Fair equality of opportunity in our actual world

Introduction

John Rawls (2001) defined the principle of fair equality of opportunity (FEO) as the requirement that

those who have the same level of talent and ability and the same willingness to use those gifts should have the same prospects of success regardless of their social class of origin, the class into which they are born and develop until the age of reason. (p. 44)

This principle, ever since Rawls enunciated it, has been battling a war on two fronts. Looking at things from the beginning of life, it must be conceded that parents have broad liberty to raise their children as they see fit, engaging in a greater or lesser degree of nurturing and advantage giving according to any one of a broad array of permissible childrearing styles. So it seems that, because of this, the social class of one’s origin must inevitably have a great influence on one’s prospects of success. And if we look at things from the other end, asking what it is that determines whether a child, taking the social circumstances of her early life as given, succeeds in life, we notice that in a meritocratic world her array of skills is the main determinant of her fortune on the job market and thus a heavy influence on her success in life. And yet there’s so much more that goes into the cultivation of a skill besides talent and ability, such as being fortunate enough to be taken under the wing of an expert in one’s desired field, and this seems to undermine FEO all over again.

What, then, could possibly be the place of FEO in a world of broad parental liberty and meritocracy, that is, in our actual world? There are actually two questions here, the specific question depending on whether broad parental liberty and meritocracy are admitted to be part of our world de jure or de facto. In this essay, I make both admissions and consequently seek to answer the two questions together by showing that broad parental liberty, meritocracy and FEO are compatible both morally and practically.

One element of my answer will be predictable and boring: broad parental liberty, meritocracy, and FEO can be made compatible by taking each of them to have merely pro tanto moral force. But having made this move there is still a great deal to be done. Most urgently, we need to formulate a political morality that contains all three elements. This requires explaining how the conjunction of broad parental liberty, meritocracy, and the FEO can constitute a coherent political outlook. And then, having (hopefully) managed that, it remains to defend that outlook against objections. Those two goals constitute my agenda here.

I begin, in ‘Broad familial liberty and meritocracy’, by arguing that broad familial liberty and meritocracy are each part of our actual world and that there is something to be said, morally, for their being so. ‘An argument for FEO’, which follows, is the heart of the essay; it contains my preferred variant of the principle of FEO and an argument for it. The version I endorse is rather narrow, dealing only with the distribution of prospects for desirable jobs and offices – and my defense of it focuses on explaining what is distinctively good about certain jobs. In support of FEO so construed I appeal to the intuition that there ought to be a fair competition for that good. The key move I make here, in contrast to other defenders of FEO, is to argue that what’s distinctively good about getting the right job is that it ensures the avoidance a particular bad – the bad of spending one’s time doing something that one finds boring.

The rest of the essay is an effort to answer some very powerful, well-formulated objections to FEO, mostly from Richard Arneson and Andrew Mason. In ‘The luck-egalitarian objection to FEO and the question of scope’, I confront the luck-egalitarian objection that asks why we should adopt a principle that allows prospects for jobs and offices to be affected by talent but not social circumstances, given that the two influences are equally arbitrary from a moral point of view. I argue in response that what counts as the right job for a person depends on what her talents are, and therefore, there is no single competition for the distinctive good of the right job but rather innumerable mini-competitions among people with similar talents. ‘The separate principle objection’, meanwhile, addresses the worry that the distribution of jobs and offices shouldn’t be regulated by its own principle but rather whatever principle in our political morality that addresses the distribution of goods per se. I respond by pointing out that the ideal of a fair competition is a non-consequentialist ideal, and therefore, it would make no sense to fold it in with a principle that addresses the distribution of goods. Finally, in ‘The childrearing objections’, I answer the objection that FEO has unpalatable implications for how parents are morally permitted to raise their children. I argue that the ideal of setting up a fair competition simply has no normative purchase on ordinary individuals; it applies only to states. Consequently, FEO has no implications at all for how parents are morally permitted to raise their children. The last section concludes.

Broad familial liberty and meritocracy

In this section, I aim to establish that broad familial liberty and meritocracy are each valid and in effect in our actual world. By ‘our actual world’, I mean developed liberal democracies. And by ‘valid’ I mean that they are each part of the true political morality – the true theory of what the state is morally obligated to do. Note that I am not claiming that justice requires broad familial liberty or meritocracy. I assume that justice is one of many elements of political morality; the state morally ought to do what is just, but it also morally ought to do other things like make sure everyone drives on the same side of the road. I am simply sidestepping here the question of whether broad familial liberty and meritocracy are just. Of course, as G.A. Cohen (2008) emphasizes in his own effort to distinguish justice from other interesting questions in political philosophy, ¹ the state should strive to be as just as it can while giving other kinds of consideration their due (pp. 284, 305). This means that in trying to establish that broad familial liberty and meritocracy are elements of the true political morality, we must beware of the possibility that they are incompatible with the requirements of justice, as that would constitute a strong counterargument.

As it happens, however, there is a strong case to be made that the state’s noninterference with broad familial liberty is a requirement of justice. Allowing family members to interact with each other as they see fit (within certain loose boundaries) is necessary if they are to realize the goods that familial life makes possible, which are some of the most valuable of all goods – namely, the spontaneous selfless provision and receipt of love, assistance, nurturing, companionship, and so on. This sort of argument has been made before, and with much greater care and rigor than I can manage here, by other theorists. ² I will simply assume that their arguments succeed in establishing that broad familial liberty is a requirement of justice. And as to the real-world situation, I will rest content pointing out that in every developed liberal democracy parents are in fact granted this broad liberty.

As to meritocracy, things are more complex. The principle, as I want to defend it, is the claim that jobs and offices should be awarded solely on the basis of the relevant qualifications. Of course, it’s difficult to define the idea of ‘relevant qualifications’. Supposing, for instance, that most of the clients of Business X are racists, Business X might turn a higher profit by making sure that all of its employees are of a certain race. This, on a broad understanding of what a qualifications is, would make race a qualification for employment in business X. ³

In spite of this problem, I’ll adopt the broad understanding. ⁴ Specifically, a qualification for a job or office will be any characteristic that would aid an individual in discharging it in a way conducive to the realizing the goals of the organization containing that job or office. We can endorse meritocracy, given this understanding of qualifications, even if we are strongly opposed to discrimination, since we can also maintain that an anti-discrimination principle is a weightier element of the true political morality.

By way of establishing that our actual world is meritocratic, I will simply note that free market societies, such as ours, must be meritocracies by and large, ⁵ since in a free market organizations are generally permitted to pursue their goals to the best of their ability, and this, by the earlier definition of ‘qualification’, requires hiring the most qualified people to fill vacancies.

Having made note of this, we are now in a position to offer a simple indirect moral argument for meritocracy. This argument begins by noting that surely some principle of respect for negative liberty is an element of the true political morality. And if we decline to interfere with individuals’ freely chosen actions, as required by that principle, one result of this will be the broadly meritocratic distribution of jobs and offices since, as just mentioned, organizations generally will, when left to their own devices, award jobs and offices to those most qualified to fill them. So meritocracy is an inevitable result of something that our political morality requires, namely respect for negative liberty. (The principle of respect for negative liberty is, no doubt, sometimes outweighed by other principles, but that need not concern us, since for the purposes of this essay it is sufficient to establish a pro tanto moral defense of meritocracy.)

Furthermore, we can make the following direct argument for meritocracy: A system that allocates jobs and offices based on qualifications promotes economic growth better than any other possible system, holding fixed the other relevant features of society such as its educational system, wage structure, and so on. ⁶ Overall, it is conducive to economic growth that organizations achieve their goals, since those goals generally include increasing the production of some good or service. If we add to this the principle that the state is morally obligated to promote economic growth ⁷ then we have in hand a complete argument for meritocracy as an element of the true political morality. Surely most of us do believe that the state has this obligation. Strong economic growth (as opposed to weak economic growth) in general has the effect of reducing inequality ⁸ and unemployment, and most of us would accept, with respect to at least one of these goals, that the government is obligated to pursue it. ⁹

Of course, these goals often come into conflict with other goals, and I do not mean to say that the former goals are always more morally urgent. Again, I am defending meritocracy merely as a pro tanto requirement of the true political morality.

An argument for FEO

In this section, I intend to defend the principle of FEO and show how it can be part of a coherent political outlook that also contains broad parental liberty and meritocracy. First, however, we need to know just what claim FEO is.

On the statement of the principle from Rawls that I quoted at the outset, it says something about how ‘prospects for success’ should be distributed. However, it must be said that Rawls is quite inconsistent on the matter of what the currency ¹⁰ of FEO is. In other places, Rawls (1999) puts forward FEO as a principle regulating the distribution of prospects for ‘culture and achievement’ (pp. 63, 64), ‘income and wealth’ (p. 64), and ‘advantaged social positions’ (p. 62). And for Rawls (1999) FEO is embedded within a broader principle, the second principle in his two-principle theory of justice, that says,

. . . social and economic inequalities are to be arranged so that they are . . . attached to offices and positions open to all under conditions of fair equality of opportunity . . . (p. 266)

thus suggesting that FEO is to directly regulate the distribution of offices and positions and regulate the distribution of social and economic inequalities only indirectly.

I do not know what version of FEO Rawls really meant to endorse, but I have argued in another essay that the most plausible version of FEO will take only positional opportunities as its currency (Sachs, 2012: 335–340). A positional opportunity is an opportunity for a good such that improving one’s opportunity to access that good necessarily comes at the cost of diminishing someone else’s opportunity to access that good. Positional opportunities are, to put it simply, opportunities to access goods that are in limited supply.

That being the case, we should reject any version of FEO that takes as its currency prospects for culture and achievement or income and wealth, since we can make more of all of those things. ¹¹ One currency that emerges as quite appropriate, by contrast, is opportunities for jobs and offices, which is pretty close to Rawls’ ‘offices and positions’. Although we can affect the number of opportunities for jobs and offices, most obviously by stimulating or depressing economic growth, opportunities for jobs and offices are in many circumstances positional-for-practical-purposes – meaning that we can find ourselves in a position where there’s nothing we can do to make more of them – which I have argued is good enough to make such opportunities a proper currency for FEO (Sachs, 2012: 340). ¹²

Knowing this much about FEO, a clearer picture of the relationship between meritocracy and FEO begins to emerge: Since meritocracy regulates the distribution of jobs and offices themselves, saying that FEO regulates the distribution of opportunities for jobs and offices is equivalent to saying that FEO regulates the distribution of prospects for succeeding in the meritocracy.

But the picture is not yet clear enough. In order to fully specify a version of FEO we have to fill in the timing variable in addition to the currency variable. That is, we have to say at what time prospects for doing well in the meritocracy should be equal. I have argued elsewhere that FEO is most plausible when understood as requiring that prospects be equal at one time, as opposed to perpetually or intermittently (Sachs, 2012: 327–335). Specifically, prospects should be equal among those entering majority at a given time. This dovetails nicely with the idea that FEO should be understood as opportunity to do well in the meritocracy, since the meritocracy is a non-issue for minors, supported as they are by their parents or some other guardian.

While I have argued elsewhere that the currency and timing variables should be filled in in the way I have suggested here, I have never before argued that FEO is a true principle of political morality. All I have done elsewhere is show that my preferred version of FEO avoids some objections that doom other versions. Having gained a better grasp of the place of FEO in a meritocratic world, I now feel able to articulate with some precision just what it is about FEO that should appeal to us. I take on this task for the remainder of this section. In ‘The luck-egalitarian objection to FEO and the question of scope’ I return to the job of making FEO more precise, arguing there for a particular stance regarding among which individuals opportunities for jobs and offices should be equal at the time of majority.

To understand the moral force of FEO, we need a better grasp on what’s good about the thing it distributes, namely opportunities for jobs and offices. The first thing to say here is that opportunities for jobs and offices are merely instrumentally good; they are good because the possession of such opportunities leads to the possession of jobs and offices themselves. Now one might ask, as Richard Arneson (1999b: 488–497, 2013: 321, 322, 326, 327) and Stephen Kershnar (2004) have, why we should adopt a principle regarding the distribution of a merely instrumental good when we could instead adopt a principle regarding the distribution of the good for which it is an instrument. I want to say something about this before discussing what is good about jobs and offices.

I am taking as meritocracy as given, based on the argument I gave in ‘Broad familial liberty and meritocracy’. This amounts to taking as given a principle for distributing jobs and offices. The argument for meritocracy was not based on an analysis of the good of a job or offices. So it cannot be accurately said that we have already adopted a principle that is suitable to the kind of good that a job or office is. Rather, the argument for meritocracy was based on (1) the assertion that meritocracy is an inevitable result of something political morality requires, namely respect for negative liberty and (2) the observation that distributing jobs and offices in a certain way is conducive to reducing inequality and unemployment combined with the claim that the state is morally obligated to pursue at least one of these goals. So the moral heart of the argument for meritocracy comes from a moral analysis of the purpose of the state (which I breezed through), not a moral or evaluative analysis of jobs and offices. ¹³ Nevertheless, jobs and offices are suited to this kind of analysis, because they are indeed good. The question is this: ‘Once we have produced this analysis, what do we do with it?’ We cannot use it to derive a principle for regulating the distribution of jobs and offices; we already have one of those. ¹⁴ The best we can do, this being the case, is to use it to derive a principle for regulating the competition for jobs and offices. Once we have noticed that jobs and offices are competed for, as opposed to simply being awarded, it becomes possible to understand the competition for them as being a potential bearer of morally good or bad qualities. So we have a nice match: We have a moral/evaluative analysis of a good (the analysis of the good of a job or office, yet to be produced) and a feature of society that regulates the distribution of that good and seems apt for regulation via a moral principle. It seems only natural to put the two together, deriving the principle from the analysis.

So much for the question of why there should be a principle regulating the distribution of opportunities for jobs and offices. We can now, at last, move on to the question of what is good about a job or office.

There are, no doubt, many right answers to this question. I will focus on establishing just one of them. My view is that if we can agree that this is one of the goods of a job or office, then we should be able to agree on FEO as formulated earlier in this section.

I begin with a clarification: the good that I want to identify is not one that attaches to jobs or offices per se; rather, it attaches to the right job or office, in a certain sense of ‘right’ that I will specify.

Next, I want to call attention to a distinctive kind of good – the good of exercising one’s refined and fulfilled talents. Rawls (1999) describes that good this way:

Other things equal, human beings enjoy the exercise of their realized capacities (their innate or trained abilities), and this enjoyment increases the more the capacity is realized, or the greater its complexity. The intuitive idea here is that human beings take more pleasure in doing something as they become more proficient at it, and of two activities they do equally well, they prefer the one calling on a larger repertoire of more intricate and subtle discriminations. (p. 374)

It is possible to ground an argument for FEO in the claim that this kind of good can be realized in the carrying out of a job or office. Robert Taylor (2004) makes an argument like this (pp. esp. 340, 341). However, an argument of this sort is vulnerable to Arneson’s (2013) insight that this distinctive sort of good can be realized by means other than the carrying out of a job or office (pp. 320, 321). ¹⁵

Fortunately, though, we can make an argument based on Rawls’ analysis of this particular good that is inspired by Taylor’s argument but immune to Arneson’s objection. Assuming one is going to work for a living during one’s adulthood, one’s job or office determines how one will spend half of one’s waking adult life. A distinctively bad outcome that this situation makes possible is being consigned to spend this time engaged in an activity that doesn’t call upon one’s refined and fulfilled talents. ¹⁶ There is a distinctive kind of soul-crushing misery that accompanies suffering this fate – the kind of misery that many early twentieth-century social critics had in mind when condemning the assembly line. ¹⁷ It is a correlate of the distinctive good Rawls identified, but is not its mere absence. In fact, a single life – indeed, a single day of a single life – can contain the realization of both the good and its negative correlate. So even if one has opportunities, outside of work, to exercise one’s refined and fulfilled talents – in hobbies, for instance – that merely counterbalances, as opposed to undermines, the misery one is likely to experience if one’s job does not make use of one’s refined and fulfilled talents. No good is so excellent such that the time spent realizing that good makes it not-bad to spend half of one’s waking adult life doing something that one finds boring.

My conclusion is that there is a particularly brutal kind of bad that one avoids only by obtaining the right kind of job or office – one that calls upon one’s talents in the way specified. ¹⁸ So in the end, I do not actually have anything original to say about the good of the right job or office; rather, I have something to say about the bad of the wrong job or office. But this will do just as well for grounding an argument for FEO.

The argument will begin by noting that if our society’s basic structure is going to inflict a serious harm on various individuals then it matters morally how it is determined who suffers that harm. Of course, the harm in this case is attached to jobs and offices, and we already have a principle for distributing jobs and offices. However, as I argued earlier, it’s plausible to suppose that our evaluative analysis of jobs/offices can be appropriately deployed to ground a principle regulating opportunities for jobs or offices. Now since jobs and offices are positional-for-practical-purposes, so is the opportunity to avoid getting harmed in the distinctive way associated with ending up in the wrong job or office. Absent special circumstance, it seems that everyone should have an equal opportunity at the age of majority – that is, before the time at which they being to engage in activities for which they can be held responsible and which might affect their opportunity – to avoid this harm. Securing an equal opportunity means putting in place a level playing field in the competition to avoid the distinctive harm of the wrong job/office. ¹⁹ That is what FEO requires, and consequently FEO inherits moral appeal from the compellingness of the idea of a level playing field.

Notice that the idea of a level playing field is applicable only when we’re talking about distributing opportunities for jobs and offices; it is entirely irrelevant to the project in which employers are engaged – namely the project of distributing the jobs and offices themselves. (Think of how the idea of a level playing field is applicable when setting up a race but not applicable to awarding medals once the race has been run.) So my defense of FEO does not commit me to aligning myself with any principle regarding the distribution of jobs and offices. The principle of careers open to talents is such a principle, the way I understand it: it requires that employers fill their vacancies in a non-discriminatory way. So while Rawls endorsed careers open to talents alongside FEO, I am not committed to doing so. I find this a relief, since Arneson raises very compelling worries about careers open to talents alongside his objections to FEO. ²⁰

The luck-egalitarian objection to FEO and the question of scope

Having argued for my preferred version of FEO, I am now in a position to respond to some very important, well thought-out objections to the principle. The first objection concerns the scope of the principle, that is, the set of individuals among whom opportunities are to be equalized. Rawls specified that opportunity should be equal among those who are equal in terms of talent and willingness to put forth effort. A principle crafted in this way requires the state to nullify the influence on opportunity of anything aside from talent and willingness to put forth effort. As Rawls (1999) emphasized, this means that the state should nullify the effects of the social circumstances of one’s birth on one’s opportunity level, which is a morally appealing goal for the state to take up given how such circumstances are arbitrary from a moral point of view (pp. 63–65). The objection, however, is that the distribution of talents is morally arbitrary as well, and so if there is a case for neutralizing the effects of social circumstances due to their moral arbitrariness then there should be a case for neutralizing the effects of talent (Arneson, 1999a: 93–97). ²¹ So a more plausible version of FEO would have as its scope individuals who are equal in terms of effort alone. Call this the ‘luck-egalitarian objection’.

A sound response to this objection emerges when we recall what the distinctive good of the right job or office is – namely, the avoidance of the distinctive bad of being stuck spending half of one’s waking adult life engaged in an activity that doesn’t call on one’s refined talents. For people with musical talent, achieving that good means having a job as a musician. For athletic people that means having a job as an athlete. And so on for the other kinds of talent. ²² Consequently ‘doing well in the meritocracy’, which is what FEO requires that there be equal opportunity for, means something different to each person depending on her talents. Therefore, there is no sense to be made of a talent-insensitive version of equality of opportunity; equality of opportunity has to be relative to talent.

This means that what we might casually call ‘the competition’ for jobs and offices is really a cluster of mini-competitions among people with similar talents. That’s why it makes no sense to propose that the principle of equality of opportunity for jobs and offices should not be relativized to talent.

I suspect that the reason why this response to the luck-egalitarian objection has not been made before is that Rawls set us down the wrong track in constructing a defense of his talent-sensitive version of FEO. Curiously, his comments, excerpted above in ‘An argument for FEO’, regarding the good of being able to exercise one’s subtle and developed talents formed no part of his argument for FEO. Rather, Rawls (1999) seems to identify the good whose distribution FEO was meant to regulate – and thus presumably the distinctive good of a job or office – as ‘authority and responsibility’ (p. 53) and ‘the realization of self which comes from a skillful and devoted exercise of social duties’ (p. 73). Although Rawls does not elaborate on these comments, one could be forgiven for thinking that Rawls takes the distinctive good of a job or office to be its affording one the opportunity to skillfully execute the duties of some publicly recognized office. This being the case, the luck egalitarian can claim, at least somewhat plausibly, that the distinctive good of a job or office is going to be closed off to certain people purely in virtue of their talent because not everyone has the necessary talent to go into public service. The luck egalitarian could even allow us to expand our notion of publicly recognized office, so that it includes other professions, outside public service, that are generally recognized as socially necessary, such as medicine, law, and teaching, and point out that we are still forced to the same conclusion: that FEO offers absolutely nothing to certain people, namely those who lack a certain set of talents. ²³

Again, however, we can respond by conceding for the sake of argument that not everyone is born with the talent necessary for eligibility for public service while insisting that everyone is born with some talent or other. In other words, no one is born to do unskilled labor (though there is unskilled labor to be done and someone has to do it ²⁴ ). So FEO, understood as requiring equal prospects for succeeding in the meritocracy, does indeed offer something to everyone: it offers them a chance, on a par with everyone else of similar talents, to access the jobs and offices that call on the skillful exercise of the developed versions of those talents.

Still, however, while (hopefully) conceding that FEO so understood offers something to everyone, a luck egalitarian is likely to worry that it nevertheless offers more to the more talented. If in speaking of ‘equally talented’ people the principle is referring to people who have the same kind of talent and the same amount of it – which is exactly how we should understand it, I maintain – then that principle will not seek to equalize prospects among those with unequal amounts of the same talent. In other words, it will allow unequal talent to develop into unequal skill. And since skill is rewarded in a meritocratic world, FEO allows for a situation in which more talented people end up in a better position than less-talented people with respect to avoiding the distinctive bad we’ve been talking about. ²⁵

To arrive at an adequate response to this worry, we need to take a closer look than we have thus far at the distinctive bad of having a job that doesn’t call upon the skillful use of one’s developed talents. One obvious aspect of this bad that I have yet to point out is that it comes in degrees. In particular, a job can call on one’s developed talents while requiring a more or less skillful exercise of them. Or to put this more straightforwardly: among challenging jobs some are more challenging than others. And how challenging a job is depends on the skillfulness of the person trying to perform it.

Consequently, there is a way in which extra skill – and by extension, extra talent – is a disadvantage. For any given job, it will be less challenging for a particular person the more of the relevant skill she possesses. So for any given job, it’s better to be one of the less-talented people to get it, all else being equal. (I assume that in a meritocratic world the least-skilled people who perform any given job are still skilled enough to perform it competently.)

Here is an illustration. Suppose there are two people who are skilled enough to be competent surgeons, and that they each get a job whose sole function is the performing of appendectomies. One of these people, I shall stipulate, has skill to spare. Performing appendectomies is, for this person, experienced only as somewhat of a challenge. For the other person, by contrast, it is the perfect challenge: she can do it well, but only by mustering all the mental and physical resources she has at her disposal. All else being equal, I submit that the latter surgeon has more fully avoided the distinctive bad of spending her working life doing something that doesn’t call on the skillful use of a developed talent.

The phenomenon that the above scenario is supposed to represent is a very real one in our society. It is all too common for highly credentialed job candidates to wind up in positions for which they are overqualified. This is what happens when higher education becomes the norm as opposed to an exception.

What we should infer from all this, I conclude, is that in a meritocratic world where FEO is in place, the more talented will indeed have better chances of getting any given job but will also get less out of any given job than the less talented. Or to put this more carefully, the more talented will have a better chance of avoiding the worst outcome – having to work a job that doesn’t call on one’s refined talents at all – but a worse chance of obtaining the best outcome, namely getting a job that makes just the right call on one’s refined talents.

The separate principle objection

Another objection to FEO begins by accepting, at least for the sake of argument, that there is a distinct kind of good that attaches to holding a job or office. I have identified that good as the avoidance of a bad – namely the bad of being stuck in a job that doesn’t call upon one’s refined talents – while Rawls, as we saw earlier, identifies that good as ‘the realization of self which comes from a skillful and devoted exercise of social duties’, and other defenders of FEO offer their own proposals regarding the identity of this good. ²⁶ The objector grants this but then asks why opportunities for this good, which presumably is just one among many, should be distributed according to their own principle. ²⁷ Call this the ‘separate principle’ objection.

This principle is often raised in the same breath as a related one – the objection to the idea that FEO should be lexically prior to whatever principle regulates the distribution of the rest of the goods. This objection is generally raised specifically against Rawls, according to whom FEO should be lexically prior to the difference principle, which regulates the distribution of other goods. ²⁸ However, this related objection can be set aside, since I have no intention of suggesting that FEO should be prior, never mind lexically prior, to any other principle.

Back, then, to the issue at hand: Why should there be a separate principle regulating the distribution of opportunities for the particular kind of good that attaches to jobs/offices? The key to answering this objection is to remember that FEO regulates the distribution of opportunities for a certain kind of good, whereas the distribution of the good itself is regulated, we are assuming, by the principle of meritocracy. ²⁹ Furthermore, there was nothing in my defense of FEO that suggested that there is something good – something non-instrumentally good, specifically – about opportunities for the distinct good of the right job/office. So FEO is not actually a principle for regulating the distribution of a good. Therefore, of course the principle that regulates the distribution of such opportunities should be distinct from the principle that regulates the distribution of goods.

This way of answering the separate principle objection naturally raises a further question: Why should we be worried about the distribution of something that is not non-instrumentally good? To answer this challenge, we need to recall what the argument was in the first place for endorsing FEO: it is a way of instantiating the moral ideal of a level playing field. Importantly, this makes FEO a non-consequentialist moral ideal. This, again, should make clear why FEO should be a separate principle instead of getting folded in with whatever principle regulates the distribution of goods. ³⁰

This way of responding to the separate principle objection forces Arneson, the originator of that objection, to confront a dilemma. If he accepts, in principle at least, that the true political morality might contain principles grounded in non-consequentialist moral ideals, then he has to argue that there is no moral appeal to the ideal of a level playing field. To my knowledge, he has never attempted to make such an argument. On the other hand, if he rejects in principle the possibility of the true political morality containing a principle grounded in a non-consequentialist moral ideal, then it will become clear that the separate principle objection is really just a veiled assertion of consequentialism in political morality – the claim that all political ideals are grounded in evaluative considerations. ³¹

The childrearing objections

I want to turn again to the relationship between FEO and broad parental liberty. FEO has been subjected to two powerful objections based on its implications for how parents should raise their children. I call them the ‘childrearing objections’.

The first childrearing objection is Arneson’s and constitutes another worry about the scope of FEO. FEO insists that opportunities be equal among those equal in talent and willingness to put forth effort. That has the effect of treating talent and willingness to put forth effort as given and beyond the reach of the demands of political morality, or at least beyond the reach of anything FEO has to say about those demands. We have already seen why it might be thought troublesome to take talents as given; this was the origin of the luck-egalitarian objection. In addition, it might be thought troublesome to take the willingness to put forth effort as given. One determinant of an individual’s willingness to put forth effort is her ambition, which is affected by how she is raised and socialized. For instance, if she is raised and socialized to think that the only appropriate role for her in society is to be a homemaker, then she is likely wind up lacking the ambition to do anything else. This seems like an objectionable outcome, and yet one that FEO cannot condemn. And as Debra Satz (2012) has pointed out, this is just the tip of the iceberg (p. 160). There are many ways in which childrearing and socialization determine one’s willingness to put forth effort, and it seems awfully troubling for a political morality to offer nothing to those who through their childrearing and socialization wind up with very little willingness to put forth effort.

I do not have anything original to say by way of response to this objection. I shall simply endorse what Satz and Arneson have already said, which is that FEO can be supplemented with other principles that can condemn at least some of these effects of childrearing and socialization or can be expanded so as to issue this condemnation itself (Arneson, 1999a: 79; Satz, 2012: 160, 161). Satz (2012), for instance, argues that a child’s winding up with little or no willingness to put forth effort is inconsistent with the liberal commitment to principles of political participation and free choice of education and that these principles require the eradication of ‘many forms of gender and racial stereotyping and stigmatization’ (p. 162). Arneson (1999a), meanwhile, suggests expanding FEO itself so as to require that ‘the education and socialization processes that influence the formation of individual ambitions are unmarred by bigotry and unfairness’ (p. 79).

The second childrearing objection is more disconcerting. Suppose we agree that childrearing practices are a significant determinant of the level of opportunity to succeed in the meritocracy that a child will have upon entering majority. If we accept broad parental liberty, as I have argued that we should, then we believe that the state should refrain from interfering with many of these practices. Once again, then, a feature of our actual world presents a problem for FEO.

Fortunately, this does not have to spell the end of the ideal of FEO. If parents took it upon themselves to forgo the activities that would otherwise give their children advantages over other similarly endowed children, then FEO could yet be achieved. So it seems that they’re obligated to do so. The second childrearing objection is simply that this conclusion seems to be an implication of FEO.

Mason, who provides the most forceful instance of this objection, anticipates the likely response: parents have either special obligations to, or special prerogatives with respect to, their children, and these obligations/prerogatives make it permissible or perhaps even obligatory for them to do the things that have the side effect of giving their children advantages over other similarly endowed children. Mason (2004) points out, by way of rejoinder, that even if we concede this it still seems that FEO implies that parents have a strong reason to refrain from these activities. Even this is still quite counterintuitive (pp. 369–372).

I prefer to make a different response to this objection, though again one that Mason anticipates. I contend that FEO grounds neither an obligation nor a reason for parents to refrain from conferring competitive advantages on their children, since FEO is a principle of political morality and therefore is binding only on states. By way of rejoinder, Mason insists that considerations of social justice apply to the basic structure of society and that the family is part of that structure. (Here Mason cites Cohen’s work.)

At this point, it becomes difficult to conduct a dialogue with Mason, since I never set out to demonstrate that FEO was an element of justice; rather, I set out to demonstrate that FEO is an element of the true political morality. But I do not want to escape Mason’s objection on a technicality, and fortunately, I have a way of making my point that does not appeal to a distinction between justice and political morality.

The most obvious reason why FEO cannot be binding on individuals is that it is based on an ideal that applies only to states. That ideal, again, is the ideal of a level playing field. It may be – and I certainly hope it is – compelling to suppose that it is the job of the state to ensure a level playing field. I see no appeal to the idea that individuals should level the playing field or even strive towards its being leveled; they are, in this metaphor, merely the players on the field. In this way, leveling the playing field is like providing national defense, securing clean drinking water, or setting aside wild areas for recreational purposes: they’re activities that do not even register on our moral radar until we begin to think about what the state morally ought to do.^32,33

This is not to deny that activities within the family constitute part of the basic structure and are thus subject to principles of justice/political morality. However, there is an important ambiguity in the idea of an activity’s being subject to principles of justice/political morality. On the one hand, it might mean that the state may legitimately regulate those activities in the ways necessary to uphold the relevant principle of justice/political morality. On the other hand, it may mean that the individuals who engage in those activities have obligations, grounded in the relevant principles of justice/morality, to moderate their activities accordingly. I am merely denying that activities constitutive of the family are subject to principles of justice/political morality in the latter sense. So in at least one sense, the former sense, I can and do concede that activities within the family, including the activities by which parents confer competitive advantages on their children, are subject to principles of justice/political morality. (Though of course I take FEO to have merely pro tanto force and therefore the question of whether the state ought to restrict parents in this way depends, for me, on the relative weight of FEO and parental liberty. ³⁴ )

Furthermore, given my claim that FEO is part of the true political morality I can say that the state does wrong insofar as it fails to regulate parental activities in whatever way is necessary to secure FEO. This allows me to make a case, based on the intuition that one ought not to take advantage of the wrongdoing of others, that when the state ought to ban a certain parental activity but fails to do so it is wrong for parents to engage in that activity. So, for instance, I could concede to Adam Swift (2004) that it is wrong of parents to send their children to private schools. But this would not amount to committing myself to the very counterintuitive claim that parents are obligated in general to forgo the activities that would otherwise give their children advantages over other similarly endowed children.

Mason has one final bullet: an appeal to a case-specific intuition. He says, ‘If, for example, male heads of families forbid their daughters from receiving a formal education, this surely violates principles of equality of opportunity even if educational institutions in the wider society do not discriminate against women’ (Mason, 2004: 372).

One thing Mason might mean is that the outcome designated as desirable by FEO could not be instantiated under the given circumstances. This would be true but beside the point; the question is who is obligated to bring about the designated outcome. What Mason must mean, then, is that the actions of male heads of families in such a society violate FEO. I admit that I have something in the neighborhood of this intuition; I have the intuition that the male heads of families in this society are violating an obligation. But I do not have the intuition that they are violating an FEO-based obligation. And I think I am not alone here. Take the average person and confront her with a vignette like this and what she is likely to say, I suspect, is that the fathers are wronging their daughters by failing to do what parents are obligated to do – for example, not compromise the open future that their children might otherwise enjoy. FEO would not enter in to it. ³⁵

Conclusion

In this essay, I have laid out a highly specified version of the principle of FEO, argued for it and defended it against objections. My version of FEO resolves certain ambiguities that remain from Rawls’ work; in particular, it says that FEO is to regulate opportunities for jobs and offices as opposed to opportunities for broader goods such as success or achievement. Furthermore, I have claimed that since FEO regulates the distribution of opportunities for jobs and offices, as opposed to jobs and offices themselves, we can understand how there could be a place for FEO in a meritocratic world: the principle should be understood as requiring equal opportunity to succeed in the meritocracy. In addition, I have offered an original understanding of what succeeding in the meritocracy amounts to and why it’s good. Succeeding in the meritocracy means obtaining the right job or office – one that calls upon the use of one’s refined talents – and is good in virtue of constituting the avoidance of a soul-crushing bad, the bad of spending one’s working days doing something that one finds uninteresting. These moves then allowed me to offer novel defenses of FEO against the most troubling challenges to which it has been subjected.