On party democracy

Issue: 124

John Molyneux

It really needs to be said that the first things to be forgotten are just the first points, the most elementary things… The first point is that there do in fact exist rulers and ruled, leaders and led… In the formation of leaders the premise is fundamental: does one wish there always to be rulers and ruled, or does one wish to create the conditions where the necessity for the existence of this division disappears?… Nevertheless, it needs to be understood that the division of rulers and ruled, though in the last analysis it goes back to divisions between social groups, does in fact exist, given things as they are, even inside the bosom of each separate group, even a socially homogeneous one…and it is mainly on this question that the most serious “errors” come about… It is believed that when the principle of the group is laid down obedience ought to be automatic…or even that it is beyond discussion… So it is difficult to rid the leaders of dictatorial habits, that is, the conviction that something will be done because the leader thinks it is correct and rational that it should be done: if it is not done, the “blame” is put on those who “ought to have” etc.1

This article examines the much disputed issue of internal party democracy in the light of the Marxist tradition and past and recent experience. It considers the challenge offered to the Marxist theory of the party by the German sociologist Robert Michels, in the belief that facing and attempting to answer this challenge yields insights into the real nature of the problem. On this basis it seeks to reformulate the issue of party democracy in a way different from that in which it is usually posed on the left, namely as a goal continuously to be striven for rather than a norm simply to be observed.

The tradition

When, in 1847, Karl Marx and Frederick Engels joined the League of the Just, an international secret society composed mainly of German artisans, which duly became the Communist League and for which they composed the Communist Manifesto, it was on condition that, among other things, “the organisation itself was thoroughly democratic, with elective and removable authorities”.2

This condition was central to the new worldview and political practice that they had persuaded the League to adopt, namely the doctrine of class struggle outlined in the Manifesto, in which the transformation of society—the revolution—was to be accomplished by the working class itself. Prior to Marx and Engels, and prior to the emergence of the modern working class, the dominant form of revolutionary organisation was the secret club or conspiracy—a model inherited from the French Revolution of 1789—which envisaged the transformation of society from above following a coup d’etat by a dedicated and enlightened few. Marx and Engels considered party democracy a crucial means for combating this, essentially bourgeois, concept of revolution. “This alone [a democratic party structure] barred all hankering after conspiracy, which requires dictatorship”.3

The commitment to party democracy was inscribed in our tradition from its inception and has been reinscribed many times since.

“Unity of action, freedom of discussion and criticism. Only such discipline is worthy of the democratic party of the advanced class…the proletariat does not recognise unity of action without freedom to discuss and criticise,” wrote Lenin.4 Leon Trotsky wrote of the Bolshevik Party:

The party took watchful care not only that its boundaries should always be strictly defined, but also that all those who entered these boundaries should enjoy the actual right to define the direction of the party policy. Freedom of criticism and intellectual struggle was an irrevocable content of the party democracy. The present doctrine that Bolshevism does not tolerate factions is a myth of the epoch of decline. In reality the history of Bolshevism is a history of the struggle of factions. And, indeed, how could a genuinely revolutionary organisation, setting itself the task of overthrowing the world and uniting under its banner the most audacious iconoclasts, fighters and insurgents, live and develop without intellectual conflicts, without groupings and temporary factional formations?5

Similarly, Duncan Hallas in 1971:

[A revolutionary socialist party] cannot possibly be created except on a thoroughly democratic basis; unless, in its internal life, vigorous controversy is the rule and various tendencies and shades of opinion are represented, a socialist party cannot rise above the level of a sect. Internal democracy is not an optional extra. It is fundamental to the relationship between party members and those amongst whom they work.6

It should be added that this is not just a question for Marxist revolutionary parties. The overwhelming majority of working class organisations—trade unions, parties, campaigns, cooperatives, tenants associations, etc—are at least established with democratic constitutions and aspirations. The working class is the socialist class because, among other things, it is the democratic class. The economic role and position of the working class, making and operating the decisive means of production, producing the bulk of its wealth, concentrated in large workplaces and towns and integrated into a global division of labour, gives it (a) the power to defeat the capitalist class; (b) the ability to inaugurate the transition to a fully classless society because it can be both the producing and the ruling class at the same time. In other words, it can create structures (the Paris Commune, soviets, factory councils, etc) that enable it to control production and govern society democratically.

The experience

And yet… And yet everyone who knows something of the history of the socialist movement and anyone who has had any experience of it over the last 30 or 40 years knows that the question of inner party democracy has been an ever-recurring issue.

The Stalinist parties provide the most obvious example. In the Communist Party of the Soviet Union every shred and semblance of democracy was extinguished by Stalinist counter-revolution. Communist parties in Eastern Europe, China, North Korea, Vietnam, etc more or less followed suit, and if the Communist parties of Western Europe had a less draconian character, not having a secret police to hand, they nevertheless were not remotely democratic.

However, for us now, the undemocratic nature of the Stalinist parties poses the least theoretical problems. It is clear that in the Stalinist states the Communist parties had become instruments of rule, not of the working class, but of new bureaucratic ruling classes and therefore had to be undemocratic in the extreme. This was the only way of overcoming the contradiction between their actual function and their declared aims.7

Something similar applies to the traditional social democratic parties with their dual character as bourgeois workers’ parties (as Lenin put it)—ie as parties with working class membership, and bourgeois leadership and policy. In these cases their mass working class bases obliged these parties to retain some elements of democracy (conferences with some real debate, elected leaders, etc) but also to develop mechanisms capable of preventing the worker majority from asserting itself. In the case of old Labour (New Labour became even less democratic) the two most important of these mechanisms were the block vote in the hands of the trade union bureaucrats and the ability of the parliamentary leadership to ignore decisions of party conference.8

More theoretically challenging is the recurrence of the democracy issue in avowedly non-Stalinist, non social democratic or libertarian organisations. Let us take a few examples drawn from throughout the history of the socialist movement. First there was the conflict between Marx and the anarchist Mikhail Bakunin in the First International. Bakunin denounced Marx’s authoritarianism and called him a “dictator over the proletariat”. Marx replied by accusing Bakunin of operating a secret society inside the International (the International Brotherhood, also known as the International Alliance of Socialist Democracy) in which there was no democracy of any kind but rather an unelected “collective and invisible dictatorship” of Bakunin.9 During debates over the split between the Bolsheviks and the Mensheviks in 1903 Trotsky attacked Lenin, saying, “In the internal politics of the party these methods lead…to the party organisation ‘substituting’ itself for the party, then the central committee substituting itself for the party organisation; and finally the dictator substituting himself for the central committee”.10 At the same time Rosa Luxemburg upbraided Lenin for his “overanxious desire to establish the guardianship of an omniscient omnipotent central committee”.11

The international Trotskyist movement in the 1930s was plagued with complaints, disputes and splits over real or alleged violations of inner party democracy, and Trotsky was frequently led wearily to lament petty bourgeois elements who wanted to debate and discuss forever. In the most important of these splits, that in the American Socialist Workers Party in 1939-40, the question of the “party regime” presided over by James P Cannon played a considerable role.12

After the Second World War the pattern on the Trotskyist left remained broadly the same. In virtually every faction fight or split, and they were far too numerous to document here, the issue of inner party democracy would raise its none too beautiful head. At one end of this particular spectrum lay the Socialist Labour League/Workers Revolutionary Party, which, from reasonably promising beginnings around 1956, degenerated into the ever more erratic personal dictatorship of its “philosopher” thug leader Gerry Healy. Healy deployed a combination of an idealist version of the dialectic and highly materialist fists to terrorise intellectuals and worker militants alike until it all fell apart in 1986, amid accusations of the systematic sexual abuse of female comrades. At the other end was the “ultra-democratic” International Marxist Group (IMG), British section on the Unified Secretariat of the Fourth International, and followers of Ernest Mandel. The IMG went so far in the opposite direction that by the 1970s there were institutionalised permanent factions (at least three at any one time), none of which had a working majority. At this point ultra-democracy seems to have turned into its opposite in that the majority of the members were unable to assert any kind of stable strategy or line.

Between these extremes lay a multitude of small Trotskyist or semi-Trotskyist groups with a wide variety of organisational practices and internal regimes. However, the general isolation of such groups from the mass of the working class meant that the tendency to become a sect, dominated in practice by a few individuals, was pretty strong.

The most promising development in these years (in my opinion, of course) in terms of its politics, growth and democratic openness was the International Socialists (IS), which later became the Socialist Workers Party (SWP), in Britain. I had joined the Socialist Labour League Young Socialists in January 1968, only to leave shortly thereafter, repelled by its terrifying authoritarianism. I joined the IS in about June of that year and at my first conference found the organisation divided into at least five (!) short-lived tendencies or factions. Despite some serious splits in the leadership in 1975-6 and 1979 (basically over coming to terms with the downturn in industrial struggle) the overall tendency was for the central committee steadily to increase its hegemony within the organisation. But there were always concerns being raised, rightly or wrongly, about the issue of internal democracy—at one point there was actually a Faction for Revolutionary Democracy (known as FRED).

The purpose of this brief and superficial survey is not to engage in retrospective “democratic” judgment (for Marx against Bakunin, with Lenin against Trotsky and Luxemburg, or 0/10 for Healy, 6/10 for Mandel, or whatever) but simply to demonstrate that the problem of inner party democracy is a recurring one and it is therefore necessary to consider its social basis.

The problem

For Marxists, as historical materialists, this ought to be the obvious starting point but this aspect of the problem has received relatively little attention. On the one hand, it has frequently been asserted that the leadership of small revolutionary group X do not constitute a materially privileged bureaucratic stratum—in the sense that the trade union and Labour leaders, or the Communist Party leaders in Stalinist states, clearly do—and that therefore it is unmaterialist to think there can be a serious problem over democracy in such a group. On the other hand, Trotsky provided a sociological explanation as to why “petty bourgeois elements” would be prone to make unreasonable demands for excessive democracy:

A worker spends his day at the factory. He has comparatively few hours left for the party. At the meetings he is interested in learning the most important things: the correct evaluation of the situation and the political conclusions. He values those leaders who do this in the clearest and the most precise form and who keep in step with events. Petty bourgeois, and especially declassed elements, divorced from the proletariat, vegetate in an artificial and shut-in environment. They have ample time to dabble in politics or its substitute. They pick out faults, exchange all sorts of tidbits and gossip concerning happenings among the party “tops”. They always locate a leader who initiates them into all the “secrets”. Discussion is their native element. No amount of democracy is ever enough for them.13

In contrast Duncan Hallas provided an illuminating account of how the degeneration of inner party democracy can result from a false perspective. He showed how the aforementioned Socialist Labour League ruined itself through its continued adherence to a perspective of imminent economic catastrophe (drawn directly from Trotsky’s 1938 writings) throughout the prolonged boom of the 1950s and 1960s:

Discussion, which is dangerous to the leadership, can be checked by hyperactivity; and this, in turn, is justified by the nearness of crash. The membership, driven at a frenzied pace, has a high casualty rate. A large proportion is always new—and therefore does not remember the non-fulfilment of past prophecies. A vicious circle is set up which makes the correction of the line more and more difficult. “Building the leadership”—which is, of course, identified with the organisation—becomes a substitute for serious political and industrial work. Serious militants are repelled and the “revolutionary youth” come to make up an ever-larger proportion of the activists. The leadership, which alone has much continuity, becomes unchallengeable and finds it less and less necessary to check its policies and practice.14

The tendency that manifested itself in grotesque form in the case described by Hallas could also manifest itself in more moderate forms in much saner organisations. The argument, however, can be reversed and used to defend complacency. Since the perspective is “correct” it can be said that there is no need to worry about democracy. If the function of party democracy is to achieve a correct perspective, the claimed existence of a correct perspective could either render democracy superfluous or constitute proof of its good health.

But this line of argument is flawed. The causal relation may flow, not from false perspective to lack of democracy, but from lack of democracy to false perspective. Also it is mistaken to base the case for party democracy solely on its providing the necessary inputs and checks to control the perspective. Democracy is also necessary to educate and train the members to argue for socialism in the class15 and to retain “ownership” of the party by its members and, ultimately, the working class.

The challenge from Michels

A different but interesting light is thrown on this whole question by Robert Michels in his book Political Parties, first published in 1911. Michels was a pupil of Max Weber and a member of German Social Democracy prior to the First World War who, after the war, became a supporter of fascism. In Political Parties Michels propounded his so-called “iron law of oligarchy” according to which any large-scale political party or organisation will inevitably be ruled by a small self-perpetuating elite at its centre: “It is organisation that leads to the domination of the elected over the electors… Who says organisation says oligarchy”.16

Hallas comments:

The equation “centralised organisation equals bureaucracy equals degeneration”…leads to profoundly reactionary conclusions. For what is really being implied is that working people are incapable of collective democratic control of their own organisations. [This] is to argue that socialism is impossible because democracy, in the literal sense, is impossible. This is precisely the conclusion that was drawn by the “neo-Machiavellian” social theorists of the early 20th century [eg Michels] and which is deeply embedded in modern academic sociology.17

Hallas is absolutely right (and in becoming a fascist Michels was, in a sense, following his own logic) but pointing out the reactionary conclusions of an argument is not the same as refuting it, and the fact is that Michels makes a powerful case, based precisely on his knowledge of the socialist movement, particularly in Germany.

Michels argues that there exists a “mechanical and technical impossibility of direct government by the masses” and observes:

It is easier to dominate a large crowd than a small audience…enormous public meetings commonly carry resolutions by acclamation or by general assent, whilst these same assemblies, if divided into small sections, say, of 50 persons each, would be much more guarded in their assent… The impotence of direct democracy, like the power of indirect democracy, is a direct outcome of the influence of number…in the great industrial centres where the labour party sometimes numbers its adherents by tens of thousands, it is impossible to carry on the affairs of this gigantic body without a system of representation. The great socialist organisation of Berlin…has a member roll of more than 90,000.18

He then describes the tendency of the representatives to establish their independence from and hegemony over the represented. He notes “the establishment of a customary right to the office of delegate” and the need for leadership felt by the masses including “the cult of veneration among the masses”.19 Michels cites especially Ferdinand Lassalle, but also Giuseppe Garibaldi, Jules Guesde and Marx; today, of course the list could be much longer and more grotesque.

Michels notes the general cultural and educational superiority of the professional leaders over the rank and file, and he comments, “No one who studies the history of the socialist movement in Germany can fail to be greatly struck by the stability of the group of persons leading the party”. 20

Moreover, he argues, the leaders actively strive to maintain their position:

As the chiefs become detached from the mass they show themselves more and more inclined, when gaps in their own ranks appear, to effect this not by popular election, but by co-optation, and also to increase their own effectiveness wherever possible by creating new posts upon their own initiative. There arises in leaders a tendency to isolate themselves, to form a sort of cartel, and to surround themselves, as it were, with a wall, within which they will admit those only who are of their own way of thinking.21

Michels analyses the role of “bourgeois elements in the socialist leadership” and “labour leaders of proletarian origin”,22 while noting the “psychological metamorphosis of the leaders”:

The average leader of the working class parties is morally not lower, but on the whole higher, in quality than the average leaders of the other parties… Yet it cannot be denied that the permanent exercise of leadership exerts upon the moral character of the leaders an influence which is essentially pernicious.23

Michels also explains why neither syndicalism nor anarchism, for all their rhetoric, are able to overcome these oligarchic tendencies, citing, among others, the familiar example of Bakunin and his secret unelected dictatorship inside the First International.24 However, underpinning all Michels’ observations, acute as many of them are, and running like a thread throughout Political Parties is what he calls “the formal and real incompetence of the mass”.25 The “iron law of oligarchy” is, for him, an iron law because the masses are inherently incapable of running their own organisations or democratically controlling their leaders:

The incompetence of the masses is almost universal throughout the domains of political life, and this constitutes the most solid foundation of the power of leaders. The incompetence furnishes the leaders with a practical and to some extent a moral justification.26

And we should be clear, for Michels this incompetence is innate and general, as is the drive of leaders to dominate. It is a question of human nature. Almost certainly, we see here the influence of Friedrich Nietzsche’s “will to power”:

The apathy of the masses and their need for guidance has as its counterpart in the leaders a natural greed for power. Thus the development of the democratic oligarchy is accelerated by the general characteristics of human nature… The desire to dominate for good or evil, is universal…every human power seeks to enlarge its prerogatives.27

The Marxist response

This, of course, is where a Marxist response to, and critique of, Michels must begin. The various facts, tendencies and patterns of behaviour observed by Michels are not universal or general characteristics of human nature but products of class society in general and capitalist society in particular. And it is precisely on this point that the Bolshevik Nikolai Bukharin took issue with him:

What constitutes an eternal category in Michels’ presentation, namely, the “incompetence of the masses”, will disappear, for this incompetence is by no means a necessary attribute of every system; it likewise is a product of the economic and technical conditions, expressing themselves in the general cultural being and in the educational conditions. We may state that in the society of the future there will be a colossal overproduction of organisers, which will nullify the stability of the ruling groups.28

But here Bukharin is treating Michels’ book, which he describes as “very interesting”, primarily as an objection to the possibility of a future classless society. However, in relation to the transition to full socialism Bukharin accepts that Michels points to real problems, though he remains confident they can be overcome.

But the question of the transition period from capitalism to socialism, ie the period of the proletarian dictatorship, is far more difficult. The working class achieves victory, although it is not and cannot be a unified mass. It attains victory while the productive forces are going down and the great masses are materially insecure. There will inevitably result a tendency to “degeneration”, ie the excretion of a leading stratum in the form of a class-germ. This tendency will be retarded by two opposing tendencies: first, by the growth of the productive forces; second, by the abolition of the educational monopoly. The increasing production of technologists and of organisers in general, out of the working class itself, will undermine this possible new class alignment. The outcome of the struggle will depend on which tendencies turn out to be the stronger.29

Clearly, writing in 1921, the question of the transition to socialism was uppermost in Bukharin’s mind, but for our purposes he is not addressing the key point. To argue that the oligarchic pressures derive from capitalism, not human nature, and will therefore be overcome post-capitalism, in the transition to socialism, is all well and good, but the revolutionary socialist party has to be built under capitalism.

Of course, it would be possible to claim that Michels’ analysis was based on the parties of the Second International—mass reformist parties built in a period of relative social peace in which a bureaucratic hierarchy of paid officials developed. In that case it would be of no relevance to small revolutionary groups, where no substantial materially privileged bureaucracy exists. However, although the anti-democratic pressures are less powerful in small revolutionary groups than in mass reformist parties, they still exist. The workers who join such groups are still products of the profoundly anti-democratic socialisation received in capitalist society, which is reinforced by every defeat we suffer at the hands of the bosses or the government.

The pressures generated by bourgeois society are also a factor in the cult-like features exhibited by some small groups. In order to maintain the loyalty and discipline of a tiny number of adherents in more or less total opposition to wider society they develop the sort of characteristics typical of small religious sects such as the veneration of “the leader” and the establishment of shibboleths. A shibboleth was originally a code word or phrase whose use distinguished the member of a group from an enemy or spy. In our context a shibboleth is a belief or doctrine whose principal function is to separate the true believer from the common herd and reinforce their loyalty. This is, for example, the function of the ban on blood transfusions for Jehovah’s Witnesses or support for Israel for the Alliance for Workers’ Liberty. With the Socialist Labour League it was the perennial call for a general strike. These sect practices are highly anti-democratic because they strongly inhibit free and rational debate of policies and perspectives.

The adoption of a false perspective, especially an exaggerated ultra-left perspective, greatly intensifies the need for a leadership cult and shibboleths, but this does not make the false perspective the sole source of undemocratic tendencies. Rather we need to understand that the social origin of real democracy in the socialist movement is the struggle of the working class. Party democracy is likely to suffer in so far as the party is cut off from that source—whether because it has set itself above the class (labour and trade union bureaucrats), because it has been driven to the margins of the class (Trotskyism in the 1930s), because it has cut itself off from the class or because the level of class struggle is low.

When I first considered the challenge posed by Michels (while researching for Marxism and the Party in the early 1970s) I concluded that, while the iron law of oligarchy was generally valid for social democratic type parties, trade unions and similar organisations, it did not apply to Bolshevik-type parties. This was because central to the Bolshevik model was the restriction of party membership to (a) those who placed the overall interests of the working class above any sectional interest (ie were internationalist, anti-racist, anti-sexist and non-sectionalist), (b) militant activists working under the discipline of party organisations. Such a membership, though necessarily a minority of the class, would not be “incompetent” and would be able to democratically control its leadership. This seemed to me sound in theory and confirmed in practice by the example of the Bolshevik Party, which was highly democratic, especially in 1917.

In making this judgement I was much influenced by Tony Cliff who argued that the trade union bureaucracy, with its conservative and undemocratic practices, rested on the passive majority of the union membership as opposed to the active minority. Chris Harman also argued in his 1969 article “Party and Class” that there was a fundamental difference between the Leninist and social democratic models of organisation. And, he argued:

Within Lenin’s conception those elements that he himself is careful to regard as historically limited and those of general application must be distinguished. The former concern the stress on closed conspiratorial organisations and the need for careful direction from the top down of party officials, etc: “Under conditions of political freedom our party will be built entirely on the elective principle. Under the autocracy this is impracticable for the collective thousands of workers that make up the party.”

Of much more general application is the stress on the need to limit the party to those who are going to accept its discipline. It is important to stress that for Lenin (as opposed to many of his would-be followers) this is not a blind acceptance of authoritarianism. The revolutionary party exists so as to make it possible for the most conscious and militant workers and intellectuals to engage in scientific discussion as a prelude to concerted and cohesive action. This is not possible without general participation in party activities…

Centralism for Lenin is far from being the opposite of developing the initiative and independence of party members; it is the precondition of this… “Discipline” means acceptance of the need to relate individual experience to the total theory and practice of the party. As such it is not opposed to, but a necessary prerequisite of the ability to make independent evaluations of concrete situations. That is also why “discipline” for Lenin does not mean hiding differences that exist within the party, but rather exposing them to the full light of day so as to argue them out.30

At that time Harman’s perspective, and mine, was to build Leninist parties of this kind in Britain and internationally, in the belief that the objective conditions had developed which made this possible. Unfortunately experience, nearly 40 years of it, has shown this to be a rather complicated business, and this in turn has implications for the question of party democracy. Between Bolshevism’s birth as a faction of the Russian Social Democratic Labour Party in 1903 and its conquest of power as a mass party in October 1917 lay only 14 years. Those years included a revolution in 1905, a catastrophic world war and the revolutionary overthrow of Tsarism in February 1917. In contrast the IS/SWP tradition has endured almost 60 years, 40 since our emergence as a mini-party in 1968, without any comparable experiences. In Russia the period of reaction (which was intense and nearly destroyed the Bolsheviks) began in 1907 and lasted five years before being swept aside by mass strikes from 1912. In Britain class struggle entered a downturn in the mid to late 1970s and, with some fluctuations and partial recovery, has remained at a low level to this day. Moreover, while the level of struggle has generally been low, there has not been any sustained or generalised repression. And, of course, similar conditions have applied for all left organisations in Western Europe and North America over this period.

Lenin devotes the early chapters of Left Wing Communism—An Infantile Disorder to expounding the relationship between Russian conditions and experience and Bolshevik organisational principles (Lenin focuses on “discipline” but, as we have seen, discipline and democracy are intimately connected):

As a current of political thought and as a political party, Bolshevism has existed since 1903. Only the history of Bolshevism during the entire period of its existence can satisfactorily explain why it has been able to build up and maintain, under most difficult conditions, the iron discipline needed for the victory of the proletariat… Bolshevism…went through 15 years of practical history (1903-17) unequalled anywhere in the world in its wealth of experience. During those 15 years, no other country knew anything even approximating to that revolutionary experience, that rapid and varied succession of different forms of the movement—legal and illegal, peaceful and stormy, underground and open, local circles and mass movements, and parliamentary and terrorist forms.31

And Georg Lukács comments:

The Bolshevik concept of party organisation involved the selection of a group of single-minded revolutionaries, prepared to make any sacrifice, from the more or less chaotic mass of the class as a whole… _Lenin’s concept of party organisation presupposes the fact—the actuality—of the revolution… Had the historical predictions of the Mensheviks been correct, had a relatively quiet period of prosperity and of the slow spread of democracy ensued, in which—at least in backward countries—the feudal vestiges of “the people” had been swept aside by the “progressive” classes, the professional revolutionaries would have necessarily remained stranded in sectarianism or become mere propaganda clubs. The party, as the strictly centralised organisation of the proletariat’s most conscious elements—and only as such—is conceived as an instrument of class struggle in a revolutionary period. “Political questions cannot be mechanically separated from organisation questions,” said Lenin, “and anybody who accepts or rejects the Bolshevik party organisation independently of whether or not we live at a time of proletarian revolution has completely misunderstood it”.32

Unfortunately the times we have lived through have not, a few exceptional moments apart, been a revolutionary period and the revolution has been “actual” only in the most abstract sense. It has therefore not been possible to restrict party membership to “a group of single-minded revolutionaries, prepared to make any sacrifice”, or even to people prepared to be consistently active in party organisations. When Harman wrote, “The party is not to be made up of just anybody who wishes to identify himself as belonging to it, but only those willing to accept the discipline of its organisations. In normal times the numbers of these will be only a relatively small percentage of the working class; but in periods of upsurge they will grow immeasurably”,33 he and the rest of us envisaged a party of many thousands, perhaps tens of thousands, growing in time of revolution to hundreds of thousands (the Bolsheviks were about 26,000-strong in February 1917 and between 200,000 and 400,000 by October), not a group of tens or a few hundred. To have restricted the membership of the SWP to the criteria of commitment required by the Bolsheviks would, in our non-Bolshevik conditions, have reduced the party to the low hundreds at best and would anyway have been false “toy bolshevism”, since such fanatical “revolutionaries” would have lost the other key pillar of Leninism: the ability to “maintain the closest contact, and—if you wish—merge, in certain measure, with the broadest masses of the working people”.34 Consequently circumstances obliged us to operate with a substantial proportion of members who were not sufficiently engaged to exercise democratic control over the party.

The question of repression and illegality plays an interesting and contradictory role here. On the one hand, conditions of illegality create huge obstacles to the proper formal functioning of party democracy, to the convening of regular meetings and conferences, to the holding of regular elections, to the provision of accurate party statistics and so on. Moreover, illegality may strengthen the claims of discipline over democracy, insofar as sometimes they can conflict. On the other hand, repression resolves the problem of an engaged membership: if the penalty for party membership is possible exile to Siberia, or imprisonment, torture and death, the passive armchair member is taken care of and members have a massive, perhaps life or death, stake in the determination of party policy. Of course, this in itself offers no guarantee of democracy, as the example of many Stalinist parties shows, but it does help with one aspect of the problem.

Another factor that has to be considered here is the long timescale involved. Gramsci commented in his Prison Notebooks that a prolonged siege is always debilitating, and many people capable of intense resistance and extreme sacrifice for a short time are unable to sustain the same commitment over a long period. To this must be added the crucial role played by the level of the class struggle. A major element in Michels’ “incompetence of the masses” is not lack of technical competence but lack of confidence, which the capitalist socialisation process breeds in working class children as surely as it breeds confidence in the children of the bourgeoisie. The principal antidote to this lack of confidence, and therefore vital for internal party democracy, is the experience of collective resistance and struggle, of organising and leading strikes, picket lines, occupations, demonstrations, workplace branches and the like (in the case of students, speaking at mass union meetings, leading college occupations, etc). Not only does participation in such activity raise members’ confidence as individuals but it also means that when they take up an issue inside the party they often do as a representative of a collective in their workplace or college. In conditions of downturn, when party members’ typical experience at work is of defeat or isolation, their confidence to challenge the party’s leadership is undermined. Even if they remain active revolutionary socialists, the feeling may develop that in addition to fighting the bosses, the government, the system, the media and probably their own union leaders, all as a small minority, arguing in their own party is just too much.

In these conditions the counterpart to a passive rank and file, a leadership that becomes accustomed to leading unchallenged, is virtually certain to develop or at least begin to develop.

Having so strongly stressed the anti-democratic pressures at work on and in any would-be revolutionary party, it is necessary also to note that there is a major countervailing tendency. In opposition to Michels’ “iron law of oligarchy” there exists, in any party or organisation whose leadership does not wield the combined sticks and carrots of state power, an almost universal “law of democracy”. In any voluntary organisation where membership does not itself confer material privilege (even the British Tory party or the Nazi BNP) there is an element of democracy in that the leadership requires the consent of the rank and file in the form of its continuing membership and support. So even right wing union leaders, such as Joe Gormley of the National Union of Mineworkers in 1972-4 or Derek Simpson of Unite today, are obliged to defend their members’ interests and respond to their demands to some degree, on pain of losing members (and with them the dues that pay the leaders’ salaries).

This democratic pressure from below is all the stronger in a small far-left socialist party, even if it remains overtly passive, because (a) the leaders are plainly not motivated by desire for material privilege, there being none on offer (though sometimes the desire to maintain material security may be a factor); (b) the rank and file are motivated overwhelmingly by conviction; and© it is not difficult for them to vote with their feet and leave. To give an example from the organisation I know best, namely the SWP, it is well known that the party’s leadership remained pretty stable through the 1990s and early part of this decade, and was seldom subject to serious challenge. But this was on condition that it ran the party within the narrowly prescribed limits of Trotskyist revolutionary socialism. If the SWP leadership had, as opposed to making strategic or tactical errors, ever clearly crossed class lines or contravened basic socialist principles or made moves to renounce Marxism, there would undoubtedly have been an outcry and, if the outcry was not rapidly successful, a mass exodus. This is not to suggest that the SWP leadership ever wanted to do any of these things. I know of no evidence of this whatsoever. Nevertheless, the objective pressure was there and it was a democratic pressure—a pressure to lead the organisation in conformity with the wishes of its members. Moreover it is clear that in general this inherent democratic pressure will, given favourable conditions, be capable of considerable expansion.

Some conclusions

What conclusions follow from this analysis? First and foremost that the case for party democracy made by all the classical Marxists and cited at the start of this article retains all its validity, but it must be understood that the achievement of this democracy in a capitalist society is far from easy.

From a Marxist perspective it is possible not only to integrate the anti-democratic pressures cited by Michels but also to elaborate on them. Capitalist socialisation, through the family and school, instills obedience and subservience in working class children, practically from birth. A “good” baby is one who sleeps and feeds to order; a “good” child is “no trouble” and does what its parents tell it. A successful pupil is one who accepts the agenda of the school and its teachers.35 A “deviant” teenager has “a problem with authority”. Ruling class children are likewise trained to obey the rules but, especially via the public schools, their education also contains a significant stress on developing leadership qualities.

The world of work is invariably hierarchical and undemocratic. Working class occupations consist overwhelmingly of following orders, ruling class ones of giving them and middle class ones of enforcing decisions from above on those below. What is completely lacking from most people’s lives is any experience of democracy other than the extremely limited business of voting every so often in parliamentary or local elections. By far the most important exception is trade unionism, which does provide some working people with the experience of saying “no” to those in authority over them, but, as we know only too well, this is a highly uneven and fluctuating process and offers an ongoing regular democratic engagement to only a minority.

The act of joining a revolutionary organisation constitutes a major rebellion against society’s conditioning but it does not eliminate it. The anti-democratic pressures continue to operate on and within the party. This is why party democracy is not something that can be guaranteed by any constitution or set of institutional arrangements (which is not to gainsay the necessity of democratic constitutions and institutional arrangements) but also requires the development and maintenance of a democratic culture based on frank and open debate in which party members are encouraged to speak their mind. Such a culture has to be embodied in institutions and practices, of course, the most important and permanent of which is the principle of the party conference or congress as the party’s sovereign body. But the precise nature of these institutions and practices must necessarily be adapted to specific circumstances and change over time.

Nor is party democracy a political or moral norm that can be established simply by the will or good behaviour of its leaders (or members). For every would-be revolutionary party internal democracy is a goal, a relationship between members and leaders, that has to be continually striven for in the same sense that the correct relationship of party to class has to be continually striven for—indeed the two relationships are intimately connected. Deficiencies in democracy, like errors in perspective and tactics, are inevitable, but not reasons for despair. The point is to correct them.

In this ongoing struggle there are no grounds whatsoever for renouncing either the Leninist concept of the vanguard party or democratic centralism. On the contrary, the Leninist democratic centralist party is both necessary for the success of the revolution and the most democratic form of political organisation.

The case for the Leninist party rests on arguments in no way limited in their validity to Russia or to the historical period of the Russian Revolution. Rather these arguments refer to features of capitalist society and the working class struggle which are pretty much universal and permanent: (a) the centralised nature of the capitalist class and its state which demands centralisation on the part of its adversary; (b) the bourgeoisie’s ideological hegemony which requires the waging of an ideological struggle in society and within the working class, a struggle that can only be conducted by a party based on ideological and political clarity; (c) the unevenness in the levels of working class consciousness, confidence, organisation and struggle which require the welding together of the most advanced elements in the class to defeat the reactionary, scab elements and increase their influence over the vacillating majority; (d) the existence, on the basis of this unevenness, of mass reformist (or Stalinist or nationalist) parties which will hold back or betray the revolution, and which must be combated by a revolutionary party which retains its political and organisational independence. Moreover, these theoretical arguments, powerful in themselves, have been confirmed in practice positively, by the role of the Bolshevik Party in the victory of the Russian Revolution and, negatively, by the defeats in Italy, Germany, Spain and elsewhere precisely for the want of such a party.

As far as democracy is concerned, all other forms of political organisation—the social democratic party, the trade union, the “loose” anarchist or autonomist federation or clique, the single-issue campaign—are subject to exactly the same anti-democratic “oligarchic” pressures from capitalist society without the same resources to resist them. No other form of organisation compares with the Leninist party in terms of its ability to equip its members with the political education that enables them to assess the overall political situation and their own party’s work. No other form of organisation practises a comparable level of intervention in such a variety of issues, campaigns and struggles, thus potentially training its members as political generalists able to hold its leaders to account. I do not doubt that SWP branches have many defects but one would only have to compare the topics discussed at a typical SWP weekly branch meeting with those at an average Labour Party ward (do they still meet?) to get the point. At the former you might get the economic crisis one week, Palestine the next, followed by fascism and the BNP the week after; at the latter it would be more likely to be the jumble sale, the local pavements and who contests which seat in the local elections, if that.

Furthermore the element of party discipline inherent in democratic centralism—the notion of unity in action in implementing party policy—far from undermining or infringing democracy, is an essential democratic provision. Without it the party could engage in the most democratic process of debate and decision making only to see those decisions come to nothing when they were ignored or flouted by the party leadership, as was routinely the case with old Labour. This element in a revolutionary party is particularly important at decisive moments in the class struggle, especially that of insurrection, when the political and psychological pressures on party leaders are most intense.36

If the case for the Leninist party remains compelling, it would, however, be wrong to identify Leninism with one narrowly defined organisational model or set of practices. Lenin himself, in his last speech to the Comintern in 1922, while insisting on the international importance of the Bolshevik experience, warned against the mechanical imitation of Russian organisational methods.37 For example, on the question of factions there have been at least two views taken in recent years by avowed Leninist Parties—namely the British SWP and the French Ligue Communiste Révolutionnaire (LCR). Both would accept Trotsky’s statement, quoted earlier, that “the present doctrine that Bolshevism does not tolerate factions is a myth of the epoch of decline”.

But whereas the SWP has permitted only temporary factional organisation, partly to try to avoid the ingrained tendency of Trotskyist groups to split, the LCR has taken the opposite view and sought to maintain unity by allowing permanent tendencies to coexist in the party. I have alluded above to the fact that this policy did not work well for the LCR’s co-thinkers in Britain, the IMG, but the LCR, with their recent electoral success and launch of the New Anticapitalist Party, have fared much better. One important objection to their model is that it might lead to paralysis of the party leadership in crucial situations where speedy and decisive action is required. Nevertheless this is a debate which can be had between Leninists and which depends on concrete circumstances.

This last proviso applies to much of this debate. The principle of the need to struggle to realise and improve democracy in the revolutionary party is permanent but the precise means of achieving it vary over time. In general what is needed is, in Gramsci’s words, organic centralism, not bureaucratic centralism: “democratic centralism that is ‘centralism’ in movement, so to speak, that is, a continuous adjustment of the organisation to the real movement”.38 This means working, at each stage in the class struggle, to establish a political culture and set of political practices that facilitate democratic debate and decision making, and develop the political knowledge and confidence of the rank and file.

This is a problem that has to be solved and re-solved not on paper but in practice—and it will never be fully resolved this side of the overthrow of capitalism. Nevertheless a clear statement of the problem and an awareness of the challenge involved may help. That has been the aim of this article.


1: Gramsci, 1970, pp143-144.

2: Engels, 1970, p195.

3: Engels, 1970, p196.

4: Lenin, 1965a, pp230-231.

5: Trotsky, 1936, chapter five.

6: Hallas, 1971.

7: Clearly I am here taking for granted the analysis of the Stalinist states and the Comintern by Trotskyists and, later, Tony Cliff in his theory of state capitalism.

8: The best analysis of how such parties work is provided by Cliff and Gluckstein, 1988.

9: See Molyneux, 1978, p28.

10: Trotsky, 1979, p77.

11: Luxemburg, cited in Molyneux, 1978, pp98-99.

12: For Trotsky’s side of this dispute see Trotsky, 1975. For Cannon’s see Cannon, 2001. Commentators from this journal’s tradition have tended to side with Trotsky and Cannon, but to have some sympathy with the opposition on the “regime” question-see Cliff, 1993; Bambery, 1987.

13: Trotsky, 1975.

14: Hallas, 1969.

15: This point was powerfully made by Isaac Deutscher: “When the European Communist went out to argue his case before a working class audience, he usually met there a social democratic opponent whose arguments he had to refute… Most frequently he was unable to do this, because he lacked the habits of political debate, which were not cultivated within the party, and because his schooling deprived him of the ability to preach to the unconverted”-cited in Hallas, 1971, p21.

16: Michels, 1968, p15.

17: Hallas, 1971.

18: Michels, 1968, pp63-65.

19: Michels, 1968, pp81, 93.

20: Michels, 1968, pp107, 117.

21: Michels, 1968, p126.

22: Michels, 1968, pp238, 277.

23: Michels, 1968, p205.

24: Michels, 1968, p327.

25: Michels, 1968, p107.

26: Michels, 1968, p111.

27: Michels, 1968, pp205-206.

28: Bukharin, 1925, chapter 8. As far as I know Bukharin is the only major Marxist to have attempted a rebuttal of Michels, though there is also a brief response in Sidney Hook, 1933, p312.

29: Bukharin, 1925, chapter 8.

30: Harman, 1971, pp59-61.

31: Lenin, 1964, chapter 2.

32: Lukács, 1970, chapter 3.

33: Harman, 1971.

34: Lenin, 1964, chapter 2.

35: Willis, 1977; Bowles and Gintis, 1976.

36: Trotsky, 1937, for an analysis of how such pressures affected even the Bolshevik leadership in October 1917. This work was also an oblique commentary on the failure of the German party to seize the revolutionary opportunity in autumn 1923.

37: Lenin, 1965b, p430.

38: Gramsci, 1970, p178.


Bambery, Chris, 1987, “The Politics of James P Cannon”, International Socialism 36 (autumn 1987).

Bowles, Samuel, and Herbert Gintis, 1976, Schooling in Capitalist America (Routledge).

Bukharin, Nikolai, 1925 [1921], Historical Materialism: A System of Sociology (International Publishers), www.marxists.org/archive/bukharin/works/1921/histmat/

Cannon, James P, 2001 [1940], The Struggle for a Proletarian Party (Pathfinder), www.marxists.org/archive/cannon/works/1940/party/

Cliff, Tony, 1993, Trotsky: volume four—The Darker the Night the Brighter the Star (Bookmarks),

Gramsci, Antonio, 1970, The Modern Prince and Other Writings (International Publishers).

Hallas, Duncan, 1969, “Building the Leadership”, International Socialism 40, first series (October-November 1969), www.marxists.org/archive/hallas/works/1969/xx/building.htm

Hallas, Duncan, 1971, “Towards a Revolutionary Socialist Party”, in Tony Cliff and others, Party And Class (Pluto), www.marxists.org/archive/hallas/works/1971/xx/party.htm

Harman, Chris, 1971, “Party and Class”, in Tony Cliff and others, Party and Class (Pluto), www.marxists.de/party/harman/partyclass.htm

Hook, Sidney, 1933, Towards the Understanding of Karl Marx (John Day).

Lenin, Vladimir, 1964 [1920], Left Wing Communism—An Infantile Disorder, Collected Works, volume 31 (Progress), www.marxists.org/archive/lenin/works/1920/lwc/

Lenin, Vladimir, 1965a [1906], “Party Discipline and the Fight Against the Pro-Cadet Social Democrats”, Collected Works, volume 11 (Progress), www.marxists.org/archive/lenin/works/1906/nov/23d.htm

Lenin, Vladimir, 1965b [1922], “Five Years of the Russian Revolution”, Collected Works,
volume 33 (Progress), www.marxists.org/archive/lenin/works/1922/nov/04b.htm

Lukács, Georg, 1970 [1924], Lenin: A Study of the Unity of his Thought (New Left Books), www.marxists.org/archive/lukacs/works/1924/lenin/

Michels, Robert, 1968 [1911], Political Parties (Free Press). An alternative version of the book is available online: http://socserv2.socsci.mcmaster.ca/~econ/ugcm/3ll3/michels/polipart.pdf

Molyneux, John, 1978, Marxism and the Party (Pluto).

Trotsky, Leon, 1936, The Revolution Betrayed, www.marxists.org/archive/trotsky/1936/revbet/

Trotsky, Leon, 1937 [1924], Lessons of October (Pioneer), www.marxists.org/archive/trotsky/1924/lessons/

Trotsky, Leon, 1975 [1942], In Defence of Marxism (New Park), www.marxists.org/archive/trotsky/idom/dm/

Trotsky, Leon, 1979 [1904], Our Political Tasks (New Park), www.marxists.org/archive/trotsky/1904/tasks/

Willis, Paul, 1977, Learning to Labour (Ashgate).