International Endowment for Democracy ���or���

Justifying Democracy

By Frank Cunningham
From Democratic Theory and Socialism. Cambridge University Press, 1987. Chapter 4, pp. 55-80.

That democracy needs to be justified indicates the low degree of it in today’s world. In some socialist circles, sad to say, use of the word except for the consumption of others generates skepticism about one’s socialist determination. This chapter addresses significant worries about democracy that socialists share with nonsocialists; later chapters address specifically socialist sources of concern.

  1. Why More Democracy?

    The principal reason to favour more democracy over less is that as social units become increasingly democratic, more people become increasingly free. If ‘freedom’ refers to the availability of options to do what one might prefer, the thesis can, of course, be challenged by rejecting this concept of freedom. This topic will be pursued in Chapters Seven and Eight, along with an examination of the objection that one ought only to favour maximizing the freedom of those in a certain group, such as the working class. Some fundamental questions of ethical theory are also set aside. One concerns the entirely self-interested person who only wishes personally to be free. Perhaps there are arguments proving that democratic progress makes any one person’s freedom secure, and that people who have the concern of others at heart are generally better off than those who do not. But such arguments will not be pursued here. 1 Nor will the ethical debate be joined about whether a prescription for freedom, or democracy, or any other goal may ever justifiably be overridden.

    Instead it is stipulated that such prescriptions at least carry a strong presumption in their favour. I recall a debate with a socialist friend who takes a far less critical attitude than I do toward the Soviet occupation of Afghanistan or the use of martial law against Poland’s Solidarity movement. ‘So,’ he said, ‘you favour that democracy be promoted in any circumstance, at any time, no matter what the surrounding conditions or the consequences?’ When I allowed that one might conceive of exceptional situations, this elicited the response that we agreed after all. In fact, we did not agree, but I realized that the dictatorship of the proletariat, so to speak, needs only a crack in the door to fill the whole house.

    To forestall this sort of thing, it is here maintained that there must be overwhelmingly conclusive reasons for limiting democracy. Defining ‘overwhelmingly conclusive’ would be difficult, though not impossible. It certainly rules out making an exception on the basis of just any reason. There are likely always some reasons that weigh in the direction of shelving democracy, and most defences of antidemocratic actions can be seen as seizing on excuses rather than seriously taking up the burden of proof. Also, defence of an overwhelmingly strong case would not give one carte blanche to limit democracy in any way, since there would still be a burden to show that a prescribed democracy-limiting measure is the least undemocratic one available.

    As to the argument favouring more democracy, it must at least be shown: a) that increasing freedom is a worthwhile goal and democracy is to be valued as a means to it; and b) that increasing democracy does in fact increase freedom. Claim a) will be resisted by those who think of democracy not as a means for maximizing individual freedom, but as a means for establishing a morally good society and who fear that people might not choose one. It will also be resisted by theorists who think democracy should be valued as an end in itself and not as a means to something else. Claim b) is challenged by those who doubt that a highly democratic society is attainable by making incremental advances. These challenges are examined before turning in ii and iii to classic critiques of democracy.


    In his book, The Justification of Democracy, William Nelson criticizes attempts to defend democratic procedures by appeal to freedom: ‘When we evaluate legislation or policy from a moral point of view, we must reject some laws or policies even when they reflect the tastes or desires of individuals. . . . Indeed, it could be argued that we must sometimes reject these laws just because they are responses to desires — the wrong desires.’ 2 Nelson’s alternative is to value democracy as a force for creating a society of people with good characters. 3

    The distance between Nelson’s defence of democracy and one appealing to freedom may not be very great; since a society of people with morally good characters would presumably be one where everyone was free to do as he or she wanted (though having good characters, they would only want to do morally acceptable things). A substantive difference between an approach focussing on the good society and one focussing on individual freedom concerns their different views about the extent to which democratic tolerance for possibly morally objectionable preferences can be sustained short of full democracy. It is hard to deny that in an imperfect world considerations of democracy and of morality will sometimes pull in opposite directions. The problem this poses may be conceived globally to question whether (imperfect) democracy is in general a suitable means for progressing toward perfect democracy. The problem can also be addressed on a case- by-case basis of conflict between the requirements of morality and of democracy. Thus regarded, the following considerations may allay Nelson’s concerns:

    1. It is consistent to favour maximizing freedom while also limiting it. This will be the case when it is necessary to limit the freedom of some to protect the freedom of others. Judgments about whose freedoms to curtail might be made on other than freedom-maximizing grounds, but they need not be. An alternative is to use the criterion of ‘progress toward complete freedom,’ (therefore toward perfect democracy).
    2. One might favour the freedom of people to do whatever they wish, while hoping that they will not wish to do certain things. Should people wish to do them anyway (and a ‘progress toward complete freedom’ criterion does not rule this out), then there is a conflict between a norm in favour of freedom and the norm on which such acts are negatively evaluated. Availability of a criterion for adjudication in such cases would allow a choice, but if none is available, one would be stymied. Neither situation is ideal, but then neither is entirely unencountered in a world where complex moral decisions must be made, and life goes on.
    3. Perhaps in some social unit the situation just described is not an exception but the rule. In this society freedom may not be something to value, and Nelson’s approach would be appropriate. But also appropriate would be that in such a social unit democracy is not to be valued. I suspect that were the human community sufficiently depraved (for instance, globally possessed of Nazi values), increasing democracy would cease to be a worthwhile collective activity. Though it might be questioned what would be worthwhile in such a world.


    These observations will be resisted by those who think that the sincere democrat must see democracy itself as an end. Thus, Seymour Martin Lipset: ‘Democracy is not only or even primarily a means through which different groups can attain their ends or seek the good of society; it is the good society itself in operation.’ 4 This view makes favouring democracy difficult unless one already accepts the value it is supposed to embody. A developed example may be found in Philip Green’s arguments for replacing the ‘pseudodemocracy, liberal capitalism’ with a genuine democracy conceived as political equality. In Retrieving Democracy he argues that ‘social equality’ (a form of material equality plus nonoppressive divisions of labour and equal access to the means of production) is necessary for ‘political equality,’ a state of affairs equivalent to ‘real democracy’ in which ‘everybody should count for one and nobody for more than one.’ 5

    Green describes a demand for political equality as ‘an ultimate desire of public opinion,’ by which he must mean that it should be ultimate, since it is unrealistic to think this a presently universal motivation. 6 This work questions whether the democrat must or should endorse such a prescription. We shall return to this topic in iv and again in Chapter Eight. Suffice it here to observe that in a highly democratic society, people recognize that protection of political equality is required, not as an ultimate end, but as a means for pursuing their various goals. Moreover, it is debatable whether a society in which democracy itself, as interpreted by Green or by anyone else, is desirable, since insisting that democracy is an end in itself limits people’s courses of action to ones consistent with whatever the goal ‘democracy’ is said to be.

    Brian Barry is too unkind when he charges those who view democracy as an end in itself with ‘giving aid and comfort to the politics of the beautiful people — the radical chic of the Boston— Washington corridor and the London—Oxbridge triangle.’ 7 Perhaps some democrats are accurately described by Barry, but others who value democracy for its own sake are more seriously motivated. An example is Roy Medvedev, who associates democracy definitionally with certain moral rights to avoid the sort of ‘class instrumentalism’ exhibited in the following judgment: ‘Democracy under capitalism is to be welcomed, even fought for, not as an end in itself, but in order to demonstrate the material basis, the real roots of political power.’ 8 Avoiding such class instrumentalism is an appropriate motive to think of democracy in noninstrumental terms, but this is neither necessary nor sufficient. It is not necessary because there are other ends which democracy serves and because good ‘democratic-instrumental’ reasons can be given to resist class instrumentalism. 9 It is not sufficient because class instrumentalists can simply reject as ‘bourgeois’ any defence of democracy making reference to values they think stand in the way of advancing working-class interests.

    The view that democracy is an end in itself is often combined with approaches that regard it as a matter of moral rights. Some fear that otherwise democracy will be sacrificed in the interests of expediency. Perhaps the association of rights-based ethical theories with anticonsequentialism is so strong that anyone friendly to democracy and hostile to utilitarianism feels it imperative to put democracy on a par with rights. But this mixes up two questions: how moral principles are to be justified (by reference only to consequences of acting on them or by something else) and how things other than moral principles are to be justified.’ 10 Another challenge concerns political equality. Green’s view discussed above reflects the opinion of many that democracy essentially involves not just participation but equal participation, and it might be thought that unless democracy encompasses the right to political equality as an end, this important value will be sacrificed in the name of democracy itself. This concern will be addressed in due course.


    A perfectly democratic society would probably be as freedom maximizing as any society could ever be. Granting, however, that perfect democracy is not attainable, it is central to the argument of this chapter that much closer approximations to it than exist anywhere today are realistic and desirable goals, and that these goals can be reached only by relentlessly working to increase democracy wherever and whenever possible. Both claims are contested. Critics who think of democracy as a form of mob rule, a tyranny of the majority, or a font of incompetence and inefficiency might agree that perfect democracy would not suffer from these things. But they may still claim that when the trade-offs typical of any imperfect democracy are made, less democracy will be preferable to more. These challenges will be the subject of subsequent sections of this chapter. However, there is a certain purchase on credibility to scepticism about imperfect democracy’s ability for improvement sometimes expressed on the political left.

    In a world short of perfect democracy, democratic advance will often involve deliberately limiting the control some people have over some aspects of their social environments. Even sanctions against stuffing ballot boxes is a limitation. It is therefore hypothetically possible that progress in democracy is best made by paternalistic denial of control to most people over most aspects of their social environments up to the golden day when they are kicked out of their protected nest into a garden of full democracy. This extreme position could be modified by allowing limited non- paternalistic measures provided they could be justified on a case- by-case basis. That is, the burden would now be placed on the antipaternalist.

    The alternate position is that progress toward full democracy requires maximizing individual freedoms except where limitations are: demonstrably necessary for progress in democracy; of short duration and minimum expanse; and imposed in such a way as to be reversible if they turn out to be a mistake. (For instance, laws protecting a minority, though clearly necessary for democratic progress, should be carefully laid down and capable of repeal.) More freedom for effective input to collective decisions, on this view, promotes democratic progress because restrictions on such freedom create a downward democratic spiral, while removal of restrictions creates an upward one. (Readers are invited to consult their own experiences and whatever lessons we can learn from history to test these claims.)

    The counterclaim that stepping backwards in democracy is required for making progress in the long run has the additional deficiency that the costs of being wrong are so great as to justify strong scepticism, despite the appeal of simplicity. Indeed, this very appeal should be cause for doubt. Any social unit will be an historically inherited blend of democratic and undemocratic habits and attitudes. It is understandable that somebody might think major progress can be made in one fell swoop, wiping the slate clean and starting again near the top. However, rather than making for progress, this ahistorical approach is apt to create a reaction against attempts to construct new democracy-promoting structures and to prompt retrenchment into antidemocratic traditions on the part of a populace which must view such Draconian efforts as contrary to their interests.

    In this work, then, perfect democracy is taken as something to be valued by somebody who, for whatever reasons, values the maximization of human freedom, and imperfect democracy is to be valued so long as one always tries to improve upon a secured level of it. Explication of the case that more democracy lays the basis for yet more will show that this is a realistic prescription, and it will afford an opportunity (in section iii of the chapter) to meet some traditional political arguments against democracy. Core concepts in this endeavour — the ‘democratic fix’ and the ‘anonymous majority’ are introduced in section ii by examining some abstract challenges to democracy.

  2. The Feasibility of Democracy

    Though most political philosophers from Plato’s time to the 18th Century were critics of democracy, they still recognized it as an option. It is only in our century that the very coherence of democracy has been called into question. The question is not about whether perfect democracy can ever be attained. In addition to obvious impediments (the complexity of overlapping and always changing social units, the fact that people’s visions often exceed their grasp, and the many sources of potential conflict in human affairs), it is possible that some people may not want to live in a perfectly democratic society. Considering this possibility creates a paradox. A society including such people could not be perfectly democratic; but excluding the possibility of such wants puts restrictions on the content of preferences. The conclusion to be drawn is not that therefore the idea of perfect democracy is incoherent, but that the possibility of a preference that could not exercise control in a perfect democracy is an additional reason to suppose that this level of democracy will never actually be reached.


    The question raised by Kenneth Arrow, among others, is whether any democratic mechanism is feasible. Arrow lays down some conditions that must be satisfied by any procedure for making a social choice which is both rational and democratic, and he shows that no procedure (or ‘rule of social choice’) simultaneously meets all of them. For example, the possibility of cyclical majorities means that majority vote would fail to meet a condition that a social choice rule will always select a single option (the ‘collective rationality’ condition).’ 11 Prodemocratic theorists approach this problem in three ways: by modifying Arrow’s conditions; 12 by indicating methods for circumventing the problems on a case-by-case basis when they arise in political practice (for instance by logrolling or arbitration in the case of cyclical majorities); 13 or simply by noting that usually workable rules are justified even if one can imagine circumstances when they would not work. 14 The latter, practice-oriented, reactions are on the right track, but they must be focussed in a certain way. .

    In criticizing the democratic component of liberal democracy, Andrew Levine notes that it will not do to show that in practice Arrow-type paradoxes seldom arise or can usually be gotten around, since it is alleged that democratic social choice rules suffer ‘incoherence’ and are thus ‘profoundly defective’ and ‘fatally flawed.’ 15 The point can be sharpened by asking what a procedure for making social choices must be to avoid such a flaw. One of Arrow’s own descriptions is typical in holding that a ‘social decision process’ or ‘constitutional rule’ should be capable of ‘selecting a preferred action out of every possible environment.’ 16 Arrow’s emphasis is on the word ‘every’ in such characterizations, and for those with a proclivity for orderliness it makes sense to say that a rule which fails in this respect is to that extent defective. The question is whether this defect is ‘fatal,’ but here intuitions differ. For the practically oriented democratic theorist to consider this defect fatal is to put unduly stringent demands on collective decisions. Indeed, such stringency would probably also invalidate decision rules commonly employed by individuals. Yet it seems strained to claim that someone who follows usually decisive procedures, making ad hoc adjustments when they fail, is thereby necessarily acting irrationally.

    An alternate way to make out a profound defectiveness case is to focus not on the word ‘every’ in Arrow’s phrase but on the word ‘preferred.’ Leaving aside the question of how individual preferences are ascertained, describing a person as having made a rational decision based in part on his or her preferences is not problematic from the point of view of the political theorist. But ascribing preferences and decisions to collections of people is problematic. Were the social choice theorist’s only task to evaluate procedures whereby social groups with their preferences attempt to make rational choices, Arrow’s problem could be met on a case-by-case basis by anyone for whom ad hocery is not anathema on general principle. However, the social choice theorist faces a more grave problem at the very first step, namely to identify group preferences themselves.

    In this version of the view about what would make a social decision process fatally flawed, phrases like, ‘the collective preference of group X’ are supposed to be defined by reference to some decision procedure that the members of X might follow, such as voting, drawing lots, or following a leader. When a putative decision-making rule of this kind is shown flawed (for instance, because it conflicts with other indispensable rules), then irrationality would be, as it were, within the very attempt of a group to formulate a preference, and hence the political theorist would not be able to identify group preferences at all, much less ascertain the rationality of actions taken to further them.

    The approach of this chapter sides with the practically-oriented defenders of democracy on the question of whether completeness is a necessary condition for rationality (of groups or of individuals), and it suggests a way to avoid the problem of identifying group preferences. The degrees-of-democracy theorist is not concerned to discover what, if anything, would constitute group rationality in the formulation of preferences. Rather, this theorist wants to know how progress toward perfect democracy may he made. ‘Perfect democracy’ is not defined by reference to collective preferences, but to individual preferences and to the availability of some means of control which people may jointly employ, where the list of such means is open ended.

    In a perfect democracy means must he available whereby negotiation can take place when consensus is not reached (or sought), but it is unnecessary to identify some means as ideal, much less as definitive of ‘negotiation.’ Activities like voting or seeking a consensus before acting are devices that may make progress toward perfect democracy or that might function as a means of negotiation in a perfectly democratic society, but they are not meant to he definitive of ‘democratic social choice.’ Thus conceptualized, many considerations advanced to solve Arrow-type problems can be valued as suggesting techniques to be employed when circumstances merit it, even if none of them constitutes an acceptable definition of ‘democratic decision.’ 17


    This approach can be applied to two more problems addressed by social choice theory: that of the free rider and of the autocompromiser. Democracy-promoting mechanisms and institutions are ‘public goods’ for those enjoying their benefits. But while it is of benefit to everyone that democratic institutions exist, no one will consider it necessary to put out any effort, even if this is only to vote, to secure it. One should expect, therefore, that nobody will make the effort with the result that nobody benefits from democracy. Reflection on this dilemma has generated quite a body of literature trying to explain why people do in Fact exert such effort.

    Autocompromising was touched on in the last chapter in discussing ‘control.’ Political theorists who study it note the way platforms of competing political parties often fail to match the first preferences of their mainstream adherents. Policy makers of each party fear that unless the party takes a stand close to the political centre, it will lose to a party that itself employs this strategy. The result, again, is a situation in which whoever wins the election, it will be on a platform that is nobody’s first choice. Rather than each party running on a platform reflecting its members’ primary values, uncertainty prompts advance compromise. 19 The problem is an instance of that studied by decision theorists as one version of ‘the prisoners dilemma,’ where choices must be made by conflicting parties in justified mutual mistrust and with severely limited options. 20

    If these sorts of ‘dilemmas’ are associated with the project of defining ‘democratic social choice,’ then similar considerations are pertinent. As part of the ongoing effort to maintain and expand democracy, however, they raise somewhat different difficulties. There is obviously something wrong with the first problem as a deduction about how people must behave, because many people do in fact exert the effort to participate in democratic procedures. While some of them may be irrational (lunatics with a deranged compulsion to fill out ballots), most do not seem either crazy or stupid, and to conclude that they must be because they are not free riders is objectionably antiempirical. Hence, considered at the most abstract level, this is a problem for theories of rationality, not for democratic politics.

    But viewed practically, the free-rider problem must be addressed by someone who wishes to make progress in democracy. 21 Low voter turnout and other forms of nonparticipation are well- known phenomena, no doubt partly to be explained by the possibility of being a free rider. The practical question for the democrat to ask is how free riding can be limited to the degree that n does not make democracy-enhancing practices impossible. One mechanism often suggested is to attach a cost to nonparticipation, for instance, fines for not voting or penalties for not attending meetings. This sort of approach is too pessimistic, since there is an alternative.


    In the 1960s some technophiles liked to talk of the ‘technological fix,’ the view that problems created by technology are best rectified by the employment of more technology. 22 We know the disastrous consequences of this view; one need not be a technophobic to realize that machines, whatever their virtues, are incapable of taking an interest in human well-being. But there are some reasons to suppose that an analogous theory of a ‘democratic fix’ is sound. The free-rider problem is created by the achievement of a certain level of democratic society, one that can provide everyone with benefits even if not everyone helps to maintain that level.

    By the democratic fix, solutions to this problem are found by increasing democracy. People who do not vote in elections are not just lazy. They are also reacting to the fact that few have effective input to the selection of candidates and that there is sparse correlation between what candidates promise and what they do. People who have an opportunity to participate in the governing of their affairs in a more direct way, for instance, in voluntary neighbourhood or issue-related organizations, but refrain, are influenced by the fact that these activities take up much time with relatively meagre results that are often later reversed. On the argument of the democratic fix, increased popular control of the conditions that impede participation is required to rectify this situation. 23

    It is noteworthy that the most extreme example of autocompromising, the prisoner’s dilemma, is an abstract model in which constraints are so severe that the democratic fix cannot be applied. The parties in question have no opportunity either to try reaching consensus or to negotiate with one another. Being players in an abstract model, the prisoners do not live in a real world where there would be more room for movement (for instance, carried on by the prisoners’ lawyers), nor is there a world outside the prison and the courts subject to democratizing measures which (the democrat argues) can in the long term increasingly minimize crime and hence the need for prisons and their constraints. This is not to deny that the practical problem of autocompromising is a real one, as the problem of political party platforms illustrates, but in this real-life example there are ways of minimizing constraints.

    The most frequently encountered response is to take antidemocratic measures. Faced with pressures to move away from stating preferred political policies, parties compromise in platforms, but do whatever they want when elected. The contrasting democratic prescription is to increase democracy. Political theorists have noted that autocompromising ought to lead to a two party-system. It is accordingly instructive to see what sorts of counteracting pressures have prevented this, as they have in most of the world where there are parliamentary systems with competing political parties.

    One hypothesis is that convergence is less likely when parties have an active, popular base of support with effective input to their decision-making structures. Examples are some labour, socialist, and communist parties with trade union support, and the Green Party in the Federal Republic of Germany, which has active support from ecological, peace, and feminist movements. To the extent that this hypothesis is sound it illustrates the democratic fix. Dependence on support by people actively pursuing goals outside of a party structure provides an incentive for it to advance policies in accord with these goals. Internal political party democracy also helps to keep leaders honest. Perhaps one can imagine circumstances within which the democratic fix is unlikely to work, and there are problems other than those of the free rider and autocompromising. But once these problems are considered practical ones, instead of challenges to the coherence of notions like ‘democratic choice,’ the impossibility of the democratic fix cannot be established in advance of concrete inquiry.

  3. Constitutions, Majorities, Minorities

    Dennis Mueller’s book, Public Choice,24 explores the question of how democracy can get started and by whom. Mueller distinguishes between methods for making decisions over the allocation of goods within a political structure and methods whereby ‘constitutional’ decisions about the principles of such a structure are made. His main concern is with constitutional decisions, where he sees problems with both unanimity and majority vote.

    Requiring unanimity can prevent a constitution from ever being agreed to or allow a small minority to blackmail the majority with its veto power. Majority decision can leave a minority feeling itself shut out or unbound by a constitution. Mueller suggests that social choice theorists are therefore attracted to social contract theory, where the most basic normative decisions are thought somehow to have been made in advance of ongoing political society. But a serious problem for contractarians concerns the locus of ‘primary citizenship’ or the question of who makes the contract: ‘If primary citizenship is vested in the inclusive polity, citizens in unfavourably endowed local communities are able to tax those living in other local communities. How then is the writing of the social contract envisaged, as a pact among all members of the inclusive polity, which sets aside certain rights for the local polities? Or do the citizens first form a pact for the local polity, and then delegate authority over some issues to the larger one?’ 25 Mueller’s challenge offers an opportunity to sharpen a degrees-of-democracy approach.


    The question of an ‘Ur-Constitution’ does not concern a degrees- of-democracy approach (though the question of deciding how inclusive an actual constitution should be and whether people should have a right to opt out is of concern). On the degrees approach, democracy never has to be gotten off the ground because it is never on it. There is always some measure of democracy, and the problem is how to increase it. People always have found themselves in social units, some partially shaped by formal constitutions, most not, but all of them more or less democratic. When it is important to construct or to alter a constitution (that is, collectively and deliberately to set out principles in accord with which further actions affecting a shared social environment will be taken), this is done against the background of these overlapping social units and of the degrees of democracy each possesses. Democracy enhancement is of the lifting by one’s boot straps variety of processes rather than of the fundamental construction variety. 26

    Nevertheless, Mueller’s considerations throw into relief some problems about the relation among the three indices of ‘more democracy’ outlined in the last chapter (number of people having control, number of aspects over which control is exerted, and the importance of aspects over which control may be had). In one respect the ‘proportion of people’ measure has no claim to being more important than the others, and a democrat may recognize that in some imaginable circumstances progress would be made by disallowing wide public input to a range of decisions, whether unanimity, majority vote, or something in between, for a considerably long period of time. This logical possibility keeps the imposition of emergency legislation or the imposition of revolutionary dictatorship on an unwilling majority from being anti- democratic by definition. However, we know, do we not, that these things are in fact antidemocratic, that is, that they contribute to regress in democracy? Not only is control limited in the short term, but well-known social, political, organizational, and psychological processes are set in motion that inhibit democracy and are hard to reverse. I have no doubts that were one ever written, a world history of democracy27 would justify thinking of it as a spiral, to employ the image of Bowles and Gintis, 28 such that motion in either an upward or a downward direction creates an impetus to keep moving in that direction. Thus viewed, the ‘proportion of people’ measure takes on special significance. To do something against the will of the majority is to set in motion dynamics that will lead to less and less democracy, even if it could lead to more democracy were one able to think these dynamics out of existence.

    Most constitutions limit a majority’s ability to force an election at any time on any issue, and the leaders of many postrevolutionary or nationally liberated states often refrain from holding elections immediately after attaining statehood. Such measures can be sorted according to whether there is popular support, measured in a variety of ways, for them. 29 In these cases intuition gives unanimity and/or majority agreement special place among ways to ascertain popular support. It is thought that were there unanimity, this would be an ideal situation, and if a majority agrees, then taking into account only the ‘proportion of popular control’ measure, the result would be at least more democratic than one agreed to by less than a majority.


    The problem addressed by Mueller that unanimity might become de facto rule by minority veto can be ameliorated by seeking criteria to identify situations where on balance more democracy is promoted than retarded when a minority has this power. But there are bound to be limitations on how well such criteria could function in an ever-changing world. There is also the question of how a choice of the criteria themselves is to be made: by unanimous consent or otherwise. However, these problems will, again, be less acute for the degrees-of-democracy advocate than for others.

    Unanimity itself admits of degree, one limit of which is veto by a minority of one, the other, consensus. Theorists of participatory democracy have done a good job of describing the difference between decision-making by consensus and by what might be called ‘voting on fixed interests.’ Benjamin Barber describes the strongest form of consensus as: ‘an agreement that arises out of common talk, common decision, and common work but that is premised on citizens’ active and perennial participation in the transformation of conflict through the creation of common consciousness and political judgment.’ 30 This form of consensus is not just acquiescence to a stubborn majority; it requires coming to agreement. Nor is it a case of people happening to agree already. Barber calls this weak form of consensus ‘substantive’ as opposed to ‘creative’ consensus. This is an apt term, signalling as it does that consensus is a process during which people’s preferences change as a result of engaging in discussion and debate with others engaged in shared projects. (In Japanese the phrase sodan kai refers to a group that strives to reach consensus by talking together.) Habermas describes the same phenomenon in his description of democracy as the ‘participation of citizens in discursive processes of will-formation.’ 31

    When there are good reasons both to seek more than majority rule and also to avoid a minority veto, the degrees-of-democracy approach will prescribe striving for consensus in this strong sense. One standard objection is that people can be expected to enter social decision-making forums and to exit from them with the same, probably selfish, preferences. This mean-spirited view of human nature has been more than adequately treated by democratic theorists and will be touched on in later chapters. A much different objection is that anything other than consensus, and in particular majority vote, fails to be democratic at all. This is Barber’s own view: ‘Majoritarianism is a tribute to the failure of democracy: to our inability to create a politics of mutualism that can overcome private interests. It is thus finally the democracy of desperation, an attempt to salvage decision-making from the anarchy of adversary politics.’ 32 This criticism of majority rule supplements that of Mueller, but both can be met.


    Writing from a degrees-of-democracy perspective, Felix Oppenheim persuasively argues that in order to maintain an acceptable level of democracy, the ‘majority principle’ itself requires constraints on what any one majority may do. 33 The point applies especially to the protection of minority rights. If majority rule is to be ongoing and secure it must avoid a situation where a minority ceases to feel bound by majority will or where restrictions on minority rights trigger a downward spiral that finally undermines the power of the majority itself. Oppenheim also notes that minority-protecting rights serve the interests of majority rule by protecting the minority’s ability to change the minds of the majority. This justifies a large number of traditional civil liberties and some equality rights. Members of a minority subject to discrimination, threatened by arrest without recourse, lacking information and political skills, or hamstrung by poverty will hardly be in a position to try educating members of the majority to their opinions; yet it is surely not in the interests of ongoing democracy that a majority should do things to keep itself ill-informed. 34

    Implicated in these considerations is that the notions ‘minority’ and ‘majority’ are more complex than usage sometimes suggests. Each is a group concept, but groups per se do not, strictly speaking, vote or otherwise make decisions. Hence, to talk of a minority changing the mind of a majority is misleading; rather it is the actual contours of the majority that change. The point is made by Carl Cohen who discusses the ‘fluctuating majority,’ 35 and by public choice theorists who insist that, in a democratic decision, who votes for what on one issue should not determine the outcome of a subsequent vote. Kenneth May called this open-ended majority ‘anonymous.’ 36

    It is too strict to make anonymity necessary for any democracy at all. Even fixed majority rule will be more democratic, other things being equal, than minority rule. But it will be less secure, less likely to be ongoing majority rule and less likely to provide fertile ground for developing other methods, like seeking consensus. Thus, the democrat will strive to protect the anonymity of the majority. The implications of this for pluralism and class reductionism will be discussed in due course. The point here is that the democrat need not be stuck with an alternative between a simply interpreted unanimity procedure and an equally simply interpreted majority rule. Each, like democracy itself, is a complex matter of degree.

  4. The Practicality of Democracy

    So far the strategy of this defence of democracy has been to displace abstract problems to the terrain of real political practice. Now, however, we must confront some practical reservations which many have expressed about democracy: that it constitutes a tyranny of the majority; that it promotes rule of the incompetent; and that it is inefficient.


    Elaine Spitz, a strong defender of majority rule, argues that the force of this criticism can be dampened if majority rule is seen as a ‘social practice,’ embodying certain values and modes of comportment, but she grants there is something to the objection. 37 In terms of a degrees-of-democracy approach, a situation where a majority oppresses a minority may well be more democratic than one in which a minority has the ability to oppress the majority, but at the same time minority rule might also be justified under certain conditions. Neither alternative can sit well with the democrat. However, once the notion of democracy as ongoing is introduced, the intuition can be captured that there is something out of keeping about both a majority tyrannizing a minority and minority rule. Promoting minority rule will require dismantling political mechanisms difficult to rebuild, thus leading to far more antidemocratic consequences. Similarly, many democratic theorists concur with the view explicated above that unconstrained majority rule is autodestructive. 38

    Here it is appropriate to return to the question of political equality raised in section i, since this principle also finds a source of justification in the anonymous majority. The point is almost made by Green: ‘Since the rule of political equality is everybody to count for one, nobody for more than one, political equality cannot exist when there are permanent or long-term minorities in the polity. . . . A minority must be nothing more than a random collection of people who lost the last vote; if that proviso does ‘lot generally hold true, then we have not political equality but majority tyranny.’ 39 This argument may he inverted to maintain that equality of participation is a requirement for prevention of majority tyranny. Democracy might he best served if those most strongly affected by the outcome of a decision have greater input to the decision than those little affected (hence the rule ‘one-person-one-vote’ ought not to be advanced as inflexibly universal). But democratic progress is not promoted if certain people always have disproportionately advantageous input.

    Clearly, the more such situations there are, the greater the possibility of a tyranny of a minority with its attendant consequences of creating cynicism and political apathy in the majority. At the same time systematic disproportionate advantage also threatens majority anonymity. Though it is misleading to say with Green that a minority or a majority on any one issue should be ‘a random collection’ (there must, alter all, have been reasons for them to have voted the same way), full anonymity would be impeded if there were a tendency for some always to he members of a majority. Those with disproportionate input will be just such people, since others will see an advantage to getting them on their side, not because of the content of the majority’s opinions, but just because they are thus privileged. 40


    Anonymous majority rule is what Green calls ‘the true idea’ of majority rule: not that everyone has equal influence, hut that no one has any influence.’ 41 The description is appropriate, hut it suggests a problem addressed by some theorists about the stability of democracy. Claude Lefort and François Furet argue that democracy requires sovereignty to be empty ‘The legitimation of [democratic] power is based on the people; hut to the image of popular sovereignty is attached that of an empty place, impossible to occupy, such that those who exercise public authority cannot claim to appropriate it. Democracy joins these two apparently contradictory principles: one that power comes from the people, the other that it is die power of nobody.’ 42

    Neither Lefort nor Furet link their idea of the empty space of democracy with the anonymous majority, and perhaps they can be criticized for supposing a theory of fixed interests. However, their concept of popular sovereignty sits more comfortably with the notion of the anonymous majority than with a doctrine of fixed interests. If the majority becomes so fixed that decisions are always made by the same configuration of people, they become the occupants of the space of sovereignty. Lefort’s concern is rather that the space of democracy can be ‘occupied’ by a minority acting in the name of the people.

    Furet thinks this happened after the French Revolution when the Jacobins assumed this role, and Lefort sees the Bolsheviks as having done the same thing after the Russian Revolution. These are intriguing hypotheses, and one can find illustrations closer to home in parliamentary governments when representatives who are barely responsive to a population authoritatively pronounce on some issue in the name of ‘the people.’ Thus, the problem is not that the majority becomes tyrannical but that to prevent its so becoming, the door is opened to unresponsive minority rule. One need not conclude that therefore all is lost, but attention is drawn to the fact that, like almost anything else worth pursuing, progress in democracy involves a certain risk. An important way to prevent the space of democracy from being misused is to ensure that a populace is aware of the importance of keeping it open. This applies both to a majority, who should have a conception of the importance of protecting minority rights at the same time that it opposes a specific minority, and to a minority, who should value abiding by the will of the majority, even while disagreeing with it. 43


    Democrats are no doubt right that those who question the people’s moral or intellectual abilities to govern their own affairs are usually elitists and snobs. But one must still confront the substance of this charge. The main counter argument must be empirical. It must be denied that people are incapable of self-government. Evidence for this is not easily found outside of participation with people who are not professional politicians in efforts at self- determination. My own experience (in left-wing politics, the peace and civil rights movements, and in my own work place and professional associations) has informed me, to be sure, of the problems to be overcome when people try collectively to direct some portions of their lives in a political, cultural, and economic environment often hostile to this task. I think it also cured me of any populist romanticism I may have had and shows that some people seem to have a special aptitude for this kind of activity.

    But it has also revealed that people concerned with issues close to them and given half a chance do not bifurcate into the few who are smart and capable and the many who can never hope to be more than good followers. People can and do rise to the occasion. This is one thing that makes voluntary organizations, for example, women’s or peace movements, tenants and environmental protection organizations, or trade unions, as effective as they sometimes are against powerful adversaries, who typically underestimate people’s ability to get their acts together. Further, where there are problems of apathy or insecurity among ‘rank and file’ members of an association, my experience has been that this is rather an effect than a cause of elitist domination of leadership by a few.

    Participation. Observations such as mine are no doubt of a highly subjective nature and not likely to convince a hostile critic of democracy. (Although it should be noted that the view often supporting such scepticism that knowledge about things social and political requires detachment and nonparticipation can be challenged.) 44 But additional, burden-shifting arguments are available to the democrat. One of these appeals to the possibility of education. Moses Finley concludes his book on democracy in ancient Greece: ‘The conviction of Socrates is not the whole story of freedom in Athens. . . . I have tried to argue that [exclusive concern to erect protections against the public] is a way of preserving liberty by castrating it, that there is more hope in a return to the classical concept of governance as a continued effort in mass education. There will still be mistakes, tragedies, trials for impiety, but there may also be a return from widespread alienation to a genuine sense of community.’ 45 Like most other democratic theorists Finley has in mind not just formal education but the education that comes through democratic participation itself.

    Proving that participation educates people for self-government demands much argumentation, but some features of a participationist argument might be noted, all supposing versions of the ‘democratic fix.’ Required for effective, ongoing and increasing participation in collective self-determination by members of a social unit is: 1) that they have the desire to participate, that is, that they are not apathetic; 2) that they have appropriate values, for example, simultaneous respect for majority rule and minority rights; and 3) that they have the skills required for self-governance, including the ability to use good judgment in negotiation, debate, coordination of self-governing efforts in overlapping social environments, and in general practical knowledge about human nature and political processes. 46

    The democrat believes that activity that increases shared control over a social environment (democratic participation), once engaged in, is found effective at solving long-run problems and yields short-term rewards, as bringing people together in common action helps to break down alienation, isolation, and a sense of impotence. Thus, unless somehow counteracted, a desire for increased democratic participation should grow as its benefits are learned. The process needs to be started, but as noted, it always does exist in some measure. Extending it requires people’s becoming involved in such things as community associations, home and school organizations, unions or caucuses in unions, national or cultural organizations, or issue-related movements. Involvement can be more or less democratic, and education in democracy requires also that the involvement be deliberately and critically democratic. A main task of democratic theory and social research is to explain why and how such things take place or fail to take place where and when they do.

    The participationist notes that the more local and modest the democratic enterprise the less difficult it is for people to engage in it and the more direct the incentive to do so. Hence there is the possibility of entering a process. Potential for increasingly acquiring the values and skills required for more democracy is found within the process. One learns how to negotiate by doing it. Success in changing a majority’s mind educates one to the role of protecting minority rights even when in the majority. The values of respect and tolerance are initially learned by people seeking the same goals but differing on means, and it can then be extended to respect and tolerance for people sharing some goals but disagreeing on others. Democracy-inhibiting prejudices (racial, sexist, or national chauvinist) thrive on mutual isolation, but begin to he called into question in the interaction that participation in joint projects Facilitates. To challenge the participationist it will not do simply to adduce examples where participation has not had the desired effects, hut further to show that there were no counteracting tendencies (fin example, opposition so strong that shared goals were thwarted and morale broken), and that impediments to the development of relevant skills and attitudes could not have been removed by more democracy (for instance, in the internal organization of a collective itself).

    Philosopher Kings. Another defence against the ‘people are incompetent’ challenge is to ask who is competent to control human affairs if not all the humans whose affairs they are. The question- begging response, ‘those who happen to have the incentive and skills necessary to rule,’ does not suffice. In the first place, there is the question of what they are to do as rulers. Controlling human affairs means doing things that will affect some society, and undemocratically responsive rulers will have to decide how to use their supposed skills and knowledge. The main alternatives seem to he: to promote satisfaction of a ruler’s own preferences no matter what effects tins has on others; to promote satisfaction of needs shared by every member of the community (to stay alive, for example); or to try maximizing preference satisfaction across the society.

    Each alternative is problematic. The first is overt tyranny. The second two alternatives are forms of paternalism and raise the question of how the ruler knows what universal needs or individual preferences are. If the ruler relics on people’s expressions of them, this approaches the democracy that was to be avoided. If the ruler has some other access to knowledge of these things, then one wants to know how this knowledge is obtained and who has it. This suggests another weakness in the elitist’s argument. Walzer addresses it in discussing Plato’s claim that philosophers should decide what is best for people. Lie comes down against the philosophers: ‘The interventions of philosophers should be limited to the gifts they bring. Else they are like Greeks bringing gifts, of whom the people should beware, for what they have in mind is the capture of the city.’ 47 The point is surely apt. Who is to judge whether decisions of supposed leaders are in people’s real interests? If the people themselves are the judges, tins elitist position again turns into that of the democrat. If the rulers are to judge, there is no way to check against bad and self-serving decisions, thus leading to tyranny.

    Supposing that either of these arguments against the elitist can be sustained, there remain two alternatives: democracy and ‘traditionalist conservatism.’ The latter view is that those people should rule who find themselves in the traditional roles of rulers, whether they are competent or not.48 Among the problems with this profoundly pessimistic viewpoint is that even if one agreed with its premises, it is too late to act on in countries which have achieved a degree of democracy that includes an ethos hostile to such blind obedience. Unless the traditionalists are to take the status quo as the ‘tradition’ to follow (in which case this would often involve more democracy than they want), they will have to be ‘reactionaries,’ advocating return to some past tradition. It is hard to see how a criterion telling one which tradition should be reverted to and consistent with traditionalism could he found.


    While pro- and antidemocrats take opposing views about whether democracy is more or less efficient, 49 it is not always made clear what democracy is supposed to be more or less efficient at. Presumably there are tasks requiring coordinated activity to be performed. More or less democracy may enter at one or both of two stages: deciding to confront a task and agreeing on means to undertake it. Relatively undemocratic situations regarding both ends and means are ones forced on people by circumstances like a natural disaster, pursued in accord with blind tradition, or dictated by a few to suit their purposes (as when a government, responding to the arms industry, determines that defence’ will be promoted by massive arms spending). Most who claim democracy to he inefficient avoid the topic of the selection of goals, as well they might.

    To argue that goal selection should be undemocratic requires maintaining that all such selection is forced or blindly traditional or else that the few who know what goals people ought to have make these decisions. Surely the first alternatives are objectionably fatalistic, and the last is a version, again, of Platonic elitism. Perhaps it is in recognition of these weaknesses that most critics of democracy concern themselves with its putative inefficiency as a means for realizing goals. The arguments are familiar. Democracy is said not to allow sufficient coordination or to be too slow and clumsy in emergencies. Or it is urged that while people may be the best judges of their ends, special expertise is required in the choice of means.

    Evaluation of the critic’s claims requires comparison of success in solving problems of joint concern carried out more or less democratically. The democrat argues that despite regular denunciations of democracy as inefficient (almost always levelled by people whose own power would be curtailed by more democracy) such comparisons tell another tale. Of course, the comparisons must be thorough ones. In a thin slice of space and time Hitler made trains run on schedule, but not many were running when he surveyed a country in rubble from his bunker a few years later. This is not the place to carry out such evaluation, but some general considerations on the democratic side can be noted. Lipset appropriately points out that when special expertise is required, a democratically ‘legitimated’ society is better able to produce and call upon qualified experts, because people are more likely to be recognized and promoted on the basis of real achievement than on superficial characteristics such as skin colour. 50 And Habermas criticizes as an unjustified, technocratic assumption that planning and citizen participation must be in contradiction. 51

    There are also the advantages often and justly referred to by democratic theorists of democratic debate and the need for accountability to advance knowledge and develop expertise. Moreover, democracy allows utilizing the practical ‘expertise’ of people locally affected by decisions. 52 The relatively rare city planners who consult residents of neighhourhoods, school superintendents who consult students and teachers, hospital administrators who consult nurses and patients, or factory managers who consult workers almost always make better decisions than the many who do not.

    ‘Emergency’ situations said to require the shelving of democracy in the interests of expediency are often suspect.53 Sometimes the emergency which requires autocracy turns out to be no more than democratic threat to autocrats themselves. Genuine emergencies, for instance, ones caused by natural disasters, economic collapse, or military invasion, are almost always better met democratically than autocratically. They demand sacrifices which people who have bad input to a decision wilt he more committed to making than if someone else has declared that they must endure them, and there is less danger that temporary checks become permanent ones. Also, means and ends have a way of turning into each other. Democratically secured commitment to seek a common goal can atrophy when there is little participation in choosing means, and relative lack of democracy in the choice of means can change a social environment in such a way as to create new problems.

    These are not meant to be more than outlines of defences for full democracy. Once they are developed the importance of seeking always to advance democracy is established, so the democrat maintains. On this premise, then, one wants to know about social units as large as entire countries what political/economic structures will best promote democracy. Chapter Six will make a case for socialism in this respect. First some key and controversial problems of definition need to be addressed.

  1. See David Gauthier’s attempt, Morals by Agreement (Oxford: Oxford University Press, 1985). In addition to the question of whether Gauthier’s approach can ground morality as he envisages it, there is an additional question of whether it can ground the egalitarian moral prescriptions of the present work’s Chapter 6. Gauthier, himself, is not concerned to endorse such prescriptions.

  2. William Nelson, On Justifying Democracy (London: Routledge and Kegan Paul, 1980), pp. 90—1.

  3. An adequate morality for Nelson is a set of principles that ‘produce benefits and prevent harm,’ that can be the fundamental charter of a well-ordered human association,’ and that can function ‘in a society of free and independent persons,’ ibid. 109— 10.

  4. Seymour Martin Lipset, Political Man: The Social Bases of Politics (Garden City: Doubleday, 1960) 403. Quoted and discussed by Felix Oppenheim, ‘Democracy — Characteristics Included and Excluded,’ The Monist 55(1): 29—59(Jan. 1971), at p. 30. (This issue of The Monist has several contributions on democratic theory.) See, too, Robert Paul Wolff, The Poverty of Liberalism (Boston: Beacon, 1968) 181, 193—4.
    It is interesting to note that in his expanded edition of Political Man, 2nd ed. (London: Heinemann, 1983), Lipset, a defender of the ‘end of ideology’ thesis as a welcomed development of modern times, criticizes the New Left of the ‘60s and ‘70s for attempting to revive antipluralistic ideological viewpoints, 524—544. My clear recollection is that the main ‘ideological’ orientation shared by streams of the New Left was that democracy is to be valued as an end in itself.

  5. Philip Green, Retrieving Democracy: In Search of Civic Equality (Totowa, NJ: Rowman and Allenheld, 1985), see pts. 1 and 4 for an overview of the argument. ‘Political democracy’ is given this short definition at 5, and is elaborated on in chap. 8. Green has many good criticisms of antidemocrats. In my view his work would be helped by abandoning his fragmentist distinction between pseudodemocracy and real democracy. Note 40 below suggests one way this approach hinders his analysis. Another problem is that it is hard to imagine what ‘democracy’ is supposed to be ‘retrieved.’

  6. Ibid. 242. Green cannot mean that this is a main motivating desire of people consistently with his lament that thanks to living in a pseudodemocracy they think of their social being more in terms of ‘competition and withdrawal into the life of privatized consumption’ rather than of ‘cooperation and participation,’ 5.

  7. Brian Barry, in Stanley Benn, et al., Political Participation (Canberra: Australian National University Press, 1978) 47. The target of Barry’s criticism is the contribution of Benn in this collection. Barry contests Benn’s contention that it is impossible due to the free-rider problem for any rational person to engage in democratic politics for any reason other than for its own sake.

  8. Roy A. Medvedev, On Socialist Democracy, ed. and trans. Ellen de Kadt (New York: W.W. Norton, 1977) 39. The instrumentalist view is expressed by John Hoffman, Marxism, Revolution, and Democracy (Amsterdam: Gruner, 1983) 87. Hoffman and I corresponded about a draft of this book, and I am cited for support in it; I am also acknowledged, in a later book by Hoffman on Gramsci, with thanks for urging him to read the Eurocommunist writers. Though John Hoffman is kind to refer to me in his books, I should note that my evaluations of Gramsci and ‘Eurocommunism’ differ from his, and the main point I made to him was that Stalinism should be explicitly addressed.

  9. For example, Medvedev also argues that democracy is necessary to cultivate genuinely capable political leaders and to allow them to be recognized, while the ‘uncompromising sectarianism’ typically associated with class instrumentalism inhibits this, Socialist Democracy 311.

  10. See the argument that respect for rights is consistent with a consequentialist ethics in Wayne Sumner, The Moral Foundation of Rights (Oxford: Oxford University Press, 1987). Here is a representative account: ‘Instrumental justifications value democracy for the way in which it translates people’s wishes into government policy. . . . Intrinsic justifications value democracy for the benefits that the act of policy-making confers directly on the participants: democracy is desirable either because political participation is an intrinsically worthwhile activity or because it develops qualities of character. . . that are valuable in other areas of life,’ David Miller, ‘The Competitive Model of Democracy,’ in Democratic Theory and Practice, ed. Graeme Duncan (Cambridge: Cambridge University Press, 1983) 133—155, at p. 151. Miller does not explain how an activity could be ‘intrinsically worthwhile,’ but supposing democratic participation to be such a thing, the prescription for more society-wide democracy could be seen as a prescription to make it possible for everyone to enjoy the direct benefits of participation. Democratic activities would then be both a means and an end.

  11. Kenneth J. Arrow, Social Choice and Individual Values (New York: John Wiley, 1951), chap. 3. Arrow summarizes his argument in ‘Public and Private Values,’ in Human Values and Economic Policy, ed. Sidney Hook (New York: John Wiley, 1951) 3—2 1. The four conditions are: 1) that a collective choice rule will always determine a unique ordering of preferences for the collectivity in question (‘collective rationality’); 2) that if everyone ranks some possible outcome higher than an alternative the rule will rank it higher as well (the ‘weak Pareto principle’); 3) that a collective choice depends only on the individual preference rankings of those in the collective (‘independence of irrelevant alternatives’); and 4) that no individual’s choice is automatically the collective choice (‘nondictatorship’).

  12. See, Amartya Sen, Collective Choice and Social Welfare (San Francisco: Holden Day, 1970) 44ff., and Duncan Black, The Theory of Committees and Elections (Cambridge: Cambridge University Press, 1971) 19.

  13. An example is Michael Davis, ‘Avoiding the Voter’s Paradox Democratically,’ Theory and Decision 5(3): 295—3l1 (October 1974). Davis points out that Robert’s Rules of Order contains straightforward provisions for dealing with cyclical majorities.

  14. This point is ably argued by N.M.L. Nathan, ‘On the Justification of Democracy’ The Monist 55(1): 89—120 (Jan. 1971).

  15. Andrew Levine, Liberal Democracy: A Critique of Its Theory (New York: Columbia University Press, 1981) 61. Levine does not suggest how the nonliberal-democratic choice mechanisms he calls for might escape this criticism.

  16. Arrow, in Hook ed. Human Values 13. Arrow defines a ‘constitution’ as a rule for making social decisions which have this capability.

  17. In his survey of ‘processes for revealing social preferences,’ Dennis Mueller discusses the following: a unanimity rule, the optimal majority, simple majority rule, cycling, logrolling, plurality rule voting, Condorcet voting, use of the Borda count, exhaustive voting, and approval voting. Public Choice (Cambridge: Cambridge University Press, 1979), chap. 3. Mueller’s discussions include methods to encourage individuals to reveal their actual preferences, a problem not discussed in the present work. See his Chapter 4.

  18. A selection of positions on this dilemma may be found in Stanley Benn, et al., Political Participation, cited above.

  19. Anthony Downs discusses what is here called autocompromizing among other places in An Economic Theory of Democracy (New York: Harper and Bros., 1957), chap. 4.

  20. In the ‘prisoner’s dilemma,’ two prisoners are separated from one another and offered the following conditional options (each knowing the other is offered them): if neither confesses to a crime they are charged with having committed together, each will receive the same light sentence; if one confesses and the other does not, the one who confesses will get off and the one who does not will receive a heavy sentence; if both confess, each will receive the same medium heavy sentence. Since neither can trust the other to refrain from confessing, they both confess with the result that they are rationally led to a situation known by each to be worse than the available alternative where each would receive a light sentence.

  21. There is a division among socialist theorists over the extent that the free-rider problem ought to be of such concern. Some, like Allen Buchanan and Jon Elster, not only take it seriously, but place the problem at the centre of their analyses of political consciousness. See Buchanan, Marx and Justice: The Radical Critique of Liberalism (Totowa, N.J.: Rowman and Littlefield, 1982) 88—102, and Elster’s, Making Sense of Marx (Cambridge: Cambridge University Press, 1985) 345—37 1. Richard Miller challenges this approach, charging that it supposes an atomistic view of people as having preferences that are independent of their role-determined characters, in his Analysing Marx: Morality, Power and History (Princeton: Princeton University Press, 1984) 63—76. Similarly, Samuel Bowles and Herbert Gintis, Democracy and Capitalism: Property, Community, and the Contradictions of Modern Social Thought (New York: Basic Books, 1986), argue that It is a mistake to hold (as is supposed in statements of the free-ride problem) that people participate only ‘to meet preexisting ends,’ because ‘preferences are as much formed as revealed in the exercise of choice,’ 138—9. There are germs of a solution to the free-rider problem here, albeit a less decisive solution than one might want.

    If peoples’ characters are importantly formed by the nature and degree of their participation in joint actions with others, and if they want to have some control over this process, then they are ill advised to absent themselves from participation. That people are in fact thus formed is a claim almost all democratic socialists are prepared to defend. That people want to participate to this end is more problematic. Miller seems to accept a doctrine of motivating objective interests (107, 143, 149), but we shall see that this is itself problematic. Bowles and Gintis (who do not employ an objective interest doctrine, 149—50) imply either that people have or that they ought to have social formation of their characters through participation as a goal. Elster or Buchanan could reply that not everyone has this aim, and that the free-rider problem is not dissolved, but highlighted, by prescribing that people participate. Still, the number of those tempted to be free riders can be reduced by encouraging recognition of the way social processes affect one’s personality and of the possibility of affecting these processes through participation. Chapter 8, section iv pursues this topic.

  22. I believe the term was coined by Alvin Weinberg, who used it to describe the superiority of technology over ‘social engineering’ (including both attempts to eliminate the causes of social problems and to change people’s values) due to the relative economy of making technological innovations, ‘Can Technology Replace Social Engineering,’ University of Chicago Magazine 59: October(1966), 6—10; reprinted in Technology and Man’s Future, 2nd ed., ed. Albert Teich (New York: St. Martin’s Press, 1977) 22—30.

  23. Studies of public apathy bear this out. For example, in explaining decreasing public participation in environmental policy formation, Steven Schatzow concludes that there is ‘a general suspicion and dislike for greater public participation’ on the part the government, and he lists a variety of realistic measures that could rectify inhibitions to participation by creating more possibilities for it. ‘The Influence of the Public on Federal Environmental Decision-making in Canada,’ in Public Participation in Planning, eds. W. Derrick Sewell and J. Coppock (London: John Wiley, 1977) 14 1—158. See, too, the essay, ‘Citizen Participation in Practice: Some Dilemmas and Possible Solutions,’ by Timothy O’Riordan in the same collection, 159—171.

  24. Mueller, Public Choice, makes use of the work of Kurt Wicksell to distinguish between ‘allocative’ and ‘redistributional’ decisions, showing that they require different voting procedures.

  25. Ibid. 268. Mueller’s distinction between majority rule and unanimity is summarized in a chart on 2 16—17. Chaps. 11 and 12 treat the book’s larger themes. And see Robert Dahi, Dilemmas of Pluralist Democracy: Autonomy vs. Control (New Haven: Yale University Press, 1982) 81—95.

  26. Michael Davis makes a similar point in criticizing Arrow. He compares a ‘deliberative’ body to a scientific community, each of which makes decisions both about means and ends in an already partially shaped environment, ‘Voter’s Paradox,’ 303. Examination of what political theorists and philosophers have had to say about real- life constitutional questions can be found in the case of Canada/Quebec in Philosophers Look at Canadian Confederation/La confederation canadienne: qu’en pensent les philosophes? , ed. Stanley G. French (Montreal: The Canadian Philosophical Association, 1979).

  27. Allusion to a ‘world history of democracy’ is not only made for heuristic purposes. Given the complex ways that the world’s societies overlap, a synoptic history sufficiently detailed to be of interest would be an undertaking of mind- boggling proportions. This does not mean, however, that historians cannot address fair-sized portions of our pasts to exhibit the things that impede or facilitate progress in democracy. Something like this has already been a principal focus of some historians.

    Thus, historians in the tradition of Charles Beard looked critically at political institutions of the United States to ask, as Richard Hofstadter observes, not what laws or policies exist, but ‘who has control and what do they want?’ in Hofstadter’s, The Progressive Historians (New York: Alfred Knopf, 1968). Barrington Moore’s, Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World (Boston: Beacon, 1966), is another example, where he tries to show how different ‘routes to the modern world’ led to either (parliamentary) democracy or to (left or right) dictatorship. Raymond Williams studies cultural history to chart the changing conceptions and evaluations of ‘democracy’ and related ideas, Culture and Society: 1780—1 950 (London: Chatto and Windus, 1958). Then again, there are the labour historians, such as Herbert Gutman whose focus is on the everyday experience of working people (not just of labour organizations and leaders) both in and outside of the workplace to produce what James R. Green calls ‘labor history from the bottom up,’ The World of the Worker: Labor in Twentieth-Century America (New York: Hill and Wang, 1980). And, of course, there is E.P. Thompson’s The Making of the English Working Class (London: Penguin, 1980), first published in 1963, which integrates a bottom-up orientation with institutional and cultural ones.

    Perhaps it is not presumptuous to suggest that historians could profit from the work of democratic theorists in pursuit of detailed histories of democracy and that much more such historical study should be undertaken. Without question, the fruit of this work would benefit democratic theory and practice.

  28. Bowles and Gintis, Democracy and Capitalism ‘Two possibilities present themselves. In one, a democratic dynamic, a democratic set of rules induces a more democratic culture, and this in turn leads actors to render the rules more democratic, further enhancing democratic culture and eventually leading, perhaps, to a highly democratic institutional equilibrium. Equally possible, however, is an antidemocratic dynamic, in which the rules promote a less democratic culture, [fostering] the progressive erosion of democratic rules,’ 187.

  29. It is true that this possibility is hypocritically appealed to by authoritarians, and a political unit of any size in which elections were indefinitely postponed would no doubt be grossly deficient in democracy. However, it should be remembered that elections are culminating events of complex processes, requiring mechanisms for identifying issues, formulating platforms, nominating candidates, carrying on debate, determining what counts as a decisive result, and providing structures for assumption of political authority after the election, not to mention holding the election itself. All this takes time.

    Shortly after the Cuban Revolution many antisocialist politicians in the United States (in the main those who had been conspicuously silent during the Batista years) loudly demanded that the new regime hold immediate elections. Making reference to the social and economic difficulties that had to be overcome to allow for elections, Castro is reported to have reminded his critics that the Cuban people were still armed and thus had the means to effect a change of government directly if they had wanted this.

  30. Benjamin Barber, Strong Democracy: Participatory Politics for a New Age (Berkeley: University of California Press, 1984) 224.

  31. Jurgen Habermas, Legitimation Crisis, trans. Thomas McCarthy (Boston: Beacon Press, 1975) 123, and pt. 3. See also William Connolly’s Appearance and Reality in Politics (Cambridge: Cambridge University Press, 1981), especially chap. 3.

  32. Barber, Strong Democracy 198.

  33. Felix Oppenheim, ‘Democracy,’ see 32—40.

  34. Oppenheim is right to conclude his article by observing that democracy ‘cannot provide a solution to all social problems,’ but he is too pessimistic, and indeed inconsistent, when he says that ‘not more, but less democracy,’ is required to better the lot of outvoted and deprived minorities in the contemporary United States, ibid. 50.

  35. Carl Cohen, Democracy (New York: The Free Press, 1973) 7 1—4.

  36. Kenneth 0. May, ‘A Set of Independent, Necessary and Sufficient Conditions for Simple Majority Decision,’ Econometrica 20(4): 680—4(October 1952). (Those of us who worked with Ken May before his sudden death in 1977 recall that he was an active democratic socialist as well as a gifted theorist.)

  37. Elaine Spitz, Majority Rule (Chatham, N.J.: Chatham House Publishers, 1984), chaps. 9 and 10. Spitz’s response is compatible with that of Oppenheim and, while she does not explicitly express it, also with the theory of the anonymous majority.

  38. In addition to Oppenheim and Spitz, the point is made by Dahl, Dilemmas 6; Joshua Cohen and Joel Rogers, On Democracy: Toward a Transformation of American Society (Harmondsworth, England: Penguin, 1983) 149—50, and Medvedev, Socialist Democracy 42, 45.

  39. Green, Retrieving Democracy 17 1—2.

  40. Green argues later in his book that an adequate defence of minority rights cannot be derived from empirical or logical considerations about the requirements for democracy, but must rather appeal directly to considerations of morality, ibid. 214. The worry Green expresses is less severe for somebody who thinks of democracy as a matter of degree. Such a one can agree that the necessity of protecting minority rights is neither a logical nor an empirical requirement for there being any degree of democracy whatsoever, while still maintaining that protection of minority rights is an empirically necessary condition for making advances in democracy.

  41. Ibid. 171.

  42. Claude Lefort, L’invention democratique: les limites de la domination totalitaire (Paris: Fayard, 1981) 92, see especially chaps. 2 and 3. And see François Furet’s Interpreting the French Revolution, trans. Elborg Forster (Cambridge: Cambridge University Press, 1981). Claus Offe, citing Max Weber, makes a similar point, Contradictions of the Welfare State, ed. John Keane (Cambridge, Mass.: MIT Press, 1984) 134.

  43. Daniel Goldstick makes this point in an exchange with Richard Woliheim over whether it is a ‘paradox in the theory of democracy’ that a majoritarian in the minority on some issue is committed to having contradictory values regarding that issue. Goldstick’s article is ‘An Alleged Paradox in the Theory of Democracy,’ Philosophy and Public Affairs 2(2): l8l—189(Winter 1973).

  44. See in this connection Y. Michael Bodemann, ‘The Fulfilment of Fieldwork in Marxist Praxis,’ Dialectical Anthropology 4: 155—161(1979).

  45. Moses Finley, Democracy: Ancient and Modern (New Brunswick, N.J.: Rutgers University Press, 1973) 103.

  46. Benjamin Barber does a good job of discussing these various aspects of the requirements for participation in his Strong Democracy, Pt. 2. In my view, his suggestions can be accepted without also accepting his problematic classification of ‘politics as epistemology,’ 166—7.

  47. Michael Waizer, ‘Philosophy and Democracy,’ in What Should Political Theory Be Now? , ed. John S. Nelson (Albany: State University of New York Press, 1983) 75—99, at p. 93.

  48. Examples are Roger Scruton, The Meaning of Conservatism (London: Macmillan, 1980) and Russell Kirk, The Conservative Mind (Chicago: Henry Regnery, 1953): the essence of social conservativism is preservation of the ancient traditions of humanity,’ Kirk, 7.

  49. ‘Efficient’ is here used as a quality of means, not ends. Given that a social goal has been set, alternative sorts of activities, institutions, or modes of organization will be ranked as more or less efficient by estimating their relative likelihood of success at attaining the goal in question and taking account of costs incurred (which costs are in turn determined by reference to other goals). In much contemporary economic theory ‘efficiency’ is used in a different sense, not to be confused with this one, to describe a certain sort of social outcome, namely one which is ‘Pareto optimal.’

  50. Seymour Martin Lipset, The First New Nation: The United States in Historical and Comparative Perspective (New York: Basic Books, 1963), chap. 6.

  51. Habermas, Legitimation Crisis 133—8.

  52. John Stuart Mill was good at describing the first of these advantages but in an elitist way singularly blind to the second. See the critique by Zillah Eisenstein, The Radical Future of Liberal Feminism (New York: Longman 1981), chap.6.

  53. In Canada a ‘War Measures Act’ suspending civil liberties was used sporadically during World War I and systematically on only two occasions — against Japanese Canadians during World War II and against Quebec nationalists in 1973 — both times inflicting unjust material and psychological harm on innocent people. The pretext in the first case was that Canadians of Japanese descent were Fifth Column threats; in the second it was declared that an ‘insurrection’ was underway. In both cases the justifying claims have been shown not only false, but known false by leading government officials at the time. See Ann Gomer Sunahara, The Politics of Racism: The Uprooting of Japanese Canadians During the Second World War (Toronto: Lorimer, 1981), and Aubrey Golden and Ron Haggart, Rumors of War (Toronto: New Press, 1971). I dare say that readers from any other country in the world can produce their own examples.

Democracy Library