Giuseppe Pellizza da Volpedo: ‘The living torrent’ (1896)

Socialism from below: a delusion

We need to go far beyond slogans, urges Mike Macnair

Revolutionary Socialism in the 21st Century has recently published both an article by Neil Davidson (‘Politics in the age of austerity: from above or below?’, spring 2015), and one by Dan Swain entitled simply ‘Socialism from below’ (summer 2015). The common idea - ‘socialism from below’ - is more widely shared.

The Socialist Workers Party, in spite of its own obviously Stalinist internal regime, also subscribes to that idea, and its Greek co-thinkers use the tag as the title of their bimonthly journal.1 The US International Socialist Organization promptly reprinted Swain’s article in its own Socialist Worker2, while Socialist Resistance has David McNally’s version on its website.3 And so on. Googling “socialism from below” (with quotes) produces 14,700 hits. The tag is derived from Hal Draper’s The two souls of socialism (originally published in 1960, though the version generally used is that of 1966).4

There are some critiques of ‘socialism from below’ available, but they have their own problems. Martin Thomas’s ‘What’s wrong with “socialism from below”’(1990s, expanded version 2003)5 is mainly concerned with defending the usefulness of small political left groups like his own Alliance for Workers’ Liberty. While I (obviously) share the view that small political groups are useful, I am sceptical of Thomas’s reasoning and think the issue is a larger one. Joaquin Bustelo’s ‘On the two souls of socialism’ (2005, republished at the North Star site 2013)6 is, if anything, more problematic, linked to the illusions in Cuba which Bustelo inherited from his time in the US SWP (not linked to the British SWP). Andy Newman’s brief comment in his 2011 review of Ian Birchall’s biography of Cliff7 plainly reflects Andy’s own move towards Stalinism.

I have said that the SWP subscribes to ‘socialism from below’ in spite of its own obviously Stalinist internal regime, but in fact it is arguable that such a regime follows from the conception of ‘socialism from below’ as interpreted by the SWP, by its co-thinkers and its ex-members. The conception is, I would suggest, at the end of the day Bakuninist in its reliance on ‘revolutionary’ direct action, counterposed to political/electoral action; and because it is Bakuninist in this way it actually logically requires something like Bakunin’s conspiratorial political practice: the ‘invisible dictatorship’.8 The aspiration to do political action in a democratic and participatory way is thus poisoned by the conception of what political action is. One can arguably make the same point about the relation between Rosa Luxemburg’s advocacy of the mass strike and of spontaneity, and the political practice of her Polish Social Democracy of the Kingdom of Poland and Lithuania (SDKPiL), a notorious top-down dictatorial sect (as I have argued before).9 Why?

Draper’s amalgam

I begin with Draper’s The two souls of socialism and its critics. The essence of his argument is the ‘two souls’ of the title: socialism ‘from above’ and ‘from below’. His summary:

What unites the many different forms of socialism-from-above is the conception that socialism (or a reasonable facsimile thereof) must be handed down to the grateful masses in one form or another, by a ruling elite which is not subject to their control in fact. The heart of socialism-from-below is its view that socialism can be realised only through the self-emancipation of activised masses in motion, reaching out for freedom with their own hands, mobilised ‘from below’ in a struggle to take charge of their own destiny, as actors (not merely subjects) on the stage of history. “The emancipation of the working classes must be conquered by the working classes themselves”: this is the first sentence in the rules written for the First International by Marx, and this is the first principle of his life work.

For Draper, ‘socialism from above’ thus starts with the early conspiratorial leftists (Babeuf to Blanqui) and the utopian socialists (Owen, Fourier and so on); is criticised by Marx; reappears in the conspiratorial form in Bakunin. The fundamental politics of the Social Democratic Party of Germany (SPD), Draper argues, was Lassalle’s statism and his (private) labour monarchism, linked in turn to the German Kathedersozialisten (‘professorial socialists’, statist and German nationalist) and to British Fabianism, and in this way to Bernstein, with Rosa Luxemburg as the prime advocate of ‘socialism from below’ in this generation. In the US, Debs offered ‘socialism from below’, Progressives and later Rooseveltians ‘socialism from above’; Stalinism, too, is a ‘socialism from above’, and Draper proceeds to hypothesise a series of subordinate ‘isms’ contained in ‘socialism from above’ which allow him to count Harold Wilson, Juan Posadas, Paul Sweezy and Isaac Deutscher among its advocates.

The fact that Draper regarded Debs as advocating ‘socialism from below’ is worth noting. In the first place, Debs was won to socialism by Kautsky’s writings - that is, the SPD politics which Draper elsewhere regards as ‘socialism from above’.10 Secondly, Debs’s socialism was electoral; here Draper’s ‘socialism from below’ is not tinged with the anti-electoralism which is a marked feature of Cliffite and related uses of the idea.

More generally, Draper is in this instance guilty of what Trotsky, in relation to the Moscow trials, called an ‘amalgam’: that he associates different and opposed political tendencies by the more or less arbitrary ‘differentia specifica’ of the claim that they all stand for ‘socialism from above’.

Martin Thomas in his critique notes the point in relation to the Owenites: however much Owen’s New Lanark was at its origin a paternalistic project, the Owenite movement engaged in activities which would entirely have satisfied Draper as being ‘from below’.

The same could be said of the mass movements animated by the classical pre-1914 social democracy, which Draper also wants to class as offering socialism ‘from above’. Here the point is a fortiori, since it is certainly the case that Wilhelm Liebknecht and August Bebel and their co-thinkers opposed the ‘labour monarchism’ of Lassalle and the Lassalleans, and took it that they were pursuing a project of working class self-emancipation, in some sense connected to that of the First International. And, in spite of the concessions to Lassallean theory made at the Gotha unification in 1875, on the issues of leadership accountability and local and sectoral self-government in the party, the unified party wholly rejected the Lassallean approach.11

The Two souls amalgam nonetheless contains - implicitly rather than explicitly - a truth, and one which is missing in the Cliffite, sub-Cliffite and post-Cliffite versions and among the critics. This truth is contained in the fact that the utopian socialists were as much concerned with an (ultimately Platonist) image of socialism/full communism as a society run by an enlightened expert elite, as they were concerned with the question of how to get to this end goal. The Cliffite version is, in contrast, all about how to get to socialism.

And Marx’s12 criticisms of the utopians, Lassalleans, and so on, were as much objections to the image of the end goal as ‘barracks socialism’, as criticisms of any ‘how to get there’ question.

Marx’s and Engels’ conception of ‘how to get there’, centred on proletarian self-emancipation and involving radical democracy in relation to the state and in the movement, flowed from an understanding of the end goal of communism, which involved the supersession of occupational specialisation, as much in the 1876-78 Anti-Dühring as in the 1845-46 manuscripts later printed as The German ideology. Hence the working class learning to self-manage in building its own class movement, specifically including cooperatives, etc, and learning to lead society through working class political action, was the initial key to the process through which the working class could first take power, and then lead society altogether beyond the existence of classes and expert elites of any sort.

This two-sided aspect - that Marx’s, Engels’ and their co-thinkers’ arguments are both about immediate tasks, and about the nature of the communism which is the basic goal - allows us to approach the problem from both ends.


To start from the immediate: learning to self-manage. A riot, like those in British cities in 2011, is an act of resistance to, or a protest against, an intolerable existing order. But in itself it does not lead anywhere except to inflicting damage on the poor’s own living areas. Rioting can be a form of action in support of collective bargaining (as were, also, sabotage and arson, before the legalisation of trade unions); but only if there is some form of ongoing collective organisation (even if very limited) lying behind it.

This is the Marx versus Bakunin point: not,primarily, the points Draper picked up on in 1960-66 about Proudhon’s or Bakunin’s anti-Semitism or their attempts to persuade Louis Bonaparte or the tsar to promote ‘anarchist’ projects. Spontaneity without organisation leads nowhere. Hence, if spontaneous direct action, revolt and resistance in itself, is to be fetishised (as in Bakunin, but not in Proudhon), there must be some organisation to play the role of the ‘bearer’ of the ideas between episodes of riot, strike or whatever. The Chartists said, “Peacefully if we can; forcefully if we must” and endeavoured the political representation of the working class - though they were still, in the end, crushed by repression.13 The Bakuninists, in contrast, sought to commit the movement a priori to validating only the strike, the riot, the insurrection, as against ‘peaceful’ and ‘legal’ forms of action. This requires that the continuing organisation have the character of a clandestine conspiracy - the ‘invisible dictatorship’ in theory, in practice the clandestine group operating within a front operating within the First International.

But the question which is posed, and not answered, by this is: even if the ‘invisible dictatorship’ is to give leadership to the spontaneous mass actions, how is the ‘invisible dictatorship’ itself to take decisions? Back to Marx and Engels versus Bakunin: in The Bakuninists at work (1873) Engels chronicles how in the Spanish revolution of that year the Bakuninists’ ‘federalism’ - meaning localism - led them to promote town-by-town action - leading to defeat in detail by the military reactionaries, who coordinated their action on a national level.14

An analogous issue related more directly to production is described by Roger Pethybridge in The spread of the Russian Revolution. Railways which ran on wood-fired steam engines had acquired plantations of trees to supply fuel. Local peasants seized the plantations as part of ‘their’ village’s land; and by doing so stopped the railways.15Some decisions can be and should be taken locally. In relation to others, attempting to take local decisions is, in fact, for the locals to decide for everyone else that there shall not be a railway, or electricity supply, or an internet, or whatever the large-scale infrastructure item involved is.

Now one might argue, as some Greens do, and as is in some respects implicit in some of Michael Lebowitz’s work,16 that to make this sort of argument is ‘productivism’: ie, that it is to accept the choice of forces of production (here meaning technologies) which capitalism has delivered for us and which it has created with a view only to its profit imperatives; that we will need to make choices about what sort of technologies we use which will facilitate human development, and therefore which will facilitate ‘local’ and ‘horizontal’ decision-making.

It is certainly true that there are aspects of capitalist technology which we will want to supersede. It is in the highest degree unlikely, however, that any process of supersession will allow a return to a real localism or ‘Small is beautiful’. At the most brutally simple level, capitalist national and world tradein food products permits massively larger total populations to be fed than was the case of its predecessors. Britain, for example, currently has a population just under 65 million; in 1600 it was under 5 million. Equally, electricity and gas supply infrastructures, where they exist, reduce the pressure of large populations on wood supplies, leading to deforestation, desertification and higher carbon emissions from the aggregate of domestic fires than centralised electricity generation.

Certainly there are a great many decisions which can be taken locally. Indeed, as long as we preserve and keep working the infrastructure and the main structures of necessary production, storage and distribution, there is a great deal which can and should be purely individual choices: just look at the range of writings, music and artwork presently available on the web, or the diversity of individual choices of personal appearance on any urban street in England. It should be an aim of socialism to expand this diversity, not to reduce it.17 But that ‘as long as’ is a big proviso. It means that there remain numerous decisions which must be made on a global scale (how to deal with human-induced climate change), a continental or a national/regional scale.

The problem is how to make these decisions. And as we have just seen, in relation to the Bakuninists, this is not just a problem for the future socialist society: it is a problem for the present practical struggle for working class interests under capitalism.

I have written about this issue before, in an exchange with Paul Cockshott in 2010.18 I do not propose to repeat the points made there (chiefly against referenda). My point here is that the distinction between ‘from above’ and ‘from below’ is utterly useless in addressing the problem. The result of adopting it is then that some unexamined form of decision-making is taken up: with the SWP, this was ‘democratic centralism’; anarchist forms tend to involve less accountable charismatic leaders, functioning like the pre-SPD forms of capitalist parties (like today’s US Democrats or Republicans).

From above

The first point to be made is that the idea of ‘socialism from above’ is illusory, for reasons originally adumbrated by Marx as far back as the 1843 Critique of Hegel’s philosophy of right [law]:

In monarchy the whole, the people, is subsumed under one of its modes of existence, the political constitution; in democracy the constitution itself appears only as one determination, and indeed as the self-determination of the people. In monarchy we have the people of the constitution, in democracy the constitution of the people. Democracy is the resolved mystery of all constitutions. Here the constitution, not only in itself, according to essence, but according to existence and actuality, is returned to its real ground, actual man, the actual people, and established as its own work. The constitution appears as what it is: the free product of men ... 19

To put the point in another way, James can only be king if his subjects are prepared to treat him as king; Joseph can only be general secretary if party members are prepared to treat him as general secretary; however much Rupert may think David would be a better Labour leader than Ed, David could only become Labour leader if members were willing to back him.

All these examples illustrate the point that Marx’s claim is about human social decision-making, not a claim about formal democracy: James is not elected at all, Joseph’s re-elections from 1929 on were the work of Cheka operations against the party, and (as has been shown this year by the surge of support for Jeremy Corbyn) a choice between the Miliband brothers is not a choice which a formally democratic process in the Labour Party, as opposed to one controlled by the parliamentary party and the apparatus, would have generated. But, as long as enough people are prepared to support the monarchy, or the general secretary regime, or the leader, they can lead.

Leaders must have people willing to follow them. What happens in revolutions is, commonly, and in the first stage, that the willingness of people to follow the existing leaders evaporates. In this country, Tim Harris’s Revolution (London 2006) illustrates the point: well before James II was forced to flee by Dutch invasion and by elite and popular mobilisations against him, his political projects were already failing due to the passive resistance of the Tories who had supported him when he first inherited the throne.

Just as monarchies can exist if they have mass support, the same is true of models of ‘socialism from above’. The reality of Stalinism is thus not just a story of evil manipulators above, but of the shared mistakes of those below who supported this project. The enlightened elite cannot lead the society without support from those below.

A particular aspect of this is that the idea of an enlightened minority conning its way into creating socialism fails. Broad masses can see well enough that the Trot fronts they are offered are just Trot fronts, however much they pretend to be old Labour or just anti-austerity, and do not bother voting for them. If Corbyn wins the Labour leadership, he will not be able to lead effectively by appealing to unity with the right20: it was already guilty of passive resistance and unattributable briefing against Gordon Brown and Ed Miliband, and so he will need to raise up support outside the parliamentary party.

Imagine, for the sake of argument, that he succeeded in doing so, and that some quirk of the electoral system led to Labour winning a majority in parliament with - say - 37% of the popular vote (as Cameron has 36.9%; or compare Syriza’s 36.3%). A Tory administration can govern with this sort of largest-minority level of support, because it has extra-parliamentary support among the civil service, the security forces and the judiciary, as well as among the controllers of private capital. A Labour administration led by the left would face at least bureaucratic, passive resistance and adverse judicial decisions on a massive scale. It would need to appeal to support in the country, to coerce the bureaucracy - and 37% largest-minority support won on the basis of an assumption that there would be no need to coerce the bureaucracy would not be enough for the purpose; just as Syriza’s 36%, even with the backing of a referendum, is not enough support to make fundamental change in Greece.21

Expressing this as a hypothesis about a Corbyn leadership of the Labour Party, or a comment about the situation of Syriza, is merely to dramatise the point: these are illustrations of the fundamental point: that a hypothetical enlightened elite cannot lead the society without support from those below for its leadership. ‘Socialism from above’ if it ever exists (and Stalinism indeed looks pretty like the old 19th century utopian or Kathedersozialist ideas of barracks socialism, Draper’s original ‘socialism from above’) exists because it is supported from below; and when it ceases to be supported from below, as happened in 1989-91, it collapses.

From below

‘Socialism from above’ depends on support ‘from below’. But it is equally true that ‘socialism from below’ inherently involves leadership and coordination ‘from above’. This is partly the point already made about the need for decisions on a scale larger than the local. But it is also about ‘leadership’ in two senses, and ‘vanguards’; hopefully, however, divested of the fetishism of permanent leadership, which is the leftist’s form of the bourgeois principle of occupational specialisation.

Homo sapiens sapiens is a zoon politikon (Aristotle): a social animal. But we are not a social species in the sense that some social insect species are. Our sociality is accompanied by strong capabilities for individuality and is communicative. Even hunter-gatherer societies, which have low levels of complexity and no class stratification, display disputes and disagreements, evidence of the transhistorical presence in human life of the contradiction of the individual and the social.22

‘Leadership’ is an aspect of the social/individual contradiction in two ways. The first is that, in relation to a new idea (whether technical, political, religious or artistic, or whatever), someone has to have it first. Of course, several people may come up with the same idea independently; but that is still not the same as everyone coming up with it simultaneously: to imagine the latter was possible would be to reject the reality of human individuality. The individual or individuals who have the idea first lead; or, if you will, they are the vanguard, the first to reach a certain point, or the advanced part. This, of course, assumes that the idea is valid, or useful, or later generally adopted; if it is not any of these things, we would not use this sort of expression.

The second aspect is that beyond a certain level of social (and technical) complexity, a practical coordination function arises to coordinate the different efforts of individuals. Meetings above a certain size need someone to chair them. Lebowitz uses the example of the orchestra and its conductor to illustrate what he damns as the ‘vanguard mode of production’, and proposes as an alternative a society of ‘associated conductors’:23 which produces the odd image of lots of people waving at each other and no music - an image instantiated in practice in left e-lists and the comments on left blogs and Facebook pages. Another perhaps-alternative is a jazz jam session. But, however well this may work in jazz (hit or miss ...), it is, perhaps regrettably, ‘no way to run a railway’. The coordination function is real, and there are strong reasons to suppose that it will not disappear in a future society.

As I said earlier, the ‘Marx and Engels’ image of the future society is not one in which either the coordination function, or the reality of some people having ideas first, disappears. It is one in which no-one exercises the coordination function, or is taken to have the role of ideas-producer, permanently: “there will no longer be any professional porters or architects, and that the man who for half an hour gives instructions as an architect will also act as a porter for a period, until his activity as an architect is once again required”.24 Paul Cockshott and Allin Cottrell in Towards a new socialism usefully use the image of jury service as a way of thinking about people getting ‘landed with’ the coordination role for a time; other possible images might include the obligation of borough citizens to serve as constable, or as sheriff, which citizens can be seen struggling to evade in English case-law around 1700.25

The case of ‘having ideas first’ is, if anything, easier: what is required is merely recognition that neither creativity, nor getting things right, is a permanent attribute of any individual, and therefore institutional choices which facilitate as far as possible the open exchange and discussion of rival ideas before any particular idea is carried into action.

The ‘Comintern model’ of the party, starting with the Second Congress, muddled up the two issues of the coordination function and the leadership role of having ideas first, and fetishised a ‘combination view’, in which the party as the advanced part of the working class not merely led the class in the sense of having ideas first, but also performed the coordination role in relation to the class as a whole; and, further, that the party politicallyrepresented the class as a whole, as against ‘backward’ particular sections of the class which disagreed with it. By logical extension, the central committee as the advanced part of the party not merely led the party, but represented the party as a whole ... so that the political bureau as the advanced part of the CC not merely led the CC, but represented the CC as a whole ... and, in turn, the general secretary as the most advanced member of the politburo not merely led the politburo, but represented ... This model has been faithfully adopted, or, indeed, caricatured, by Trotskyist organisations like the SWP, and has been explicitly defended in its full Stalin-era form by writers in the ‘Spartacist’ tradition.26 It is, in essence, built on the bourgeois and petty-bourgeois image of occupational specialisation, critiqued in the passage quoted above from the Anti-Dühring.

This fetishistic model has produced ‘anti-vanguardism’ as a natural attempt at negation of it. But ‘anti-vanguardism’ is characterised by the fact that it accepts as good coin the ‘Comintern model’, which glues together different types of leadership and fetishises the glued-together role as a sort of occupational specialisation - but proceeds to attempt to negate this glued-together fetishism as a whole, without accepting either the need for the coordinating role or the need for individual and minority initiative. Instead what is proposed is various forms of ‘network’, ‘horizontalism’, and so on.

Since, however, the need for the coordinating role, and the inevitability of individual and minority initiative are not artificial ‘vanguardist’ constructions, but features of homo sapiens sapiens as a particular sort of social animal, this refusal simply fails.

It may result in a straightforward collapse: thus the 1960s Socialist Review group/International Socialists were advocates of ‘socialism from below’ and of ‘Luxemburg against Lenin’, but around 1970-73 this approach simply collapsed into a hard-line ‘Comintern model’, leaving ‘socialism from below’ as a sort of political tic in the modern SWP and its descendant groups, which at most indicates strikism.

Or, paradoxically, ‘anti-vanguardism’ may turn into an instrument of labour bureaucracy control, by characterising minority groupings as would-be vanguardists or ‘self-appointed theorists’: so in the ‘social forums’ - ‘anti-vanguardist’ but controlled by the Brazilian Workers Party ... and so, presumably, in Lebowitz’s attachment to the Chávista project. In this sort of ‘anti-vanguardism’ the coordinating role is half-admitted, but the role of minority initiative denied as ‘vanguardist’. It might as well be decried as ‘permanent factionalism’ (SWP) or ‘Trotskyite wrecking’ (old-style Stalinism).

My point is not to say that we (CPGB) have all the answers to the notorious problem of how to do workers’ organisation and decision-making on scales larger than the local, without falling into regimes of bureaucratic legalism (Labour, trade unions) or bureaucratic centralism (SWP, and so on). It is merely to assert that real problems are posed, and trying to address them using the slogan, ‘socialism from below’, is completely useless and, by its uselessness, merelyanother route back to bureaucratisml



1. See https://en.wikipedia.org/wiki/Socialist_Workers%27_Party_%28Greece%29.

2. http://socialistworker.org/2015/07/16/socialism-still-comes-from-below.

3. http://socialistresistance.co.uk/socialism_from_below_by_david_mc.htm.

4. www.marxists.org/archive/draper/1966/twosouls/index.htm.

5. www.workersliberty.org/story/2003/09/30/whats-wrong-socialism-below.

6. www.thenorthstar.info/?p=8036.

7. http://socialistunity.com/cliff-in-perspective.

8. Marx’s viewpoint documented at H Draper Karl Marx’s theory of revolution ... ; anarchist defences are cited at https://en.wikipedia.org/wiki/Invisible_dictatorship.

9. Weekly Worker August 16 2012, albeit with some caution.

10. www.marxists.org/archive/debs/works/1902/howi.htm.

11. RH Dominick Wilhelm Liebknecht and the founding of the German Social Democratic Party (Chapel Hill 1982) is a detailed treatment.

12. And Engels’, as Draper’s massively fuller work in Karl Marx’s theory of revolution made clear.

13. J Saville 1848: The British state and the Chartist movement Cambridge 1990.

14. www.marxists.org/archive/marx/works/1873/bakunin, section III.

15. London 1972, pp12-13.

16. Eg, in the late chapters of The contradictions of ‘real socialism’ (New York 2012); or the final section of ‘Understanding the Critique of the Gotha Programme’, chapter 2 of The socialist imperative (New York 2015). Compare with the present point David Laibman’s critique of the latter piece, ‘Quotology, stages and the posthumous anarchization of Marx’ Science and Society No78, pp281-86 (2014).

17. Much of what is available is poor; but that is just Sturgeon’s Law in operation (‘90% of everything is crap’).

18. ‘Representation, not referendums’ Weekly Worker June 30 2010. Cf also Ben Lewis, ‘Referenda and direct democracy’ Weekly Worker Sept 18 2014, on Kautsky’s 1893 Parliamentarism, direct legislation by the people and social democracy.

19. www.marxists.org/archive/marx/works/1843/critique-hpr/ch02.htm.

20. ‘Jeremy Corbyn calls for party unity after warning from rival Andy Burnham’ The Guardian Aug 1 2015.

21. I leave aside the point that Syriza cannot face down the ‘institutions’ on the austerity issue, which would still be true if the party had 80% support and a real ‘plan B’ alternative: the country is just too small and marginal for socialism in Greece alone to mean anything more than socialism in Cornwall alone.

22. CR Boehm Hierarchy in the forest Cambridge MA 1999; S Roberts Order and dispute New Orleans 2013, chapter 5.

23. The contradictions of ‘real socialism’ New York 2012, passim.

24. Anti-Dühring part 2, Political economy, section 6, ‘Simple and compound labour’.

25. Eg, City of Oxford case (1689) 2 Ventris 106; City of London v Vanacre (1699) 5 Modern 438.

26. J Seymour Lenin and the vanguard party (1977-78) incorporates much of the anti-Trotskyist argumentation about Bolshevik history from the late 1920s and pays not the slightest attention to Trotsky’s arguments on the issue in The Third International after Lenin.