I. Introduction
Education is “teleological”: it has aims or goals. It is also value-directed or “normative”: its aims are deemed worthy of pursuit. It is commonplace to distinguish between relatively specific and relatively general educational aims. As examples of the former, children should have reading, writing, and effective oratory skills, as well as knowledge of cultural history, literature, elementary natural science, mathematics, art, and music. As examples of the latter, it has been proposed that children should develop into critical thinkers and autonomous agents. Education’s relatively general aims have frequently been thought of as its ultimate aims. This is apparently because although relatively specific educational aims may vary considerably depending upon contingencies such as economic or cultural ones, the ultimate aims seem more basic, stable, and universal. Footnote 1 These aims play a pivotal role in regulating and structuring moral and other types of normative education.
Which candidates reasonably qualify as education’s ultimate aims, and how is their putative privileged status of being ultimate to be justified? Addressing these challenging questions, Harvey Siegel proposes that “critical thinking is, at a minimum, ‘first among equals’ in the pantheon of educational ideals. . . . [and] is rightfully seen . . . more dramatically, as the ultimate educational ideal.” Footnote 2 He advances the following argument for this interesting proposal:
Consider a case in which . . . [some ideal, such as critical thinking] conflicts with some legitimate other. In such a case, one might argue that the other should override critical thinking in this instance. And perhaps so it should. But it requires rational argument, and appeal to reason, in order to make the case for the preferability of the rival ideal to that of critical thinking. And such appeal is, of course, an appeal to, and an honoring of, the latter ideal itself. Consequently, an overriding of critical thinking by a rival educational ideal at one level requires acknowledgment of the reign of critical thinking at the next highest level. In this way critical thinking must preside over and authorize the force of its rivals. Footnote 3
Contrast Seigel’s favored candidate of education’s ultimate aim and its proposed defense for being ultimate with John White’s approach. Footnote 4 According to White:
Personal autonomy is a central liberal value. It rests on an even more fundamental value in human life—personal well-being. Autonomous well-being is only one variant of the more general concept, given that people can flourish or not flourish in non-liberal—for example, traditional-tribal—as well as liberal societies. Footnote 5
White’s strategy traces its ancestry to Aristotle. In Book I of the Nicomachean Ethics, Aristotle advances what seem to be “bedrock” or foundational values or aims that supposedly carry their justification on their sleeves. The bedrock aims that Aristotle identifies concern what is good for both individuals and society at large, and are closely tied to what it is to flourish as private persons and as public citizens. In the early sections of the Nicomachean Ethics, Aristotle says that the good life is the life of flourishing and virtue. To achieve a state of well-being, proper social institutions are necessary. The political setting must enable people to cultivate the peculiarly human excellences, the moral and the intellectual virtues, that are necessary for the good life. Indeed, Aristotle proposes that the state should actively encourage people to inculcate the virtues in order for its citizens to flourish. Their flourishing, in turn, ensures that the political order itself flourishes as well. Footnote 6
White’s Aristotelian approach assumes that education’s rock bottom aim is to raise our children in such a fashion that, of the many different ways in which their lives could turn out, each child gets a life that is good in itself — that is intrinsically valuable — for him or her. On this approach, identifying and justifying what we may call education’s “specific, substantive, and ultimate aims,” such as ensuring that our children develop into autonomous agents, or virtuous agents, or critical thinkers, requires exposing how these aims are essentially associated with personal well-being.
I want to contribute to the literature on ultimate aims by casting doubt on whether we can hope to achieve what may reasonably be taken to be various specific, substantive, and ultimate aims. An outline of the primary reasoning for this skeptical position distills to this: Achieving these aims requires that we be free in a sense of “free” to be explained. Factors beyond our control ensure that we are not free in this sense. Hence, achieving these aims is largely beyond our reach.
Taking my cue initially from White’s insightful ruminations on the Aristotelian approach, I assume that among the specific, substantive, and ultimate aims of education are: (a) Uprightness. Children should develop into morally upright agents, agents disposed to fulfill their moral obligations and to avoid doing what it is morally impermissible for them to do. (b) Virtuosity. Children should develop into morally virtuous agents, agents disposed to perform actions that manifest virtue. (c) Appraisability. Children should develop into morally responsible agents, agents disposed, minimally, to perform actions for which they are not morally blameworthy, and, desirably, to perform actions for which they are morally praiseworthy.
To stave off a preliminary worry, the pronouncements of uprightness — what it is obligatory for agents to do — need not conflict with those of virtuosity. Consider, for instance, the following strategy to develop a sort of Aristotelian ethics of virtue that confirms compatibility. In this view, the primary normative status of an act (whether it is morally permissible, impermissible, or obligatory) is a function of the net amount of virtue its agent manifests in performing it. Assume that we have a list of all the virtues and the vices. Assume, moreover, that for any possible action, there is a number that represents its “virtue value”: take the total amount of virtue its agent would manifest if this agent performed it and subtract from it the total amount of vice its agent would manifest if this agent performed it. This sort of virtue ethical theory would then say: an act is permissible if and only if no alternative has a higher virtue value than it has.
In what follows, I first show that obligation requires avoidability — no act is morally obligatory for one unless one can both perform, and refrain from, performing it (Section II). Next, I introduce and defend a principle (Motivation/Ability) that links motivation to do something with ability to do it. This principle, together with the thesis that obligation requires avoidability, sustains the conclusion that we have a moral obligation to perform actions only in a narrow range of cases (Section III). Finally, I exploit this conclusion to tease out pertinent implications regarding education’s ultimate aims (Section IV).
II. Perfect Virtue and Diminishing Obligation
I begin with an extreme case of “perfect virtue.” Features of this case will be exploited to draw lessons for more mundane cases. The objective is to show that as one becomes more virtuous insofar as one is committed to doing what is seemingly obligatory for one, what one may take to be obligatory for one will not in fact be obligatory for one. Roughly put, one’s strength of character diminishes the range of things that are obligatory for one.
Attaining this objective calls for a slight digression into the freedom requirements of moral obligation. Elsewhere I have defended the view that obligation, permissibility, and impermissibility all require freedom to do otherwise. Footnote 7 It will suffice for the objectives of this essay to summarize the crux of this complex defense. The defense invokes the primary principle, Kant’s Law: If one has an overall (as opposed to merely a prima facie) moral obligation to do A, then one can do A; and if one has an (overall) obligation to refrain from doing A, then one can refrain from doing A.
Kant’s Law captures the commanding thought that moral obligation requires freedom or control. Unfortunately, few have reflected on control requirements of moral permissibility and impermissibility. It is just as compelling that permissibility and impermissibility require freedom or control as it is that obligation requires control. After all, what would sustain the view that it can be impermissible (or permissible), as of some time, for one to do something at that time, for example, if one cannot, that is, if one is not free to, at that time, do it then? Wouldn’t deontic judgments regarding moral impermissibility and permissibility be empty or trivial if it could be the case that it can be impermissible (or permissible) for one to do something that one cannot do? One should take it, then, that in the absence of good reasons to believe otherwise, impermissibility and permissibility require control. Precisely what control do they require? I have proposed that, minimally, just as the moral “ought” entails the “can” of ability and opportunity, again, unless there are compelling reasons to think differently, so do moral impermissibility and permissibility. Footnote 8
Now argue as follows. If it is impermissible for one to do something, one ought not to do it (from principle Equivalence: “ought not” is equivalent to “impermissible”). If one ought not to do something, one can refrain from doing it. Hence, if it is impermissible for one to do something, one can refrain from doing it. But it is also true that if it is impermissible for one to do something, one can do it. So, there is an alternative possibilities requirement for “impermissibility.”
Regarding obligation, if it is obligatory for one to refrain from doing something, it is impermissible for one to do it. Furthermore, if it is impermissible for one to do something, one can do it (from “impermissible” implies “can”). Therefore, if it is obligatory for one to refrain from doing something, one can do it. But it is also true that if it is obligatory for one to refrain from doing something, one can refrain from doing it. Hence, “ought” entails “can” and “can refrain from.”
If impermissibility and obligation require alternative possibilities, there is little reason to deny that permissibility, too, requires alternatives. We may conclude that there is an alternative possibilities requirement for the truth of judgments of moral impermissibility, obligation, and permissibility. Footnote 9
An objection to this sequence of reasoning and a reply to it will help to set the stage for brief discussion on whether the “can” in Kant’s Law denotes a “specific” or a “general” ability. The objection, capsulized, is that just as Harry Frankfurt’s examples impugn the “principle of alternate possibilities” (persons are morally responsible, blameworthy, for example, for doing something only if they could have done otherwise), so these examples impugn the principle that it is obligatory (impermissible or permissible) for a person to do something only if the person could have done otherwise. In a template of such an example concerning responsibility, in its initial stage, Augustine is morally blameworthy for pushing a child off the cliff to her death. Next, this stage is modified so that something precludes Augustine from doing anything incompatible with pushing the child but without in any way interfering in Augustine’s actually pushing, as it turns out. A mind reader, Ernie, who can tell what Augustine is about to do, will do nothing if he detects some reliable sign Augustine displays that he, Augustine, is about to push the child, but will force Augustine to push the child if he discerns the reliable sign that Augustine is about to refrain from pushing. Ernie never intercedes because Augustine proceeds exactly as before. Since without Ernie Augustine is morally blameworthy for pushing, and since in the modified stage Augustine behaves no differently, he is blameworthy for pushing here too, even though he could not have done otherwise. Footnote 10 Hence, the principle of alternate possibilities is false.
We may assume that in stage 1, although Augustine pushes the child, he could have refrained from doing so. We may also reasonably assume that in this stage it is impermissible for him to push the child. At least intuitively, in stage 2, it is also impermissible for Augustine to push the child; he ought not to have pushed the child even though he could not have refrained from pushing. So, it is contended, it is false that “impermissibility” implies “can refrain from.” A slight modification in the example seems to show that “ought” does not entail “can refrain from” either. Suppose Augustine ought in Stage 1 to comfort the child. In Stage 2, he fulfills what seems to be this obligation, but he could not have done differently because of ever vigilant Ernie, who, this time around, wants Augustine to comfort the child.
However, it is not clear that this way of rejecting the principle that impermissibility (or obligation) requires avoidability is cogent. Notice that Kant’s Law, a power or control principle, is pertinently like the following highly credible control principle.
Blame /Control: One is blameworthy (or, more generally morally responsible) for performing an action only if one can perform it.
This principle simply affirms the connection between control and blameworthiness (or moral responsibility). It expresses the widely held and plausible view that one is morally responsible for an action only if one has control regarding it. Notably, Frankfurt examples do not undermine this principle. The principle of alternate possibilities - regarding blameworthiness:
PAP -Blame: One is morally blameworthy for having done something only if one could have done otherwise,
is a conjunction of Blame/Control and
Control: One can perform an action only if one can refrain from performing it. Footnote 11
It is Control that provides the alleged link between blameworthiness and alternative possibilities. However, Frankfurt examples undermine Control.
If Frankfurt examples leave unscathed the principle that blameworthiness requires control (Blame/Control), they should leave unscathed the principle that obligation requires control (Kant’s Law) or that impermissibility requires control:
Impermissibility /Control: It is impermissible for one to do something only if one can do it.
The link between obligation and alternative possibilities is provided not by Control but by Impermissibility/Control and Equivalence. Recall the apt argument: if one ought not to do something, it is impermissible for one to do it; if it is impermissible for one to do something, one can do it; therefore, if one ought not to do something, one can do it. Together with the principle that if one ought not to do something, one can refrain from doing it, we derive the principle that obligation requires alternatives. Just as Frankfurt examples do not impugn the principle that impermissibility (or obligation) requires control, so they do not impugn Equivalence.
There is, then, this significant difference between PAP-Blame and the principle that obligation requires alternatives: Regarding the former, essential to the link between blameworthiness and alternative possibilities is principle Control, but Control is false as Frankfurt examples confirm. Regarding the latter, essential to the link between obligation and alternative possibilities are Impermissibility/Control and Equivalence (and not Control). Frankfurt examples threaten neither of these principles. Or, minimally, if one grants that Frankfurt examples leave untouched the principle that blameworthiness requires control, one should also grant that they leave untouched the principle that impermissibility (or obligation) requires control. Moreover, it is implausible to suppose that Frankfurt examples undermine Equivalence.
Next, for another objection, reconsider the variation of the Frankfurt example in which Augustine fails to do what he seemingly ought to do: refrain from pushing the child. The objection has it that we can consistently preserve the view that even in Stage 2 of this example, Augustine has a moral obligation not to push the child, and insist that obligation requires avoidability. Rather than working with a highly restricted reading of “can” in Kant’s Law, what some call the “can” of specific ability, the sense in which an agent can do what she ought to do is meant to be more general. Footnote 12 General abilities like this need not be limited to what one is able to do given one’s motivations, or circumstances, just as they are. On this proposal, we preserve an “ought” implies “can” and “cannot” principle, but we are just more permissive about the reading of the ability claim in Kant’s Law.
A proper assessment of this objection awaits candidate accounts of the general ability (or, more expansively, the broader ability) at issue. However, sidestepping details of any such account, there does seem to be a hard-to-pin-down although intuitive distinction between specific and general abilities, and we can work with this intuitive understanding to respond to the objection. Positioning the objection in context, intuitively, although Augustine does not have the specific ability to refrain from pushing the child in his Frankfurt predicament, he retains the general ability to refrain from pushing. Ernie, together with his mind-reading machinery, does not in any way affect this general ability because he never shows his hand. If the ability in question that “can” expresses in Kant’s Law is this sort of general ability, one might propose that Augustine (in Stage 2) of the pertinent Frankfurt example has both the (general) ability to push and to refrain from pushing the child (although in his actual circumstances he lacks the specific ability to refrain from pushing the child). Consequently, we can hold on to the view that if one ought to do (or avoid doing) something (if, for instance, Augustine ought to refrain from pushing the child), one can do it and one can avoid doing it (when it is understood that “can” expresses general ability) even if it turns out that one lacks the specific ability to do or to refrain from doing that thing. Footnote 13
This way of understanding the “can” in Kant’s Law has high costs. Here, I expose only one of them. Footnote 14 Let “S G-can do A” express the proposition that S can do A when “can” expresses general ability. Then, to introduce further terminology, S G-can do A does not entail S can bring about A. (Augustine G-can hold on to the child even though in his circumstances he cannot bring it about that he holds on to the child; remember, with Ernie on the scene it is unalterable for him that he push the child off the cliff.) If we let “S S-can do A” stand for the proposition that S can do A in the specific ability and opportunity sense of “can” that entails that S can bring about A, then S G-can do A does not entail S S-can do A. Roughly, if one has the general ability to do A, it does not follow that in one’s situation, one can bring about A. Construing “can” as “G-can,” Kant’s Law now says:
G -Kant’s Law: If S morally ought to do A, then S G-can do A (i.e., S has the general ability to do A); and if S morally ought not to do A, then S G-can refrain from doing A.
To facilitate discussion, let “G-Normative Theory” stand for some normative theory concerning obligation which, in conjunction with relevant facts, implies that Augustine ought to hold on to the child in his Frankfurt situation even though he cannot bring about his holding on to the child. We may suppose that advocates of G-Normative Theory accept G-Kant’s Law because this theory entails G-Kant’s Law.
Now ponder this case. Ted has come down with a disease. Doc’s best course of treatment is to give two consecutive doses of medicine A. The next best course of treatment is to give two consecutive doses of medicine B. Anything else leads to death. Footnote 15 Suppose treatment is to begin on Monday. Intuitively, Doc ought to give A on Monday and then again on Tuesday. Augment the case with a counterfactual intervener who is hell-bent on Doc’s giving two consecutive doses of B. Even if he wanted to, Doc couldn’t give A on Monday followed by giving A on Tuesday. Assume that, being morally imperfect, Doc freely gives B on Monday (in the sense of “free” — whatever it precisely is — in which it is true that Augustine freely pushes the child off the cliff). If G-Normative Theory is true, as of Sunday Doc ought to give two consecutive doses of A (even though he S-can’t give A on Monday or on Tuesday). This is partly because on Sunday Doc has the “weak” alternative of giving A on Monday in the sense that he has the general but not the specific ability to give A on Monday (he can’t bring about his giving A on Monday); and the “strong” alternative of giving B on Monday in that he has the specific ability to give B on Monday; he can bring about his giving B on Monday. Briefly, if G-Normative Theory is true, as of Sunday, it is obligatory for Doc to give A on Monday because Doc has both the weak alternative to give A on Monday and the strong alternative to give B on Monday, and the former alternative is overall better than the latter insofar as if it were performed — if Doc could bring about his giving A on Monday — the world would be overall better. But if Doc ought to give A on Monday, then it is impermissible for him to do anything else. So on Monday, it is impermissible for him to give B on Monday even though he S-can give B on Monday. This result is hard to swallow for, given that Doc cannot bring about his giving A on Monday (or on Tuesday for that matter), as of Monday in all the best worlds accessible to him — worlds he can make actual — he gives two consecutive doses of B. In short, Doc can only cure the patient by first giving B on Monday; so how can any plausible moral theory yield the result that as of Monday, it is morally forbidden for Doc to give B on Monday when he S-cannot give any other medicine on Monday? To retort merely that Doc has the general ability to give A on Monday is pretty lame.
Another example helps to highlight a similar, highly counterintuitive result of G-Kant’s Law. In Scene-1, a child is drowning. You S-can easily rescue her; in your situation, you can jump into the water and conduct her to safety. Assume that you have an obligation to save her. Scene-2 is just like the former except that, unbeknown to you, there is a diver in the water between you and the child who will prevent you from saving the child; he will kill you if you even attempt to enter the water. Proponents of Kant’s Law will say that, as of the time you have not entered the water and the diver is keeping watch, you have no obligation to save the child (although you may think you have such an obligation). In contrast, presumably, proponents of G-Kant’s Law will claim that as of this time, you do have an obligation to save the child: although, as of this time, you S-cannot intentionally save the child, you G-can intentionally save the child. You have the general ability to save the child because the diver has not intervened and your germane general abilities remain unscathed. Again, this appeal to general abilities seems irrelevant. For all rhyme and reason, with the diver poised to kill you, you are just as relevantly “incapacitated” as you would have been if you were tethered to your seat and, thus, were unable to get to the child.
It appears, then, that understanding “can” to express general ability in Kant’s Law will result in unpalatable consequences.
Let’s now revert to our discussion on character and obligation. With his typical wit and moral insight, in a passage (laced with irony) Mark Twain pronounces, “I am morally superior to George Washington. He couldn’t tell a lie. I can and don’t.” Footnote 16 When Luther broke with the Church of Rome in the sixteenth century, he made the now famous declaration: “Here I stand. I can do no other.” A mentally healthy, loving mother might claim: “I couldn’t but rescue my child from the monstrous waves.” Whether someone can in fact be constitutionally or psychologically incapable of performing various actions is controversial. Far less contentious is that the sort of person one is, or broadly, one’s character, influences the types of action one can or cannot perform. The extent of this influence is debatable. Suppose it was literally true of Washington that he could not bring himself to lie, given his disposition to be honest. Then since obligation requires avoidability, and he could not have prevented himself from telling the truth, it is false that, on the relevant occasion, it was obligatory for him to tell the truth on that occasion. Imagine a George Washington endowed with sundry virtues — generosity, loyalty, temperance, and so forth — and that the influence of these virtues on his conduct precluded him from doing various things at various times. For example, he could not lie nor betray a friend (unless forced to choose between exemplifying one or another of these virtues, for instance, lying to remain loyal). Then the range of obligations for Washington narrows considerably. As one becomes more virtuous insofar as one cannot refrain from doing what it is seemingly obligatory for one to do, this thing that is ostensibly obligatory for one will not in fact be obligatory for one. Furthermore, what is relevantly true of our hypothetical paragon of virtue is true of many others as well. Our devoted mother finds it unthinkable not to rescue her child from the violent waters. If she could not have refrained from saving her child, then, again, it is false that on the relevant occasion it was obligatory for her to save her child then.
I have assumed that Washington could not but have told the truth and the mother could not but have saved her child. What would support this assumption, and why believe that, generally, conscientious people are incapable of doing what they take to be impermissible?
III. A Constraint on Being Able to Do Otherwise
The burden of this section is to support the second premise of the following argument for the thesis that our range of obligations is considerably narrower than we may have hitherto presumed: (1) For any action, A, one has a moral obligation to do A only if one can refrain from doing A. (2) For any A, one can refrain from doing A only in a narrow range of cases. Hence, (3) for any A, one has a moral obligation to do A only in a narrow range of cases.
To motivate the second premise, I introduce a principle that links motivation to perform an action with ability to perform that action. Action requires some sort of pro-attitude, that is, something that has motivational force. It is a plausible conceptual claim that it is impossible for a person to perform an action without having some pro-attitude — or desire, for short — to perform that action. For example, beliefs standardly construed do not have motivational force. Here is a first stab at the principle:
Principle Motivation/Ability-1: If S believes that it is morally impermissible as of t1 for S to perform A at t2, and S lacks any desire to perform A at t2, and in the relevant temporal interval (between t1 and t2), S cannot acquire a desire to do A at t2, then as of t1 S cannot do A at t2. Footnote 17
The principle’s basic idea is straightforward: If you believe that it is morally impermissible for you to do something, and you do not have any desire to do it (you have no motivation to do it), and, moreover, you cannot in the relevant time interval acquire any motivation to do it — you cannot, that is, acquire any pro-attitude to do it — then you cannot do it. This principle should not be conflated with the following (principle Motivation/Ability-2) that is false: If you believe that it is morally impermissible for you to do something, and you do not have a desire to do it, then you cannot do it. As John Fischer and Mark Ravizza explain, Motivation/Ability-2 has counterexamples. Here is one:
Just about anybody can summon up the worry that he is not free to do otherwise . . . This worry can then generate some reason (perhaps, a desire) to do otherwise simply to prove that one can do so. Thus, barring special circumstances . . . even an agent who actually does not have any desire to do other than A can have the power to generate such a desire (during the relevant temporal interval). And insofar as: (i) the agent can generate some desire to do other than A, (ii) the agent can try to act on this desire, and (iii) if he were to try to act on this desire, he would succeed, then we believe that the agent can (during the relevant temporal interval) do other than A. The leading idea here is that there is no reason to suppose that agents generally lack the power to generate (in some way or another) reasons to do otherwise, the power to try to act on those reasons, or the power to succeed in so acting. Footnote 18
Now for some refinements, first, it seems that an agent who satisfies the conditions in the right-hand side of Motivation/Ability-1 may be able to A accidentally. If so, Motivation/Ability should be taken to be a principle about A-ing intentionally. Footnote 19 This is all for the good, especially since it is moral obligation that is of concern. To amplify briefly, one is indirectly obligated regarding something just in case one is obligated regarding it by way of being obligated regarding something else; one is directly obligated regarding something just in case one is obligated regarding it but not indirectly so. As Michael Zimmerman cautions, direct obligations are restricted to intentional actions that one can perform; not so with indirect obligations. Footnote 20 Suppose Ted has a direct obligation to turn on the furnace. He cannot, however, turn it on without activating the relevant mechanisms of which he has no knowledge; so he (indirectly) ought to activate these mechanisms. This indirect obligation is one that, in his circumstances, he can perform, but not intentionally. We may take the relevant parts of the argument in the previous section to support the view that direct obligations are restricted to actions that one can both intentionally perform, and refrain from performing. As I will argue, if one frequently cannot intentionally refrain from performing various actions, then it is false that it is obligatory for one to perform these actions.
Second, Motivation/Ability should also include the clause that if S believes that it is morally impermissible at t1 for S to perform A at t2, in the relevant temporal interval between t1 and t2, S cannot cease believing that it is morally impermissible for S to perform A at t2. The resulting principle is:
Motivation /Ability-3: If S believes that it is morally impermissible as of t1 for S to perform A at t2, and S lacks any desire to perform A at t2, and in the relevant temporal interval (between t1 and t2), (i) S cannot acquire a desire to do A at t2, and (ii) S cannot cease believing that it is morally impermissible for S to perform A at t2, then as of t1 S cannot intentionally do A at t2.
Third, the principle as it stands is insensitive to “side-effect” actions. Footnote 21 Suppose you know that by B-ing you will A; furthermore, you have no desire to A, you can’t acquire a desire to A over the span at issue, and you intentionally B. As you intentionally B, you do A as a “side-effect” action (even though you cannot acquire motivation to A in the relevant temporal interval). There are two ways to handle such actions. First, one might plausibly think that one does not intentionally perform the side-effect action; although one intentionally B-s, one does not intentionally A. However, suppose one believes to the contrary that in such cases, one does intentionally A. Then restrict Motivation/Ability to cases in which it is false that you know or even believe that by B-ing, you will A:
Motivation /Ability: If (a) S believes that it is morally impermissible as of t1 for S to perform A at t2, (b) S lacks any desire to perform A at t2, (c) in the relevant temporal interval between t1 and t2, (ci) S cannot acquire a desire to do A at t2, and (cii) S cannot cease believing that it is morally impermissible for S to perform A at t2, and (d) it’s false that A is a “side effect action,” that is, it’s false that S knows or believes that by B-ing at some time S will A at t2, S has no desire to A, and S intentionally B-s, then as of t1 S cannot intentionally do A at t2.
At the heart of Motivation/Ability is the thought, roughly, that one can’t intentionally do something if one has no reason or cannot acquire a reason to do it. This invites the following objection: the freedom to act crazily. Some existentialists have proposed that it is conceivable that one can do something even in the absence of any reason to do it — what is rationally inexplicable is not impossible. Footnote 22 If we take a desire (as we have broadly characterized it) to be a constituent of any reason, then this sort of existentialist concern speaks against Motivation/Ability. Maybe the existentialists are right — perhaps there are occasions on which we can act crazily. Even so, this sort of freedom to do other than what one reasonably or rationally does is not the sort of freedom that obligation demands. The two-way control that obligation requires — the freedom to do and to refrain from doing something —is, minimally, two-way intentional control. Acting crazily does not give us this sort of dual intentional control.
Motivation/Ability may next be mobilized to show that the range of obligations is narrow. Imagine that the devoted mother in our example above, Ann, must attempt to rescue her son immediately to prevent him from drowning. She believes that she ought to save her son, and it would be impermissible for her not to do so. Can she refrain from saving him? Safely assume that in most, if not all, such situations Ann has no desire to refrain from saving her child. Given Motivation/Ability, the pertinent question is whether she can, during the relevant time interval, acquire the desire to not save him (Dr).
If she can acquire Dr, she can do so intentionally or unintentionally. Regarding the first option, she cannot intentionally acquire a particular desire unless she has a conscious thought concerning that desire. (Compare: if one intentionally wants to acquire a desire to quit smoking, one cannot do so unless one has some conscious representation of that desire.) In the situation, she does not have any such thought about the desire to refrain from saving her child. In addition, she cannot intentionally acquire the thought on the basis of any choice she makes, perhaps by engaging in practical reasoning with the goal of acquiring the thought. To acquire the thought in this way, either she must have the thought already, or she performs some other action as a result of which the thought in question unintentionally occurs to her. The first apparent option is not an option. Regarding the second, as Michael Zimmerman explains, it is very unlikely in any situation, for one immediately, or during the relevant temporal interval, to acquire the thought. Footnote 23 For, first, the number of independent conscious thoughts any one of which an individual is in principle capable of having at one time is extremely large; and second, the maximum number of independent conscious thoughts all of which an individual is capable of having at one time is very small. So, if all thoughts were equiprobable, the likelihood that on some occasion a particular one of them should occur would be exceedingly low. Second, however, not all thoughts are equiprobable; some are far more likely to occur than others, given the agent’s past history, character, circumstances, and so forth. But even if this is so for some thoughts, it will not be so for many others. Indeed, it cannot be; for, given that at most some independent thoughts can occur at once, the increased likelihood of one thought tends to render others less likely. Footnote 24
Regarding the second option — Ann’s unintentionally acquiring the desire to refrain from saving her child — often we do acquire desires unintentionally. For example, strolling in the park on a hot, sunny day, you chance upon an ice cream cart. You find yourself with a desire for a strawberry sundae. The cart, or your perception of it, is a stimulus that triggers your acquisition of the relevant desire. Or sometimes we unintentionally acquire desires on the basis of the way we feel. On hearing about the fall of the stock market, you are doleful. Your feeling of impending doom causes you to acquire a desire to puff on your pipe. Your acquisition of this desire is unintentional. Reverting to the scenario in which the son is drowning, it is not true that there is some stimulus, relevantly analogous to the ice cream cart or Ann’s perception of it, that might contribute to her unintentionally acquiring a desire to refrain from saving her child (Dr). Nor is it true that Ann has relevant feelings that trigger her acquiring Dr. One cannot, however, dismiss the possibility that Ann acquires Dr by sheer chance. But the probability of Ann’s doing so, as Zimmerman cautions, is exceedingly low in as much as the pool of relevant candidate thoughts is very large and the number of thoughts that can be had at any one moment is very small.
In the passage reproduced above, Fischer and Ravizza propose that just about anyone can muster the worry that he is not free to do otherwise. We may readily agree with Fischer and Ravizza that there are contexts, such as the one discussed in their passage, in which this is true. But it should also strike us that there are many contexts in which we would not be able to summon up the thought that we are free to do otherwise. The mother’s context is just of this sort. To explain, presumably, intentionally to summon up the thought that Ann is free to do otherwise, she must have this conscious thought. But in this context, Ann does not have any such thought. Nor does it seem plausible, for the reasons already canvassed, that Ann can intentionally acquire the germane thought by any choice she makes. Similarly, for the reasons previously discussed having to do with the very large pool of candidate thoughts and the very small number of thoughts that can be had at any one time, it is not credible to suppose that Ann can unintentionally acquire the conscious thought that she is free to do otherwise.
So, it is reasonable to think that Ann cannot (during the relevant temporal period) acquire the desire to refrain from saving her son. Recall Motivation/Ability:
Motivation /Ability: If (a) S believes that it is morally impermissible as of t1 for S to perform A at t2, (b) S lacks any desire to perform A at t2, (c) in the relevant temporal interval between t1 and t2, (ci) S cannot acquire a desire to do A at t2, and (cii) S cannot cease believing that it is morally impermissible for S to perform A at t2, and (d) it’s false that A is a “side effect action,” that is, it’s false that S knows or believes that by B-ing at some time S will A at t2, S has no desire to A, and S intentionally B-s, then as of t1 S cannot intentionally do A at t2.
Concerning the elements of this principle’s antecedent, Ann satisfies (a) with the trivial qualification that she lacks any desire to fail to perform a certain action (she lacks the desire to refrain from saving her son). She also satisfies (b), (ci) and, we may safely assume, (d) as well. What about (cii) though? Is it credible to suppose that she cannot, during the relevant temporal interval, cease to believe that it is morally impermissible for her to refrain from saving her son? Surely so; for, first, unless the circumstances are extraordinary, we simply cannot shed deeply ingrained beliefs over at least a short temporal span. Second, assuming we are rational, we would need to acquire reasons to shed deeply ingrained beliefs under ordinary conditions. It seems that an argument parallel to the one that shows that Ann cannot, during the relevant time span, acquire the desire to refrain from saving her son, can be deployed to show that she cannot, during this time span, acquire reasons to shed her belief that it is impermissible for her not to save her son.
With its antecedent satisfied, Motivation/Ability generates the result that Ann cannot, during the relevant temporal period, refrain from saving her child. One noteworthy, intermediary conclusion is apt: If Ann cannot but save her child, then this act which one initially may plausibly take to be obligatory for her is not in fact obligatory for her insofar as obligation requires control, and such control entails that one could have done otherwise.
We highlight one more observation. Ann is precluded from being able to refrain from saving her son by factors beyond her control. She is unable to refrain from saving her son because she is unable intentionally or unintentionally to acquire the desire (Dr) to refrain from saving him during the relevant temporal span. We take each of these options in turn. Ann is unable intentionally to acquire this desire because she lacks the conscious thought concerning it, and, moreover, she cannot acquire this thought. In many situations Ann will not have any conscious thought concerning desire Dr because of the sort of person she is. Furthermore, in many situations, the sort of person she turns out to be is inextricably associated with factors beyond her control, such as her genetic makeup, the sorts of parents she had, the education she received as a child, and the social culture in which she was nurtured. In her situation, Ann cannot intentionally acquire the conscious thought to possess desire Dr because the candidate pool of relevant thoughts is very large and the number of thoughts that can be had at any one moment is very small. Both of these factors are beyond her control.
Ann is unable unintentionally to acquire desire Dr because there is no relevant stimulus that would trigger her acquiring Dr, or she has no apt feelings that would contribute to her acquiring Dr, or the probability of acquiring Dr by chance is excessively low. Each of these things — lack of a proper stimulus, apt feelings, or the low probability — and, as one credibly suspects, any other reasonable candidate as well, is something not within her control. So, we may now draw a second, intermediary conclusion: In many situations Ann is unable to refrain from saving her child because of factors beyond her control.
Finally, it appears that there is nothing special per se about the sort of example — Ann’s saving her child — in that many seemingly obligatory acts are acts of the sort that their agents are unable to refrain from performing in the relevant temporal interval. For instance, suppose you very strongly believe that you ought to contribute to feeding the starving children in East Africa, or you ought to aid the elderly person to cross the street, or you ought to show gratitude to your colleague for her staunch support over the years, or you ought to be loyal to your spouse, and so on. Given pertinent circumstances and the sort of person you are, if you believe it is impermissible not to perform each of these acts, and you have no desire not to perform them, then, it may well be far more probable that in the relevant temporal interval, you cannot acquire the desire to refrain from performing them than that you can acquire the desire. And, again, the previous discussion sheds light on why, if you cannot acquire the germane desires, your inability to do so may well be something that is beyond your control. Drawing on our two intermediary conclusions, we may infer the following: Many acts that one initially may plausibly take to be obligatory (or for that matter, impermissible or permissible) for one are not in fact obligatory (impermissible or permissible) for one insofar as obligation requires freedom to do otherwise, but factors beyond our control preclude us from being able to do otherwise.
It is worth dwelling on an instructive objection before joining this discussion with educational aims. Suppose Ann has made herself into a certain sort of person; she has taken deliberate steps to see to it that she would protect or come to the aid of her children (or others) when called for. Furthermore, suppose she has molded herself to become the kind of person who cannot do (or refrain from doing) various things (such helping her children in germane circumstances). This sort of character formation should not undermine obligation or impermissibility; even if she cannot refrain from saving her child, it is still obligatory for her to save her child because she has freely made herself into the kind of person who cannot but rescue or attempt to rescue her child from impending danger.
Whether agents freely mold themselves, or to what extent they do so, to become the adults they are is controversial. But I will not dwell on this concern. Rather, the objection can be parried by recognizing that it is a variation of the “self-imposed impossibility” objection to Kant’s Law, and appreciating that this objection leaves much to be desired. Imagine you promise Fred that you will return a book not later than Friday morning that you borrow from him on Monday. In virtue of promising, assume that you ought to return the book by Friday. Now suppose that on Wednesday, you sell the valuable book because you are short on cash. Against Kant’s Law, some may object that you still have an obligation to return the book by Friday even though, as of the time you have sold it, you cannot return it. However, friends of Kant’s Law may rejoin that what is in fact true is that prior to selling the book, you ought to return it by Friday in part, because you can, before selling it, keep your promise. After selling it, though, and so no longer being able to return it, it is not true that you ought to return it by Friday. Of course, one can maintain that you ought, up to the time of selling the book, to return it later, and as this (remote) obligation is not satisfied, a wrong is done. But this is consistent with holding that as of the time that you sell it, your returning the book is not obligatory for you; moreover, as of this time you acquire new obligations, such as the obligation to apologize and make reparation. Notice, in addition, that obligations do expire. If you ought to return the book not later than 9:00 a.m., and it is now 9:02, even if you have not sold the book this obligation has expired. At 9:02, you no longer have an obligation to return the book at 9:00, and you do not have this obligation because as of 9:02 you cannot return the book at 9:00. If obligations can expire after the proper time of discharge (9:00 a.m. in our example), why suppose that they cannot expire before the slated time of discharge if the relevant inability has set in prior to this slated time? It is, then, perfectly in order to conclude that once you have sold the book, and thus the relevant inability sets in, your obligation to return it by Friday expires. Footnote 25
For whatever bizarre reason, suppose you have made yourself into the kind of person who will not return a borrowed book on time. The prior considerations lend support to the view that once the pertinent inability sets in, you do not have an obligation to return the book even if it is assumed that before this inability sets in you did have such an obligation.
IV. Joining the Issues: Educational Aims and Freedom to Do Otherwise
The conclusion that many acts which one initially may plausibly take to be obligatory (impermissible or permissible) for one are not in fact obligatory (impermissible or permissible) for one has disturbing implications for child education. I discuss some of them in this section.
A. The defeat of uprightness
I assumed in the first section that one of education’s goals is uprightness: children should be brought up to be agents who are disposed to fulfill their moral obligations and to avoid doing what it is morally impermissible for them to do. Suppose Ann, the devoted mother, received a fairly conventional education. During the course of her upbringing, she attained several moral beliefs and values. Assume that over time she acquired dispositions to perform actions she deemed to be obligatory for her and to avoid performing actions she nonculpably regarded to be impermissible for her. Still, given the reasoning in the previous section, on several occasions on which she performed actions she took to be obligatory for her, she fulfilled no obligation at all (because she could not have avoided doing what she did on those occasions). A sobering thought is that educators, whether parents, teachers, or friends, could have done nothing to see to it that she did end up fulfilling obligations: as obligation requires avoidability, and, on many occasions, she could not have done otherwise owing to factors beyond her control, and, moreover, nothing anyone else could have done would have made it the case that she could have done otherwise on these occasions, what is seemingly obligatory for her on these occasions is not obligatory for her no matter what education she received.
B. The defeat of appraisability
I also assumed in the first section that another of education’s goals is appraisability: Children should develop into agents who are disposed, minimally, to perform actions for which they are not morally blameworthy, and, desirably, to perform actions for which they are morally praiseworthy. Restricting discussion to moral blameworthiness, I argued (in Section II) that impermissibility, just like obligation, requires alternative possibilities. Owing to factors beyond our control, on many occasions we cannot do otherwise. Hence, on many occasions it is not impermissible for us to do what we do on these occasions. To introduce another character, imagine that Sally always does what it is prudentially best for her to do even if so doing requires that she do what it is morally impermissible or what she nonculpably deems it is morally impermissible for her to do. Consider an occasion on which she does what appears to be in her long-term self-interest to do but is seemingly morally impermissible for her — she A-s. Assume this time around that, relevantly like Ann, she cannot refrain from A-ing on this occasion because she cannot, during the pertinent time span, acquire motivation to refrain from A-ing. Is she morally blameworthy for A-ing? Confine attention merely to one element of blameworthiness: its conceptual association with impermissibility.
Some think that one is blameworthy for an action only if it is impermissible for one to perform it. Others think that one is blameworthy for an action only if one nonculpably believes that it is impermissible for one to perform it. (Yet others may endorse both of these principles). If the first is true, then since blameworthiness requires impermissibility, and impermissibility requires avoidability, the person, like Sally, who cannot refrain from doing what is seemingly impermissible will not be blameworthy for such deeds. If the second is true, and presuming the first is not, then a person may erroneously but nonculpably believe that many actions she cannot refrain from doing are impermissible for her to do, and so may be blameworthy for doing them (or at least, so assume). However, if rational and enlightened about her mistake concerning the moral statuses of her actions, her enlightenment would diminish blameworthiness; she would not be blameworthy for pertinent seeming wrong doings.
One may remark that if one of our aims is to see to it that children develop into agents who are not blameworthy for their actions, education is not needed to attain this end; for the most part, left to their own devices children will turn into agents who are frequently not to blame for their conduct. If the germane arguments in the last two sections are cogent, this is indeed so. However, the relevant moral is that no matter what educators do, children will mature into agents who are neither praiseworthy nor blameworthy for much of their behavior, and this should give us reason to pause about whether it is rational to privilege appraisability as an ultimate end of education.
In sum, reduction in what is obligatory or impermissible for a person, owing ultimately to our frequently not being able to do otherwise, together with requirements of blameworthiness that link this species of agent-appraisal and impermissibility, markedly constrains the range of what we are blameworthy for, or would constrain this range if we were not deluded about whether our actions were obligatory, permissible, or impermissible. A similar sort of argument, which I will not develop here, that appeals to the principle that praiseworthiness is essentially tied to obligation or permissibility or nonculpable belief in obligation or permissibility, establishes that the range of things for which we are praiseworthy is considerably narrow as well.
We may conclude that the educational goals of uprightness and appraisability are for the most part unachievable. Alternatively, if these goals are largely beyond our reach, we should reconsider whether these proposed goals should indeed be bona fide goals of education.
C. Spreading the Net: More is at Stake
As we remarked, Siegel takes critical thinking to be education’s ultimate aim. As there is no obvious association between our capacity to think critically and freedom to do otherwise, the sort of argument marshaled to show that uprightness and appraisability are endangered will not threaten critical thinking. Similarly, one may believe that educating for virtuosity should remain squarely in our sights: we should strive to ensure that our children develop into virtuous agents. Despite its being true that frequently we cannot do otherwise, seemingly nothing precludes our acting from generosity, kindness, loyalty, and so forth, if we are appropriately trained and nature cooperates. Indeed, one may think that most of education’s putative ultimate goals remain unaffected by the argument in the previous section. Regarding moral education, Nel Noddings and Michael Slote write:
There seem to be three main philosophical theories of morality (or four, if we separate virtue ethics and communitarianism) that could potentially influence current understanding of moral education. Virtue ethics and mainstream communitarianism would naturally encourage a form of moral education in which schools and parents would seek to inculcate good character in the form of specific (labeled) habitual virtues. Kantian/ Rawlsian rationalism/liberalism would seemingly encourage moral education to take the form of developing certain capacities for moral reasoning and certain very general principles [derived from a general duty of respect for the autonomy and dignity of every person] that can be applied to different moral dilemmas or decisions. Finally, an ethic of care would most naturally see moral education as a matter of children’s coming to an intelligent emotional understanding of the good or harmful effects of their actions on the lives of other people as well as deepening understanding of defensible ways to live their own lives. Care involves caring for oneself as well as others. Footnote 26
Simplifying somewhat, Noddings and Slote suggest that the aims central to the “liberal paradigm” of education are tied to personal autonomy, moral reasoning, and critical thinking, whereas those at the heart of the “nonliberal” paradigm are affiliated with good character, moral sentiment, and caring relationships involving benevolence and kindness. Aims of the first sort are concerned with encouraging self-conscious and conscientious attention to one’s own goals, values, and choices, and promoting obedience to universal (moral) rules and principles. Aims of the second sort are associated with enforcing spontaneous other-directed reactive attitudes and feelings, and inculcating particular acts of caring.
White, who builds on the liberal position R. S. Peters, P. H. Hirst, and R. F. Dearden enunciate in the 1960s, depicts liberal education and its goal as follows:
Education aims at promoting pupils’ personal well-being. In a liberal-democratic society…this will include personal autonomy. Footnote 27
Autonomy depends on the existence of options. Education cannot supply these, but it can make students aware of them. Its job is partly to open up horizons on different conceptions of how one should live — ways of life, forms of relationship, vocational and nonvocational activities. But a broad understanding of options is not enough. Autonomous agents also need to understand themselves. They need to interpret their major goals and establish priorities among them, and to discern possible psychological obstacles arising to their self-directness....They need also to be equipped with qualities of character. They have, for instance, to be able to withstand pressures to conform to what authority or public opinion want them to do. For this they require the critical independence of thought to asses others’ arguments, as well as the moral courage to stand up for their own views. Exercise from as early an age as practicable in making choices and reflecting on these is a further requirement—as is a whole-heartedness of commitment to activities of their own choosing. Footnote 28
In contrast to emphasizing values of individual choice and personal well-being, nonliberal educational theorists call attention to community values, traditions, and good habits and they underscore the desirability to care for other people and to act out of immediate concern for the welfare of others.
It would seem that even if frequently we cannot do otherwise ultimately owing to factors beyond our control, and so, uprightness and appraisability as goals of education are imperiled, surely, the components of an ethics of care, for example, or some of the goals White underscores — autonomy — for instance, can and should remain as fundamental educational goals. For these other goals, allegedly, remain unaffected by the sort of argument that casts doubt on uprightness and appraisability. However, this is overly optimistic as I now argue.
White plausibly suggests that although education cannot supply the options that autonomy — a putative ultimate goal of education — demands, education can make students aware of such options. If nurtured to become autonomous agents, when the time comes our children can make intelligent choices among these options; they can select among “different ways of life, forms of relationship, [and] vocational and nonvocational activities.” Depending upon one’s view of autonomy, if, owing to factors beyond one’s control, one cannot on many occasions do other than what one does on those occasions, one may judge that one’s autonomy concerning one’s pertinent choices on those occasions is severely compromised. Elaborating, when we make choices, that is, when we make decisions, we perform mental actions. The argument developed in Section III for the view that, usually, we cannot do otherwise does not discriminate between mental actions and overt ones. Hence, even if we have the option to choose among different ways of life or vocational activities, we will normally not be able to make choices regarding these ways of life or activities other than the ones we make owing to factors beyond our control. Supposedly, the picture White has in mind when he addresses pertinent options is not merely that education makes us aware of these options. He also envisions that if we are fortunate to have such options, under mundane circumstances our natures would not preclude us from making a choice other than the one we made. But this image is distorted. For example, how Sally turns out to be will depend partly but vitally on her childhood upbringing and her genetic constitution. As I previously explained, such factors play a substantial role in curtailing choice. Or suppose, as Noddings and Slote propose, an ethics of care sees moral education (partly) as a matter of deepening children’s understanding of defensible ways to live their own lives. We have an analogous concern here as the one just discussed in selecting among, for example, different vocational activities. Contrary to what would presumably be desirable, one would frequently be unable to make a choice concerning caring for others or oneself different from the one that one actually made. Footnote 29
It should be stressed, however, that whether lack of freedom to do otherwise per se undermines autonomy is controversial. Many semicompatibilists, for example, who believe that determinism is incompatible with freedom to do otherwise have defended the position that determinism is, nonetheless, compatible with autonomy or moral responsibility. Footnote 30 The problem regarding educational aims that I wish to raise is a somewhat different one.
One of education’s objectives is to channel decisions in various directions. Presumably, we hope that a person like Sally would develop the disposition to do what is morally required of her (assuming it was possible for her to fulfill her obligations) rather than to develop the disposition (she has) to do what is prudentially required of her. An ethics of care would, again, discourage our children from making decisions regarding others that are in conflict with central edicts of such an ethics. White would want our children to mature into agents who make decisions consonant with withstanding “pressures to conform to what authority or public opinion want them to do.” Briefly, even if they cannot refrain from doing or choosing as they do or choose, it is desirable that as they acquire various cognitive and other psychological capacities, children make certain sorts of decision; abbreviating, it is desirable that they make “apt” decisions. But how do we get maturing children to make such decisions? The answer that readily comes to mind, of course, is to educate them: we “open up horizons on different conceptions of how one should live”; we get them to appreciate the advantages and pitfalls of the alternatives; we strive to convince them of the virtues of autonomy, and so forth. Presenting them with such options should, hopefully, influence the choices they make even if, frequently, they cannot refrain from making the choices they do. The extent of this influence, however, is difficult to discern. It is not contentious that factors beyond their control affect the choices they make. But how much control do others, especially, educators have over these factors when practices such as indoctrination are ruled out? It seems none at all or very little over some of them, such as genetic constitution; presumably more over others. It is, then, controversial just how much influence educators have in channeling children’s decisions in certain directions. Perhaps it is minimal, and the rest is simply up to luck — to factors beyond one’s control.
One may advance the following objection. Suppose the skeptical arguments developed above are sound and that, for example, uprightness is generally unattainable since most people most of the time are not able to do otherwise. How much in the end should this affect education’s aims? Isn’t there room here for a kind of innocuous illusionism of the following sort: for practical purposes, let those of us charting the course of education assess its ultimate aims under the (illusory) belief that uprightness for most people is attainable. Admittedly it really is not attainable, but if we teach and aim as if it were, we would be more likely to generate educated adults who, though they are not able to act otherwise, on many occasions, act as if they are seeking to do what they morally ought to do. More so, in those situations in which they are able to act otherwise, they would be disposed to believe for good reasons that what they ought to do is what morality requires of them. Why would not an aim so specified be enough for counting among education’s ultimate aims?
I prefix the reply to this objection with the following remarks. First, I have assumed that lack of freedom to do otherwise is no hindrance to acquiring the capacity for critical thinking. Presuming we educate for critical thinking, we would hope or maybe it should be an aim of education that, as they mature into critical thinkers, children come to realize that, roughly, given their natures, frequently they will not be able to do or choose other than what they do or choose; and so, frequently, they will not do anything it is morally permissible, impermissible, or obligatory for them to do. Moreover, more often than not they will be neither morally praiseworthy nor blameworthy for much of their conduct. Remember, White proposes that autonomous agents need to understand themselves. “They need to interpret their major goals and to establish priorities among them, and to discern possible psychological obstacles arising to their self-directedness.” In striving to understand themselves, children who become critical thinkers should come to appreciate that in certain respects, the moral playing field is fairly level: no matter what one’s religious outlook, or cultural background, or racial denomination, or economic standing — no matter what one’s endowments or fruits of a Rawlsian natural lottery — regarding moral appraisability (being praiseworthy or blameworthy for conduct) and moral uprightness, we are all much more alike than different.
Second, the relatively specific aims of education, such as acquiring reading, writing, and oratory skills, and having knowledge of elementary natural science, literature, art, and so forth, remain unaffected by my skeptical argument.
Third, should virtue appraisals (or at least some of them) remain intact even if we are unfree to do otherwise, we should be mindful of the rationale for educating for virtue. Footnote 31 Particularly, the rationale should not be that the person who acts out of kindness, generosity, and so forth, will be disposed to fulfill her moral obligations and perform actions for which she is not blameworthy. Nor is it transparent that, necessarily, the virtuous agent will live a life that is good in itself for her. For example, if a form of hedonism as an account of personal well-being is true — if, roughly, the good life is the pleasant life — then a “successful” criminal may live a life that is highly good in itself for her. Regarding educating for virtue, perhaps one will be able to develop the rationale that it is (intrinsically) good that one be virtuous, or the world will be better if, generally, people acted from virtue. But then this would affect what are taken to be the ultimate aims of education: maybe the aim of something like social or world betterment would replace the proposed aim of personal well-being.
This last remark points the way to responding to the objection under consideration that advocates mild illusionism. We value truth; we desire to believe the truth (or so we hope). Perhaps educators are not doing anything morally objectionable by attempting (if they do) to turn children into morally upright agents. But if the preceding skeptical arguments are sound (as the objection grants), then if they value truth, educators should stare the facts in their face: more often than not, those they are guiding into adulthood are not going to fulfill their obligations (or do wrong) because, more often than not, they will have no obligations to fulfill (or violate). As the objection queries, what would be amiss if we adopted mild illusionism: if we simply carried on as if (roughly) our natures did not preclude us from having obligations? If we continued to educate for uprightness, appraisability, and so forth, full well knowing that these goals remain largely unattainable? This sort of pretense, though, would usher in this problem: what would be the rationale for educating for, say, uprightness? One couldn’t, of course, duly propose that the fulfillment of obligation is intrinsically good, and this supplies us with a powerful incentive to go along with the pretense. More often than not, there would be no obligations for us to fulfill (or to violate). In the previous paragraph, I alluded to other rationales. One concerned the personal well-being of agents: if we educated for uprightness even realizing that this goal is largely beyond our reach, our children would end up living lives that were overall intrinsically good for them. However, as noted, it is not in the least obvious that living as if one had moral obligations, and conducting oneself in a way in which one would if one were morally conscientious, is conducive to personal well-being. It is not pellucid that morality contributes to personal well-being in the way in which, for instance, enlightened self-interest does not. A second rationale is that if we were to behave in ways in which we would if we were morally upright agents, the world would be better off. Perhaps there might be something to this claim. But then we should be honest with ourselves: we should educate for world betterment. What educators would do if they were aiming, generally, for world betterment might diverge considerably from what they would do if they were aspiring to turn children into, for example, morally upright agents. In addition, if world betterment were the goal educators are aiming for, then they should own up to another challenge: it is far from clear that the world would in fact be better if, acquiescing in mild illusionism, educators aimed at turning our children into morally upright agents.
Reflecting on this objection from mild illusionism, maybe some will suggest that obligation is not what we really we care about. We just want people to do certain things in certain situations, whether or not these things are rightly regarded as obligations. This objection raises two important and complex issues. Here, I simply expose pathways to be explored to respond to questions that these issues generate.
The first intricate issue concerns, roughly, the “value” or importance of various normative assessments, such as morally deontic assessments of permissibility, impermissibility, and obligation, and responsibility assessments of praiseworthiness and blameworthiness. Assessments of obligation are “act-focused”; they are concerned primarily (though not exclusively) with appraisals of actions (omissions, or states of affairs). In contrast, assessments of responsibility are “agent-focused”; in the first instance they are appraisals of agents, and are only derivatively appraisals of the conduct for which agents are praiseworthy or blameworthy.
Regarding assessments of obligation, if someone were to claim that we want Augustine not to steal the pears, and we want him to aid the needy, one important justification for the desired conduct is the presumption that it is morally impermissible for Augustine to steal, and that it is morally obligatory or permissible for him to help those in need. This sort of justification would have no place in a world in which no judgments of obligation, permissibility, or impermissibility were true. If the sort of skeptical argument previously developed concerning morally deontic assessments is sound, we could not rightly counsel our children not to steal on the basis that it is impermissible for them to steal. This “deontic rationale” to refrain from stealing would have to be replaced with some other rationale, perhaps some of the ones discussed in the previous response to the objection from mild illusionism — rationales having to do with, for instance, personal well-being or social betterment.
Like assessments of obligation, evaluations of praiseworthiness and blameworthiness have several dimensions of importance. An intelligent discussion of these dimensions calls for a book-length investigation. Here, however, we have the benefit of drawing from elements of the lengthy, pertinent literature in the free will debate: what would life be like if no one were ever morally responsible for any of one’s conduct? In intriguing work, Derk Pereboom, an optimist of sorts, develops the position that a conception of life without the freedom that responsibility requires would not be devastating to morality or to our sense of meaning in life, and in certain respects it may even be beneficial. Footnote 32 Others are less sanguine. Footnote 33 Needless to say, this is not the place to adjudicate this debate.
The second issue is whether different moral concepts could pick up the slack if we were indeed (largely) bereft of responsibility or obligation. This is, again, an enormously complex issue that should properly be discussed elsewhere. Here, I simply indicate two lines of inquiry. The first, diligently pursued by Pereboom, seeks to uncover “analogues” (moral or otherwise) that replace or serve the function of the sorts of things imperiled if no morally deontic or responsibility appraisals were true. So, for instance, it appears that forgiveness — something central to good, interpersonal relationships — presupposes that the person who is forgiven is forgiven for something it was impermissible for her to do and for which she was morally blameworthy. If no assessments of obligation or responsibility were true at our world, then it would seem that forgiveness would be endangered. Pereboom argues that an analogue of forgiveness would survive: think of the analogue as a sort of decision to continue the relationship despite something bad having befallen the one who is forgiven, and recognizing that the relationship has been impaired as a result of the relevant actions of the forgiven). Footnote 34 Even if no agent is morally responsible for any of her conduct, the agent can still choose to restore a previously impaired relationship with moral resolve, and the willingness to apologize for her untoward behavior. The second dissenting line of inquiry urges that there are no adequate “moral replacements” for at least some (or much) of what would be lost if no assessments of obligation or responsibility were true. So, for example, against Pereboom, Dana Nelkin has recently argued that the roles played by forgiveness and Pereboom’s “relationship restoration” are fundamentally different. Footnote 35
Tying some ends together, whether we merely want people to do various things in various situations, independently of whether these things are appropriately regarded as obligations or whether people are morally responsible for them masks complex issues. My own inclination (supported by preliminary argument) is that much of value in our lives would be lost if we were indeed deprived of obligation or responsibility). Footnote 36
In conclusion, if we equate something’s being a matter of luck (roughly) with something’s being beyond our control, then the gist of this paper may be summarized in this way: Owing to luck, we frequently cannot do otherwise. If we frequently cannot do otherwise, then far less turns out to be morally obligatory or impermissible for us than hitherto believed. Reduction in what is obligatory or impermissible, together with specific requirements of moral responsibility, markedly constrains the range of what we are morally blameworthy or praiseworthy for or would constrain this range if we were not in error about whether our actions were obligatory, permissible, or impermissible. This result should impel us to reconsider whether uprightness — being disposed to do what is obligatory and to avoid doing what is impermissible — and appraisability — being disposed to avoid conduct for which we are blameworthy, and, desirably, to engage in behavior for which we are praiseworthy — should qualify as reasonable ultimate aims of education. Furthermore, owing ultimately to luck, educators may have significantly less influence than they might believe in affecting or directing the sorts of choices that children make or will make as they journey through life.
