In this very well-argued book, Sean Ingham proposes to replace conceptions of democracy such as rule by the will of the people or rule by the majority by the idea that democracy is government under the control of many majorities. This idea is designed to overcome the difficulties posed by social choice theory for the popular will account as well as the simple majoritarian account. I find this idea quite attractive as it seems to give us a way to characterize rigorously at least one dimension of the idea of democracy as political equality.
Ingham’s foil is an account of popular control as popular will, which is understood in terms of majority rule. The problem with this account arises when a society faces more than two options and there is no Condorcet winner among the options. If every leading option is rejected by some majority, so that there are majority cycles encompassing all the leading options, it is hard to see how there can be a popular will in favour of any option. And if the preference profiles that undermine Condorcet winners arise sufficiently often in a society, democracy as rule by popular will is an idle wheel and it cannot give an account of the normative grounds of actually existing democracy (36).
Against Gerry Mackie’s thesis that profiles of preferences in which there are no Condorcet winners are rare (Mackie 2004), Ingham argues that the evidence for this thesis is at best thin (50–4) and that it is highly likely that a community of people who disagree about policy along a number of dimensions will display preference profiles that trigger majority cycles (43–50).
But Ingham also rejects William Riker’s (Reference Riker1982) pessimistic inference from the pervasiveness of majority cycles to the idea that popular control is illusory and proposes an alternative account of popular control. The difficult job this account of popular control is supposed to perform, then, is to give an intuitively plausible conception of popular control that can still affirm the reality of popular control in the context of pervasive majority cycling. Hence the main activity of the book is to provide a conceptual elaboration of this conception of popular control. I think Ingham goes a long way in establishing such a conception and so his book is a significant achievement.
Let us look at the four elements of this account. First, a majority has some control over the social choice between x and y when and only when it is the case that were the majority to have a shared preference for x over y and were that preference sufficiently intense, then the majority would be able to stop y from being chosen. He calls this the ‘preference condition’ (62). This condition does not imply that the majority that favours x over y always excludes y. It would exclude y if its preference for x over y were sufficiently intense. The key here is that this allows for each of many majorities to have control over an issue even when only one of them gets what it wants.
Second, a majority can have more or less control on two different dimensions. On one dimension, a group has more control the less intensely it has to care about the issue to block the option not preferred. If it has to care very intensely for x over y in order for it to block y, then it has little control. If it does not have to care very intensely for x over y in order to block y, then it has a lot of control. In other words, the less the majority has to give up in order to achieve what it wants, the more control it has. The importance of intensity of preference already involves a crucial departure from the classical Arrowian social choice approach (62). On dimension two, a person or group has more control to the extent that it has control over a greater range of values. So, a person or group could have some control to the extent that it is able to block alternative y when it prefers x to y. But the total amount of control is very limited if that is the only issue in which it is able to block an alternative.
Ingham asserts it is not sufficient for democracy that every majority in a society have some control over some outcomes of political decision making. This condition could well obtain in some authoritarian societies. Having some control can involve merely minimal control concerning x over y, when the person or group must have very intense preferences for x over y before it is able to block y, for example, if it is willing to risk life and limb in order that x prevail over y. This kind of control may obtain for disfavoured groups even in an authoritarian society.
Third, a society is under minimally democratic popular control when and only when every majority has some sufficient level of control over outcomes, or, in other words, no majority has to sacrifice basic needs to get what it wants (106). Presumably the gap between the control of the more powerful and the less powerful must not be too great. We might think that many modern democracies fit this minimal characteristic. Fourth, egalitarian democracy realizes political equality at least in those areas in which it has legitimate authority, when equally sized groups have equal control (156). Equal control of groups over a particular issue implies that each group has the ability to get what it wants given the same intensity of preference over the alternatives. If one group needs to care more about the issue for it to get what it wants, it has less control (158). This notion will require a conception of interpersonal comparisons of preference intensity and hence involves a major departure from the classic framework of social choice.
Ingham defends his account of popular democratic control in a variety of ways. He argues that the counterfactual element of control is necessary for it to fit common intuitions about control (100). But he also defends it on the grounds that it gives an account of popular control that is egalitarian while applying even when there are pervasive majority cycles. Hence it is a conception of popular control that can play a substantial role in evaluating modern societies (101).
This conception of popular control can apply even in cases where there are pervasive majority cycles because it does not require that majorities have control only when they get what they want. A majority can have equal control over some issue and yet not get what it wants because the other majority puts a great deal more store by the result. Hence, all majorities can have control even if no Condorcet winner exists (111).
It is worth noting how far this conception of popular control takes us from the popular will account of democracy. In a sense, democracy here is no longer defended in terms of the expression or formation of a popular will at all. Those who have been attracted to Rousseauian conceptions of democracy will not be happy with the results here.
I think this idea of popular control constitutes an important achievement. But I have some questions about the central ideas.
Welfarism
One worry about the preference condition of control is that it is satisfied when a person gets what he wants because he has intense preferences for those things, so that the condition of control can be satisfied even when the person has no control (intuitively speaking) at all. For example, a person, completely incapable on her own of getting what she wants, may get what she wants because she has a benevolent master who happens to know what she wants and how much she wants it. In the purest case, a benevolent despot simply knows what people want and realizes their desires in proportion to how much they matter to them. A condition of equal control can then be satisfied to the extent that all persons have their desires satisfied in proportion to the degree of importance they attribute to those desires.
Normally, we say persons have control over an outcome when something they do or can do affects the outcome. The outcome is up to them. To be sure, the degree of control I have over some issue is in significant part measured by how much I have to give up in order to get the outcome I want. But as it is stated, I am not sure I can agree that it can give us the whole of the idea of control.
It is interesting that early in Chapter 4, the preference condition is defined in terms of action (61) while the official definition on p. 62 does not mention action at all. And then the worry above arises.
Condorcet problems
Another worry concerns how the approach solves cases where there are no Condorcet winners and there is a kind of equal control. Take the standard majority cycle case:
In this case A and C prefer x > y, A and B prefer y > z, and B and C prefer z > x. Each majority has control because were its preferences sufficiently intense, then it would be able to exclude the dispreferred outcome. Let us suppose, in addition, that each majority has equal control and they also have equally intense preferences. Then it is not clear what the outcome is going to be. It looks like it could be any of the three alternatives and there is nothing to select between them. As a consequence, it is not clear there would be any control over what options win or get blocked. It looks like it would be a kind of random selection. But this contradicts the initial assumption that there is control.
To be sure, this case will be rare; but the trouble is that it raises a conceptual issue about control. On the counterfactual conception of control, each group has control. But in the particular case, no one seems to have control in any intuitive sense.
Equal control and political equality
One problem about the idea that rule by multiple majorities captures equal control concerns the ways in which preferences are formed. The whole account relies on the idea that people’s preferences are the primitive of the picture. Control is defined in terms of how intense one’s preferences must be in order to get what one wants. But if a society is structured in such a way that the preferences of one group are formed in a particularly problematic way while the preferences of the other group are formed in a less problematic way, the latter group have a more genuine control over how the society is structured than the former group. For instance, if the society is structured so that some people have a harder time forming reasonable preferences that are connected with their genuine interests while others do not, then there is a sense in which those who are burdened with inferior epistemic conditions for shaping their preferences lack some kind of control vis-à-vis the society they live in. This is of course a traditional worry relating to issues about false consciousness or ideology but it suggests a kind of concern about the understanding of control. Those who are subjected to processes of preference formation that are problematic in this way might be thought to have less control than others over how the society is organized. The challenge here is that the notion of control must be at least partly understood in terms that are independent of the preferences people have and not entirely in terms of how intense their preferences must be.
This concern does not have to be incompatible with Ingham’s approach. One can have a conception of the distribution of control, in a narrow sense that is suited to persons with already formed preferences, while another, broader conception of control concerns the institutions relating to how preferences are shaped. And the narrower system could be thought of as a component of the larger part.
Still, the broader notion of control or power is interestingly distinct. It would include the ability to learn about one’s interests and develop more refined conceptions of those interests as well as the opportunity to influence the development of other people’s conceptions of their interests. This broader account would need to be a part of a complete conception of political equality.
Consider the claim that indifferent or incomplete preferences on the part of citizens do not detract from their control. Ingham argues that the indifferent slaveholder who chooses not to exercise control, nevertheless has control (86). The problem is that there may be deeper structural reasons for why some citizens do not have suitable preferences over policies such as environmental conditions that shape preferences, as Anthony Downs suggests. This could involve a lack of control.
All of this requires us to ask, what is a preference? Preferences may consist in deeper rankings of social states. Citizens may have preferences over social states but not know how to advance those preferences because they are very poorly informed. This is not how Ingham is using the expression ‘preference’. He writes in terms of preferences over policy and he suggests that sometimes people just don’t have preferences or they are indifferent. If we understand preferences in this latter way then misinformed and undeveloped preferences may sometimes work to the detriment of the person who possesses them. When people act on these preferences, they may not be exercising genuine control even if they actually affect policy.
To be sure, if we understand preferences in a deeper way, then we would need an account of what they are. And we would then have to come up with an account of how citizens can develop the kind of understanding that is necessary to advancing these preferences. Political control would be partly a function of how citizens can come to have such an understanding.
So, we have a bit of a dilemma. Either we describe the theory in terms of a capacity to advance fairly superficial preferences and we then need an account of how they arise in order to have a complete conception of control. Or we understand the account in terms of deeper preferences and then ask how citizens are to figure out how to know what these are and how to advance them in the political system. Either way, we need a significant supplement to the notion of equal control Ingham has provided.
Equal control as political equality
Ingham’s idea of control seems to be a conception of control ‘at the margin’ that is, in a particular situation given the particular preferences of others. The question is, is there a way to scale up this notion of control in a way that is global in character? Political equality, as we normally understand it, has this global character. That is, we think of people as political equals not merely or even usually in particular situations, but on the whole. It is often the case that people, having used up their political capital, are not in situations of equal control with others. If the use of the equal power were exercised on an issue on which they put a great deal of store and with the knowledge that they will have less power later or on other issues, then the subsequent inequality would be legitimate. But it is legitimate only if the overall relations between the political actors are on a footing of equality.
I am not sure how the notion of control can be used to generate this kind of equality on the whole. I would like to see how the notion of equal control as Ingham has developed it in this work can give us an account of overall or global equality of control.
Conclusion
In sum, Sean Ingham’s book is an important contribution to our understanding of the normative foundations of democracy. There remains much work to be done. The welfarism of the approach needs to be tempered in order to give us an account of political equality. And there are a number of puzzles about the conceptual structure of the account. We need to understand how the idea can be scaled up to help us understand the political equality of citizens on the whole and how it can be fitted together with a normative account of the proper conditions of preference formation in democracy.
Acknowledgements
I would like to thank Johanna Thoma for helpful comments on an earlier draft of the paper and to Sean Ingham, Itai Sher and David Wiens for helpful discussion.
Financial support
Funding from the ANR Labex-IAST is gratefully acknowledged.
Thomas Christiano is Professor of Philosophy and Law at the University of Arizona and a Visiting Fellow of the Institute for Advanced Studies in Toulouse. He is co-editor of Philosophy, Politics and Economics (Sage) and co-director of the Center for Ethics, Markets and the Law at the University of Arizona. He has been a Visiting Fellow at All Souls College, the National Humanities Center and the Princeton University Center for Human Values. He has written a couple of books on the philosophical foundations of democracy and has written many articles on democracy, distributive justice, international justice as well as fairness in economic markets.
