Hostname: page-component-cd9895bd7-dk4vv Total loading time: 0 Render date: 2024-12-26T05:23:34.394Z Has data issue: false hasContentIssue false
  • English
  • Français

Further Reflections on “The Elitist Theory of Democracy”

Published online by Cambridge University Press:  01 August 2014

Robert A. Dahl
Robert A. Dahl*
Affiliation:
Yale University
Get access Rights & Permissions [Opens in a new window]

Extract

An interest in the roles, functions, contributions, and dangers of leadership in popular regimes is not, of course, new among observers of political life. This has, in fact, been an ancient and enduring interest of political theorists. It is possible, however, to distinguish—at least in a rough way—two different streams of thought: one consisting of writers sympathetic to popular rule, the other consisting of anti-democratic writers.

It has always been obvious to practical and theoretical observers alike that even where leaders are chosen by the people, they might convert a democracy into an oligarchy or a despotism. From ancient times, as everyone knows, anti-democratic writers have contended that popular governments were unlikely to provide leaders with wisdom and virtue, and insisted on the natural affinity between the people and the despot. These ancient challenges by anti-democratic writers were, I think, made more formidable in the course of the last hundred years by critics—sometimes ex-democrats turned authoritarian when their Utopian hopes encountered the ugly realities of political life—who, like Pareto, Michels, and Mosca, contended that popular rule is not only undesirable but also, as they tried to show, impossible. The failure of popular regimes to emerge, or, if they did emerge to survive, in Russia, Italy, Germany, and Spain could not be met merely by frequent assertions of democratic rhetoric.

Fortunately, alongside this stream of anti-democratic thought and experience there has always been the other. Aware both of their critics and of the real life problems of popular rule, writers sympathetic to democracy have emphasized the need for wisdom, virtue, and self-restraint not only among the general body of citizens but among leaders as well.

Type
Research Article
Information
American Political Science Review , Volume 60 , Issue 2 , June 1966 , pp. 296 - 305
Copyright
Copyright © American Political Science Association 1966

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 See, for example, his comment: “It is popular leaders who, by referring all issues to the decision of the people, are responsible for substituting the sovereignty of decrees for that of laws.” See also his comments on “the particular causes of revolution and change” in democracies, e.g. “In democracies changes are chiefly due to the wanton license of demogogues.” The Politics of Aristotle, trans, with an introduction, notes, and appendixes by Barker, Ernest (Oxford University Press, 1952), pp. 168, 215Google Scholar.

2 In the Discourses, after noting that two “principi virtuosi” in succession were sufficient to conquer the world—Philip of Macedon and Alexander the Great—Machiavilli goes on to say: “Il che tanto più debba fare una republica, avendo per il modo dello eleggere non solamente due successioni ma infiniti principi virtuoissimi che sono l'uno dell'altro successori: la quale virtuosa successioné fia sempre in ogni republica bene ordinata.” Libro Primo Dei Discorsi Sopra La Prima Deca di Tito Livio, XX, in Tutte Le Opere Storiche E Letterarie di Niccolò Machiavelli, Mazzoni, and Casella, (eds), (Florence, B. Barbéra, 1929) p. 90Google Scholar.

3 He lamented, for example, that men of true merit (virtù) are sought for in difficult times but in easy times it is not virtuous men who are most favored but those with riches and the proper relations; in peaceful times, other citizens who are jealous of the reputation of the virtuous want not merely to be their equals but their superiors. Book 3, XVI, Ibid., p. 224. See also ibid. XXVIII, 239ff.

4 Du Contrat Social, Cho. IV. “De la démocratie,” pp. 280–81. Note his comment: “Il est contre l'ordre naturel que le grand nombre gouverne et que let petit soit gouverné,” p. 280 (Paris, Éditions Garnier Frères, 1962.Google Scholar)

5 Ibid., p. 282.

6 On Liberty and Considerations on Representative Government, ed. by McCallum, R. B. (New York, Macmillan, 1947), pp. 215–216Google Scholar.

7 May I register a dissent seemingly so minor that I fear it will appear to be nitpicking: the appropriateness of the label? I realize that Professor Walker has taken the expression “the elitist theory of democracy” from Lipset; but even if Lipset may have had his reasons while writing a preface to the major work of Michels for applying this phrase to Weber, Schumpeter, Parsons, and James Burnham, that is not a good reason for stretching it, as Professor Walker does, to cover others. I, for one, object to being labelled “elitist” not only because—as I hope to show—it would be inaccurate in implication even if it were a neutral term, but even more so because in our language and in our society it is unavoidably, I think, a pejorative, even a polemical epithet.

To substitute epithet for argument was, I am sure, not Professor Walker's intention. Nonetheless, to stick the label “elitist” on someone is to discredit half his argument without saying another word. Moreover, precisely because the term “elite” carries many of the connotations that Professor Walker and most others are inclined to read into it—no matter how much an author may try to sterilize the term by definition—I have generally avoided the term in writing about American politics. Like David Truman and V. O. Key, I have used terms rather more descriptive and discriminating, so it seems to me, such as political leadership, political strata, and the like. It is revealing, incidentally, that in the index to V. O. Key's Public Opinion and American Democracy the only reference to “elite” reads as follows: “elite: see political activists.” I suggest that this difference in the choice of words is more than a mere matter of taste or distaste for certain labels. It also reflects a conviction on the part of Key, Truman, myself and others that “elitist” interpretations of American political life are inadequate both empirically and normatively. The extent to which Professor Walker has misunderstood the orientation of the late V. O. Key, Jr., is best indicated by the work that Key was writing at the time of his death, and that Professor Walker could not have read, of course, when he wrote his article. “The perverse and unorthodox argument of this little book,” Key wrote, “is that voters are not fools. To be sure, many individual voters act in odd ways indeed; yet in the large the electrorate behaves about as rationally and responsibly as we should expect, given the clarity of the alternatives presented to it and the character of information available to it.” The Responsible Electorate (Cambridge: Harvard University Press, 1966), p. 7Google Scholar.

8 Why, by the way, not also: Campell, Converse, Miller, Stokes, Downs, Sartori, Almond, Verba, Kornhauser, Lasswell, Lane, Tingsten …? In short: among writers who have examined questions of leadership and participation, who is not eligible for Professor Walker's list?

9 Here are some of them, with italics added to emphasize the difference between what they say and what Professor Walker says they say: “The democratic citizen is expected to be well informed about political affairs…. By such standards the voter falls short. Even when he has the motivation, he finds it difficult to make decisions on the basis of full information when the subject is relatively simple and proximate; how can he do so when it is complex and remote? …” Voting (Chicago: University of Chicago Press, 1954), p. 308Google Scholar. “How could a mass democracy work if all the people were deeply involved in politics? Lack of interest by some people is not without its benefits, too…. Extreme interest goes with extreme partisanship and might culminate in rigid fanaticism that could destroy democratic processes if generalized throughout the community…. Some people are and should be highly interested in politics, but not everyone is or needs to be. Only the doctrinaire would depreciate the moderate indifference that facilitates compromise.”: ibid., pp. 314–315; “The classical political philosophers were right in the direction of their assessment of the virtues of the citizen. But they demanded those virtues in too extreme or doctrinal a form. The voter does have some principles, he does have information and rationality, he does have interest—but he does not have them in the extreme, elaborate, comprehensive, or detailed form in which they were uniformly recommended by political philosophers”: ibid., p. 322.

10 Like many other writers on politics, including Rousseau, Lipset's writings contain statements which, quoted out of context, might seem to offer faint confirmation for the interpretation offered by Professor Walker. But I do not see how, for example, Lipset's Chapter on “Elections: Who Votes and Who Doesn't?” in Political Man could be regarded by anyone who reads the whole chapter as yielding the sentence in Professor Walker's essay: see especially pp. 181, 186.

11 The third citation is to an article by Morris-Jones, which I do not have at hand as I write. If it turns out that he has correctly interpreted Morris-Jones, would that justify his interpretation of Berelson et al and of Lipset?

12 Incidentally, while we may have recently emphasized the conditions of democratic “stability” too much, and the conditions of democratic change too little, I doubt whether anyone who remembers the failure of “stable” democracies to emerge in the USSR, Italy, Germany, and Spain will ever find it in himself to scoff at writers who focus on the conditions of democratic stability. What such writers are likely to have in mind when they think of democratic “instability” is not cabinet changes nor even piddling differences in regime but the possibility of democratic failures eventuating in brutal dictatorships in comparison with which even the worst polyarchy will seem like the promised land.

13 In Politics, Economics, and Welfare, Lindblom and I wrote: “Polyarchy also requires a relatively high degree of political activity. That is, enough people must participate in the governmental process so that political leaders compete for the support of a large and more or less representative cross section of the population….

“Admittedly this is a rather imprecise formulation; in what follows we shall attempt to refine it a little. But one cannot be very precise…. In practice, moreover, even in one country the extent of political activity varies enormously from one policy-making situation to another, from complete apathy to widespread activity. Then, too, political “activity” is itself a difficult kind of behavior to measure. The number of variables is large, including the number of people involved, the intensity with which they pursue their goals, the type of activity they indulge in, the political position and location of those who are active, their status, degree of control over others, and so on….

“… In a very large number of important governmental decisions only a small minority of the electorate expresses or apparently even possesses any definite preferences at all among the alternatives in dispute. And it is equally safe to say that very little specific national policy is ever a product of an expressed preference for a specific alternative by an overwhelming majority of the electorate….

“… In practice, then, the democratic goal that governmental decisions should accord with the preferences of the greater number of adults in the society is extraordinarily difficult to approximate, and rarely, if ever, is it closely approximated….

“… This discrepancy between polyarchy and democracy arouses anxieties among those who wish to approximate democracy more closely, and rightly so. Keeping this fact in mind, let us suggest some general lines of approach to the question of the level of political activity required as a precondition for polyarchy.

“A considerable measure of political inactivity is not by itself a sign that the democratic goal is not being roughly approximated by a polyarchy….

“… The question, then, is not so much whether citizens are active but whether they have the opportunity to exert control through activity when they wish to do so….

“… Therefore the problem is not so much one of insuring that every citizen is politically active on every issue as it is one of insuring that all citizens have approximately equal opportunity to act, using ‘opportunity’ in a realistic rather than legalistic sense….

“… Equal opportunity to act is not, however, a product merely of legal rights. It is a product of a variety of factors that make for differences in understanding the key points in the political process, access to them, methods of exploiting this access, optimism and buoyancy about the prospect of success, and willingness to act. Some of these factors probably cannot be rationally influenced given the present state of knowledge and techniques. Three that to some extent can are income, wealth, and education. A fourth that may become important as knowledge increases is personality….

“… Nevertheless, many policy decisions cannot actually reflect any specific preferences of the greater number. About the most that can be said for polyarchy is that, if the opportunities for political action are kept open to a representative section of the adult population, specific policies will rarely violate highly ranked, intense, stable, and relatively broad preferences of the greater number for a longer period than about the interval between elections….” (pp. 309–314.)

14 Thus in Politics, Economics and Welfare, Lindblom and I offered “Seven Basic Ends for Social Action”: freedom, rationality, democracy, subjective equality, security, progress, and appropriate inclusion. Of democracy we wrote as follows:

“The democratic goal is twofold. It consists of a condition to be attained and a principle guiding the procedure for attaining it. The condition is political equality, which we define as follows: Control over governmental decisions is shared so that the preferences of no one citizen are weighted more heavily than the preferences of any other one citizen. The principle is majority rule, which we define as follows: Governmental decisions should be controlled by the greater number expressing their preferences in the ‘last say.’

“Democracy is a goal, not an achievement. The main sociopolitical process for approximating (although not achieving) democracy we shall call polyarchy. The characteristics of polyarchy, its prerequisites, and its significance as a device for rational social action on economic matters are discussed in a later chapter. If democracy is one of our goals and if polyarchy is a process for approximating that goal, it follows that we must also value polyarchy as a means. But here we are concerned with the democratic goal itself.” (p. 41)

15 (p. 71.) Though I have always tried to write lucidly, I am increasingly appalled by incontrovertible evidence of my inability to do so. If Professor Walker interprets as normative theory what I (and, as I believe, others) wholly or primarily intended to be empirical theory, others have reversed the process by interpreting my ventures in normative theorizing as if I were describing the American political system. Despite the clear warning contained in the sentence just cited, the fact that my model of “polyarchy” and my description of “The American Hybrid” are in the same book, though in separate chapters, is evidently enough to lead to their being confounded. Cf. Agger, Robert E., Goldrich, Daniel, and Swanson, Bert E., The Rulers and the Ruled (Wiley, New York, 1964), pp. 93 ffGoogle Scholar. Incidentally, my guess, supported by some data, is that if a number of “democracies” were measured by the standards of polyarchal performance described in the model, the United States would be found to rank well down the list. A highly innovative attempt to undertake such a ranking is Neubauer's, DeaneOn the Theory of Polyarchy: An Empirical Study of Democracy in Ten Countries (unpublished doctoral dissertation, Yale University, 1965.Google Scholar)

16 Many of the factors Professor Walker advances as possible explanations for varying degrees of political involvement-apathy, or political participation-nonparticipation, will be found in the pioneering article, published in 1954, by Lipset, , Lazarsfeld, , Barton, , and Linz, , “The Psychology of Voting: An Analysis of Political Behavior,” Handbook of Social Psychology (Cambridge, Mass., Addison-Wesley, 1954), Vol. II, pp. 1124–1175Google Scholar. A decade later, Angus Cambell presented a compact, suocinct, and (to me) potentially powerful explanatory theory that takes into account much of the work in the interval: “The Passive Citizen” in Rokkan, Stein, ed., Approaches to the Study of Political Participation (Bergen, The Chr. Michelsen Institute, 1962)Google Scholar.

17 Although reported crime rates are, for a variety of reasons, notoriously unreliable indices of actual crime, investigation conceivably might turn up some connections. However, as to the relation between suicide rates and voting rates, an examination made at my request by Edward R. Tufte of readily available evidence shows that in the United States the relationship, if any, runs counter to Professor Walker's conjecture. Taking each state as a unit, there is a positive and not a negative relationship between suicides and voting turnout. The correlations are 0.24 with turnout in presidential elections and 0.34 with turnout in off-year elections for governor and Senators. Mississippi has the lowest turnout and the second lowest suicide rate. Rhode Island, which has the lowest suicide rate, has a high voting turnout, while Wyoming, which has about the same turnout as Rhode Island, has the second highest suicide rate. California and New York had almost exactly the same turnout in the 1960 Presidential election, but the suicide rate in California is 16 per 100,000, putting it in the highest group, while New York at 9.7 per 100,000 is among the lowest. Professor Walker's hypothesis would imply that the suicide rate among Negroes is higher than among whites; in fact, for the United States the rate among whites (11.4) is more than twice that among Negroes (4.5). In fact the rate among non-whites is lower than among whites in all states except six with few non-whites. In eleven Southern states the correlation between suicide rates and voting in Presidential elections (which according to the conjecture should of course be negative) is positive and moderately high: 0.47; with off-term elections for governor and Senators, it is lower but still positive: 0.36. The correlation of suicide rates with voting in off-term elections for governor and Senators is practically identical in North and South; with voting in Presidential elections, the correlation in the North is almost non-existent but negative: 0.17. The data are from Dublin, Louis I., Suicide (New York, Ronald Press, 1963), pp. 218–219Google Scholar, and Jacob, Herbert and Vines, Kenneth (eds.), Politics in the American States (Boston: Little Brown, 1965), pp. 40, 46Google Scholar. My strong impression is that if the hypothesis were checked against comparative data, it would run into similar difficulties. For example, the Scandinavian countries have similar voting rates; yet while the suicide rate is high in Sweden and Denmark, it is low in Norway. Italy, which has astonishingly high turnout, has a low suicide rate, etc.

18 I have in mind particularly Almond, Gabriel and Verba, Sidney, The Civic Culture (Princeton: Princeton University Press, 1963)CrossRefGoogle Scholar; and Converse, Philip E. and Dupeux, Georges, “Politicization of the Electorate in France and the United States,” Public Opinion Quarterly, 26 (Spring, 1962), 11–13CrossRefGoogle Scholar; and Rokkan (ed.). Approaches to the Study of Political Participation, op. cit.

19 Heberle, Rudolf, Social Movements (New York, Appleton-Century-Crofts, 1951), p. 6Google Scholar.

20 Professor Walker may have been somewhat misled because he has looked for studies of “social movements” under the wrong headings. Standard texts on political parties and pressure groups have for decades contained descriptions of farmers' organizations, the labor movement, the NAACP, etc., under such headings as “pressure groups” or “interest groups.” They have also treated third parties, sometimes extensively. E.g., the third edition of Key's, V. O.Politics, Parties, and Pressure Groups (New York, Thomas Y. Crowell Co., 1952)Google Scholar, which had chapters in Part I, Pressure Groups, on “Agrarianism” that included a section on “Cycles of Agrarian Discontent: The Nature of Political Movements”; “Workers,” “Business,” and “Other Interest Groups,” including “Racial and Nationalist Minorities.” See also, Chapter 7, “The Party Battle, 1896–1952” and Ch. 11, “The Role of Minor Parties.”

21 The best known work is, of course, The New American Right, ed. by Bell, Daniel (New York: Criterion Books, 1955)Google Scholar, and the “expanded and updated” version The Radical Right (New York: Doubleday, 1963)Google Scholar. In 1963, The Journal of Social Issues devoted an entire issue to “American Political Extremism in the 1960's” (Vol. 19, April, 1963)Google Scholar. And see the results of a direct attempt to interview people at a San Francisco Regional School of Anti-Communism by Wolfinger, R. E., Wolfinger, B. K., Prewitt, K. and Rosenhack, S., “America's Radical Right: Politics and Ideology,” in Ideology and Discontent, Apter, David (ed.) (New York: The Free Press of Glencoe, 1964)Google Scholar.

22 Comparable, for example, to Kirchheimer's, Otto “Confining Conditions and Revolutionary Breakthroughs,” this Review, 59 (12, 1965), 964–974Google Scholar; or Lorwin's, Val R. “Labor Organizations and Politics in Belgium and France,” in National Labor Movements in the Postwar World, Kassalow, E. M. (ed.) (Evanston: Northwestern University Press, 1963), pp. 142–168Google Scholar; and his “Reflections on the History of the French and American Labor Movements,” Journal of Economic History (03, 1957), 24–244Google Scholar.

Submit a response

Comments

No Comments have been published for this article.