Negations of democratic centralism

Mike Macnair completes his series of articles dealing with the issues raised by the collapse of the US International Socialist Organization

In the first two articles in this series I tried to explain how the 1920-21 version of ‘democratic centralism’ was based on rational - but in hindsight mistaken - decisions by the Russian Communist Party (Bolsheviks) in 1918-21, and has subsequently become merely an ideology of ‘left’ versions of managerialism.1

In the third, I looked at the original adoption of ‘democratic centralism’ in the Russian Social Democratic Labour Party in 1905-06, its roots in the Social Democratic Party of Germany before that time, and its probable original meaning as involving the concept of an individual membership party, the rejection of ‘labour monarchist’ regimes of directly elected ‘leaders’, and as regulating relationships between centre and branches, and between elected committees and their electors.2

This fourth article has as its purpose, on the one side, to carry the argument to a slightly higher level of abstraction than the concrete history, and conversely, on the other, to return it to the more immediate or concrete issue of alternatives to ‘democratic centralism’ (in the RSDLP sense and its SPD roots, not in the 1920-21 sense).

It is not a full explanation, but merely as partial polemic about some issues which are related to the claim that the collapse of the International Socialist Organization in the USA is a failing of ‘democratic centralism’ in general.

Capitalist rule and bourgeois parties

I begin in what may seem to be an odd place: how the capitalist class rules, and how this affects the form of the bourgeois political parties. I begin here because the early political movement of the working class inherits this form, and because there is a periodic hankering for it among the modern left (an example used in the arguments about the collapse of the ISO is Hal Draper’s arguments of the early 1970s3).

‘Bourgeois democracy’ is an oxymoron. If capital was free to choose without external constraints, its form of rule would be a cooptative oligarchy: the forms of the Venetian republic, the ‘regents’ of the 17th-18th century Netherlands United Provinces - and the modern City of London Corporation, though this has an elective form: one in which businesses greatly outnumber individual voters.4

Even in such a form, capitalists would usually govern through paid agents - lawyers and the like. Unlike feudal aristocrats, capitalists do not get legitimacy from their ability to govern, but from their character as ‘wealth creators’ (their profits). Governing is a side issue, which they often hand to paid agents.

In practice, the capitalist class is a small minority, whose growth unavoidably implies a substantial working class. It needed the support of (at least) the petty bourgeoisie - small business and the urban middle classes more generally - to win power in the first place. It continues to need their support (and/or some degree of alliance with the rural classes) to hold the working class in subordination.

The standard form of the pre-SPD capitalist political party reflects the second and third of these characteristics: paid agents, and the winning of petty bourgeois support. It consists of a combination of two basic elements. The first is a group of ‘professional politicians’ at the centre - the agents referred to above - in some form of parliamentary caucus, and/or social clubs, with their journalistic hangers-on and related interventions in the press. This is the bit which is normally the locus of national-level corruption. The second element is analogous local groupings of the small capitalists and petty bourgeois, with local attorneys, and so on, in local clubs and societies, loosely affiliated to the national grouping and linked by its ‘brand’ (‘Whig’, ‘Tory’ and so on). This is the locus of local corruption. It works as such because the horizons of the petty bourgeoisie as a class are inherently local.

This works as a social form, and hence as an effective political form, because it is the form of the linkage between the professional politicians (attorneys and the like included) and the large capitalists at the centre, and the small capitalists and petty bourgeois in the localities.

To imitate this as a form of working class political organisation is therefore delusive. To create such a form without the money inputs of the large and small capitalists would be merely to create a form which would be unable to act effectively. On the local scale, as on the national scale, the loose coalition form could not mobilise enough small contributions in money and in unpaid work from working class people to allow it to outweigh the contributions of capital on the other side.

Political action

This lesson of the need for an organised and subscription-paying individual membership had, in fact, already been learned in relation to trade unions well before the first adoption of an individual-membership model for a workers’ political party - which was the work of Lassalle’s 1863 Open letter and the formation of the Allgemeiner Deutscher Arbeiterverein (ADAV).

The other principal immediate element of the idea of the Open letter was the need for working class independent political action. On this front, Lassalle argued ideas which he held in common with Marx, Engels and their co-thinkers, and which went back to the ideas of the Communist League of 1848-52: that the working class must first form an independent political party to fight for universal suffrage, in order to pursue its class ends.

As to what the class should do with universal suffrage once it had it, Lassalle’s proposals were very different from those of Marx and his co-thinkers. Indeed, Lassalle counterposed his proposals to trade unionism, where Marx and Engels at the same period endeavoured to work with trade unionists in the First International.

But the basic point of political action as the immediate task was common to Marxists and Lassalleans. It was opposed alike by the Proudhonists, who counterposed the construction of cooperatives,5 and by the Bakuninists, who counterposed the idea of the general strike and insurrectionary overthrow of the state, and damned Marxists and Lassalleans alike on the basis that “All the German socialists believe that the political revolution must precede the social revolution. This is a fatal error.”6

Karl Marx wrote a letter to Friedrich Bolte in November 1871, explaining criticisms of the Proudhonist and Bakuninist opponents of working class political action, as well as of the Lassalleans. At the end of the letter, he wrote:

NB as to political movement:

The political movement of the working class has as its object, of course, the conquest of political power for the working class, and for this it is naturally necessary that a previous organisation of the working class, itself arising from their economic struggles, should have been developed up to a certain point.

On the other hand, however, every movement in which the working class comes out as a class against the ruling classes and attempts to force them by pressure from without is a political movement. For instance, the attempt in a particular factory or even a particular industry to force a shorter working day out of the capitalists by strikes, etc, is a purely economic movement. On the other hand, the movement to force an eight-hour day, etc, law is a political movement. And in this way, out of the separate economic movements of the workers there grows up everywhere a political movement - that is to say, a movement of the class - with the object of achieving its interests in a general form, in a form possessing a general social force of compulsion. If these movements presuppose a certain degree of previous organisation, they are themselves equally a means of the development of this organisation.

Where the working class is not yet far enough advanced in its organisation to undertake a decisive campaign against the collective power - ie, the political power of the ruling classes - it must at any rate be trained for this by continual agitation against and a hostile attitude towards the policy of the ruling classes. Otherwise it will remain a plaything in their hands, as the September revolution in France showed, and as is also proved up to a certain point by the game messrs Gladstone and co are bringing off in England even up to the present time.7

The basic conception of working class political action is also reflected in Marx’s posthumously published 1874 Conspectus of Bakunin’s statism and anarchy:

[Bakunin]: If there is a state [gosudarstvo], then there is unavoidably domination [gospodstvo], and consequently slavery. Domination without slavery, open or veiled, is unthinkable - this is why we are enemies of the state. What does it mean, the proletariat organised as ruling class?

[Marx:] It means that the proletariat, instead of struggling sectionally against the economically privileged class, has attained a sufficient strength and organisation to employ general means of coercion in this struggle. It can, however, only use such economic means to abolish its own character as salariat, hence as class. With its complete victory its own rule thus also ends, as its class character has disappeared.8

This issue has a surprisingly intimate connection with the issue of ‘democratic centralism’ and the ISO’s collapse. The ISO - like the rest of that part of the left which maintains commitments to ‘new left’ ideas - abandoned the Marxist idea of ‘political action’ in favour of an ‘anti-parliamentarism’; which is, in effect, a variant on Bakunin’s argument. The route to this view was a vulgarisation of the (already weak) arguments of Rosa Luxemburg in The mass strike, and a fetish of her (and the wider German SPD far left’s) attempt to pursue a ‘policy of the offensive’ through mass action in 1910-11. A vulgarisation, because the proposed offensive in 1910-11 was precisely around the constitutional issue of suffrage, not around some economic strike, or some popular liberal idea, or piece of lesser-evilism.

The result is that, in place of “the object of achieving its interests in a general form, in a form possessing a general social force of compulsion” and of training the working class “by continual agitation against and a hostile attitude towards the policy of the ruling classes”, we have offered to us - by the Socialist Workers Party in Britain, the US ISO and all sorts of similar organisations - the promotion of any ideas which happen to be currently popular in the hope that they may result in street and/or workplace mobilisations.

On the one hand, this has precisely the consequence that the workers’ movement and the left remain a ‘plaything in the hands’ of the capitalist politicians, who can drag it from tail-ending them, in fetishising one bit of media outrage, to fetishising another and opposite one.

We can, in fact, see precisely this at work in the identity-politics aspect of the ISO collapse: tailing successively mobilisations round racial and gender identity politics, the organisation became so identity-political that it could not survive the shock of having its own ‘MeToo’ scandals.

This was already a visible problem in the 1850s-60s, since capitalist and other political managers already deployed one or another distraction to demobilise workers’ movements, split strikes, and so on. It was this experience which made Lassalle’s Open letter in Germany, and the idea of the First International in Britain and France, attractive. The effect of moving into suffrage campaigning was very visible in Britain, where its result was both a limited extension of the suffrage in 1867 and the 1871 attempt to legalise trade unions.

On the other hand, if we suppose that the task of a ‘revolutionary party’ is not to push forward political proposals, but to promote street mobilisations and strikes, then ‘democratic centralism’ is necessarily driven towards bureaucratic centralism. The reason is that there is inherently no space available for tactical divergences and local or sectoral creativity. If the whole point of ‘politics’ is to get the maximum turnout on the latest anti-racist demo (or whatever), someone at the centre has to take micro-management decisions, and everyone else has to play the role of followers (or rentacrowd).

This was already visible before 1914 in Rosa Luxemburg, Leo Jogiches and their co-thinkers’ Social Democracy of the Kingdom of Poland and Lithuania, and in Daniel de Leon’s Socialist Labour Party - both organisations committed to direct actionism, both plain bureaucratic-centralist sects. Indeed, there is a sense in which it was already present in Bakunin’s ‘invisible dictatorship’: the idea of the spontaneous revolutionary movement, by refusing the need for ordinary decision-making processes, generates hidden centralism.9 Since the emergence of the new left in the later 1950s, and more particularly since organisations of this type of a significant size developed in the late 1960s-70s, the apparent paradox has been stunningly visible, that the advocates of ‘socialism from below’ turn out to be particularly driven to bureaucratic centralism.

Against labour monarchism

Particularly driven to bureaucratic centralism - not uniquely drawn. I already referred last week to Lassalle’s argument, applied after him by Schweitzer, for concentrating the multiple wills of the members of the ADAV into a single will of the elected president of the party. This particular argument was a sub-Hegelianism. But the idea that parties, and other organisations, must have single leaders - or directly elected single officers answerable to the membership rather than to some leading committee - is a much more widespread belief.

It grows out of the norms of bourgeois management and government: all the way back to John Lambert’s arguments in the 1650s for the necessity of a ‘single person’ executive (ie, Oliver Cromwell as lord protector), through the invention of the ‘prime minister’ as a substitute monarch, and still reflected in the legal requirement that every registered political party have a ‘leader’ under the rules made in Tony Blair’s anti-democratic Registration of Political Parties Act 1998 and its successor, the Political Parties, Elections and Referendums Act 2000.10 Behind this very long-standing commitment to one-man management is the prima facie tyrannical character of the sphere of production itself under capitalism, as Marx famously remarked in Capital Vol 1:

The consumption of labour-power is completed, as in the case of every other commodity, outside the limits of the market or of the sphere of circulation. Accompanied by Mr Moneybags and by the possessor of labour-power, we therefore take leave for a time of this noisy sphere, where everything takes place on the surface and in view of all men, and follow them both into the hidden abode of production, on whose threshold there stares us in the face ‘No admittance except on business’. Here we shall see not only how capital produces, but how capital is produced. We shall at last force the secret of profit making.

This sphere that we are deserting, within whose boundaries the sale and purchase of labour-power goes on, is in fact a very Eden of the innate rights of man. There alone rule Freedom, Equality, Property and Bentham. Freedom, because both buyer and seller of a commodity - say of labour-power - are constrained only by their own free will. They contract as free agents, and the agreement they come to, is but the form in which they give legal expression to their common will. Equality, because each enters into relation with the other, as with a simple owner of commodities, and they exchange equivalent for equivalent. Property, because each disposes only of what is his own ...

On leaving this sphere of simple circulation or of exchange of commodities, which furnishes the ‘free-trader vulgaris’ with his views and ideas, and with the standard by which he judges a society based on capital and wages, we think we can perceive a change in the physiognomy of our dramatis personae. He, who before was the money-owner, now strides in front as capitalist; the possessor of labour-power follows as his labourer. The one with an air of importance, smirking, intent on business; the other, timid and holding back, like one who is bringing his own hide to market and has nothing to expect but - a hiding.11

In reality, one-man management is an ideology or apologetic representation of the right of the capitalist to hold the worker in subordination. The ‘single person’ cannot manage affairs without collaborators. The absolute Roman emperor required his consilium or later consistory (the latter imitated by the papacy); the ‘single person’, Cromwell, needed a protectoral council; the Stalinist bureaucracy after Stalin was able to speak about ‘collegial management’.12 Very recent British political events have shown that Theresa May, in spite of her formal electoral mandate, was unable to carry her cabinet with her after a series of parliamentary defeats, and was forced to resign. The Führerprinzip - the ‘principle’ that there has to be an individual leader - proves to have inherent limits discreetly hidden by the media’s fetishes of leaders.13


In spite of this ideological aspect, we are here at the core of the issue. We need to organise to conduct collective political action. This need has two sides, which are not obviously in tension with each other, but are, in fact, in tension in a more complicated way.

The first side is that we need to have means of taking collective decisions about what to do in our common actions. The second is that we need to actually mobilise the labour, skills, creativity and resources of the members of a party (or a campaign) in order for that party to be able to counterweight the money resources available to big capital on the international and national scale, and to small capital on the local scale.

We need to have means of taking collective decisions about what to do in our common actions. Disagreement is normal for human beings and hence for workers’ organisations. Without decisions, there can be no common actions. This is on its face a completely banal point. The question is what means of deciding.

For some purposes snap decisions do indeed require that we delegate power to one individual, if we are to get a decision at all. A single example from my own experience is a mass picket-line in Oxford in the 1970s, which was attacked by the far right. There needed to be (but was not) a chief steward to decide how we should respond to this attack.

At the opposite extreme, there are some issues which it is educational to discuss without the need for a collective decision following the debate (for example, whether there was ever an ‘Asiatic mode of production’; or whether Soviet Russia was ‘state capitalist’). And others - for example, redrafting a party programme - can be conducted usefully over a prolonged period, before eventually going to the vote, rather than being forced to a quick decision.

Further, there are a variety of possible methods of decision-making beyond delegating authority to an individual - general members’ assemblies, delegate conferences, elected committees, committees appointed by sortition (juries) and so on.

It is a common error among the left to suppose that there is some particular vice in the ‘democratic centralist’ method of the election of committees, which in turn have control over individual officers; and that it would be better to decide by direct election of officers by the membership. But the latter is to make vesting authority in individuals the dominant mode of decision-making - it is hard to get rid of directly elected officers and hence hard to control their abuses.

The objection to ‘vanguardism’ is somehow not seen to apply to the direct election of officers. But in fact this method of proceedings is more ‘vanguardist’ than the election of leading committees: it is transparently managerialist.

It also drives towards the sort of undesirable forced choices posed in constitutions with directly elected presidencies: ‘Vote for the crook [Jacques Chirac], not the fascist [Jean-Marie Le Pen]’ in France in 2002; ‘Vote Clinton to defeat Trump’ in 2016 - and get Trump. Equally, Neil Kinnock was the ‘realistic left’ candidate for Labour leader in 1983 - and proceeded once elected to witch-hunt the Militant Tendency.

This second objection holds all the more for the idea of making all decisions through referendums. We in this paper have written at length against referendums in the last few years, and I do not propose to repeat more than the elementary point that the referendum cannot handle any question which requires more than a ‘yes’ or ‘no’. And, even then, such questions - like ‘Should the UK leave the EU?’ - can turn out to involve ulterior questions which make the apparent simplicity of the referendum a scam.

The collapse of the ISO actually provides another example. The process of decision-making took the form of prolonged all-members conference calls - which will undoubtedly have been dominated by those who for one reason or another were most willing to push their particular agendas - followed by an online poll: that is, a referendum.14 Now it might be that a different method of decision-making would still have produced the same result. But it is also possible that another method of proceeding would have allowed those who wanted to salvage some form of collective action out of the failure of the ISO to have caucused, formulated proposals and progressed towards organising. The referendum - even if it is not to become a ‘neverendum’ - is actually a disorganising method of decision-making. I note merely that this was already one of Karl Kautsky’s objections to it in 1893.15

If the direct election of officers is transparently managerialist, referendums are non-transparently managerialist. The question is - who writes the question put to the voters (or the members, and so on)? Usually, this will be held out as a ‘technical’ problem: given to the political manoeuvrer ‘back-room boys’, but in reality it is a decisive issue.

Equally non-transparently managerialist is decision-making by consensus. A real requirement of consensus - that is, that no decision can be taken without unanimity - implies that very few decisions can be taken at all. This is fine for the property-owning class and their managers, who are thereby set free to take their decisions at the expense of the rest of society, but useless for a workers’ organisation, which demands common action. More commonly, ‘consensus’ means, as in the World and European Social Forums, decision-making in which certain favoured groups have a veto. This is practice means non-transparent decision-making by the full-time apparat of the favoured groups: of the Brazilian Workers’ Party in the World Social Forums, of Rifondazione Comunista in the early European Social Forums, of the London mayor’s office in the London ESF.

‘Consensus’ is, in fact, merely an extreme version of the entrenched rules and supermajority arrangements found in the constitution of the USA, the European Union treaties and so on. Such arrangements are favoured by capital because these ‘counter-majoritarian’ devices work, precisely, to secure veto powers for property-owners and their managers.


I have already said that a workers’ political party fundamentally needs to be able effectively to mobilise its members: it is its members’ work, energy and creativity, as well as their mere numbers, which can counterbalance the monetary and other resources of big and small capital.

From one angle, this point is not in tension with the need for centralised decision-making; as I have said, no decision implies no common action. From another, however, there is a tension. The point is that we need to mobilise, as I said, members’ work, energy and creativity. Merely showing up as a member of rentacrowd is doing no more than the capitalist parties have done before the workers’ movement emerged, and continue to do in the spectacles of Republic and Democrat conventions and Tory conferences.

As long as workers’ party organisation is conceptualised as copying these practices of leaders and their mere followers, the capitalist parties will inevitably win - capital, behind them, has the resources. For the workers’ party to mobilise its members’ capabilities, the party has to be, and be grasped as, the members’ party - not their leaders’ party.

The consequences are not and cannot be expressed juridically as fixed rules, but only politically as constitutional conventions. The reason is that fixed juridical rules are by their nature the private property of the legal profession as a group, and/or of the formal adjudicators as individuals. Witness (for example) the nature of the Labour Party’s disciplinary procedure and the totally arbitrary quality of which rules are enforced and which are not.

What conventions, then? I draw attention to two in particular. The first is that - as I have already said - disagreement is normal. The watchword is ‘Freedom of discussion, unity in action’. The existence of disagreements and efforts to persuade others do not constitute undesirable ‘factionalism’. Nor are sharp expressions of disagreement ‘abuse’ or ‘sectarianism’. A party should seek to draw out and clarify its disagreements in the decision-making process. Conversely, dissentients have a duty to raise their views within party channels, rather than merely walking out.

This point carries with it a second. The common left idea that sectarianism is produced by separation from ‘the mass movement’ is displayed as a delusion by the profound sectarianism of the left within the Labour Party and trade unions. The idea that to reject ‘small parties’ would be to reject ‘sectarianism’ (Louis Proyect) is a branch of the same error. It actually produces merely a dynamic towards sects of one member and managerialist control by full-time trade union officials, etc.

The second convention is that central decision-making is for central tasks. It does not imply central control of local or sectoral tasks, except insofar these become central (as Schippel, by supporting naval budgets, became not merely a local Reichstag member, but an icon of the capitalists’ argument for a loyalist socialist party).

Making the party the members’ party needs not just democratic, central decision-making, but also the ability of the members to self-organise and self-publish locally and sectorally. Democratic centralism thus has implications for the election of local full-timers - by the locals, not by the centre; and for the division of money resources between centre and localities, which was one of the persistent issues debated in the SPD’s organisation debates, because if the centre is starved of resource there can be no central action - but the localities also need resources to be able to act autonomously.

‘Efficiency gains’ may seem to be available from managerialism as an alternative to democratic centralism (one, well-designed and glossily printed, leaflet for the whole country, perhaps?), but the costs outweigh the gains.

Through the combination of central and local decision-making the membership are mobilised in their party and can outweigh the capitalists’ resources. Though imitating the capitalists - whether by direct managerialism or by indirect managerialism through individualistic or ‘constitutionalist’ blocks on collective decision-making - the workers’ organisation tends to be demobilised and its solidarity broken up.

From this point of view the collapse of the ISO is not a warning against democratic centralism, but against its negations.

  1. ‘Transparency and solidarity’ Weekly Worker April 4; ‘Full-timers and cadre’, April 25.

  2. ‘Reclaiming democratic centralism’ Weekly Worker May 23. I should have credited Ben Lewis’s ‘Sources, streams and confluence’ (Weekly Worker August 25 2016) as providing a slightly different take on the same history, but somehow lost the reference. Apologies to Ben!

  3. Eg, ‘Toward a new beginning - on another road’ (1971): www.marxists.org/archive/draper/1971/alt/alt.htm.

  4. A convenient summary can be found at https://en.wikipedia.org/wiki/City_of_London_Corporation.

  5. Cf H Draper Karl Marx’s theory of revolution Vol 4: Critique of other socialisms New York 1990, chapter 5; and also my review of Iain McKay’s collection of Proudhon’s writings, ‘No guide to revolution’ (Weekly Worker July 19 2012).

  6. ‘A critique of the German Social Democratic programme’ (1870): www.marxists.org/reference/archive/bakunin/works/1870/letter-frenchman.htm.

  7. www.marxists.org/archive/marx/works/1871/letters/71_11_23.htm. The “September revolution in France” refers to the fall of the regime of Napoleon III in September 1870, when some ‘left’ radicals were included in the republican provisional government to give it a spurious appearance of political breadth. The “game messrs Gladstone and co are bringing off in England even up to the present time” was the Liberals holding themselves out as backers of the workers, most immediately through the Trade Union Act 1871, which attempted to legalise trade unions (Tory judge Brett J in the 1872 Gas workers’ case merely invented a new theory of the illegality not covered by the act in order to defeat its effects).

  8. www.marxists.org/archive/marx/works/1874/04/bakunin-notes.htm.

  9. www.marxists.org/reference/archive/bakunin/works/1870/albert-richard.htm. I am entirely aware that this topic is the subject of a large and violent debate between Marxists and anarchists: for a single example, see my ‘Bakuninist hatchet job’ (Weekly Worker February 18 2016) and René Berthier’s response: ‘About Mike Macnair, “social democracy and anarchism” and hatchets’ (www.anarkismo.net/article/29301). My point is that ‘direct actionists’ are driven towards forms of bureaucratic centralism; and that the ‘invisible dictatorship’, though it never got sufficiently significant to be an actual bureaucratic centralism, was Bakunin’s partial recognition of something which he otherwise simply failed to recognise.

  10. Political Parties, Elections and Referendums Act 2000, section 24 (1).

  11. Chapter 6: www.marxists.org/archive/marx/works/1867-c1/ch06.htm.

  12. Consilium or consistory; see F Millar The emperor in the Roman world London 1992, pp83-131; Cromwellian council: P Gaunt, ‘“The Single person’s confidants and dependants”? ‘Oliver Cromwell and his protectoral councillors’ Historical Journal Vol 32, pp537-60 (1989); bureaucracy: see, for example, M Kozlov, ‘The relationship between collegial and one-man management in Soviet state administration at the present stage of development’ Soviet law and government Vol 3, pp3-12 (1964).

  13. Equally misleading are Lenin’s arguments for “one-man management” in Soviet industry in 1918-20. One example (chosen merely for being high up among Google results for the phrase) is ‘Speech delivered at the Third All-Russia Trade Union Congress, April 7 1920’ CW Vol 30, pp502-15. The problems these arguments addressed were, in reality, specific to Russian backwardness, not symptoms of issues general to the construction of socialism.

  14. http://socialistworker.org/2019/04/02/the-isos-vote-to-dissolve-and-what-comes-next.

  15. B Lewis, ‘Kautsky on referenda’; and Karl Kautsky, ‘Direct legislation by the people and the class struggle’ (Weekly Worker March 31 2016).