Will Morrisey Reviews

Maurice Duverger: Political Parties: Their Organization and Activity in the Modern State. Barbara and Robert North translation. London: Methuen and Company, 1967. Originally published in 1951.

 

Born in 1917, trained as a jurist, Maurice Duverger began his political engagement as a student at the University of Bordeaux in the late 1930s. He joined the Parti Populaire Français, a Right-wing answer to the Popular Front, the ‘one big Left’ organization of those days. By the end of the war, he had shifted his allegiance to the Left and had founded a political science program at the university, where he now taught. At the end of his long career, he was a member of the European Parliament, where he represented the Democratic Party of the Left, a Marxist organization, originally the Italian Communist Party, which abandoned Lenin’s dictatorship-of-the-proletariat model and almost immediately reinvented itself after the collapse of the Soviet empire. (Duverger himself was defending Stalin as late as 1956, in opposition to Nikita Khruschev’s rather limited set of reservations expressed in his now-famous ‘Secret Speech’ to Party officials.) Political Parties is by far his best-known work among European and American political scientists, sharing with many of them a neo-Marxist analysis of politics. The contemporary school of ‘developmental political science’ derives much of its approach from Duverger.

Marxism presents a problem, as Duverger evidently understands. Marx claims that his socialism, unlike all others, is not utopian/’idealist’ but scientific/empirical. But although in Capital he adduces reams of empirical evidence in support of his theoretical claims about the ‘historical dialectic,’ there remains what Duverger calls “a vicious circle,” as seen in the attempt to understand political parties: “a general theory of parties will eventually be constructed only upon the preliminary work of many profound studies; but these studies cannot be truly profound so long as there exists no general theory of parties.” How, then, to know about such matters as “the evolution of party structures, the number and reciprocal relations of parties, the part they play in the State”? Duverger provides a scientist’s answer. His ‘theory’ won’t be so much a theory, at this point, as a sort of hypothesis, or what he calls a “vague, conjectural, and of necessity approximate” general theory of the parties in the attempt “to introduce objectivity into a field where high feeling and special pleading are the general rule,” along with “a general plan of the field of study.” 

“In recent years the Marxist conception of party as class, taking the place of the Liberal idea of the party as doctrine, has given a new direction to research.” Duverger endorses the “dualism” of class conflict as “very appropriate,” but “there are many more shades of social stratification than this rough manicheism suggests.” The ‘neo’ in his neo-Marxism gives more weight to the mind than Marxists had been inclined to do: “the ‘bourgeoisie’ and ‘proletariat’ do not perhaps constitute two classes, defined in strictly economic terms, but they characterize two states of mind, two social attitudes and two ways of life.” This distinction, which hints at one dimension of Aristotle’s definition of a regime, “throws light on certain problems concerned with the structure of parties.” The formula revises Marx only if “states of mind” do not derive more or less directly from socioeconomic classes, if they function as independent variables in one’s explanation of such sociopolitical phenomena as parties. To his credit, Duverger wants to restore the political to the consideration of political parties, although he doesn’t go far enough, as will be seen.

It is therefore helpful of Duverger that he immediately addresses the question of the definition and origin of parties. By a party he does not mean a faction; factions have existed more or less since the beginning of political life. By a party, he means an organization that combines “parliamentary groups” and “electoral committees.” Parliamentary groups came first, followed by the electoral committees; only then did politicians establish “a permanent connection between the two.” Sometimes, a parliamentary group was based upon doctrinal affinities, at other times, a geographical region or a profession. Philosophical societies, churches, military veterans’ associations, secret societies, industrial and commercial groups: all of these types of organizations generated political groupings in parliaments. It wasn’t until the beginning of the twentieth century that such groups arose outside of parliamentary institutions; these were the Socialists. And in England, even one of the major Socialist parties was first organized in Parliament, whereas the Labour Party was organized during the 1899 Trades Union Congress. Since then, most new parties in republican regimes have originated outside the legislatures, although in “countries new to democracy” the older origin story had prevailed, at least until 1950. Obviously, such organizations as the Solidarity movement in Poland began, as the name suggests, as social movements outside of the sham legislature of the Communist regime. 

Duverger identifies three “sociological types” of parties. “Middle-class” parties recruit their candidates from a relatively narrow segment of the population, with emphasis on “outstanding people,” usually members of the legislature; they are decentralized; and they are organized by caucuses. Socialist parties recruit their candidates from a wider segment of the population; as ideologically-animated organizations, they emphasis “political education” of members and crystallize as top-down bureaucracies arranged with branches connecting the center to the periphery of the membership, which simplifies the task of “political education.” In the “totalitarian” (Communist and Fascist) parties, even more centralized and arranged with “cells” ruled by an autocrat or small set of autocrats from the center, the elites impose doctrines upon the rank and file. Duverger carefully identifies “several types of parties” that are “outside this scheme”: the Catholic and Christian Democratic parties of Western Europe, which combine organizational traits seen in the pre-twentieth century parties with traits of Socialist parties; Agrarian parties, which take a variety of organizational forms and in any case are now rare; and the parties formed “around an influential protector, clans formed round a feudal family, camarillas united by a military leader”—as often seen in the non-Western world and in “pre-1939 Central Europe,” often places emerging from imperial rule.  In a “camarilla” party, the ‘big man’ social structure is the understandable basis for any wider political activity, given the structure of societies with little or no experience of ‘modernization.’ Duverger reserves a unique place for the Labor parties because their origin within labor unions and cooperatives has lent themselves to “a pattern of ‘indirect structure’ which will require special analysis.”

A party organization might be “direct”—with no prior organization within civil society, such as a labor union or a church acting as a source of doctrine and structure—or “indirect,” where prior social organization did exist. The least ‘ideological’ of the European Socialist parties—England’s Labour party and the Scandinavian parties—incline toward “concrete reforms,” not doctrinal conformity. Duverger finds some connection between the degree of prior centralization of the state and the centralization of parties, as seen in France. He does not claim too much for this point, however. This is not “a sociological law,” but rather “a basic tendency, combining with many other factors liable to attenuate or invalidate the results.” One reason for this is that “a party is not a community but a collection of communities, a union of small groups dispersed throughout the country…and linked by coordinating institutions.” These disparate groups come in “four main types”: the caucus, the branch, the cell, and the militia.

Caucuses predate the formation of parties in Duverger’s sense of the term. “In Marxist terms they are the normal political expression of the middle class,” seen in nineteenth century Europe: Conservative parties, consisting of “aristocrats, industrial magnates, bankers, even influential churchmen”; and Liberal parties, consisting of “tradespeople and lesser industrialists, civil servants, teachers, lawyers, journalists, and writers.” Caucuses are quite decentralized because both of these types originated in disparate elements of the middle class and, in the case of the Conservatives, alliances with the aristocracy. 

As the term implies, branches are “less decentralized” than caucuses, connected to the trunk, incapable of surviving without the trunk. “The branch is extensive and tries to enroll members, to multiply their number, and to increase its total strength.” That is, it registers social equality, ‘democracy’ as Tocqueville calls it. “In practice you only need to wish to belong to be able to do so.” “The branch appeals to the masses.” It is “a Socialist invention.” Crucially, it is more democratic than the caucus in terms of extension, in terms of membership, less democratic than the caucus in terms of command and control, which is held by a small group in the ‘trunk’ or center of the structure. Catholic parties and Fascist parties have adapted the branch as parts of their organizations, organizing their members along religious and nationalist lines, respectively. Because religion and nationalist cross ‘class’ lines, both of these parties “generally succeed in attracting to themselves some proportion of the working-class masses,” making them serious rivals to Socialists. 

Cells are “a Communist invention.” They differ from branches in two ways. Branches are geographical, local, even if connected to the center; cells are based upon occupations: factory workers, shop workers, office workers, government administrators, and so on. Although in some sparsely populated areas where there are not enough persons in any one occupation to form a cell you might find an “area cell,” “the real cell is the workplace cell which unites party members working in the same place.” In the United States, for example, one would find Communist cells within the public schools, others in the same area but located within one or more factories. 

Finally, there are militias, private armies “whose members are enrolled on military lines, are subjected to the same discipline and the same training as soldiers, lie them wearing uniforms and badges, ready like them to march in step preceded by a band and flags, and like them ready to meet the enemy with weapons in physical combat,” always “ready to hold themselves at the disposition of their leaders.” Duverger cites the obvious examples of Hitler’s Storm Troops and Mussolini’s Shock Troops, many of them veterans of the Great War. “Just as the cell is a Communist invention, so the militia is a Fascist creation,” well-suited to Fascist doctrine, “that mixture of Sorel, Maurras, and Pareto, which affirms the predominance of the elites, of the activist minorities, and the necessity of violence to allow them to conquer and to retain power.” Duverger associates Fascism with middle classes that intend to dominate the working classes “by opposing with force of arms the strength of the masses,” especially in states and regimes too weak to use their own forces to prevent working-class rule. He tends to minimize the ‘Socialist’ element in National Socialism. If he were alive today, he would likely include the Iranian Revolutionary Guard as a contemporary example of a party militia, with such allies as Hamas, Hezbollah, and the Houthis as militia-based political parties.

Duverger next considers the linkages among caucuses, branches, cells, and militias. This is “essentially a political question and therefore of prime importance” because “the arrangement for linking and relating the primary groups of the party have a profound influence upon its militants, upon its ideological unity, and the efficacy of its action, and even upon its methods and principles.” Here is where the regime comes in, although Duverger does not use that term. He somewhat more cumbersomely remarks that “political articulation tends to model itself upon the articulation of administration in the state,” the “pattern of a hierarchical pyramid coinciding with the official territorial divisions,” combined with the electoral system of the country, which influences the degree to which a party can ‘articulate’ within the ruling institutions. This notwithstanding, Duverger regards the intra-party elements as basic, since Liberal parties are less well-defined or ‘articulated’ than Socialist parties and “most Catholic parties, Communist and Fascist parties featuring an “even more precise, rigid, and strong” ‘articulation.’ It might be remarked that such parties are typically regimes in waiting, however, and so are organized in accordance with the regime type their ‘leaders’ aim to establish. Socialist, Communist, and Fascist parties are proto-regimes, which makes the regime analysis fundamental. The political questions of “prime importance” turns out to be: ‘What kind of regime do you want, and why?’

With regard to linkages among the elements of parties, “there is frequent confusion between vertical linkage and centralization, and between horizontal linkage and decentralization.” “Vertical” and “horizontal” links refer to “ways of coordinating the basic elements of which the party is made up.” “Centralization” and “decentralization” refer to “the way in which power is distributed among the different levels of leadership.” One might expect vertical linkage and centralization to comport with one another, horizontal linkage and decentralization the same, but this is not always so. For example, the French Socialist Party is decentralized but vertically linked, the British Conservative Party centralized but horizontally linked. Moreover, decentralization and centralization themselves come in several forms: local, ideological, social, and federal. A party might be decentralized in some of these aspects, centralized in others. The matter of centralization, Duverger adds, may affect the viability of a given state and its regime. Looking at the United States, with its (then) decentralized parties, he writes: “It is a serious matter that the greatest nation in the world, which is assuming responsibilities on a worldwide scale, should be based on a party system entirely directed towards very narrow local horizons.” At the time Duverger was writing, Charles de Gaulle was making a similar point, criticizing the parties and parliamentarianism generally and advocating a new republican regime with a much stronger executive. 

Given the four elements of parties, membership and ‘leadership’ or more properly rulership must also be considered. Membership varies from party to party, with “the introduction of universal suffrage” fostering “the development of Socialist parties,” except (at the time) in the United States, where socialist motifs have been co-opted by the Democrats within their broad coalition, at the time including civil rights advocates and racial segregationists. “It was the Marxist conception of the class party that led to such massive structures” in Europe, although it must be said that the United States already had very broad-based parties, if not parties according to Duverger’s definition of the term, and those parties fought a civil war over the character of the regime. The Socialist parties “made it possible to free the working class from the tutelage of middle-class parties,” with independent sources of funding, including the funding of newspapers, funding made possibilities made possible by the ‘mass’ character of the parties—again, very much like the structure of the American parties in what Tocqueville had already identified as the world’s “sample democracy.” Showing his Marxist colors, Duverger classifies (in both senses) the Fascist parties as middle-class and therefore “Conservative” parties, although it really makes no sense to say that they were anything other than regime parties intent on overthrowing all of their rivals, including those advocating a return to monarchies with Church establishments. He will argue that Fascist parties “lose their pure mass-party characteristics” over time, but then so did the Communist parties.

Duverger frames things differently, however. Are we “still dealing with true mass parties” or with “a gradual evolution towards a new conception, a third category: devotee parties, more open than cadre parties, but more closed than class parties,” as seen in Lenin’s Bolshevik parties and in the Fascist parties. In both cases (although Duverger seems reluctant to admit it fully, regarding the Communists), we see an elite cadre ‘leading’ a mass membership. “Even within the party” (he means the Nazis but surely also the Bolsheviks and the Chinese Communists) “there can be found concentric circles corresponding to different degrees of loyalty and activity.” In all “mass” parties, participation also may be classified in three circles, the broadest being the electors—who may or may not be ‘card-carrying’ members of the party—followed by the “supporters”—who not only vote for the party but actively advocate on its behalf—and the “militants”—those who organize and operate the party apparatus while formulating the propaganda the supporters mouth and the “general activities” in which they partake. “Insofar as” these “inmost circles” actually represent the outer circles, one can call such parties democratic (as in Lenin’s famous phrase, “democratic centralist”); “otherwise this series of concentric circles is to be defined as an oligarchy.” In terms of legislative/parliamentary institutions, parties are oligarchic to the degree that the inner circles control the elected representatives. In the Soviet Union, notoriously, the Communist Party dictated votes, not the electors. “Measurement of the disparity between electors and members is thus equivalent to measuring the degree of oligarchy which impregnates the systems that we term democratic.” 

The more ‘totalitarian’ parties have “general aims: they provide complete and coherent systems of thought about society; they aim at a total organization of national and even international life.” Some have hit upon the “brilliant idea” of forming “alongside the party…a series of ‘doubles'” or satellite organizations: youth groups, women’s organizations, sports clubs, clubs for veterans, literary clubs, and so on. Such groups bring in non-party members who, with exposure to “discreet propaganda,” may be recruited as new party members.

As for party ‘leadership,’ Duverger rightly considers it “democratic in appearance and oligarchic in reality,” except in Fascist parties, which “are bold enough to confess in public what others practice in secret.” He also notices that parties whose leaders are elected democratically very often do not choose people like themselves. “Country folk do not choose country folk as their parliamentary representatives,” usually preferring lawyers “because they consider them to be more capable of defending their interests in parliament.” By contrast, Communist oligarchs often choose working-class men to join them. In this, Duverger sees a conflict between “two conceptions of representation, one juridical founded upon election and delegation, the other technical and founded upon a de facto similarity between the masses and those who govern them.” He pauses to wish for “a truly scientific democracy,” as indeed a disciple of Marx might do, “in which parliament would be made up of a true sample of the citizens reproducing on a reduced scale the exact composition of the nation,” perhaps some combination of the Athenian lottery and a Gallup poll. 

In sum, by 1951 “two essential facts seem to have dominated the evolution of political parties since the beginning of the century: the increase in the authority of the leaders and the tendency towards personal forms of authority.” This trend, it must be said, has continued, quite in opposition to the hopes and expectations of the eminent sociologist Emile Durkheim, who “saw in the weakening of power and in its progressive ‘institutionalization’ the fundamental characteristics of democratic evolution.” However, it must be said that democracy has, following Durkheim and as anticipated by Tocqueville, also seen a very substantial, concurrent increase in institutionalization—namely, in the formation of ever-more-extensive bureaucracies or ‘administrative states.’ Indeed, Communist ‘leaders’ such as Stalin and Mao have not only decried bureaucracy but purged it, although it usually returns after the pitchfork stops prodding, the modern state requiring some form of regularized rule in order to hold itself together.

Summarizing his hypotheses on political parties, like a dedicated historicist Duverger identifies “three phases in the evolution of parties”: the “domination of parliamentary representatives over the party” is the first phase, a condition “of relative equilibrium” between the two groups is the second, and “finally the domination of the party over parliamentary representatives” is the third. Each of these phases “correspond[s] to a certain type of party,” with the historical trend going from liberal to socialist to communist or fascist. In his own career, Duverger obviously attempted to ride that supposed trend. Other politicians and political intellectuals have done the same.

Parties exist within regimes that feature party systems. There are, of course, one-party systems, but most systems include two or more major parties, sometimes a plethora of fairly small parties. They are all systems in the sense that they all have parts that interact with one another; even one-party systems see interactions among factions within that party. To understand these systems, one must first consider “similarities and disparities that can be discovered in the internal structures of the individual parties that make up the system” and then distinguish elements not merely similar but common to all parties, such as numbers, size, alliances, geographical location. “A party system is defined by a particular relationship amongst all these characteristics,” no matter the number of parties in the system. They are class structure (landed aristocracy versus industrial and commercial capitalists, capitalists versus working class), ideology (which “are never simple epiphenomena in relation to the socioeconomic structure,” as Marx and Engels claimed), and—most important among the technical causes—”the electoral regime.” While it is true that the party system “exercises a vital influence upon the electoral regime,” the reverse is also true. Class structure and ideology are not technical causes, and Duverger regards them as the “most decisive” drivers of party formation itself.

The party system exerts influence on the electoral system in several ways. A two-party system “favors the adoption of the simple-majority single-ballot form” of electoral system, a multi-party system reinforces proportional representation of parties, a one-party system favors party-rigged elections or no elections at all. “The party system and the electoral system are two realities that are indissolubly linked.” And again, the reverse of these causal pathways is also true. Duverger once again cautions his readers that “these very general propositions define only fundamental tendencies; they are far from including all the influences of the electoral system on party systems” but serve “solely as a first working hypothesis.”  

A two-party system can endure only if both parties differ only in “secondary aims and means,” with “a general political philosophy and the fundamental bases of the system are accepted by both sides.” But when the two parties dispute “the very nature of the regime and the fundamental concepts of life,” the conflict “assumes the aspect of a veritable war of religions”—religions themselves being regimes, it might be added. “This is equivalent to saying that the two-party system is inconceivable if one of the two parties is totalitarian in structure.” As a Marxisante thinker, Duverger inclines to dialectical conflict, insisting that “the center does not exist in politics,” but at most consists of a ‘Center-Left’ and a ‘Center-Right.’ “The fate of the Center is to be torn asunder, buffeted and annihilated” because “the dream of the center is to achieve a synthesis of contradictory aspirations,” a power that exists only in the mind never in practice. To foster a relatively moderate, if not centrist, political regime, the simple-majority single-ballot system works best, as it inclines toward two parties, both of which usually need to capture the unstable and contested middle ground in order to succeed in elections. 

By contrast, a multi-party system can endure much sharper regime divisions and ideological conflicts. This system goes well with proportional representation. “Proportional representation” in legislatures “always coincides with a multi-party system: in no country in the world has proportional representation given rise to a two-party system or kept one in existence.” This does not mean that proportional representation gives rise to an ever-increasing number of parties, since ‘regime’ parties, ideological parties, tend to encourage and indeed enforce discipline upon their members, discouraging the habits of mind and heart that lead to schism. 

Single-party systems arise when monarchy meets democratization. (Once again, Tocqueville regards despotism as one possible political outcome of social equality; he had Napoleon as his prime example.) Once in power, however, the supreme party “must conquer the natural passivity of the masses, their fundamental conservatism, in order to win them over to the changes that have been undertaken,” as when Stalin undertook to liquidate the ‘Kulaks’ or wealthier peasants. Further, the party “must overcome the tendency to inertia and conservatism of its own members,” comfortably ensconced in their ruling offices. Duverger denies that the Fascist regimes do this, however, after they have seized power. Fascism, he claims, “does not lay hands on the economic and social structure” of the regime. But ‘laying hands’ on someone or something is one thing, co-opting it another. The Nazis may have allowed industrialists to continue in power, but their ingenuity went into the construction of military equipment and gas chambers. Their efforts were redirected.

“The development of parties has brought about a profound transformation in the structures of political regimes.” The tyrants of contemporary one-party state “bear only a remote resemblance” to previous tyrants, while “modern democracies,” with their “several organized and disciplined parties,” differ noticeably “from the individualist regimes of the nineteenth century, which depended on the personal interplay of members of parliament who were very independent of each other.” If a democratic regime is one “in which those who govern are chosen by those who are governed, by means of free and open elections,” the existence of parties has “greatly changed” them. Parties have inserted themselves between electors and representatives, the people and parliament, since “before being chosen by his electors the deputy is chosen by the party” and “the electors only ratify this choice.” Representatives now receive “a double mandate”—first from the party, then from the electors. Who, then, does the representative represent? After all, “every party system constitutes a frame imposed upon opinion, forming it as well as deforming it.” True, a party system is “the result of public opinion,” but public opinion gets shaped by the party system, “as that is shaped by the circumstances of history, the evolution of politics, and a whole combination of complicated factors amongst which the electoral system plays a dominant part,” along with ballot procedure. So, for example, if I live in a regime with a two-party system, I will likely vote for a candidate whose opinions and policies are somewhat remote from my own, simply because there are only two ‘practical’ choices before me, whereas in a multi-party system I may well find a candidate more to my liking and, on the other extreme, in a one-party system I will have no choice at all—maybe not even the choice of whether or not I vote. Such party-formulated opinion is now “raw opinion,” since “parties create public opinion as much as they express it.” For example, Marxists make much of ‘class consciousness,’ “but class consciousness does not exist unless a party exists to awaken and develop it,” as the higher officials of Marxist parties seek to mold public opinion, granting it leeway only under duress. Duverger goes so far as to claim that parties “give skeletal articulation to a shapeless and jelly-like mass.”

Having exaggerated, he does relent. “Public opinion, electoral system, and party system…form three interdependent terms which are not unidirectional in the influence of each upon the other.” Modify the electoral system and you modify the party system; modify the party system and you modify the way in which opinion will be expressed; modify the electoral system and you will modify “the representation of opinion” in the regime.

Part of the hesitation or confusion arises in his assumption that “the development of parties has burst the bonds of the old political categories inspired by Aristotle or Montesquieu.” According to Duverger, “the classical contrast between parliamentary, presidential, and National Convention regimes can henceforth no longer serve as the pivot for modern constitutional law” because Kemalist Turkey, Soviet Russia and Nazi Germany “were profoundly similar” despite the fact that Turkey initially practiced a National Convention regime, the Soviets a “semi-parliamentary regime,” and the third “a semi-presidential regime.” The real modern distinction is between single-party, two-party, and multi-party systems. That is indeed a real distinction, but there are serious problems with the argument. First, and most obviously, Aristotle is no modern and Montesquieu is. This means, among other things, that such categories as ‘parliamentary,’ ‘presidential’ and ‘National Convention’ don’t immediately fit the Aristotelian categories of rule by the few, the one, or the many, respectively. More significantly, there is no reason why Aristotle would in the least be fazed by modern regimes, whatever the terminology now applied to them may be. He might well need to think carefully about the modern state, an entity he never saw. But Aristotle always looks not only at institutions but at the persons who rule—their number and their virtues and vices. Modern political parties quite obviously lend themselves to his approach. Montesquieu’s system of regime classification, which adapts Machiavelli’s simpler classification (the one, the many—republics are any regime that is not rule of the one), could also readily account for political parties, although he would be profoundly disappointed at the existence of ‘totalitarian’ ones.

Duverger additionally claims that a two-party system in which one party controls all three branches of government makes the existence of those branches a “constitutional facade” for the rule of “the party alone.” That is true if the party is the real regime in such a system, as when a Communist party rules, but in actual republics, which distribute genuine legal protection to minority parties, that isn’t true, or is not true in the same way. Aristotle would easily see the difference because he would consider the real regime, not the paper regime, and for him not only are a regime’s ruling institutions the ones that really rule, not the constitutional facades but the regime itself consists of institutions, rulers (one, few, many and good or bad) but a way of life and a purpose, all what in Duverger’s language should be considered distinct and interdependent variables. 

Duverger concludes with a very important observation, even if is not a simply true one. “The organization of political parties is certainly not in conformity with orthodox notions of democracy. Their internal structure is essentially autocratic and oligarchic: their leaders are not really appointed by the members, in spite of appearances, but co-opted or nominated by the entire body; they tend to form a ruling class, isolated from the militants, a cast that is more or less exclusive.” If so, and if parties moreover “impose a prefabricated mold upon” public opinion, then why do the ways of life and purposes of political parties (in 1951, say, in the United States and the Soviet Union) differ so sharply? “All government is oligarchic,” he claims, following Rousseau. For Aristotle, oligarchy means the rule of the few wealthy men, whereas aristocracy means the rule of the few virtuous men. Meaningful distinctions can be drawn between these regimes and between the four others, distinctions that affect that actual lives of the persons who live under them. Duverger more or less admits as much by conceding that “the Marxist distinction between nominal liberties”—such civil liberties and freedom of speech and of religion, the right to vote, and other ‘paper guarantees’— and “real liberties”—a guaranteed job, economic equality—is “only partly correct” because although “the political liberties recognized by Western regimes remain a formality for a large section of the masses for lack of an adequate standard of living, of adequate education, of social equality or of an adequate political equilibrium,” such formal liberties “may become real liberties,” someday—presumably, if neo-Marxist socialist parties take charge. “All government is by nature oligarchic, but the origins and the training of the oligarchs may be very different, and these determine their actions.” Replace Lincoln’s formula—government of, by, and for the people—with the formula, “Government of the people by an elite sprung from the people.” Unfortunately, this formula would make the regime that featured George Washington and his colleagues inferior to the one that featured Mao Zedong. Similarly, when Duverger writes that “a dictatorship with a single people’s party tending to create a new ruling class is nearer to democracy than party-less dictatorship of the personal or military type which strengthen the feudal powers in their control.” But was the Stalin regime really more democratic than that of Nicholas II? Is one’s local hereditary oligarch, calling himself an aristocrat, really any less democratic than one’s local commissar, commoner though that commissar would be? And, as Aristotle saw, can ‘democracy,’ however defined, at least sometimes be a bad regime?