fantasy is the fantasy of total novelty, and a glossary that corrects the first while licensing the second has done half its work. Contradictions end, when they end, the way institutions end: partially, unevenly, with survivals that embarrass the theory of the ending for decades afterward, and the analyst’s vocabulary should be built to describe that untidiness rather than to promise cleanliness in either direction.

A borderline case graded in full: the state that must feed what it restrains

One further application, worked at length because it sits exactly on the instrument’s edge, will show the constitution test handling a case where the correct verdict is a refined partial rather than a clean pass or fail. The case is the capitalist state as analysed in a line of theory running from James O’Connor’s account of the fiscal crisis to Claus Offe’s studies of the welfare state. The claimed contradiction, stated in its strongest form, is this. The state in a capitalist order depends for its revenue, its employment figures and its electoral survival on the continued private accumulation of capital, which it does not control and must not frighten. The same state is charged, by democratic pressure and by the functional needs of accumulation itself, with restraining, taxing, educating, healing and compensating the casualties of the accumulation it depends on. Its expenditures must therefore grow in two directions at once, toward subsidising accumulation and toward repairing accumulation’s effects, while its revenue is drawn from the very process both expenditures burden, and O’Connor’s celebrated formulation was that the state must perform two mutually contradictory functions, accumulation and legitimisation, on a fiscal base that the performance of each function erodes.

Grade the claim by the protocol. Nomination succeeds: the poles are not abstractions but specifiable mechanisms, the state’s fiscal dependence on taxable private activity and its expenditure commitments to social provision and legitimation. Unity succeeds: a single budget and a single institutional order contain both. Constitution is where the refinement begins. The state as such is not constituted by capital in the way the wage labourer is constituted by the wage relation; states predate capitalism, survive in non capitalist orders, and possess coercive and administrative substance of their own, so the removal experiment, applied to the state in general, fails. Applied to the specifically capitalist form of the state, the experiment bites harder: a state that finances itself by taxing an accumulation process it does not direct, that is constitutionally barred from commanding the investment on which its revenues depend, and that must nevertheless answer electorally for the employment and welfare that investment decides, is a figure that cannot be specified without the capital relation that both empowers and handcuffs it. Constitution therefore holds at the level of the form, exactly as it did in the grading of social production against private appropriation, and fails at the level of the institution in general. Pressure, at the level where constitution holds, is exhibited rather than asserted: each function, performed well by its own standards, raises the cost of the other, since generous legitimisation spending increases the fiscal claim on accumulation while vigorous accumulation support increases the casualties requiring repair, and the squeeze appears in experience as the chronic fiscal stress and the recurring politics of austerity and overload that the theory was built to explain.

The verdict, then, is a second register contradiction of the positional kind, internal to a historically specific state form, with its scope marked in both directions: narrower than its enthusiasts claim, since it says nothing about states as such, and stronger than its critics allow, since no paraphrase in terms of competing priorities captures the regenerative structure, the way each budget round reissues the squeeze because the form that generates it remains in place. The borderline case also yields the protocol’s final lesson. A partial verdict is not a failure of the test but its characteristic success, since the interesting claims in this literature are almost never pure. They are constitutive at one level of abstraction and external at another, positional in one period and merely tense in a second, and a method that can say exactly where the constitution stops has given the researcher a map of where further evidence would change the grade. That is all that any instrument in the human sciences can honestly promise, and it is considerably more than the vocabulary’s ordinary usage, which knows only the two verdicts of profundity and dismissal, ever delivers.

The word across the traditions

A glossary entry owes its readers a map of how the term’s meaning shifts as it crosses the tradition’s internal borders, because a reader who learns the word from one school and applies it in another’s texts will misread both. In the orthodox codification that descends from Engels’s Anti Dühring and the Dialectics of Nature through the Soviet textbooks, contradiction is one moment of a universal dialectic governing nature, society and thought alike, illustrated by examples drawn from mechanics, chemistry and biology as readily as from economics, and its register is cosmological: the unity and struggle of opposites as a law of everything that exists. The precision cost of that register is the one Colletti invoiced, since natural examples are almost invariably real oppositions, magnets and charges and action and reaction, and a doctrine trained on them imports their logic into social analysis unmarked. Marx’s own usage, by contrast, is overwhelmingly institutional and economic, and the distance between the master’s practice and the codification’s scope is one of the open secrets of the scholarship, to be handled by noting it rather than by either denying it or treating it as a scandal that settles every question it touches.

In the Western Marxist line that runs from Lukács through the Frankfurt School, contradiction migrates toward the theory of consciousness and culture while keeping its constitutive skeleton. Lukács makes the contradiction between the contemplative stance imposed by reification and the practical standpoint of a class constituted to abolish it the hinge of an entire philosophy of history, a use that passes the constitution test at the level of class positions while adding a philosophy of the subject that Marx’s economics neither supplies nor requires. Adorno radicalises the term in the opposite direction, toward non resolution: negative dialectics keeps contradiction as the index of a world whose concepts fail their objects, and refuses in advance the reconciliations, Hegelian or political, that would declare the failure healed. In the Maoist development, already bounded above, the term becomes an instrument of strategy, a way of ranking the oppositions in a concrete situation so that action can be concentrated where the structure is most load bearing. In the analytical school, finally, the term is retained, where it is retained at all, as a courtesy title for precisely specified mechanisms, with writers such as Jon Elster demanding that any alleged contradiction be restated as a counterfinality, a fallacy of composition, or a conflict among agents with defined preferences, and accepting the restatement as the analysis itself. The map’s lesson is that contradiction names a family held together by the constitutive core and shading off, at every border, into neighbouring notions that other vocabularies handle as well or better. A reader equipped with the table and the test can locate any given author’s usage on the map within a page or two of encounter, which is the practical definition of having learned a technical term.

Two border crossings deserve individual mention because they generate disproportionate confusion in examinations and in journalism alike. The first is the crossing into ordinary political English, where contradiction has completed its slide into a synonym for inconsistency between statements, so that a politician is said to contradict himself when his Tuesday speech denies his Monday one. That usage is legitimate English and has nothing to do with Marx, but its dominance in the ambient language means every Marxist use is heard against it, and part of the teacher’s task, addressed in the persona layer below, is to make the homonymy explicit before it does its damage. The second crossing is into the vocabulary of systems and policy analysis, where internal contradiction is used for any design whose components impose incompatible requirements, an aircraft that must be both light and armoured, a benefits system that must be both generous and cheap to administer. These are tradeoffs, sometimes tragic ones, and the closing sections of this article are concerned with keeping them distinct, since a tradeoff between requirements imposed on an artifact by its designers involves no poles constituting one another through a social relation, however painful the choice between the requirements may be. The pain is real and the category is different, and a vocabulary that cannot register the difference will call everything a contradiction and analyse nothing.

A researcher’s protocol: six checks before the word is used

The cumulative argument of this article can be compressed into a working protocol, offered to any reader who writes as well as reads, and it doubles as the standard against which the article’s own claims should be judged. The first check is nomination: name the two poles in plain institutional language, without the word contradiction doing any work in the naming, and refuse to proceed if either pole can only be described as a side, a force or an aspect of the problem. The second check is unity: state the single relation or process within which the poles stand, and be able to say what would count as being outside it. The third check is constitution: run the removal experiment on each pole in institutional detail, and record honestly whether the dependence is mutual, one sided, or absent, since one sided dependence yields relations of domination or dependency that are important in their own right and misnamed as contradictions. The fourth check is pressure: show that each pole, operating normally and even virtuously by its own standards, strains the conditions the unity needs to persist, and distinguish that strain from frictions caused by incompetence, scarcity or ill will. The fifth check is register: declare which of the four uses is in play, tendencies, positions, form against content, or real opposition, and forfeit any inference that belongs to a different register. The sixth check is substitution: replace the word with its plainest paraphrase and read the passage again, keeping the word only where the paraphrase loses the regenerative, constitutive content the analysis actually uses. A claim that passes all six checks has earned the strongest term in the tradition’s methodological vocabulary. A claim that fails one check has located its own correct description, which is a gain and not a defeat, since a precisely named conflict or tradeoff can be argued about, evidenced and settled in ways a vaguely certified contradiction cannot.

The protocol has one further virtue worth recording: it makes disagreement productive. Two analysts who dispute a contradiction claim usually talk past one another, one defending the word and the other attacking it. Two analysts armed with the six checks dispute a location instead: the nomination, the unity, the removal experiment, the pressure, the register, or the paraphrase. A disagreement with a location can be settled by evidence addressed to that location, and a dispute that can be settled is the difference between a research tradition and a debating society.

The protocol also disciplines reading, and not only writing. Faced with a celebrated text, the six checks convert reverence or suspicion into a sequence of answerable questions, and the history of the concept suggests that the conversion is overdue by about a century. Marx himself passes the checks at his best with an ease that explains the vocabulary’s survival: the working day chapter nominates, unifies, constitutes, pressures and declares its register within a few pages, and the paraphrase test fails there exactly as the defence requires, because no paraphrase of the chapter’s central claim preserves its point about settlements that regenerate dispute. The codifications that followed him pass less often, and the journalism that invokes him passes least of all, which is the distribution one would expect if the term were a genuine instrument that most of its inheritors use as a emblem. None of this requires the reader to become a partisan of the instrument. It requires only the recognition that the instrument exists, that it can be told apart from its emblematic counterfeit, and that telling them apart is a learnable skill with rules that this article has tried to state in full.

Contradiction and its near neighbours: a last set of distinctions

The closing summary of correct usage promised by the glossary template can be given as a set of distinctions among the words that crowd this neighbourhood, since most errors with contradiction are at bottom errors of neighbourhood. Conflict is the genus of opposed aims or claims among agents, and it asks nothing about constitution: enemies in a war are in conflict, and each would flourish in the other’s absence, which is precisely what the strong contradiction forbids. Tension is a state of strain within a system, registerable without any poles being named, and it is the honest word for most situations in which the poles have not yet been identified. Tradeoff is a relation among requirements or goods under scarcity, where advancing one dimension costs another, and its poles are values or specifications rather than positions in a relation. Paradox is a property of statements or descriptions, an apparent absurdity that closer analysis dissolves or deepens, and it belongs to the logic of accounts rather than to the structure of the accounted world. Crisis is an event or phase, a moment when pressures become unmanageable by routine means, and it stands to contradiction as weather stands to climate: related, sometimes dramatically, but a different category, with the series guide to crisis theory carrying the economic development of the distinction. Real opposition, finally, is counterpoise among independent forces, and it is the correct name for the fourth use wherever that use is doing honest work.

Against that field of neighbours, correct usage of the headword can be stated in three sentences that a reader may treat as the article’s portable residue. Use contradiction where two poles of one relation constitute each other and, by operating normally, strain the relation that constitutes them. Use the neighbour words everywhere else, and let the analysis gain the credibility that restraint purchases. And when reading, extend to every author, beginning with Marx, the same discipline: grade the usage by the test, keep what passes, rename what fails, and notice how much of the tradition’s reputation for obscurity was a failure of grading rather than a property of the world it described. The word, so disciplined, is neither the master key its devotees advertise nor the empty incantation its detractors dismiss. It is a good tool with a narrow jaw, and this article’s whole labour has been to measure the jaw.

Three more worked examples, graded

The instrument earns confidence by being used on material beyond its home examples, so three further canonical cases are graded here, with the grades argued rather than announced. The first is the contradiction at the heart of the commodity, between use value and value, which the first chapter of Capital presents as the germ cell of the whole analysis. Run the checks. The poles are specified: a thing’s usefulness, grounded in its physical properties and in human needs, and its value, grounded in the abstract labour socially necessary for its production and expressed only in exchange with another commodity. The unity is the commodity itself, the elementary form of wealth in societies where the capitalist mode prevails. Constitution is the interesting step. A product made purely for use, consumed by its maker, is not a commodity and carries no value in Marx’s technical sense; value exists only where products confront one another in exchange. Conversely a pure value with no use value is impossible within the analysis, since nothing useless commands exchange, a point Marx states as a condition: a thing that nobody wants is not a bearer of value whatever labour it cost. Each pole therefore conditions the other’s existence in the form under analysis, the labour producing the thing must be at once concrete labour making this useful object and abstract labour counting as a fraction of society’s total, and the famous dual character of labour is exactly the constitution clause rendered as a doctrine of production. The pressure audit finds real strain: production for exchange subordinates usefulness to saleability, so that what gets produced is decided by the value pole while needs answer to the use pole, and the periodic discovery that warehouses of value coexist with unmet needs is the pressure surfacing in experience. Grade: a genuine contradiction of the first use, intertwined with the third, since the value form also presents the social character of labour as a property of things. The example matters beyond its elegance, because it shows the test certifying a case whose poles are aspects of an object rather than groups of people, confirming that the instrument tracks relations and not merely class antagonisms.

The second case is the contradiction between social production and private appropriation, the summary formula Engels gives in Anti Dühring and Socialism: Utopian and Scientific for the whole capitalist order. Production, on the account, has become social in its organisation: hundreds cooperate in factories, thousands across supply chains, under a technical division of labour that makes every product a collective achievement whose individual authorship is untraceable. Appropriation remains private in its form: the collective product belongs to the owners of the separate capitals on whose premises and under whose contracts the cooperation occurred. Nomination and unity are easy. Constitution is subtler and must be argued with care. Socialised production does not depend for its existence on private appropriation in the abstract, since cooperative and public forms of organising the same technical processes are conceivable and historically attested at various scales; what depends on private appropriation is the specifically capitalist form of social production, in which cooperation is assembled by capitals through the wage relation and dissolved by them when valorisation fails. The formula therefore names a contradiction internal to the capitalist organisation of a socialised labour process, not a contradiction between social production as such and property as such, and grading it requires holding that scope exactly. Pressure is exhibited in the analysis’s characteristic phenomena: crises of overproduction in which socially organised plenty confronts privately limited demand, planning inside each firm of increasing sophistication set against anarchy between firms, and the recurring political demand that an organisation already social in fact be made social in form. Grade: a genuine first use contradiction within its proper scope, and a standing lesson in how slogans expand beyond their scope in transmission, since the formula is routinely quoted without the capitalist qualifier that its own logic requires.

The third case is deliberately chosen from the hostile direction: the alleged contradiction between democracy and capitalism, a staple of both socialist argument and its liberal mirror image. Grading must begin by refusing the slogan form, in which two abstract nouns are set against each other with no relation specified. Specified properly, the candidate poles are political equality, one citizen one vote with formal equality before the law, and economic inequality of the kind that converts into unequal political weight through campaign finance, media ownership, lobbying capacity and the structural dependence of governments on private investment decisions. The unity would be the liberal capitalist state considered as one institutional order. Constitution, however, fails in the strict form: political equality in its modern form predates mature capitalism in aspiration and has survived in orders with very different economic structures, and capitalist production has coexisted historically with franchise restrictions, authoritarian states and democracies alike, so neither pole owes its existence to the other as capital owes its existence to wage labour. What the case actually contains is a standing tension, consequential and well evidenced, between a legitimating promise and a distribution of power that strains the promise, together with real oppositions among organised interests that fight over the strain. Grade: not a contradiction under the test, and the grade is a service rather than a slight, because the tension analysis that replaces the slogan is more precise about mechanisms, about the channels through which money becomes influence and the countervailing channels through which numbers become influence, and precision about mechanisms is what a research programme, or a reform programme, can actually use. The grading exercise generalises: the test’s negative verdicts are among its most valuable outputs, since each one converts an incantation into a researchable structure.

Are contradictions necessary for change?

A question the long tail of inquiry presses constantly, and the glossary must answer it without either dogma or evasion, is whether change itself requires contradiction, so that a structure without internal contradiction would be changeless and the discovery of a contradiction is the discovery of change’s motor. The strong version of the claim descends from Hegel through the codifications, where contradiction is the root of all movement and self movement, and it should be rejected in its strong form on plain empirical grounds. Structures change under external pressure constantly: climates shift and agricultural orders built over centuries adapt or perish without any internal pole having constituted their undoing; technologies arrive from outside a given relation and reprice everything within it; conquerors, plagues and accidents write history in a hand no immanent analysis dictated. A method that must redescribe every external shock as the working out of an internal contradiction in order to save its thesis has purchased consistency at the price of content, and the redescription is always available, which is exactly the problem. The weak version of the claim is different and defensible: that the changes a social structure undergoes through its own normal operation, the changes it generates by being faithfully itself, are unintelligible without its internal contradictions, since faithful operation is precisely what the poles’ pressure describes. On the weak version, contradiction is necessary not for change as such but for a specific and theoretically privileged kind of change, endogenous transformation, and the analyst who claims that a transformation was endogenous owes the contradiction analysis that the claim implies. Stated with that scope, the necessity claim becomes a research discipline rather than a metaphysical banner, and it directs attention exactly where this series habitually directs it: to the difference between what happens to a structure and what a structure does to itself.

The companion question, whether contradictions guarantee the direction of change, has been answered in passing by the grammar of outcomes and is worth one direct paragraph. They do not. Pressure underdetermines direction because direction is set by organisation, by ideas, by the state, by the international field and by the accidents of leadership and timing that overdetermination gathers into a conjuncture. The same certified contradiction, between capital and the labour it must both employ and cheapen, has issued historically in social democratic settlements, in authoritarian modernisations, in neoliberal offensives that shifted its costs onto new populations, and in long stagnations that transformed little beyond the distribution of insecurity. A method that claimed to derive the outcome from the contradiction would be refuted by that list. A method that claims to identify the pressure every outcome had to answer, and to explain why settlements keep reopening, is confirmed by the list’s common shape. That is the scale of claim the disciplined vocabulary supports, and it is not a small one: most social theory cannot say even that much with a straight face.

The word in the texts: an itinerary

Because the glossary standard requires the textual source of a term to be checked rather than assumed, a brief itinerary of the word’s principal habitats in Marx’s writing closes the analytical part of this article. The itinerary confirms the fourfold sorting from an independent direction: the uses cluster by text and by period, and the strong uses belong overwhelmingly to the mature economic work. In the writings of the eighteen forties the word does its loosest and its most rhetorical work. The Holy Family and The German Ideology wield contradiction polemically, against opponents whose systems are shown to contradict themselves, a usage closer to the logical sense than to anything in Capital, since the point is that a doctrine has asserted inconsistently. The Poverty of Philosophy, for all its immaturity, contains an early structural use against Proudhon, whose attempt to keep the good side of every economic category while discarding the bad side Marx answers with the claim that the categories are the theoretical expression of relations whose sides belong together, an anticipation of the constitution doctrine in polemical dress. The Communist Manifesto elevates contradiction into historical narrative: bourgeois society is presented as a sorcerer unable to control the powers it has summoned, and the crises of overproduction are described as the absurdity, the word is close to the text’s own, of an epidemic of too much wealth. The Manifesto’s usage is the second register raised to world history, prophetic in cadence and constitutive in content, and readers who meet the term there first should be warned that the pamphlet is the doctrine in its most compressed and least argued form.

The Grundrisse of 1857 and 1858 is the term’s laboratory. In those notebooks Marx works out the contradiction between production and exchange, between the drive to develop the productive forces without limit and the narrow basis of consumption relations, between labour time as the measure of value and a production process that progressively reduces the labour time embodied in each product, the celebrated fragment on machines in which capital is described as working on its own dissolution through the very development it compels. Notebook prose is not finished doctrine, and the glossary discipline grades it accordingly: the formulations are exploratory, some abandoned, some carried into Capital in tightened form, but the constitutive structure is present in unmistakable outline, and any account of the concept that begins and ends with Capital’s published chapters misses the workshop in which the tool was forged. Capital itself distributes the four uses with the unevenness already mapped. Its first chapter works the form register at full stretch, value and use value, abstract and concrete labour, the fetishism of the form. Its chapters on the working day and on machinery work the positional and tendency registers, and its third volume, as noted, settles for long stretches into the honest counterpoise of real opposition, tendency against counteracting influence, with the contradiction vocabulary thinning exactly where the analysis becomes a balance sheet of forces. Theories of Surplus Value, the least read of the major works, contains some of the most instructive uses of all, because there Marx tests his predecessors’ contradictions as diagnostic instruments: Ricardo’s system is shown breaking down at the points where its categories cannot accommodate the relation between capital and labour, and Smith is shown oscillating between two incompatible determinations of value, so that contradiction in another thinker’s system becomes evidence about the structure the system was trying and failing to grasp. That reflexive use, contradiction as a symptom in a text that betrays a contradiction in the world, is a fifth, methodological usage that the fourfold table deliberately folds into its neighbours, since a glossary must sort by the thing named and not by the use made of the naming. The itinerary’s net verdict matches the audit conducted earlier. The vocabulary has a mature core where it is exact, a polemical penumbra where it is emphatic, and a notebook workshop where its exactness was earned. Criticism that samples only the penumbra refutes a pamphleteer. Criticism that samples the core must do the work Colletti did, which is why his essay, and not the pamphlet level objections, remains the reference point.

Unity of opposites, negation, and the limits of the imagery

Two companion phrases cling to the headword and generate a disproportionate share of its mystique, and a precision article must price them. The unity of opposites, elevated in Lenin’s Philosophical Notebooks into the very essence of dialectics, the splitting of a single whole and the cognition of its contradictory parts, states in imagistic form the constitution doctrine this article has defended in analytical form: a whole whose parts are what they are through their opposition, so that knowing the whole means knowing the opposition as internal rather than as an accident between finished things. Priced soberly, the phrase contributes the reminder and nothing more. It does not license the inference, common in the codifications, that every unity must contain exactly two opposites, tidily paired, when real structures present triads, networks and asymmetries that the pairing habit forces into costume. Nor does it license the cosmological extension, in which magnets and chemical affinities are enlisted as instances of the same law that governs classes, since physical poles do not constitute social positions and the analogy, pressed, becomes the fourth use wearing the first use’s uniform. Negation, the second companion phrase, requires the same pricing. In Hegel’s machinery a determination is negated by its opposite and the negation is itself negated into a richer unity, and Marx’s economic work contains structures that fit the pattern loosely, money negating the barter form’s limits, machinery negating manufacture’s craft basis, cooperation negating the isolated labourer. Fitted loosely is the operative phrase. The pattern is a narrative template that can be laid over almost any three stage history, which is why the triad of thesis, antithesis and synthesis, a formula Marx never used as a method and which belongs to the textbooks rather than to the texts, has done such damage to the tradition’s reputation among careful readers. The series treats that myth in its own article, and the present entry’s ruling is brief: wherever a negation story is told, the glossary questions apply unchanged, name the poles, state the unity, run the removal experiment, and let the imagery go if the anatomy fails. A small logical frontier should also be marked for completeness. A minority position in modern logic, associated with Graham Priest’s defence of dialetheism, the view that some contradictions are true, has claimed Hegel and occasionally Marx as ancestors for a paraconsistent logic in which a proposition and its negation can both hold without explosion into triviality. The claim is contested even among its sympathisers, and nothing in Marx’s economic analysis requires or suggests it, since the analyses proceed by distinguishing aspects, levels and tendencies in a manner entirely compatible with classical logic. Readers who meet the dialetheist appropriation should treat it as an interesting annex built by others, not as a load bearing wall of the house under inspection.

The imagery question has a practical corollary that completes this article’s analytical work. Metaphors in a technical vocabulary are scaffolding: legitimate while the building rises, and an obstruction when they are left standing and mistaken for the building. Unity, struggle, negation, motor, engine: the contradiction literature is scaffolded throughout with such figures, and the figures are not lies, since each gestures at a real feature of the constitutive structure, the belonging together, the pressure, the transformation. The glossary discipline is not to banish the scaffolding but to keep the accounts straight, so that at any point in an argument the writer can say which parts of a claim are figure and which are structure, and a reader can demand the conversion. Where the conversion succeeds, the figure retires gracefully and the structure stands in plain words: poles, unity, constitution, pressure, outcomes in the plural. Where the conversion fails, the figure was the whole claim, and the claim should be retired with it. With that rule in hand, the four uses sorted, the two great objections heard and partly sustained, and the outcomes grammar fixed, the vocabulary of contradiction is in the best condition a glossary can leave a vocabulary: smaller than its legend, larger than its obituary, and ready for work.

Worked corrections: five sentences repaired

Precision is best taught on specimens, and this closing analytical section takes five sentences of the kind that circulate in essays, broadcasts and textbooks, and repairs each one by the method of the article. The first specimen runs: the multinational contradicts itself by preaching sustainability while expanding extraction. Analysis: the poles proposed are a public commitment and a commercial practice, and their opposition is a gap between profession and conduct, the moralising misreading corrected in an earlier section. No unity constituting both poles has been named, since the commitment could be withdrawn tomorrow by a press office while the extraction continued unaltered in kind, and the practice could be sold to another operator while the preaching continued. The removal experiment therefore dissolves the alleged contradiction into a charge of hypocrisy, which may well be justified and is a different charge, assessable by evidence about sincerity and effect rather than by structural analysis. Repaired, the sentence becomes either a hypocrisy claim in plain words or, if the writer can show that the firm’s licence to operate depends constitutively on the credibility of the commitment while its returns depend on the extraction, the beginning of a genuine form against content analysis, in which the green presentation is a necessary appearance of the relation between the firm and its regulators and consumers. The repair costs a paragraph and buys an argument.

The second specimen: the contradiction between town and country is as old as civilisation. Analysis: two populations with different occupations and interests are named, and an antiquity is claimed, but no single relation constituting both as poles has been specified, and through most of the history invoked the town and the countryside were related by tribute, trade and migration among parties each perfectly capable of existing in the other’s absence, as self sufficient rural worlds without cities and garrison towns without agricultural hinterlands of their own both attest. What the sentence gestures at becomes analysis only when a specific unity is supplied: under capitalist agriculture, for instance, where the countryside is reorganised as a department of urban capital, supplying food, labour and raw materials under prices and credit terms set in the city, while the city’s reproduction depends daily on the countryside it subordinates, the poles do begin to constitute one another, and the pressure, the drain of labour and value against the dependence on supply, can be audited. Repaired, the sentence names its period and its mechanism, and shrinks its claim to the size of its evidence, which is what repair means.

The third specimen: her argument contradicts itself, because she praises the market in chapter two and condemns its outcomes in chapter nine. Analysis: this is the logical sense doing legitimate work in ordinary English, and the only question is whether the alleged inconsistency survives the distinction of respects. Praising a mechanism for its allocative properties and condemning the distribution it produces assert no proposition and its negation, since the respects differ, and the appearance of contradiction is manufactured by summarising two qualified judgements into two slogans. The specimen is included because its repair teaches the level discipline that the whole article depends on: before any contradiction, logical or dialectical, is announced, the respects must be matched, the level stated and the poles fixed, and most announced contradictions in intellectual controversy dissolve at exactly this step, leaving a disagreement about weights and values that could have been argued directly. The tradition’s own controversialists are not exempt from the lesson, and a reader who applies it even handedly will find Marx’s polemics, for all their force, occasionally manufacturing the same effect against opponents whose qualifications have been summarised away.

The fourth specimen: the gig economy is a contradiction, because the worker is both free and controlled. Analysis: beneath the slogan stands analysable material, and the repair is instructive because the sentence fails in its wording while its neighbourhood contains a pass. Free and controlled are predicates attached to one subject in different respects, contractual independence in respect of hours and acceptance, algorithmic direction in respect of pacing, pricing and discipline, and predicates in different respects contradict nothing. The constitutive structure nearby is the one graded in an earlier section: a platform constituted by a workforce it does not employ, and workers constituted in their market access by a platform they do not own, with the pressure arising as the platform cheapens and intensifies the labour it depends on while disclaiming the dependence. Repaired, the sentence abandons the both and formulation, which belongs to the rhetoric of paradox, and states poles, unity and pressure in declarative prose. The general rule the specimen teaches deserves explicit statement: any sentence of the form X is both A and not A is a promissory note, redeemable only by specifying the respects, and a writer who cannot specify them has written a riddle rather than an analysis.

The fifth specimen: the contradiction between the generations is deepening, as younger workers reject the settlement their elders accepted. Analysis: generations are cohorts, not poles of a relation; the young do not owe their existence as young workers to the old, nor the old to the young, and cohort succession is a biological and calendrical fact that no social relation constitutes. Yet the specimen, like the fourth, has analysable material in its neighbourhood, since inside pension systems, housing markets and public debt the cohorts are genuinely bound into single structures whose terms set their interests against one another in specifiable ways: a pay as you go pension scheme constitutes current contributors and current beneficiaries as poles of one transfer relation, each unintelligible as such without the other, and pressure arises from the scheme’s own arithmetic as ratios shift. Repaired, the sentence descends from the generations, a demographic atmosphere, to the scheme, an institution with poles that can be named and a pressure that can be computed. The five repairs share one moral, which is the article’s whole doctrine in miniature. The word contradiction fails upward or downward by specifiability: wherever the repair can name poles, a unity, a constitution and a pressure, the term is restored to work; wherever the repair ends with atmospheres, cohorts, values and slogans, the term is honourably discharged and its neighbours, conflict, tension, tradeoff and hypocrisy, take over a workload they can actually carry.

For students and exam candidates

For students and exam candidates, the four uses are the answer structure and the logical distinction is the mark carrying move. Open any answer on this topic by stating that a dialectical contradiction is not a logical contradiction, since tendencies and poles are not propositions, and you have already separated your script from the majority that define contradiction as opposing forces and stop there. Then deploy the four uses with one example each: opposed tendencies, productivity against labour dependence; opposed positions, the working day; form against content, the wage form; real opposition, the counteracting influences on the profit rate. Introduce the constitution test as your evaluative instrument, apply it to one strong case and one weak case, and concede, with the literature’s own audit, that much published usage fails the substitution test. That concession, precisely placed, reads as command of the material rather than as disloyalty to it. Practise the removal experiment on capital and wage labour until the institutional detail comes without hesitation, since examiners reward the candidate who can show a pole disappearing when its opposite is withdrawn, and that single demonstration distinguishes an answer about dialectics from an answer that merely recites its slogans.

For teachers and seminar leaders

For teachers and seminar leaders, the live problem is that the logic objection arrives in every philosophy class within the first ten minutes, usually from the strongest student, and it deserves an immediate answer rather than a postponement. The answer that works is concessive and taxonomic: grant that nothing in Marx licenses asserting a proposition and its negation, show Kant’s distinction between contradiction and real opposition on the board, and then set the class the removal experiment on capital and wage labour as a practical exercise, since students who have tried to describe a capitalist without wage labour, in institutional detail and without cheating, have learned the constitutive point in their hands rather than in their notes. Colletti’s essay makes an excellent seminar antagonist for advanced groups, precisely because it is written from inside the tradition and cannot be dismissed as incomprehension. Guard the session against the slide among the four uses by returning, whenever a contribution blurs them, to the table’s last column and asking which specific objection the contribution is answering.

For researchers, historians, and journalists

For researchers, historians, and journalists, the Colletti challenge is the reference point for any serious treatment, and the constitution test is offered as the working reply that concedes its taxonomy. Historians should treat contradiction claims in their sources as hypotheses about structure that require the poles, the unity and the pressure to be evidenced separately, and should be especially wary of retrospective certification, in which an outcome that occurred is used to prove that the contradiction said to produce it must have been deep. Journalists meet the word mostly in its degraded senses, as inconsistency or as looming breakdown, and the discipline of this article applies with full force to the newswire register: before printing that an institution faces a contradiction, establish whether the institution faces opposed poles that constitute it, or merely a tradeoff, a scandal, or a bad quarter, and let the noun in the copy reflect the finding. Researchers designing new work will find the six check protocol a serviceable referee, since a contradiction claim that survives it specifies its own evidence, and one that fails it specifies, in the manner of its failure, the more accurate concept under which the work should proceed.

Frequently Asked Questions

Q: What does contradiction mean in Marxism?

In its strong sense it names a relation between two poles inside a single structure, where each pole exists only through the other and their mutual pressure drives change in the structure itself. Capital and wage labour are the standard case: neither class can be defined without the relation that joins them, and their opposed claims over the working day and the product arise from that relation rather than from anyone’s temperament. In weaker senses the word covers opposed tendencies within one process, a social form that misrepresents its content, and forces in simple counterpoise. Because the senses differ, a careful writer states which one is meant, and a careful reader asks. Where no poles, no unity and no mutual dependence can be shown, the honest words are conflict, tension or tradeoff, and using them costs an argument nothing it actually possessed.

Q: Is a dialectical contradiction a logical contradiction?

No, and the distinction is the foundation of everything else on this topic. A logical contradiction asserts a proposition together with its negation, in the same respect at the same time, and such an assertion cannot be true. A dialectical contradiction asserts no proposition at all. It describes a structure in the world: opposed tendencies within one process, or two positions constituted by a single relation, pressing against one another. Capital’s need to raise productivity and its need to employ labour are both real, and stating both involves no logical offence, because tendencies are not sentences and cannot contradict one another in the logician’s sense. Confusion arises because one English word serves both discussions, and because loose summaries compress aspect distinctions into slogans of the form A is not A, which neither Marx’s analysis nor sound method requires anyone to assert.

Q: What are examples of contradictions in capitalism?

The disciplined examples are those that pass the constitution test. Capital must economise on labour through machinery while depending on living labour as the source of surplus value, a tension inside one process. Capital and wage labour contest the length of the working day and the division of the product from positions a single relation creates. The wage form presents all labour as paid while the working day divides into paid and unpaid portions. Production is organised socially across vast cooperative networks while its product is appropriated privately. Each example names poles, a unity and a pressure. Loose lists that add any unwelcome feature of the system, from advertising to traffic congestion, without that anatomy are not examples of the concept but of its decay into a synonym for problem, and they weaken every strong example they accompany.

Q: What is the difference between contradiction and conflict?

Conflict is the wider and weaker notion: opposed aims or claims among agents, each of which could exist and flourish without the other. Armies in a war, firms bidding for a contract and neighbours disputing a boundary are in conflict, and the removal of one party leaves the other intact. Contradiction in the strong Marxist sense adds two conditions that conflict does not require. The opposed poles must belong to a single unity that constitutes them both, so that removing one pole abolishes or transforms the other, as removing wage labour abolishes the capitalist as such. And their opposition must arise from the normal operation of that shared relation, regenerating after every settlement, rather than from a contingent clash of purposes. Many situations described as contradictions are conflicts, and calling them conflicts loses nothing. The narrower word is reserved for the cases where separation or victory would dissolve both terms of the opposition.

Q: What is a real opposition?

Real opposition is Kant’s category, from his 1763 essay on negative magnitudes, for the relation between two real and positive grounds whose effects cancel or outweigh one another. Two equal forces pressing a body from opposite sides, a credit balanced against a debt, a pleasure offset by an equal displeasure: nothing impossible is asserted, and the outcome, rest or a measurable resultant, is entirely determinate. Kant set real opposition against logical contradiction, whose outcome is not a resultant but sheer nothing, the unthinkability of a predicate both affirmed and denied. The category matters for Marxism because a large part of what Marx analyses consists of real oppositions, tendencies and counteracting influences in counterpoise, most famously in the chapters on the rate of profit. Colletti’s critique argues that Marxist writing conflates real opposition with Hegelian contradiction, and grading any example begins with asking which of Kant’s categories it truly inhabits.

Q: Are contradictions necessary for change?

Only for one kind of change, and the qualification is the answer. Structures change from outside constantly, under climate shifts, invasions, inventions arriving from elsewhere and plain accident, and none of that requires internal contradiction. What requires contradiction is endogenous change, the transformation a structure generates by operating faithfully according to its own rules: capital cheapening the labour it depends on, a wage relation regenerating the dispute it settles. The strong claim that contradiction is the root of all movement, inherited from Hegel through the codifications, should be rejected as metaphysics, since it can redescribe any external shock as internally ordained and therefore excludes nothing. The modest claim survives every test: to show that a change came from within a structure, exhibit the constitutive poles whose normal pressure produced it. Necessity belongs to the explanation of a kind of change, not to change itself.

Q: What is the principal contradiction?

Within a situation containing many contradictions, the principal contradiction is the one whose development conditions the development of the others, so that its transformation would alter the whole field. The notion is systematised in Mao’s 1937 essay On Contradiction, where identifying the principal contradiction is presented as the first task of analysis and of strategy, since effort concentrated on a secondary opposition leaves the structure that generates the situation untouched. The concept is a ranking claim and should be handled as one: it must be argued from the structure of the case, it can be mistaken, and its history includes use as a label conferred by authority in advance of analysis, which converts it from a finding into an instruction. Ranking also never abolishes the subordinate contradictions, since the doctrine of overdetermination holds that the subordinate ones often decide when and how the principal one becomes practical history.

Q: Can a contradiction be resolved without revolution?

Yes, in several distinct senses that should not be merged. Some contradictions are non antagonistic, arising between poles that share a fundamental interest, and the tradition itself holds that these can be resolved inside the existing unity by criticism, adjustment and institutional reform. Others are managed rather than resolved: factory legislation reconstituted the working day contradiction within legal frontiers, and collective bargaining and welfare institutions metabolise class pressure continuously without abolishing the relation that generates it. Still others are displaced, their costs exported to other populations or to the natural world, which quiets the home structure without touching its constitution. Resolution in the strict sense, where the poles cease to exist in their constituted form, is the rarest outcome and the only one that requires a transformation of the underlying relation. Much sterile argument dissolves once the parties state which outcome, resolution, management, displacement or adaptation, a given claim is actually about.

Q: Why do critics say contradiction is a misused word?

Critics press three counts. The semantic count holds that the word already has a precise logical meaning, and that borrowing it for tendencies and interests creates an equivocation dialectical writing exploits, shuttling between senses as objections approach. The methodological count holds that a term covering tendencies, interests, appearances and forces can be stretched to absorb any unwelcome fact, so that failed predictions are redescribed as contradictions displaced or deepened and the theory insures itself against refutation. The deflationary count proposes a substitution test: replace contradiction with conflicting tendencies and see what is lost, arguing that in most published usage nothing is lost except an air of profundity. Serious defenders concede much of this. The word is ornamental in a majority of its occurrences, and its prestige is borrowed from a constitutive core where substitution genuinely fails. The misuse charge is therefore best read as a grading instrument rather than as a refutation of every use.

Q: What is an antagonistic contradiction?

An antagonistic contradiction is one whose poles cannot both persist within the existing unity, so that the opposition tends toward the unity’s rupture and the transformation of the poles themselves. The contradiction between a slave owning class and the enslaved, or between a colonial power and a colonised population organised for independence, is the type case: no settlement internal to the relation satisfies both poles’ constitutive claims. Mao’s development of the doctrine pairs it with the non antagonistic contradiction, between poles sharing a fundamental interest, which can be handled by discussion and adjustment without rupture. Two cautions belong to the glossary. The classification is a claim about a structure, to be evidenced like any other, and not a mood about how fierce a dispute feels. And its political history includes classification by authority, in which labelling a dispute antagonistic licensed the treatment that followed, so the term should always be asked for its evidence about poles and unity before its consequences are discussed.

Q: How does a Marxist contradiction differ from a logical contradiction?

The two differ in subject, in test and in consequence. The subject of a logical contradiction is a set of propositions: a claim and its negation, asserted in the same respect, which cannot jointly be true and which logic handles by rejection of the set. The subject of a Marxist contradiction, in the disciplined usage, is a set of relations: poles constituted by one unity, whose pressure logic alone cannot assess, because relations are not true or false but stable or unstable, reproductive or self undermining. The test differs accordingly. A logical contradiction is detected by inspection of meanings. A Marxist one is detected by the removal experiment, asking whether each pole survives the other’s abolition in institutional fact. The consequences differ most of all. A logical contradiction must be removed before reasoning can proceed. A Marxist contradiction can persist for centuries as the normal working condition of an order, managed, displaced and adapted to, and its persistence refutes nothing in the analysis that identified it.

Q: What should a researcher check before calling two trends a contradiction?

Six checks, in order. Name the two poles in plain institutional language, without the contested word doing work in the naming. State the single relation or process that contains them, and what would count as standing outside it. Run the removal experiment on each pole in detail, recording whether the dependence is mutual, one sided or absent. Exhibit the pressure each pole exerts on the unity’s own conditions by operating normally, distinguishing structural strain from friction caused by incompetence or scarcity. Declare which of the four uses is in play, tendencies, positions, form against content or real opposition, and forfeit inferences belonging to other uses. Finally, perform the substitution test, replacing the word with its plainest paraphrase and keeping it only where the paraphrase loses constitutive or regenerative content. A pair passing all six checks has earned the term. A pair failing one has located its own more accurate description, which is the outcome good method should welcome.

Q: Why is the capital and wage labour relation called constitutive rather than external?

Because each term of the relation is what it is only through the other, which is the mark of an internal or constitutive relation. A capitalist, in Marx’s technical usage, is not simply a rich person or an owner of assets. The capitalist is the owner of money and means of production who purchases labour power and sets it to work for the expansion of value, a figure who cannot be specified at all in a world without sellers of labour power. The wage labourer is correspondingly not simply a poor person or a worker. The wage labourer is the person separated from independent access to means of production and therefore obliged to sell the capacity to work, a figure constituted by the same relation from the other side. Remove either pole and the other does not persist minus a partner; the proprietor working his own assets and the dismissed worker facing a locked factory are different social figures altogether. Buyer and seller at a market stall, by contrast, meet externally and part unchanged in kind.

Q: How does overdetermination complicate a single-contradiction account?

Overdetermination, a term Althusser adapted from Freud, holds that every contradiction in a social formation exists inside a conjuncture of other contradictions that condition its effects, displacing some pressures, condensing others and setting the tempo at which any of them becomes historical. The Russian revolution is the standing example: rupture arrived not where the capital and labour contradiction was most developed but where national, agrarian, military and dynastic contradictions had accumulated at a weak link. The complication falls on inference rather than on inventory. A certified contradiction between two poles states a real structural pressure and states nothing, by itself, about that pressure’s dominance, visibility or imminence, since dominance is settled by the whole network at a given moment. Single contradiction accounts fail not because the pair they name is unreal but because they read timetables off a structure that supplies none. The remedy is conjunctural analysis: each contradiction tested on its own, their interactions shown, and the fusion at any rupture reconstructed chain by chain.

Q: What separates a contradiction from a mere tradeoff or tension?

A tradeoff sets two goods or requirements against one another under scarcity, as when an aircraft must be both light and armoured or a service both generous and cheap to administer, and its terms are specifications chosen by designers or values held by decision makers. A tension is a felt strain within a system whose poles may not yet have been identified at all. A contradiction, in the disciplined sense, requires what neither of the others does: two poles of one relation that constitute each other and strain the relation by operating normally, so that the opposition regenerates after every settlement. The diagnostic is the removal experiment plus the regeneration test. Remove one term of a tradeoff and the other proceeds happily; settle a tension and it may never return. Remove one pole of a contradiction and the other pole ceases to exist in its analysed form, and settle its disputes as skilfully as you like: the relation reissues them, because issuing them is part of how the relation reproduces itself.