Choosing reasons to choose: Chang, Murdoch, and Walden on quandaries
I wrote last summer about Ruth Chang’s concept of will-given reasons, which she contrasts with what she calls world-given reasons. In that post, I was interested in the idea because Laura Callahan uses it in her account of how we respond to permissivism about epistemic rationality, which she dubs epistemic existentialism. In this post, I want to say more about Chang’s notion, and the problems it is intended to solve; and I want to do that by discussing Kenneth Walden’s recent paper ‘The Poets of Our Lives’ in The Journal of Philosophy, which grapples with the problem that motivates Chang.
The quandary of Sartre’s student
Chang is interested primarily in cases in which the world-given reasons fail to determine which of two courses of action you should take. Take Jean-Paul Sartre’s famous example of his student, who faced the decision, during the Second World War, whether to go England to join the Free French Forces, or to stay with his mother. There were strong considerations, or reasons, in favour of each and strong considerations, or reasons, against each. Were he to fight, he might help to defeat the occupying Nazi regime and avenge his brother, who was killed as the German army advanced in 1940; but equally he would certainly be absent from home, and that would cause his mother great distress to add to the betrayal of her collaborator husband and the death of her older son; and, furthermore, he might be killed, which would be even worse for her. Were he instead to stay, he would certainly give his mother great joy; but he would leave others to fight for the country’s freedom, which he might consider dishonourable, and he would be unable personally to avenge his brother, and he might wish not only that his brother is avenged, but that it is he who does it.
For Chang, these are the world-given reasons. The term is a little confusing, since some of them are reasons only because the student values certain things, such as his mother’s happiness, vengeance for his brother, the freedom of France from the occupying forces. So it isn’t really the world alone that gives the reasons; there are facts fixed by the world that, because of his values, give him reasons in favour or against certain courses of action. But let’s stick with the term. The point is that these world-given reasons are not decisive. If they are reasons the student is able to compare with one another, then, when he weighs them in the balance, they come out equal; or we might think he simply can’t compare them—these are the sorts of cases that typically interest Chang. In either case, the world-given reasons give out before determining a verdict: the student has no more reason to pursue one course of action than he has to pursue the other; and that either because he has equal reason for both, or because the reasons are incomparable. Walden calls such situations quandaries.
What is he to do? In decision theory, the situation would be described either by saying the student is indifferent between the two options, or by saying he neither strictly prefers one to the other nor is indifferent between them, so that his preference ordering is simply incomplete and offers no comparison of them. We would then say that each option is permissible. And we would say that the student must, in Edna Ullmann-Margalit and Sidney Morgenbesser’s phrase, ‘pick’ between them—that is, plump for one of them while acknowledging the other was also permissible. But, while I will end up defending this view, I want to consider a worry about it that Kenneth Walden raises and tries to answer. He first considers Chang’s answer and rejects it, then an answer inspired by Iris Murdoch and rejects that too, and finally argues for an answer inspired by Kant’s approach to aesthetic judgment in the third Critique. I want to argue that the Kantian answer doesn’t do what Walden would like it to do, and we must make our peace with picking.
The problem of quandaries
Here is Walden’s concern (substituting in Sartre’s example):
These choices are normative in the sense that they require a normative judgment about which option should be selected. To make [his] choice correctly, [the student] cannot just “pick or plump” to go [to England and join the Free French Force]. In other words, [he] cannot choose without thinking that the option [he] is choosing is correct—that it is appropriate, fitting, or called-for. This is a feature of the structure of the choice. A selection without such a judgment is a choice ill-made by the standards of the problem [he] faces. (279)
A nitpicky point: it seems to me that, even if Sartre’s student picks between staying with his mother and going off to fight, in the Ullmann-Margalit and Morgenbesser sense, he will think the option he chooses is correct and appropriate and fitting. After all, both are permissible relative to his preferences; neither is favoured by the balance of reasons, and so neither is forbidden. So neither could be incorrect or inappropriate or not fitting. And so it seems he’ll judge each correct and appropriate and fitting. So what I think Walden really means here is that the student cannot choose without thinking the option he chooses is called-for. That is, he cannot choose without thinking that the option he chooses is not only permitted but also required. It’s this that creates the puzzle: the world-given reasons make neither required; but he must consider one required before making it; it is not sufficient that he consider the option he chooses to be permissible; but how can he do this without deluding himself about what reasons there are?
Chang on quandaries and will-given reasons
Chang’s response to this worry is to say that he can, of his own free will, create reasons that tip the balance in favour of one or the other. These are will-given reasons. According to Chang, we create such reasons, for instance, when we consciously and intentionally decide to commit romantically to someone we’ve been involved with for some time. We needn’t announce this commitment publicly, though of course often we’ll want to, at least to the person to whom we’re committing. For Chang, we need only decide ‘internally’ to do it, and we thereby change the normative landscape; we create a reason to do things that previously we had little or no reason to do. For instance, before committing to someone, I might have had a weak general reason to donate my kidney to them should they need it, but afterwards I have a strong specific reason to do so. Similarly, if Sartre’s student commits to staying with his mother, he thereby creates a reason to do that, and that reason might be strong enough to tip the scales in favour of that course of action. If it is strong enough, he can then truly judge that he must stay with his mother, thereby answering Walden’s requirement that he must think his choice is called-for or required of him.
In my earlier post, I noted that the problem with this proposal is that it isn’t at all clear that we take such internal acts of commitment to have the sort of normative force that Chang requires them to have. Suppose you decide to commit to someone you’ve been seeing for a while; they need a kidney; according to Chang, you now have reason to donate one of yours; but, as you reach for the phone to tell this person the good news so they can inform the hospital, you change your mind; you no longer wish to commit to them. On Chang’s account, you’ve acted against what you have reason to do. Your earlier commitment gave you reason to donate your kidney, and so you have reason not to change your mind about that commitment because that will likely make you not donate. Yet I submit: you do nothing wrong in changing your mind. And, if that’s right, that suggests your commitment carries no normative force.
Walden raises similar worries, but adds a further concern that arises from his claim that Sartre’s student’s choice is normative: if we are concerned that picking between staying with his mother and joining the Free French Force is not an appropriate response to a choice of this magnitude, we should surely be concerned that picking between committing to staying with his mother and committing to joining the Free French Force. But that is what Chang would have him do. The choice between commitments to different courses of action is normatively underdetermined in exactly the way that the choice between the different courses of action is.
Murdoch on really seeing
Walden contrasts Chang’s account with one he finds in Iris Murdoch. On this account, you could see all the particular reasons in favour of one action or another, and yet still fail to really see the value of the actions. What Sartre’s student needs to do is spend more time considering the two options with a view to really seeing them as they are. By doing this, certain conventional associations and distortions will evaporate and he’ll be able to see the choices for what they really are, and this might well reveal to him that one is better than the other. Here is Richard Wollheim saying something like this about appreciating a painting:
For I came to recognise that it often took the first hour or so in front of a painting for stray associations or motivated misperceptions to settle down, and it was only then, with the same amount of time or more to spend looking at it, that the picture could be relied upon to disclose itself as it was. (Painting as an Art)
But of course if sitting longer with the decision will lead Sartre’s student to see the value of his actions as they really are—just as sitting before a painting for two hours reveals it to Wollheim as it really is—and if seeing them as they really are shows him that one or other is better, then this is not a quandary. His reasons in fact do tell between them; it just requires a bit of extended reflection to reveal this.
So this won’t work. In genuine quandaries, Murdoch’s proposal won’t help. There is nothing really there to be revealed.
Walden on Kant on aesthetic judgment
Walden’s alternative is this: The student does not create a reason where there was none before merely by a single act of picking a commitment to a particular course of action, as Chang suggests. But he might create such a reason by exercising his creative faculties on the collection of reasons in favour of the two possible courses of action, and thereby bringing those reasons together in a particular way that breaks the tie between them. By doing this, Walden contends, the student does not simply add another reason in favour of one of the courses of action; rather, as he puts it, he “creates a new conception of the normative import of [his] reasons that [he] can regard, despite its artifice, as a fitting response to those reasons” (295).
As I mentioned above, Walden wants to appeal to Kant’s account of how we form aesthetic judgments to spell out how our creative faculties achieve this alchemy—in particular, 5:314-7 in The Critique of the Power of Judgment. The crucial feature of this account is what Kant calls the ‘free lawfulness’ of such judgments: they are not compelled by the features of the object we judge; but they are not completely unconstrained by those features either—they are based on those features, but no deterministic law connects those features to the judgment. I judge this poem beautiful because of features of it—the features of it justify my judgment—but those features don’t determine that I must find it beautiful—the features do not compel my judgment.
Here is Walden describing the process:
Say that I am looking to buy one of two paintings. To make my decision, I examine the paintings and find many discrete things that each does well and poorly. The muse’s robe in Painting A is magnificently diaphanous. Painting B has spectacular shadow work. The nymph’s smile in Painting A is devilishly wry. After noting all these merits (and some flaws), I take my reasons for favoring each painting to be on a par. And so I am in a quandary. Seeing my frustration, a friend suggests I consider which painting is more beautiful, a question I had somehow hitherto forgotten. So I stand back and take in both paintings. In doing this I note all the same merits and demerits I have already enumerated, but I try to orient them within a larger assessment of each painting’s overall beauty—an exercise that naturally requires significant imagination. Eventually I decide that A is the more beautiful and buy it. (293)
Now, Walden says that I do not thereby simply come to discover some extra reason I hadn’t included beforehand—the reason given by the greater beauty of painting A, or by my judgment that it is more beautiful—that now tips the balance in favour of A:
the beauty of A encompasses those properties—it is grounded in them—and the normative support beauty lends to the choice of A likewise encompasses the reasons provided by them. (293)
Similarly, Walden thinks, Sartre’s student might make a judgment about the two courses of action available to him by exercising his creativity and imagination to draw together the reasons that previously existed but which failed to tell between them.
[He] should approach the possible lives that would follow from each of [his] options as [he] might approach a work of art—as something whose constituent parts can be arranged into a normatively significant shape by our creative powers. [He] should not do this as a way of producing or discovering a new balance-tipping reason. [He] should do it as a way of creatively reengaging with the reasons that led [him] into a quandary in the first place. [He] should do it in the hope that this reengagement, this inventive vision, might produce a way of appreciating those reasons’ collective force. (294)
Free choice and permitted choice
Return to the example of the paintings. I agree that the beauty of A is not a reason that existed beforehand; it is not a pre-existing reason I simply discovered by attending to the various components of the painting—the wryness of the smile; the precision with which the robe’s appearance is rendered. But it is simply another reason. It’s just that I added it to the pile by coming to judge A more beautiful than B. I created this reason by coming to make the aesthetic judgment. It’s admittedly a judgment about a global property of the painting, whereas the examples of the other reasons—the smile, the robe, perhaps also the brushwork and the composition—were local properties; but global properties give reason just as much as local ones—as Jenann Ismael notes, we often value lives that collect into a whole the same distribution of happiness but in a different order and so with a different shape, and this is a global property of the life.
But that’s a small quibble and I don’t think it affects Walden’s proposal, for the newly-formed aesthetic judgment about the paintings’ comparative beauty provides exactly the reason that I need to judge that I must choose painting A—and so if the choice between the paintings is normative in Walden’s sense, and so requires that I judge whichever I choose to be the one I must choose, this reason will allow me to make that judgment truthfully. And, in the case of Sartre’s student, should he come, through such a creative act, to judge the action of staying with his mother to have some normative property he didn’t previously judge it to have—it is the loyal choice or the virtuous or caring one, perhaps—he will have created the sort of reason that now allows him to judge that he must stay with his mother, and thus can choose to do this in a way that respects the normativity of the choice.
But I have a different worry about Walden’s proposal. When we look in greater detail, it seems to me that it is not a genuinely distinct alternative to Chang’s or Murdoch’s. Think again of the case of the paintings. What exactly happens when I take the friend’s advice and ask whether the painting is beautiful? I return to look at the paintings. I stand before both, and I wait. One of three things might now ensue.
First: I have dispositions within me that deterministically take as stimuli the particular features of the paintings—the smile, the robe, the brushwork, the composition—and give out as manifestation my judgment that A is more beautiful than B.
Second: I have non-deterministic dispositions within me. Again, these take the particular features as stimuli; but they don’t inevitably lead to the particular judgment that A is more beautiful than B. On another occasion, they might have led me to judge in the opposite order. For each aesthetic judgment in which they might manifest, there is some probability they will. This time they lead me to judge A more beautiful than B.
Third: I have non-deterministic dispositions within me, and I fill the gap between the stimulus and whichever aesthetic judgment I will make by making a free choice—I freely choose to judge A more beautiful than B.
On the first, surely my reason for choosing painting A—namely, that I judge it more beautiful—was in fact present prior to my contemplating the painting with a view to determining its beauty. By gazing upon it, and letting that deterministic disposition manifest, I discover the reason that was there all along. So this case is akin to Murdoch’s, and it won’t help for the same reason Murdoch’s doesn’t help. In the case of a genuine quandary, there isn’t decisive reason to do one thing or the other, even once all those reasons are revealed to the chooser as they really are. Similarly, if Sartre’s student didn’t judge the act of staying with his mother to be the virtuous one beforehand, but he was deterministically disposed to so judge it upon reflection, then the reason that judgment provides is there already before he judges and the reflection simply reveals it to him. So it isn’t a quandary.
On the second, this isn’t quite true. The aesthetic judgment about the paintings’ comparative beauty isn’t fixed before I sit down to contemplate them, and Sartre’s student’s judgment that staying with his mother is the virtuous act isn’t determined before he reflects upon it. But in this case, it is simply chance that makes me judge one way rather than another about the paintings’ comparative beauty, and Sartre’s student judge one way or the other about the virtuous act. And surely choosing on the basis of a judgment made by chance is no better than picking, which Walden thinks does no justice to the nature of the choice. Indeed, Walden considers a case in which someone flips a coin to make such a choice and says explicitly that it is not appropriate for a normative choice.
On the third, we choose freely to judge A more beautiful, and Sartre’s student freely chooses to judge staying with his mother the virtuous course of action. But of course, because of Kant’s observation that these choices are not normatively determined—for instance, multiple different aesthetic judgments would also have been justified on the basis of the features of the painting I take in as I contemplate them—this is simply a case of picking. And that brings us back to the concern about Chang’s proposal. Picking a commitment to prefer painting A to painting B can’t be an appropriate response to the problem of choosing between A and B, if simply picking between A and B is not appropriate. Similarly, picking an aesthetic judgment concerning the relative merits of A and B can’t be an appropriate response to the problem if simply picking between A and B is not appropriate. And similarly, picking the judgment that staying with his mother is the virtuous act out of all the judgments his reasons permit him to make can’t be an appropriate response to the situation of Sartre’s student if simply picking between his mother and the Free French Forces is not appropriate.
Choosing to value
I should clear something up before I wrap up. I do think that, whether I come to my aesthetic judgment about the paintings in the first, second, or third way described above—that is, by a deterministic disposition, a non-deterministic one, or a free choice—once I have made that judgment, I have constructed for myself some new values. After the judgment is made, I now genuinely value painting A more than I value painting B. And that value does genuinely give me reason to choose A over B. After all, many things to which I assign a value I do so because of a process like this. That process is further in the past, but it’s often no different. Those original reasons for and against painting A and for and against painting B—the reasons that failed to tell definitively in favour of one painting over the other—might very well be reasons for me precisely because in the past I’ve constructed my values in one of the three ways described—I’ve been disposed to construct them that way, or I’ve done so by chance, or I’ve freely chosen to so construct them. And so if those values constructed in the past are sufficient to underpin the original set of reasons that gave rise to the quandary, then the newly-constructed value is also sufficient to underpin a new reason. So my complaint about Walden’s approach is not that the process of forming an aesthetic judgment that he describes is ineffective. It’s that it’s not sufficiently different from picking or tossing a coin or simply discovering pre-existing values. And so if the choices are genuinely normative, it won’t help.
So I think in the end we must make our peace with the fact that, in genuine quandaries, we must pick or let a random process determine our choice. That is, if we feel that choices with such high stakes should be made in ways that allow us to say, once we’ve made them, that we had to make them this way—that is, if we take such choices to be normative in Walden’s sense—I fear we’re wrong.
Easy for me to say, of course, because I don’t have that intuition in the first place. I suspect this is partly because, when I think of quandaries, I think first of cases in which the indecision arises because we don’t know enough and not because we haven’t constructed a sufficient rich system of values that allows us to compare all options. For instance, suppose I have a very well specified set of political values, but I have very sparse evidence about how to achieve the political goals I’ve set, and so the choice between entering politics and joining an activist group is a quandary for me—my evidence just doesn’t settle which would be the best means to my precisely constructed ends. In this case, despite the high stakes and momentous nature of the choice, unless it’s possible for me to gather evidence that I’d expect to settle the matter, picking between them seems the appropriate reaction. Constructing further values won’t help, since I’ve got a very rich set of values, so there likely aren’t any gaps to fill. So I must simply plump for one or the other, and pursue it in the knowledge that there was equal reason to pick the alternative. And, as in the epistemic case, so in the case in which the quandary arises from gaps in our values.
I suspect this is just a rather disconcerting feature of our condition. We would very much like to meet decisions of great moment with a way of choosing that does justice to their magnitude. It seems inappropriate to have to use the same strategy, namely picking, that we use when faced with two different cereal boxes in a supermarket and no reason to favour one or the other. Something similar happens in the case of transformative experience—or perhaps it’s just an instance of a quandary. Laurie Paul contends that we would like to make this choice by imaginatively projecting ourselves into the situations to which each option might lead us—the situation of being a parent, for instance. But, in cases of transformative experience, that isn’t possible. There, as here, I think we must make peace with the fact that we simply cannot make the decisions in the ways we’d like to. That is just not open to us.