Reply to Wiens

Some theories of justice purport to require actions of people that are not only unlikely but also contrary to widespread and deep-seated motivations. Examples include things like voting for a party whose policies are against one’s self-interest, or paying one’s employees a higher wage than would be dictated by maximising the bottom line, or exercising much less favouritism towards one’s own family and friends. Suppose that because of strong and widespread self-interested, or partial, or other motivations, people would either not make an effort or even if they did they would abandon the effort without succeeding. In that sense, we might allow, in an idiomatic phrase, that they can’t bring themselves to do these things. Now if that, in turn, entailed that they are unable to do the things, then there would be a strong case for denying that they are morally required to do so. This is because many will accept (and David Wiens and I both accept for purposes of argument) that no one is morally required to do anything he or she is unable to do. In Human Nature and the Limits (If Any) of Political Philosophy (Estlund 2011, hereafter ‘Human Nature’), I argued that being unable to bring oneself to do something (in the specified sense) does not entail being unable to do it. Therefore, theories of justice that purport to require things of people that they cannot bring themselves to do are not thus refuted. In this short note, I respond to David Wiens’ argument that this is mistaken (Wiens, 2016).

My principal aim in ‘Human Nature’ was to refute what I call the ‘human nature constraint’, which states:

A normative political theory is defective and thus false if it imposes standards or requirements that ignore human nature—that is, requirements that will not, owing to human nature and the motivational incapacities it entails, ever be satisfied. (p.208)

Refuting that would require finding one or more counterexamples, and I proposed several cases in which agents (such as messy Bill, who is a lazy sort who cannot bring himself to take his trash all the way to the curb) tend to stop trying to perform some action owing to morally flawed motivations. I went on, in light of those cases, to argue for the stronger claim that no (non-pathological) motivational incapacities count as or establish inability. ¹ It is this stronger claim that Wiens challenges. He presents a subset of motivational incapacities that he sees as counterexamples to my claim that ‘can do’ is never refuted by ‘can’t will’. He argues that cases of failed repeated good faith efforts are cases of ‘can’t will’ that do refute ‘can do’, even when the agent only fails because she stops trying. As he writes, ‘good faith motivational incapacities undermine abilities’ (Wiens, 2016: p 334). The crux of the issue, then, is about the nature of ability.

An agent who sets out to do something but will (certainly or probably) fail might seem to be unable to it. The plausibility of that thought suggests reliance on a simple success-conditional of the following form, tying the idea of ability to the ideas of success and failure:

A person is able to φ if and only if were she to try she would (tend to) succeed.

It seems fair to say that even a person who will not φ only because she will, at some point in the endeavor, stop trying, counts as failing to φ. The simple success conditional then, arguably, implies that she is not able to φ.

I should pause to point out that neither Wiens nor I mean to assert and defend the general success-conditional approach to ability, much less any specific version of it. That is not how that approach becomes relevant. In ‘Human Nature’, I merely pointed out that this is one familiar conception of ability according to which ‘can’t will’ need not, if suitably specified, be seen as entailing ‘can’t do’, as I will explain below. ² But the general conditional approach is more integral to the broader issue than that, as I now see things. The view I argue against says, in effect, that even if a person tries to do a certain action she might fail to do it owing to her own contrary motivations. Such an agent, even if she were to try to φ, would not succeed in φ-ing, precisely because her contrary motives will lead her to leave off before completion. Any plausibility I find in that idea –and I do find some – relies on the observation that it is a kind of attempt that will fail, along with the thought that this condition, somehow worked out, is constitutive of inability. Without that success-conditional idea (which may or may not be ultimately defensible) in the background, I fail to see why the view would be tempting. ³ So, the broad conditional approach to ability is assumed by the view I criticize. I argue that it can be accepted for the sake of argument. I then argue, as I will explain, that only an implausible version of that conditional approach would let ‘can’t will’ entail ‘can’t do’.

What would remain of my argument if we bar any appeal to a success-conditional approach to ability? In that case, my only support for the claim that ‘can’t will’ does not entail ‘can’t do’ would rest on intuitive cases in which a person is obligated to do something that they can’t bring themselves to do, combined with the assumption that ‘ought implies can’ (OIC). ⁴ If someone rejects that latter principle, then I am offering nothing further against them. There should be no surprise in this. The core of my argument was that the idea that ‘can’t will’ entails ‘can’t do’ is seriously challenged by cases in which a person is morally required to do something even though she cannot bring herself to do it. I focused on the challenge arising from motivational deficiencies that are themselves morally problematic. If those cases of requirement are granted, then one can save the entailment between ‘can’t will’ and ‘can’t do’ only by giving up OIC. Having clarified the polemical situation around the success-conditional understanding of ability, it should be clear that Wiens and I both adopt the broad success-conditional approach for purposes of argument. The question, then, is which of several versions of it is more defensible, to which I now turn.

The simple version (1), above, counts too many agents as unable to do things. Wiens and I agree about this. One reason is that at least if a person will give up very lightly, then the fact that she will fail does not show that she is unable to succeed. In an example of Wiens’, Danny is a person who, even if he sets out to write a book, will not end up writing a book, but only because he will abandon the project as soon as it requires hard work. Lightly abandoning a goal is not enough to establish an inability to achieve it, but only a lack of motivation to achieve it.

One way to refine and improve the success-conditional in the face of such cases is, as I observed in ‘Human Nature’, to disqualify the case of no longer trying from counting as a disabling factor. So, the following success-conditional might seem to better capture the true connection between ability and success, by, for example, denying that Danny is unable to write a book:

A person is able to (can) [φ] if and only if, were she to try and not give up, she would tend to succeed. ⁵

Danny will not succeed even if he tries to write a book because he will give up. But it might yet be that he will fail only because he will give up, and that otherwise he would tend to succeed. In that case, the revised conditional no longer counts him as unable to write a book in that case, and this is an improvement. Since there is nothing to prevent persistent efforts to write a book from succeeding, he is able to write a book.

This is not as narrowly tailored as it might be, and Wiens is right to point this out. This revised success conditional excludes not only lightly giving up (addressing Danny’s case), but all cases of giving up. Suppose that Claudia, if she tries to write a book, will not give up lightly. She makes, in Wiens’ phrase, a ‘good faith effort’. (This is not further defined, but the intended idea could be equally expressed by saying that she makes an earnest effort, or that she does not give up lightly. The exact boundaries will not matter for our purposes.) But, while she will stick with the hard work in a way that Danny would not, she will eventually stop trying in the face of weightier motivations than mere laziness. Suppose, for example, that she is already getting disappointing feedback, and/or that she has a great fear that any book she writes will be bad, or at least that it will be badly received by readers, something she very much does not want to endure. Now, crucially, it is not that this makes her creative juices dry up. Nor is it that these fears so grip her that her fingers tremble and she can’t type (more on such cases below), or anything like that. Those would not be cases where it is only her giving up that prevents her success. Rather, despite being willing to work hard, and still having ideas, and still being able to type, she abandons the project of completing a book because she dreads or fears the results.

The point is this: unlike Danny, it is not plausible to say that Claudia gives up lightly. But since it is only her giving up that accounts for her not succeeding, the revised success conditional still counts her, and not just Danny, as able to write a book so long as the only reason she fails is that she gives up trying. If that were found implausible, then, as Wiens points out, there is an intermediate success-conditional that would treat Claudia and Danny differently. He proposes,

A person is able to (can) φ if and only if, were she to repeatedly make good faith attempts to… φ, she would tend to φ successfully ⁶ (Wiens, 2016: p.340).

On this approach, Danny is able despite the fact that he will fail, but Claudia, because she will fail despite not giving up lightly, counts as unable to write a book.

Should we, as Wiens believes, opt for (3), which counts Claudia as unable to write a book? The fact that even if she sets out to write a book she will fail does not show that she is unable, as Wiens agrees, because that is also true of Danny, who he allows might well be able nevertheless. And nothing is settled philosophically by the fact that some ordinary speakers might say that Claudia, given her worries about rejection, can’t write a book. Other speakers would say that of course she can; she just needs to keep working despite her doubts. Neither of these settles the matter. Ordinary language apart, no argument would be required, I suppose, if it were obvious to every philosophically alert reader that Claudia is unable to write a book. But there are many philosophers on the other side of the question. ⁷

Wiens evidently favours (3), as I say, since otherwise he could not conclude that my arguments are mistaken for relying on (2). Wiens writes, ‘My claim is this: if (3) is the most plausible specification of Estlund’s analysis of ‘can’ (it seems it is), then good faith motivational incapacities undermine abilities’ (9). This is both a mere conditional, and also an assertion, perhaps only as an aside, of the antecedent. In any case, he does not offer any argument for (3) over (2), or any criticism of (2) so far as I can see. Still, Wiens is certainly right that these are two conflicting ways of understanding inability in terms of a success-conditional, and he can charitably be read as pointing out only that I have given no argument in favour of (2) over (3). This would be important, since I would need (2) to be superior for my broader purposes in the article. As mentioned earlier, I argued that, at least on a plausible analysis of ability along the lines of (2), many things that people ‘can’t bring themselves to do’ (or, for short, ‘can’t will’) would nevertheless be things they can do. Therefore, even if no one is required to do anything they can’t do, they might yet be required to do things they can’t bring themselves to do. But, as we can see from Wiens’ piece, if (3) is at least as plausible as (2) then many cases of ‘can’t will’ might indeed be cases of ‘can’t do’. They would thus be requirement-blocking, contrary to the main arguments of my paper. The challenge might, then, be this: why isn’t (3) at least as plausible as (2)? That is, even assuming some success-conditional account is correct, why isn’t it at least as plausible to count only Danny, but not Claudia as able to write a book, as it is to count them both as able?

In favour of (2) over (3), I will now offer two categories of troubling implications that (3) alone appears to have. While Wiens does not specify what it takes for an effort to be ‘good faith’, or when we should say that someone does or does not give up lightly, I presume that in the following cases, the giving up is not light and there is no question about whether the effort is serious or in good faith.

First, consider Tal who is a lawyer. He wants to write a book, but that would mean giving up his law career. He thinks about it long and hard and decides to set out to write a book even at the cost of his law practice. After some months of hard work and long hours as a writer, he is not deterred. But eventually Tal’s thinking changes or comes into better focus. His desire to be a lawyer eventually leads him, with great reluctance, to quit writing the book. (If repeated failure matters, suppose that later he tries and fails again (maybe even twice more) to write a book rather than being a lawyer.) (3) says that Tal, since he only gives up after a good faith effort – that is, only in the face of weighty and potent motives to the contrary – is unable to write a book. But that is not plausible. Tal has simply decided, reluctantly and in a personally difficult choice, not to write a book, and would probably do so again.

Second, consider a case where (3)’s implication of inability implausibly implies the absence of a moral requirement. Janna correctly believes that she is morally required to keep her large financial investments away from corporations in some evil enterprise, such as apartheid South Africa. She can afford the losses this would cause her but would be much better off without them. For a long time she persists with such divestment even in the face of mounting opportunity costs, forgoing tens of thousands of dollars, but eventually she gives up and invests in South African companies, contrary to her own correct moral views. Since she gives up (repeatedly, if you like) but not lightly, having made a good faith effort, according to (3) she is unable to keep her investments out of South African companies, and, being unable, is not required.

Now in the face of these examples, the ‘good faith effort’ view might be traded for a view in which it is not the earnestness of the effort, not the lightness or seriousness of the giving up that matters, but rather some privileged category of motives that are somehow special in a way that is disabling. There is some slight suggestion of this different approach in Wiens’ example of Maggie, so we ought to consider it. She ‘resolves to tell her spouse (and/or other confidants) that she opposes the dictator and wants to help organise a public protest’ (p.343). ‘What prevents her from ultimately expressing dissent is the fear that grips her at the climactic moment’ (p.344). It might be that the import of being ‘gripped’ by the fear is not only that it is a non-light motive but also that it is, in some additional but unexplained way, agency-interrupting. This would introduce, as I said, something other than earnest or good-faith effort. We should put aside cases where the motive, say fear, disrupts Maggie’s breathing or vocal coordination or upends the social setting making it literally impossible to say her piece. In those cases, she does not fail only because she stops trying, but for other reasons that continued trying would not overcome. We must also not be swayed by the fact that certain fears and their consequences might raise the stakes in a way that give the agent important reasons not to go ahead. Perhaps doing so under such fear would be grotesquely painful. But the reasons to act are not in question, only the ability. ⁸

There is a loose category of motives that we might treat together here, namely, motives contrary to some action that are such that persistence in the face of those motives pushes the agent towards collapse as a functioning agent. We need only conjure the mental states that might accompany such extreme conditions, such as mental states that produce erratic breathing or hyperventilation, sweating or chills, trembling or muscular weakness, involuntary groans or crying. Even if these physical symptoms do not physically disable the person from performing the act (which in many cases they would), the basic integrity of the agent’s processes of judgment and deliberation is in question.

We do not need sensational examples, but only the example of a very high ladder. Most people will walk up some number of steps with little hesitation. But every agent eventually approaches rungs that they are reluctant to climb. If they do climb them anyway, most people nevertheless reach rungs that they will not climb. But before that, there is the lower range in which many agents very much do not want to climb further, and will not do so, at least not without some good reason. They cannot bring themselves to do it, we often say. On my view there is, so far, no basis for saying that they are unable to do it. And the plausibility of this position is not damaged by the fact that there is, for most people, also the higher range in which ability might be compromised by the inability to even think straight. Above some point, the agent’s fear of climbing further may no longer be compatible with the ability to climb, as the basic mental processes of genuine agency might no longer be intact, the person being so grievously gripped by fear.

If Maggie is so-gripped by the dangerous prospect of expressing dissent, she may not be able to express dissent, but this would not be owed merely to the fact that she has contrary motives that move her (though not lightly) to quit trying, like Claudia. Psychological breakdown cases (as we might call them), and the cases in which the fear or other motive causes some other disabling effects such as hyperventilation or seized muscles, are cases where the success-conditional (2) is violated. Further efforts would not succeed. I cannot tell which way Wiens was understanding Maggie, but it doesn’t matter. If she is in such a breakdown scenario, account (2) plausibly grants that she is unable to express dissent, otherwise it does not. Again, Wiens assumes but does not argue that Maggie is unable, perhaps taking it to be obvious. I submit that it is no more obvious than was the case of Claudia.

The question was never whether cases of interfering motives are ever closely associated with cases of inability, as they clearly are in cases of psychological breakdown or muscular effects. It could be misleading to say, as Wiens does, that I argued that ‘motivational incapacities never block moral requirements’ (p.4; see also p.1.). The view that I mean to defend and which Wiens criticises is the view that an agent does not count as unable to do something simply in virtue of being such that were she to set out to do it, she would not be sufficiently motivated to carry through. That is not challenged by the fact that motives such as fear or excitement can lead an agent to lose balance, to go physically weak, to be unable to think straight or even to pass out. There is no doubt that such cases will often entail an inability to perform certain actions. ⁹

Turning to a final matter, I argued in ‘Human Nature’ (p.224) that if our motivational tendencies were, as such, requirement blocking (by way, for example, of entailing inabilities to do otherwise), then justice could not forbid cruelty if it so happens that we harbour motivations that tend to lead us to be cruel. I argued that this would be absurd even if such motives were part of human nature itself. ¹⁰ Call this argument of mine the cruelty challenge. Wiens points out that if the problem with our nature were that we gave in to cruel inclinations lightly, without a good faith effort to resist them, then his preferred account (3) would not trigger the absurd implications about inability and blocked requirements. But, as he acknowledges in an endnote, this maneuver would not be available ‘[i]f… people typically do not refrain from engaging in cruel… acts despite repeated sincere efforts to overcome the relevant motivational deficiencies’ (p.351). ¹¹ If humans tend to give into cruel motives, though only after earnest efforts such as Claudia’s to hold firm, then, on the view (3) defended by Wiens, justice would not proscribe cruelty. Having acknowledged that this may be a bullet, he does not say clearly whether he wishes to bite it. He grants only that ‘[i]f… we can’t stomach the thought that cruel or selfish dispositions might constrain the demands of justice’, then we would need some other, so far unsupplied argument in order to avoid that unpalatable conclusion (p.351).

Much hangs on whether there is some such adequate argument. So far as his own arguments go, Wiens’ acknowledges that his account would have this justice-shrinking implication so long as humans were to have potent cruel motives. For those of us who would indeed hope to avoid removing cruelty from the kinds of behaviour that justice forbids even if humans tend to be (though not lightly) cruel, Wiens, in an endnote, refers us to a suggestion offered for other purposes by Peter Graham. The suggestion, as Wiens writes, is that ‘“positive” duties – duties to perform some φ – are subject to the “ought implies can” principle, while “negative” duties – duties to refrain from performing some φ – are not’ (p.351). Wiens continues, it would then not apply to the ‘negative duty –a requirement to refrain from cruelty’. Leave aside the fact that Graham himself says that he merely ‘floats’ this ‘highly speculative’ idea which ‘[p]erhaps… is correct. Perhaps not’ (Graham, 2011: 378–379). In any case, it would not serve the purpose for which Wiens invokes it. The reason is that the plausible moral requirement not to be cruel is, in part, a duty not only to refrain from cruel action, but also a duty not to cruelly refrain from performing certain actions, such as easily rescuing someone from vicious ridicule under certain conditions. Not to do so – to just stand there – would often be cruel. If, as I take this to show, Graham’s proposal cannot be employed to avoid the cruelty challenge (and it was not proposed for that purpose), then there is no other route on offer by which Wiens’ preferred account of ability can avoid it. The cruelty challenge does not afflict the account that I defend and which he rejects, namely (2). There is no prospect of a person’s behaving cruelly while trying not to do so and not giving up, so no prospect on that account of people being unable to avoid behaving cruelly. Contrary to Wiens’ proposal (3), since it is within our abilities even if we have potent cruel motives, refraining from (active or passive) cruelty may be – and surely is – required by morality and justice.