State formation, parties and democracy. Studies in comparative European politics

[p. 69]

II.
Political parties

[p. 71]

chapter five | parties, elites, and political developments in western europe

Introduction^*

This chapter is in many ways an exercise in the impossible. It treats exceedingly complex social phenomena that are tacitly reduced to a common denominator by the use of deceptively simple concepts such as ‘party’, ‘elite’ and ‘political development’. It threatens to fall victim to what someone once called the ‘propinquity fallacy’: because Europe is one geographic area, it is supposed that its political experiences can be lumped together in one general treatment. This study deals, not with one political system at a given time, but with widely different systems through time - a somewhat strange exercise for one who is thoroughly skeptical of such over-general constructs as the ‘European continental political system,’^1. or of the superficial assumption that European societies followed a similar political course. The addition of the s after ‘political development’ in the title is therefore deliberate. It is my profound conviction that true analysis will pay detailed attention to variation in political developments between European states as well as within each of them.

At the same time, this recognition makes the task for a single political scientist well nigh hopeless. He has intimate knowledge of a few systems of government at the most, and is likely to read his own rather than true conclusions into those systems he does not know so well. What follows here consists therefore mainly of a series of eclectic remarks - usually termed hypotheses, but more honestly called impressions. At the outset of the journey I seek refuge in that excellent, if worn-out, defense that it is up to those who really know the individual political systems concerned to test, prove, or more likely to disprove, my generalizations.

The starting point of this chapter is the proposition that European states fall prima facie into at least three distinct groups: (1) countries which developed slowly from oligarchies into consistently stable democracies: e.g. Britain, the Scandinavian countries, The Netherlands, Belgium, and Switzerland; (2) countries which have undergone serious reversals in political regime, whereby democratic constitutions have given way to autocratic or even totalitarian systems of government: e.g. France, Germany, Austria, and Italy; and (3) countries with authoritarian regimes of a somewhat traditional nature, in which democratic groups tend to form at most an underground or exiled opposition: i.e. Spain and Portugal.^2.

It is much easier to say what factors are not responsible for these differences

* This essay first appeared as a chapter in LaPalombara, J. and Weiner, M. (1966) (eds.) Political Parties and Political Development, Princeton: Princeton University Press. I am grateful to the late Otto Kirchheimer, Val R. Lorwin, and Joseph LaPalombara, who criticized an earlier draft of this paper.

1. Gabriel Almond's analysis (1956) in ‘Comparative Political Systems’, Journal of Politics, Vol. 18 is a considerable advance over earlier writings that ascribed ‘Continental’ politics to such institutional factors as proportional representation and assembly government. Even so, his statement: ‘The Scandinavian and Low Countries stand somewhere in between the Continental pattern and the Anglo-American’, still betrays a similar attitude. Why should France, Germany, and Italy be more ‘continental’, than Holland, or Switzerland, or more ‘European’ than Britain? One wonders whether the description of the smaller European democracies as ‘mixed’ is not in fact an elegant way of saying that they apparently have some Anglo-Saxon virtues in addition to a number of ‘continental’ vices.

2. As the main focus of this chapter is on problems of political development, I shall not deal further with the latter group in the pages that follow.

[p. 72]

in political development than to indicate their actual causes. Obviously there is no immediate relationship to differences in stages of economic development. In the group of stable democracies there are countries that underwent the industrial revolution relatively early (Britain, Belgium) and countries which entered the modern industrial era late (e.g. Norway and The Netherlands). Similarly, German industrial development came relatively early and in full force; yet here the lapse into totalitarian dictatorship was the most gruesome ever. Moreover rapid economic development did not save the French Fourth Republic, nor does it stabilize political conditions in Italy.

Consequently it will be necessary to probe deeper and to seek for other factors that are often of an historical nature. The main variables on which this paper will center are: (1) the importance of the earlier elite setting; (2) the degree of coincidence or disparity between political and economic developments; (3) the ‘reach’ or ‘permeation’ of (democratic) parties as against other power holders in various European societies; and (4) the cleavage lines of the party system itself.

It is not suggested that these factors are sufficient to give a satisfactory explanation of the very complex and diverse processes of development which European countries underwent, whether generally or individually. They have been selected primarily because of their interest for comparative purposes in accordance with the terms of reference set for the papers in this volume.

The importance of the earlier elite setting

Differences in political development before the Nineteenth Century

The great complexity of the relationships among various social classes and status groups in European society, as that between state and society generally, has tended to be confused by the cliché assumption (found typically in college textbooks as well as the Communist Manifesto) that there was a ‘natural’ evolution in Europe from feudalism through absolutism and bourgeois revolution toward modern democracy. This view is an egregious simplification for a variety of reasons.

First, it pays far too little attention to the fact that the term ‘feudalism’ is used to describe fundamentally different structures in medieval Europe. The political relationships among king, aristocracy, clergy, cities, and peasantry as well as the economic relationships among landowners, burghers, artisans, peasants, and serfs showed great variation. If the starting point differs how one can expect linear or even parallel developments afterwards?

Second, present-day European states originated in very different ways. Roughly speaking, one can divide these states, according to the manner in which political unification came about, into four groups: (1) those in which effective centralization came early and with relatively little tension (e.g. Britain, Sweden); (2) those in which centralization came early but against considerable resistance (e.g. France); (3) those in which centralization came late but fairly gradually (e.g. The Netherlands, Switzerland); and finally (4) those in which central political power was established only as a consequence of considerable political violence in the nineteenth century (Germany, Italy).

[p. 73]

Of these four groups (1) and (3) had eventually rather similar characteristics. There was at no time a violent clash between political and social realities. Central power enmeshed itself gradually into the social system, and both regional and social groupings in turn achieved a growing influence on the center, thus making for a society which was both truly integrated and fairly pluralistic in nature.

Things were different in France, Germany, and Italy. There central control tended to be imposed by military and bureaucratic power. Hence the state came to some degree to hover above the society; the ruled came to feel themselves subjects rather than citizens, and to regard authority with a mixture of deference and distrust rather than as a responsive and responsible agency in which they had a share.^3.

Third, differences in the manner of social and economic development, even before the nineteenth century, tended to strengthen this contrast. In Britain and The Netherlands economic development ever since the middle ages was relatively free from state intervention. Autonomous economic development tended to make the newly rising bourgeoisie a much more powerful challenger of the powers that were than equivalent groupings could be in, say, Colbertist France or Cameralist Prussia. In the latter countries the state took a much more active hand in economic development, and in the process bureaucrats tended to become more managerial while the bourgeoisie tended to become more officialized than was true in Britain or the Low Countries. In the latter case, civil freedoms and a measure of responsible government preceded the establishment of a powerful central bureaucracy; in France the new social forces were eventually powerful enough to revolt, but in the process they succeeded only in building up safeguards against the bureaucracy rather than absorbing it or making it fully accountable; in the German Reich, finally, liberal groups failed to seize power and fell prey to the stronger hold of the Polizei- or Beamtenstaat.

Finally, European societies experienced different effects from the religious wars and their aftermath. In some countries the religious composition of the population remained homogeneous (whether Roman Catholic or Lutheran). There the church often remained for long an appendage of the upper classes; if this assisted them in their bid for the support of more traditionally oriented lower-class elements, it also tended eventually to provoke both fundamentalist and anticlerical protest. In other countries (notably in Switzerland, The Netherlands, Britain, and parts of Germany) various religious groups contested with one another until they finally reached some measure of tolerance or accommodation. In this way religious pluralism^4. and religious dissent often provided the spearhead of political resistance against entrenched elites, ultimately forcing recognition of the limits of state power and of the justice of individual and corporate rights.

3. In Germany most political forces submitted fairly rapidly to the existence of the new Reich, in contrast to the situation in Italy, where Church resistance and regional opposition continued for a much longer time. This made the existence of the new Italian state for a long time more precarious but may, on the other hand, have provided a safety valve that prevented nationalist unifiers from going to the extremes experienced in Germany.

4. The fact of religious pluralism seems more important than the particular religion in question. Whereas the Catholic Church in the Latin countries identified closely with vested social interests and alienated large sections of the population in the process, Dutch Catholics turned into a distinct protest group that pleaded for a separation of church and state, represented a considerable challenge of the outs against the dominant Liberal bourgeoisie, and maintained an effective hold on lower-class groups. Similarly, Lutheranism tended to be much more an instrument of vested authority in Germany than in Switzerland, while in Scandinavia it played the dual role of both maintaining an official religion and inspiring fundamentalist protest against a too modernist sphere in the central cities. Calvinism too was in practice much more nonconformist in some societies than in others, depending on whether its hold was strongest on existing elites or on lower-class elements.

The transition to modern democratic politics

Through such factors (and others such as the incidents of geography and war)^5. some political systems in Europe had hardened along autocratic lines by the eighteenth century, while others had maintained or even strengthened a pluralist setting that, however oligarchical, allowed a measure of political influence to a variety of political and social groups. This vital difference was to affect the establishment of

5. Geographic factors made certain European societies more secure from foreign attack than others: the insular position of England, the mountains of Switzerland, the rivers and canals of The Netherlands made these countries to a large extent immune against invasion on land. Consequently there was no urgent need for them to develop large standing armies. This had profound consequences for domestic political and administrative structures. In Lord Esher's telling phrase, the Navy often proved ‘a constitutional force’, while an army was more readily ‘a royal force’. Similarly, the early development of a citizen-militia in Switzerland was a great deal removed from the compulsory militarization of Prussian Untertanen.

[p. 74]

political parties during the nineteenth century in at least two respects: in the ease with which they became a recognized part of the political system, and in the role which they came to assume within it.

In Britain, the Low Countries, Switzerland, and Sweden, conciliar forms of government, whether in cities or in the center, had a long and honorable tradition. The style of politics tended to be one of careful adjustment, of shared responsibility, of due respect for ancient privileges. Attempts at absolute kingship eventually broke on the concerted strength of particularist interests, whether corporative, regional, or social. As the political order was in a very real sense built upon parts, the idea that men could reasonably be partisans found ready recognition even before the age of formalized party politics. There never was a ‘monochromatic, unicentric world,’ in Sartori's sense, to form an obstacle to the formation of parties.^6.

In these countries the view that government was somehow a trust toward the governed had old roots, however elitist actual systems tended to be until late in the nineteenth century. Intra-elite competition, being a recognized and even institutionalized phenomenon, made it easier to weather the ‘crisis of participation’. Conflicts between towns, between town and country, or among various religious groups created certain links between clashing oligarchies and sub-political groups below them. Competing elites sometimes sought lower-class support to strengthen their position, thus granting the lesser orders a political title and whetting their political appetites. Conversely, new claimants could exercise some influence on an oligarchical system simply by the threat of potential support to one or other side within it. Once some social groups were granted a measure of influence, this tended to provoke further demands from those yet further down until, finally, the burden of proof in the suffrage debate came to rest on those who defended restriction rather than on those who advocated extension of suffrage. Some upper-class groups came to doubt their own title, while most came to realize that fighting democracy might be more dangerous to their social position than democracy itself. Thus both pressures from competing élites downward and concomitant pressures from sub-elites upward made for a competitive gradual extension of democratic rights.

This process was facilitated by the circumstance that it came about in slow, evolutionary ways. Neither in political theory nor in actual behavior was there an abrupt transition from elite politics to mass politicization. Political newcomers were slowly accommodated. At any one time they tended to be given at most only part-power enough to give them a sense of involvement and political efficacy but not enough to completely overthrow the evolving society. Older political styles that had been developed to guarantee the rights of aristocrats or hauts bourgeois were thus more easily transferred. The ‘political domain’, to use Neumann's term,^7. expanded only slowly. Since at any one time the political stakes were relatively modest, the upper classes were less afraid and the lower classes less threatening. Older and newer elites were thus held more easily within the bounds of one constitutional, if changing, political system that neither alienated the one into reactionary nor the other into revolutionary onslaughts on it. In time, however, the over-all political order could thus become more truly responsive to the demands

6. Giovanni Sartori (1976) Parties and Party Systems; A Framework for Analysis, Cambridge: Cambridge University Press.

7. Sigmund Neumann (1956) Modern Political Parties, Chicago: University of Chicago Press, p. 404.

[p. 75]

of a wide variety of political groups within it. In 1867 Bagehot thought ‘dignified parts of government’ necessary to keep the masses from interfering with the ‘efficient government’ of the few; a century later the many were efficiently using those very same ‘dignified parts of government’ to secure substantial concessions to themselves.^8.

Developments were very different in those societies where power was heavily concentrated by the end of the eighteenth century. In France royal absolutism provoked truly revolutionary resistance of a much more drastic and upsetting character than appeared in the English Civil War of the seventeenth century, let alone in the Glorious Revolution of 1688. If the king called on the droit divin to claim absolute power, so did liberal thinkers of the Enlightenment on the basis of the nation or the people. From the outset a leading strand of French democratic thought became therefore ‘totalitarian’ in Talmon's sense,^9. becoming highly suspicious, for instance, of corps intermédiaires between the individual and the state. If in the countries described earlier pluralism seemed the natural corollary of liberty, in the latter it was often regarded as the prolongation of inequality and privilege. The traumas of the French Revolution created lasting and bitter divisions in French society. Articulate political groups continued to harbor fears and suspicions of one another, doubting one another's intentions and having different views of the legitimacy of past regimes and present institutions. Paradoxically, in that European country where popular sovereignty was proclaimed first and most explicitly no governmental system ever rested on a universal basis of popular support and respect. Traditionalist groups continued to be politically strong, and the newer bourgeoisie and the rising working classes came to be divided in their respective allegiances. Democratic regimes met with a continuous threat from the right. Democratic groups suspected the state even when they were nominally in control of it. This in tum made it more difficult for successive regimes to achieve their goals or to capitalize on their positive achievements and to gain legitimacy and lasting adherence throughout the nation.

In Prussia, and later in the German Reich, the bureaucratization and militarization of the society had gone much further than in France before the existing power division was challenged. In France democratizing forces generally triumphed, however precariously. In Germany the Kaiser-Junker-Army-Bureaucracy^10. complex was for a long time strong enough to manipulate the new social forces rather than to have to adjust to them. From the outset large sections of the new industrial capitalist classes were drawn into the existing power cluster; this left the fate of German liberalism to the faltering hands of mainly professional and intellectual groups rather than to a strongly unified economic class. In most European countries bourgeois elements had triumphed sufficiently to occupy key positions in the political system before the real onslaught of the working classes was felt. In Germany, on the other hand, the existing power groups were powerful enough to maintain themselves against both, even offsetting bourgeois demands for responsible parliamentary government with a careful weaning and manipulation of the new working classes. Typically, a democratic breakthrough came not of its own strength but only in the aftermath of lost wars. The explicit democratic articles of

8. Walter Bagehot (1867) The English Constitution, (1952 World's Classics ed., Oxford: O.U.P. pp. 4ff). The fact that most of the countries here treated have remained monarchies would seem to have been a consequence more than a cause of these developments.

9. J.L. Talmon (1965) The Origins of Totalitarian Democracy, London: Praeger.

10. For a sophisticated analysis of the way in which the Prussian Junkers maintained their social position and lost their political independence by their submission to the ‘new social factor... the state power’ see Joseph Schumpeter (1960) ‘Social Classes in an Ethnically Homogeneous Environment,’ in Imperialism and Social Classes, Cleveland, Ohio: Meridian Books, pp. 144ff.

[p. 76]

the Weimar Constitution were to become the hallmark of success of democratic forms on paper at the expense of social substance.

The Different Role of Parties

The role of parties in European countries varied considerably with such substantial differences in actual political development. In countries where modern mass democracy evolved slowly from a pre-existing pluralist society, various regional, social, and ideological groupings tended to form what might be called ‘protoparties’ at a rather early stage. Consisting of informal groupings seeking to obtain preferential treatment for themselves and the definite interests which they represented, they tended to fill certain functions of the modern political party (such as interest aggregation and to a lesser extent political recruitment), but not others (such as political socialization and political mobilization). As the increasing power of parliamentary assemblies tended to bring such groupings nearer to the effective decision-making centers, organization came to be at a premium. Similarly, when new social claimants came to exert pressure for representation, organization outside the parliament became not only profitable but essential for political survival. The process of party formation tended to spread therefore from existing competing elites downwards, but this veiy process also facilitated a reciprocal movement. Party organization itself created many new elite-posts even if only at sub-parliamentary levels. Second-rung leaders who provided an essential link with important elements in the expanding electorate had to be accommodated, and some in time fought their way to the top. Party competition for various groups in the electorate made some existing parties more responsive to new demands, while new social groupings came to imitate and expand existing forms of party organization.

In countries where autocratic regimes prevailed longer the development of parties showed different characteristics. Autocracy in its more explicit forms was incompatible with free party organization. Instead factionalism and a limited measure of interest representation tended to predominate. Democratic stirrings could take form only in intellectual protest movements or outright conspiratorial activity. Thus even some of the earlier democratizing movements, both on the liberal and on the socialist side, showed strong influences of secret societies.^11. In the more limited autocracies that the constitutional lawyers of another day used to call constitutional monarchies (as distinct from parliamentary monarchies) a measure of party organization could come about more easily. Even traditionalist political forces had eventually to resort to at least nominal electoral processes; but in their case parties were not so much the cause as a symptom of effective political power. Certain bourgeois and professional groups sought to use the parliamentary benches for a measure of oppositional politics that was often ineffective for lack of courage and organization. Further to the left, certain Weltanschauungsparteien showed tighter organizational forms and ideological programs; their verbal fervor tended to be symptomatic, however, of their weak position in the present. They made up for their lack of current influence with the vista of an utopian future, and could be ‘wholistic’ in their ideological claims precisely because they had little chance ever to be confronted with the compromises that partial power entails;

11. Note, for example, the role of the Free Masons in building up early Liberal and Radical parties, and the influence throughout Europe of Mazzini's La Giovane Italia. See Guglielmo Negri (1964) Three Essays on Comparative Politics, Milan.

[p. 77]

only a more basic political and social revolution could change their role in a fundamental fashion.

Finally, under conditions of more democratic rule the political role of parties became more important. But often past divorce from active political power continued to hinder them in the exercise of their nominal functions, while at the same time their somewhat timid hold on governmental power was endangered by the hidden sabotage or open competition from anti-democratic groups. We shall return to this point when we shall discuss the ‘reach’ or ‘permeation’ of democratic party-systems.

Coincidence or disparity between political and socio-economic development

The complex processes which we have come to denote in shorthand as the Industrial Revolution exercised a massive influence on political developments throughout Western Europe. Everywhere the self-contained political life of separate small communities was broken up, a development which freed the individual from older political bonds, allowed for the growth of wider, if often less compendious, political loyalties. Social and economic changes created considerable turmoil, which furnished the raw material for new political alignments. State and society grew more closely together as the scope of central power expanded, while simultaneously many new social forces came to exercise strong pressures for specific government action. In the process many new links were forged between the state and its citizens through the expansion of administration and the establishment of a great number of new political groups. The modern political party itself can be described with little exaggeration as the child of the Industrial Revolution.

It would be a mistake, however, to draw conclusions too easily about specific causal determination, for in practice socio-economic changes differed greatly from one country to another and within different regions of a single country. Furthermore the political effects of seemingly similar socio-economic changes varied according to the specific political settings in which they made themselves felt.^12. It seems useful therefore to consider the effects of economic development according to at least three criteria. When did economic development start? How fast did it come? What political effects did it have on various social strata? The criteria of time and tempo give four logical possibilities according to the following scheme:

Figure 5.1: Modes of socio-economic development

12. See earlier sections in this chapter.

[p. 78]

Of these four possibilities, II is not a real one; early European economic development germinated slowly. I is more representative for European experience. Without further proof it may generally be postulated that political strains were comparatively easy to cope with in this case: the very slowness of the process of socio-economic change gave the political system considerable leeway in meeting social and economic changes and in adjusting itself to them; these changes themselves were at any one moment also less drastic, hence less upsetting.

Somewhat similar considerations apply to III, but here a new factor enters. While in I social, economic, and political changes tended to move concurrently, a certain disparity between political and socio-economic changes could arise in countries in which economic change came late. Once new political ideas, new institutions, and new political techniques had developed in certain countries (like the United States or Great Britain), they could not but influence similar groupings in economically less developed societies as well. Political factors could therefore acquire a much greater autonomous momentum in the latter case. Thus the attempt was made to transfer certain institutional devices long before social realities showed corresponding changes. Ruling elites might deliberately concede the forms rather than the substance of democratic institutions to divert political unrest and maintain their own positions virtually intact. An elected parliament might be allowed, but not responsible government. Or a wide franchise might be granted, but only after adequate care was taken that this conveyed little power - the weighting of votes, the refusal of the private ballot, and slated apportionment of seats being particularly useful expedients in this respect. Rather than providing an effective lever in the hands of the masses, such ‘democratic’ reforms could paradoxically develop into a measure of plebiscitary control over them. This could result in an enduring alienation of sizable sections of the population rather than in their permanent integration in an effectively responsive political system.

France provides the classic case in Western Europe of such a disparity between political and socio-economic changes. At a very early moment the country was caught in the Whirlpool of mass politics. The principle of popular sovereignty was recognized long before a politically articulate people could make its will felt. Hence almost a century after the French Revolution the country could still live up to Laboulaye's description of France as ‘a tranquil people with agitated legislators’.^13. Agitation in the Carrousel de Paris, not being very meaningful in terms of a large number of social and economic variables in France, could not but prematurely disillusion French citizens with politics as such. To quote Philip Williams' description of the situation that prevailed in France until quite recently, ‘her atomized, small-scale structure promotes political individualism, strong local loyalties, and a political psychology more adapted to resistance than to positive construction. It reinforces the old tendency to incivisme, the lack of civic consciousness which makes so many Frenchmen regard the state as an enemy personified in the tax-collector and the recruiting sergeant.’^14. But this in turn meant also a lack of sufficient incentive for political change. To quote Williams again, ‘It was because there was no majority for action in the country that there was no pressure strong enough to overcome the resistance which found so many points of advantage in the

13. John E.C. Bodley (1900) France, New York, 1, 57.

14. Philip Williams (1958) Politics in Post-War France, London, p. 3.

[p. 79]

constitutional framework.’^15. When finally massive social and economic changes did come, these were consequently not easily channeled along earlier established institutional and political lines.

Finally, IV, in which economic changes are both fast and late, offers the greatest political difficulties of all; all the problems of III are repeated and compounded by the state of insecurity and flux which is inherent in rapid social and economic development itself. In Europe such conditions are found only in certain regions and usually within the bounds of a more comprehensive, stable, articulated political system. Not so in the developing countries, where politicization far outstrips socio-economic changes, and where these social and economic changes themselves, if forthcoming at all, only add to the discomfort of a body politic already weakened in other ways.

15. Ibid. p. 8.

Social classes and economic development

What were the political consequences of economic development (or the lack of it) for the various social classes in Europe? Obviously such a question can in the context of this paper be answered only in the most general, that is misleading, terms. Even the concepts we use, such as aristocracy, bourgeoisie, peasantry, middle classes, working classes, are not really satisfactory; they are indefensible (but necessary) simplifications of social categories and social divisions that are in reality very complex. The following discussion contains therefore, only a very rough sketch, and a highly impressionistic one at that.

First then, the effect of economic development on the nobility: In certain countries (e.g. The Netherlands^16. and Switzerland) the position of the nobility as against that of burghers and independent peasants tended always to be weak and to grow weaker as capitalism expanded. In other countries, notably in Britain, and to a more limited extent Germany, old aristocracies adapted themselves relatively successfully to the new facts of industrialization. This assisted them in their bid for continued political influence (even though other factors made for different attitudes toward democracy). A positive stance in favor of economic development and a paternalistic rejection of the extremes of Manchester Liberalism by both Tory squires and Prussian Junkers made it easier for the conservatives of both countries to maintain a measure of liaison with a significant section of the rising working classes (as well as to retain considerable rural support), which in turn facilitated the establishment of conservative mass movements in both countries.

In contrast, French and Italian aristocracies did not excel in economic initiatives and so tended to be anachronistic, their remaining political power resting more exclusively on traditional resources like their hold on the church, the land, the military, or administration. The gap between them and the rising bourgeoisie tended to grow wider than in either England or Germany. Or, to be more precise, the continued influence of the aristocracy divided the new bourgeoisie into those who adjusted themselves to the style of living of their continuing ‘betters’^17. and another section that sought to fight such influences. The bourgeoisie as a whole was therefore less easily credited with political ability or economic skills than were their Dutch or Swiss or English counterparts.

16. In the Province of Holland the Estates consisted in the seventeenth century of 19 members: one representative of the nobility, and 18 delegations of cities, manned by burghers.

17. In Prussia the same phenomenon occurred, much to the distress of Max Weber, who was angered by the tendency of ‘an amalgamation between a landed aristocracy corrupted by money-making and a capitalist middle-class corrupted by aristocratic pretensions.’ See Reinhard Bendix (1962) Max Weber - An Intellectual Portrait, New York: Anchor Books, p. 40. The greater political and economic prestige of the German upper classes presumably lessened revolutionary resistance to them, contrary to the situation in France and Italy, where revolutionary sentiment may have been fanned by the low prestige accorded to traditionalist political and economic elites.

[p. 80]

This fact influenced the political reactions of other groups in the population. For one thing, it helps to explain the large influence of professional and intellectual groups in these societies (considered by some observers to have been the outstanding characteristic of the politics of the French Third Republic)^18. which could not but strengthen the tendency toward the highly ideologized politics that seems typical of political societies in which political claims outstrip underlying socio-economic realities.

It also had an unfavorable effect on the relations between bourgeois and worker. In the Latin countries the patriarchal family firm long remained the characteristic form of economic enterprise. The Patron was a far cry from the revolutionary bourgeois of the ‘Communist Manifesto’. A low esteem for his economic qualities reinforced the defeatist outlook of the proletariat,^19. already skeptical of politics for reasons which we discussed earlier. A vast gap tended to develop not only between employer and worker but also between the professional socialist politicians who took part in the parliamentary game and the generally syndicalist masses who rejected all party action. This weakened both. It eventually assisted the Communist encadrement of the French and Italian working classes; and it goes far to explain the checkered results of both democratic institutions and social reform policies in France and Italy.

The evolution of working-class politics in most other European countries stands in considerable contrast. There the industrial revolution generally developed more thoroughly and effectively. At the earlier stages, social dislocations tended to produce a ‘hump of radicalism’. But the rapid growth of large-scale industries and urbanization soon laid the foundation for well-organized trade unionism and concurrent social-democratic action, quickly shifting from empty revolutionary phraseology to more immediate short-term goals within the existing socio-economic systems.^20. If this strengthened democracy in systems which were already democratizing themselves, it weakened the incipient stirrings of democracy in those societies in which modernization took place largely under continued autocratic auspices (as in Germany).

Somewhat similar factors influenced the political position of the peasantry. In all European countries the relative importance of the agricultural sector declined as economic development proceeded; but whereas in some countries this process caused relatively little political tension, in others it provoked violent conflict. In some countries strong protectionist policies both symbolized and maintained the power of certain agricultural groups; whether these were large landowners (as the Prussian Junkers) or a large mass of generally inefficient small farmers (as in France) depended on earlier developments in land tenure and social organization generally. In other countries the reduction of the agricultural sector went on at a much faster pace. But simultaneously foreign competition, self-help, and government policies facilitated modernization. Thus Danish and Dutch farmers^21. managed well. Typically, Scandinavian agrarians often cooperated with socialist parties, in contrast to France, where sizable blocs of peasant votes turned to rightist or Communist extremists; their mood was mainly one of apolitical malaise instead of one of definite expectation of positive action.

18. See for example T.B. Bottomore (1964) Elites and Society, London, pp. 64ff.

19. See Val Lorwin's (1958) ‘Working Class Politics and Economic Development in Western Europe’, American Historical Review, Vol. 63, pp. 338ff and S.M. Lipset (1964) ‘The Changing Class Structure of Contemporary European Politics’, Daedalus (Winter 1964), pp. 271-304.

20. Lipset, ibidem.

21. Dutch peasants, though split among the religious parties and the Liberal party, have nevertheless formed a powerful interest group across party lines. In addition, in the interwar period, and again since 1963, a small Poujadist-oriented Peasant Party has represented a more extremist protest-vote.

[p. 81]

Speaking more generally, economic development has caused the decline of some groups and the rise of others. The specific nature of the complex underlying processes has often been confused by the facile use of the hazy notion of the ‘middle classes’. One should at least distinguish between more traditional elements, like the artisanat, the retail traders, and small-scale employers who form the residue of social and economic developments, and the ‘new’ middle classes of technical, managerial, administrative, and professional people, who are rather their result. Political attitudes have tended to differ correspondingly. While in France, for instance, Poujadism tended to find its main support among the earlier groups, Gaullism has tended to appeal more strongly to the ‘new’ middle class. The rise of the latter has tended to make for a new dimension in the political controversy between right and left also in other countries, forcing both traditional socialists and traditional conservatives to take note.

The ‘reach’ or ‘permeation’ of the party system

Partly as a consequence of historical factors European parties have differed greatly in the extent to which they have permeated and enveloped other political elites. In some countries the role of parties has become all-pervasive; in others the parties have penetrated far less successfully to the mainsprings of political power. Substantial differences are also encountered in the extent to which parties have become true integrating agencies between political elites on the national and on the local scene. In this section the ‘reach’ of a party system is briefly analyzed along the following three dimensions: the extent of involvement of traditional political elites in the party system; the measure of absorption of new political claimants; and the degree of ‘homogenization’ which parties provide between national and local political elites.

Party systems and traditional elites

In European societies the relationship of traditionally powerful political elites and the party system seems to have taken one of three forms: they have participated from the outset, slowly learning to share power with newer groups; they have participated in the party system but only half-heartedly and with reactionary intentions; or they have stayed outside altogether, seeking to maintain their influence through other power structures (notably the military, the bureaucracy, business, or the church). The precise developments depended greatly on the way parties originated and the specific nature and extent of the democratizing process.

As we have noted some European parties were in many ways the outcome of earlier institutionalized conflict on the elite level; factions hardened increasingly into substantial political organizations as these conflicts spread from the elites downward into an ever widening circle of political actors. Though older elites were eventually confronted with new parties outside their control, they never came in immediate conflict with the party system. This facilitated the transition from oligarchical to polyarchical forms of government.

In other European countries parties were first established in opposition to autocratic regimes that forbade or at least restricted the scope of party conflict.

[p. 82]

Eventually in these countries too older elites found it necessary to participate in electoral processes. But parties established under their auspices tended to remain little more than outward appearances, democratic fig leaves, so to speak, for entrenched power positions that had their real basis elsewhere. Consequently right-wing parties in various European countries came to assume basically different attitudes toward the rules of the game of democratic party-politics. The acceptance of the substance of democratic ideals and practices is still the clearest criterion with which to distinguish Scandinavian or British Conservatives from, say the right in France, Weimar Germany or present-day Italy.^22.

In the latter the constant presence of potentially or actually antidemocratic parties within the party system has hindered the effective working of democratic politics; it has narrowed the range of democratic rule; it has caused disillusionment to spread to other potentially more democratic groups; and it finally eroded the very existence of democratic regimes.

The ‘reach’ of the party system over against other traditional political elites is revealed most clearly in its relation to the permanent bureaucracy. Bureaucracies have been far more responsive to the party system in some countries than in others. Much has depended on historical relations and the specific characteristics of the ensuing party system.

Thus it was of profound significance whether an articulated party system developed before, after, or concurrently with the rise of a bureaucracy. In France and Germany powerful bureaucracies were built up as social control-mechanisms long before non-bureaucratic social groups had learned to use the weapon of political organization to secure influence. Ever since, parties have had difficulty in obtaining full control and, to this day, bureaucracies have tended to enjoy a distinct political existence.^23. In Britain, on the other hand, the build-up of a modern civil service occurred after non-official social groups were securely in political control; ever since, the civil service has loyally accepted control by party ministers. Many other European countries would seem to fall between these two cases. State bureaucracies developed earlier than in Britain, but non-state groups were strong enough to make their weight felt simultaneously and ultimately to prevent them from becoming uncontrollable elements in the body politics. To use a somewhat simple metaphor, the British Civil Service was from the outset below party; the French and German bureaucracies were to a very real extent above it; in other cases parties and bureaucracies tended to be on one line. In systems where certain parties tended for long to have a hegemonic position they often staffed the bureaucracies after their own image; thus Liberal dominance made the Dutch bureaucracy long a Liberal perquisite, and in somewhat similar fashion the Democrazia Cristiana is at present heavily represented in the Italian bureaucracy. Alternation between parties could lead to an attempt to take the bureaucracy out of politics (as in Britain), but also to competitive politicization by rival parties. Coalition politics has often led to a careful distribution of administrative ‘fiefs’ to rival parties, as in present-day Austria and Belgium, or to balanced appointments of rival partisans not only at the ministerial level but also in cabinets du ministre, or even in established administrative posts.^24. Generally speaking, bureaucratic

22. These differences are apparent in the European Assemblies. Many Northern and Western Liberals have been hesitant to join the ‘Liberal’ caucus. Many Northern ‘Conservatives’ have stayed outside any grouping, finding it impossible to join the right, the left really being somewhat nearer to their beliefs, though not their label.

23. Bottomore (1964) p. 82, quotes a letter from the founder of the French École Libre des Sciences Politiques, Emile Boutmy, dated February 1871, in which he points to the need for an administrative elite in the following terms: ‘Privilege has gone, democracy cannot be halted. The higher classes, as they call themselves, are obliged to acknowledge the right of the majority, and they can only maintain their political dominance by invoking the right of the most capable. Behind the crumbling ramparts of their prerogatives and of tradition the tide of democracy must encounter a second line of defense, constructed of manifest and useful abilities, of superior qualities whose prestige cannot be gainsaid.’

24. See Val R. Lorwin (1962) The Politicization of the Bureaucracy in Belgium, Stanford: Center for Advanced Study in the Behavioral Sciences.

[p. 83]

traditions, fortified by political and legal doctrines, have prevented such devices from degenerating into the full excesses of the American nineteenth-century spoils system. Contrary to traditional belief, they have worked not too badly in those systems in which the party system itself was reasonably cohesive and effective. In a segmented society like the Netherlands, carefully balanced political appointments would even seem to have smoothed the relations among the parties and between politicians and bureaucrats. They have given parties the certainty that their views were taken into consideration at the beginnings of policy formation and in the details of policy execution; they have provided officials with a new avenue by which to obtain political support for administrative concerns; they have thus acted as brokers between officials and politicians and between various parties, softening political conflict in the process.

The party system and new claimants

As in their relation toward older elites, party systems have differed in their responsiveness to the claims of new groups seeking political representation. In European history the outs at the lower end of the scale have been either lower-class groups (notably the working classes and the peasantry) or religious protesters (e.g. Dutch Calvinists and Catholics, English Non-Conformists, the Norwegian Left). Again, the relation between these new claimants and the party system took any of three forms: their absorption into a preexisting party system which gradually came to widen its appeal; the formation of special parties; or their continued exclusion from the party system.

Robert Dahl^25. has suggested in the case of the United States that the non-appearance of a special working-class party was due to a considerable extent to the fact that representative government and a wide franchise were introduced before an urbanized proletariat came to exert new demands on the system. Hence parties and political techniques suitable to the operation of parties were evolved in time to grapple with this new challenge and to accommodate labor in the existing system. In contrast, in Britain representative government came early, the urban proletariat next, and general suffrage only at the end. While developments were such as to keep new rising groups within the constitutional order, the existing parties were not elastic enough to accommodate the rising demands of the working classes. In Germany, urbanization and the general suffrage preceded representative government, thus sterilizing political party activity into necessarily ineffective attitudes. With somewhat similar ideas in mind, Stein Rokkan has asked for further study of the interesting relationship between franchise extension, special electoral arrangements (such as weighting of votes, privatization of electoral preferences, proportional representation versus other electoral systems), and the mobilization of new groups into the political system.^26. These studies must then be further related, I suggest, to such factors as the earlier elite-setting, and the extent of disparity between political and social and economic development (also in their regional variations) to account for the measure of actual involvement of the out-groups in one political framework.

Generally speaking, then, not the establishment of special parties representa-

25. See Robert A. Dahl (1966) (ed.) Political Opposition in Western Democracies, New Haven, Yale University Press, Ch. 13.

26. See his (1961) ‘Mass Suffrage, Secret Voting and Political Participation’, European Journal of Sociology, Vol. 2, pp. 132-52.

[p. 84]

tive of the lesser groups of society but only their psychological identification with the political order and the responsiveness of that order, in turn, to new demands can serve as the true measure of the relation between the party system and new claimants. A responsive political order may ensure an effective political participation of new claimants without the establishment of special parties. Special parties, on the other hand, can both integrate and isolate according to the reaction of other parts of a party system. Thus Dutch Calvinists and Catholics established highly segmented political and social organizations but jointly rose to power and in the process ensured the integration of their clienteles into the political system, actually making it more integrated, responsive, and democratized.^27. The same cannot be said, it seems to me, of the Norwegian Christians or of various parties composed of nationality interests in the Austrian-Hungarian Empire. The uncritical use of the term ‘fragmentation’, then, does not bring the analysis much further if attention is not paid at the same time to the question whether a division of a political system into a number of quite distinct spiritual and political groups ultimately means the break-up of one society, or rather the growing of roots of very different groupings in one constitutional order. To use Sartori's terms, seemingly fragmented systems can in practice be centrifugal or centripetal,^28. and only exact sociological analysis can reveal which is ultimately the case.

Just as older elites in certain cases stayed outside (if not above) the party system, so various groups of society remained outside or below it even after the general franchise was introduced. As suggested earlier, one cause may have been a disparity between a strong politicization of the working classes and the granting to them of the means of effective political action. In the Latin countries anarchism and syndicalism were strengthened by the acute feeling that party and parliamentary activity could achieve little in practice. Vested interests may so continue to dominate the parliamentary scene that even their nominal voters may feel manipulated rather than active participants. This has been for long true of Italy, for instance, and still is to a lesser extent of most European countries.

If we combine the first and the second paragraph of this section, the ‘reach’ of a party system in relation to various groups in the society might be visualized as follows. Most removed (though not necessarily antagonistic) would be those political groups which are outside or below the system altogether; by definition they are politically unorganized. Following them are conscious anti-system groups that reject the existing political order but have some measure of group identification (e.g. the syndicalists, even though they rejected party organization and put their trust in spontaneous rather than institutional leaders). A somewhat closer participation is found among those who organize in political anti-system parties but with the deliberate aim of participating in order to destroy; in practice, however, the very act of participation tends to create certain vested interests in the system (cf. Robert de Jouvenel's famous dictum that there is more in common between two deputies one of whom is a revolutionary than between two revolutionaries one of whom is a deputy). Anti-system parties may therefore show a wide range between outright rejection and near-acceptance, and their influence may become so great that their presence becomes a significant variable within the system. Somewhat further on

27. See H. Daalder (1966) ‘The Netherlands: Opposition in a Segmented Society,’ in Dahl (1966) Ch. 6.

28. Giovanni Sartori, (1976) and (1966) ‘European Political Parties: The Case of Polarized Pluralism,’ LaPalombara, J. and Weiner, M. (1966) (eds.) Political Parties and Political Development, Princeton: Princeton University Press, Ch. 5.

[p. 85]

The area of effective democratic party government is restricted to that of the Government and loyal opposition parties - hence the ‘reach’ of a democratic party system is measured by its total proportion of the political pyramid. This proportion is very different in various European societies: while in some it is nearly coterminous with the whole (England, Sweden) in others it occupies only a relatively small area (present-day Italy, Weimar Germany, and French Fourth Republic). Proportions are far from stable moreover. On the one hand, isolationist and anti-system parties may gradually be domesticated into the system (e.g. the Nenni Socialists). On the other hand, increasing opposition to the system (as measured by a proportionate increase in the strength of anti-systems groups and anti-system parties), and mounting indifference (as apparent in an increase of subsystem groups) may narrow the area of democratic politics. Anti-system parties may seek deliberate involvement to discredit democratic politics and thus to increase both anti-system and subsystem groups. This is the reasoning behind traditional Communist tactics and seems to reflect fairly accurately the situation in France before 1958, when anti-system parties, anti-system groups, and political malaise made the area of democratic politics shrink to such an extent as to make it practically ineffective.
Figure 5.2: Governing elites

the road to involvement are those isolationist parties that have no chance to gain even part-power but continue to organize definite subcultural groups that wish their voices to be heard (even if with little chance of their being taken into account). Next in the scale would come opposition parties that effectively compete for office, proximity to power being the criterion with which to measure real involvement. Here again there is considerable scope for variation; whereas some are natural ‘outs,’ others are semi-government supporters. Finally some governing parties, tied most closely to the existing system, the extent of their dominance being the measure of their effective control. A simplified representation of this scheme is given in Figure 5.2.

[p. 86]

Party Systems: the center and local realities

The central-local axis provides yet another dimension by which to measure the permeation of party systems. Increased interaction between the center and the localities greatly affected the formation and organization of parties. Generally speaking, a two-way process took place: political forces in the center sought to extend their political bases by mobilizing political support over wider geographic areas, while political groups in the periphery organized to promote regional interests with the center.

This two-way movement resulted in very different situations. In some cases, a fundamental nationalization of politics led to a far-reaching ‘homogenization’ between politics at the center and in regional areas; such a movement was facilitated by the absence of strong economic or cultural regional cleavages, by good communications, and by the entry of issues that helped to nationalize politics (e.g. class). In other cases ethnic, linguistic, religious, or geographic barriers prevented such osmosis from taking place. Politicization tended to strengthen centrifugal rather than centripetal tendencies (separatist political movements like that of the Irish in nineteenth-century Britain or of nationality groupings in the Austrian-Hungarian Empire, forming their logical extreme). In yet different cases politics at the center and in the localities tended to remain in highly differentiated spheres, with only minimal linkages between them. Although this did not threaten national existence, it complicated national politics. Again, a comparison between Britain and France offers an instructive contrast.

In Britain leaders and labels penetrated relatively early from the center into the constituencies, thus drawing national and local elites into one reasonably unified system. Although certain regional sentiments and interests continued to have some importance, to this day providing British parties with distinct pockets of regional strength, they were not such as to fragment the decision-making process at the center. The essence of British politics is therefore national politics, and British parties are above all national political organizations. In France, on the other hand, local concerns long continued to dominate the choice of national parliamentary personnel. This caused a curious paradox: provided the local representative showed due respect for local sensitivities and interests, he was, on the national scene, as far as his constituents were concerned to a considerable extent a free agent. The French Chamber became therefore very much ‘La Maison sans Fenêtres’, a meeting place of local interests and individual personalities rather than of cohesive, integrated national political parties. Nationalization of politics occurred therefore more easily on the level of ideological debate than of political will, of political oratory rather than of effective national political organization. For the rest, the French Chamber tended to be more highly sensitive to interest groups (pressuring MPs in their local base) than to issues of more national importance. This accentuated the cleavage between the French bureaucracy (feeling itself the self-appointed guardian of France in a truly Parisian way) and the Chamber, stronger in resisting the executive than in dominating it, more ready to veto than to formulate national policies.

The ‘homogenization’ of politics between the center and the localities is therefore an important factor in the politics of both. An effective linkage helps to le-

[p. 87]

gitimize the national political order. Where links are absent, alienation is likely to ensue. The character of the party system is an important variable in this process. Parties can be agencies of both integration and disintegration. They assist national integration if they serve as genuine brokers between disparate regional or social interests (without losing their national existence in the process). They are likely to strengthen centrifugal forces, on the other hand, if they become the passive tools of sectional interests. Paradoxically, synthetic unifiers who seek to identify their own sectional interests with that of the one and indivisible nation can contribute as much to such disintegrating tendencies as those who deny the existence of one political community from over-particularistic concerns.

The cleavage lines of the party system

European countries reveal considerable differences according to the character and the intensity of the cleavage lines that form the basis for political conflict and political organization. These differences are partly due to objective differences in social structure; certain social cleavages did exist in some countries but not in others (e.g. ethnic diversity). They depend further on the circumstance of whether and to what extent particular cleavages were effectively politicized; factors such as religion or class have been much more exploited in some political systems than in others. Finally, considerable variations also exist in the persistence of cleavage lines in the party system. Whereas some issues have been of only passing importance and have subsequently fallen out of the political domain, others have remained characteristic dividing lines long after their original raisons d'être has been forgotten. In this way the particular history of past political controversy has continued to exercise a substantial influence on political loyalties and on the way in which new issues are focused and processed. Therefore only careful historical, sociological, and political analysis can do full justice to the distinct qualities of any given political system. It follows that it is much easier to categorize a number of cleavages that seem to have been historically important in European political development than to evaluate their importance for political stability or effective decision-making.

In early days David Hume considered ‘factions from interest’ and ‘factions from affection’ as the most normal cases, proclaiming, unlike Burke, the rise of a new category of ‘factions from principle’: ‘the most extraordinary and unaccountable phenomenon that has yet appeared in human affairs.’^29. Generally speaking, the most important dividing lines in Europe have tended to be: class or sectional interest (the landed versus the moneyed interests; parties representative of sections of industry or commerce, labor, or agriculture); religion (Modernists versus Fundamentalists, Catholics versus Protestants, Clericals versus Anticlericals, Anglicans versus Non-Conformists); geographic conflict (town versus country, center versus periphery); nationality or nationalism (ethnic minority parties, extreme-nationalist movements, and parties having their real allegiance to another national state, etc.); and regime (status quo parties versus reform parties, revolutionary, or counterrevolutionary parties).

The difficulty of qualitative analysis of the importance of cleavage structures

29. See David Hume (1953) Political Essays, (ed.) Charles W. Hendel, New York, pp. 80-81.

[p. 88]

comes out in the exaggerated attention paid to quasi-mechanical factors, such as the number of cleavages, or whether they run parallel to or cut across one another. Both English and American literatures seem to be based often on the a priori notion that the political universe is by nature dualistic, so that two-party systems are the self-evident political norm. This view is reinforced by Duverger's analysis, which attempts to reduce the explanation also of multiparty systems to a ‘superposition of dualisms’,^30. to the non-coincidence of dividing lines in the body politic. While ‘bargaining parties’ in a dualist system are likely to ensure both stability and the orderly solution of successive issues, so the Standard argument goes, a multiparty system leads perforce to fragmentation and immobilisme.

This view is based on a slender empirical basis. Britain and the United States have two parties, their politics are apparently satisfactory to the theorist; ergo, a two-party system is good. In contrast, France, Weimar-Germany, and Italy had many parties, their politics were unsatisfactory; hence a multiparty system is a lesser if not an outright degenerated form. This type of reasoning then leads to the curious term of ‘working multiparty systems’^31. - phenomena that are apparently somewhat akin to ‘the boneless wonder’ of Barnum's Circus. Such a view testifies to an insufficient awareness of the political experience of a host of smaller European countries (such as Belgium, The Netherlands, Denmark, or Switzerland) that have successfully governed themselves for generations under complex multiparty systems. Would it not be possible on the basis of the politics of these countries as confidently (but equally subjectively) to assert that the best political system is one in which all important social groupings have occasion to have themselves politically represented in separate parties, which can then use the forum of parliament and coalition government to reach the politics of compromise?

The confusion is clearly revealed by our tendency to hold two conflicting theories with equal conviction. On the one hand, we argue that politics is best served by a constant dualistic regrouping of political forces in distinct majority-minority positions. On the other hand, we hold with equal conviction that a political system can quickly be brought to the breaking-point if a number of cleavages come to run parallel to one another - for instance, if conflicts about religion, nationality, and class each make for the same division of society. Whereas we point at one time to the crisscrossing of cleavage lines as the main source of political inefficiency, we assert at another moment that only adequate cross-pressures, which offset tendencies toward increased polarization, can make for a working political community. It is to this variable that we look to explain why Flemish and Walloons, why Capital and Labour, why Clericals and Anticlericals can continue to cooperate in feasible political systems. I suggest that this paradox cannot be explained unless new variables are also taken into account.

Of crucial importance is not only the severity and incidence of conflicts, but also the attitudes political elites take toward the need to solve them by compromise rather than combat. Such attitudes are deeply rooted in political culture, itself the product of complex historical factors that differ greatly from one country to another. Traditional leadership styles, the traumatic memory of past conflicts (which may either perpetuate conflict, or cause parties to draw together), a realistic sense

30. M. Duverger (1954) Political Parties, London, pp. 229ff.

31. Sec Gabriel A. Almond (1960) The Politics of the Developing Areas, Princeton: Princeton University Press and Dankwart Rustow (1956) ‘Scandinavia: Working Multiparty Systems,’ in Sigmund Neumann (1956) pp. 169ff.

[p. 89]

of what can be reached through political action and what not, the presence of substantial or imaginary common interests, the extent to which party leaders are more tolerant than their followers and are yet able to carry them along - all are important. Unfortunately they are evasive of systematic analysis except in a specific context.

Parties and political elites at the present time

The relation between parties and political elites in present-day Europe may now be briefly discussed in three steps of increasing generality: parties and the ‘Iron Law of Oligarchy’; the differences in influence among different party elites; and the influence of party elites (as a genus) over against other political elites.

Parties and the ‘Iron Law of Oligarchy’

The ‘Iron Law of Oligarchy’ is both an analytical statement and a somewhat emotional political theory. In the following paragraphs we shall discuss each of these two aspects in relation to European experience.

From the analytical point of view the oligarchy model of political parties is generally buttressed by what seem to me to be three false arguments: first, the confusion of inequality of influence (as among leaders, militants, and voters) with oligarchy; second, the determinist fallacy which sees too direct a link between the social origins of politicians and class bias, in their politics; and third, the delusion of indispensability, which wrongly deduces an exclusive power position for those who fulfill functions that are socially indispensable. A short word on each of these: Robert Michels' analysis of the various factors (both technical and psychological) which give leaders in mass organizations considerable power over their followers has rightly become a classic in political science literature. The real proof of oligarchical domination is, however, in other directions. Are leaders always unified when they are subject to pressure from below? Are they virtually unaccountable and unremovable? Are they free agents instead of brokers seeking to reconcile various conflicts in society? To what extent and under what circumstances can they dispense with considerations of mass interest? That few men take many decisions is not proof of oligarchy, but that they are able to take any decisions they care to take, even disregarding strong political objections.^32.

Unfortunately European political science is richer in noisy debate on such issues than in concrete research. Consequently it engages all too often in ‘yes’ or ‘no’ arguments rather than in careful enquiry as to the actual degrees of influence exerted by leaders as against followers. Research proper^33. would presumably uncover great differences from one party to another and from one party system to another. It would reveal, one suspects, that leaders have far greater freedom in some matters (as foreign policy) than in other fields that impinge more directly on the daily lives of vast numbers of people. Furthermore, relations cannot be static; whereas at certain moments leaders will dictate policies, at others they will bend to explicit or even implicit demands from lower down. Parties, in other words, are almost certainly agencies of elite-recruitment and elite-maintenance, but they also serve as transmission belts for pressures from lower down. Most European

32. See Robert A. Dahl (1958) ‘A Critique of the Ruling Elite Model’, American Political Science Review, Vol. 2, pp. 463-9, and (1963) Modern Political Analysis, Englewood, N.J.: Prentice-Hall, Ch. 5. To break away from the tautology: ‘an elite is present when there is inequality of influence; there is inequality of influence, hence we live under an elite,’ Raymond Aron has proposed in his paper (1963) ‘La Classe Dirigeante: Mythe ou Réalité?’, rapport introductif, Association Française de Science Politique, November 1963, pp. 5-6, to distinguish between ‘classe dirigeante,’ defined as ‘une minorité délimitée, cohérente, consciente d'elle-même, qui prend les décisions majeures, use et abuse de sa situation privilégiée pour exploiter ou opprimer les masses’ and ‘catégories dirigeantes,’ for ‘les minorités, qui dans la société, occupent des positions telles ou exercent des fonctions telles qu'elles détiennent un pouvoir dans leur sphère propre et ont chance d'avoir une influence sur la minorité qui exerce le pouvoir politique au sens étroit du terme (le pouvoir central ou étatique).

33. There is great need in Europe for studies along the lines of Dahl's (1961) Who Governs?, New Haven: Yale University Press.

[p. 90]

parties would seem to be comparatively open agencies that allow for a great deal of intra-elite conflict as well as for the rise of new elite groups in competition to older ones. Parties work, moreover, generally in a democratic environment that permits publicity and criticism by competitors and outsiders and that forces actual accountability to independent electoral groups; this cannot but blunt oligarchical proclivities.

As to the second false test of oligarchy, a rapidly growing series of publications on the social background of parliamentarians, ministers, and party members^34. has confirmed that great inequalities continue to exist between various social classes. Upper- and middle-class elements are highly over-represented in all parties (including the explicitly working-class ones). In some countries there are signs that inequalities persist (or even increase) rather than decline. It seems an aprioristic sociological determinism, however, to conclude much more from this than the obvious - that European society is still far removed from equality of opportunity as among various classes (notably in matters of education). Social origin may but need not determine political sympathies. On the contrary, many politicians of working-class origin have been more conservative than socialist renegades from the upper classes. Politics is an autonomous process that certainly is affected by class factors but is not causally dependent on them. Theoretically a political elite (and above all competing political elites) composed almost exclusively of a large number of upper-class persons can still be fully responsive to pressures from below.

Lastly, elitist theory is often marred by what might be termed the delusion of indispensability. It is proved to the satisfaction of a theorist that a certain social group fulfills an indispensable function in society. It is concluded from this that the group has (or could have) sole control; for instance, that by withdrawing its services it could bring society to a standstill. In this way different observers have pointed to entrepreneurs, finance capitalists, bureaucrats, the military, the working class, or even the peasantry (in the physiocratic sense) as the true elite of any given society. Little attention is paid to the question whether such groupings are ever sufficiently cohesive to exploit the full power resources of their seemingly strategic social position, and what actual countervailing powers there are to stop them from even considering this. Similarly, many observers have jumped all too readily from the correct view that leadership is indispensable in large organizations (as in society at large) to the incorrect one that this gives a monopoly of power to any particular leadership circle.^35.

The ‘Iron Law of Oligarchy’, then, is defective as an empirical theory or even as an heuristic tool. But the extent of its popularity is at the same time a significant yardstick by which to measure differences in democratic realities in Western Europe. Most elitists have come from societies where democratic politics has had an uneasy life. Mosca and Pareto reasoned mainly from the precarious background of Italian politics before 1914. Michels wrote on the basis of a German Social Democracy whose lack of power caused them to envy as well as to try to emulate Prussian rulers. Sorel's anti-parliamentary writings on the need for a spontaneous elite mirror above all the frustrations of a French intellectual banking on a mythical working class. And even Ostrogorski wrote as much to bring a message as to study

34. E.g. the studies of J.F.S. Ross and W.L. Guttsman on Britain, Mattei Dogan on France, Kaufmann et al. on Germany, Pesonen on Finland, Sartori on Italy, etc., as well as the interesting symposium edited by Jean Meynaud (1961) on ‘The Parliamentary Profession,’ International Social Science Journal, Vol. 13, pp. 513-619.

35. An early trenchant criticism of Michels is contained in a book by the Dutch Socialist and criminologist, W.A. Bonger (1934) Problemen der Demokratie, Amsterdam.

[p. 91]

scientifically the processes of caucus politics in Birmingham or in American cities. Neo-elitists are, again, most frequently found in postwar France (M. Waline's Les Partis contre La République, 1948, being an early specimen, later followed in far more subtle terms by Duverger^36. and many others) and among certain social scientists in the United States who feel increasingly disenchanted with traditional eager hopes for democratic reform. Against this, very few thinkers in stable democracies, such as England, Switzerland, the Low Countries, or the Scandinavian countries, can be identified with those illustrious names. Even R.T. McKenzie's Political Parties does little but pay lip service to Michels and Ostrogorski.

Why this connection between elitism and societies with uneasy democratic institutions? As a political theory it is pervaded by an atmosphere of despair in political action. Elitists have a low opinion of politicians; according to Ostrogorski they are worse than either Cain or Harpagon. Elitists look at parties with equal distaste; according to Waline no self-respecting Frenchman could honestly subscribe to any of the French political parties before 1940.^37. Such pessimism about the present is reinforced by the use of the absolute yardstick of democracy as direct popular rule. Proof that even in generally accepted democratic systems men are far from equal comes therefore as a moral shock, still traceable in the grim delight that disappointed idealists, now turned ‘realists,’ continue to show at every new piece of evidence of obvious fact. In countries where the political order is effectively responsible to a wider range of political forces there is hardly the same temptation to engage in powerless invective against politicians and uncritical adulation of non-party elites; people can act and deem this sufficient.

36. It is interesting to note that in 1947 Duverger described Michels' Political Parties still as ‘a rather superficial little book’ (in ‘Die Politische Parteien and die Demokratie,’ Der Wähler, Schriften der Deutschen Wählergesellschaft, Erstes Heft), but in 1951 as ‘an excellent little book’, Les Partis Politiques, Paris, p. x.

37. M. Ostrogorski (1903) Democracy and the Organization of Political Parties, London p. 632; Waline, M. (1948) Les Partis contre La République, Paris: Éditions Rousseau, p 34.

Differences in influence among various party elites

The political power of party elites differs according to the internal structure of each party and its power in the party system.

Duverger has provided what is by far the most detailed and refined analysis of differences in internal party structure; he has carried the work as far as can be done short of further detailed analysis of particular parties. It would be invidious therefore in the context of this chapter to seek to add to his rich exposition except for one comment. It seems to me that his distinction between ‘internally’ and ‘externally’ created parties, however valuable as a starting point for analysis, is in danger of being overworked.

In the first place, not all ‘internally’ created parties answer to Duverger's implicit model of the French Radicals; in the process of time many middle-class parties have greatly extended their organization outside parliament. They could do so - unlike the French middle-class parties - because the nationalization of politics had proceeded much further in other European countries than in France. Thus the Dutch Catholics and Calvinists (and Christian-Democratic parties generally) have put on increasingly effective organizational drives that have intensified reciprocal action between parliamentary and extra-parliamentary elites. Secondly, many ‘externally’ created parties have tended to loosen up as they have approached office. Coalition tactics have often required a high degree of discretionary authority for party leaders; this has decreased the freedom of action of party segments outside

[p. 92]

the immediately dominant circle and has therefore reduced the difference from ‘internally created’ parties. Thirdly, it seems that in Europe ‘parties of social integration’ (in Neumann's sense)^38. are losing ground under the impact of the manifold social processes responsible for the process of de-ideologization that now occupies so much of our attention. Therefore in the short run at least the importance of the distinction seems to be declining. Whether the professionalization and the de-ideologization of politics will ultimately lead to a resurgence of irrationalist ‘parties of total integration’ is another matter.

Following earlier writers, Neumann has considered a party in office or in opposition the basis for his distinction between ‘parties of patronage’ and ‘parties of principle’.^39. Whereas office gives party elites access both to the traps and the trappings of power, opposition encourages the posture of uncommitted principle. Ceteris paribus, the value of this distinction would increase with the measure of ‘predominance’ (in Sartori's terminology) or ‘hegemony’ (in the words of LaPalombara and Weiner) that particular parties have. Long tenure has made certain parties almost indistinguishable from the formal state apparatus. Conversely, remoteness from office has increased the ideological element in those parties which we termed earlier isolationist parties.

The influence of the leaders of opposition parties stands prima facie in inverse ratio to the strength of their governmental opposite numbers. This simple statement, however, covers relations of great complexity. Government parties have differed considerably in their willingness or ability to exploit even nominally hegemonic positions. Lack of conviction, expediency and internal conflict has often posed serious obstacles in the way of doing so fully. The same factors have given opposition leaders a wedge with which to penetrate a seemingly solid government front. They have sometimes not been satisfied to play only the parliamentary game and have used other power resources, like bureaucratic connivance, interest group pressure, or mass propaganda campaigns, to thwart government action. At other times and places government and opposition parties have often formed a tacit condominium; certain issues are removed from the party game by mutual consent, but on the condition of regular consultation. The political process, in other words, is considerably more complex than a simple opposition or juxtaposition of political parties suggests, and consequently the analysis cannot stop with parties, as we have to do here.

38. Neumann (1956) pp. 404-5.

39. Ibidem, p. 400.

The influence of party elites as against that of other political elites

Perhaps the best measure to distinguish the relative hold of party elites on a political system as against that of other elites is to ask how far positions of political influence can be obtained through, as opposed to outside, party channels. So defined, this question is the obverse of the earlier one relating to the ‘reach’ of a party system. Where party systems are comprehensive, safely anchored in the main power positions of a society, and reasonably stable over time, the role of parties in the recruitment of political elites, or at least in their legitimation, is by definition considerable. In contrast, where parties operate on a narrow focus, where their position toward other groups in society is precarious, where the party system

[p. 93]

is generally unstable - there party elites occupy positions of doubtful permanence, other elites finding different loci of power and threatening to replace parties by whatever means at their disposal (e.g. the bouleversement of the French elite of the Fourth Republic in 1958).

Even in systems where the reach of the party system is wide and its stability considerable it would be wrong to conclude, however, that party elites enjoy a monopoly of political influence. In the first place, the stability of particular parties does not necessarily mean a stability of their elites; internal change-overs may considerably affect their personnel and their policies even though clienteles and labels remain much the same. In the second place, the wider the scope of party-controlled political activity, the more likely it is that elements of a very diverse nature will enter into it, thus introducing institutional, personal, and interest conflicts within the life of each party.^40. The superficial image of a tight elite breaks down, when intra-party as much as inter-party conflicts provide the arena in which the most disparate political conflicts are being fought out.

Perhaps this is one key with which to explain seemingly contradictory developments in present-day Europe. In many countries, especially the more settled ones, we may witness, on the one hand, an increasing penetration of party activities in society (e.g. by a further politicization of the bureaucracy, by an increase in party-tied interest groups, a closer control of the mass media, etc.). On the other hand, there are equally definite signs of a lowering of the temperature of party conflict, a de-ideologization of party life, a professionalization of party activity, and a bureaucratization of organized politics. Stein Rokkan has spoken in this context of ‘an intriguing process of historical dialectics’:

the extension of the suffrage increased the chances for a status polarization of national politics [thus raising the temperature of politics and increasing the role of party], but this very polarization brought about a proliferation of sectional and functional organizations which in turn tended to soften the overall strains in the system and reduce the level of polarization. What we tend to find is an accumulation of forces making for a narrowing of the alternatives for national politics, a fragmentation of the net-works of policy-influencing organizations, and a consequent decline in the importance of the decisions of the electorate-at-large. This may tend to lower the level of general political participation and to alienate from politics sizable sections of the once enfranchised citizenry, leaving the basic decisions to a bargaining process between interest organizations, parties and agencies and departments of the national bureaucracy.^41.

This seems an exceedingly interesting aperçu; but alas, we must also underwrite the author's final lines:

We know far too little about the dynamics of these developments and we need to do much more to facilitate co-operation and co-ordination of studies of these problems in different countries.^42.

40. It is interesting to note that the amount of direct interest-representation in parliamentary parties seems to differ considerably from one country to another (e.g. the studies on ‘The Parliamentary Profession’ in International Social Science Journal, Vol. 13, 1965, pp. 513-619). While in some countries interest groups become increasingly involved in internal party politics, in others they seem to keep parties consciously at arm's length. Perhaps the reason for such differences may be sought in differences in the political role and power of various assemblies, the measure of direct access interest groups have to the administration, the power and specific structure of individual parties, and differences in formal nomination procedures. The increasing importance of government action for interest groups has inevitably raised their concern with politics. But at the same time their need for constantly satisfactory relations with government, of whatever party, has put a premium on avoiding too close involvement with any one of them. See on this point Kirchheimer's subtle analysis in LaPalombara and Weiner (1966).

41. Rokkan (1961) p. 152.

42. Ibidem.