Please cite the published version of this paper. This version may contain typographical and typsetting errors. Published as Cox, Ryan. 2022. “Democracy and Social Equality.” Journal of Ethics and Social Philosophy 23 (1). https://doi.org/10.26556%2Fjesp.v23i1.1890
Recent egalitarian theorizing has drawn attention to the importance of a conception of social equality understood in terms of all members of a society relating to each other as equals.1 On such a conception of social equality, there appears to be a particularly intimate connection between democracy and social equality. Democracy—understood in terms of all members of a society having an equal opportunity to influence political decisionmaking—appears to be necessary for social equality—understood in terms of all members of a society relating to each others as equals. Democracy appears to be an important constituent of social equality.
While egalitarian theorists have often assumed that there is an intimate relation between democracy and social equality, only recently have they attempted to clarify the exact nature of the relation between social equality and democracy and to argue that such a relation obtains. Earlier egalitarian theorists seem to have held that while social equality was intimately related to democracy, democracy only bore a contingent and causal relation to social equality—democracy, perhaps, was seen to be more likely than other forms of government to lead to social equality. Many recent egalitarian theorists, however, have wanted to argue that the relation between democracy and social equality is more intimate than this. They have wanted to argue that democracy is necessary for social equality and that it is an important constituent of social equality.2 For instance, Niko Kolodny has argued that democracy is “necessary for full or ideal social equality” (Kolodny 2014b, 308) and that it “is a particularly important constituent of a society in which people are related to one another as social equals” (Kolodny 2014a, 196); Elizabeth Anderson has argued that hierarchies of command are constitutive of social inequality and has investigated the ways in which democratic forms of government turn relations of rulers and subjects into relations of agents to principals, where “democratically accountable officeholders” do not “constitute a social hierarchy in the sense of distinct classes of rulers and ruled” (Anderson 2012, 43–44, 48); Samuel Scheffler has argued that “[t]he ideal of a society of equals” is “subject to the presumption that each participant in an egalitarian relationship is equally entitled to participate in decisions made within the context of the relationship” (Scheffler 2015, 38); and, more recently, Daniel Viehoff has written that he is “sympathetic to the theory that equality is a constitutive component of certain non-derivatively valuable relationships” (Viehoff 2019, 8).
The question of the nature of the relation between democracy and social equality bears directly on the issue of the nature of justifications of democracy in terms of social equality. Many democratic theorists seek non-instrumental justifications of democracy over alternative forms of government.3 If democracy—understood as requiring political equality—is necessary for full social equality, then insofar as we have an interest in relating as equals to those whom we stand in ongoing social relations with, we will have a non-instrumental pro tanto reason to support democracy over alternative forms of government. Thus we have the beginnings of the kind of justification of democracy which goes beyond the standard instrumental justifications of democracy, and may favour democracy over the alternatives, even if democracy does not turn out to be the instrumentally best form of government.4
My aim, in this essay, will be to argue that the relation between democracy and social equality is, at best, a contingent and causal relation. It follows that there can be no non-instrumental justification of democracy in terms of social inequality, so understood.5 While I agree with egalitarian theorists that we have an interest in relating as equals to those whom we stand in ongoing social relations with—that is, I agree with the egalitarian assumption behind this kind of argument for democracy—and I agree that there is an intimate relation between democracy and social equality, I believe that it is a mistake to think, as recent egalitarian theorists do, that democracy is necessary for social equality and that democracy is an important constituent of social equality. My case against the constitution thesis—the thesis that democracy is necessary for social equality and is an important constituent of social equality—will involve presenting a series of cases which show how democracy and social equality can come apart. This, of course, is an appropriate methodology for assessing claims of necessity and constitution. However, I will not rest my case entirely on such judgements about these cases. By presenting an alternative account of the relation between democracy and social equality, one on which democracy is merely contingently and causally related to social equality, I will attempt to explain away the initial appearance of a necessary and constitutive connection between democracy and social equality. I will argue that a better explanation of what is problematic about inequalities in power and de facto authority in the political context is that they can be used to establish or reinforce social inequalities either through the direct exercise of that power, or can be used by those with perceived greater power and de facto authority to extract greater consideration from those with less power and authority.6
The discussion is structured as follows. I begin, in section two, by locating the topic of social equality and examining the argument for thinking that democracy is an important constituent of social equality and is necessary for full social equality. In section three, I argue on the basis of a series of examples against the claim that democracy is necessary for full social equality. I argue that these examples show that the relation between democracy and social equality is, at best, causal and contingent and use the examples to motivate the alternative extraction theory. In section four, I compare my argument for this conclusion with a related argument by Richard Arneson. In section five, I briefly discuss the implications of my conclusion for non-instrumental justifications of democracy.
I will begin, in this section, by locating the topic of social equality. It is important for my purposes that we have an independent, pre-theoretical grasp of the notion of social equality and of its normative significance. I will then make some general remarks about the structure of claims about social equality and the normative foundations of social equality before turning to recent analyses of the notion of social equality. With these theoretical and pre-theoretical understandings in hand, I will then examine the case for thinking that democracy is necessary for social equality, and for thinking that democracy is an important constituent of social equality.
The notion of social equality that is of interest to us is one that is best introduced in terms of the notion of members of a society relating to each other as equals. Egalitarian theorists often speak of social equality in such terms. For instance, Elizabeth Anderson writes that egalitarians are committed “to creat[ing] a community in which people stand in relations of equality to others” (Anderson 1999, 289). One way of understanding talk of “relations of equality” is in terms of how members of a society relate to each other. Do the members of a society relate to each other as equals or do they relate to each other as inferiors to superiors or superiors to inferiors? Standing in relations of equality to each other, on this understanding, is a matter of relating to each other as equals as opposed to superiors or inferiors. This understanding is in play when Samuel Scheffler writes that egalitarians should care about “the establishment of a society of equals, whose members relate to one another on a footing of equality” (Scheffler 2015, 21). And it is this understanding which is in play when David Miller writes:7
[Equality] identifies a social ideal, the ideal of a society in which people regard and treat one another as equals—in other words, a society that does not place people in hierarchically ranked categories such as classes. We can call this … kind of equality equality of status or simply social equality. (Miller 1999, 232)
I assume that we each have a good pre-theoretical understanding of the notion of relating to each other as equals involved in social equality. We are able to identify paradigmatic cases where some members of a society relate to each other as superiors to inferiors. We immediately recognise the relation of lord to peasant, of patricians to plebians, of Brahmins to untouchables, of Blacks to Whites (in the reconstruction era South, at least), as relations of superior to inferior. Not only do we immediately recognise these relations as relations of superior to inferior, but we immediately recognise the normative significance of these relations: they are relations we have good reason to avoid. It is by reflecting on such paradigmatic examples, and variations on such examples, and using our pre-theoretical understanding of the notion of relating to each other as equals, that we can start to build a theory of social equality. What is it about these paradigmatic examples of social inequality which make them examples of social inequality?
We will examine an attempt to answer this question in a moment. But before doing so I want to make a few remarks about the structure of claims about social equality which should be common ground among theories of social equality. At the heart of the notion of relating as equals is the notion of relating to each other in a society. The notion of ‘relating’ to each other is a capacious one, and it includes all of the ways that members of a society may relate to each other in a society, from how they interact personally to how they structure their physical environment. I take it to be a task of a theory of social equality to say which of these ways of relating are ways of relating as equals, and which are ways of relating as superior to inferior. It is plausible that some ways of relating to each other will only count as ways of relating as superior to inferior in certain contexts or in certain circumstances. So, for instance, relating to another as one with greater power or authority, may not be a way of relating as superior to inferior in certain contexts, but may be in another context. As we are about to see, recent egalitarian theorists hold that relating to another as one with greater power or authority does amount to relating as superior to inferior, at least in the political context.
As I just said, we immediately recognise the normative significance of relations of social inequality. We recognise that they are relations we have good reason to avoid. I agree with egalitarian theorists that the reasons we have to avoid such relations are non-instrumental reasons. We have reasons to avoid such relations for their own sake. It may well be that the claim that we have good reason to avoid social inequality is moral bedrock. It may be the kind of claim that does not admit of further defense. I suspect, however that the claim that we ought to relate to each other as equals is somehow grounded in the idea that we are, in some important sense, equals. We ought to relate to each other as the equals we in fact are.8 In the absence of now discredited views that some human beings are naturally superior or naturally inferior, most now believe that all human beings are morally equal, and, as such, ought to relate to each other as the equals we are. As a consequence, our social relations—how we relate to each other in society—ought to reflect this basic fact about us and we ought to relate to each other as the equals we are. It is a familiar point, however, that relating to one another “as equals” may involve a great deal of unequal treatment. Our question is whether political equalities are the kind of equalities which are required for the kind of treatment as equals, relating to each other as equals, required for social equality. I mention this foundation of the normative significance of social equality only to point out that it is plausible, whether we take it to be moral bedrock or not, that we have good reason to avoid social inequality for its own sake.9
Having located the topic of social equality, let us now turn to recent attempts to answer the question of which ways of relating to each other in particular contexts or particular circumstances are ways of relating as equals and which are ways of relating as superior to inferior.
Recent egalitarian theorists have attempted to analyse social equality and inequality, to isolate those features of the paradigmatic cases which make them cases of social equality or social inequality. Perhaps the most sustained attempt to offer such an analysis in the recent literature comes from Niko Kolodny. Whereas theorists like Anderson have drawn attention to important differences between different kinds of social hierarchies and their characteristic varieties—hierarchies of domination and command, hierarchies of esteem, and hierarchies of standing—Kolodny has been the most explicit about attempting to isolate the features of these different kinds of hierarchies which make them problematic cases of social inequality.10 It is partly for this reason that I will focus on Kolodny’s analysis here. Another reason for focusing on Kolodny’s account is that, as we will see below, Kolodny foresees a particular kind of objection to analyses of social equality in terms of equalities of power and authority and modifies his account in order to avoid this kind of objection. Although I will mainly focus on Kolodny’s analysis, I take the arguments of the next section to tell against any attempt to analyse social equality in terms of equality in power and de facto authority.
As I mentioned above, we can begin to develop an account or analysis of social equality by reflecting on paradigmatic cases of social inequality, drawing on our pre-theoretical understanding of relating as equals, and attempting to isolate the features which make them cases of social inequality. This is the approach taken by Kolodny. Since it is difficult to distinguish judgements about social equality, and relating as equals, from judgements about distributive equality or distributive justice, Kolodny considers an example of a “society administered by a class of ascetic warriors” in which distributive justice is perfectly administered, and the warriors themselves in no way benefit from the distribution they administer: each is given their due, including the warriors themselves, and any benefits which accrue to them in virtue of having the authority they do is offset by their depriving themselves of other advantages. Kolodny writes that, even in such a society, where everyone has been given their due: “there is an obvious sense in which [the ascetic warriors] constitute a superior social stratum, occupy a higher position in the hierarchy” (Kolodny 2014b, 295).11 Reflecting on this example, Kolodny asks “[W]hat is present in the societies that we have described….that might account for the intuitive presence of social inequality?” (Kolodny 2014b, 295). According to Kolodny “[i]t seems to have to do with the following” (Kolodny 2014b, 295–96):
Some having greater relative power (whether formal or legal, or otherwise) over others, while not being resolutely disposed to refrain from exercising that greater power as something to which others are entitled.
Some having greater relative de facto authority (whether formal or legal, or otherwise) over others, in the sense that their commands or requests are generally, if not exceptionlessly, complied with (although not necessarily for any moral reasons), while not being resolutely disposed to refrain from exercising that greater authority as something to which those others are entitled.
Some having attributes (for example, race, lineage, wealth, perceived divine favor) that generally attract greater consideration than the corresponding attributes of others.
These are plausible candidates for being the features of the example in virtue of which it is an example of social inequality.12 Since it will be important for the discussion to follow, I want to draw attention to the “while”-clauses in (i) and (ii): “while not being resolutely disposed to refrain from exercising that greater power/authority as something to which others are entitled”. According to Kolodny, “What social equality requires is that ‘natural’ power be regulated by the right dispositions” (Kolodny 2014b, 296 n. 8). The thought here is that some having greater relative power over others may not be seen as problematic if those who have the greater power are resolutely disposed to refrain from exercising that power as something to which others are entitled, for in such a case the inequality in power will be regulated by the right dispositions. It is only when those with greater power are not so disposed that greater power becomes problematic. As we will see, this qualification is essential to Kolodny’s view and we will be able to distinguish a simple view of the relation between inequalities of power and de facto authority and social inequality—a view which takes such inequalities themselves to suffice for social inequality—and Kolodny’s more complex view—a view which holds that such inequalities, in the absence of the relevant dispositions, and in certain circumstances, suffice for social inequality.
These are the three characteristic features of social inequality then: asymmetries in power, asymmetries in de facto authority, and asymmetries in consideration.13 These asymmetries are characteristic of cases where members of a society fail to relate to each other as equals. Now, we might ask whether each asymmetry is necessary for social inequality, or whether any asymmetry is independently sufficient for social inequality. Kolodny suggests that the consideration component of social equality “need not be essential to the argument that a concern for social equality implies a concern for democracy” (Kolodny 2014b, 298). He thinks that the power and de facto authority components “may suffice for the argument” (Kolodny 2014b, 298). In other words, according to Kolodny, asymmetries in power and de facto authority—in certain contexts—might suffice for social inequality—that is, they might suffice for a failure to relate as equals. Of course, asymmetries of consideration are nonetheless central to our understanding of social inequality, and it seems that asymmetries here, while perhaps not necessary, may independently suffice for social inequality. This is a point to which we will return later.
Now that we have located the topic of social equality and have a candidate analysis of social inequality in hand, we can ask how democracy is supposed to relate to social equality so understood. As I said earlier, whether a way of relating to each other is a way of relating as equals or as superior to inferior is often a matter of context. Relating to another as one with greater power need not always amount to relating as superior to inferior. After all, there are inequalities in power and de facto authority across many contexts of our social lives, from churches to universities and workplaces, and we do not find these inequalities in power and authority problematic, and we do not think they suffice for a kind of social inequality.14 In order to argue that democracy is necessary for social equality, and that democracy is an important constituent of social equality, we need to know what it is about the political context which explains why inequalities in power and de facto authority suffice, in this context, for relations of social inequality? According to Kolodny, the “three features of political decisions” which are relevant here are (Kolodny 2014a, 226):
According to Kolodny, then, insofar as these are features of the political context, inequalities in power and de facto authority will suffice for relations of social inequality in the political context.15 These features do plausibly explain why inequalities in power and authority are problematic in the political context but not in other contexts.
With these claims in place, we are now in a position to connect democracy and political equality to social equality: since subjection to political decisions is non-voluntary, and political decisions are treated as having final authority, and political decisions involve the use of force, if there are asymmetries of political power and authority, then there is at least some failure to achieve the ideal of social equality.16 So the thought is that once we bear in mind these features of the political context, we will see that asymmetries of political power and authority (in the absence of certain dispositions) suffice for, or constitute, a form of social inequality. Under such conditions, those with greater power relate to those with less as superiors to inferiors. Insofar as we have an interest in standing in relations of social equality to those we share ongoing social relations with, we will have at least a pro tanto reason to support democracy over alternative forms of government, since only democracy, by definition, involves equalities in (opportunities for) relative power and de facto authority.
While I find much to agree with in the egalitarian argument, I believe that the relation between democracy and social equality is not as tight as many egalitarians think it is. In this section I will begin by presenting an example which, I will argue, shows that mere inequalities in power and de facto authority do not suffice, even in the political context, for social inequality. This example tells directly against the simple view of the relation between democracy and social equality mentioned in the previous section. It does not, however, tell directly against Kolodny’s version of the argument, but it does set the scene for an examination of Kolodny’s more complex view. On the basis of an additional series of cases, I will argue against Kolodny’s more complex view. I will argue that inequalities of power and de facto authority, together with the absence of certain dispositions, do not suffice for social inequality.
Consider a futuristic society which conforms to Kolodny’s conception of a social democratic utopia: all members of the society have an equal opportunity to influence political decisions and all relate to each other as social equals—no one is “above” or “below” anybody else. Using the wonders of modern technology, this society is able to ensure that voting on a range of issues is as easy as going to the local store. Indeed, it does this by making digital voting machines widely available throughout the society. In this society the voting machines work perfectly, and everybody’s vote is registered exactly once.
Now consider an almost identical society, which differs from Kolodny’s social democratic utopia in only one respect: unbeknownst to everyone in the society, there is a serious fault with the voting machines. They register the votes of almost half of the citizens—those with names beginning with the letters A through M, say—correctly, but they register the votes of the other half of the citizens—those with names beginning with other letters, N through Z, say—twice. A consequence of this, let us suppose, is that the outcome of votes in this second society never differ from those in the first hypothetical society. Because there is nothing suspicious about the outcomes of the decisionmaking process, nobody ever stops to wonder whether there is a fault with the voting machines and nobody ever comes to know that the votes of some are counted twice.
It is fair to say that those with names beginning with the letters N through Z in the second society have greater power and de facto authority than those with names beginning with the letters A through M. To use Kolodny’s vocabulary and image, some have greater contributory influence, where their greater influence can be modelled in terms of their applying a vector of force which has a greater magnitude than the vector of force applied by others in determining the result of the vote (Kolodny 2014a, 200). So, the two societies certainly differ with respect to the distribution of power and de facto authority. Now, the crucial question for us is this: do these societies also differ with respect to how their members relate to each other? Do some relate to others a superiors to inferiors? By hypothesis, members of the first society relate to each other as equals. Do members of the second society relate to each other as equals, or do some relate to others as superiors to inferiors? It seems to me that there is no difference between the two societies with respect to social equality: even though some have more power and de facto authority than others in the second society they do not thereby relate to those others as superiors to inferiors.
This kind of example is a problem for the simple view of the relation between democracy and social equality but not for Kolodny’s more complex view. The case involves the features which Kolodny claims make inequalities in power and de facto authority problematic: subjection is involuntary; the authority is final; and force is involved. It involves, after all, political decisions. Yet, even with these conditions in place, mere inequalities in political power and de facto authority do not suffice for social inequality. We seem to have an example on our hands where there are inequalities in political power and de facto authority and yet there is no social inequality. Political equality is not necessary for full social equality. The example does not, however, immediately tell against Kolodny’s more complex view, since, as we saw, it is important for Kolodny that inequalities in power and de facto authority be accompanied by the absence of particular dispositions. For Kolodny, social inequality involves some having greater relative power “while not being resolutely disposed to refrain from exercising that greater power as something to which those others are entitled” (Kolodny 2014b, 295). There is nothing in our example to suggest that those who have the greater power are not resolutely disposed to refrain from exercising that greater power as something to which the others are entitled. Perhaps they are. (This is a point I will return to below).
Before moving on to raise problems for Kolodny’s more complex view, I want to consider whether the conclusion I have drawn from the example can be resisted and whether the simple view of the relation between democracy and social equality can be defended. How might my conclusion be resisted? Suppose someone disagrees with my verdict that there is no difference in how people relate to each other between the cases in the relevant sense. How might they do so? Well, it should be immediately conceded that the members of the second society do not relate to each other in exactly the same way as members of the first society do in a very broad sense of “relate to each other”: even though they do not know it, some members of the second society relate to others as those with more power and de facto authority. The question for us, however, is whether this is a way of relating as superior to inferior, or a way of relating as equals. And here it might be thought that this just is a way of relating as superior to inferior, since having more power and de facto authority is characteristic of superiority. But this borders on question-begging. There may be a sense in which some relating to others as those with more power and de facto authority is a way of relating as superior to inferior. But, this itself does not seem to be incompatible with them relating as equals in the relevant sense.
It would be disappointing if our disagreement here amounted to nothing more than a verbal disagreement about the application of the expression “social equality”. I think that we can make progress, however, by considering each of the characteristic features of social inequality identified by Kolodny independently. It should be very clear that in the examples provided there are no asymmetries of consideration in the relevant sense. In terms of consideration, members of the second society certainly relate to each other as equals. Since asymmetries in consideration are a large component of what we object to when we object to the idea of relating as superior and inferior, this is what explains why we do not find the second society particularly problematic. Now let us ask whether there is anything particularly problematic about the asymmetries of power and de facto authority in the second society. On the face of it, there does not seem to be anything particularly problematic here.17 Insofar as relations of social inequality are meant to be problematic, this would suffice to show that there are no relations of social inequality here. Alternatively, we might concede that there does seem to be something problematic here, but then we might attempt to explain away the appearance. We might argue that our concern with asymmetries of power and de facto authority stems from our concern to avoid asymmetries in consideration. It is primarily because greater power and authority can be used to extract differences in consideration, that we object to it. But for it to be so used, those who have it need to know that they have it, and those who do not need to know that they do not. Since those with more power and authority do not know that they have it in the second society, we correspondingly find the asymmetry either not problematic at all, or far less problematic than in cases where there is a risk of the asymmetries being used to extract differences in consideration.
What the example seems to support, then, is what I will call, the extraction theory of the bad of inequalities in power and de facto authority: inequalities in power and de facto authority are problematic not because, in certain circumstances, they constitute social inequalities, but because, in certain circumstances, they can be used to extract greater consideration for those with greater power. A view like this is suggested by Michael Walzer when he writes:
It is not the fact that there are rich and poor that generates egalitarian politics but the fact that the rich ‘grind the faces of the poor,’ impose their poverty upon them, command their deferential behavior. Similarly, it’s not the existence of aristocrats and commoners or of office holders and ordinary citizens (and certainly not the existence of different races or sexes) that produces the popular demand for the abolition of social and political difference; it’s what aristocrats do to commoners, what office holders do to ordinary citizens, what people with power do to those without it. (Walzer 1983 xii–xiii)
The extraction theory can be contrasted with constitutive theories on which the bad of inequalities in power and de facto authority, in certain circumstances, constitute or suffice for social inequalities. The example of the faulty voting machines seems to directly support the extraction theory over a simple constitutive theory of the bad of inequalities in power and de facto authority. Insofar as we find inequalities in power and authority problematic, it is because they can be used to extract greater consideration for those with greater power. However, since the example does not tell against Kolodny’s more complex version of the constitutive theory, it does not support the extraction theory over Kolodny’s theory. I will consider Kolodny’s theory in more detail in the next section. But before doing so, I want to fill out the extraction theory a little more in order to have something concrete to compare Kolodny’s theory with.
In order to fill out the extraction theory a little more, it might be helpful to return to Kolodny’s statement of the third characteristic feature of cases of social inequality. Kolodny’s original statement of the consideration condition is as follows:
- Some having attributes (for example, race, lineage, wealth, perceived divine favor) that generally attract greater consideration than the corresponding attributes of others.
According to the extraction theory one of the relevant attributes here might be perceived greater power. Some having perceived greater power might generally attract greater consideration than the corresponding attributes of others. The extraction theory effectively subsumes the inequalities in power condition under the inequalities in consideration condition.18 Something like this view is suggested in the following passage from Miller:
… social equality is a matter of how people regard one another and how they conduct their social relations. It does not require that people be equal in power, prestige, or wealth…. What matters is how such differences are regarded, and in particular whether they serve to construct a hierarchy in which A can unequivocally be ranked as B’s superior. (Miller 1999, 239)
For Miller, then, social equality is “a matter of how people regard one another and how they conduct their social relations”, that is, it is a matter of consideration. Inequalities in power, prestige, or wealth, according to Miller, are not in themselves problematic from the point of view of social equality. Rather, the issue is how these differences are regarded. Do those with perceived greater power, prestige, or wealth, generally attract greater consideration? Do those with perceived greater power, prestige, or wealth, use this perception to extract greater consideration? According to the extraction theory, differences in consideration will be largely due to those with greater power using that power to extract greater consideration from those with less power. It may well be that, in addition to this, those with greater power simply attract greater consideration without those with greater power using their power to extract such consideration. Even in such cases it may be possible to speak of greater consideration being extracted from those with less power. I will say something about the difference between merely attracting greater consideration, and that consideration being extracted from others in a moment.
In order to fully fill-out the extraction theory, more will need to be said about what consideration is such that some having attributes which generally attract greater consideration than the corresponding attributes of others constitutes a form of social equality. Following Kolodny, we can get an initial fix on the notion of consideration by thinking of those responses that social superiors, as social superiors, characteristically attract (Kolodny 2014b, 297). Such responses would include responses of showing certain forms of deferential respect, fawning and toadying, bowing and scraping (Walzer 1983 xiii), other forms of deference, and efforts to ingratiate or curry favour (Kolodny 2014b, 297). We tend to recognise responses that social superiors, as social superiors characteristically attract, when we see them. It is very difficult to say exactly what characterises this class of responses, however. They are all inherently ways of relating as inferior to superior. They all involve treating someone as superior, even if those who respond in these ways do not believe that the other is somehow above or superior. But what is it about these responses which makes them inherently ways of relating as inferior to superior? How do they differ from other forms of positive responses which are not inherently ways of relating as inferior to superior? As Kolodny notes, the class of responses which constitute consideration is partly characterised by the fact that although the basis of these responses is generally some narrow accidental feature of the person, the response is generally focussed on the person themselves and their interests and claims (Kolodny 2014b, 298). Moreover, the responses are practical matters that deal with the person and claims of the person and have an agent-neutral character—the relevant feature is seen as calling for the same response from everyone (Kolodny 2014b, 298). However, as Kolodny notes, these conditions are not sufficient to characterise the relevant notion of consideration, and a full account remains elusive.19 My own view is that relating as equals in our social interactions is largely a matter of experiencing ourselves as being equals in our social interactions. Some ways of responding to others involve an experience of subordination, or inferiority. Here I am taking seriously Walzer’s claim that “the experience of subordination—of personal subordination, above all—lies behind the vision of equality” (Walzer (1983) xiii). This experience is not a response to a perception of superiority in another, rather, it belongs to the form of response itself. It may be especially grating to have to respond to someone with perceived greater power in a way that is accompanied by an experience of subordination or inferiority when one firmly believes that the person with perceived greater power is not superior to oneself. It is this experience of inferiority, I think, which justifies talk of the extraction of greater consideration. The consideration is felt as being extracted from the inferior and granted to the superior.20 We might think of class of responses which constitute consideration as practical responses to the whole person which are based on perceived accidental features of the person and which have an agent-neutral character and are accompanied by an experience of inferiority on the part of those whose responses they are. What I have offered here is an initial sketch of what a theory of consideration might look like and how it might be developed. Given my aims in this paper, I will not attempt to develop the theory in any more detail here. For my purposes, it is enough to have a clear view of the alternative approach that I want to contrast Kolodny’s theory with.
To sum up the argument to this point. I have argued that the example of the faulty voting machines poses a problem for the simple view of the relation between democracy and social equality. Inequalities in power and de facto authority appear to be compatible with full social equality even in the political context. I have just suggested that any lingering sense that these inequalities are problematic may be explained on the hypothesis that we think that such inequalities are problematic because, when those who have greater power know that they do, they can use their greater authority and power to extract greater consideration from others, thus undermining social equality. Now, someone like Kolodny could well agree with the conclusions reached so far. He could agree that mere inequalities in power and de facto authority do not themselves suffice for social inequality. And he could agree that such inequalities are often problematic because they can be used to extract differences in consideration. However, Kolodny will want to say something stronger also. He will want to claim that such inequalities, together with the absence of certain dispositions, suffice for a kind of social inequality. If so, then there would be a reason, over and above the fact that such inequalities can be used to extract greater consideration, for objecting to inequalities in power and authority. And we would have a less empirical and contingent reason in favour of democracy over the alternatives.
What does Kolodny mean when he says that social inequality involves some having greater relative power over others “while not being resolutely disposed to refrain from exercising that greater power as something to which those others are entitled”? We can start to get a feel for what he means by reflecting on why, in the example of the faulty voting machines, even though half of the society does exercise its greater power and de facto authority, they may nonetheless possess the disposition Kolodny is referring to. In the example of the faulty voting machines, half of the society exercises its greater power and de facto authority unknowingly. They are disposed to exercise their greater power unknowingly. But this is perfectly compatible with their being resolutely disposed to refrain from exercising it knowingly. It may be that if those with greater power discovered that they had it, they would immediately stop exercising it and seek to have equality of power restored.
This explains why the example of the faulty voting machines does not pose an immediate problem for Kolodny’s view. For all that we have said, members of the society might be resolutely disposed to refrain from exercising their greater power as something to which others are entitled. Now, at this stage, we might wonder what would happen if we simply stipulated that members of the society are not so disposed. Consider, then, the first society, the society without faulty voting machines. Suppose each member of that society would happily exercise greater power over others, if they could get it. But as things stand, there is no means for them to get more power, and they are not particularly interested in trying to get more power. While this may reflect badly on their character, it is doubtful that merely not being resolutely disposed to refrain from exercising greater power over others would suffice for a form of social inequality. Suppose that we carry this condition over to the society with faulty voting machines, again supposing that members of the society are not aware of the fault. Do we not now have an example where there are inequalities of power and de facto authority, together with the absence of a disposition to refrain from exercising such greater power and authority? Well, yes and no. There is a sense in which those with greater power lack a resolute disposition to refrain from exercising it, they are just not in a position to exercise the power knowingly, since they do not know that they have it. But it would be uncharitable to understand Kolodny’s position this way. It would be more charitable to understand Kolodny as holding that those who have the greater power are such that (i) they are in a position to exercise it knowingly and (ii) they are not resolutely disposed to refrain from exercising their greater power as something to which others are entitled. This gives a deeper reason for thinking that examples like the faulty voting machines, which turn on the absence of knowledge of greater power, do not pose an immediate problem for views like Kolodny’s. For Kolodny, greater power is problematic, in part, when those who have it are in a position to exercise it knowingly, and yet are not resolutely disposed to refrain from doing so for the right reasons.
While Kolodny’s theory is not immediately threatened by the examples we have considered so far, I think that there are related examples that do pose an immediate problem for Kolodny’s view. Consider a variation on the faulty voting machines example where the fault is due to a devious programmer. The programmer studied political philosophy for many years, and, being unable to find an academic job, turned to programming, and became involved in programming software for voting machines. After a few years of running his social experiment with the faulty voting machines, he decides to change things up a bit. He secretly contacts those whose votes are counted twice, letting them in on this fact. Since he explains his role, those who he contacts have good reason to take him at his word. Now, the programmer is careful to ensure that each of those he contacts does not know who else’s votes are counted twice. And those whose votes are counted twice have good reason to think that if they tried to discover who else had greater power, they would come undone, and the whole system would be undermined. Suppose, as in the variation on the example we were discussing a moment ago, members of this society are not resolutely disposed to refrain from exercising their greater power knowingly. Indeed, the programmer may have explained to them that, since those with greater power have basically been randomly selected, their exercising their greater power makes no difference to the outcome of the voting procedure. In light of this, those with greater power may think that it does not really matter that they have and exercise greater power. They exercise it more from indifference than malice.
We now have an example where all of Kolodny’s conditions for social inequality are met: subjection is involuntary; the authority is final; force is involved; and some have greater power and de facto authority while not being resolutely disposed to refrain from exercising that power and authority as something to which others are entitled. Yet, intuitively, there is no more social inequality in this society than in our original society. There is no relevant difference in how members of this society relate to each other. There is no reason to think that those who choose to exercise their greater power in the circumstances hold a malevolent attitude to other members of their society. Those who know they have greater power and de facto authority are in fact rather lonely in their knowledge, since they know that as soon as they seek to share their knowledge, the whole system will unravel. It is their lonely secret. Moreover, in light of the fact that they also know that their having this greater power makes no difference to the outcomes, it would be hard for them to develop a sense of superiority over others. Having greater power is nothing to them.
If it is agreed that there is no more social inequality in this society than in our original society—the one in which the voting machines work as they should—then we have a counterexample to Kolodny’s position as stated. There are, however, two paths of response open at this stage: a proponent of the constitutive theory might resist the verdict about the case, or they may argue that a small adjustment can be made to the theory in order to avoid this kind of example while still retaining the spirit of the theory. I will examine each kind of response in turn.
There are two ways that a proponent might try to resist my verdict about the case. They might just bluntly state an opposed intuition about the case. Or they might argue that the case is sufficiently similar to cases where we judge that there is a kind of problematic social inequality involved. Suppose someone bluntly states an opposed intuition. Then we can consider similar arguments to those we considered in the previous section. We might ask what is meant to be so problematic about the circumstances in this society? There may be a lingering sense of social inequality here. But again, we should consider what is meant to be problematic about this combination of greater power and the absence of certain dispositions. The example under consideration is designed both as an example where Kolodny’s conditions are met, but also as an example where we have ensured that those with greater power and authority are not in a position to extract greater consideration, even though they know they have greater power and authority. By hypothesis, if they were to try to extract greater consideration, the whole system would come undone. So they cannot use their greater power to extract greater consideration. Even if we think that merely having greater power while lacking the relevant disposition is problematic, it is nowhere near as problematic as when that greater power can be used to extract greater consideration. So, again, we should consider an error theory of any intuitions about inequalities in power in the absence of certain dispositions amounting to social inequality. Indeed, once we are considering such dispositions, we should be even more tempted by an explanation in terms of the extraction theory. For, it is plausible to think that those who would not be resolutely disposed to refrain from exercising their greater power, in ordinary cases, would also not be disposed to refrain from extracting greater consideration on the basis of their having greater power. There is a risk that this line of reasoning may simply lead to a clash of intuitions about the case, however. So let us now examine how a proponent of the constitutive theory might offer an argument to resist the verdict about this case.
A proponent of the constitutive theory might argue that the case at hand is sufficiently similar to a kind of case which is clearly problematic and which clearly involves social inequalities. They may conclude from this that the case at hand involves social inequalities, but perhaps to a lesser degree. The kind of case the proponent of the constitutive theory will appeal to will be a case involving rule by a secret society, or the Illuminati, for instance.21 To bring this case as close as possible to the case at hand, we might imagine that the Illuminati has hired our programmer to introduce the bias into the voting system, and to have the votes of members of the Illuminati counted twice. And, of course, every other member of the society would be left in the dark. Now, we certainly would find such an example problematic. There is something objectionable about secret rule by the Illuminati. And the example has similarities to the case at hand. Those with less power are not aware that those with greater power have greater power, and if their power were to be discovered, the whole system would collapse. Given that we find this example so problematic, why should we not find the original example of the devious programmer problematic? Well, there is an important disanalogy between the example of the Illuminati and the original example. Since the members of the Illuminati are mutually aware of their having greater power, and can communicate their intentions with nods and winks without being noticed, they can be said, collectively, to exert a kind of intentional control over the decisionmaking process that the individuals with more power, taken collectively, could not be said to exert, in the case at hand. I suspect that this is what drives our intuitions about the Illuminati example. And since it is missing in the original example of the devious programmer, our intuitions about the Illuminati example should not be taken to bear on the original example.
An interesting question remains, however, as to whether we think that the Illuminati example is problematic because it involves relations of social inequality, or whether we think it is problematic for some other reason. Do members of the Illuminati relate to other members of the society as social superiors? My sense is that they do not. Here is why: it does not seem to matter for our verdict about what is problematic about the Illuminati example whether members of the Illuminati are members of the society they control, or whether they are members of another society, ruling the original society remotely. So the issue does not seem to be one about how members of a society relate to each other at all. Now, it might be said that, nonetheless, when those with greater power who are members of a collective which controls the decisionmaking process, are members of the same society that they control, they are thereby socially superior. This is fine, but it amounts to a stipulative use of ‘socially superior’, and what is problematic about this instance of social superiority would be shared with cases of foreign rule, and is not explained in terms of a failure to relate as equals. In any case, since those with greater power in the original example are not members of a collective which exerts a kind of intentional control over the decisionmaking process, this line of reasoning does not provide an argument against our verdict about this example.
The second way a proponent of the constitutive theory might respond to the example of the devious programmer is by arguing that a small adjustment can be made to the theory in order to avoid this kind of example while still retaining the spirit of the theory.22 The proponent of the constitutive theory might think that the lesson of the first example of the faulty voting machines is that to have greater power or de facto authority over others in the relevant sense—that is, in the sense that is constitutive of social inequality—it is not merely enough that one have greater contributory influence. It requires that one know that one does. In response to the example at hand, they might argue similarly, that to have greater power or de facto authority over others it is not enough to know that one has greater contributory influence, those who have lesser contributory influence must know that you do. In other words, perhaps the kind of inequalities in power which are constitutive of social equality are inequalities of power which are common knowledge. At the very least, it might be thought that in order for inequalities in power to be constitutive of social inequality, those with less power must have an inkling that some have more power than they do. So why not amend Kolodny’s statement of the conditions which constitute social inequality as follows:
Would this amendment not be largely in the spirit of the constitutive theory and save it from the counterexample? Is it not just obvious that this is what a proponent of the constitutive theory should say? My view is that it will not do simply to amend the theory in this way. What the constitutive theorist needs is that inequalities in power themselves be at least in part constitutive of social equality. They can amend the theory by claiming that it is inequalities of power together with an inkling on the part of those at the losing end of the equalities which constitute social inequality. But this opens them to two objections. First, why should adding knowledge or suspicion on the part of those at the losing end to inequalities in power suddenly amount to a failure of social equality? What does mere suspicion of inequalities of power add to inequalities of power such that together they suddenly constitute social inequalities? Would not a better explanation be that the suspicion of inequalities of power are causally relevant to social inequality? Would not those who suspect that they have less power than others fawn over those with more power and give them greater consideration? Second, once this possibility is noticed, there is a threat that inequalities of power themselves will drop out of the picture as irrelevant. Either it is the suspicion of greater power that constitutes the social equality, or it is the suspicion of greater power that is causally relevant to social equality. Either way, there is no work for actual inequalities in power to do in constituting social inequality. I will now present an example which, I think, shows that inequalities in power do, in fact, drop out of the picture as irrelevant in either case.
Consider a final variation on the faulty voting machine example. This time, begin with a society where, by design, the system is meant to be a system of plural voting. In this society it is common knowledge that the votes of some are counted twice while the votes of others are counted only once. Moreover, and predictably, those with more power use their power to extract greater consideration, so there is significant social inequality supported by the system. Those with more power relate to those with less as superiors to inferiors. But now here is the twist. Our devious programmer has been at it again, and while it is widely believed that the votes of some count for more than the votes of others, this is not in fact the case, and everyone’s vote is counted only once. So one means by which those with perceived greater power might have actually had greater power has been closed off. This is not to say, however, that in this example everybody has equal power. It may well be that those with perceived greater power nonetheless do have slightly greater informal power, since they will be able to influence the decision-making of others through their perceived greater power.
Let us now ask if things would be worse from the point of view of social equality, if the programmer had done his job correctly, and programmed the plural voting scheme. That is, let us ask if things be worse if there were greater informal inequality of power and authority in the society. If the programmer had done his job correctly, then, in addition to the social inequality due to the extraction of greater consideration, and the purported inequalities due to whatever greater power the perception of greater power gives to those who are perceived to have greater power, there would be additional social inequality constituted by some having, through the formal voting system, greater power and de facto authority while not being resolutely disposed to refrain from knowingly exercising such power and authority. But, on the face of it, things would not be worse from the point of view of social equality if this were the case. All of the damage is already done by the widespread belief that some have more power and authority. No further damage is constitutively done by there actually being additional asymmetries in power and authority accompanied by the absence of certain dispositions. On the plausible assumption that this is not simply a case of over-determination, we should conclude that inequalities of power and de facto authority do not suffice in the circumstances for additional social inequality.
This case confirms the suspicion raised at the end of the previous section that once we add knowledge or belief to the picture, the asymmetries in power due to actual differences in contributory influence drop out of the picture as irrelevant. And this should lead us to see the earlier example of the devious programmer in a new light. There we introduced the hypothesis that it was really what those with greater power could do with power that we objected to, and which led to social inequality. Now we can see that they need not actually have greater power. There only needs to be a mutual perception that they have greater power.
This completes my case against both the simple view of the relation between democracy and social equality and Kolodny’s more complex view of this relation. Neither mere inequalities in power and de facto authority, nor inequalities in the absence of certain dispositions, suffice, in the political context for social inequality. When those lacking a disposition to refrain from exercising greater power have greater power over others but are not in a position to extract greater consideration, there is no threat to social equality, and merely having greater power while lacking the relevant disposition does not suffice for social equality. Moreover, when some are in a position to extract greater consideration on the ground of their supposedly having greater power and authority, their actually having greater power and authority, and their lacking the disposition to exercise it, does nothing constitutively to further undermine social equality.
At this stage it might be worth briefly returning to consider Kolodny’s motivating example, the society of ascetic warriors. Why might we have been initially inclined to agree that in such a society there was a problematic form of social inequality? Well, I suspect that it is partly because the example is under-described. Kolodny is eager to describe the case as one in which the requirements of distributive justice are met. But we know little else about how members of the society respond to rule by the ascetic warriors. Perhaps we would not judge the example to be problematic if it came closer to our examples of the faulty voting machines, and there was some factor which ensured that the warriors would not attract greater consideration in virtue of their perceived greater power. Perhaps we judge the example to be problematic only because we assume that their having perceived greater power would attract greater consideration. We cannot conclude immediately on the basis of considering the example that it is the inequality in power and de facto authority which constitutes the problematic social inequality. We need examples like those we have been considering which attempt to tease apart inequalities in power and de facto authority from inequalities in consideration. I have argued that when we consider such examples, we see that it is ultimately the inequalities in consideration, inequalities which can be extracted by those with perceived greater power from those with lesser power, which constitute the problematic forms of social inequality.
In this section I compare the argument given above with a related argument offered by Richard Arneson in the context of discussing Philip Pettit’s work on non-domination.
The argument I have just given is structurally similar to an argument briefly sketched by Richard Arneson in a discussion of republican freedom or freedom as non-domination (in the broader context of a discussion of social equality). As Arneson notes, a central theme in Pettit’s work on freedom as non-domination is that freedom as non-domination is instrumentally valuable as a means to avoiding certain relations of social inequality (Arneson 2010, 46). For Pettit, the connection between domination and social inequality is mediated by common knowledge. Here is the crucial passage from Pettit:
Domination is generally going to involve the awareness of control on the part of the powerful, the awareness of vulnerability on the part of the powerless, and the mutual awareness—indeed, the common awareness among all the parties to the relationship—of this consciousness on each side. The powerless are not going to be able to look the powerful in the eye, conscious as each will be—and conscious as each will be of the other’s consciousness—of this asymmetry. Both will share an awareness that the powerless can do nothing except by the leave of the powerful: that the powerless are at the mercy of the powerful and not on equal terms. The master-slave scenario will materialize, and the asymmetry between the two sides will be a communicative as well as an objective reality. (Pettit 1997, 60–61)
For Pettit, then, domination is a means to social inequality.23 When it is common knowledge that there are relations of domination, these relations can be used to extract differences in consideration, in social status. Having drawn attention to this aspect of Pettit’s view, Arneson then writes: “This comment on the badness of dominating power raises the question: suppose the [conditions for domination] obtain but nobody knows this is so, or nobody knows but the dominant party, and domination is never exercised, so none of the envisaged bad consequences occur” (Arneson 2010, 46). The first part of what Arneson is asking us to suppose is similar to what we supposed in the initial version of the faulty voting machines example: “suppose nobody knows this is so”. The question Arneson then asks is whether, on this supposition, relations of domination would be non-instrumentally bad. He then reports the intuition that they would not be, and in light of the foregoing discussion of the faulty voting machines example, we can agree.24 Arneson draws the following moral: “Pettit’s discussion calls attention to the fact that one might object to inequality of power without prizing equality of power per se” (Arneson 2010, 47).
In the previous section, I argued that Kolodny’s position is not directly susceptible to this style of argument. For Kolodny, asymmetries in power and de facto authority are not problematic in and of themselves. They are problematic only in conjunction with the absence of certain dispositions. We have seen, however, that this style of example can be extended to target Kolodny’s position and we ended up with a view very similar to Pettit’s here. We concluded that asymmetries in power and de facto authority are problematic primarily because they can be used to extract additional consideration for those with perceived greater power. When it is common knowledge that they obtain, asymmetries of power and de facto authority can be used as means to social inequality. This is a point well-made by Pettit in his discussion of the instrumental value of non-domination.
It is striking that Kolodny does not comment on this aspect of Pettit’s view when he compares his own approach to Pettit’s. In Kolodny’s discussion of Pettit on freedom as non-domination, Kolodny suggests—rightly in my view—that our concern with non-domination may not be so much with being under an arbitrary and alien will itself, but with being “on the losing end of an asymmetry of power with another person” (Kolodny 2019, 106). Given the passage I quoted above, it may be plausible to re-interpret Pettit as being concerned more with asymmetries of power rather than with being under an arbitrary and alien will.25 However, reinterpreting Pettit’s concerns in this way would see asymmetries in power and de facto authority as potential means to social inequality, and not as sufficing for or constituting social inequality in the political context. After all, Pettit insists that inequalities in power would be problematic even in the presence of a robust disposition not to knowingly exercise them. From our point of view this looks plausible, since those with greater power could still use that power to extract greater consideration, even if they were in fact resolutely disposed not to knowingly exercise that power.
Where does our discussion leave the egalitarian argument for democracy? Well, in my view, the foregoing does not have any significant implication for the conclusion of the argument, namely, that we have reason to prefer democracy over other forms of government. Nor does it give us any reason to think that there could be no good argument from considerations about social equality for democracy. It rather has implications for the kind of argument we might hope for here.
For Kolodny, since democracy is necessary, in a strong sense, for social equality, insofar as we have an interest in social equality, this gives us a reason to prefer democracy over alternatives to democracy. There is no need for an empirical argument to the effect that democracy is, contingently, as things are here and now, the only available means to avoiding social inequality. It is enough for social inequality, given the nature of the political context, that there be differences in power and de facto authority in the absence of certain dispositions. We can know from the armchair that inequalities in political power and de facto authority, in certain circumstances, suffice for and partly constitute social inequality.
The examples above suggest that the case for democracy on the basis of considerations of social inequality will depend, in large part, on the empirical premise that inequalities in power and de facto authority will be used to extract greater consideration for those with greater power. Now, in my view, this empirical premise is very plausible. But to establish it, we would need to do more than just reflect on hypothetical examples. We would need to investigate the conditions which allow those with greater power to extract greater consideration. It may well be that there are ways of arranging society which would allow some to have greater power and authority without thereby being in a position to extract greater power. Then our argument would not necessarily support democracy. But it is plausible to think that, given the significance of political power, and the importance of political decisions—something highlighted by Kolodny’s focus on finality, force, and involuntariness—the temptation to extract greater consideration would be too great to resist, and that, unless we were not to relate to each other in society at all, greater consideration could easily be extracted from those with less power by those with more power.26