Marxism and Anarchism

Both Marxists and anarchists perceive the state as an instrument of domination that must be abolished for a truly free society to exist.

In practice, however, the two movements have pursued very different strategies with respect to the state.  Marxists, in general, have tried to take control of the state and use it to implement one or another form of socialism[1], whereas anarchists attempt to operate outside of and in opposition to state authority at all times.

We can understand this difference in terms of all sorts of factors – values, opportunities, personal biography, and so on. As a historical materialist, I do not think people’s actions are so much driven by their ideas; our ideas usually are driven by our actions. But I find it interesting to understand this difference between Marxists and anarchists as a theoretical difference – indeed, as an epistemological difference.

Marx theorizes the state as an expression of class domination. As part of the superstructure of society, it has at best a limited autonomy from its base in class relations.[2]

In other words, the state has no or not much “life of its own”. It expresses the interests of the class that controls it.  Therefore if the working class can seize control of it, the state can be made to express the interests of workers and to dominate the capitalists — until capital and class difference can be abolished entirely, leaving the state with no further social function.  This domination of capital by labour, prior to the achievement of communism properly so called, is what Marx means by the ‘dictatorship of the proletariat’[3].

I am not well read in anarchist theory, but my understanding is it disagrees with Marx on precisely this point. According to anarchists, the state is a force unto itself.  It does have a life of its own. As much as the state collaborates with class domination it does not reduce to class domination, even in the last instance.

This implies that a dictatorship of the proletariat is impossible.  Any group that seizes control of the state will find itself constrained by the logic of state power, in the same way that a capitalist who decides to run his business benevolently is still constrained by the logic of profit and loss.

For social anarchists, this means (a) that one must abolish the state at the same time, with the same revolutionary gesture, that one abolishes capital, and (b) the revolutionary movement must develop a new mode of politics that does not depend on or reproduce within itself the authority of the state.

One can understand the intense, often bitter feuds between Marxists and anarchists as a conflict over revolutionary strategy defined by this elementary theoretical difference.

[1] In case it needs to be said, this is far from being the only means by which Marxists have pursued revolutionary change.

[2] Marx expresses this view directly in The German Ideology and The Manifesto of the Communist Party, and it animates his analysis throughout The 18th Brumaire of Louis Bonaparte.

[3]  Such a ‘dictatorship’ would be more democratic than the status quo. Marx observed that “The executive of the modern state is but a committee for managing the common affairs of the whole bourgeoisie”. He wrote that in 1848, long before the enfranchisement of working people, when those words were literally true even by liberal standards.  Still, at best today the liberal-democratic state is a battleground where labour and capital fight for control, and on which capital has the more advantageous position.

13 thoughts on “Marxism and Anarchism

  1. I think you’re right. Here’s Saul Newman (“Anarchism, Poststructuralism, and the Future of Radical Politics,” SubStance #113, Vol. 36, no. 2, 2007, p. 13): “The central contribution of anarchism to radical political thought lies in its rejection of the state and all authoritarian forms of politics, its critique of Marxism, and its commitment to a libertarian and egalitarian ethos. In particular, the innovativeness of anarchism lies in its theorization of political power—namely the power of the state—as an autonomous field of power relations and a specific site of political struggles that was analytically separate from, and not determined by, the capitalist economy or class relations. This was why the state could not be trusted to “wither away” after the revolution: as an abstract machine of domination that had its own logic and rationality, it would only perpetuate itself through the guise of the workers’ state. Therefore, by breaking the absolute structural link that Marxism had established between the political and the economic, anarchism performed a vital theoretical operation—one that foreshadowed later poststructuralist and post-Marxist interventions.”


    • Thanks! Yes, I don’t think I’m saying anything too original. I stumbled on another post that makes a similar point, here: although that post turns into a polemic on the side of Marxism. Thanks for the Saul Newman reference; I’m going to look that up.

      I find it interesting to think of this in terms of epistemic strategies: given complex information about a domain of phenomena, e.g. human social life, what framing assumptions do we make? Do we assume holism, that it all fits together somehow? Or a single dialectic behind everything, as in Marxism? Or multiple dialectics? Or no overarching systems or dialectics at all, just a bunch of local institutions, discourses, narratives, bumping into each other? Or no emergent structures at all, just individuals acting from subjective motives? I’m in the middle of that list, seeing things in terms of multiple dialectics.


  2. Reblogged this on stolen truth and commented:
    Tulisan menarik tentang Marxisme dan Anarkisme. Semuanya sama, menentang keberadaan negara yang hanya jadi alat dominasi. Bedanya, Marxisme meski menolak keberadaan negara, masih tetap mencoba merebut kekuasaan negara sebagai salah satu strategi transisi untuk mengimplementasikan Sosialisme. Anarkis, tidak meminjam sedikitpun kekuasaan negara. Mereka seutuhnya menolak keberadaan negara setiap waktu.


  3. Not sure if it helps get Marxists and Anarchists into bed, but as a strictly theory thing: do you know the work of Nicos Poulantzas? He tried to transcend the dilemma between the horns of the crude and long-discredited instrumental-expressive conception of the State found in Marx’s texts on the one side, and the dubious Anarchist conception of the State as the incarnation of a pure will to power that is at once a deviation from the ways of Nature and an enormous threat to them on the other. The error in either conception, as he saw it, is that they each conceive of the economy and/or the State as theoretically and practically self-sufficient phenomena when they should properly be viewed as mutually-constitutive “regions” of a unitary mode of production. To take the Marxist idea of an articulated mode of production, considered as the unitary theoretical object of its own proper science, seriously would entail recognizing that neither the State nor the economy can even in abstraction be regarded as water-tight and self-sufficient entities. Class relations are constitutive of the State in the strong sense, i.e. present in its workings *by definition*; mutatis mutandis, the economic constitutively entails political relations as both its existential conditions of possibility and theoretical conditions of conceivability.

    The State thus certainly cannot be seen as the incarnation of a pure and primordial will to power; indeed, according to Poulantzas, there can be no such thing as a truly general theory of the State of any kind in the same way that there exists a general theory of chemistry, of biology, etc. (although there certainly can be a legitimate “regional” theory of the State and its role in the MOP). But- by the very same token- Marxist science does not and cannot provide a general theory of the economic in this sense, since its penultimate theoretical object is not “the economy” (as it is for mainstream economics), but the mode of production (of which the economic is but one region, albeit the determinant one). The upshot, under these terms, is that capitalist production can in no way be conceived as something that ontogenetically starts off life as pure economic activity and subsequently grows political appendages (as Marx often suggests, in spite of himself, in his muddled expositions of “base” and “superstructure”). On the contrary: the modern State has always, as a matter of both fact and definition, maintained a constitutive presence in modern economic life (contra a poorly-informed mythology dear to both the modern Left and Right alike).

    For Poulantzas, the key to clear and rigorous thinking about these things is to aggressively give pride of place to class struggle in the Marxist analysis, not the spatial metaphor of base-superstructure, which is wrong-headed and in any case irrelevant (i.e. the political and the economic are anatomically conjoined by the class relation, which itself is conceptually defined in terms of exploitation; this covers the valid theoretical intent behind such misleading Marxist formulations as “base-superstructure” or Althusser’s “determination in the last instance by the economic”).


    • Hi! I’ve been meaning to reply and have just been very busy. I know of Poulantzas by reputation and I find his position appealing but I’ve never gotten around to reading him directly. I’ve also heard Marxists quite hostile to Poulantzas articulate a similarly broad conception of class struggle and of the mode of production, in which the mode of production includes all material relations in a society including not only direct production but the distribution, use, and consumption of all material goods. I think the English school epitomized by E.P. Thompson took this view, although I might be off a little. There too the implication would be that the state, rather being a superstructure that rests on top of the mode of production, is part of the mode of production. What I’m commenting on here, what I find striking, is that Marxists of quite different and mutually hostile theoretical orientations seem to have decided at about the same time to deliberately broaden the notion of class in a way that obviates the base-superstructure distinction.

      I’m very sympathetic to this move. Certainly I’ve always found it dissatisfying to equate ‘material social relations’ or even ‘relations of production’ with ‘the economy’. However, I’m going to make a case that this doesn’t quite resolve the epistemological disagreement between Marxists and anarchists that I’ve described.

      I think that it is a feature of a complex system that its complexity can be resolved according to more than one epistemological strategy. When we study society we encounter, in the first instance, all sorts of relations: relations of production, of reproduction, of consumption, of force, of communication, of simulation, and so on. Moreover, all of these relations are implicated in all the others: there is no relation of production, for instance, that is not also (that cannot also be understood in terms of) a relation of force, of communication, etc. So any theorist who proposes, with sufficient skill and nuance, to reduce the complexity of society to the dynamics of a single relation has an initially plausible, even irrefutable, case to make: to the extent that any one type of relation is implicated in all the others, those others can be parsed according to its logic, understood as expressions of its dynamics.

      One might think the solution would be not to reduce any relation to any other. But I think we find this just to be too confusing. Maps necessarily have to contain less information than the terrain they refer to; science must engage in some reduction to be more useful to us than our immediate perceptions. And we tend to find, in practice, intuitively, that some types of relations bend more to the imperatives of others than vice-versa – although which way we see this constraint operating seems to vary quite widely among observers, and might vary systematically by social location and by the vector of one’s movement through social relations (by one’s “purposes”, etc.).

      All of which is to say that I think Poulantzas’s choice, defensible on its own terms, would not satisfy everyone. Including the state within the domain of class struggle still entails an epistemological reduction of those relations distinctive to the state as an institution (relations of force, of representation, etc.) to the logic of relations of production – even if we do this by acknowledging, e.g., policing, or military action, or legislation as kinds of production. At least, it does so as long as the concept of ‘relations of production’ or ‘class relations’ has any kind of specificity.

      This is where we go well beyond the kinds of claims that can be adjudicated “true” or “false”, I think. A Poulantzian scheme will produce explanations that can be fitted to the empirical data, tolerably well, relative to particular standards of validity (some of which come from the theory itself and some of which are supplied by the wider historical conjuncture). Evaluating a sufficiently robust theory, one moves beyond questions of “is it true?” to questions like “what does it tell us?”, “what does it leave out?”, and above all, “what does it help us do?”.

      An anarchist might well object that subsuming force relations within relations of production might give us certain insights, but would also obscure important information or render unaskable questions we would otherwise want to ask. Ultimately, the anarchist would be concerned that Poulantzas’s epistemology does too little to inoculate us against, or help us to supercede, the tendency of state domination to reassert itself even in the midst of socialist revolution.

      At least this is my view of the situation. Sorry if I have presented it didactically … does this make sense? Have I addressed your thought?


  4. Re:

    “I think that it is a feature of a complex system that its complexity can be resolved according to more than one epistemological strategy.”

    “Maps necessarily have to contain less information than the terrain they refer to; science must engage in some reduction to be more useful to us than our immediate perceptions”.

    “This is where we go well beyond the kinds of claims that can be adjudicated “true” or “false”, I think…Evaluating a sufficiently robust theory, one moves beyond questions of “is it true?” to questions like “what does it tell us?”, “what does it leave out?”

    Et cetera argumentation esp. paragraphs 6 and 7:

    The inexorable logical implication is that the Anarchists (and, for that matter, anybody for whom the political has an irreducible “life of its own”) were in the right: Marx’s thesis that the State just withers away in a classless society is a classic case of the “fallacy of misplaced concreteness”, which confounds partial and theoretically-delimited abstractions with the entirety of the concrete. What Marx should have said is that political domination *of one economic class* by another will of necessity disappear under Socialism, but not necessarily political domination *as such*. Marx’s theoretical map makes visible only those streets of the political that lead from and to the social dynamics of production; it amounts to a set of directions that does not make all the city streets visible. That is, the Marxist conception of the State is defined only in terms of those attributes of the State salient to the play and reproduction of relations of production, in the same way that the typical sociological concept of the “personality system” maps out only those aspects of the mind relevant in an immediate theoretical sense to the social-cultural system; neither is concerned with politics or mind as such. It follows that Marx’s account of the State as an engine of class struggle and reproduction can be no more valid as a truly general theory of power than the sociological concept of the personality system is valid as a general theory of psychology.

    As you said, any science can and indeed must engage in some form of reduction for its own purposes; it is perfectly legitimate for historical materialism to concern itself only with those aspects of the political that can be traced to and from the social dynamics of production. But it would be needlessly greedy reductionism to maintain that the relationship of various actors to the means of production is the primordial ontological truth of every conceivable human dispute. (It would be like a sociologist arguing that the central nervous system is the artifact of socialization). There is no a priori reason to think that the State altogether withers away when economic classes do


    • I’ve been meaning to reply for a while.

      “Marx’s theoretical map makes visible only those streets of the political that lead from and to the social dynamics of production; it amounts to a set of directions that does not make all the city streets visible.”

      I like this very much, both for what it tries to say and for the elegance of the formulation. My first impulse was to agree. But considering it further, and extending your analogy, it strikes me that in any typical city all streets interconnect so that all streets do lead (indirectly at least) to and from any given point in the city. If one employs a mapping technique in which streets appear more or less prominently according to how directly they connect with any one given point in the city, i.e. how few or how many necessary turns onto a new street lie between them and the focus, then one can construct any number of such maps using any number of foci. This might serve as a rough analogy for how, in a sufficiently complex system, one can map all of its elements as an expression of any, with differing results.

      If we accept this notion, then it seems to follow that we evaluate competing maps not in terms of what is included or left out in an absolute sense, but what is highlighted and what is lost amidst the clutter, what is simplified and what is rendered obscurely complicated. In other words, I suppose, different maps either increase or decrease the labour required to construct different analyses and – crucially – made different types of interventions into the phenomena.

      (I actually think one could construct a theory explaining the central nervous system as an artifact of socialization; it just would be too damned cumbersome to use for the sorts of purposes for which we employ psychological theories.)

      This is me being relativist again, although there’s a non-relativist reading of what I’m saying (i.e. if we ontologize complexity, which I suppose is actually the more obvious and straightforward reading).

      So I mostly agree with you that

      the Marxist conception of the State is defined only in terms of those attributes of the State salient to the play and reproduction of relations of production, in the same way that the typical sociological concept of the ‘personality system’ maps out only those aspects of the mind relevant in an immediate theoretical sense to the social-cultural system; neither is concerned with politics or mind as such

      but I would throw the qualifier “directly” in to a few places, especially the final clause. Proceeding from Marx’s work one can analyze the aspects of state politics that are only indirectly concerned with relations of production, since even indirect relevance is still relevance and almost nothing the state does is without some consequence for class struggle. But one has to work harder to show and to analyze certain effects and dynamics of state power in this way than if one takes some other kind of relation, like representation, or force, as one’s point of departure – although the reverse is also true.

      I’m not trying to one-up you, theoretically (I think); just taking your statements and pursuing, out loud, the train of thought they provoke … thank you for these very stimulating comments.


  5. If you’re interested in a recent effort to “resolve” the Marxist attempts to understand the relationship between capital and the state, you might think of checking out the latter chapters of Nitzan and Bichler’s “Capital as Power.” I don’t think they have all the answers, but their book does try to address what they see as a gap in Marxist theories of the state: that they all, despite their considerable differences, take capital and the state to be distinct entities. The authors make an argument that in contemporary capitalism, they are not. In fact, they attempt to take another step and replace the concept of the mode of production with that of a “mode of power,” since their argument is that Marxism has focused too much on “production” (understood as the sphere in which accumulation takes place, and thus primary) to the detriment of other vehicles through which the accumulation of value in fact occurs.


  6. Pingback: The Revival of Marxism? No, just the revival of Marxists « Trade Onion

  7. Marx came to reject the idea that the state could be used to implement socialism after the Paris Commune. In the 1872 preface to the Communist Manifesto it is stated the Commune proved ‘the working class cannot simply lay hold of the ready made State machinery, and wield it for its own purposes”.

    For Marx, since his early writings, the aim was the merging of political society with civil society. this developed through out his writing and lead to, what he described when criticising Bakunin, the belief in a new form of state radically different from the current one and based on the idea of the commune.

    Marx rejected Bakunins criticism of him by argueing that bakunin waas criticising the Borugeois state and he was proposing somethign different. The bourgeois state is based upon imposing and putting into practise bourgeois ideas, becoming a minafestation of bourgois conciousness. In effect it is the dictatorship of the bourgoeisie. A workers state would be based upon putting into practice and manifesting workers conciousness and in effect is the dictatorship of the proletariat. When looked at in the context of Marx’s writings on civil society and his early writing on the state (for example on the Jewish Q”uestion) this appears to be more in line with what anarchists such as Malatesta or Kropotkin argue for, with the comune being the centre of power and minimal organisation to unite communes.

    Some Marxists proposed this andwhile anarchists have always backed bakunins criticisms, they also appear to have more in common with Marx’s idea, when looked at in detail, than against it. However, other Marxists have interpreted his work differently. In particular Lenin argued for the building of the political state (rather than the abolishing of it) and then focused on this as a defence for any future revolution.


    • Thanks very much for this; these are important points. It would be unwise to portray the relationship between Marxism and anarchism in terms of a binary opposition. Both labels embrace broad and diverse praxical traditions; some extensions of Marx’s work lead to positions that overlap substantially with anarchist positions.

      I’m struck by something you point out: Marx’s view that the bourgeois state is a manifestation of bourgeois consciousness while a worker’s state would be a manifestation of workers’ consciousness. Assuming you’re correct about this (this is an area of Marx’s thought I’m not intimately familiar with), then this point brings up again a problem I have with Marx’s theory, one that I talked about in an earlier post.

      For on the one hand, in The German Ideology and elsewhere, Marx emphasizes that consciousness is the product of practical life-activity. I have always taken this to change society, one must focus on changing the practical life-activity of human beings; do that, and changed consciousness will follow. But Marx assigns a decisive historical importance to class consciousness, and the actual focus of his politics, and that of most Marxists, seems to be on helping the working class achieve class consciousness. I’ve always been puzzled by this.

      In particular, how can the state be a worker’s state by virtue of manifesting workers’ consciousness? What are the practical social relations by means of which this happens?

      If the answer is that the workers’ state has a radically different institutional structure than the bourgeois state – e.g. because it comprises a decentralized network of semi-autonomous communes, for instance – then one needs in turn to explain why such an arrangement is more favourable to labour than to capital. For this I think one needs a theory of political relations as practical, material relations in their own right, which is something I haven’t come across in my admittedly incomplete exploration of Marx’s work and the Marxist tradition.


  8. Pingback: Marxism and Genocide | The Practical Theorist

What thoughts do you have?

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s