Robert Michels, the iron law of oligarchy and dynamic democracy
On 5 April 2018, the online site of the Merriam‐Webster dictionary reported that lookups for a certain term had risen by 3,000% over the past week, placing it at the number one spot in its “Trending Now” rubric (Merriam‐Webster, 2018). The rise was associated with US Attorney Robert Mueller's probe into Russia's meddling in the 2016 US Presidential Elections, and how he had taken the unusual step of interviewing Russian businessmen with links to Putin who had travelled to the USA at that time – businessmen who now faced potential sanctions. The word being searched for, which was the descriptor used to denote these businessmen, was “oligarch” – “Russian Oligarchs” – and the dictionary helpfully explained that it was drawn from the Greek “oligarchēs”, made up of “olig” (“few”) and “archēs” (“ruler”). An oligarch is a “member or supporter of an oligarchy”, and “oligarchy” is “a government in which a small group exercises control especially for corrupt and selfish purposes”.
The greatest theorist of modern oligarchy is Robert Michels (1876–1936). It is he who, in his classic 1911 text On the Sociology of the Party System in Modern Democracy, coined the phrase the “iron law of oligarchy” (Michels, 1962, p. 356). Michels is often paired with Gaetano Mosca (1858–1941) and Vilfredo Pareto (1848–1923), and collectively they are known as the “elite theorists of democracy”, although they have gone by other appellations too: “Machiavellians”, “theorists of minority domination” or again “sociological pessimists” (Burnham, 1943; Dahl, 1989; Linz, 2006; Lipset, 1962; Stuart Hughes, 1961). Writing at the turn of the twentieth century, they were the first who, in a modern context, tried to grapple with the fact that, although we live in a democracy, it is still the few who rule. In contrast to the ancient Greeks, these thinkers were addressing the question of minority rule within a specifically modern setting, one marked by the spread of universal suffrage and the rise of the modern, highly centralised, bureaucratic and disciplined mass party to organise it – two novel developments at that time (Michels, 1962, p. 334). That context – universal suffrage and political parties – is still the context we operate in today, so that, although much has undeniably changed since then, in many ways this setting, and its problems, remains our own. Indeed, although Marxists had been talking about the state as the “executive arm of the bourgeoisie” since the mid‐nineteenth century (Michels, 1962, p. 346), many of the terms the “elite theorists” first coined – “ruling class” (Mosca), “circulation of elites” (Pareto) and the “iron law of oligarchy” (Michels) – are still the terms through which we try to articulate our own politics today; that is to say: it is Mosca, not Marx, who gives us the term “ruling class”.
The aim of this article is to return to these thinkers – in this specific case Michels – to see whether their ideas can help shed light on our current political predicament. From Occupy Wall Street in 2011 and its slogan of the 99% versus the 1%, to 2016 and the Brexit referendum in the UK, where Leave campaigner and politician Michael Gove declared that the people had had “enough of experts”, and subsequent Prime Minister Theresa May using her Conservative Party Conference speech to attack the “rootless cosmopolitan elite”, to finally the US Presidential Election of the same year and Trump's campaign against the “establishment” and claiming that he will “drain the swamp”, the relationship the “few” entertain with the “many” has been forcefully brought back onto the political agenda.
The first section will offer an account of Michels’ “iron law” by placing it within the context of turn‐of‐the‐century European socialism and syndicalism, paying particular attention to Michels’ relationship with the German Social Democratic Party, the largest, richest and most powerful socialist party of that time. The second section will turn to evaluating the “iron law” itself – notably through the criticisms Max Weber, Michels’ mentor, opposed to it in a letter of December 1910 thanking him for an advance copy of the book – suggesting that it is perhaps not as ironclad as it might at first appear. And although Michels will later convert to Italian Fascism, his Sociology of the Party ended on a cheerier note, with Michels articulating how democracy will naturally give rise to two “palliatives” – an increase in education and competition between different oligarchies – something that has often been overlooked in the secondary literature, and which the third section will explore.
In conclusion to his masterpiece, Michels begins to articulate, through two metaphors, what might be termed a “dynamic” theory of democracy, one grounded in continually challenging elite rule. If it is always the few who rule, then democracy must be understood as the movement that continually challenges the extent of that rule. This conception of democracy is, the article will argue in the fourth section, a highly original and stimulating one, and one that can help us think about our own political situation. It is furthermore away from the more static institutional or procedural accounts of democracy we currently have, and the conclusion will attempt to address some of the questions and issues that arise from such a new conception of democracy.
1 MICHELS AND THE IRON LAW OF OLIGARCHY
Michels is best remembered today – if at all – as the theorist of the “iron law of oligarchy”. A protégé of Max Weber, he wrote his masterpiece, On the Sociology of the Party System in Modern Democracy: Investigations into the Oligarchic Tendencies of Group Life (Zur Soziologie des Parteiwesens in der modernen Demokratie. Untersuchungen über die oligarchischen Tendenzen des Gruppenlebens), in 1911. That book was first translated into Italian in 1912 by Alfredo Polledro as Sociologia del partito politico nella democrazia moderna: studi sulle tendenze oligarchiche degli aggregati politici, and then from Italian into English in 1915 by the British Communist translating couple Eden and Cedar Paul as Political Parties: A Sociological Study of the Oligarchy Tendencies of Modern Democracy – the change of title, one might surmise, to give the book more visibility. Its thesis can be found in the subtitle – the oligarchic tendencies of group life – and Michels presents the “iron law of oligarchy” in conclusion to the book as follows: “reduced to its most concise expression, the fundamental sociological law of political parties…may be formulated in the following terms: ‘It is organisation that gives birth to the domination of the elected over the electors, of the mandataries over the mandators, of the delegates over the delegators. Who says organisation, says oligarchy’” (Michels, 1962, p. 365).
The specific political party Michels had in mind when he wrote the book was the German Social Democratic Party (SPD), which, alongside the German Labour Union, was at that time the largest, richest, best organised and most powerful socialist party in the world (Michels, 1962, p. 357). By the time Michels published his Sociology of the Party in 1911, the SPD had three million members and in 1912 won one‐third of all votes, thereby becoming the biggest German political party of the time (Angaut, 2015, p. 547). From a Marxist perspective, one Michels shared, this meant that the SPD was the most advanced party en route to the forthcoming proletarian revolution, and thus represented the future development of the others, leading him to neglect developments within other socialist parties in Europe at that time (Beetham, 1981, p. 91). The SPD claimed to be organised on a democratic basis, and that if it were to come to power, it would rule in a democratic manner: the party was, in essence, a “state in miniature” (Hands, 1971, p. 170). The problem was that although the SPD dominated the Reichstag, power was still in the hands of the Chancellor and the Junkers in large part due to the “three‐class franchise” in operation in Germany, and especially in Prussia (Beetham, 1977, p. 6). The “democratic party” was in contrast to the older conservative parties, which were organised – and therefore ruled the state – in a highly oligarchic fashion. The SPD would bring a “democratic revolution” to the state itself, transforming it into a “democratic state” (Femia, 2001, p. 97; Hands, 1971, p. 156; Michels, 1962, pp. 50, 335).
What made Michels’ critique so devastating was not solely that it came from one of its own officials and sympathisers, but that he sought to demonstrate that in its internal ruling it was no different to the older, oligarchic, parties the SPD so decried (Michels, 1962, p. 339). The reasons the SPD was no better than the other parties were twofold. Influenced by Gustav Le Bon's psychology of crowds (Beetham, 1977, p. 14; Michels, 1962, p. 205), Michels posited two “psychological” explanations for the iron law of oligarchy. The most important is the “psychology of organisation itself, that is to say, the tactical and technical necessities which result from the consolidation of every disciplined political aggregate” (Michels, 1962, 365). This is not what we might immediately recognise today as a psychological reason, and indeed the emphasis on organisation clearly carried with it Weber's stamp. But Michels’ desire to talk of psychology as a “science” stemmed from his recent encounter with Pareto's Les systèmes socialistes (1902), in which the latter attempted to found a new science of mankind based on the persistence of certain constant psychological traits, which he dubbed “residues” (Beetham, 1977, p. 14). Following Le Bon and Pareto's lead, Michels viewed the mass as generally immobile and passive, in need of a leader to guide them and towards whom they felt gratitude (Michels, 1962, p. 364). Indeed, throughout the book, Michels tried to show how in general the masses, even when organised within a party, were apathetic about the running of their own affairs – committees set‐up to organise the day‐to‐day running of the party were systematically unattended – nor indeed, to Michels’ surprise, did they seem particularly interested in debating the finer points of revolutionary praxeology, preferring instead to go listen to their heroes speak (Michels, 1962, pp. 88, 105, 110; Hands, 1971, p. 162). He explains attempts to run referenda within the party as abject failures because of the “incompetence of the masses and lack of time” (Michels, 1962, p. 309; Lenski, 1980, p. 7; Linz, 2006, p. 51).
The principle of oligarchy in modern democratic parties, therefore, arises from the “technical indispensability of leadership” (Michels, 1962, p. 364). As Michels puts it: “at the outset, leaders arise spontaneously; their functions are accessory and gratuitous. Soon, however, they become professional leaders, and in this second stage of development they are stable and irremovable” (Michels, 1962, p. 364). In other words, every efficient organisation needs a hierarchical – and permanent – bureaucracy with a division of labour and a chain of command. This is both a technical – to ensure the smooth running of the party through a process of delegation – and a tactical necessity – democracy is too slow a decision‐making process to react to political events. It is that bureaucracy that will form the ruling oligarchy, such that the end result is that there is an inverse proportion between the size of an organisation and democracy: the larger and more complex an organisation is, the less democratic it will be: “where organisation is stronger, we find that there is a lesser degree of applied democracy” (Michels, 1962, p. 71).
It should be clear that Michels has what he calls a “Rousseauian” understanding of democracy, namely that the people, or in this case the members of the party, in some sense directly rule (“applied democracy”; Hands, 1971, p. 158; Michels, 1962, p. 73; Mommsen, 1987, p. 127). The concept of representation was foreign to him, as S. M. Lipset correctly saw: Michels and the Machiavellians “prove the impossibility of democracy within a larger polity by definition, by seeing any separation between leaders and followers as ipso facto a negation of democracy” (Lipset, 1962, p. 34).
The second reason for the iron law of oligarchy Michels attributes to what we would more easily recognise as a directly psychological phenomenon: “oligarchy derives, that is to say, from the psychological transformations which the leading personalities in the parties undergo in the course of their lives” (Beetham, 1977, p. 13; Michels, 1962, p. 365). What Michels meant by this is that the growing professionalisation of the party/labour union leads to the creation of a distinct class of bureaucrats, leaders and politicians who are separated from the rest of the party members they represent. As they live different lives, their psychological make‐up is different from the regular party members. These party officials are recruited, in the case of the German socialist party, from the proletariat itself, and, to a lesser degree, from the intellectual bourgeoisie. The end result is that both of these groups become déclassé, at least from a conventional Marxist class analysis: Michels likens the class of the party official to the level of a petty bourgeois, leading to a certain embourgeoisement of the proletarians and a fall in class for the intellectuals (Hands, 1971, p. 161). As Bakunin put it, “there can never be a worker's government, only a government of ex‐workers” (Beetham, 1977, p. 16; Beetham, 1981, p. 85). The case is accentuated in the children of these officials, who tend to have quite bourgeois upbringings, leading to a reproduction of elites that in effect creates a party “caste” (Michels, 1962, pp. 279, 301, 338).
The party officials henceforth no longer belong to the same class as their former colleagues they claim to represent, meaning their interests will differ. Most importantly, their loyalties will no longer be directly with their past comrades, but now lie with the party itself, which provides them with a living. They will come to believe, as Michels puts it paraphrasing Louis XIV, “Le Parti, c'est moi” (Lenski, 1980, p. 7), and the party will become a “state within the state” (Femia, 2001, p. 195; Michels, 1962, p. 335). As such for them the survival of the party will always come first, over and above any demands from the regular members of the party, whether economic or ideological. The simple reason, as anticipated above, is that the party is no longer a means but has become an end in itself (Michels, 1962, p. 338). Michels discusses this in the context of World War I, explaining that the European socialist parties did not oppose the war, as their ideology would have suggested, because the parties were dependent upon the national framework for their own existence. He writes: “the outcome of this regressive evolution is that the party is no longer regarded as a means for the attainment of an end, but gradually becomes an end‐in‐itself, and is therefore incapable of resisting the arbitrary exercise of power by the state when this power is inspired by a vigorous will” (Michels, 1962, p. 358). The raison d’être of the political party is to control the state; thus, it cannot vote against a policy that would harm it.
There are three different resources that, according to Michels, ensure the leaders keep control of their party. These are as follows: (a) Officials have superior knowledge, in that they are privy to much information that can be used to secure assent for their programme; (b) they control the formal means of communication, because they dominate the organisation's press (parties still had their own newspapers at the time), and as full‐time salaried officials, they can travel from place to place presenting their case at the organisation's expense, where their position enables them to command an audience; and (c), they have skill in the art of politics, in that they are far more adept than nonprofessionals in making speeches, writing articles and organising group activities (Lipset, 1962, p. 16).1
An oligarchy thus rules the SPD. A “small group exercises control”, to return to our opening definition, because a party needs a permanent bureaucracy – intermediaries (Michels, 1962, p. 278) – in order to function, and that bureaucracy becomes permanent and irreplaceable over time, and comes to dominate all proceedings. The oligarchy rules if not for corrupt – Michels is quite clear they rule with the best intentions (Linz, 2006, p. 54), but it is the logic of organisation itself that perverts their original intentions in what has become known as “goal displacement” (Hands, 1971, p. 167; Linz, 2006, p. 40) – but for “selfish purposes”, in that they put the survival of the party, which provides them with their livelihood, above all other considerations, whether ideological or socio‐economic. Their salaried dependence on the party turns them into déclassé petty bourgeois, removed from the (class) interests of their former working colleagues or more bourgeois intellectuals.
2 MICHELS AND WEBER
Although considered a classic of political sociology, Michels’ Sociology of the Party has given rise to a number of criticisms (Beetham, 1977, p. 4; Cook, 1971, p. 773; Hands, 1971, p. 155). A recent piece has looked at Marburg, where Michels was active as an organiser, speech‐giver and agitator – he even ran as the Social Democratic candidate in the neighbouring, highly rural, Lauterbach‐Alsfeld to test the field, what is known in German as a Zählkandidat (he won 8.5% of the vote) – and concludes that the small membership there was dominated by small artisans and independent craftsmen (Bonnell, 2011, pp. 23–35). This was thus very far from the type of significant industrial working class to be found in larger cities such as Berlin, meaning that Michels’ first‐hand experience of the labour movement and trade‐union organisation was quite limited, and may help to explain why the “spontaneity” of syndicalism was more appealing to him: France and Italy, where syndicalism dominated, did not have much of an organised labour movement either, and it is quite revealing that although Michels regularly contributed to the French and Italian syndicalist journals, he wrote quite infrequently for the German equivalent, Einigkeit (Cook, 1971, p. 782). Moreover, he seems to have grown quickly disenchanted with the “slow boring through thick boards”, as Weber later characterised political activity. The Marburg membership itself, which comprised a significant number of printing workers, had a natural deference to university‐educated intellectuals, whom they relied upon to write the books they would print, which the city industrial proletariat did not, and might have biased Michels’ view in terms of the docility of the membership towards their officials, Michels included.
But already these types of criticisms had been picked up by Weber himself, who wrote a letter to Michels in December 1910 thanking him and praising him for the book, which had been dedicated to him, and offering a number of objections (Michels, 2015, pp. 535–540). That letter and those objections – Michels’ first – still contain most of the criticisms levelled at Michels today. Aside from the two points raised above – how there is a difference, as Weber puts it, between “academic revisionists” (i.e. Michels) and “syndicalist leaders”, and how the industrial proletariat is more defiant in the big cities – three points stand out.
The first concerns the conservative basis of organisation, which had been one of Michels’ conclusions: “political organisation leads to power. But power is always conservative” (Michels, 1962, p. 333). To that Weber objected that both the “power of the Trust Directors has a revolutionary effect, the power of the Jacobins did too” (Michels, 2015, p. 538–539; Scaff, 1981, p. 1281). Perhaps, but Michels might have responded that neither of these were political parties in the modern sense, i.e. highly centralised and bureaucratised mass parties. Sociology of the Party was first published in 1911, but by the second German edition of 1925, an event Michels could not ignore had occurred, namely the 1917 Bolshevik Revolution (Beentham, 1981, p. 91). In the preface to the second edition, Michels tried to tackle that head on, explaining that Bolshevism was not a democratic movement (Michels, 2015, p. 32). That, needless to say, is a point of contention, but one might conceive how Lenin's restricted card‐carrying and unitary vanguard party, with its emphasis on a top‐down “democratic centralism” of a cadre of “professional revolutionaries” leading the masses – the reason for their break with the Mensheviks, who favoured looser party discipline and a larger base – might still be interpreted through the lenses of the “iron law of oligarchy”.2
Nevertheless, what Weber had put his finger on was the fact that the leadership of the party might be more revolutionary than the membership base. This brings us to our second point, namely whether the interests, and specifically the economic interests, of the base and the leadership must necessarily align (Michels, 2015, p. 539; Segre, 2001, p. 110). One need not be a Marxist to see how the interests of the officials, who, dependent for their livelihood on the party and dominated by craftsmen‐bourgeois types, might differ from those of the broader membership, composed of 54% unskilled factory workers, but does this imply an irreconcilable clash of interests (Cook, 1971, p. 791)? Michels is at his best when pointing out the tension between the official revolutionary ideology the party propagated and its much more conservative rule, yet, and in the same manner the party leadership like the Bolsheviks might be more revolutionary than its base, it would be a mistake to think that the base must be more revolutionary than its leadership. Michels believed that because he had swallowed whole the Marxist view that the proletariat were the revolutionary universal class, and that they were therefore being betrayed by their leadership (Beetham, 1977, p. 11; Michels, 1962, p. 351). But if they are not, then beyond the propaganda, perhaps the leadership was in fact responsive to their more reformist demands (Linz, 2006, pp. 49–51).
Michels actually concedes this point when he explains that the leadership will resort to demagogy to keep the masses on side, instead of pursuing the revolution their ideology demanded of them (Day, 1965, p. 427; Michels, 1962, p. 173). But it turns out that the move away from the revolutionary platform to focus on improving living conditions was in reality in line with the desires of the rank and file itself (Cook, 1971, p. 793). Indeed, Michels fully recognises that organisation “is the weapon of the weak in their struggle with the strong” (Hands, 1971, p. 164; Michels, 1962, p. 61). There is a debate here of course between revolutionary versus reformist trade unionism, and whether the role of leaders is precisely to instil the revolutionary consciousness into the masses – a view Michels seemed to have shared (Beetham, 1977, p. 17). But the concession on the notion that the membership at large preferred reform rather than revolution has led to someone like Day to argue, somewhat counterintuitively, that Michels should in fact be understood as the theorist of party democratisation (Day, 1965).
Certainly, this depends on where one starts from. If the starting point is a small equal organisation – Cassinelli estimates this as having to be lower than 1,000 members (Cassinelli, 1953, p. 782) – then the move to a much larger organisation will undoubtedly reduce the equal distribution of democratic power across the membership that Michels so well diagnosed. However, as Weber points out, democratic parties are often founded by intellectuals, that is to say, an “aristocracy” (Michels, 2015, p. 536): if the starting point is rather a charismatic leader trying to found a party around him – whether it is Lassalle, Liebknecht or Bebel, as Michels himself admits (Michels, 1962, pp. 93–5, 117) – then the bureaucratisation of the party might indeed integrate more people into the democratic process, and that membership might prove to be more socially plural than the original faithful (Day, 1965, 423).
Party democratisation was, thirdly, ultimately Weber's view too (Mommsen, 1987, p. 126). His main objection to Michels was that he had a too unequivocal view of “domination” [Herrschaft], which he considered to be more ambiguous (Mommsen, 1987, p. 130). For Michels, the officials in the party dominated their members, but for Weber, domination is extensible both ways. Seemingly anticipating Foucault, Weber explained that every human relationship has elements of domination, sometimes reciprocal ones. Channelling Sieyès, he gives the example of the shoemaker: “is the shoemaker that makes my boots necessarily my ‘master’?”, he asks. “In a sense the shoemaker dominates me, but in another I dominate him” (Michels, 2015, pp. 593–640; Scaff, 1981, p. 1282): if the shoemaker dominates Weber through his knowledge of how to make a shoe, Weber also dominates the shoemaker in that he pays him to make it. So if party officials dominate members because of their superior technical expertise, in another – and perhaps lesser – sense (voting, participating in debates, paying their membership fees etc. to draw on the “paying” analogy), the members also dominate the officials. This explains the move from revolution to reform within the party, which is in response to the desires of the members.
Thus, Weber, in contrast to Michels, welcomed the bureaucratisation of the SPD, which he thought would see it abandon its revolutionary phraseology to concentrate instead in trying to concretely advance the plight of the working class, which he supported. He also thought that it would lead to the integration of the party, and thereby also its members, into the political system of the German Reich, which would be beneficial to all involved: it would mean the SPD, alongside the Empire, would reach a degree of political “maturity” he so cherished (Michels, 1962, p. 340; Mommsen, 1981, pp. 100–116).
At this point, Weber, of course, was still looking for political parties and the parliamentary system to provide the type of leaders Germany needed. As such that parties should be “oligarchic” in the sense Michels described posed no real threat to him: leadership was a fact of life, and indeed Weber was quite concerned to ensure the freedom to act of the political leaders he was looking for, so the least they were constrained by their party, the better (Mommsen, 1981, pp. 110–111). But over time, Weber became convinced that the domination bureaucracy exerts over politics was becoming too preponderant, and that the future choice was between a “leadership democracy” and a “leaderless democracy”, leading him to advocate a “plebiscitary democracy” in which a popularly elected charismatic leader should be given substantial presidential power to be able to break out from the “iron cage of modernity” (Green, 2010; Mommsen, 1981, pp. 112–115). That, in the end, started to look quite similar to Michels’ later endorsement of Mussolini: Michels thought bureaucracy made his Rousseauian direct form of democracy impossible, whilst Weber thought bureaucracy jeopardised the type of leadership democracy he advocated. Both came to look for solutions outside the realm of bureaucracy to attain the ideal they were striving for (Scaff, 1981, pp. 1281–1284).
3 DEMOCRACY'S TWO PALLIATIVE
There is no question that in his later life Michels rallied to Mussolini, something it is beyond the scope of this article to fully explore. Yet the first edition of Sociology of the Party – or at least the revised 1915 English translation of it – ended on a cheerier note. There Michels writes that, although the ideal government would be an “aristocracy of persons at once morally good and technically efficient”, it is nonetheless true that “as a form of social life we must choose democracy as the least of evils” (Michels, 1962, p. 370). Recognising this – that democracy, whatever its faults, is still better than aristocracy – will help argue against a return to aristocracy once the scoria of aristocracy recognised. That scoria is, of course, the iron law of oligarchy itself.
The ideological tendency of democracy towards criticism and control;
The effective counter‐tendency of democracy towards the creation of parties ever more complex and ever more differentiated – parties, that is to say, which are increasingly based on the competence of the few (Michels, 1962, p. 370).
Thus, democracy, selon Michels, on the one hand, involves an increase in education of the masses, which leads to an increase in their ability to criticise and control their leaders: “a wider education involves an increasing capacity for exercising control”. The task of social education is, therefore, to “raise the intellectual level of the masses, so that they may be enabled, within the limits of the possible, to counteract the oligarchic tendencies of the working‐class movement” (Michels, 1962, p. 369). On the other hand, democracy will lead to the development of other, ever more complex and differentiated parties that will effectively cancel each other out.
It is important to note here that Michels not only believed that democracy will lead to the development of more parties, but that development also holds for within the different parties themselves: his account of the reigning oligarchy is not monolithic (Beetham, 1981, p. 90). Writing on attempts towards decentralisation within national parties, he explains: “while they suffice to prevent the formation of a single gigantic oligarchy, [they] result merely in the creation of a number of smaller oligarchies, each of which is no less powerful within its own sphere” (Michels, 1092, p. 202), and Michels readily admitted that there could be “prolonged struggle for dominion between two factions” within the party (Michels, 1962, p. 102). So competition exists between – and within – parties (Beetham, 1981, p. 94; Hands, 1971, p. 171; Linz, 2006, p. 60; Michels, 1962, p. 339).
Although these two claims, and particularly the second, appear to prefigure much post‐war Schumpeterian conceptions of competitive democracy, they have been almost systematically overlooked.3 Indeed, Michels seems to anticipate his near‐contemporary's theory of minimalist democracy when he writes “the democratic system is reduced, in ultimate analysis, to the right of the masses, at stated intervals [elections], to choose masters to whom in the interim they owe unconditional obedience” (Michels, 1962, p. 217; Schumpeter, 2010).
Part of the reason might be that Michels himself abandoned them in his 1925 reworking of the conclusion: whilst the discussion of the pedagogical effects of democracy are still present, the two palliatives as explicitly listed above have disappeared, and there remains no trace of the thought of different parties effectively counter‐balancing themselves (Michels, 2015, pp. 522–528). This omission seems to underline the passage of a Michels still sympathetic to democracy in 1915 to a leading fascisant thinker a decade later. Yet the standard English translation of Sociology of the Party remains the 1915 Political Parties edition, so there is no reason for Anglophone commentators to have missed them, nor indeed see that Michels would later drop them out.
Nevertheless, Robert Dahl, in his classic Democracy and Its Critics (1989), criticises Michels, whom he labels, with Mosca and Pareto, “theorists of minority domination”, for having extrapolating from his study of political parties to defining the whole political system as oligarchic. Dahl objects: “but even if we grant that political parties are oligarchical, it does not follow that competing political parties necessarily produce an oligarchical political system” (Dahl, 1989, p. 276). But this is precisely what Michels was advocating with his second palliative: that democracy would lead to many different competing parties that would in effect counter‐balance one another, not to mention competition within the political parties themselves. In fact, there is a remarkable similarity between the two “social means of control” Dahl posits, namely elections and competition between parties, and Michels’ two palliatives, especially when we consider that the role elections are meant to play in Dahl's “polyarchy” is to express the “voice” of the people: perhaps not too far from the educational empowerment of Michels’ first palliative, which was meant to lead to intellectual criticism and control. In any case, both Dahl and the 1915 Michels saw the salvation of the democratic system in the competition between political parties.
In 1915, Michels thought that democracy would naturally give rise to palliatives within the system itself – something he seemed to have abandoned in favour of Fascist charisma by 1925. And although in his earlier phase he also thought that political parties would be a natural breeding ground for democratic leaders, Weber would soon turn to plebiscitary democracy to save modern politics. There is, however, a difference between the 1915 Michels and the later Weber, which we might characterise as the difference between an internal and an external conception of political salvation: for Michels, democracy would find from within the means to regulate itself, whereas with Weber it would be charismatic leaders who would break, from the outside, the iron cage of bureaucracy. As Mommsen puts it: “‘plebiscitarian democracy’…served as a counterweight to the bureaucratisation of the apparatuses of power” (Mommsen, 1987, p. 132). So, for Weber, it is exogenous factors that would push back against the tyranny of bureaucracy, whereas for Michels bureaucracy and oligarchy endogenously produced its own palliatives.4
That the “iron law” should admit certain palliatives within its bosom that militate against it suggests that the law is not as ironclad and inflexible as we might have first thought: Sartori nicely suggests that the law should be renamed a “bronze” law (Linz, 2006, p. 63; Sartori, 1987, p. 149). Indeed, Cook has pointed out that “the English version ‘who says organisation, says oligarchy,’ is a noticeably stronger formulation than the original German which reads ‘Wer Organization sagt, sagt Tendenz zur Oligarchie’”, namely that the law is more a tendency than an “iron” law itself. Note too that the original 1908 article was titled “Die oligarchischen Tendenzen des Gesellschaft”, i.e. the emphasis here is again on tendencies (Cook, 1971, p. 787; Hands, 1971, p. 157). Michels will posit the “fundamental problem of politics as a science” at the end of Sociology of the Party, not as to whether “ideal democracy is realisable”, but rather “to what point and in what degree democracy is desirable, possible and realisable at a given moment”; that is to say: to what degree can democracy, understood as the two palliatives, push back against the iron law. Oligarchy as such is not static, but we should instead talk of different degrees of oligarchy: more or less; or, to put it another way, more or less democratic. As Linz has helpfully suggested, democracy and oligarchy should not be understood as pure dichotomies but rather “polar tendencies on a continuum” (Linz, 2006, p. 38), and Beetham captures it well when he concludes that Sociology of the Party works best when “its iron laws are recast in the form of more pliable tendencies” (Beetham, 1981, p. 99).
4 DYNAMIC DEMOCRACY
There are, as had often been pointed out, a number of conceptions of democracy in existence in Michels’ work (Angaut, 2015, p. 561; Hands, 1971, pp. 158–9). But two in particular stand out.5 The first is the Rousseauian form of direct democracy that will admit no representation, and that he finds impossible to fulfil: for the epigraph to his “Final Considerations” of Sociology of the Party, Michels will quote from the Social Contract – “to take the term in its fully rigorous meaning, there has never existed a true democracy and one will never exist. It is against the natural order of things that the great number governs and that the small number be governed” – a claim he will repeat over the course of his writings (Michels, 1962, pp. 73, 364). The second are the two palliatives he identifies democracy naturally produces in response to the iron law of oligarchy. The question Michels thus asks is: if it is always the small number who rule, if there is an iron law of oligarchy, then what can democracy mean if popular sovereignty is only ever a myth, if the “mass will never rule except in abstracto” (Michels, 1962, p. 366)?
after the old man's death the sons dig everywhere in order to discover the treasure. They do not find it. But their indefatigable labour improves the soil and secures for them a comparative well‐being. The treasure in the fable may well symbolise democracy. Democracy is a treasure that no one will ever discover by deliberate search. But in continuing our search, in labouring indefatigable to discover the undiscoverable, we shall perform a work which will have fertile results in the democratic sense. (Michels, 1962, p. 368)
The democratic currents of history resemble successive waves. They break ever on the same shoal. They are ever renewed. This enduring spectacle is simultaneously encouraging and depressing. When democracies have gained a certain stage of development, they undergo a gradual transformation, adopting the aristocratic spirit, and in many cases also the aristocratic forms, against which at the outset they struggled so fiercely. Now new accusers arise to denounce the traitors; after an era of glorious combats and of inglorious power, they end by fusing with the old dominant class; whereupon once more they are in their turn attacked by fresh opponents who appeal to the name of democracy. It is probable that this cruel game will continue without end. (Michels, 1962, p. 371)
A number of points are in order to make sense of what Michels is trying to articulate. The first is to say that much like his two palliatives, the first metaphor of the field is substantially reduced in the second 1925 Kröner edition of Sociology of the Party: thereby underlining the shift in Michels’ thinking from seeing how democracy might internally resolve itself, to the demand for an outside leader to take control of it (Michels, 2015, 526). The second is to say that much like the two palliatives, these two metaphors have been little‐remarked upon in the secondary literature, and in the rare instances they have been, they have often been quoted without further elaboration.6
This article submits that what Michels is here trying to articulate is what might be called a dynamic theory of democracy. It is one that identifies the continual challenge to elite rule as the true location of democracy: that it is in the movement to challenge the oligarchy where democracy is in fact to be found. Democracy's true location is thus not where it is usually thought to lie: it is neither to be found in institutions or principles, but to be located in the movement itself. This displacement means democracy is not an end point, but a continuous movement: the never‐ending challenge to elite rule which, even though it never fully achieves its aim, nevertheless through this challenge is able to offer certain democratic benefits. It is therefore not the by‐product of this struggle, but the struggle itself: it is the democratic benefits that are the by‐products. In many ways, this is how one might read the German SPD that Michels studied: although it never succeeded in achieving its own ideology of democratic revolution, nevertheless in striving towards it, it achieved real welfare benefits for its members. This might be a pessimistic, elitist, or reformist theory of democracy, but it is a theory of democracy nonetheless.7
This dynamic conception of democracy stands in sharp contrast with many of what we might call the static theories of democracy that prevail today. In his own declared masterpiece Democracy and Its Critics, Robert Dahl, the doyen of American democratic theory, listed seven set of conditions for a country to be recognised as a polyarchy (Dahl, 1989, p. 233). These conditions are static – they do not change – it is on how well the different political systems match these seven criteria that they are judged. One might say the same about Rawlsian‐inspired theories of justice: the whole point of Rawls’ two principles of justice, at least in its earlier incarnation (Bonotti, 2017), is to stabilise at the theoretical level – “ideal theory” – the principles that would underpin a “well‐ordered” society, and the next step would then be to go about trying to implement them (Rawls, 1999). Although he has shown, throughout his life, an interest in and support for political movements, the “siege model” of democracy Habermas developed in Theory of Communicative Action is rather static too: although it also expressed the notion of challenging elite rule, a “siege” nonetheless brings more to mind a blockade or a defensive position, particularly when contrasted with waves constantly breaking against the shoal, and in Between Facts and Norms, Habermas will later move to a much more procedural “sluice” model (Habermas, 1986, 1997). None of this is to deny that these thinkers do not allow for movement within their theories, but it is to point out that they do not identify democracy with movement itself. Finally, Sheldon Wolin's “fugitive” democracy, although associated with movement, appears at the margins of the “inverted totalitarian” system he theorised, and represents rather a momentary escape from it, rather than being the centre of democracy, as Michels thought it should (Wolin, 2016).
The one prominent theory this dynamic theory of democracy appears to echo is Michels’ near‐contemporary Joseph Schumpeter's theory of “minimalist” competitive democracy (Schumpeter, 2010, pp. 241–251). We have seen above that the notion of “competition” is in fact a central – if overlooked – component to Michels’ theory of democracy at least since the 1915 English edition of Sociology of the Party. Schumpeter, however, who was influenced by Weber and Pareto (Beetham, 1985; Isaac, 2012), seems to have been unaware of Michels’ work. But bringing Mosca and Pareto back into the fold can help shed light not solely on Michels but also his theory of democracy. There is little doubt that the “iron law of oligarchy” was influenced by Mosca's idea of “the ruling class”. As he writes in Sociology of the Party: “Mosca, who declares that no highly developed social order is possible without a “political class,” that is to say, a politically dominant class, the class of a minority” (Michels, 1962, pp. 342, 353). Through Mosca, Michels compares the struggle between the aristocracy and democracy as “two groups of dancers executing a chassé croisé in a quadrille”, as the struggles between an “old minority, defending its actual predominance, and a new ambitious minority, intent upon the conquest of power, desiring either to fuse with the former or to dethrone and replace it” (Michels, 1962, 342).
There is no essential contradiction between the doctrine that history is the record of a continued series of class struggles and the doctrine that class struggles invariably culminate in the creation of new oligarchies which undergo fusion with the old. The existence of a political class does not conflict with the essential content of Marxism, considered not as an economic dogma but as a philosophy of history. (Michels, 1962, p. 354)
Mosca and Pareto might also be construed as proposing a dynamic conception of democracy: Sartori writes: “as for Pareto, there is nothing inherently undemocratic in his law of the ‘circulation of elites’” (Sartori, 1987, p. 47). For Pareto the notion of the “circulation of elites” speaks for itself: the emphasis of accounting for social change is, here again, located in movement and not stable institutions. Mosca has a theory of history – a facet of his thought overlooked by Michels – that posits the continuous development of different “social forces” that arises because of new technological, social, economic, legal, military etc. phenomena, and in the Elementi di scienza politica (1932) came out in favour of a liberal representative regime, based on a legal codification of checks and balances, as the best way to harmonise these competing social forces: a system he dubbed “juridical defence” (Mosca, 1960). Because of the ever‐developing social forces, and because he thought the leaders of these new forces would be incorporated into the political class, Mosca's model is also a dynamic one, and one that ultimately finds its best expression in a representative regime.
5 CONCLUSION
Michels has often been dismissed simply as a “synthesiser”, as the least “original” of the “Machiavellians” (Beetham, 1981, p. 85; Mommsen, 1987, p. 121; Stuart Hughes, 1961, p. 251). In Consciousness and Society, Stuart Hughes writes: “there is a quaint justice in the fact that it was the least original among the trio of neo‐Machiavellians who found Fascism the least troubling” (Stuart Hughes, 1961, p. 272). Certainly, Pareto tried to offer an all‐encompassing psycho‐scientific account of the world, whilst Mosca, with his notions of “ruling class”, “political formula” and “juridical defence”, might have proposed a more comprehensive theory of politics. But even if it was left to Michels to apply Mosca's “ruling class” or Pareto's “circulation of elites” – and Michels was able to keep on good terms with both, a feat in itself (Beetham, 1977, p. 13) – to modern political parties (Stuart Hughes, 1961, p. 256), he was amongst the first, alongside Moisei Ostrogorski, not only to do so, but also to fully capture that novel development, which neither Mosca nor Pareto did in its entirety. Moreover, and again perhaps not to the same depth and extent as Weber (Stuart Hughes, 1961, pp. 250, 262–263), Michels did offer a theory of bureaucratisation, offering a strong thesis on the link between organisation and oligarchy.
So what do we learn from Michels? A number of points spring to mind. First, that any sort of direct democracy, in a modern setting, is practically, apart from on a very small scale, impossible. Second, that any type of organisation will naturally produce a class of leaders – a “political class” or “ruling elite” will arise because of the “iron law of oligarchy” – more or less independent from those they are responsible to. When we think about politics, this sounds about right: politicians are more or less independent from voters. How much is an empirical question, and this is not to deny that numerous factors, including the structure of organisation, might serve to reduce that independence (Hands, 1971, p. 169; Linz, 2006, p. 62). But we are here always talking of degree, never of a complete reduction.
Perhaps Cassinelli got it right when he wrote that by the “law of oligarchy” was meant: “the executive or leadership activities in an organisation are free from control by the other activities; or, putting it another way, the people who hold positions of authority within an organization are not checked by those who hold subsidiary positions within the organization”. Yet that does not mean that “freedom from control does not mean that the leaders can completely ignore the actions and desires of the lower ranks in the organization” (Cassinelli, 1953, pp. 778–779). This seems to capture Michels’ sentiment well when he writes in Sociology of the Party: “the thesis of the unlimited power of the leaders in democratic parties requires, however, a certain limitation…the old leader must therefore keep himself in permanent touch with the opinions and feelings of the masses to which he owes his position” (Michels, 1962, p. 172).
But ultimately it still is, as our opening Merriam‐Webster definition put it, the “few who rule”. Jeffrey Winters in his recent Oligarchy has criticised Michels for supposedly having obscured the fact that since its first theorisation with Aristotle oligarchy meant the rule of the rich and not simply the rule of the few (Winters, 2001, p. 26). This might not be entirely consonant with the dictionary definition, but what it does allow us to see is that for Michels it is through bureaucratisation that riches can be acquired: that it is through the institution of the party that members can rise up to a petty bourgeois existence and guarantee a better quality of life for their children. Extended to the rest of politics, the implications are clear.
Lenski has also highlighted the predictive power of Michel's “iron law of oligarchy”. Distinguishing between analytical and normative elitism, he posits that analytical elitists have a “theory with remarkable powers of prediction”, citing Michels’ view that “the problem of socialism is not merely a problem of economics…[it] is also an administrative problem” (Michels, 1962, p. 350) as anticipating the rise of the nomenklatura in eastern Europe and Soviet Russia after the war (Lenski, 1980, pp. 2–3, 8; Linz, 2006, p. 26): in the earlier literature, it was quite common to compare Michel's Sociology to the Yugoslav former second‐in‐command to Tito Milovan Djilas’ The New Class (Cook, 1971, 786; Lenski, 1980, p. 5; Lipset, 1962, p. 20).
If Sociology of the Party strikes us as a little dated today, this is for two reasons. The first is that although Michels does offer considerable empirical data and engages in the type of statistical analysis that is recognisable as the beginnings of the type of work social scientists engage in today (Michels, 1962, pp. 88, 106, 257–258, 268–269, 310–311; Femia, 2001, pp. 92, 101) – thereby rightly making him one of the founders of political sociology (Linz, 2006, p. 3) – the book overall comes across as a little more impressionistic and journalistic. This might be because, as Linz points out, Michels, unlike Weber and Durkheim, did not collect his own data – perhaps for lack of time, family pressure or economic concerns – nor was he able to develop an original methodology akin to Durkheim's, or possess the synthesising and systematising abilities of Weber (Linz, 2006, pp. 20–21). But the second is that the questions he raised concerning party organisation and democracy are those social scientists (Lipset, Parsons, Duverger, C Wright Mills) and democratic theorists (Schumpeter, Dahl, Bobbio, Sartori, Aron) have been arguing over ever since (Lipset, 1962, pp. 20–39; Michels, 1962, pp. 63–68).
Moreover, Michels, with his theory of “dynamic” democracy, offers us a means through which to conceptualise how to continually apply pressure to make the iron law of oligarchy more pliable. As such he offers us not solely an empirical study of the possibility of democracy within political parties and political systems, but also an aim – what both Mommsen and Sartori have called an ethical “yardstick” (Mommsen, 1981, p. 109; Mommsen, 1987, p. 128; Sartori, 1987, p. 159) – that we may never reach, but from which, in striving to reach it, democratic benefits will accrue. This combination of realism (awareness of the facts) and idealism (value pressure upon the facts) is precisely the account of democracy Giovanni Sartori proscribes in his The Theory of Democracy Revisited (Sartori, 1987, p. 164). And although the peasant's sons will never find the non‐existing buried treasure in the field, their work is not in vain, as from having been tilled the land will have been made richer, thereby making them also richer.
As such, Femia's view that Mosca, Pareto and Michels are the greatest exponents of the “futility” thesis – that true democracy where a majority rules over a minority will never be achieved – must be accepted with considerable reserve: that if it is the case that a minority will always rule, that does not mean things cannot change, and indeed for the better (Femia, 2001, pp. 9–10). Femia himself concedes the point when he mentions – one of the few to do so – both Aesop's fable and Michels’ “palliatives”, requesting that these reflections be deepened, and he concludes: “pursuit of the unattainable is not always a waste of time; the futility thesis, correctly understood, need not be a counsel of despair” (Femia, 2001, p. 109). Indeed, for Femia the futility thesis, when correctly understood – that if democracy is not impossible it is at least imperilled – has in fact made “the most profound contribution to democratic theory” (Femia, 2001, p. 15).
Bringing Mosca and Pareto into the conversation can help deepen the “dynamic” conception of democracy, as Michels on his own cannot carry the whole weight of the theory. Beetham, for instance, ridicules the parable of the sons digging for the treasure in the field as pertaining to the domain of “gross self‐deception” (Beetham, 1977, p. 19). Do those who are challenging elite rule need to truly believe in democracy to be able to engage in their work? Already from Michels’ second metaphor for democracy – of waves breaking against the shoal – we see this need not be the case: that a new elite will naturally arise to denounce the aristocratic slide of the old elite, and that that challenge will continue without end. Pareto had already suggested in his Systèmes socialistes that new elites will ally themselves with the people to challenge the old elites, and Mosca allows us to see that new elites, based on new social forces, will inevitably arise to challenge the few (Femia, 2001, p. 88). It is in these moments that the cursor of the iron law might be dragged towards the democratic side. Of course true democracy is never achieved, and the people, in Michels’ account or Pareto's, are betrayed, but for the system to work all that is needed is the appearance of a new elite that feels excluded from power to want to challenge the old one for its place within the ruling class.
One might further ask how are we to know whether such a challenge will be a democratic rather than an authoritarian one? Again, here Pareto comes to hand with his theory of the “foxes” and the “lions”, his two types of elites. “Lions” rule through force and are more conservative, emphasising unity, homogeneity, faith and centralisation, whereas “foxes” are characterised by combinazioni: deceit, cunning, manipulation and co‐optation, and theirs is a decentralised, plural and sceptical rule, uneasy with the use of force (Femia, 2001, p. 72). Already here we have a means to distinguish what type of new elite is challenging the old, and thus whether the law will likely be made firmer or not.
All of the above suggests a much richer and broader account of democracy than the one offered by Schumpeter's “minimalist” theory of democracy. If the latter benefits from the simplicity of focusing on competition through elections – a theory, we have seen already in existence at least in Michels some thirty years before – it cannot offer a means of discerning between different types of elites, nor, indeed, whether the iron law of oligarchy is to be made more pliable or not. Neither it is capable for accounting for change within the regime itself.
What we also see here are theories – sometimes even advocacy – of democracy. What the “elitists” offer is to think about the meaning and definition of modern democracy in a world of the “fact of oligarchy”. That meaning they found in the continual challenge by a new elite of the old. Thus, instead of labelling Mosca, Pareto and Michels “elite theorists of democracy”, we might consider labelling them “democratic theorists of elitism”, in Natasha Piano's happy turn of phrase (Piano, 2019). In any case, we still have a lot to learn from them.
ACKNOWLEDGEMENTS
These papers were first presented at a conference in Cambridge, “Elites and Democracy in Modern Political Thought”, which took place in CRASSH on 7 December 2017, and at a subsequent APSA panel “Elites, Political Parties, and Democracy's Discontents” on 31 August 2018. We would like to thank everyone who participated in both, alongside two anonymous referees for their helpful comments. Thanks also to Nadia Urbinati for putting this special issue together, and to audiences at the Yale Center for Representative Institutions in April 2018, the Nottingham “Engage” Series and the Political Theory Workshop at Sheffield in March 2019 for specific comments to this paper.
Biography
Hugo Drochon is Assistant Professor in Political Theory at the University of Nottingham.
