Welcome to the second Ethical Theory and Moral Practice discussion here at PEA Soup! This time, we’re discussing Ian Carter‘s new article, “Equal Opportunity, Responsibility, and Personal Identity”, which can be downloaded here. Kasper Lippert-Rasmussen kicks things off with a critical précis, which appears immediately below. Please join the discussion!
Précis: “Equal Opportunity, Responsibility, and Personal Identity”
Kasper Lippert-Rasmussen
Suppose two persons – Jerry and Ronnie – justly enjoy equally good opportunities at some point in time (T1). Suppose that they make different choices for which they are responsible and that as a result thereof, at some later point in time (T2), Jerry ends up very badly off, while Ronnie enjoys a life of luxury. On what we might call the starting gate interpretation of equality of opportunity, there is nothing unjust about this situation. Suppose, however, that T1 was when Jerry and Ronnie were both 21-year-olds, and T2 when they are 71-year-olds. Given this fact and given the huge gap between their conditions at T2, the starting gate conception of equality of opportunity comes under pressure. Could it really be the case that equality of opportunity embraces the huge inequality between 71-year-old Jerry and similarly aged Ronnie on account of their differential choices 50 years earlier? Call the objection that equality of opportunity has this implausible implication the harshness objection to equality of opportunity. The aim of Ian Carter’s brilliant and wonderfully perspicuous piece is to develop a principled version of equality of opportunity that is immune to the harshness objection.
Some will dismiss the harshness objection, because they think we have a duty to respect people’s agency and that this includes respecting people’s exercise of their agency in relation to the intrapersonal distribution across their lives. People who take this view, might add that while we might not be required as a matter of justice to help badly off 71-old Jerry, we might have reasons of compassion etc. to do so. Others will dismiss the harshness objection, because they reject equality of opportunity in the first place, e.g., they might think that justice requires that everyone has enough at any given time. Ian’s piece is written for the benefit of those who find neither of these responses attractive – “ambivalent egalitarians” (p. 3) he calls them. Ambivalent egalitarians see both the attraction of respecting people’s agency, including how they exercise this agency in relation to intrapersonal choices, and the attraction of the view that justice requires us to address the situation of very badly-off people even when these people’s predicament is the result of their past responsible choices. I suspect Ian’s intended audience is large.
Basically, there are two ways of rescuing equality of opportunity from the harshness objection. First, one can redistribute ex post, thereby in effect limiting both the degree to which people are liable to bear the costs of the past choices and limiting the degree to which people are entitled to enjoy the fruits of their past choices. Second, one can restrict the stakes ex ante by limiting the initial option sets available to agents, e.g., by imposing mandatory insurance. Specifically, one can eliminate items in the option set that might result in agents being much worse off at T2 as a result of choices made at T1.
Ian defends the ex ante solution. His defense rests on two central premises. First, he assumes that equality of opportunity is grounded in the moral equality of persons and that in turn is grounded in a requirement of opacity respect. Second, he assumes a Parfitian reductionist conception of personal identity over time. Let me briefly describe each of these two premises starting with the first.
Ian assumes that equality of opportunity is something people are due, because they are agents and because respect is “an appropriate response” (p. 5) to agency. Some might note that even though most human beings are agents, they have agential capacities to varying degrees and, thus, that one might reasonably ask why people endowed with greater agential capacities are not due greater respect than people with lesser agential capacities (or, if respect is incapable of admitting of degrees, why the appropriate response to people with greater agential capacities is not some other response than the appropriate response to people with lesser agential capacities). This is where Ian’s notion of opacity respect becomes relevant. On Ian’s account, once people’s agential capacities exceed a certain minimum threshold, they are entitled to be treated with opacity respect. Opacity respect in relation to a person involves “refraining from taking into account the degree to which she possesses [agential capacities] above a minimum qualifying threshold” (p. 4). Treating people as opaque amounts to “adopting an external perspective, not ’looking inside’ them, and in this sense treating them as opaque” (p. 4). Equality of opportunity not only is a way of respecting agency – it gives agents “the freedom to form and carry out plans” (p. 5) – it is also a way of respecting agents that, specifically, embodies opacity respect – it does not require us to adopt an internal perspective to ascertain people’s varying level of agential capacities above the minimum threshold in order to determine if some are due greater freedom than others. So much for the first premise in Ian’s argument. I now move on to the second, i.e., the one pertaining to personal identity.
On one view – call it the further fact view – what makes a person at one time identical to a person at a later time is the fact that, say, the person has the same soul – e.g., some Cartesian, immaterial substance – at these two times. The further fact view naturally implies that the proper object of respect is the “agent considered as a single, temporally extended unit… your respect for my agential capacities surely involves allowing me to decide for myself how to distribute my resources over that time-span” (p. 6). On this view, the requirement that 71-year old Jerry should not bear the costs of the risky choice he made when he was 21 amounts to not respecting him as a single, temporally extended unit.
Instead of the further fact view, Ian adopts a Parfitian reductionist conception of personal identity over time: “a person’s existence ‘just consists in the existence of a brain and body and the occurrence of a series of interrelated physical and mental events” (p. 8). Crucial to this account are the notions of connectedness and continuity. My selves at T1 and T2 are psychologically connected if, say, at T2 I remember an experience I had at T1, or if at T1 I intend to do something and at T2 I execute that intention. Connectedness between earlier and later selves are a matter of degree. Typically, as time goes by one’s past and one’s present selves become less and less connected. Indeed, at some point in time they may not be connected at all. This is where continuity enters the picture. My selves at T1 and T3 might be continuous, even if they are not connected in any way, e.g., if my selves at T1 and T2 are sufficiently connected, and my selves at T2 and T3 are sufficiently connected, even if my selves at T1 and T3 are not. By way of illustration: At T2 I remember some of my experiences at T1, and at T3 I no longer remember any of my experiences at T1 even though I remember some of my experiences at T2. As Ian puts it: “Continuity, not connectedness is what explains the common belief in personal identity over time, for continuity obtains in virtue of overlapping chains of strong connectedness and is transitive: even if A is not strongly connected to C, A can be continuous with C” (p. 8).
Ian notes that agential items “such as intention, commitment, and the recollection and carrying out of plans” (p. 9) are crucial to psychological connectedness over time or, as he puts, crucial to the “diachronic integration” of different stages of a person’s life. (While agential capacities are crucial they are, as Ian notes, not the only items that matter. Passive items like memories matter too and, arguably, because opacity respect is tied to agency it does not apply to this aspect of diachronic integration.) Next, Ian notes that diachronic integration is a “scalar property”, one can be more or less diachronically integrated over time depending on the degree to which one’s selves at different times are connected. But this observation brings into play Ian’s response to the fact that different people have different degrees of agential capacities. Different people enjoy different degrees of diachronic integration, but opacity respect, diachronically applied, requires that once the degree of connectedness between time-slices of a person surpasses a certain minimum threshold, opacity respect for persons as extended in time “tells us to abstain from taking into account degrees of diachronic integration above the minimum threshold” (p. 10). Hence, in contexts where opacity respect is appropriate, e.g., in the relation between citizens and the states, the state should not, say, hold people who are more connected to their past selves liable to a greater degree for their past choices than people who are less connected to their past lives to a lesser degree. To do so would amount to violation of the requirements of opacity respect.
How do these premises help us to defend a version of equality of opportunity that retains respect for persons as temporally extended and yet avoid the harshness objection? On the one hand, because of the Carterian element of opacity respect it can explain why “persons are due respect as temporally extended agents” and in virtue thereof why “they are due equality of opportunity” (p. 11). On the other hand, because of the Parfitian reductionist element it can accommodate the intuitive force of the harshness objection. Still, as Ian acknowledges there seem to be two different ways to go.
First, one might go for an ex post solution where liability for past choices diminishes over time and uniformly across persons. The first aspect reflects the Parfitian reductionist element and the fact that psychological connectedness tends to diminish over time. The second aspect, i.e. the uniformity across persons aspect, reflects the Carterian element, i.e., opacity respect applied to temporal integration.
Second, one might go for an ex ante solution where we “limit the scope of application of liability-responsibility by limiting the initial opportunities of earlier selves” (p. 13), to wit, by eliminating risky options. On this view, unlike the ex post solution, liability for past choices does not diminish over time. As Ian puts it: “continuity is sufficient for liability” (p. 13). This solution reflects Parfitian reductionism in that the rationale for restricting the scope of liability responsibility is reductionism. If personal identity over time consisted in some further fact over and above psychological continuity (and absence of branching), then, possibly, I may now have an unlimited right to distribute my future opportunities. But given that there is no such further fact, it is fitting that this right of mine is not unrestricted. The ex ante solution also reflects Carterian opacity respect, because to apply this view, we can retain a fully external perspective on people.
Ian thinks the second solution is the correct one. Why is that? Basically, he thinks that the first solution has “morally perverse implications” (p. 12):
“If you have loaned me €1000 and I have failed to repay the sum by the agreed date, the size of my debt to you will, on [the ex post solution], diminish with time… the greater the period of procrastination, the less I will be liable to repay you and the greater the portion of the bill that should be picked up by society as a whole. Moreover, since I am aware of this rule I will have an incentive to hold out as long as possible before repaying the debt. The perversity of these implications suggests the appropriateness of maintaining the same degree of liability-responsibility regardless of temporal distance between the liable self and the self that performed the relevant actions” (p. 12).
The ex ante solution avoids these perverse implications and, accordingly, this asymmetry – that the ex post solution has “morally perverse implications”, and that the ex ante solution has not – suggests that the ex ante interpretation of equality of opportunity is the correct response to the harshness objection.
While Ian’s piece is a valuable contribution to the literature, it might not be the last word on the harshness objection. In closing, I want to suggest four points, which might merit further discussion.
First, some advocates of the harshness objection will see Ian’s proposal as, at best, a partial solution. For Ian focuses on risky choices with high stakes and which are distant in time. But at least on some influential versions of the harshness objection, it applies also to risky choices that are almost simultaneous with the time at which the relevant calamity that the risky choice might result in will materialize, e.g., Marc Fleurbaey and Elizabeth Anderson’s motorcyclist who is involved in an accident shortly after deciding to take a ride without wearing a crash helmet. It is unclear how appeals to a reductionist conception of personal identity can help explain, why one does not have the moral authority to take a risky choice that might result in a very bad outcome shortly thereafter.
Second, Ian’s ex ante solution does not specify that risky options are assessed differently depending on the amount of time that passes from the risky choice to the time at which the risk materializes. Even if continuity is, as Ian puts it, “sufficient for the right to distribute future opportunities” (p. 13) that still leaves open that, in view of how psychological connectedness diminishes over time, the contours of that right should reflect this fact. Specifically, one might suggest that, on the most plausible version of the ex ante solution, the longer into the future that the risk will materialize, the more willing we should be to eliminate that risky option from the initial set of options. This suggestion may be seen (and is intended) as a friendly amendment to Ian’s ex ante solution. However, it may also be seen as a potential objection for reasons related to my first point. It suggests that on the most plausible version of the ex ante solution, equality of opportunity does not avoid the harshness objection with regard to quite recent, risky choices involving high stakes.
Third, I have some doubts about Ian’s argument for why the ex post view has perverse implications (and, thus, by implication his argument for preferring the ex ante to the ex post view). Note first that Ian refers to perverse implications. I take it that the two implications he has in mind is, first, the perverse incentive to procrastinate and, second, the resulting imposition of a bill on society as a whole. I set aside the second perverse implication, partly because the ex post solution as such is silent on who should pick up the bill, partly because, if the borrower knew the rules, then presumably so did the lender, in which case the lender might have insisted on safeguards to prevent procrastination. Focusing then on the first perverse implication, I did not see why Ian is entitled to assume that the borrower is “aware of this rule” (p. 12). Moreover, the principle of diminishing liability-responsibility is a moral principle and does not as such affect what incentives people have. True, a legal rule reflecting this moral principle might bring into existence the relevant incentive, but now perversity seems to adhere in the legal rule, not the moral principle. Finally, I am unsure how the relevant perverse incentive, if such a thing exists, can be avoided given the Parfitian reductionist element in Ian’s account of personal identity. Even if the amount that Ian is supposed to repay does not diminish over time, he still has an incentive to procrastinate given that his later self who will have to pay the money will be connected to his present self to a lesser degree.
Finally, I wonder how central Ian’s account of the basis of moral equality is to the solution he proposes. Suppose people are not moral equals. Suppose, say, that people’s (unequal) moral standing results from their specific level of agential capacities such that moral standing is a scalar thing. Suppose, finally, that people for that reason are not entitled to equal initial opportunities, but to opportunities that fit their level of moral standing, e.g., if you have slightly greater agential capacities than I have, you are entitled to a slightly better initial opportunity set than I am. Call the relevant theory the starting gate interpretation of inequality of opportunity. Even if we presuppose this theory, something like the harshness objection still seems to get a grip, e.g., I can make risky choices and much later end up much worse off than you in a way that, despite the fact that my entitlements regarding initial opportunity sets were lesser than yours, we might find objectionable from the point of view of the relevant ideal of inequality of opportunity. In response to this problem, one can defend an ex ante solution analogous to the one that Ian proposes, i.e., one that consists in eliminating certain risky items in our initial opportunity sets in view of a reductionist view of personal identity over time. But if opacity respect is not doing any essential work in this Carterian response to the sketched harshness of objection to the relevant ideal of inequality of opportunity – as, surely, it does not, since ex hypothesis moral status is not based on opacity respect but on the degree to which one enjoys agential capacities – it is difficult to see how it can be doing any essential work in Ian’s response to the harshness objection to the ideal of equality of opportunity on the starting gate interpretation.[1]
[1] I thank Ian Carter for useful feedback on this précis.
Hello everyone. Apologies for the long comment. I have three points, and feel free to ignore any or all of them depending on what is most interesting! And as an aside, I shared Kasper’s worry about perverse incentives, to which one might add the point that all sorts of moral precepts contain similar “perverse incentives,” like the moral precept “you don’t have to repay moderate debts if you come down with a debilitating fatal illness,” which perversely incentivizes people to hold off on paying debts just in case they get very sick. That’s no argument against the moral precept, though!
First point: on page 7 the article suggests that if we move away from seeing people as temporally unified, we’ll start to care more about welfare than freedom, because freedom concerns possibilities for future action, and absent continuity with one’s future self, possibilities for future action are no longer a meaningful concern. I am not sure this is accurate. Possibilities for future action are no longer a meaningful prudential concern, but I can have concerns beyond prudential concerns, including concerns that support my present freedom.
If for instance I want to be free to do X at time t for the sake of some other person’s welfare at time t+n, I don’t see how my other-focused desire weakens our duty to respect my freedom. If we assume a further fact view about personal identity, it’s fine to say that peoples should be free not only for their own sake but for the sake of allowing them to do things for, or to coordinate with, other people, including people in the future. If we can say this if we accept a further fact view, why can’t we say it if we accept a reductionist view, minus the part about being free for one’s own sake?
More broadly speaking, it seems to me that the truth of reductionism might require us to re-describe values we took to be prudential in non-prudential terms, but this doesn’t mean we’d have to give up these values. Couldn’t freedom be accommodated via this kind of redescription?
Second point: I briefly thought the paper was going to go in a different direction. Here’s the argument I thought it was going to make: people who make imprudent choices are likely not going to end up very integrated with their future self. Someone with prudential concern for their future self would not make imprudent choices, and one’s future self, looking back on an imprudent choice, is liable to be antagonistic towards one’s earlier self, rather than seeing the earlier self as entirely integrated with one’s future (that is, present) self.
So, we have an argument for restricting imprudent choices: there is likely to be little integration between the imprudent person and their later self if this person is presently inclined to acting imprudently, and thus for the sake of the later self we might restrict these imprudent choices. Any thoughts on this argument? We do have to assume the level of integration will drop below that which is sufficient for opacity respect, or modify/give up opacity respect, but I for one am happy with either option (and it sounds like maybe Kasper is also happy with giving up opacity respect, in light of his last point).
Third point (related again to Kasper’s last point): what if I am an “integration savant” – a person who arranges my life so that I’m certain that my past, present, and future selves are and will be highly integrated? I’m very good at predicting and planning for the future, very good at owning my past choices, very inclined to do these things I am good at, and so on. Maybe I also have religious commitments or other personal commitments which suggest that a high degree of integration is both necessary and choiceworthy for me. Is it really respectful to me if we keep me from making certain choices for my future and if we justify this restriction by reference to how normal people behave? I’m not normal! I’m a savant! I want society to make an exception for me. I don’t want my talents to be shielded behind opacity respect any more than a gifted artist wants their talents to be shielded behind the attitude of someone who just says “oh, that’s very expressive” when confronted with any half-decent work of art.
Or consider a nearby argument: there is someone else who is good at predicting integration. They paternalistically make decisions for their child that conflict with the state’s paternalistic restrictions. So for instance an Amish parent paternalistically pulls their kid out of school early. The parent argues that experience shows that Amish people, like the kid, tend to be integrated such that they are totally fine with this supposedly imprudent choice made during childhood. Can the state prevent the parent from doing that on the basis of the importance of education for people generally, even though in this case the parent (let’s say) knows enough about their kid’s future to justify this supposedly imprudent choice?
More broadly speaking, I worry about how we’re supposed to come up with societal standards for what counts as imprudent that will be strong enough to rule out the sorts of imprudence covered by the argument in the article but not so strong as to block people from living lives outside the norm (like religious lives of certain kinds, or the lives of child actors who grow up into adult actors, and other things like this).
Many thanks to PEA Soup for organizing this discussion, and to Kasper for his careful summary and generous comments. The four points at the end of Kasper’s précis raise some interesting and difficult questions. I’ll take the first two points together and then post separate comments on the third and fourth.
Kasper’s first point is that my response to the harshness objection seems to apply only where the choice in question has high stakes AND the bad outcome occurs in the distant future with respect to the time of the choice itself. This may be true, but I don’t think it stops me accommodating Fleurbaey’s and Anderson’s view that the motorcyclist ought to be required to wear a helmet. The accident occurs shortly after the choice not to wear a helmet but, if the accident has lasting effects, such as disablement, the “very bad outcome” referred to by Kasper extends far into the future. The same is true of the 21 year old in a luck-egalitarian regime who blows her entire initial endowment in a casino (the example with which I opened my article). The gambler and the motorcyclist seem to be on a par in this respect. Doesn’t the weight of the harshness objection, for “ambivalent egalitarians”, depend very much on the lasting effects of the choice in question? Isn’t that why it seems so “harsh” to make the risk-taker pay the whole cost of her choice in the examples that are typically put forward? If so, Kasper’s “friendly amendment” in his second point, to the effect that “the longer into the future that the [cost of a failed] risk will materialize, the more willing we should be to eliminate that risky option from the initial set of options”, indeed looks plausible.
At this point the worry for the ambivalent egalitarian seems if anything go in the opposite direction: since virtually all risks seem at least potentially to have knock-on effects in the long term, should people’s freedom to take risks not be reduced to a bare minimum at all times? Taking the harshness objection too seriously might lead to the virtual abandonment of equality of opportunity in favor of equality of outcomes. To counter this worry I appeal to our respect for choosers as temporally extended persons, an idea which is itself based on the deeper moral attitude of opacity respect (applied diachronically so as to “pull individuals together” in the eyes of others). I try to make these two perspectives compatible in the following way: on the basis of opacity respect, we adopt an external perspective, recognizing persons as equal and temporally extended, we accord them substantial initial opportunities, and we hold them fully responsible for the choices that they actually make; and on the basis of an empirical generalization about the less-than-full diachronic integration of individuals, we limit the initial opportunity to place heavy costs on temporally distant selves. I’d like to think this construction holds water as a matter of principle, even though I didn’t say where exactly the lines should be drawn.
Regarding Kasper’s third point: in light of Kasper’s perceptive remarks about the incentive to procrastinate, I’m wondering if my reference to that incentive wasn’t a red herring. I included it in order to emphasize the counterintuitive nature of the implication I was pointing to, but perhaps on reflection I should have left it out. The key point that I wanted to make was that it seems perverse to reward someone for procrastinating in the honoring of their commitments. (The person we would be rewarding IS the same as the person who took on the commitment, IF we have ascribed to the same biologically continuous individual the range property of temporally extended moral personality – see p. 12, fn 9 in my article). One can make this point without claiming that the diminishing-responsibility view creates an incentive to procrastinate or that the diminishing-responsibility view is unique in creating such an incentive.
At a more basic level still, it does seem to me plausible that liability-responsibility (e.g. responsibility for a debt incurred) is inherited step-by-step from temporally proximate past selves, and that the plausibility of this step-by-step inheritance of our responsibilities is not affected by the adoption of a reductionist perspective on persons, assuming the proximate selves to be strongly connected.
I’ll come back to Kasper’s remarks about incentives if any useful responses occur to me.
I turn now to Kasper’s fourth and final point. Opacity respect is of course doing essential work, in my account, in favoring equality of opportunity over inequality of opportunity. This much isn’t contradicted by Kasper’s fourth point. Without opacity respect people wouldn’t be equal, nor therefore would they be owed anything equally. What Kasper’s point denies is the idea that opacity respect is doing any essential work in my discussion about ex ante restrictions of opportunity, which has to do with the relation between opportunities and outcomes. Basically, if Kasper’s fourth point goes through, then I have unwittingly written two papers rather than one: a first paper about equality vs inequality, extending the notion of opacity respect diachronically so as to justify the notion of “temporally extended moral personality” understood as a range property; and a second paper about opportunities vs outcomes, justifying the ex ante restriction of initial opportunities on the basis of people’s less-than-full integration over time. Kasper is suggesting that the first paper has no bearing on the second.
This last point of Kasper’s is clever and also raises deep issues concerning the role of opacity respect in my account of equality of opportunity. I’ll need to think more about it, but here are two tentative responses.
First, my account does seem to differ from a reductionist non-egalitarian account not only in terms of acceptance or rejection of equality but also in terms of the importance it attributes to initial opportunities. On the reductionist non-egalitarian account, there will be people who are given very few initial opportunities to plan for the future, because they show such a low degree of diachronic integration (or at least, there will be people of whom we would say with hindsight that they SHOULD have been given very few such opportunities). These are the people who live from day to day, or from week to week, and so meet only the minimal threshold of diachronic integration. They might include people who share Galen Strawson’s strong aversion to unifying “narratives”. My view says that we should respect these people as temporally extended moral persons because we should view them as integrated (i.e. as meeting the minimum threshold of integration and therefore as possessing the relevant range property), and that we should therefore accord them substantial initial opportunities. This difference seems to me to go beyond the mere difference between equality and inequality.
Second, the reductionist non-egalitarian position imagined by Kasper (a position that includes respect for temporally integrated agents understood as a matter of degree) doesn’t itself seem to address the harshness objection. Yet the harshness objection is basically about opportunities vs outcomes, not about equality vs inequality. If Fleurbaey’s motorcyclist is really extremely highly integrated (he really cares immensely about feeling the wind in his hair, the risk is very carefully calculated, he thinks all the time about, and is strongly connected to, all of his future selves that might come about in all of the various possible outcomes of his risk), then Kasper’s reductionist non-egalitarian will presumably let that motorcyclist take the risk on the grounds that he is highly integrated, and the motorcyclist will then pay the full cost if he has an accident (assuming Kasper’s reductionist non-egalitarian agrees with me about liability-responsibility). The reductionist non-egalitarian might then feel the force of the harshness objection, but I can’t see how she can respond to that objection except by appealing in a pluralist spirit to some exogenous value. On my respect-based account, by contrast, someone who is very high up on the scale of degrees of diachronic integration and someone who is only just above the minimal threshold are accorded the same initial opportunities. My account confirms the status quo existing in many countries today (not, of course, regarding equal starts but regarding the use of crash helmets). And whatever you might think of that in terms of fairness to the more integrated vis-à-vis the less integrated, it does qualify as a response to the harshness objection and, more generally, to the ambivalence felt by many respect-based egalitarians when contemplating starting-gate equality of opportunity.
Hello Danny, and thanks for the stimulating and challenging comments! I’ll give them some thought and try to get back to you later. First, on your aside about incentives: as you’ll have seen above, I concede Kasper’s third point (for now, at least!). This said, I wouldn’t like to generalize this concession into the claim that any incentive-based argument is as good (or bad) as any other. After all, many moral and political philosophers do think incentives are a legitimate concern in evaluating principles (Rawls on the difference principle and Buchanan on secession immediately spring to mind), and that it matters that some incentives are stronger than others. For example, perhaps one reason we don’t take very seriously the perverse incentive you point to, created by the principle “you don’t have to repay moderate debts if you come down with a debilitating fatal illness”, is that the incentive will normally be quite weak.
And, just to argue against myself for a moment, a problem people sometimes point to in the idea of mandatory insurance is that of moral hazard.
I do think the moral hazard idea is relevant, and I think both Rawls and Buchanan are talking about institutional design (or as Kasper puts it, ‘legal rules’) rather than more fundamental moral principles when they invoke incentives, so I’m not sure Kasper’s original point (which is that perverse incentives are an issue for institutions to deal with, not for moral theory to worry about) might still stand.
Or, if one disagrees with my reading of Rawls or Buchanan, one could take the GA Cohen route with Rawls, or the basically everybody but Buchanan route with Buchanan, and argue that incentives shouldn’t even be in the picture in the first place when working out the relevant moral principles. It’s up to institutions to handle (as best they can) whatever potential perverse incentives exist, and maybe institutions can work around them, or maybe there will always be some moral hazard or some other issue with whatever solution we pick if we want our solution to respect the moral facts of the matter. C’est la vie, etc.
But, I don’t want to belabor the incentives point too much!
Danny,
Thanks again for the three points in your first comment. Here are some thoughts on each.
1. It’s true that, if you exist at t but not a t+n we can still be concerned about your freedom at t to perform actions that occur at or just after t (but before t+n), and that those freedoms, and the various consequences after t+n of their being exercised, can be important to you for non-prudential reasons. That much is certainly true, so the extreme reductionist can accommodate that much concern for your freedom. But we (and you) won’t be concerned about your freedom at t to do actions occurring at t+n or later than t+n.
Any statement about freedom implicitly contains two time indexicals: the time of the freedom possessed, and the time of the action one is said to be free to perform. As agents, we make plans, including long-term plans, which include so-called “complex actions” comprising combinations of many more-or-less spatio-temporally specific actions extending into the future. Thus, on the common-sense view our freedom at any one time consists in a vast action tree of alternatives extending into the future. If I come to an end before t+n, we cannot meaningfully say “I am free now (at t) to perform the specific action x occurring at t+n”. I am of course free/able to make plans about how future people ought to act, but their actions won’t be my actions, and their freedoms are not my freedoms.
2. I need to think more about your second point, but the alternative argument you present here seems to presuppose a perfectionist position on diachronic integration, such that high levels of diachronic integration are a good thing and to be promoted. Is that right? Where you suggest that we could restrict imprudent choices “for the sake of the later self” I take you to be saying that we can be concerned with the degree of integration of the later self (with the earlier self). I’d like to avoid that perfectionist premise (see top of p. 10). My argument instead suggests that we can be legitimately concerned with the degree of freedom of the later self.
On “giving up opacity respect”: I believe Kasper is happy to give up opacity respect, but only because I’ve talked with him about it on previous occasions. I read his last point as a purely logical one that identifies the possibility of a reductionist non-egalitarian who dispenses with opacity respect AND with any concern for basic equality, but is still moved by the harshness objection and can still adopt my response to it.
If you do give up on opacity respect, then the onus is on you to find an alternative explanation and justification of the idea of basic equality, unless you simply reject basic equality and deny that people are due equality of anything at all as a matter of entitlement. (In the latter case this might be because you are a purely instrumental egalitarian, or because you believe greater recognition respect is due to the more talented/virtuous/integrated than to the less talented/virtuous/integrated, or both.)
3. I’m not sure whether your third point is limited to questioning my extension of opacity respect to cover degrees of diachronic integration, or whether it amounts to a more general questioning of opacity respect – and therefore applies also to basic agential capacities at any one time. (If opacity respect covers only basic agential capacities then it doesn’t necessarily veil the talents of the great artist, but I take it you were only using that example as an analogy.) In the article I argued that the two perspectives – synchronic and diachronic – are inseparable from the point of view of opacity respect. This won’t move you unless you already accept opacity respect within the synchronic perspective (I argued at greater length for the latter in my 2011 Ethics article). If you do follow the argument where it leads, there may indeed be some bullets to bite. If you instead deny the consequent (“I don’t like that implication, therefore I reject opacity respect”), then see the previous paragraph (at the end of my point 2, above). For my part, although my main concern here is distributive equality, I share the view of relational egalitarians like Anderson that there is something basically disrespectful about evaluations (especially by the state) of basic agential capacities when made with a view to establishing the nature of people’s basic rights, and that this holds regardless of where a particular person’s basic capacities lie on the scale from the minimum threshold upwards.
In answer to the question you pose at the end of your paragraph presenting the Amish example (“Can the state prevent the parent … on the basis of the importance of education for people generally …?”), I would say “Yes”, because education of children has to do with the FORMATION of moral agents, whereas I was assuming recognition respect to apply to fully formed moral agents (bearing in mind that the process of formation is gradual and that recognition respect becomes appropriate in degrees during that process). However, this answer works only because your example involves a child.
I guess the following comment is somewhat tangential to this discussion, inasmuch as it queries the validity of the ‘harshness objection’ as a criticism of a theory of justice. So this places me in the group Kasper refers to in the following passage:
“Some will dismiss the harshness objection, because they think we have a duty to respect people’s agency and that this includes respecting people’s exercise of their agency in relation to the intrapersonal distribution across their lives. People who take this view, might add that while we might not be required as a matter of justice to help badly off 71-old Jerry, we might have reasons of compassion etc. to do so.”
Justice is widely seen as being only one in a plurality of basic moral values/virtues. Other members of this plurality include honesty, generosity, courage, tolerance and – not least – compassion. There obviously can be actions that instantiate the presence (or absence) of more than one of these values in one and the same act. But that is no reason to conflate them. Countless historical events and literary works confirm the distinctness of each of these values. Indeed, the common expression ‘tempering justice with mercy’ is itself such a confirmation. There is a kind of conceptual imperialism involved in subsuming those other values under the requirements of justice.
The duties which justice imposes on us are generally thought to be ones which should be legally enforced. We don’t ordinarily go to the time and expense of consulting our lawyers when the interpersonal conflicts in which we’re occasionally engaged arise out of perceptions that our opponents’ courses of action are deficient in regard to those other values: such deficiencies are not regarded as morally warranting legal prohibition. Which is very far from saying that they’re morally permissible. Rather, the appropriate responses to them are usually thought to range from mild reproaches and reprimands to disassociation and outright stigmatization. Of course aid should be given to the motorcyclist whose injury is due to his having chosen not to wear a helmet. To do otherwise is morally indecent. But it’s not unjust.
Hi Hillel,
No doubt the Fleurbaeys and Andersons of this world will agree with you that “of course aid should be given to the motorcyclist whose injury is due to his having chosen not to wear a helmet”, but will disagree when you add that the state should NOT send an ambulance to save that motorcyclist. I take it you would make this additional claim, unless you think that the state can have just grounds for conscripting citizens’ help in such cases. Anyhow, this is just an exchange of normative intuitions, and neither Fleurbaey nor Anderson satisfactorily address your conceptual point, as far as I’m aware.
Personally I’ve never been very happy with the expression “harshness objection”, precisely because it implicitly suggests an appeal to the value of welfare and the virtue of compassion. One of the aims of my article was to find an alternative way of capturing the discomfort felt by many egalitarians about such examples – one that appeals to respect for agency. The thought was: if I could succeed in this aim, then I could bring on board those who, to cite Kasper again, “dismiss the harshness objection, because they think we have a duty to respect people’s agency”. This seemed to me possible on the basis of a more nuanced account of personal identity, as long as we remain sensitive to the danger of such an account leading us in the direction of Parfitian welfarism (see the “dilemma” presented in sec. 3 of my article and the exchange above with Danny Weltman about welfare vs freedom).
Thinks: isn’t it possible to use the term “harshness” to refer to a failure to be fully just? When we say that a judge gave a “harsh sentence”, do we always mean that the sentence was “strong on justice but short on compassion”? Or might we sometimes mean that the judge’s interpretation of justice itself was too strongly punitive?
We might, perhaps, say that the sentence was overly punitive as a matter of justice because we think the judge was being over-simplistic in her interpretation of various component parts of the concept of justice as applied to the particular case. If so, it doesn’t seem absurd to say that the view that it is just for the state not to send the ambulance to save the motorcyclist is over-simplistic in its implied interpretation of justice. We might say this, for example, if we think that view presupposes a naïve conception of agency or personal identity. On this basis, it might be claimed that the “harshness objection” raises a question of justice after all.
Hi Ian. Some thoughts:
1) You note: “If I come to an end before t+n, we cannot meaningfully say “I am free now (at t) to perform the specific action x occurring at t+n”. I am of course free/able to make plans about how future people ought to act, but their actions won’t be my actions, and their freedoms are not my freedoms.”
This is true, but can we get from here to the conclusion that, if reductionism is true, we should care less about my freedom and more about welfare? You claim that we can, because “we assign people rights to individual freedom because we respect them as temporally extended agents” (page 7 of the article). But why think this is the reason we assign people rights to invididual freedom? So for instance a utilitarian assigns people rights to individual freedom for the sake of promoting utility, not because the utilitarian respects anyone as a temporally extended agent. Similarly, can’t a reductionist assign people rights to individual freedom for some reason other than that someone is a temporally extended agent? What if I assign people rights to individual freedom for the sake of other people’s rights to individual freedom, for instance?
2) I took the alternate approach to be independent of any value attached to diachronic integration. If there is high integration, then there’s nothing to worry about, because someone will act prudentially. If there is a low level of integration, then prudence won’t cause someone to protect the interests of their future selves, and they’ll act in risky ways, such that we have a reason to step in for the sake of those future selves. So, the concern isn’t to make sure people end up highly integrated, but to make sure that people who are not going to be highly integrated don’t end up trashing the lives of future people. The people who do end up highly integrated won’t tend to take these risks in the first place.
3) I had in mind Kasper’s last point, and your gloss of it that you might be offering two arguments, the latter of which doesn’t have much to do with opacity respect. My point was the positive version of the Strawson example you offered in your reply. You point out that we wouldn’t want to overly restrict Strawson’s options on the basis of his minimal integration. Fine. My worry was about people on the other side of the spectrum, who are so good at and inclined towards integration that they’re asking for permission to be allowed to take risks. In your reply you point out that this doesn’t deal with the harshness objection, which I think is right. In response someone taking this line would maybe want to say at least two things:
A) Some of the harshest bits in the harshness objection come from people who are rendered so badly off that it’s hard to see how they could be living a good life at all, let alone a good, highly integrated life. This is the case with a motorcyclist, I think, who ends up very bad. We would expect someone who is highly integrated to avoid those sorts of risks, and instead aim for risks like the gambler. If they end off very poor, they’ll do much better, we might think, than most poor people, because they’ll think “I chose this, I endorse this, this is where my life has been aiming” and so on. So maybe it’s not going to be harsh if we don’t compensate integration savants for their ill fortune.
B) Well-being is, perhaps, a rather pliable thing. An integration savant will do a good job finding well-being in situations they end up in, so long as these are situations that are a result of their free choice (as opposed to the paternalistic restrictions of their society).
I’m picturing some libertarian individualists chafing at paternalism and asking how it could conceivably be a form of respect to keep them from living their lives the way they are capable of living them. They’re not hot-headed young motorcyclists, they’re just, for instance, gamblers who are fine with living a monastic life if they don’t strike it rich.
There are of course plenty of reasons one might reject A) and B), but I think some people might want to endorse them.
One response you could give is just to deny that people can be this integrated, or, better, that anyone can know that they will be this integrated in the future. But maybe opacity respect blocks that response.
Thanks, Danny, for your further thoughts! Unfortunately I have a lot of teaching right now, so the following responses are not only slow in arriving but also probably incomplete. Still, here goes.
1. I certainly agree with you that there can be other, non-respect-based reasons for valuing freedom. In case you’re interested, I discussed these in ch. 2 of my book A MEASURE OF FREEDOM (1999). In the article we’re discussing, I restricted my attention to a respect-based perspective (for reasons given at p. 3 of the article and in my response to Hillel above). So you may be right that the claim I made on p. 7 is too sweeping. Still, I think the general point in my first response to you, about trees of alternative courses of action spreading out increasingly into the future, holds for all perspectives on the value of freedom: the point is that, if persons persist in time for much shorter periods than those assumed by the common-sense view, then there is simply less freedom in the world for us to be interested in. There will be the same number of persons at any one time, but the sum total of freedom won’t be the same, even if we sum over generations.
2. If your alternative proposal isn’t to promote diachronic integration as a value, it starts to look closer to mine. Except that mine rests only on an empirical generalization to the effect that people are not fully integrated over time (opacity respect rules out particular assessments). From there it gets to the idea that respect for persons, understood as temporally extended over whole biological lives, needn’t entail an exclusive concern with their INITIAL freedom; it’s not based on the claim that there will be future persons (supervening on the same biological life) with claims against the earlier persons. That claim would rest on a more straightforward reductionist perspective, whereas I try to get from the reductionist perspective, via opacity respect, to a view that nevertheless treats all normal persons as being temporally extended over whole biological lives.
3. Regarding your case of the savant: I said in my earlier response that for some people there may be bullets to bite, and this may be one of them. That said, it would of course be a very strange libertarian, whether of the left or the right, that favored state assessments of individuals’ prospects for diachronic integration in order to decide on how much freedom they should get initially and how much later on in their lives. Libertarians typically assume the further-fact view of personal identity, so the question of levels of integration doesn’t arise for them. Rather, they think everyone should be treated like your savant. If they simply assert this, it seems to me that they are begging the question against my argument.
If you disagree with those libertarians about everyone being appropriately treated like the savant, and you want instead to make a distinction between the savant and less integrated people in how they are treated in terms of basic rights to freedom, you must advocate a principle that presupposes the kinds of assessments just mentioned. And if you agree with me that there is a sense in which those assessments are themselves disrespectful of persons’ agency, then we simply have a conflict between two possible meanings of respect.
My argument is supposed to be a way of getting close to the libertarian view – inasmuch as I do want to treat people as temporally extended wholes – but without assuming the “further fact”. We start from an empiricist, reductionist point of view on persons, but we adopt an external point of view as a matter of respect (in virtue of an attitude that becomes appropriate in the presence of a minimum of agential capacities), and this explains why we tend nevertheless to treat persons as temporally extended over whole biological lives.
Regarding your final two sentences: I do give that response in my article, in the second paragraph of p. 14. I think the objection itself is blocked by opacity respect, which means that the response to it is given arguendo. I guess the article could have been clearer on this point.