By In Discussions, Normative Ethics Comments (30)

Politics, Philosophy & Economics online discussion: Juliana Bidadanure’s ‘Making Sense of Age- Group Justice with a précis by Paul Bou-Habib

Welcome to our online discussion  of Juliana’s Bidadanure’s recent publication in Politics Philosophy & Economics ‘Making Sense of Age-Group  Justice: A Time For Relational Equality?’. The paper is available through open access here.  Paul Bou-Habib is starting us off with a critical précis.  David Axelssen (LSE), Paula Casal (Pompea Fabra), Kasper Lippert-Rasmussen (Aarhus), Martin O’Neill (York) and Christian Schemmel (Manchester) will follow with their comments.  This  is going to be a very interesting exchange, so please join in!

Here’s Paul:

People who believe that our social and political institutions should aim at securing equal resources and opportunities for all citizens, or equal welfare for all citizens, must resolve a question about time. Should our social and political institutions aim for equality between the complete lives of citizens or between certain temporal parts of their lives, or both?

The view that favors equality only between the complete lives of citizens is known as complete lives egalitarianism. One powerful justification for this view is the possibility of compensation between different temporal parts of our lives. How we fare at any one moment in time can, at least sometimes, compensate for, or can be compensated for by, how we fare at other moments. This seems to suggest that our claims on each other shouldn’t always depend on how we fare at any one moment in time, but on how we fare over all of the moments that can compensate for each other during our lives, or, assuming that all the moments of our lives are of this kind, over our complete lives.

One objection to complete lives egalitarianism is that there may not be sufficient psychological continuity between all of the selves that live through the moments of a complete life. That objection should lead proponents of complete lives egalitarianism to specify their view so that it favors equality between the long (though not necessarily complete) periods of people’s lives during which they live through psychologically continuous temporal parts.

Juliana Bidadanure’s excellent article discusses a different objection to complete lives egalitarianism. This is the objection that it favours unacceptable inequalities between people at given moments in time. To illustrate the objection, she begins her article with Dennis McKerlie’s memorable example of a city block that contains a condominium complex and a retirement home (see page 7 of his book, Justice Between the Young and Old). Residents of the condominium are “middle aged, middle class, affluent, and happy.” The retirement home “is old, overcrowded, and badly managed. Its residents receive adequate medical care, but their situation offers them little dignity and little opportunity for anything that approaches happiness.” Suppose the old once lived in the condominium and the young will live in the retirement home (and suppose other things are equal). If we believe we must compare people over their complete lives, we might have to conclude that there is nothing troubling about the current inequality between the young and the old. Yet, intuitively, it does seem troubling.

Bidadanure’s main claim (p. 235) is that a certain kind of egalitarian can account for what is troubling about the simultaneous inequality that is illustrated in McKerlie’s case only if that kind of egalitarian makes a specific assumption. The kind of egalitarian Bidadanure has in mind is a comparative egalitarian – that is, anyone who believes inequality is bad because it involves a person’s being worse off than others. (Non-comparative egalitarians believe we should support whoever is worse off, not because inequality, as such, is bad, but because the moral importance of benefitting a person is greater the worse off she is in absolute terms. This view is also known as prioritarianism.)

Bidadanure argues that comparative egalitarians can properly account for what’s troubling about McKerlie’s example (and other examples of age group inequality one could think of) only if they assume that social and political institutions must prevent certain morally problematic relationships from arising between people, such as relationships of domination, marginalization and status hierarchy.

To justify this claim, Bidadanure needs to show two things. First, that this assumption about the importance of preventing problematic relationships (let’s call it, the relational assumption) does indeed enable comparative egalitarians to explain what’s troubling about McKerlie’s example. She shows this by pointing out that the old in his example are spatially segregated from the young, and that spatial segregation is often a prelude to marginalization. Their relative poverty can also be associated, she explains, with unequal status and unequal levels of respect. Comparative egalitarians can explain what’s worrying about the age group inequality in McKerlie’s example by pointing to these aspects of it.

Secondly, she needs to show that comparative egalitarians cannot explain what’s troubling with McKerlie’s example on the basis of any other assumptions (i.e. apart from the relational assumption). She considers several alternatives. The first is one that McKerlie examines and rejects in Chapter 4 of his book. This is an assumption that inequality between simultaneous segments of different people’s lives is bad in itself. What matters isn’t only that people have equal amounts (of whatever it is we take to be the relevant equalisandum) over their complete lives, but that they have equal amounts of it, also, over simultaneous segments of their lives. Bidadanure rejects this proposal on the ground that there is no non-arbitrary way in which to demarcate the lengths of the segments of different people’s lives that must be simultaneously equal. Is it each month of their lives, each year, each decade, or some other period?

A second alternative to relying on the relational assumption is to rely on what Axel Gosseries has called a continuous sufficiency principle – i.e. a principle that requires that all persons continuously remain above some minimum threshold throughout their lives. This alternative, as Bidadanure acknowledges, is also noted by Paula Casal, and I have also argued that a variant of continuous sufficiency can be grounded in the idea of the dignity of persons. Bidadanure points out that the only version of the continuous sufficiency principle that is available to comparative egalitarians is one that involves the relational assumption that she says is indispensable. If the continuous minimum threshold is defined in absolute terms – i.e. in non-comparative terms – it is unavailable to comparative egalitarians. If it is defined in comparative terms, so that the “minimum threshold” is set as a proportion of what the better off have at any given moment – then McKerlie’s example may well meet that requirement: what the old have in his example, she says, “is within the normal opportunity range of the elderly in their society, and they have had the opportunity to live a normal lifespan” (p. 250). The only remaining version of the continuous sufficiency principle is that the minimum threshold is set as a “relational threshold” – i.e. a threshold that protects people against certain problematic relationships. But that, of course, is just the relational assumption that Bidadanure says comparative egalitarians must endorse.

Suppose Bidadanure’s central claim is right. Why does it matter? Apart from telling us about the attitude comparative egalitarians should take to age group inequality, she points out that her claim has significance for a broader debate between “relational egalitarianism” and “distributive egalitarianism”. Relational egalitarianism is the view that we should care about equality or inequality because of their relevance for the emergence of certain problematic relationships between citizens. (By “distributive egalitarianism” I think Bidadanure means the same thing as what I above called “comparative egalitarianism”.) Bidadanure believes that if, as she has argued, distributive (or comparative) egalitarians must endorse a relational assumption, then, two things follow. First, relational egalitarianism cannot be reduced to distributive egalitarianism in some way – i.e. it is a genuinely distinct form of egalitarianism. (I am not entirely sure why this is an implication of her claim but I will explain the reason that I think Bidadanure has in mind in the next paragraph.) The second is that distributive (or comparative) egalitarianism is insufficient as a full statement of the kinds of commitments egalitarians should endorse. Bidadanure doesn’t believe that egalitarians should dismiss distributive egalitarianism. She thinks there are good reasons for caring about how certain important goods are distributed between citizens. So, the overall implication she takes her main claim to have is that we should endorse a hybrid kind of egalitarianism: “we should supplement the distributive view with a relational component” (p. 238). In other words, distributive egalitarians should combine the view that some equalisandum should be distributed equally over the complete lives of citizens, with a relational assumption that requires that citizens be protected, throughout their lives, from various problematic relationships.

Bidadanure’s article contains several, complex arguments. I won’t be able to discuss all of them here. Instead, I will raise two objections. The first objection, which I will mention only briefly, is about the broader implications of her main claim for the relationship between distributive and relational egalitarianism. Distributive egalitarians might argue that relational egalitarian concerns can be re-written as distributive concerns so long as we specify the equalisandum that everyone must enjoy equally in a complex way so that it includes the relational goods of non-domination, inclusion and status. But – and here is what I said earlier I think Bidadanure has in mind when she says that relational egalitarianism cannot be reduced to distributive egalitarianism – the core concern of relational egalitarianism is that people should enjoy these relational goods continuously throughout their lives, not that they should enjoy equal aggregate amounts of them during their lives (whatever that would mean). However, a continuous concern for equality between people is just simultaneous equality, which, as we noted earlier, is implausible. So, the core concern of relational egalitarianism cannot be re-described in distributive terms.

I am not sure this argument works because distributive egalitarians might argue that the relevant equalisandum that should be equalized between simultaneous segments of people’s lives should be restricted to the goods of non-domination, inclusion and status and that when it is restricted in that way, simultaneous equality becomes much more plausible. This argument allows distributive egalitarians to maintain that relational egalitarianism reduces to distributive egalitarianism and that distributive egalitarianism is sufficient on its own.

I want to spend more time discussing a second objection, which is about Bidadanure’s hybrid view (complete lives egalitarianism plus a relational assumption). The hybrid view doesn’t seem to account for all cases of troubling age group inequality. The elderly face hardships that don’t have anything to do with the kinds of problematic relationships to which relational egalitarianism is sensitive. Some elderly people face serious illness and severe physical pain. Some struggle far more on a daily basis under financial strain. Yet, even if the young will face those same hardships, and we assume their complete lives are equal, it still seems wrong to ignore them. One could try to make the case that those hardships always translate into problematic relationships with others, but that that is true in all cases seems a stretch. And besides, even if one could show, in a given case, that illness brings, say, marginalisation, what the elderly reasonably want help with isn’t just their marginalisation, but their illness. We don’t necessarily do the right thing by an elderly person whose illness has caused her to be marginalised, if we end her marginalisation, while ignoring her illness (even if she will have had as good a complete life as us). The problem, in other words, is that it is difficult to see how we can explain all troubling cases of age group inequality with a hybrid view that combines complete lives egalitarianism only with a relational assumption. Bidadanure may be right that a relational assumption should be included in our deliberations about what different age groups owe to each other, but the hybrid view is still missing something.

I think what’s missing is a non-comparative principle of one kind or another that explains why we have reason to worry about people’s conditions at specific times independently of the quality of their whole lives, and independently of whatever relationships they share with others.

Consider two possibilities. The first, which Bidadanure does not discuss, is endorsed by McKerlie in Chapter 5 of his book. McKerlie believes we should endorse prioritarianism – i.e. the view that the moral importance of assigning benefits to people depends on their level of welfare, in absolute terms, prior to their receiving those benefits. More specifically, McKerlie thinks we should endorse two kinds of prioritarianism: complete lives prioritarianism, which tells us to prioritise support for persons with worse whole lives, and time-specific prioritarianism, which tells us to prioritise support for persons who are worse off at specific times. He believes we should try to strike some sort of balance between these two kinds of prioritarianism in cases where they conflict.

Time-specific prioritarianism provides one possible explanation for why the inequality between the young and the old in McKerlie’s example is troubling. It is troubling not for the comparative reason that the old are so much worse off than the young during that specific time, or for any relational reason, but simply because they are so much worse off at that time than they ought to be.

Still, I don’t think we should endorse time-specific prioritarianism. If, as prioritarians believe, the basis for a person’s claims is his level of welfare, then it is difficult to see why it is his welfare at specific moments that should matter, as opposed to his welfare over his complete life. Suppose someone was better off than me in the past, and that that situation is exactly reversed right now. His past superiority in welfare seems to compensate him for his being worse off than me now, and there does not seem to be a reason to give him priority when allocating available benefits between us.

I think a more promising possibility is a non-comparative principle that is grounded in the dignity of persons. A person’s dignity is a separate reason for being concerned about her from her welfare. So, it can tell us to show concern for her at specific times independently of whatever levels of welfare she has enjoyed at other moments in life. Furthermore, the kind of concern it tells us to have for her is not restricted to the kinds of relationships she shares with others at that specific time. People need to be treated with dignity, but they also need to be able to live with dignity, and we are unable to live with dignity when we are incapacitated, in various ways, by illness or pain. Dignity thus grounds the kind of non-comparative principle we need in order to account for all troubling cases of age group inequality.

But what about Bidadanure’s objection, that comparative egalitarians, who care about inequality as such, cannot coherently endorse a non-comparative principle? That objection does not seem decisive to me. We can be comparativists at the level of our complete lives while endorsing the idea that people should be treated in certain ways continuously throughout their lives. We can coherently say that inequality at the level of our complete lives is bad as such, and that, while inequality is not bad as such at specific times, it is bad that people fall below some non-comparative threshold at specific times. And we can say this compatibly with also saying that it is important that people be protected against various problematic relationships throughout their lives. The view that emerges would be even more hybrid than the hybrid view Bidadanure defends. But, as Bidadanure shows us so well in her article, there is nothing wrong with hybridity as such. The concerns that preoccupy different egalitarians may not always conflict with each other. They may even be indispensable parts of a complete view.

Love this idea? Nominate it for the Annual PEA Soup Awards!

, , , ,

30 Responses to Politics, Philosophy & Economics online discussion: Juliana Bidadanure’s ‘Making Sense of Age- Group Justice with a précis by Paul Bou-Habib

  1. Juliana’s paper is really excellent and stimulating. Among the big issues that it raises is whether egalitarian justice has a distributive as well as a relational component. I think a hybrid view roughly of the kind Juliana proposes might well be attractive. However, in this opening comment (which I post before dashing off to do some teaching which will keep me busy until mid-afternoon) I want to mention three comparatively minor points.

    First, Juliana raises some doubts about previous attempts to characterize the distinction between distributive and relational egalitarianism. Specifically, she questions the following attempt (by me) at offering a distinction between the two positions: ‘The latter [distributive egalitarians] take the distribution of goods to matter, from the point of view of justice, independently of its effect on social relations. The former [relational egalitarians], by contrast, contend that distribution matters only instrumentally in virtue of its impact on social relations and the degree to which these are suitably egalitarian’ (p. 238). She finds this cut problematic, because relational egalitarians care too about distributions as constitutive of social relations or as having a certain expressive value, which might be (in)compatible with social relations of a suitable egalitarian nature.

    Here are two responses: a concessive and a non-concessive one. First, exegetically speaking it is noteworthy that Elizabeth Anderson does write something along the lines Juliana suggests. So from an exegetical point my cut might not be a happy one. But, and that is my non-concessive point, if distributions of a certain sort are constitutive, in part, of suitably egalitarian social relations, the distance between the distributive and relational egalitarianism becomes much smaller. For instance, relational egalitarians will have to engage in the ‘equality of what?’-discussion, since doing do so is necessary for identifying the social relations required by relational justice. Yet, this is a discussion Anderson and Samuel Scheffler normally dismiss as misconstrued. Also, and more importantly, many of the objections to the distributive paradigm that social relational egalitarians put forward would seem to apply to relational egalitarianism, if that ideal has a distributive component as a constitutive component.

    Second, while Juliana mentions the possibility of a complete lives version of relational egalitarianism which would allow that X’s dominating Y at t1 might counterbalance or at least mitigate Y’s dominating X at t2, her view is that the natural form of relational egalitarianism is time-relative: ‘Phases of domination, marginalization, or segregation cannot be thought to cancel out diachronically. These phases are nonderivatively offensive: they matter to us precisely and directly for the relationships they contain or are likely to contain’ (p. 246). If this were not so, it is unclear why McKerlie’s condominium case motivates a relational egalitarian element – perhaps as opposed to a time-relative relational element – in the account of egalitarian justice. At least, this is so on the counterfactual assumption that we all grow old and thus all come to be dominated, marginalized, and segregated in old age (to an equal degree). However, I find it intuitively clearly that, at the least, reverse direction inequalities in social relations can mitigate one another as it were. For suppose X dominates Y for the first 40 years of their lives and the relation of domination is then reversed for the next 40 years. Even if we think this is not perfect from the perspective of relational egalitarianism, at least it is less bad than if X dominated Y for their entire lives. Would you not prefer this situation about rather than one in which X dominates Y for their entire lives even if everything else is equal and if these were your only two options?

    Could one defend a more ambitious thesis, i.e. that complete lives relational egalitarianism is true? That depends on why we think egalitarian social relations matter. Relational egalitarians might not agree about this and, as Patrick Tomlin has argued, are often not very clear about why egalitarians social relations are good. At least some relational egalitarian views might pull in the direction of a complete lives version. Take for instance David Miller’s view, which is heavily inspired by Michael Walzer’s idea about different spheres of justice. On this view, there is nothing objectionable about large inequalities within a particular sphere of justice provided this inequality is counterbalanced by inequalities in the reverse direction in other spheres. To use a somewhat pathetic example (‘pathetic’ in more than one way): there nothing objectionable about your being much richer than I am provided that there are other spheres of justice where I do much better than you. Roughly, I can still think of myself as your equal even though you do much better than I do in the economic sphere, because there are other spheres where I do better than you. If we care about social relations, because we care about people being able to justifiably see themselves as equals, this suggests that relational egalitarianism might well take the complete lives form. Specifically about McKerlie’s case: young people will not think of themselves as superior to old people, because they know that some day they will be in their shoes and old people will not think of themselves as inferior, since they knew that once they were in young people’s (much nicer) shoes. Obviously much more needs to be said at this point and it is one of the many virtues of Juliana’s piece that it brings out the need for clarifying the relational ideal of equality with regard to time.

    The third and last point I want to make in this opening remark concerns stereotyping. ‘To provide a better illustration of what it means for an age group to be treated in a relationally inegalitarian fashion’ Juliana quotes some surveys documenting how young people – people in their 20’s – are stereotyped as being associated with ’crime, deviance, delinquency, drug and alcohol problems, sexual promiscuity and general disorderliness’ (p. 247). On my view — and I am not sure that I am saying anything here which conflicts with what Juliana says — the relational ideal of equality cannot explain the wrongness of stereotyping. So to return to Juliana’s case: I suppose there are also positive stereotypes about young people, e.g., that they are energetic and more inclined than old people to think in novel ways and come up with new solutions to problems. Similarly, there are negative and positive stereotypes of old people. In principle at least, we could imagine a situation where young and old people relate to one another as equals despite stereotyping each other (as well as themselves), e.g., members of each group might think ‘Admittedly, qua my being young/old I am not so good at this and this, but s/he qua being old/young is not so good as that and that. We’re different, but none of us is better than the other all things considered’. If we think that there is a problem about this situation, this problem might not be one that has anything to do with egalitarian justice – whether distributive or relational — at all. Rather, we might object to stereotypes on other grounds, e.g., that people who stereotype are unable to see each other as what they really are or the like.

    Admittedly, the situation I am speaking to is one in which people are stereotyped to an equal degree across age groups, where the one Juliana is speaking to is one in which people are stereotyped to an unequal degree. However, perhaps the right response to someone pushing this line of response is to say that not being (negatively) stereotyped is an advantage – generally at least — and that there is no reason why distributive egalitarians might not include this advantage in their relevant equalisandum. This, however, opens up another and much larger debate about whether the concerns of relational egalitarians – or at least some of them – can be captured within the distributive ideal of equality.

  2. Anca Gheaus says:

    I’d like to congratulate Juliana on this important paper, which was a pleasure to read. And to support – and complement – one of the worries raised by Paul on whether relational egalitarianism can be expressed as/reduced to a form of distributive egalitarianism. Paul writes:
    “distributive egalitarians might argue that the relevant equalisandum that should be equalized between simultaneous segments of people’s lives should be restricted to the goods of non-domination, inclusion and status and that when it is restricted in that way, simultaneous equality becomes much more plausible.”
    I think Paul is right that, if distributive egalitarians are concerned with the goods of non-domination, inclusion and status then simultaneous equality doesn’t seem problematic in the ways in which it would be with a different equalisandum. Moreover, enjoying these goods may require some kind of simultaneous equality. This is because goods such as non-domination and inclusion may display a non-excludability feature, such that it is not possible for some individuals in a society to enjoy it while others don’t. If some individuals in my society are dominated or excluded then, even if *I* am not, I fail to enjoy the good of living in a non-dominating and inclusive society. The precise identification of the relational goods in question is very important for making a convincing case that relational egalitarianism can be expressed as/reduced to distributive egalitarianism. It is of course possible for some people to have more power or higher social status than others – power and status are excludable. If we take turns dominating and excluding each other we may end up with equal amounts of power and status over our complete lives, and zero equality of status and power. But it’s more likely that the goods to be enjoyed are not only power or status as such, but the instrumental and non-instrumental goods of *enough equality* of power and status to preclude domination and marginalisation. Distributive egalitarians with this agenda will need to ensure the kind of continuous equality endorsed by relational egalitarians.

  3. In her highly illuminating paper, “Making Sense of Age-Group Justice”, Juliana argues that getting a prosper grasp of justice between generations requires a hybrid view encompassing both distributive, prudential, and, most controversially, relational considerations. Juliana argues, as this last component, that a just society must also ensure that different generations living at the same time relate to each other in an egalitarian manner – i.e. that the young are not subject to domination and stigmatized or that the elderly are not excluded and marginalized. I think Juliana is right that a hybrid view is the right way to go and, especially and pace Paul’s introductory precis and Anca’s comment, that a purely distributive view cannot capture all of what can be intergenerationally wrong from the point of view of justice. But I think that they are right in asking Juliana to say more about why this is the case and what it is, precisely, that cannot be redescribed in distributive terms.

    Here is one thing that I think distributive views overlook and, so, maybe this can serve as an attempt at a defense of what Juliana calls “the irreducibility thesis” – the idea that relational concerns cannot be reduced to a distribuendum / equalisandum (something to be distributed/equalized) – I would very much like to hear her thoughts on the matter, though.

    Both Paul and Anca (and Kasper mentions this possibility too in his recent book, Luck Egalitarianism – and Anca elaborates more on her view in her article, “Hikers in flip-flops”) mention the possibility of redescribing the relational component of egalitarian justice in terms of some stuff that must be distributed or equalized. Paul writes “distributive egalitarians might argue that the relevant equalisandum that should be equalized between simultaneous segments of people’s lives should be restricted to the goods of non-domination, inclusion and status and that when it is restricted in that way, simultaneous equality becomes much more plausible” and Anca follows up “If some individuals in my society are dominated or excluded then, even if *I* am not, I fail to enjoy the good of living in a non-dominating and inclusive society.”

    But I think Juliana is right to insist that there is a separate relational component that cannot be captured in this way. Let us set aside for the moment that it is an awkward and unnatural way of describing relationships to speak of them in terms of goods; as something that one individual “has” to different degrees (Imagine the complaint “How come I have 4 non-domination and she gets to have 6?”) – it looks to me like rendering something unrecognisable just to make it more easily measurable and comparable. But setting that aside (which, apparently, I had a harder time doing than I thought), I think a society is worse if some people relate to others in a disrespectful way, if they see another group as inferior, if they are dominating, or if they have a stigmatizing attitude towards others. And I don’t think this is reducible to the being related to in a disrespectful way, the being dominated, the being viewed as inferior, etc. There is something bad, inegalitarian, and unjust about a society in which such intentions, attitudes, and outlooks prevail. And I think this is bad regardless of how it impacts people’s shares and opportunities (although, of course, I think it is worse if it does).

    But perhaps this is not what Juliana means – or perhaps (which seems likely to me) there is more to the relational component?

  4. One very nice aspect of Paul’s critical précis of Juliana’s excellent paper is that, just as Anca points out here, it brings into clear focus the issue of whether a relational egalitarian view can be expressed in the language of distributive egalitarianism. Distributive egalitarians believe that a commitment to equality as a political value involves believing that individuals should have an equal amount of some good or set of goods, where this good or goods are what G. A. Cohen called an ‘equilisandum’. Much ink has been spilled by distributive egalitarians in attempts to spell out exactly what this equilisandum might be. Relational egalitarians on the other hand — writers such as Scanlon, Anderson, Scheffler and Young — hold that the egalitarian concern for distributive outcomes is derivative of a deeper and underlying concern with how people relate to each other. Relational egalitarians hold that what is central to the value of equality is a vision of society in which people see themselves and others as equals, and relate to each other under conditions marked by equal standing and status, and without objectionable forms of oppression or domination. On this relational view, distributive outcomes matter, but their significance is derivative. In the possibly inelegant terms I used in my 2008 paper, “What Should Egalitarians Believe?”, the relational egalitarian view is a form of ‘non-intrinsic egalitarianism’, which sees the value of reduced socioeconomic inequality as being explained derivatively, with reference to this underlying ideal of social relations.

    I am somewhat uneasy about Paul’s terminology of ‘comparative egalitarianism’, which he defines as involving a belief that “inequality is bad because it involves a person’s being worse off than others”. My unease turning in one part on the connotations of alternative egalitarian views being “noncomparative”, and in another part on his claim that “noncomparative” egalitarianism is equivalent to prioritarianism. It seems to me that prioritarianism covers a broader range of possible views, marked by a shared belief in what one might call “the diminishing marginal significance of greater benefits”, and that such views will only be contingently egalitarian, and would not mandate in all cases that benefits must go to the worse-off, for example where a proportionately larger benefit could be given to others. I shall assume, though, that despite this unease, Paul’s category marks out a useful class of views, perhaps equivalent to those falling under Derek Parfit’s category of ‘telic egalitarianism’.

    At any rate, what matters here is that the relational egalitarian view is not, on Paul’s terms, a variety of ‘comparative egalitarianism’, because it does not hold that (distributive) inequality is bad *because* it involves x being worse-off than y in terms of some good or bundle of goods, but instead holds that (distributive) inequality is bad because it undermines egalitarian social relations (where ‘undermines’ here might be understood causally or expressively, or in some other derivative sense).

    The main point that I want to make in these remarks is not that we need take a stand on whether the relational egalitarian view can somehow be recast or reconfigured in distributive egalitarian terms, where the list of goods comprising the equilisandum now includes non-domination (conceived as an attribute that an individual has, rather than a feature of relationships per se), status or inclusion. My point is that, even if we managed to come up with some extensionally equivalent reformulation of the relational egalitarian view into the language of distributive egalitarianism, it is not remotely clear why we should want to do so. For what would have driven our attempt to expand the set of goods within the equilisandum would have been our prior acceptance of the relational egalitarian insight regarding what really matters about inequality. The task of re-expressing the simple central claims of relational egalitarianism in some notationally equivalent distributive language might be a task that presents some interesting intellectual challenges, much as it was no doubt an interesting intellectual challenge for Ptolemaic astronomers to add a sufficient profusion of epicycles in order to maintain a geocentric model of the workings of the solar system. But once distributive egalitarians are engaged in such an effort — and here, I think, I agree with Juliana — they have already granted to the relational egalitarian a substantive point of normative concern with regard to how we should best understand the value of equality.

    I’ll close with a question for Juliana. One way of reading her paper is as an immanent critique of distributive egalitarianism, using the example of justice between age-groups to show how (non-epicyclic forms of) distributive egalitarianism lacks the resources to deal with troubling cases of inequality between groups. Perhaps due to her starting point, which begins where the distributive egalitarian view begins, she ends up travelling only so far as supporting a ‘hybrid view’, under which any plausible distributive view would need at least to supplement itself by accepting central relational egalitarian claims about the value of equality. This seems to me to be right, but only insofar as we might have independent good reasons to start from a distributive egalitarian starting point. (And, as I’ve suggested above, I think it’s of only secondary significance whether the distributive egalitarian accepts this modification to their view via the healthy mechanism of straightforwardly conceding ground to the relational egalitarian, or the more surreptitious mechanism of epicyclic redescription. 🙂 ) But I’m left with a question as to why Juliana should start from that distributive position at all. Why not go further, and accept the view, which seems to me to become less assailable the longer one considers it, that there’s actually nothing to be said for the view that Paul describes as ‘comparative egalitarianism’. Why not accept that, while it is difficult to overestimate the significance of overcoming socioeconomic inequality, this significance can be explained entirely via a mix of ‘non-intrinsic’ and relational egalitarian concerns, deontic concerns with the fairness of collective institutions, and nonegalitarian humanitarian concerns (under which I’d include Paul’s remarks about dignity)? Why bother with a ‘hybrid’ stop-gap, other than as a polite way of leading distributive egalitarians gently away from their misguided attachment to a misleading conceptual picture of equality and its significance? To put my question to Juliana as concisely as possible: why stop here?

  5. HI all, I’ve just seen that I had initially posted this on the wrong page, so here again:
    I really enjoyed reading Juliana’s excellent paper again, as well as Paul’s really helpful precis. I’d have two points for discussion:
    1. The first concerns the significance of Juliana’s argument about the importance of caring about some forms of synchronic relationship (in)equality for the general debate between relational and distributive egalitarians. Here, Juliana’s claim is that her argument helps to see how the two camps come apart not only in marginal cases, like the luxury car example (p. 237). But it seems to me that if we understand the difference between these two general approaches as mainly about “different kinds of reasons to care about inequalities” (p. 238), as Juliana proposes to do, with reference to Scanlon, then we can quite easily find lots of different non-marginal cases where the two camps come apart. For example, if relational egalitarians care mainly about respect/recognition/non-domination/status, and distributive egalitarians care mainly about fairness, especially of a luck egalitarian sort, then they come apart in disability cases: if relational egalitarians care about non-marginalisation, non-domination, and social inclusion, then they will strictly require these for the disabled, but they will not compensate them for additional losses in well-being, which might be very significant (perhaps even more so if one becomes disabled during the course of your life). An example in the opposite direction: if you responsibly make very bad choices leaving you badly off, then you might also lose a lot of opportunities for political influence (even if some form of sufficiency is still guaranteed), but most relational egalitarians will want to make sure that you continuously retain some form of equal substantive opportunity for political influence. So they will, to that extent, in effect compensate such persons for the effects of their own bad choices. If this how we understand the difference between the two approaches, Juliana’s argument about synchronic relationship equality is another illustration of this difference in underlying reasons, or values, but there seems to be no difficulty in coming up with other non-marginal examples for it.
    If, on the other hand, we focus on whether claims typically made by one camp can be formulated in the language typically employed by the other (treating people a certain way vs. making sure they get equal amounts of some good), then we might be able to formulate concerns about non-domination, status, etc. in distributive language, and that might work for the example of synchronic age group inequalities, too (whether or not that coheres well with other concerns distributive egalitarians typically have). I take it that this relates to the first objection to Juliana that Paul raises in his precis. I am not sure whether formulating these concerns in this way really does work, for non-domination and status; it obviously depends on how we spell these out in detail. But, in any case, the point would seem to be once more a general one.
    So my question to Juliana here would be whether she would be happy to accept that the case of synchronic age group inequalities is merely one of many possible illustrations of the difference between distributive and relational egalitarianism, or whether she would hold that there are still reasons to think that this example is somehow of special importance for this general debate.
    2. The second point concerns a possible comeback of the distributive “simultaneous segments”-approach. You show beautifully (pp. 242ff) how overall inequality here arbitrarily depends on the length of the relevant life segments chosen. But I wonder whether some arbitrariness in numbers might be acceptable as long as we have good reasons to focus on a certain way to carve up the life segments, based on their content. Take the example of reaching adulthood: Choosing a certain age for the onset of adulthood is quite arbitrary (17, 18, 21?). But the category of adulthood itself is not arbitrary. Adults really differ from children. So, if the segments chosen corresponded to certain relatively discrete phases of life that typically exhibit some different characteristics, and where different capacities come to fruition (when these phases go well), might that allay the arbitrariness problem? You refer to “children, young people, formed adults, and the elderly” as roughly the relevant age groups (p. 240); if we took these as the generally significant stages of life, this could be a reason to focus on their typical duration (which, in this case, is not the same for all four stages), and not on “smaller units of time” (p. 243), or bigger ones, when assessing synchronic inequalities. I am wondering whether there is any promise in such an approach.
    If so, it might actually also inform relational approaches on the question of which kinds of relationships should be egalitarian, in and across these different life phases. Juliana refers of (negative) stereotyping of the young that clearly seems objectionable (p. 247). But could there be reasons to think that certain relational inequalities – not domination, but some inequalities of status, perhaps – are acceptable between age groups? Example: generally, even in an egalitarian society, formed adults might hold more of the more desirable positions commanding more status in society, based (when things go well) on the capacities they typically exhibit. Children and the young aspire to filling them when their time comes, and the elderly look back contentedly on having filled them. This way of making the point might be a bit idyllic, perhaps, and there might well be many objections to the proposal (it might not be very liberal, for once), but I wonder whether it could be an avenue worth pursuing. Has it been tried and rejected in the literature (I am one of the egalitarians whom Juliana rightly chides for not having paid enough attention to intergenerational justice, so far)?

  6. Anca Gheaus says:

    Does it make conceptual sense to express the claims of relational egalitarianism in the language of distributive justice (with a properly identified equalisandum)? If so, is this a worthwhile enterprise – or just some kind of sophisticated intellectual game with no real import?

    David asks the first question, noting it is awkward and unnatural to describe relationships in terms of goods. I agree; but the “reductivist” move doesn’t aim to describe relationships in terms of goods, only to describe what is valuable about relationships in terms of goods. (Or, more precisely, what is the value of relationships to individuals.) We do this all the time, successfully, when we say that we enjoy somebody’s friendship or when people who live in more egalitarian societies tell you they enjoy the moral and practical effects of the egalitarian ethos. Now, perhaps it is wrong to pursue relationships with the primary aim of enjoying the goods they generate, and relationships cannot be reduced to these goods. But even so, if you want to know what it is about (egalitarian) relationships that makes them relevant to justice it seems to me that it’s entirely appropriate to consider the goods they generate.

    Martin raises the second question. Assume it turned out that concern with fairness (the distributive egalitarian’s core value) does not make you blind to the importance of egalitarian relationships (the relational egalitarian’s core value); quite the opposite, once you see how egalitarian relationships contribute to people’s lives going well, your concern with fairness makes you more responsive to these relationships. Wouldn’t this be a very good piece of news for the prospects of building an ecumenic egalitarian agenda? (Of the kind, perhaps, that Kasper advocates in recent work.)

  7. Kasper asks in his comment “For suppose X dominates Y for the first 40 years of their lives and the relation of domination is then reversed for the next 40 years. Even if we think this is not perfect from the perspective of relational egalitarianism, at least it is less bad than if X dominated Y for their entire lives. Would you not prefer this situation about rather than one in which X dominates Y for their entire lives even if everything else is equal and if these were your only two options?” He goes on to suggest that complete lives relational egalitarianism might be true – that we might well imagine that X and Y relate to each other as equals knowing that they will switch places at a later time.

    Maybe Kasper is right here that egalitarian relations could exist in this way. I find it hard to imagine in anything but the abstract, though, and this might actually highlight another important difference between (most actually existing) relational and distributive views. On an abstract level, we might be able to imagine relations, in which people are able to not be bothered or affected by being dominated for 40 years if they know that they will be the dominators for the next 40. But only in the abstract. In real life, 40 years of domination and unequal relations will have a very fundamental impact on people’s lives (presumably, both on the dominated and the dominators).

    Distributive theorists (often, though not always) seek a form of abstract precision, which isolates different elements and values – for example, by redescribing relationships in terms of goods and holdings in order to better be able to compare how people are faring. In doing so, they achieve one kind of precision, which maybe we could call abstract or fundamental precision. In this way, you become able to say more precisely who is worse off and in what way (what equalisandum they are lacking).

    Relational theorists (often, though not always) might say that this fundamental precision comes at a cost. It becomes less easily recognisable to the people suffering from the relevant injustices; in abstracting away from the particulars, the inequalities become something different, something simplified and something that no longer looks like what makes relations of domination, oppression, or marginalization wrongful. They aim at a different kind of precision; one that tries to expose the experience of inequality to the reader, so she is better able to understand how it is to be in it. They aim at diagnosing this experience because they are (often, and most explicitly in the case of Elizabeth Anderson and Jonathan Wolff) concerned with making argumentative, moral tools for actual political agents seeking to overcome these inequalities and injustices. I think an important source of disagreement can be found in this difference of aim and approach.

  8. Below follows a few replies to David’s interesting thoughts. 1) David discusses my the changing places scenario in relation to a whole lives version of relational egalitarianism. He notes that “In real life, 40 years of domination and unequal relations will have a very fundamental impact on people’s lives (presumably, both on the dominated and the dominators).” I think he is right, but even so this is not an objection to what I call the whole lives version of relational egalitarianism. At least, it is not if egalitarian justice is not all one should care about, but one should also care about welfare independently of egalitarian justice. If so, there are strong welfare-based reasons against the scenario, but that it is not saying that it involves any defects from the relational egalitarian point of view.
    2) The language of goods is not precise per se and there could be goods where, at most, we can say that people are roughly equal in terms of that good. Some relations that some relational egalitarians care can be quite precise (or, perhaps better, it can be determined in a very precise way whether they obtain or not), e.g., the relation of political equality embodied in the one-person-one-vote principle.
    3) I suspect that some distributive egalitarians would say that they too try to make sense of the complaints of justice voiced by real-life egalitarians. For instance, I suspect G. A. Cohen would think (rightly so in my view) that his critique of the incentives argument elucidates the reservation many real-life egalitarians have felt, but probably not been able to articulate very well (or at least not as well as Cohen does), in response to suggestions to the effect that they should endorse tax cuts for rich people.

  9. Paula Casal says:

    Hola todos and thanks for an engrossing discussion! (Sorry I was away preparing a student for his viva).

    I can see many people being attracted to Juliana’s hybrid view, and some real cases make it particularly persuasive. One case is an African society studied by Mary Douglas in The Lele of the Kasai. Lele men are equal from a total life perspective, but very stratified by age groups. They have long childhoods, short working lives, and an early, pampered retirement. For relational egalitarians, however, Lele society is hell ─ even leaving gender aside. There are glaring differences in status between age groups, and retired Lele openly boss others around. Clearly, distributive total-life egalitarianism is not enough, but what else is needed?

    Lele society suggests it is relational egalitarianism, but consider this other case: in what seemed to me a justified violation of total-life equality, I once momentarily left in a Tenerife hospital, a friend, then stable, who had nearly drowned that morning. I did so to bring drinks to a well-off German octogenarian local nurses could not understand. She was at no risk of domination, and the movements of her jewel-laden arm matched her authoritarian “Ich habe Durst!!” tone. But she was thirsty, and thus below sufficiency for once. I couldn’t ignore her need. Juliana’s principle is insufficient to justify my action as the lady’s status was intact, and unnecessary, as one can appeal to sufficiency instead, as Paul notes above. If so, is it clear we need such relational egalitarian principle elsewhere?

    A principle we clearly need is one that distributes resources well within lives, not leaving too little for retirement, as in McKerlie’s Unequal City, or too much, as the Lele do. (In fact, Douglas’s thesis was that the Lele’s intra-life distribution explained their poverty and backwardness). But once the intra-life distribution is inefficiency and insufficiency-free, do we need a more complex hybrid?

    Juliana may grant that relational equality is insufficient to justify some of our convictions but still believe it is necessary because its absence puts people at risk of undesirable relations. Noting risks, however, may not justify adding new principles. Wilkinson and Pickett have identified over a dozen undesirable effects of inequality, and we need not invent a principle to condemn each.

    Juliana may also note these risks to stress the coherence of her hybrid, whilst admitting its components may conflict. Indeed. Simultaneous relational equality may require sacrificing (i) an inefficiency/insufficiency-free intra-life distribution, perhaps giving the elderly of Unequal City worse overall lives or (ii) total-life equality. For example, if a wife cooks and cleans to supports her husband’s career for 5 years, and then her husband does likewise, both get careers and total-life equality. But if we insist on permanent equal cooking, or revert to equal cooking before the decade ends, neither spouse obtains a career, or only he does: prudence, equality, or both are sacrificed. Juliana could then make an “other things equal” claim, and say that if both can have careers and cook together, this is better than taking turns. Perhaps, but this option will be unavailable, for example, when total-life equality requires simultaneous inequality, because of, eg., responsibility considerations. For example, if Sun spends his life sunbathing while Moon moonlights, but then Sun receives half of Moon’s pension for the sake of retirement relational equality, distributional total-life equality is sacrificed.

    So, total-life egalitarians would probably prefer to appeal to principles of intra-life distributions than appeal to simultaneous or relational inequality. (Sorry for the length, and thanks again!)

  10. I am quite in line with Anca on whether some of the concerns of relational egalitarians have can be cashed out in terms of goods — e.g. the good of not being dominated or subjected to public shaming — that might not have figured prominently in the literature of distributive equality but might well have done so — there is nothing in, say, the view that people should be equally well off unless inequalities are warranted by how they exercise their responsibility that is incompatible with being concerned with how well off people are in terms of not being dominated and not being subjected to public shaming.
    Martin writes: ‘My point is that, even if we managed to come up with some extensionally equivalent reformulation of the relational egalitarian view into the language of distributive egalitarianism, it is not remotely clear why we should want to do so. For what would have driven our attempt to expand the set of goods within the equilisandum would have been our prior acceptance of the relational egalitarian insight regarding what really matters about inequality.’ I am not sure I follow this. Surely, one can subscribe to a certain theory, be persuaded by a certain putative counterexample, and then learn that one is mistaken to see the counterexample as a counterexample, because there are version of one’s theory to which the relevant counterexample is not one. This seems unproblematic. Perhaps Martin’s real point is that there is no real difference between a version of distributive equality and an extensionally equivalent relational theory. If so, things cut both ways and we might as well say that it is not ‘remotely clear’ why anyone should insist on using the relational vocabulary. More generally, I think clarifying whether the insights of relational egalitarians can be accommodated by some forms of the distributive ideal is important. One reason is that we want to know whether relational egalitarians object to particular forms of the distributive ideal or whether they object to distributive equality per se. Saying the former does not remove disagreement, but it makes us understand it better taxonomically speaking.

  11. Martin asks: ‘Why not accept that, while it is difficult to overestimate the significance of overcoming socioeconomic inequality, this significance can be explained entirely via a mix of ‘non-intrinsic’ and relational egalitarian concerns, deontic concerns with the fairness of collective institutions, and nonegalitarian humanitarian concerns (under which I’d include Paul’s remarks about dignity)?’ This is a fair question and I think Juliana hints at an answer in her piece (but she is in a better position than I am to say something about this matter.) Some egalitarians think that inequalities across non-overlapping generations (or, if you like, cohorts) matter, e.g., that we might have a duty not to bring about a future where different and non-overlapping generations will be very unequally well off. This distributive view seems difficult to explain by a mix of ‘non-intrinsic and relational egalitarian concerns. Of course, one might deny that there is such a thing as justice between generations (none of whom might be socially related to one another in any sense of social relations that relational egalitarians care about), but, at least, this case shows that the concern for distributive equality is not derivative.

  12. Christian writes: ‘if relational egalitarians care mainly about respect/recognition/non-domination/status, and distributive egalitarians care mainly about fairness, especially of a luck egalitarian sort, then they come apart in disability cases: if relational egalitarians care about non-marginalisation, non-domination, and social inclusion, then they will strictly require these for the disabled, but they will not compensate them for additional losses in well-being, which might be very significant (perhaps even more so if one becomes disabled during the course of your life). An example in the opposite direction: if you responsibly make very bad choices leaving you badly off, then you might also lose a lot of opportunities for political influence (even if some form of sufficiency is still guaranteed), but most relational egalitarians will want to make sure that you continuously retain some form of equal substantive opportunity for political influence. So they will, to that extent, in effect compensate such persons for the effects of their own bad choices.’
    Christian’s claim is a conditional one, so I agree. However, I do not think one has to construe distributive and relational equality in the way Christian does. First, some friends of distributive equality are resourcists. They too will not compensate disabled people for loss of welfare. Second, some relational egalitarians care about responsibility. Elizabeth Anderson, for instance, thinks that it is ok for justly convicted criminals not to have the same opportunity to participate in civil society like the rest of us. Presumably, this implies that it is ok for them not to enjoy equal substantive opportunity for political influence. As I see it, the issue of responsibility is not an issue that separates distributive and relational egalitarians per se. (Perhaps Christian agrees because he writes ‘MOST relational egalitarians will want to make sure that you continuously retain some form of equal substantive opportunity for political influence’ – my emphasis). Relational egalitarians might think treating one another as equal involve a hefty dose of consequential responsibility and distributive egalitarians might be outcome egalitarians.

  13. Just a word on Paula’s excellent example: ‘For example, if a wife cooks and cleans to supports her husband’s career for 5 years, and then her husband does likewise, both get careers and total-life equality. But if we insist on permanent equal cooking, or revert to equal cooking before the decade ends, neither spouse obtains a career, or only he does: prudence, equality, or both are sacrificed.’ As I see it – and perhaps as Paula sees it – the example does not show that simultaneous relational egalitarianism is false. At most it shows that to realize that ideal we might have to sacrifice other values. Note also that the case nicely illustrates why non-complete lives relational egalitarians might face an arbitrariness problem which is analogous to the one Juliana convincingly argues that segment distributive egalitarians face – that any delimitation of the segments inequalities between which are objectionable is bound to be arbitrary (though Christian has some interesting thoughts on this). For presumably it would implausible to say that the wife and the husband must cook any meal together. Surely, it is acceptable if the wife makes breakfast and the husband dinner (they eat out for lunch). But if so, would it not be fine if the wife cooks on Monday, the husband on Tuesday? And so on and so forth. If we extrapolate this line of reasoning we will end up with complete lives relational egalitarianism (wife cooks for 30 years and then the husband cooks for the next 30 years or the reverse), so somewhere in this chain of reasoning non-complete lives relational egalitarians will have to object. However, where they should do so seems entirely arbitrary. Avoiding arbitrariness might force one to embrace complete lives relational egalitarianism in the light of the implausibility of time-relative relational egalitarianism (at least how it is cashed out here in terms of co-cooking of any mean being mandatory).
    (My commute ends soon, so it might be a while before my next post:-))

  14. Dear all,

    Thanks a million for all your comments! It’s great to see that you have enjoyed the paper and that it spurred lively reactions! Since I am writing a book on the topic of age group justice, all those comments are very motivating and helpful in deciding which issues to spend more time thinking about. So, I look forward to our continuing discussion!

    I am sorry it is taking me a while to reply, but I am in California and so 9 hours behind you – I was sleeping while you were criticizing me. How brutal! ☺

    I will start by replying to Paul’s excellent Précis. I will post again later today to reply to the other comments (and you will probably only read some of those Thursday morning your time).

    1 – Paul’s first objection was about the possibility to re-describe my argument within the distributive egalitarian paradigm. Why can’t distributive egalitarians, he argues, equalize some goods (the goods of non-domination, inclusion and status, for instance) over simultaneous segments and other goods over complete lives?

    Distributive egalitarians could indeed try to make some kind of qualitative distinction between the goods that ought to be equalized over complete lives and the goods that we should equalize at given simultaneous segments. I think this is a reasonable alternative. Some resources like healthcare resources and income may be subject to diachronic rationality, but basic respect and non-domination would not, for instance. On this view, the distributive paradigm does not need to be supplemented; we should just acknowledge that there is a qualitative distinction between what must be distributed diachronically and what must be distributed synchronically.

    3 responses to this:

    A- I don’t think this objection threatens the view that the best reasons for rejecting some synchronic inequalities are relational. The relational conception would provide the rationale for separating goods that must be distributed over complete lives, at shorter segments, or continuously. We need freedom from oppression to be accessible continuously at any point in our life, but we need healthcare resources and income to be accessible fairly over our complete lives, for instance. I think that I would be very happy if my paper could give additional reasons to convince people who don’t see the relational egalitarian view as a constitutive part of egalitarian justice to attempt to integrate those concerns better within their distributive accounts. So this does not threaten the view that relational egalitarianism is essential, but it would threaten my view that distributive egalitarians cannot re-describe this adequately. That’s a problem for me. I agree with David that it is often awkward to bend relational concerns as much as possible so that it would fit within the distributive paradigm. I am just not convinced (like David and Martin) that we should force those concerns into the conceptual boxes of distributive equality. I prefer a hybrid account that does not risks losing part of what matters to egalitarian justice. So the two additional objections below will tell you why I think this option runs into problems.

    B – I think there is a problem with your first objection, Paul (And you will tell me what you think because I may be missing something). Having or not having access to goods and opportunities, almost always has an impact on whether people will be able to relate as equals without hierarchies of status, power or rank. Opportunities for jobs, housing, income and healthcare all have a fundamental impact on synchronic relational equality. If our ability to stand as equals before others is dependent on the bundle of resources and opportunities we have access to throughout our lives, then it’s not clear how distributive egalitarians will separate the goods that must be distributed over complete lives from those that must be distributed over simultaneous segments. This may be a reason to not force my relational concerns into a distributive framework. We may instead accept that all resources and opportunities must be distributed fairly (and thus diachronically), but that we should also be concerned when this fair distribution leaves it open for people to oppress each other at given points in time. These would remain 2 non-reducible concerns.

    C- Last little comment on your 1st objection. I don’t think that the arbitrariness problem goes away simply by equalizing non-domination (as opposed to opportunities say) over simultaneous segments. The problem is linked to the use of segments itself, not simply to the change of currency. As soon as you use segments (which you have to do if you are primarily concerned with distributing resources fairly) you lose the concern for continuity – that no one shall ever be dominated or be in a position to dominate others. And you open yourself to the arbitrariness problem: OK domination at T1 matters, but isn’t it trumped by the aggregated non-domination over complete lives? If we care about what’s fair, then complete lives should be the focus, no? This is the arbitrariness problem.

    2- Paul’s second objection is that there is something missing is my hybrid account. An element that is non-comparative. For instance a sufficiency threshold that ensures a life in dignity.

    I fully agree with Paul that the synchronic relational view + the complete lives view won’t get us all we need here. But I offer a different complement grounded on a slightly revised account of prudent planning (drawing on Daniels). As you see in the conclusion of my paper I point at a second principle: “treating age groups fairly by making sure institutions are prudent (principle 2)” In the book I am writing, I spend a chapter discussing what prudent planning entails and I conclude that it should lead to two prudential principles which I call lifespan sufficiency and lifespan efficiency. The lifespan sufficiency threshold I appeal to is described like this: “Institutions must distribute resources between age groups in a way that ensures a normal opportunity range throughout people’s lives – i.e. members of any age group should be maintained above two thresholds: (1) an absolute threshold defined by appeal to basic human needs (or negatively avoidance of human deprivation); (2) an age-relative threshold set at the level of what counts as a ‘normal’ set of opportunities for a given age group in a given society at a given time.” I think the first level of this threshold is very similar to what you point at here (and to what Paula Casal points to as well). So I agree with you that something like this is needed. And I also agree with you that accepting this commitment to sufficiency does not make my call to secure relationships of equality at any point redundant.

    An important difference between this prudential threshold and yours though is that the prudential threshold is described by appeal to what is ‘normal” and ‘age-relative’ it is thus at least partly comparative. The question Daniels asks is which inequalities between age groups are just? And in order to answer that question, he appeals to the prudential lifespan account described in my paper (p250-252). What is prudent intra-personally becomes what distributive fairness between age groups requires (within the constraint of complete lives equality). So sufficiency here is defined partly comparatively by looking at what other age groups have and whether a given distribution of resources is prudent or not. This way, what counts as ‘enough’ partly depends on what other age groups have.

    Waw – that was long, sorry. I will be shorter in my responses to the other comments. Thanks a lot again Paul, and I look forward to your next comment!

  15. Dear Anca,

    Thanks so much for taking part in this discussion. Your Hikers in Flip-flops paper recently gave me a lot of food for thought!

    As you say commenting on Paul’s first objection: “The precise identification of the relational goods in question is very important for making a convincing case that relational egalitarianism can be expressed as/reduced to distributive egalitarianism.” I agree with you that identifying those goods is key for distributive egalitarians to offer a response. And I have doubts this is possible for the reasons I mention in my comment to Paul: if all goods have a relational dimension, how are we going to decide which good to allocate simultaneously from the goods that ought to be distributed diachronically? I think that can only be done successfully by both appealing to prudent planning (which is entirely compatible with the distributive framework), AND by accepting a concern for continuous relational equality that we REFUSE to reduce to the rationality of diachronic fairness. As you conclude: “Distributive egalitarians with this agenda will need to ensure the kind of continuous equality endorsed by relational egalitarians.”

    I now have to go to a faculty meeting. I will respond to the other comments later in the afternoon CA time).

  16. Paula Casal says:

    A brief note while Juliana is away 🙂 I like Kasper’s slippery slope version of the cooking case, but maybe Juliana could use David’s remark that “40 years of domination and unequal relations will have a very fundamental impact on people’s lives“ and reply: the line should be drawn at the point where periods are sufficiently long to have an important impact on relational values like status, domination and so on.

  17. bouloshabib says:

    Thanks Juliana for that reply and thanks to others for those interesting points. I have several comments about the discussion so far.

    The first is a brief point about the reducibility of relational egalitarianism to distributive egalitarianism (this is a point that the discussion has already covered in some detail, so I won’t cover it much). My first objection to your article Juliana was that I couldn’t see why the relational egalitarian concern that people not be dominated, excluded or regarded as inferior in status, couldn’t be re-described as the concern that they be simultaneously equal in non-domination, inclusion and status. Martin says, and I think you agree, that for distributive egalitarians to redescribe their view in that way is for them to concede, in effect, to relational egalitarians. As he writes, “what would have driven our attempt to expand the set of goods within the equalisandum would have been our prior acceptance of the relational egalitarian insight regarding what really matters about inequality.”

    I think Martin’s reply assumes that a distributive egalitarian is someone who holds a narrow equalisandum. I don’t understand why expanding the equalisandum makes one a relational egalitarian. Perhaps relational egalitarians attribute an unfairly narrow equalisandum to distributive egalitarians and then claim that expanding it takes distributive egalitarians into relational territory. That wouldn’t be fair. David and Juliana add to Martin’s point that it is awkward to fit relational concerns into distributive boxes (as Juliana puts it). I am not sure about that. True, it is awkward to say that the reason I must not dominate others, or exclude them from society, is that “I should ensure that a distribution of equal non-domination or inclusion in society obtains between them and me”. But it is not awkward to make the same point in different words, for example by saying: they should be as included in society as I am, and they should be as free of arbitrary interference as I am.”

    My second point is about Juliana’s suggestion that the kinds of continuous non-comparative concern that she and I both think we must show people can be justified by appealing to the prudential lifespan account (which says, very roughly, that people should be given the resources at any given stage of life that they would have allocated to that stage had they been able to spread a fair lifetime share of resources across all the stages of their life). Juliana suggests that the prudential lifespan account can justify a principle of “lifetime sufficiency”. I am not sure it can. There are stages of life in which it is so expensive to raise people above a sufficiency threshold, or in which resources will only do very little to improve one’s below-sufficiency condition, that the resources needed would be better used in other parts of their lives. I sometimes ask my students, for example, whether they would prefer spending a lot of resources on their care should they suffer from severe Alzheimer’s in very old age or whether they should use very few resources on their care and divert the saved resources to earlier parts of their lives. Many of them say the latter. But when I ask whether they think our society should divert resources way from people with severe Alzheimer’s in old age to younger people, they reject the suggestion. I agree with them. So did McKerlie in his books, “Justice Between the Young and the Old”. He gives this example: “A nicely furnished television room does little for the mentally incompetent, and the people who could enjoy it may often be reluctant to use it, because of the presence of others. Yet to many of us, the effort and the expense is worth making, despite the limited benefits.” (p. 47) The prudential lifespan account can’t explain that conclusion. So I think that if we want to endorse a non-comparative continuous sufficiency principle we have to do it independently of the prudential lifespan account.

  18. Paula Casal says:

    I agree with Paul that we cannot expect continuous sufficiency to be always a by-product of prudence.

    And I’d be interested in reading more reactions to Juliana’s view that relational goods must be distributed in one way, and other goods, eg health, in a different way. Does this offer the strongest case against Unequal City and other cases of simultaneous inequality? The case seems stronger if, instead, we claim that all the goods ever mentioned in instrumental arguments for equality are arguments for simultaneous rather than total-life equality.

    For example, it has been argued that a society that, during a certain period, exhibits large resource inequalities will pay a price for this in terms of losses of community, fraternity, trust, legal impartiality, democracy, law and order, sustainability, mental health, physical health, longevity, and increases in crime, violence, drugs, domination, stigmatization and teenage pregnancies. If we can appeal to the whole list, the case against Unequal City is stronger than if we appeal merely to those goods that are described as relational.

    One may also argue that all the goods in the list are relational in some sense. Even health losses are often linked to unequal status, unequal responsibilities, stress and other relational shortcomings. So, why treat some goods in the list asymmetrically and contrast distributive and relational egalitarianism rather than total-life equality and instrumentally-justified simultaneous equality?

  19. I am back with quite a few comments! And I have to thank you all again. It was wonderful to read all your sharp and challenging posts.

    I have capitalized your names in the Posts so you can see easily where I am responding to your comments.

    This first post is about how I can defend my synchronic relational view from the charges that 1) relational equality ought to apply over complete lives; and (3) that the relational view may face an arbitrariness problem too.

    (1) relational equality ought to apply over complete lives

    KASPER says “For suppose X dominates Y for the first 40 years of their lives and the relation of domination is then reversed for the next 40 years. Even if we think this is not perfect from the perspective of relational egalitarianism, at least it is less bad than if X dominated Y for their entire lives. Would you not prefer this situation about rather than one in which X dominates Y for their entire lives even if everything else is equal and if these were your only two options?”

    I don’t think I ever deny that the swapping-cases world is better than the whole-life castes world. What I am doing however is point to what is missing for that society to be adequately egalitarian. The society described is wrong from the point of view of complete lives equality (whether relational or distributive) AND from the point of view of synchronic relational inequality. So I can agree that it is in one way worse. What I argue, however, is that a world where X dominates Y for 40 years and then Y dominates X for 40 years is not clearly better than a world where X and Y are unequal over their complete lives, but never in a position to dominate each other.

    DAVID says “There is something bad, inegalitarian, and unjust about a society in which such intentions, attitudes, and outlooks prevail. And I think this is bad regardless of how it impacts people’s shares and opportunities (although, of course, I think it is worse if it does).” DAVID I fully agree with you. This is exactly what I mean and how I wish to reply to KASPER’s X/Y example too.

    The more difficult concern you raise KASPER, however, is that relational concerns could apply over lifetimes rather than continuously. We have had this discussion before and I think that this is a very important objection that I am spending quite a bit of time thinking about at the moment.

    KASPER’s objection is that the conflict that I am seeing between complete lives equality and synchronic relational equality can happen between complete lives relational equality and synchronic relational equality. And yet, he says, I don’t say why we should value synchronic relational equality over lifetime relational equality.

    I have two answers to this problem. My first answer is to say: perhaps then I should replace the complete lives egalitarian principle in my hybrid with a relational complete lives egalitarian principle. Like I said before, I am not worried about internal conflicts, I just want to (1) show that we cannot reduce the problem to a matter of complete lives equality and (2) explain the normative significance of some synchronic inequalities. So I should not be more bothered by conflict between diachronic fairness and synchronic relational, on the one hand, than by a conflict between relational equality applied at both temporalities. The aim of this paper was to show that distributive egalitarians have to accept that something is missing and that the best way to capture the remaining concern is synchronic relational equality. Whether one also decides to replace their complete lives principle by a relational one is up to them, but I doubt you want to do that ☺.

    However (and this replies to your question I think MARTIN), I don’t want to entirely give up on distributive fairness over complete lives either. As you say KASPER, and as I hint at in the paper, I think that the best way to express the requirements of justice between birth cohorts is to say that fairness requires that we should make sure successive generations are not made worse off through no fault of their own. They should have access to approximately equal sets of opportunities. This seems to be a very intuitive and important principle to me. My only complaint is that we also need principles of age group justice and the approximate equality of opportunity between cohorts principle does not tell us much about which inequalities between age groups are objectionable. (In response, to MARTIN, I would simply add that distributive fairness over complete lives can be ‘grounded’ on relational concerns. It may be the best expression of what it means to treat birth cohorts as equals. My second principle appeals more explicitly to relational concerns since they were missing in the existing framework and it is a core part of age group justice – but I don’t see why distributive fairness could not be grounded on the basic relational commitment too).

    My second answer (which I put forward in the paper) is that there are no reasons for relational egalitarians to focus exclusively on the diachronic perspective, whereas there are reasons for distributive egalitarians to do so (reasons of fairness as I explain in the paper). The distributive approach aggregates goods diachronically to decide who is worse off. To decide whether the distributive inequality between two individuals today is fair, I need to know the big (diachronic) picture. The relational view does not need to do that. It does not need segments to apply itself onto. What it does is describe relationships that are objectionable whether they are at a given time or over time. The relational wrong is even more relevant if it has time to develop and cluster over time of course, but the relational wrong is there whenever it happens. By contrast, a distributive inequality at a given segment CANNOT be deemed unjust, on the distributive view, if it generates compete lives equality (unless we appeal to a sufficiency threshold, prudence or my synchronic relational principle – and in the case of the unequal city, I argue, only the later will help).

    (3) Does this leave us with an arbitrariness problem as KASPER suggests?

    So I said that we should ensure non-domination continuously. But what if a wife and husband dominate each other only during short periods of time and those phases equalize over time?

    I like PAULA’s reply: “the line should be drawn at the point where periods are sufficiently long to have an important impact on relational values like status, domination and so on.” This is an interesting response. I think that when we care about ensuring equal relationship continuously, we will be attentive to the way they operate, cluster and are built over time. So the relevant temporality is not an instant. We object to relationships of inequality as they are continuously and we investigate which are really about domination and unequal status and which aren’t.

    It is important to note though that the example PAULA gave in her previous post of a husband and wife taking turn at developing their careers seems compatible with synchronic relational equality to me. Insofar as we consider that home/care work needs not be exploitative and degrading per se. However, if you change this case and now the husband and wife take turn at dominating each other: for 5 years the wife makes all the decisions and gets to use her husband as a slave and has all the power and then they swap – then we are in a case where we should insist that synchronic relational equality is undermined largely regardless of whether people switch position or not. And whether this benefits their careers overall is partly beside the point.

  20. Dear Christian,

    In reply to your two excellent and challenging comments.

    First, on whether the example I give is of special importance for this debate (or whether we can easily find other examples like this). KASPER has already offered a few replies. What I want to add is that I do think the age group case is special. This is because even though there are many non -marginal cases (as you point out) they are not many (1) where both ‘camps’ share intuitions on the unattractiveness of the case; (2) distributive egalitarians won’t bite the bullet that it is a non-problematic inequality after all; (3) distributive egalitarians (at least I argue) cannot explain what is wrong without committing themselves to the (continuous) relational view. I think few distributive egalitarians would say ‘the unequal city example is just fine’. In the cases you give disability and opportunities for political influence, I think the two camps would not have the same intuitions on given examples, and as KASPER has responded, both camps would have explanations for why they do or don’t matter.

    Second, on your important comment that discrete stages of life (childhood, young adulthood, formed adulthood, old age, etc) may solve partly the arbitrariness problem distributive egalitarians have. After all, those stages are not arbitrary. You say it very nicely: “I wonder whether some arbitrariness in numbers might be acceptable as long as we have good reasons to focus on a certain way to carve up the life segments, based on their content.”

    My answer is as follows:

    The appeal to those life stages does not solve the arbitrariness problem as such, I think. All we are trying to do is find an egalitarian reason to explain our reluctance to accept the unequal city as just. Wellbeing egalitarians may develop a non-arbitrary story about discreet phases of life (infancy, youth, adulthood and old age) with distinctive importance for an overall good life. However, this does little to solve the arbitrariness problem. The issue here is that we must explain why synchronic inequalities between segments matter, when complete lives are however equal. Explaining why parts of lives matter to overall welfare does little to explain why inequalities at Tx should be the relevant concern rather than inequalities at smaller or larger segments.

    Discrete stages of life play a very important role in defining what prudent planning requires for instance. Once we have agreed that goods must be distributed fairly over people’s complete lives (that they should all get their fair share of opportunities for instance), the next question is clearly how should we allocate this fair share over people’s lives as a whole. Should we give it all as they start their life, or should we leave enough at any point? What is the best way of doing so? Daniels has his prudential answer. I argue that the Unequal city may be compatible with prudent planning and so cannot explain what is wrong. I add the synchronic relational view in this context.

    You are right that childhood is a tricky case, and so I would gladly accept that synchronic relational equality between children and adults is not an obvious goal (even if I think we should do our best to relate to children in a non-dominating manner). But I think that synchronic relational equality between adults of all ages is important. Infantilizing or marginalizing young adults or elderly people, as is often the case, should be dreaded. I know we all feel that way when we see elderly people being treated as though they were children. I know we feel that way when we hear the media stereotyping young people as lazy and self-entitled. And I think we are right to feel that way, but that we should also worry of those that we see as ‘adapted’ to the place in the lifespan of an age group – the veneration and political influence of older members of some societies is one of these. I would add that accepting differences between age groups does not have to lead to accepting relational inequalities between them. Take the example of the absence of young people below the age of 30 years old in parliaments. According to the Inter-parliamentary Union: “When “young” is defined as under 30, only one country, Norway, breaks the 10 per cent barrier. Two thirds of single and lower houses of parliament have 2 per cent or fewer young parliamentarians. All upper houses have less than 6 per cent, with three quarters electing no young parliamentarians at all.”

    Should we accept this given that we all age and given that \young people are different in that they are less likely to have as much experience (political experience) for instance as older MPs? I don’t think so. Accepting their political marginalization seems mistaken to me – it signifies that we hold them as political inferiors and reinforces the idea that politics is the task of the wise old father or grandfather. I think that there is a lot of continuity between our acceptance of status inequality and these examples of marginalization by age. The exclusion of the young in parliaments is not unfair (if we follow the principle of complete lives equality), but it is still unjust from an egalitarian perspective (if we follow the synchronic relational principle).

  21. Dear Paul,

    Very briefly (because I have been so long in almost all my previous comments):

    I like your response to MARTIN, DAVID and I, as well as your attempt at re-expressing the concern in ways that are not awkward.

    But my two other responses to your first objection still stand right? The first was in essence: “If our ability to stand as equals before others is dependent on the bundle of resources and opportunities we have access to throughout our lives, then it’s not clear how distributive egalitarians will separate the goods that must be distributed over complete lives from those that must be distributed over simultaneous segments. This may be a reason to not force my relational concerns into a distributive framework.” I think this is also a point that PAULA wanted us to discuss more. So I would love to hear what you have to say in response to her last comment and to this point I make.

    The second, in essence: “I don’t think that the arbitrariness problem goes away simply by equalizing non-domination (as opposed to opportunities say) over simultaneous segments. The problem is linked to the use of segments itself, not simply to the change of currency. As soon as you use segments (which you have to do if you are primarily concerned with distributing resources fairly) you lose the concern for continuity – that no one shall ever be dominated or be in a position to dominate others. And you open yourself to the arbitrariness problem: OK domination at T1 matters, but isn’t it trumped by the aggregated non-domination over complete lives? If we care about what’s fair, then complete lives should be the focus, no?”

    Same answer to KASPER to his question: Can’t we just integrate those worries within the distributive framework? If some forms of domination are bad, then let’s just add them to the relevant equalisandum. For instance, KASPER says that if stereotyping is a problem, then we can just add it to the list of what needs to be distributed.

    Again, my problem here is that you would have to aggregate stereotyping over complete lives. You would not have egalitarian reasons to stop stereotyping young people or elderly people, as long as the goods and bads of stereotyping are distributed fairly. This is what I am finding very unsatisfying.

    On whether Daniels’s view can in fact generate lifespan sufficiency, I agree with you that it is a difficult and important debate (and I really liked your paper on this and draw on it quite a bit on my chapters on the complete lives view and on prudence – which I would love you to have a look at at some point!!). I would simply say that I don’t take prudent planning to be simply about efficiency. It is about distributing resources in a way that makes a life go as well as possible overall. This, for me, necessarily generates a commitment to lifespan sufficiency. Granted, the case of the very elderly is tricky. If I am supposed to lead a life of normal length, why should I leave resources for my very old age? However, behind the veil, I know that I may leave a longer life (I make my decision based on an average life expectancy so I know I may leave longer right?), so I think it makes sense as a planner to want to ensure a decent life (or in your words a life in dignity) for as long as I live.

  22. Dear Paula,

    I loved your example of a society that is extremely unequal in terms of how it treats its young members!

    On your equally interesting second example at the hospital. I find it hard to believe that the elderly lady’s status was intact in your example actually. Was she being infantilized by the nurses? Or was she looked at by the nurses with a despising attitude (with rolling eyes)? You said the nurses could not understand her, and I think also that there is a very particular kind of vulnerability that comes with not being understood. The kind precisely that makes you vulnerable to status threat! If so, I think your reasons to step in may have been related to domination after all!.

    But, let’s assume it had nothing to do with trying to prevent disrespect, infantilization, or domination. Then perhaps indeed it was just because of her basic need. As I said in my response to Paul, I agree that a sufficiency threshold is part of an adequate hybrid. Mine is hidden in the prudential requirement (principle 2). But I agree that my interpretation of the requirements of prudent planning may face issues..

    Now, do we still need the hybrid? PAULA rightly points out that we don’t need principles for all negative effects of inequalities. In response, I would simply say that relational inequalities are not undesirable effects. I try to avoid talking about how the elderly are made to ‘feel’ in the unequal city case. In fact, I think that whether they feel bad or whether their wellbeing is undermined is not what I am concerned when I criticize this case from the point of view of relational equality. Insofar as they are stigmatized, infantilized, marginalized, or segregated, we, as egalitarians can see that something is wrong regardless of its effects. And here, the non-comparative sufficientarian threshold will only get us far enough, I think, as I try to argue in the paper.

    Also, on your last post. I think I was not clear in my reply to Paul. I do not believe that “relational goods must be distributed in one way, and other goods, eg health, in a different way.” I think that this view does not work precisely for the reasons you mention. But I took it that this was Paul’s first suggestion in his Precis when he says: “distributive egalitarians might argue that the relevant equalisandum that should be equalized between simultaneous segments of people’s lives should be restricted to the goods of non-domination, inclusion and status and that when it is restricted in that way, simultaneous equality becomes much more plausible.” So I was trying to first express what this would entail before criticizing it a couple paragraphs below: “Having or not having access to goods and opportunities, almost always has an impact on whether people will be able to relate as equals without hierarchies of status, power or rank. Opportunities for jobs, housing, income and healthcare all have a fundamental impact on synchronic relational equality. If our ability to stand as equals before others is dependent on the bundle of resources and opportunities we have access to throughout our lives, then it’s not clear how distributive egalitarians will separate the goods that must be distributed over complete lives from those that must be distributed over simultaneous segments. ” This is why I don’t believe that separating goods in two categories (those that must be distributed over complete lives and those that must be distributed at simultaneous segments) will not work.

    But I may have misunderstood PAUL’s view though (or your comment PAULA)! PAUL – it would be great if you commented on this when you have time tomorrow!

    I cannot wait to read you all again tomorrow!

  23. This is a really wonderful discussion, and I’m only sorry that I’ve been too busy with other commitments here to be able to engage more extensively. I will, though, just add one very small rejoinder to Juliana’s response to my question of why she’s minded to stop at a ‘hybrid view’ rather than to reject what Derek Parfit called ‘telic egalitarianism’ (and what Paul calls ‘comparative egalitarianism) altogether. Juliana’s response is that, while she wants to hold on to a commitment to distributive fairness between birth cohorts (over complete lives), she nevertheless thinks “distributive fairness over complete lives can be ‘grounded’ on relational concerns. It may be the best expression of what it means to treat birth cohorts as equals”.

    Now it seems to me that, when we talk about the relationship between distributive and relational egalitarianism, there tend to be two different issues in view. One is an issue of translatability, and one is an issue of normative priority. To put it another way, there’s one question which is to do with whether the central claims of relational egalitarianism can be captured in the language of comparative or distributive egalitarianism, as regards the specification of a suitably broad or complex equilisandum, but the more important issue, it seems to me, is the other question, which is whether the relational or the egalitarian view gets to the normative heart of the matter, as regards why we should care about equality and inequality.

    So, to return to Paul’s characterisation of ‘comparative egalitarianism’, it is the view that “inequality is bad because it involves a person’s being worse off than others”. Now, relational egalitarians will often agree with this claim, if read flatly, and hold that some inequality is bad because it leaves one person worse-off than another. But they will think that this is just what one might call an intermediate or non-fundamental normative claim, given that there is a further explanation that can be given of what makes this distributive inequality bad, and that explanation will itself make references to the relationship between that inequality of condition and an underlying moral concern with egalitarian social relations. Inequality could be bad because it makes some worse-off than others, but that might just be a first step towards an underlying relational view.

    In order for Paul’s characterisation to mark out a distinct view, then, and not just a truism that relational egalitarians would be happy to accept, it needs to be read as suggesting not only that inequality is bad because it leaves some worse off than others, but that this is a non-derivative or fundamental account of the badness of inequality. (Compare with Parfit’s characterisation of telic egalitarianism, according to which “it is *in itself* bad if some people are worse off than others”.) Parfit’s characterisation marks out telic egalitarianism as a view that is distinct from relational egalitarianism, as it asserts the intrinsic badness of inegalitarian distributions. If we insert a tacit *in itself* into Paul’s characterisation of ‘comparative egalitarianism’ then it is equivalent to Parfitian telic egalitarianism; if we do not insert a tacit *in itself*, then Paul’s ‘comparative egalitarianism’ states a claim that all egalitarians can trivially accept, and does not mark out a distinct view. I therefore would think that Paul would be happy to accept the tacit commitment to non-derivativeness, and for comparative egalitarianism and telic egalitarianism to be the same view.

    This brings us back to Juliana’s response to my question, regarding why she wants to stop with a ‘hybrid view’ rather than rejecting comparative egalitarianism altogether. Her response has the surface appearance of being concessive towards the claims of distributive egalitarians, in that she accepts that we can have reason to accept comparative egalitarian claims, but this surface appearance is misleading; in fact, she’s being far more radical than I was minded to be in undermining the comparative egalitarian position. This is because Juliana also suggests that these claims may in fact simply be derivative of an underlying relational egalitarian view. So in fact, despite surface appearances, Juliana’s response is much less concessive than it might have appeared. My suggestion was that comparative or distributive egalitarianism was a distinct kind of egalitarian view and one that is merely mistaken (I was, so to speak, reading Paul as tacitly including Parfit’s ‘in itself’, and as marking out a fundamental rather than a derivative account of the badness of inequality). Juliana, on the other hand, isn’t so much allowing that distributive or comparative egalitarianism is a distinct view that’s merely mistaken, but she’s instead suggesting that it’s a view with merely the appearance of distinctiveness, which is liable to collapse into relational egalitarianism under further analysis.

    My concise question to Juliana was: why stop here? The answer, I now see, is that she hadn’t really stopped here at all, and that her apparently measured and conciliatory approach towards the distributive or comparative egalitarian view masks a radical underlying rejection of that whole way of thinking about the normative badness of inequality. 🙂

  24. Paula Casal says:

    Uuuh, a non-distributivist wolf in hybrid sheep’s skin! :)

  25. This is a response to Juliana’s reply some posts ago, which was helpful for me in clarifying that her aim is not to adjudicate between synchronic and complete lives relational egalitarianism but simply to show that distributive equality must be supplemented with some form of relational view — synchronic version, complete lives version or something else. An ambitious project even if it generates a new question about the nature of this supplementary component.

    In relation to Paula’s example of wife and husband taking 5 years turns doing all the cooking and cleaning Juliana writes: “However, if you change this case and now the husband and wife take turn at dominating each other: for 5 years the wife makes all the decisions and gets to use her husband as a slave and has all the power and then they swap – then we are in a case where we should insist that synchronic relational equality is undermined largely regardless of whether people switch position or not. And whether this benefits their careers overall is partly beside the point.” I definitely agree with the last point. The concern for welfare (assuming careers is a proxy for welfare) is an important concern but it is not a concern of egalitarian justice. In most cases, it might well favour egalitarian relations over inegalitarian relations, but in some cases it will not. I am less convinced by what comes before this remark in the quote. To my mind life-time domination is much worse that a scenario where two parties take turns in dominating each other. In fact, if the parties agree to take turns to dominate each other, they might enjoy one/the important good normally secured by relating to one another as equals — to wit, that they can justifiably see themselves as each other’s equal. To determine whether such a scenario is in any way bad from the point of view of egalitarian justice, we will need to know more about what exactly is bad about domination (even if consented to and even if the parties change places on a regular basis). Perhaps some relational egalitarians think that there is a deontological restriction against domination (even if consented to and even if mutually beneficial for both of the part-time dominators/part-time dominatees). Note also that this scenario is a very unusual case of domination, so if relational egalitarians propose their theory as something that articulates the complaints of real life egalitarians — something David suggested and which I think he is right in thinking is the self-understanding of Anderson and Scheffler and perhaps others — perhaps they should have an open mind about such a case not being unjust at all.

  26. Dear Kasper,

    Just a clarification. You say that my “aim is not to adjudicate between synchronic and complete lives relational egalitarianism but simply to show that distributive equality must be supplemented with some form of relational view — synchronic version, complete lives version or something else.”

    That’s not exactly what I said. I said that I am not defending the complete lives distributive view over the complete lives relational view (at least in this paper). And that maybe we could do without distributive fairness over complete lives if we had relational complete lives egalitarianism (whatever that would look like). However, I do think that the complement that’s missing in the complete lives distributive account is continuous relational equality (or synchronic relational equality – people must be free from domination at any point), and not complete lives relational equality. What is missing in the Unequal city case is synchronic/continuous relational equality. Whether that may in some cases conflict with complete lives equality (however understood) is clear. But I am saying that the concern for synchronic relational equality is important and non-reducible to the diachronic approach (relational or not).

    Then you say:”To my mind life-time domination is much worse that a scenario where two parties take turns in dominating each other.” And I think you are right of course… This life-time domination case is failing BOTh the complete lives equality requirement AND the synchronic relational requirement. How could anyone prefer it! I am just saying that a distributive complete lives view cannot explain what is STILL wrong when people take turns.

    Moreover, the relevant scenario is rather this: if a group has dominated another for 10 years and we can now act and do something about it. Do we give the power to dominate to the former dominated? Or do we make sure no-one is ever able to dominate the others? If you go for the second option, you subscribe to synchronic/continuous relational equality. If you go for the first option, you go for pure complete lives equality. I think in that case we will have good reasons to think that synchronic relational equality should be secured even if it conflicts with diachronic fairness. In other cases we may go the other way.

  27. Dear Juliana,
    Re your aim in the paper and what you write about it here: You’re right. I misunderstood what you wrote.
    Re life-time one-way vs taking-turns domination: I still think that the way to make progress on this matter — ‘Do we give the power to dominate to the former dominated? Or do we make sure no-one is ever able to dominate the others?’ — is to try to say something about what makes relating to one another as equals valuable. Perhaps not more can be said that it is always unjust if someone dominates another whatever the circumstances — no account given, this is just a rockbottom intuition –, but to my mind this is unsatisfactory in the same way as saying that it is always bad if at any given moment someone is worse off than another. I do not think it is intuitively clear that we should always avoid domination. Suppose the husband has always dominated the wife in relation to choice of holiday destination for the first 40 years of their marriage. Is there not a case for making the wife dominate the husband for the last 10 years rather than making them compromise for the last 10 years of their marriage EVEN IF were they to compromise they would select the same destinations as the wife would unilaterally select? Would we not find it objectionable if the husband said “Ok, let us compromise for the last 10 years”. At least on some occasions I think people think like this in their personal relations (hope I do not manifest myself as a petty minded person).

  28. Thanks so much Kasper for pushing me on this important point! I think that you are right that much more needs to be said about what it means to ‘relate as equals’. And I will definitely spend some more times trying to think more about what it means ‘over time’ while knowing that one will age and change position for instance. ( And I do not think you manifest yourself as a petty minded person 🙂 )

    MY RESPONSE HOWEVER:

    I would stick to the principle that ‘no one should be in a position that allows them to dominate others or be dominated by others at any point”. I still don’t think that it “is unsatisfactory in the same way as saying that it is always bad if at any given moment someone is worse off than another.” This is because, it is UNFAIR to discount what has happened before and what will happen next. So, we have reasons of distributive fairness to never be satisfied with the exclusive focus on a segment that is shorter than a complete life. There are no such reasons on the relational principle I put forward – it is not about fairness, it is about establishing communities where no one is in a position to dominate or be dominated by others. Of course, domination at an instant may be no domination at all since it may do nothing to status and power. But ‘status and power domination’ let’s call it, needs to be avoided for a community of equals. And here two wrongs don’t make a right. If a group is being dominated, the proper egalitarian response (if we only follow this synchronic relational principle) is to end the domination and establish non-dominating relationships rather than replace it with a reversed form of domination. One way I would NOT define continuous domination is a series of micro-dominations that cancel out over time – this is ‘domination fairness’ and that’s not what I think is missing in the complete lives egalitarian account.

    In your example, if you think that the wife ought to have the full power now to decide what they will do for their holidays (to compensate for the 40 years when she had no power) what you are doing is equalizing opportunities for something (decision or domination) over a life course. These are reasons of fairness – whether or not domination is included in the equalisandum. And, as you know, I don’t think our egalitarian reasons are all reasons of fairness. So, if we remain in the paradigm of fairness, complete lives are indeed the only non-arbitrary segment. If we step away from reasons of fairness, there are no reasons why aggregating opportunities for domination over a life course is what we should be concerned with. You may still think that this is part of the puzzle (and of course I don’t deny this since I have a hybrid account) but it is not the full story. And insofar as the synchronic principle is not grounded on fairness, I don’t need to buy the “but you should aggregate domination over time’ bullet.

    When we are deciding what to do in a case where those two principles conflict, we may decide that fairness may be more important than synchronic non-domination. Deciding on a holiday may not be such a big deal from the point of view of status and power domination so we may decide to promote fairness over continuous relational equality in that specific case. In other cases, we may decide that our reasons of synchronic/continuous relational equality are stronger than our reasons of fairness.
    I hope that was clear and helpful!

  29. Paula Casal says:

    I would never think of you as petty minded, Kasper. The example refers to something unimportant, but it is the idea behind it that matters, and the idea makes a lot of sense.
    I was also wondering what people think of this case. A mother who is a distributive egalitarian plans to leave a fourth of her house to her 4 similarly well-off daughters. The daughters never see each other because they live in different neighborhoods, regions or countries and, so relational concerns do not arise among them. But since the same wealth has different significance in different parts of the world, should she leave those in rich areas more, for the sake of their relational equality with their neighbors?

  30. Hi All, after being somewhat derailed by the start of the semester two weeks ago, I have now finally found some time to read all the posts here properly, and think about some of the many very interesting points in it. Below some comments on a few of them.

    ANCA/KASPER: I agree that it is more interesting, and promising for distributive egalitarians, “to describe what is valuable about relationships in terms of goods” and incorporate these goods into the metric, rather than simply aiming “to describe relationships in terms of goods” (ANCA). I have two points here:
    1. At least some relational egalitarians who see some kind of equal non-domination/non-hierarchy as a, and perhaps the primary, requirement of justice (Anderson and Scheffler sometimes write like that, though not always; Pettit in recent works on republican social justice does; I am also inclined towards that position) do not regard it as necessary to spell out the value of non-domination in terms of other goods. They think that guaranteeing non-domination just expresses the right kind of respect for peoples’ fundamental status as moral equals, independently of the question of which further goods (such as planning certainty, or self-respect, understood purely psychologically as confidence about one’s worth) non-domination may generate. To make this position non-mysterious, in comparison to the more straightforward focus on goods generated, a lot of work then has to go into specifying what exactly non-domination is, and what it ranges over (see for example Pettit’s latest works; I intend to contribute to these complications further in the future…).
    2. ANCA writes that goods “such as non-domination…may display a non-excludability feature”: “If some individuals in my society are dominated…then, even if I am not, I fail to enjoy the good of living in a non-dominating…society.” However, if the latter were the relevant good, that would seem to have some strange implications. One is that it doesn’t seem right to care about your domination primarily because I no longer live in a non-dominating society because of it. Or, if that puts things in too selfish a manner, it still implies that you and I have both exactly the same kind of claim of justice to the removal of your domination. That doesn’t seem right to me. You have to deal with somebody else having the capacity to arbitrarily interfere in your choices and projects; I don’t.

    JULIANA, thanks for your replies to my two points. To the first, I say “thank you, I see”. As regards the second, two points:
    1. You reject the idea of non-arbitrarily delineated temporal life segments – by reference to their typical content (childhood, adulthood, and so on) – as a possible basis for a distributive egalitarian comeback, by saying “explaining why parts of lives matter to overall welfare does little to explain why inequalities at Tx should be the relevant concern rather than inequalities at larger or smaller segments”. I am not sure I follow this. Is the point that even if we have a plausible content-based way of cutting up life segments, this does nothing to justify a focus away from complete lives to these segments for the distributive egalitarian (or to having both)? I guess what a distributive (well-being based) egalitarian could say is that these segments of life are important not only because of the contribution they make to complete lives welfare, but in themselves – you just can’t be compensated, or at least not fully, for a childhood that is much worse than somebody else’s by having a much happier old age than that person. A worse/better childhood matters non-reducibly, as it were, because of the specific goods it instantiates (when it goes well).
    2. You also seem to reject my suggestion that a relevant life segments story could be of use for relational egalitarians, as well, in that it could help us find out which relational (in)equalities we should regard as objectionable between different age groups, and which ones might not be that problematic, after all (a point also raised by KASPER). Your main example in your response to me, absence of the young in parliaments, seems to be about political domination – older people will not be forced to take the interests of the young properly into account when the latter are absent from the most important decision-making bodies. I think that’s right, but I was more interested in broader questions of societal status. PAULA made a great real-life example (the Lele) of massive such status inequalities, but she also argued that such cases are objectionable for different reasons, and advocated focusing on the effects. Against that, and in other responses, you seem to insist on comprehensive relational equality all the way through (between all age groups). But don’t we have to determine the relevant relational egalitarian aim in an age-specific fashion, at least sometimes? Sometimes there seems to be a strong case for some form of ‘adaptation’. Take marginalization, for example: if full inclusion in a society means having access to stable paid work, then many old people will be marginalized (and eventually all). But not having access to stable paid work typically means very different things for the young and the elderly. It seems to me that here we need a notion of marginalization that is age-sensitive, otherwise the concept can’t do the work we would like it to do, in terms of signalling social problems to which our conception of justice ought to respond. This links in with DAVID’s comment on the central importance of diagnosing actual social problems for (at least some forms of) relational egalitarianism.

    JULIANA, KASPER, MARTIN on going for a hybrid view vs. going relational ‘all the way down’:
    I was also initially unsure whether you, JULIANA, argue for a hybrid view primarily for polemical purposes (‘even if you are a distributive egalitarian, you need to endorse some relational egalitarianism’) or whether you also think, more positively, that there is something worth retaining in a (intrinsic? See MARTIN’s interventions) distributive egalitarian framework. You say that you want “approximate equality of opportunity between cohorts” – successive generations should not be made worse off through no fault of their own- and that that is best accounted for by fairness-based distributive egalitarianism. But you also say there may be a relational “grounding” for that principle, if I understand you correctly: it may be “the best expression of what it means to treat birth cohorts as equals.” I’m interested in that second alternative: does this mean it can be grounded in relational concerns if we find ourselves in some kind of social relationship with these other birth cohorts, within which we then have to treat them as equals? This would be to couple ‘relational’ as we have been using it in this thread – concerned about the content/structure of certain relationships – with ‘relational’ in the sense in which it is used in the global justice debate – existing social relations ground concern of justice (I think there is a natural affinity between these two, though, pre-empting KASPER :-), they don’t always *have* to go together). If that is the right way to construe this grounding, then it would seem to imply that the requirement of equal opportunity for future generations becomes weaker, in principle, or covers fewer matters, as our social relationship with them becomes more tenuous – over time. If not, I am not sure what the relational grounding could be, and how it would be indeed distinct.

    JULIANA, PAULA, and KASPER, on the interesting point of “compensatory domination” (for the petty minded :-), happy to count myself as part of that group) in order to balance out, in the name of fairness, what happened before in a couple: isn’t the solution here to end domination, but to award the partner who had suffered domination by the other before some sort of compensation, as a matter of historical justice? And that compensation could take the form of power. If that is the reason, which both parties know (and know the other knows), for letting one partner now unilaterally decide some things for a specified period of time, and if the things s/he can now decide are circumscribed, and bear appropriate reference to what happened to him/her before, then this would seem sufficiently non-arbitrary to me to no longer count as an instance of domination. I don’t think we have to choose between fairness and ending domination, at least not in this case. What this brings out is that domination isn’t simply unequal power over some domain of choice. It also raises the question to which extent the concept of non-domination may be moralized, and how much of a problem it is if it is.