Who Are the Revisionists?

by Dimitris Moustakas, Nov. 16, 2026

Dimitris Moustakas argues that, contrary to being a dirty word, revisionism should be seen as core to the project of scientific socialism.

The use of the term ‘revisionism’ in Marxist circles has always annoyed me. It just doesn’t have a blanket negative connotation in normal use. I was somewhat familiar with the term ‘revisionist historiography’, and sometimes that was high-quality scholarship; sometimes it was low-quality scholarship, or even garbage. And in science, revising theories is an ordinary and necessary part of producing better knowledge. Most proposed revisions need not be correct for this to be true.

So I could never understand why I’m supposed to react negatively to hearing that someone is a revisionist. Whenever I pointed this out, I got a reaction that amounted to ‘well, you know what we mean by that’, and that’s where the discussion ended. But, frankly, I never knew what they meant by that.

So in this article I plan to take my revenge upon all that incomprehensible emotivist shouting and declare that:

Revisionism is good.

And not only that. I’m going to do what all these people—and a surprisingly large part of radical theory—seem to hate: tell you exactly what I mean by it. The horror.

Now, I’m using revisionism in a straightforward descriptive sense. Revisionism is the revision of an established body of theory, doctrine, or interpretation. A revisionist argues that we should change or drop something we currently accept. Scientists do this all the time: a revision may concern a small detail, but what looks like a minor empirical problem can also force them to rebuild much of a theory. A tradition that calls itself scientific has to be prepared to do the same. A proposed revision can also turn out to be completely wrong, but all of that is business as usual for a science.

Marxists have nevertheless managed to turn revisionist into an insult.

The precise content of the accusation has varied enormously. Bernstein was a revisionist. Kautsky was eventually a revisionist—or, for some, he always was but just hid it very well. Khrushchev and Tito were revisionists. And for Maoists, well, even the chair I’m sitting on right now is technically a revisionist. Who counts as a revisionist depends on which Marxists you ask, and they often include each other. Even the subject of the accusation changes: it might be about a strategy or an economic theory, a philosophical disagreement, what someone thinks of a socialist state, or which faction they belong to.

I’m not going to settle those disputes or defend everyone accused of revisionism. Some were wrong, some were politically awful, and some abandoned positions I consider essential. In general, I don’t particularly care for these discussions, but I would like to know what the accusation is supposed to establish. What was their sin: proposing a false theory, giving up on socialism, or disagreeing with somebody they weren’t supposed to disagree with? Calling all of this revisionism mixes scientific, philosophical, and political questions without explaining what was wrong with someone’s position. We still have to work out whether the proposed revision is true, politically desirable, or even remotely intelligent.

For the long-term prospects of our political project, I think a narrower question matters much more: what attitude should a political tradition which calls itself scientific take towards revisionism? Scientific socialism needs people who challenge the accepted theories of the broader movement, however fundamental those theories are—or are thought to be. They must be able to do that without having to defend their commitment to socialism for questioning something held sacred. Making this possible within a political movement that also needs to act together raises difficult questions about how it should be organised. The short answer, then, is that it should expect revisionism, allow it, and often welcome it.

Our political commitments and our efforts to understand society are connected, but we cannot answer an empirical question by appealing to our desire for socialism.

I came back to this question while reading Donald Parkinson’s recent article on the controversy over ‘Western Marxism’. He criticises a habit in Marxist debates: treating an argument as politically suspect instead of examining whether it holds up. In discussing Althusser, he notes the latter's complaint that the politics of the cult created a suspicion towards theoretical novelty which seriously damaged scientific research. Parkinson's concern is to defend an intellectual culture in which Marxists can pursue truth even when the result is politically inconvenient.

In pursuit of this broader goal, my concern here is with a category Marxism has retained which already contains much of this problem: revisionism. If changing an accepted proposition is itself politically or doctrinally suspect, independently of the reasons offered for changing it, then one has already introduced the wrong criterion into a scientific question.

But even though Parkinson's article motivated me to pick this question up again, my main inspiration came from another essay. Adam Schaff's 1971 ‘What Does It Mean to “Be a Marxist”?’ helped me clarify my thoughts on the matter. Don’t get me wrong, he’s not getting it right. On the contrary, he fails to answer the question in his title, along with several more interesting questions he raises along the way.

The essay is nevertheless useful because Schaff sees almost every problem that bothers me, and then demonstrates how difficult it is to solve any of them while retaining the category he is trying to save. He rejects a Marxism consisting of absolute fidelity to the classics. Theories become obsolete, and fear of being called a revisionist can deter inquiry; Schaff recognises both problems and repeatedly insists that a scientific Marxism must be open to change. And then he spends much of the essay trying to work out how much change is too much.

Schaff’s problem cannot be solved. It can only be dissolved. Most of the difficulty disappears once we stop treating agreement with a theoretical tradition as a virtue in scientific inquiry.

Schaff's Problem

Schaff begins from an obvious fact about Marxist polemics. “Marxist” and “revisionist,” he writes, are often deprived of their scientific meaning and instead function as “emotive descriptions for immediate targets in the political struggle.” His project is, at least partly, an attempt to rescue the terms from this usage.

Schaff’s definition has a subjective component: one must consciously identify with Marxism. Its objective component requires knowing the views of Marx, Engels and the other classics, and upholding them except where subsequent developments in science and society have invalidated or modified them.

I have several problems with this definition, but there is a difficulty that should be immediately visible.

If Marxism is a science, why should the point of departure be fidelity to the views of particular people? Scientists obviously have intellectual traditions. They inherit concepts and questions, work within research programmes and often identify with broad theoretical schools. They also read important predecessors, though not as often as people seem to think. I see no problem with any of that. What I object to is making membership in a scientific community depend on the continued correctness of particular propositions.

Schaff initially calls a revisionist a Marxist who ‘advocates certain appropriately qualified alterations in Marxist theory’. He accepts that Marxist theory must change when it conflicts with reality, so ‘appropriately qualified’ has to do the work of separating legitimate changes from illegitimate ones. In his account, a legitimate change responds to such a conflict; an illegitimate revision lacks that justification.

On that definition, revisionist is an unusually elaborate way of saying wrong. Suppose I argue that the evidence requires a substantial revision and some other guy thinks the existing theory still holds up. If we investigate and discover I was wrong, I’ve made a scientific error. I don’t see what we gain by attaching an emotionally charged label to it.

Realising the issue, Schaff insists that error must not be confused with revisionism, because “there can be no search for truth without the risk of error.” He even points out that fear of the accusation can discourage creative inquiry. So a proposed revision can fail his test without making someone a revisionist. He has, by now, painted himself into a corner as every qualification has to be qualified. Now, he says, whether the accusation applies depends on the person’s “system of views regarding some major part of Marxism.” He wants us to consider a person’s views as a whole, rather than condemn an isolated mistake on an insignificant point. That is a non-solution in my view. To distinguish revisionism from an ordinary scientific error, he now has to decide which parts of Marxism are fundamental and explain why mistakes about those deserve an extra accusation.

If genuine inquiry requires the freedom to make mistakes, we cannot reserve that freedom only for minor questions. If one knows that getting a fundamental question wrong may threaten his standing as a Marxist, he has a reason to leave it alone and work on something where a mistake will remain an ordinary error. The more important the question, the more he has to worry about those consequences. Schaff’s qualification leaves people afraid to investigate the questions that could matter most, despite his own warning about what that fear does to inquiry.

Later in the essay, he tries to establish the limits beyond which changes would amount to abandoning Marxism altogether. He repeatedly insists that Marxism is open to revision, including revision of its fundamental theories. Anyone seriously investigating an issue has the right to propose changes, and no individual or institution enjoys special authority to decide the matter. But there is a limit: modifications must be distinguished from the rejection of the basic theories that constitute Marxism as a system. Someone who rejects those theories can no longer claim to be a Marxist.

There is a perfectly intelligible question here about the identity of an intellectual tradition. If someone rejects Marx’s political economy, his account of historical development, his philosophy, and most of his political conclusions, it may be a stretch to call their views Marxist. But this tells us only how those views relate to Marx and the ‘classics’, not whether they are correct. Schaff eventually draws the same distinction. He says that his discussion of these boundaries has set aside the question of truth and falsehood, and insists that “science knows no taboos whatsoever.” Even a fundamental proposition must be criticised if it conflicts with scientific truth. For Schaff, rejecting certain fundamental propositions means ceasing to be a Marxist, even if there are good scientific reasons to reject them.

But Schaff also wants to preserve revisionism as an accusation within a political movement that claims to be scientific. Deciding who has the right to call themselves a Marxist has consequences for people in that movement. A researcher may be free on paper to reject any proposition and still have to worry about whether they will be accepted as a Marxist afterwards or face blanket rejection for abandoning this or that ‘fundamental’ proposition. Saying that science has no taboos does not remove that pressure when the movement makes theoretical fidelity a condition of belonging.

Schaff’s boundaries may tell us what to call our views, but they cannot give us a scientific reason to keep a proposition we have good reasons to reject. We should reject it even if that means abandoning Marxism as a distinct theoretical system by Schaff’s definition. And if rejecting false claims eventually leaves nothing recognisably Marxist, preserving the name is no reason to keep those claims.

Someone who proposes a mistaken revision is open to scientific criticism, but challenging a proposition central to Marxism does not justify adding a separate accusation. We cannot tell researchers that whatever they discover must leave us with something recognisably Marxist. Reality has no obligation to preserve Marxism.

What Is a Revision?

The obvious objection is that I’m being deliberately annoying. How do you make this definition fit what Marxists have historically meant by revisionism? That’s the neat part: you don’t. The historical use of the word is part of what I’m objecting to.

The word has been used to collapse several different kinds of disagreement which should be kept apart.

First, there is doctrinal revision. Marx said X, and somebody says not-X, or modifies X enough that the result is no longer recognisably Marx’s position.

Second, there is scientific revision. A proposition currently accepted by some scientific community, Marxist or otherwise, is challenged because somebody believes a different account better explains the evidence.

Third, there is political revision. A person or organisation changes its programme, strategic commitments, political objectives, or conception of socialism.

Of course, a disagreement can involve more than one of these kinds of revision. If Marx made a claim about capitalist concentration, challenging it can be both a doctrinal and a scientific revision. But knowing that someone has departed from Marx’s position still tells us nothing about whether they are wrong about concentration.

We can see the problem already in the Bernstein controversy, where claims about capitalist development and arguments about socialist strategy came under the same heading of ‘revisionism’. He argued that capitalist concentration was proceeding unevenly and that some expectations about its development had been overstated. We can investigate whether he was right about that; pointing out that he departed from Marx’s account does not answer the question. He also argued that social democracy should organise workers and pursue democratic reforms rather than base its tactics on an expected economic collapse. These arguments were connected, as we would expect: what we believe about the world affects what we think we should do. But we still need to examine whether his strategic conclusion follows from his premises and whether the proposed strategy is any good.

Abandoning socialism altogether would be a different matter again. If someone does give up on socialism, they are no longer a socialist, but that judgement concerns their political commitments. It does not answer the empirical or strategic questions we were asking. These questions can arise together in someone’s intellectual development without becoming the same question, and shouting revisionist does nothing to help us answer any of them.

The scientific case is particularly simple. If a revisionist is someone who proposes revising an accepted scientific claim, then revisionism is normal. A scientific tradition that does not generate revisionists is dead.

The Right to Be Wrong

Welcoming revisionism does not mean assuming that revisionists are generally right. A person arriving with a new theory still faces a burden: established theories usually explain something, have held up against a body of evidence, fit with other things we know, or at least prove more successful than the alternatives proposed so far. But having to support a proposed revision is not the same thing as having to justify the act of proposing it.

The evidence supporting an existing theory does not rule out discovering that some part of it is wrong. But allowing it to be challenged also means allowing revisions that turn out to be mistaken, since the challenge has to be made before its merits can be assessed. That is why Schaff is right that the possibility of error is essential. One cannot tell researchers: you are free to challenge our theory, provided you turn out to be right.

That freedom has to extend to fundamental questions too. If a community tolerates mistakes about peripheral questions but makes someone’s intellectual or political standing depend on accepting its fundamental propositions, it gives researchers a reason to avoid challenging those propositions. Anyone who thinks such a norm would have no effect on inquiry has an implausible theory of human beings.

The effects of such limits, stated or unstated, will be obvious. There is a big chance that researchers will avoid some questions altogether. And even when they investigate them, they have reasons to hedge politically inconvenient conclusions more heavily than the evidence warrants, or to explain away difficulties in accepted theory because rejecting it would threaten their standing. Challenging a central proposition therefore takes political courage as well as scientific work: researchers also have to bear the consequences the community attaches to disagreement.

You don’t need a censor or a bureaucrat hunting heresies to produce bad science; these incentives can do it quite effectively by themselves. Treating revisionism as a scientific vice does more than make us judge a dissenter unfairly after the fact: it changes the conditions in which claims are produced.

Have Some Standards

If agreement with Marx is not the standard, we are left with the need to find some way of distinguishing a serious revision from bullshit. Fortunately, other people have spent quite a lot of time thinking about how to do that. It is called science.

I do not think Marxists require a distinct Marxist epistemology or a unique method which separates us from ordinary mortals. On the contrary, claims made by socialists should generally be evaluated according to the standards of the relevant scientific disciplines. I realise that is perhaps a bit too boring for some people. But that’s life.

Some principles that follow from such a stance are the following: Arguments should be clear enough that we can identify their conclusions. Assumptions should be made explicit. Empirical claims should be answerable to evidence. Causal explanations should identify plausible mechanisms (that one is pretty hard to accept, I know). Models should make clear what they simplify. Quantitative claims should use appropriate statistical methods. Historical claims require historical evidence. Competing explanations should actually be compared rather than dismissed because of their political origin.

Of course, there is no method that can guarantee that a claim is true. But with the use of these standards we can improve our chances of getting it right and have public criteria for identifying and arguing about mistakes.

Jon Elster rejects the idea that a special “dialectical reason” separates Marxists from everyone else. Erik Olin Wright gives a concrete example of what ordinary scientific standards offer in his discussion of analytical Marxism. Defending the use of explicit models, Wright argues that formalisation forces the theorist to specify assumptions, parameters and mechanisms. Behind informal causal explanations, he argues, there is normally an implicit model anyway. The difference is the extent to which researchers are prepared to “put their cards on the table.”

And I think that’s a great epistemic standard to have: put your cards on the table. It helps your readers, and most of all yourself, answer the following questions: Which part is an empirical proposition and which part is a moral or political judgement? What would we expect to observe if the explanation were false? What rival explanations exist? How much confidence should we place in the conclusion? Under what conditions would you change your mind?

To pre-empt a counter-argument, I realise that not every question permits an answer to all of these or other sensible scientific questions. For example, historians cannot rerun past events under controlled conditions. Social systems can contain feedback mechanisms, selection effects, strategic interactions or a host of other issues which, together with problems of measurement, can make causal inference difficult. It is also true that most useful models make assumptions everyone knows are literally false in order to isolate mechanisms. It is also a fact that some evidence is inevitably observational and it can be the case that some important questions remain too poorly understood for strong conclusions.

None of this is devastating, but it does mean that serious scientific work has to make its uncertainties visible. Putting your cards on the table lets others assess what a model explains, rather than dismissing it simply because its assumptions are unrealistic.

To focus on a particular side of scientific work, modelling, it should be noted that leaving things out is often the point of a model. We simplify in order to isolate relations we want to understand. If you think a model leaves out something important, explain what difference this omission makes. It is not helpful to just point out that society is more complicated than the model. All that this accomplishes is to tell us something we already knew.

Kieran Healy discusses this problem, among others, in ‘Fuck Nuance’. What interests me here is his criticism of demands for more complexity. We can keep adding culture, institutions, power and whatever else is said to matter, without specifying how any of them affect the explanation. The result can be worse science: the ‘vocabulary’ of our theory can grow, but it becomes harder to work out what the theory claims or what would show it to be wrong. As Healy puts it, “We lose information by adding detail.”

Making our assumptions and claims explicit helps other people find mistakes in the argument. A good theory should make itself easy to attack. This sounds backwards only if theoretical sophistication is confused with the ability to survive criticism by becoming harder to specify. If nobody can work out exactly what your theory predicts, which evidence would bear against it, which causal mechanism it proposes, or even what the crucial terms mean, then its immunity to refutation is not a sign of depth. “If you are equally good at explaining any outcome, you have zero knowledge.”

There are Marxist traditions in which this is treated almost as a virtue. A simple objection is answered by invoking mediation, totality, contradiction, appearance and essence, or by accusing the critic of isolating one element of a dialectical whole. The explanatory content of these answers is left unspecified or, in many cases, is obviously missing. What is this “totality”, which relations matter, and how do they affect the conclusion? What process/mechanism have we identified by calling it “dialectical”? Calling a counterexample an “appearance” does nothing to explain it away, and calling an account a “false abstraction” does not tell us what it has left out or why that matters. The outcome is that we are left waiting for an argument clear enough to assess whether its reasoning holds up, let alone whether its claims are true.

Have some standards, for fuck’s sake.

“Mainstream” Science

By appealing to the standards of the relevant scientific disciplines, I primarily mean their methods. Apparently this needs explaining, because I keep being told about something called ‘bourgeois statistics’. Contrary to what people may hope for, there is no socialist probability theory. An objection to using statistical inference or game theory needs to explain what is wrong with the method or its application to the question at hand. Scientists argue about the details of these applications constantly, but calling a method ‘bourgeois’ avoids the argument and all it does is provide political cover for an unscientific dismissal.

The methods defended here can be criticised and improved, but someone proposing an alternative needs to explain why it would give us better knowledge. Claiming that Marxism has its own way of knowing does not answer that question (and unless some extraordinary evidence is presented, it cannot be taken as true). As David-Hillel Ruben puts it, “The difficulty in telling science what the world must be like is that the world may not turn out to be that way.”

Scientific results matter too. When researchers who know the evidence and the competing explanations subject one another’s work to criticism and converge on a conclusion, their agreement gives us a reason to take that conclusion seriously. That does not mean the current scientific establishment is infallible; a dissenter may turn out to be right. But the possibility that the experts are wrong is not itself a reason to conclude that they are. We can recognise the weight of their agreement while defending someone’s right to challenge it and investigate whether it holds up.

Scientific institutions can also create incentives to defend accepted views and discourage challenges. But mainstream science has repeatedly, contrary to many people’s expectations (including mine occasionally), demonstrated that it can identify errors, abandon established explanations and improve its own methods, even if this often takes far longer than we would like. That record gives us a reason to take it seriously despite its institutional failings. Evidence that those failings are distorting a field’s research should still affect our confidence in its conclusions. But we have to investigate the problem: calling a result ‘bourgeois’ does not establish what is wrong with it.

What bothers me is how selectively Marxists sometimes apply this scepticism. Conclusions inherited from our own tradition receive extraordinary deference, while conclusions from other fields meet extraordinary suspicion when they are politically inconvenient. If we are prepared to question an established scientific result, we should certainly be prepared to question something produced within the Marxist tradition. These conclusions have to be assessed by the same standards, including when doing so gives us reasons to abandon them.

Let the Theories Die

One consequence of all this is that some propositions traditionally associated with Marxism will not survive. And that’s okay. Take the labour theory of value for example. It is false.

I am not going to argue that claim at length here. I have little enthusiasm for revisiting the debate, and my conclusion does not depend on a single objection. Samuelson shows how competitive prices and the profit rate can be calculated from production conditions and real wages without first calculating labour values. Steedman constructs cases in which joint production (this means that one process produces several different commodities) allows positive prices and profits alongside negative surplus value.

There are also the explanatory successes of alternative theories of prices, including experimental markets in which prices converge towards the levels predicted by supply and demand. The arguments showing that exploitation can be understood without the theory also remove a reason for treating its rejection as politically disastrous. Taken together (this is of course hardly exhaustive, I just scratched the surface), these give me ample reason to regard the question as settled. Nor do I think a theory has been rescued when its defenders preserve the name by giving up its substantive explanatory claims.

What interests me here is what follows from rejecting it. Calling the rejection revisionist does nothing to defend the theory, because departing from a Marxist position and being wrong are different things.

Rejecting the labour theory of value does not mean abandoning the socialist critique of capitalism. Capitalist property relations, class power and the organisation of production remain things we need to understand and change. Shapiro and Stiglitz’s account of unemployment as a means of disciplining workers makes no use of labour values. Exploitation does not become unintelligible either. Cohen and Roemer, in different ways, explain why an account of exploitation need not depend on the labour theory of value. Particular arguments may have to be abandoned or reconstructed, but that does not dispose of the case for socialism.

There is also the broader point that making socialism depend upon the survival of a particular theory is a serious mistake. If an argument we have relied on fails, we need to work out what follows from that failure. We cannot preserve it because we are attached to the political conclusions we used it to support. Our reasons for those conclusions have to withstand criticism too. That may require changing more of our views than we initially expected, including views about what socialism should look like or how to achieve it.

The same can apply to theories of imperialism, socialist planning or class consciousness (the examples are semi-randomly selected). Whether any particular account should be retained, revised or abandoned has to be assessed on its merits. Its importance to the Marxist tradition cannot settle the question.

That importance does, however, give us a reason to examine it carefully. If other explanations or political strategies depend on a mistaken claim, protecting it allows the mistake to affect those conclusions too. The more we have built on a proposition, the more we need to know whether it holds up.

Why Scientific Socialism?

At this point the obvious Schaffian question returns. How much can we discard before scientific socialism ceases to be Marxism?

G. A. Cohen gave what I think is basically the right response in the introduction to the 2000 edition of Karl Marx's Theory of History. Cohen defines scientific socialism, “at least among other things,” as the study of the nature of and route to socialism using “the most advanced resources of social science” within the frame of a socialist commitment. He then regrets that scientific socialism came to be named after Marx and asks what a physicist would say if someone demanded to know whether she was “still a Galilean.” Physics progresses by contradicting many propositions held by its founders. If it could not do so, it would not progress.

Cohen recognises the differences between physics and social science. His analogy concerns the expectation that a science should preserve its founders’ conclusions. Developing the inquiry they began can require rejecting things they believed.

Yet Marxists routinely ask the analogous question. Does this abandon Marx? What did he say in his letter to ‘insert random name’ about this? Is modern geology in line with Marx on Russian soil? Would Lenin write three books calling you a renegade if you published your idea? Or, on a more serious note, at what point have we changed enough propositions that we are no longer Marxists?

I can concede that these questions can be important and/or interesting when we concern ourselves with issues about intellectual genealogy. But as questions about what is true of the world, they are worthless.

The conclusion of Cohen's argument can be a bit provocative: had the label scientific socialism stuck, we might have avoided some of this. Marxism's name makes a progressive body of inquiry sound more like a doctrine organised around a founder. I doubt the name is the main cause of the problem, but I think Cohen is right that it encourages this way of thinking and has certainly been a factor in the unfortunate intellectual tendencies of the tradition. If Marx contributed to a science of society, its development should make fidelity to his particular conclusions progressively less important. Marx’s work may still contain neglected ideas, questions others fail to ask, and theories worth reconstructing. But developing those contributions does not require preserving his conclusions when there are good reasons to reject them.

Scientific socialism is a better description of the project because it names its two commitments: pursuing socialism and trying to understand the world using the best scientific tools available. The political commitment does not establish scientific conclusions. Scientific inquiry does not by itself establish our ultimate political values, but what we learn may require us to revise a proposed socialist institution, reconsider how to achieve socialism, or abandon a conception of it that turns out to be impossible. Whatever follows politically, claims about the world still have to be assessed on their merits.

The Political Problem

Schaff is at his most interesting when he recognises why Marxists find this so difficult.

Marxism is not merely a collection of scientific propositions. It has historically functioned as the ideology of a political movement. But political movements, especially ones facing incredible odds against them, need commitment. Doubt is hardly what they want to encourage. They want members to believe that their actions matter, their goals are worth pursuing, their comrades can be trusted, and that sacrifice is justified. The expectation that getting to socialism will involve hard times and hard choices reinforces the perceived, or real, need for ‘ironclad’ beliefs. But scientific inquiry requires a willingness to question accepted theories, and its findings can themselves give us new reasons to doubt them. Even a strong scientific consensus can leave good reasons to investigate alternative views. That openness is difficult to reconcile with the demand for ironclad beliefs.

Schaff therefore observes that politicians will naturally be more cautious about ideological innovation than scientists. Changing principles, goals, and accepted beliefs may weaken the convictions that bind a movement together. He thinks this creates a real tension between theoretical innovation and political responsibility. I agree with him that this tension is real.

Wright makes much the same point: the certainty that sustains revolutionary commitment can conflict with the skcepticism scientific inquiry requires. Revolutionaries also need to learn from their mistakes, and scientists need strong commitments. But this does not remove “the tension between the intellectual rigidity encouraged by dogmatism and the intellectual flexibility required to learn new things.”

Where I disagree with Schaff is that his demand for political responsibility still puts pressure on scientific inquiry to accommodate the movement’s needs. In my opinion there is no politically responsible version of a false empirical claim. If a policy does not work, it will not begin working because admitting failure would demoralise the membership. Suppressing the criticism will tend to handicap the movement for much longer. If an economic theory is wrong, it remains wrong even if a correct analysis would provide ammunition to an opponent. If a socialist state is experiencing a crisis caused by its own institutions, a scientist does not get to move the causal coefficient because imperialist newspapers would enjoy the result. In my view, this has been devastating for the movement. It encourages unjustified confidence in policies and analyses whose failures can produce precisely the demoralisation we were trying to avoid.

Political work raises questions that we might otherwise neglect, and acting on a theory gives us evidence with which to assess it. But a theory’s usefulness to a political organisation does not establish that it is true.

Consider a party that has committed itself to a strategy and whose leadership has publicly defended the analysis behind it. Research supporting that analysis is welcome. Research undermining it creates a problem for both the strategy and the authority of the people defending it. If the theoretical position is also part of the group’s identity, criticism can easily be taken as an attack on the group itself. There is then a strong incentive to defend the position because of what abandoning it would mean politically, regardless of whether the criticism is justified.

Nobody has to issue instructions about what researchers are allowed to find; the expectation that they should exercise ‘political responsibility’ can do much of the work. Nor does an organisation need state power to create these pressures. A journal or a reading group can also make adherence to a theoretical position a condition of being taken seriously as a socialist. The consequences are different, but the problem remains: disagreement about a claim becomes grounds for questioning someone’s political reliability. A formal freedom to disagree does little to resolve this if exercising it is treated as evidence against you.

A movement that wants to make use of science has to be prepared for scientific work to challenge positions it currently defends. Its members must be able to investigate those positions and argue that they are wrong without having to defend their commitment to socialism every time they do so.

Theoretical Unity

This brings us to a larger organisational problem, which I can only touch on here.

Theoretical or ideological unity is routinely treated as a virtue in the socialist and communist movement. I think almost all interpretations of these terms are a huge mistake.

I am not disputing that political organisations require unity of some kind. If, for example, an organisation decides to run a candidate and half its members actively campaign against that candidate, it has a problem. But the way political and scientific questions are settled is very, very different. A congress can decide what an organisation will do. But it cannot decide what is true.

Everyone will agree on paper, but organisational practice in Marxist circles routinely ignores this distinction. Party congresses, even in organisations that claim to reject ideological unity, adopt documents containing detailed analyses of how capitalism or imperialism produces this or that outcome. These documents should establish what we are going to do. The underlying theoretical analyses should not be among the positions we are being asked to adopt: a policy can be right even when the explanation offered to justify it is wrong. Members who agree on what to do may disagree about why it will work. Requiring them to adopt the explanation as well makes that theoretical disagreement into a disagreement with the organisation’s position, even where it creates no obstacle to acting together. Besides, the bargaining involved in producing these documents can turn competing plausible analyses into an implausible or even internally inconsistent compromise.

That is one reason, among others, why I think socialists should aim principally at programmatic and not theoretical unity.

A programme establishes the political objectives around which people agree to act together. But members can arrive at those conclusions in wildly different ways. I’ve seen it again and again: unofficial factions with even antagonistic theoretical approaches agreeing on how to deal with the issue at hand. They may even disagree about why the policy they both support will work in the organisation’s favour.

Members must also be able to organise around disagreements that lead them to propose changes to the programme itself. They can work together to develop an alternative analysis, publish it and try to persuade the organisation to adopt it. Accepting a programme as the basis for common political work cannot mean accepting that its contents are beyond revision.

Allowing them to disagree individually while forbidding them to organise around that disagreement restricts their ability to develop and argue for an alternative. The demand for political unity still constrains theoretical inquiry, even if everyone formally retains the right to their own opinion.

This gives us a reason, directly connected to the scientific aspirations of the movement, to regard factions as a normal part of its life.

Revisionism

One way of answering the accusation of revisionism is to say that the dogmatists are the real revisionists. Marx and Engels understood that theories must change, so surely the people who refuse to change anything are the ones betraying their teaching. I understand the appeal of this answer, but it keeps revisionist as an insult and makes us argue over who deserves it. I don’t see why we should accept that starting point.

This also gets at much of my motivation for writing this article. I’m sick and tired of discussions about socialism turning into Marxology, Leninology, or an argument over whichever ‘classic’ happens to support someone’s position. Someone makes a claim about the world and ends up having to defend themselves by finding the right quotation. Their opponent finds another quotation, and now we’re discussing the correct interpretation of an approved text. Even the case for questioning an established theory ends up depending on finding a passage that gives us permission.

It is a theological way of arguing, and it gives people a completely wrong understanding of how science works. A quotation can bring a good argument into a discussion, but we still need to examine the argument and explain why it holds up. That applies just as much to the work of great scientists as to anyone else. Their standing comes from what they managed to explain and demonstrate. When we cite them, that is the work we should be drawing on. I want us to be able to discuss whether an argument is good without first having to establish that the right ‘classic’ would approve of it.

That is what I mean by saying revisionism is good. Scientific socialism needs people to challenge its accepted theories. Their arguments have to stand up to scrutiny, but they must be able to get things wrong without having to defend their commitment to socialism. If an argument does stand up, we have to be prepared to change our minds, including when that means giving up something we have defended for years.

Liked it? Take a second to support Cosmonaut on Patreon! At Cosmonaut Magazine we strive to create a culture of open debate and discussion. Please write to us at submissions@cosmonautmag.com if you have any criticism or commentary you would like to have published in our letters section.

About
Dimitris Moustakas –

One of many contributors writing for Cosmonaut Magazine.