2. Non-responsibility-sensitive Conceptions of Relational Egalitarianism

Relational egalitarianism is, as mentioned, a large family of theories that differ along many different dimensions, including whether (un)equal relations are personally or impersonally (dis)valuable.Footnote ⁵ Relevant to our purposes, Lippert-Rasmussen distinguishes between a responsibility-sensitive and a non-responsibility sensitive conception of relational egalitarianism:

In investigating our question – whether equal relations can be unjust according to relational egalitarianism – let us start with outcome relational egalitarianism. Usually, relational egalitarians argue that those with equal moral standing must relate as equals. To exemplify, Lippert-Rasmussen (Reference Lippert-Rasmussen2018: 170) says that ‘as a matter of fact, we are one another’s moral equals and in relating as equals we honour that fact, and this is what grounds the ideal of relational egalitarianism’ (see also Anderson Reference Anderson1999; Schemmel Reference Schemmel2012, Reference Schemmel2021). This expresses the following outcome, or non-responsibility-sensitive, view:

According to this view, justice requires that people who are moral equals also relate as moral equals. To relate as moral equals has at least two components: agency and interests. If two people are to relate as moral equals, they must take each other’s interests and autonomy to be equally important at a fundamental level (Anderson Reference Anderson1999; Scheffler Reference Scheffler, Fourie, Schuppert and Walliman-Helmer2015; Hojlund Reference Hojlund2022: 57–58). When Sexist treats Woman in a sexist manner, they fail to relate as moral equals because Sexist thereby treats Woman’s agency and interests as if they, at a fundamental level, are less important than their own agency and interests.Footnote ⁷ This relation would be unjust according to non-responsibility-sensitive moral relational egalitarianism.

But moral standing is not the only dimension in which relational egalitarians believe that moral equals must relate as equals. They also argue that moral equals must relate as social equals. To exemplify, Anderson (Reference Anderson1999: 313) says that ‘[Relational] Egalitarians base claims to social and political equality on the fact of universal moral equality’. Similarly, Kolodny (Reference Kolodny2014: 300) says, ‘Insofar as we are to have ongoing social relations with other moral equals, we have reason to relate to them as social equals – that is, in a way that deliberately avoids whatever asymmetries in power, authority and consideration would constitute relations of social superiority and inferiority, motivated by a concern to avoid these relations at such’. Finally, Viehoff (Reference Viehoff, Sobel, Vallentyne and Wall2019: 18) says, ‘If we are all moral equals, matter equally, etc., then social status hierarchy is objectionable because it treats us as if we were not’.

Whereas moral standing has to do with interests and wills in a fundamental sense, social standing is different.Footnote ⁸ I follow Kolodny (Reference Kolodny2014: 295–296) in taking social standing to consist of three components: power, de facto authority and consideration.Footnote ⁹ He understands power in a wide sense to include both formal and informal power.Footnote ¹⁰ For instance, adults have more formal power than children in having the right to vote. In this sense, adults have a superior social standing to children. De facto authority has to do with the degree to which one’s commands or requests are generally complied with. That, in the relationship between parent and child, the parent’s requests are generally complied with by the child, but not vice versa, means that the parent has more de facto authority than the child – and that they, in this sense, relate as social unequals. The final dimension, consideration, has to do with the extent to which one has attributes – such as race, lineage, wealth – that attract consideration from others. In a white supremacist society, a white person attracts greater consideration on behalf of their race than a black person. In this sense, the white person stands as a social superior to the black person. These three dimensions can be satisfied to different degrees. The larger the differences in power, de facto authority and consideration, the larger the inequality in the social relation. If there are no differences in power, de facto authority and consideration between Adam and Bert, they relate as social equals. As I said, I will assume this understanding of what it takes to relate as social equals for now. I return to this issue in section 3.

Thus, moral equals must not only relate as moral equals; they must also relate as social equals, according to many relational egalitarians. We can add this to the above view:

Assuming this view, it is clear that equal relations between moral equals are not unjust. However, this view does not tell us how moral unequals should relate to each other. It only tells us how moral equals should relate to each other. One option with regard to moral unequals would be the following:

This view adds to non-responsibility-sensitive relational egalitarianism that moral unequals must also relate as moral and social equals. Why might relational egalitarians want to say that moral unequals should also relate as equals? One reason might be that the reasons for why it is good that moral equals relate as equals also entail that it is good that moral unequals relate as equals. For instance, some relational egalitarians argue that egalitarian relationships are impersonally valuable. O’Neill (Reference O’Neill2008: 130) has most prominently put forward this argument, saying, ‘the existence of these kinds of social relations [egalitarian social relations] should itself be seen as intrinsically valuable, independent of the positive effects that such relations may have for individual welfare’. Even if, in a racist society, an inegalitarian relationship between a black person and a white person may be better for the parties, an egalitarian relationship between them would still be impersonally valuable. Similarly, in a sexist society, an egalitarian marriage would be impersonally valuable even if, for the parties, an inegalitarian relationship would be better, for example because they would thereby not face social sanctions qua not living up to the social norms in society.

If one believes that there is impersonal value in egalitarian relationships between moral equals, it seems natural to also believe that there is impersonal (dis)value in (in)egalitarian relationships between moral unequals (Bengtson and Lippert-Rasmussen Reference Bengtson and Lippert-Rasmussen2023: 403–404).Footnote ¹² After all, one does not have to say that there is as much impersonal value in the latter kind of relationship as in the former. It is perfectly consistent to say that even though egalitarian relationships between moral unequals do not have as much impersonal value as egalitarian relationships between moral equals, they are still impersonally valuable. So this would be one reason why relational egalitarians might want to say that moral unequals should also relate as moral and social equals.

According to broad non-responsibility-sensitive relational egalitarianism, there can be no unjust equal moral or social relations. Since justice according to this view requires both that moral equals and moral unequals relate as moral and social equals, there are no beings left – at least not beings with moral standing – between whom an equal relation may be unjust. Only unequal relations can be unjust. Indeed, any unequal moral or social relation between persons (and persons and animals, for that matter)Footnote ¹³ would in fact be unjust according to this view.

Now, one might argue against broad non-responsibility-sensitive relational egalitarianism that it is implausible that moral unequals should relate as social equals. To flesh out this objection, we may start with the claim that an entity that has moral status matters morally for their own sake. Some entities, such as a rock, do not have moral status. This means that there must be a reason which explains why the former entity has moral status and the rock does not. There must be some property, or properties, which this entity holds and in virtue of which it has moral status which the rock does not hold (Floris Reference Floris2022: 4). Arneson (Reference Arneson and Steinhoff2015) argues that the most plausible such property is rational agency capacity.Footnote ¹⁴ Entities with moral status have the capacity for rational agency. They can ‘identify available courses of action [they] might take, discern reasons for and against the options, weigh and assess the reasons [they] discern, deliberate and make choices, carry out the action chosen, and do all this not simply for a single decision problem at a time but with respect to long-term plans of action and projects [they] might undertake’ (Arneson Reference Arneson and Steinhoff2015: 33–34).

Moral unequals are thus entities that possess rational agency capacity to different degrees.Footnote ¹⁵ Assuming this understanding, parents and (small) children are clearly moral unequals (Floris Reference Floris2022). But most people do not believe that parents and (small) children should relate as social equals, i.e. that they should have the same degree of power and de facto authority in the relationship (the same goes for adult humans and animals). Most believe that the parent should have more power and de facto authority than the child, for example to decide when the child is to go to bed and what the child is to eat. Thus, one might argue, broad non-responsibility-sensitive relational egalitarianism implies that parent and child, being moral unequals, should relate as social equals and this is implausible because it means that they should have equal power and de facto authority. To avoid this, one may instead support:

According to this view, parent and (small) child, being moral unequals, should not relate as moral and social equals. Thus, it avoids the objection pressed against broad non-responsibility-sensitive relational egalitarianism. But this also means that according to narrow non-responsibility-sensitive relational egalitarianism, equal relations can be unjust. Since it prescribes that moral unequals must relate as social unequals, it is unjust if parent and (small) child, being moral unequals, relate as social equals.

One may find this intuitively objectionable. Is it not a bit too much to say that it is unjust if parent and (small) child relate as social equals? It may not be once we notice that this may actually be what is best for the (small) child. Clearly, a two-year-old does not know what is best for them, at least long-term, because they do not yet have a fully developed self with attached life plans and interests. If parent and two-year-old child were to relate as social equals, it would mean that the child would have as much power and authority as their parent in deciding what to eat, when to go to bed, whether to use the tablet etc. We can easily see that this would be bad for the child; that the child would be a worse chooser than their parent with regard to promotion of the child’s interests. In that sense, it does not look objectionable that the parent and child ought to relate as social unequals, as Narrow Non-responsibility-sensitive Relational Egalitarianism prescribes, since it might actually promote the interests of the child that the parent has more power than them. Or so a defender of narrow non-responsibility-sensitive relational egalitarianism may say.Footnote ¹⁷

At this point, one might object that the parent-child example does not in fact show what I intend it to show. One may agree that it is wrong for a parent to conduct their relationship with their child on a footing of equal power and authority (i.e. that they relate as social equals) when this is detrimental to their child’s fundamental interests, regarding which they have fiduciary duties. However, it is not obvious that the wrongness is explicable with reference to relational egalitarian considerations. It seems plausible, instead, that relational equality puts pressure on the parent-child relationship to be made egalitarian, but that this pressure is constrained by the fundamental interests of the child in being treated paternalistically. So perhaps the relationship should simply be as egalitarian as is consistent with promoting the interests of the child. If so, then the parent-child example does not seem to show that equal relations can be unjust on relational grounds, as opposed to being wrong on nonrelational grounds having to do with lack of promotion of the child’s interests (cf. the clarification in relation to the first question in the Introduction).Footnote ¹⁸

I have the following responses. First, the fact that the parent and child relating as social unequals may in fact promote the interests of the child is appealed to on behalf of Narrow Non-responsibility-sensitive Relational Egalitarianism to show that it may in fact not be an implausible implication of the view that it implies that parent and child ought to relate as social unequals. But it is not an argument for why parent and child must relate as social unequals on this view. The reason they should relate as social unequals, according to this view, is that they are moral unequals (the view says that moral unequals ought to relate as social unequals). Another way of making this point is to say that even if it happened to not be in the child’s interests to relate as a social inferior to their parent, Narrow Non-responsibility-sensitive Relational Egalitarianism would still say that they should relate as social unequals precisely because they are moral unequals. Of course, one may disagree with this particular verdict. But this would simply speak against the view. It would not show that there was not a relational concern at stake.

Second, perhaps a cleaner case is to consider a relationship between an adult human and an animal, say, a dog.Footnote ¹⁹ Many believe that animals, including dogs, have a lower moral status than adult humans (see e.g. Jaworska and Tannenbaum Reference Jaworska and Tannenbaum2014; Kagan Reference Kagan2019).Footnote ²⁰ If this is true, it would be unjust if an adult human and a dog were to relate as social equals according to Narrow Non-responsibility-sensitive Relational Egalitarianism, for example if they had the same degree of power and authority. Since they are moral unequals, they should relate as social unequals. And again, this should be the case, according to this view, irrespective of whether the dog’s interests would be better promoted if they related as social equals instead. One may find this implausible. But, again, this would be a reason to reject the Narrow Non-responsibility-sensitive view (and would not be something that shows that what is at stake is not relational concerns).

So much for the non-responsibility-sensitive conceptions of relational egalitarianism. We have seen that according to broad non-responsibility-sensitive relational egalitarianism, no equal relation can be unjust. According to narrow non-responsibility-sensitive relational egalitarianism, on the other hand, some equal relations are unjust. For instance, it is unjust if moral unequals relate as social equals. Interestingly, this means that even among conceptions of relational egalitarianism that do not take responsibility into account, there is disagreement as to whether equal relations can be unjust.

Before I move on, I would like to address the following worry. Sure, it is a logical possibility that there is a non-responsibility-sensitive conception of relational egalitarianism in which equal relations can be unjust. However, this possibility seems removed from what relational egalitarians have defended and for that reason is not particularly interesting.Footnote ²¹ I have two responses. First, it is not true that relational egalitarians have not defended a view along these lines. Bengtson (Reference Bidadanure2022) argues, assuming that animals and human adults are not moral equals, that it is unjust if they relate as equals since moral unequals should relate as unequals. So a non-responsibility-sensitive conception of relational egalitarianism does not seem removed from what relational egalitarians have defended, or at least discussed. Second, even if it were true that no relational egalitarian had defended a view along these lines, it may still be valuable to find out what committing to such a view would imply. For one thing, it illustrates the breadth in relational egalitarianism: that relational egalitarianism can mean many different things. For another, it shows that if one wants a relational egalitarian view in which equal relations can be unjust, one does not have to commit to a responsibility-sensitive view.

3. Responsibility-sensitive Conceptions of Relational Egalitarianism

Let us now turn to responsibility-sensitive conceptions of relational egalitarianism. As mentioned earlier, Lippert-Rasmussen (Reference Lippert-Rasmussen2018: 7) presents the following conception:

According to this view, unequal relations which are not due to differences in exercises of responsibility are unjust. It is unjust if Man and Woman relate as social unequals in a sexist society since this, let us assume, has nothing to do with them exercising their agency in different ways. It is simply that the norms in this society prescribe that men are socially superior to women.

May some equal relations be unjust according to luck relational egalitarianism? Notice that this view is negative in the sense that it deems it unjust if relations are unequal for reasons other than differential exercises of responsibility. But at the same time, it is silent on whether it is unjust that relations are equal assuming differential exercises of responsibility. It only tells us that relations must not be unequal except for differences in exercise of responsibility. So it does not provide an answer to our question of whether some equal relations are unjust according to relational egalitarianism. Since this is a negative view, let us refer to it as Negative Responsibility-sensitive Relational Egalitarianism (instead of luck relational egalitarianism). We may then distinguish this view from the following view:

This view is a relational egalitarian view in the sense that it assumes a baseline of equality (cf. Olsaretti Reference Olsaretti2002: 396).Footnote ²² If there are not differential exercises of responsibility, relations should be equal. Some have recently argued that we should make relational egalitarianism responsibility-sensitive. Indeed, in a recent paper, Schmidt (Reference Schmidt2022) argues that responsibility practices are a constitutive aspect of flourishing, egalitarian relationships in the first place. And, moreover, that society must be organized in a way that takes this fact into account. He presents the following example:

He imagines that this takes place in Equalia, ‘a society bent on minimizing relational inequalities’. In this society, ‘institutions are set up such that Petra’s lack of resources will be compensated to the point where she again stands in perfect relational equality with all other citizens’ (Schmidt Reference Schmidt2022: 1376). As Schmidt explains, one of the problems with Equalia’s institutions is that they facilitate costless reneging. ‘Someone engages in costless reneging’, Schmidt (Reference Schmidt2022: 1388) explains, ‘when she acquires obligations to other participants, does not discharge such obligations, bears no significant costs as a result, and it would be appropriate to hold her responsible (in the attributability sense) for acquiring the obligations and for not discharging them’. Because Equalia’s institutions are not responsibility-sensitive, they let Petra ‘get away’ with burning 90% of her salary. In this case, Petra engages in costless reneging against the other citizens in Equalia. To see why this leads to objectionable relational equality, we need to introduce a distinction between the following two forms of relational (in)equality:

To see the difference between these views, suppose that Xavier exploits Yoel for 30 years after which Yoel exploits Xavier for 30 years (cf. Lippert-Rasmussen Reference Lippert-Rasmussen2019: 155). In this case, Xavier and Yoel relate as equals from the perspective of diachronic relational egalitarianism, but not according to synchronic relational egalitarianism since at no point in time in their lives do they relate as social equals (see also Bidadanure Reference Bidadanure2016).

Equalia’s institutions are set up to always secure synchronic relational equality. If Petra’s burning 90% of her salary leaves her destitute, Equalia’s institutions secure that she is compensated to an extent that she is no longer destitute such that synchronic relational equality is restored (at least to that extent). In this way, Equalia’s institutions enable costless reneging such that there are no consequences for Petra in not taking into account other people’s interests.Footnote ²³ Those who engage in costless reneging thus treat other people in society as lesser participants in their cooperative undertaking, thereby creating diachronic relational inequality. As Schmidt (Reference Schmidt2022: 1388) says, ‘more generally, when we separate distributions from responsibility entirely, we might set up social structures that enable costless reneging and thereby diachronic relational inequality’.

We can see how this is unjust according to positive responsibility-sensitive relational egalitarianism. When the system facilitates costless reneging, everyone does not relate to others in accordance with their exercise of responsibility. Given that Equalia compensates Petra, she relates, in the synchronic sense, as a social equal to the other citizens of Equalia, according to Schmidt: that is what the institutions in Equalia secure. But they have exercised their responsibility quite differently: Petra has irresponsibly burned 90% of her income while they have not. According to positive responsibility-sensitive relational egalitarianism, the relationships should reflect this such that Petra ought to relate as an inferior to the other citizens – justice provides a reason not to compensate her.Footnote ²⁴ That is, the equal relations between Petra and the other citizens are unjust.

Schmidt presents another case which is relevant for our purposes. Consider:

Schmidt uses this example to show how the social context matters to how people relate to each other. In this case, the discrimination Carlton faces results in a relational inequality between him and his white friends. As Schmidt (Reference Schmidt2022: 1389) concludes, ‘[Carlton shows that] relational inequality can be generated between two groups (or between individuals) when a third party treats them in a way that expresses unequal status’. But if objectionable relational inequality can be generated between two groups, or individuals, when a third party treats them in a way that expresses unequal status, we might also think that objectionable relational equality can be generated between two groups, or individuals, when a third party treats them in a way that expresses equal status. Money Burner might be an instance of this. When Equalia’s institutions compensate Petra to secure synchronic equality between Petra and other citizens, their treatment may be said to express equal status in the synchronic sense and be objectionable for this reason (even if it might also express unequal status in the diachronic sense), especially if intentions are part of what determines what an act expresses (Scanlon Reference Scanlon2008: 53; Lippert-Rasmussen Reference Lippert-Rasmussen2018: 77). The treatment expresses that Petra and other citizens relate as synchronic equals, but it should not since they should not relate as synchronic equals given their differential exercise of responsibility – at least according to positive responsibility-sensitive relational egalitarianism. Some might disagree with what this particular example expresses. That is fine with me since it does not challenge the general point in this paragraph: if objectionable relational inequality can be generated between two groups, or individuals, when a third party treats them in a way that expresses unequal status, objectionable relational equality can be generated between two groups, or individuals, when a third party treats them in a way that expresses equal status. Thus, we have now seen two ways in which equal relations may be unjust according to positive responsibility-sensitive relational egalitarianism.

Stemplowska (Reference Stemplowska, Knight and Stemplowska2011) has also argued that relational egalitarianism should be responsibility-sensitive. She argues that ‘we should recognize that being owed respect as a social equal is conditional upon respecting others as one’s moral equals. That is, only those who respect equality of moral status should themselves be recognized as being owed respect as social equals’ (Stemplowska Reference Stemplowska, Knight and Stemplowska2011: 131).Footnote ²⁵ Suppose that Criminal treats Victim in a wrongful manner, for example by beating them up. In doing so, Criminal fails to respect Victim as their moral equal, i.e. they fail to treat Victim’s interests as fundamentally of equal importance to their own. But if Stemplowska is right, this means that Criminal should not ‘be recognized as being owed respect as social equals’ (or at least that Criminal is not owed such recognition). Indeed, as Stemplowska (Reference Stemplowska, Knight and Stemplowska2011: 133) says, ‘the proponents of the ideal of equal social respect [relational egalitarianism] would not want to (and, in fact, do not)Footnote ²⁶ argue that even the perpetrators of hideous crimes must be respected as social equals by others’. In this way, how people should relate to each other in terms of social standing depends on how they have exercised their responsibility in relation to treating others as moral (un)equals. This is in line with, although narrower than, positive responsibility-sensitive relational egalitarianism: justice requires that Criminal and Victim relate as social unequals since Criminal has failed to respect Victim as their moral equal.

At this point, one might object in the following way. When I laid out what it takes to relate as moral and social equals in section 2, I seemed to think of equal relations in terms of some fixed formula, for example equal power, such that any power inequality makes a relationship inegalitarian. However, it is not clear that relational egalitarians must be committed to the view that inequalities of power necessarily make a relationship inegalitarian. A relational egalitarian may hold, for instance, that to relate to Criminal as a moral equal is to impose appropriate punishment on them, and such punishment may be to grant Criminal less power, for example in the form of (temporary) disenfranchisement. On this view, imposing the power inequality is part and parcel of relating to the offender as an equal. More generally, relational egalitarians endorse various ‘tempering factors’ (Kolodny Reference Kolodny2023: 98) that render power inequalities morally unproblematic, or at least less problematic. This seems to be a different view of what an egalitarian relationship is. Assuming this view, it is not clear how the Criminal-Victim example provides an example of an unjust equal relation.Footnote ²⁷

I have the following responses. First, I do not mean to deny that relational egalitarians endorse various tempering factors. Indeed, I mention some of them in the next section when presenting Kolodny’s view of why the political domain is particularly important for how people relate to each other (see also Kolodny Reference Kolodny2023: 97–101, 125–126). Second, as I have emphasized throughout the paper, relational egalitarianism is not merely one view. There are many different conceptions of relational egalitarianism. And my aim in this paper has not been to defend any particular conception. My aim has been to investigate whether equal relations can be unjust on different conceptions of relational egalitarianism. So I agree that a relational egalitarian could support the view laid out in the objection, i.e. that to relate to Criminal as a moral equal is to hold them accountable and thus to impose appropriate punishment on them. The relevant question for our purposes is: can an equal relation be unjust on this view? One way to argue for an affirmative answer is as follows. Here, the distinction between relating as moral equals and relating as social equals, which I laid out in section 2, becomes important. If we assume that relating to Criminal as a moral equal is to hold them accountable and thus to impose appropriate punishment on them, then this may at the same time affect how Criminal relates to others as social equals. As we will see in the next section, equal political power is, according to Kolodny and other relational egalitarians, a particularly important constituent part of relating as social equals. In other words, if X and Y have unequal political power, X and Y relate as social unequals. Now, if the appropriate punishment in terms of relating to Criminal as a moral equal is (temporary) disenfranchisement, then holding Criminal responsible entails that Criminal relates to others – those with more political power – as a social unequal. Ensuring equal moral standing in this case requires unequal social standing. The view requires that we relate to Criminal as a moral equal and a social unequal. Suppose now that we failed to hold Criminal responsible. In that case, Criminal would not have less political power than others and would therefore, all else equal, relate to others as a social equal. In that case, we would not realize what the view required, namely that we relate to Criminal as a moral equal (by holding them responsible for what they did) and a social unequal (by granting them less political power than others). In that case, both the unequal moral relation and the equal social relation would be unjust. If so, that would be an example of how there could be an unjust equal relation on this view.

Third, even if we assume that the suggestion in the previous paragraph is false – indeed, even if we assume that on the view in question, no equal relations can be unjust – this would not threaten what I have been arguing in this paper. It would, in that case, simply point to a responsibility-sensitive conception of relational egalitarianism in which equal relations could not be unjust (the view would therefore be different from Stemplowska’s view laid out above). And this would be in line with my argument that whether equal relations can be unjust cuts across the distinction between responsibility-sensitive and non-responsibility-sensitive conceptions of relational egalitarianism.

To sum up this section, we have seen that responsibility-sensitive conceptions of relational egalitarianism give different answers to whether equal relations can be unjust. Negative responsibility-sensitive relational egalitarianism is simply silent on this question: it only tells us that it is unjust if relations are unequal for reasons other than differential exercises of responsibility. According to positive responsibility-sensitive relational egalitarianism, equal relations can be unjust. In total, we have investigated four different conceptions of relational egalitarianism, two which are responsibility-sensitive and two which are not. Surprisingly, we have seen that whether equal relations can be unjust cuts across the distinction between responsibility-sensitive and non-responsibility-sensitive conceptions of relational egalitarianism.

4. Unjust Equal Relations: Further Implications

In this section, I will continue with the two conceptions of relational egalitarianism according to which equal relations can be unjust. I want to do so to explore the further implications of committing to such a view. In that sense, the investigation should hopefully be useful in answering the question of whether we should prefer a conception of relational egalitarianism according to which equal relations can be unjust, or one in which they cannot. I cannot hope to fully settle this question in this paper, but I will point to some of the implications that follow if one supports a relational egalitarian view according to which equal relations can be unjust.

According to prominent relational egalitarians, equal political power is a particularly important constituent part of relating as equals. Kolodny (Reference Kolodny2014; see also Anderson Reference Anderson1999; Viehoff Reference Viehoff2014; Peña-Rangel Reference Peña-Rangel2022) makes this argument. He explains that equal political power is a particularly important constituent because political decisions (i) cannot usually be escaped at will; (ii) have final de facto authority; and (iii) involve force. (i) is important because ability to escape a decision at will makes a difference to what would otherwise be an unequal relation. Suppose Slave can exit their slave contract. If so, the relation between Master and Slave is not unequal in the way it would be if the slave could not escape the contract at will. Indeed, ‘the freer one is to exit what would otherwise be a relation of social inferiority, the less it seems a relation of social inferiority in the first place’ (Kolodny Reference Kolodny2014: 305). And, according to Kolodny, the thing with political decisions is that one cannot usually escape them at will (in the same way that one can escape non-political decisions).

With regard to (ii),

Equality when it comes to higher-order decisions eliminates, or at least moderates, the inequality which exists between Lord and Servant when it comes to lower-order decisions. Political decisions are important for relational equality because they are the highest-order decisions, as it were. They have final de facto authority. There are two implications we can draw from this. First, if there is inequality in influence over political decisions, this inequality cannot be moderated by equality in influence at a higher-order level (since there is no higher-order level). Second, if there is equality of influence with regard to higher-order decisions, this moderates the threat to relational equality posed by unequal influence over lower-order decisions. As Kolodny (Reference Kolodny2014: 306) says, ‘the threat to social equality that hierarchy would otherwise pose, one might say, is moderated by the fact that whatever hierarchy there may be is ultimately regulated or authorized from a standpoint of equality’.

With regard to (iii) – that political decisions involve force – Kolodny (Reference Kolodny2014: 307) notes that force is particularly important to how people relate ‘because, as a contingent matter, the power to use force is the ‘final’ power … [in the sense that it is] the power that usually determines the distribution of other powers’. That Smart is smarter than Brute is of no use to Smart if Brute can subject them to force: ‘One cannot reliably have superior powers of other kinds over others where they have superior powers to subject one to force’ (Kolodny Reference Kolodny2014: 307). This is to say that symmetry in the capacity to use force is important for relational equality. And political decisions characteristically involve (the threat of) force.

Let us assume Kolodny is right that, for these three reasons, equal political power is a particularly important constituent part of relating as equals.Footnote ^{28 ,} Footnote ²⁹ If so, deviating from equal political power must be a particularly important constituent part of relating as unequals. If Victim has more political power than Criminal, that inequality in political power is a particularly important constituent part of their unequal social relation. So this shows that if we want to turn an unjust equal relation into an unequal relation, the most effective way of doing so may be to grant less political power to the one who ought to stand as a social inferior, in casu Criminal. I am not saying that this is necessarily the only way to do so. But note that if Kolodny is right, any way of trying to make the equal relation unequal by intervening in the non-political domain would still leave an important equality intact: the equality in the higher-order domain that is constituted by equal political power. To avoid this, we would have to intervene in the political domain. This shows, interestingly, that, according to relational egalitarian views in which equal relations can be unjust, sometimes we have a pro tanto reason to grant those who should relate as unequals unequal political power.

One may object to this argument in the following way. Suppose we did grant Criminal less political power than Victim. Would that not constitute unjust unequal relations between Criminal and other people in society (who would have more political power than Criminal)? I have three responses to this objection. First, note that the argument points to a pro tanto reason to grant Criminal less political power than Victim. We have a pro tanto reason to grant Criminal less political power than Victim because they ought to relate as social unequals. But insofar as this would also lead to unequal relations between Criminal and other people in society, it may be that, all things considered, it would be better not to grant Criminal less political power than Victim. But this does not mean that the pro tanto reason goes away. It simply means that we have to weigh that reason against other reasons when making an all things considered judgement of how to distribute political power.

Second, we could imagine situations in which it would not constitute unjust unequal relations to grant Criminal less political power than Victim, for example if half of the people in society were criminals and the other half were victims. In this case, we might have both a pro tanto reason and an all things considered reason to grant Criminal (indeed all the criminals) less political power than Victim (indeed all the victims).

Third, perhaps the unequal relations between Criminal and other people in society that may be constituted by giving Criminal less political power are not objectionable. After all, Criminal has acted against the norms of relational equality in the community by treating Victim as their moral inferior. But if they have acted against the norms of relational equality, they might thereby have forfeited their relational egalitarian claims of standing as a social equal to others as well (at least for a while).Footnote ³⁰ If so, there may be other reasons for why Criminal should not relate in an unequal manner to other citizens, but it would not be unjust that they did so (cf. Stemplowska Reference Stemplowska, Knight and Stemplowska2011: 133). The upshot is that assuming a conception of relational egalitarianism in which equal relations can be unjust, we may sometimes, for reasons of justice, have to deviate from a system of ‘one person, one vote.’

Another implication of accounts in which equal relations may be unjust is the following: there may be situations in which relational egalitarian justice requires what is bad for individuals.Footnote ³¹ Sometimes unequal relations may be bad for the individuals involved, for example because they would get too caught up in the inequality or because it would lead to inappropriate thoughts of superiority – thoughts including too much superiority, as it were – for the one standing as a superior. In some of these cases, however, conceptions of relational egalitarianism in which equal relations may be unjust may still imply that an equal relation would be unjust. Note, first, that this is not a special characteristic of conceptions of relational egalitarianism according to which equal relations may be unjust. It is a consequence of non-consequentialist theories in general – as opposed to consequentialist theories – that they entail that justice may not always require what will have the best (personal) consequences. Second, conceptions of relational egalitarianism according to which equal relations cannot be unjust may have a similar implication. Such conceptions may in some situations imply that justice requires equal relations even if unequal relations would be better for the individuals involved. Think for instance of the parent-child example mentioned earlier. In that case, an unequal relationship may clearly be beneficial for the child. Moreover, as Lippert-Rasmussen (Reference Lippert-Rasmussen2018: 10) explains, Burke famously argued that

We must not forget that just as equal relations can sometimes be beneficial, even if unjust, unequal relations can also sometimes be beneficial, even if unjust. In this sense, conceptions of relational egalitarianism in which equal relations can be unjust are not different from those in which they cannot.

A third implication of conceptions of relational egalitarianism in which equal relations can be unjust is that they in one sense become similar to some conceptions of luck egalitarianism, to wit, in the sense that equality may be unjust. With regard to relational egalitarianism, in the sense that equal relations may be unjust. With regard to luck egalitarianism, in the sense that equal distributions may be unjust. Let me unfold this. Distributive theories of justice claim that justice has to do with distributions. According to the most prominent distributive theory of justice, luck egalitarianism, individuals’ distributive positions should reflect only their relative exercise of responsibility (Lippert-Rasmussen Reference Lippert-Rasmussen2015). Albertsen and Midtgaard (Reference Albertsen and Midtgaard2014) convincingly argue that luck-generated equalities and inequalities should be treated similarly: that luck egalitarianism in this sense is symmetric.Footnote ³² Just as it is unjust that Adam is worse off, distributions-wise, than Bertram through brute luck, it is unjust that Carr is as well off as Derek through brute luck. They introduce the following case to illustrate their view:

They argue that the resulting equality between Prudent and Lazy is unjust since ‘the effects of differential brute luck should consistently be neutralized’ (Albertsen and Midtgaard Reference Albertsen and Midtgaard2014: 338). Their main argument for this symmetrical view of luck egalitarianism is that it would be unfair not to neutralize the effects of brute luck in cases of equal distributions. When people, assuming a background of equal opportunities, exercise their responsibility to different degrees, they are not in a position in which they can justifiably demand that equalizing measures be set up (Albertsen and Midtgaard Reference Albertsen and Midtgaard2014: 337). As they explain, ‘the profound unfairness of equality in the presence of differential exercises of responsibility is that it amounts to asking some to bear the costs of others’ voluntary choices … it countenances exploitative cost displacement’ (Albertsen and Midtgaard Reference Albertsen and Midtgaard2014: 340). Note that there is a striking similarity between this argument and the relational egalitarian argument, put forward by Schmidt (Reference Schmidt2022), discussed earlier, to wit, the problem of costless reneging. Just as instances of costless reneging may lead to unjust equal relations, instances of exploitative cost displacement may lead to unjust equal distributions. In this sense, both relational egalitarianism and luck egalitarianism, and for strikingly similar reasons, entail that equality may be unjust.

This result is significant in itself. But it is also significant since it points to an interesting similarity between relational egalitarians and luck egalitarians. According to some relational egalitarians, relational egalitarianism and luck egalitarianism are competitors – indeed, are incompatible (of course, this could still be true even if there are important structural similarities between luck egalitarianism and relational egalitarianism; settling whether this is so goes beyond the scope of this paper). Take for instance the following remarks from Anderson, arguably the most prominent relational egalitarian:

As she ends by saying, there is disagreement at the fundamental level between luck and relational egalitarians: according to the former, distributions are fundamental; according to the latter, relations are fundamental.

Others have pushed back on Anderson’s argument that luck egalitarianism and relational egalitarianism are incompatible in this sense. They have argued for a pluralist conception of justice which includes both distributive and relational concerns (see e.g. Lippert-Rasmussen Reference Lippert-Rasmussen2018; Moles and Parr Reference Moles and Parr2019; Mulkeen Reference Mulkeen2020). The argument presented here – that both relational egalitarianism and luck egalitarianism, and for strikingly similar reasons, entail that equality may be unjust – may (not: necessarily do) provide further support for the argument that luck and relational egalitarianism, at least on some conceptions thereof, are not competitors nor incompatible.