The Moral Value in Promises.
Holly Smith poses a challenging moral problem.(1) She offers examples that appear to show that the moral significance of promising can be nefariously exploited. Her leading example is this:
Nepotism Dean Baker is able to fund just one adjunct faculty position. Baker knows that if she assigns the position to the Classics department, they will regard Baker's cousin as the best candidate and hire him. Baker also knows that History is the one department with a considerably greater need for an adjunct than that of Classics. It happens that although Baker can promise the position to Classics, she cannot promise it to History. However, she can allocate it to either department. Wanting her cousin to have the job, Baker promises the position to Classics and later keeps that promise.(2) (155)
Nepotism is intended to be a case in which Baker misuses the moral effect of promising. Yet Smith defends a consequentialist principle about promises that initially seems to endorse Baker's conduct in Nepotism:
Principle I: An agent morally ought, as of a time, to perform just those combinations of making and keeping promises that are part of the most morally valuable course of action that the agent has available as of that time.(3) (166)
On the supposition that there is always positive moral value in promise keeping and negative moral value in promise breaking, Smith argues convincingly that Principle I implies that often Baker morally ought to make and keep the objectionable promise. Given that supposition, Baker's keeping the promise to Classics would in some instances be valuable enough to render the moral value of this course of action higher than that of her next most valuable alternative--not making the promise and assigning the position to History. Yet it seems clear that Baker morally ought not to make and keep this promise. It is no better than a promise to engage in an act of irresponsible nepotism.
Smith's solution to this problem--the Moral Costs solution--is an evaluative thesis. Smith proposes that keeping a promise per se has no positive moral value, but breaking a promise per se does have negative moral value. Smith distinguishes the underlying value of an act from its full moral value. The underlying value is the part of the moral value of the act not including any value that it may have in virtue of its being a promise making or a promise keeping. The full moral value of the act includes these latter values too, if any. In Nepotism the underlying value of assigning the adjunct position to History is significantly higher than that of assigning the position to Classics. Making the promise has no underlying value. The Moral Costs proposal allows Smith to hold that there is also no positive moral value in keeping the promise to Classics. So the underlying values determine the moral values of Baker's best two alternatives--promising and assigning the position to Classics or, rather, simply assigning the position to History. Thus, Smith can conclude that the most morally valuable course of action for Baker to take, as of times prior to promising Classics the position, is to allocate it to the History department. Principle I then implies that Baker morally ought to assign the position to History and not promise it to Classics (171-72).
So far, so good.
This solution is not finally tenable. Part of the example is that Baker promises the position to Classics. As Smith acknowledges, the combination of moral and evaluative views she defends implies in some versions of this case that, as of times after Baker has made this promise, she morally ought to keep it. The basis in Smith's position for this implication is that Baker's breaking the promise would incur the negative moral value that the Moral Costs proposal attributes to all promise breaking. In some instances that cost outweighs the gain in underlying value from giving the position to the neediest department. Principle I then implies that Baker morally ought to keep her promise to Classics, though she acted wrongly in making it.
Smith accepts this implication of her views. She observes that there are times when wrongly promising makes it morally obligatory to do as promised, and she classifies the relevant instances of Nepotism as being among those times (177, 177 n. 31).
Further specification of the example can render this moral assessment intolerable. The example that results from adding the following elaboration we can call Sheer Nepotism:
Baker makes the promise cynically for the purpose of adopting a posture of being morally committed to the Classics appointment. Baker intends merely to facilitate securing her cousin's appointment. The cousin is not morally entitled to this. Baker realizes that she is sacrificing the fulfillment of a greater academic need. The Classics department would not be harmed by Baker's promptly retracting the promise. Classics is sufficiently attached to its own narrow interests to be unwilling to release Baker from the commitment. Baker is on the verge of retirement. Her breaking of this promise would not tend to make the college administration unreliable in the eyes of the faculty.
In Sheer Nepotism Baker had no good excuse for making the promise, her retracting it would not do any harm, and only by assigning the position to History would she discharge her legitimate academic responsibilities. Under these circumstances Baker morally ought to break the promise.
The Moral Costs proposal attributes negative value to Baker's breaking of the promise, no matter what turns on keeping it and no matter why it would be broken. So in some instances of Sheer Nepotism, the proposed cost of breaking the promise would outweigh the greater underlying value of assigning the position to History. In these cases the Moral Costs proposal has the result that keeping the promise has highest moral value. Principle I then mistakenly implies that Baker's cynical promise, once made, morally ought to be kept.
Smith discusses a problem of this sort (177-78 n. 31). She attributes to Robert Cummins the thought that Baker's allocation of the position to Classics is not morally right from any temporal perspective. Smith suggests the following procedure in order to accommodate this judgment within a moral view committed to Principle I. In morally assessing Baker's alternatives as of a time before Baker makes the promise to Classics, the moral values involved are as the Moral Costs proposal asserts them to be. In particular, as of the initial time negative value attaches to all promise breaking per se. This negative value helps to secure the result that Baker is not enjoined by Principle I to make the promise, since it makes this promise costly to break. The new idea is that in assessing Baker's alternatives as of a time after the promise, no moral value is attributed to breaking the promise per se. This has the effect of allowing the underlying value of reallocating the position to History to be morally decisive as of these later times, according to Principle I.
This solution plausibly identifies the moral status of Baker's alternatives, both before and after making the promise. The solution is based on mysterious claims about value, however. There does not appear to be any sort of moral value that promise breaking could have in relation to times before the promise is made, but not in relation to later times when the promise might be kept or broken. To raise this problem with a question concerning the case at issue: How can any sort of negative value that promise breaking has render it morally wrong in advance for Baker to make the promise, and yet later have no influence on the moral status of Baker's actually breaking the promise?
The solution to be defended here avoids this difficulty, while supporting the consequentialist moral perspective embodied in Principle I. The solution's central claim is that the morally relevant values in promise keeping or breaking arise from the exercise of virtues or vices. First we should see that the way is paved for this approach by part of Smith's defense of the Moral Costs proposal, once it is suitably extended.
3. Asymmetry Denied
Smith regards the Moral Costs solution as going against ordinary moral thinking about promise keeping (175, 181). She argues against this ordinary thinking. She contends that the keeping of a promise per se does not contribute any morally relevant value. We shall see that Smith's reasons for denying moral value to promise keeping per se have analogues that argue equally well that breaking a promise is not intrinsically morally bad.
Smith takes up four likely bases for attributing moral value to promise keeping. She points out that they rely on contingent effects rather than something intrinsic to any such act. It is at most a contingent consequence of keeping a promise that it gains another's trust, that it generates self-respect, or that it sustains a socially valuable institution. The fourth likely basis is that keeping a promise is an important accomplishment when it requires resisting temptation. Smith notes that in many cases no such temptation is present to be resisted (181-83). Thus, nothing here shows any noncontingent moral goodness in promise keeping.
These bases, and Smith's responses, have exact counterparts with respect to breaking promises. It is at most a contingent consequence of breaking a promise that it loses others' trust, diminishes self-respect, or undermine a socially valuable institution. And any evil in giving in to temptation cannot be intrinsic to promise breaking, because such temptation is not always present.
Another defense that Smith considers of ascribing moral value to promise keeping per se begins with the claim that we have more reason to act as promised than we would if the act had not been promised. This difference in reasons is explained by hypothesizing that promising gives higher moral value to performing the promised act than it would have had if not promised. The defense concludes that this higher value attaches to promise keeping, and this is what makes for the stronger reason to do as we promise.
In response Smith observes that people do not think that the fact that some intended act would keep a promise is a reason to promise to do something that would be done anyway. Smith offers the example of Smith herself making and then keeping a promise to her secretary to go grocery shopping on a day when she intends to go anyway. Keeping such a promise does not seem to improve things. Smith infers that it is doubtful that promise keeping per se adds positive value. She suggests that the greater reason to act as promised can be accounted for by the fact that breaking a promise is bad, just as the Moral Costs proposal implies (183-84).
It is important to add a detail to Smith's example of the promise to her secretary. It should be added that when Smith shops on the promised day, she simply does as promised. She complies with the promise, but does not thereby exemplify integrity, or deference to morality, or in any other way show virtue. This addition helps us to concentrate on any moral value in the keeping of the promise on its own. Only with this assumption is it quite clear that making and keeping the promise is no sort of moral improvement in the situation.
When the situation really is otherwise analogous, this same neutrality is found in making and then breaking a promise. Suppose that Smith promises her secretary to shop on Saturday, and then shops on Sunday instead, as Smith had every intention of doing. This does not seem bad, when certain other things obtain that are improbable but needed to have an exact analogy. The secretary is quite likely to suppose that if Smith thinks it worth making the promise to him, then there is something about the promising or what is promised that matters somehow to him, even if he cannot guess what. So if the promise is broken and he discovers this, he will be troubled. This bad effect is a contingent but likely accompaniment of such a promise. Similarly, if someone bothers to make this promise, and then breaks it, the breaking is likely to manifest some vice. Probably the person would have some bad purpose for deceiving the secretary about the day of shopping. We should stipulate away all such potential disanalogies. We can assume that the promise breaking is attributable to an innocent confusion about which day was promised, and that the secretary promptly and permanently forgets the promise as soon as it is made. Then the bad contingent features likely to be involved in breaking the promise are not present. This version of the example is the one that is relevantly analogous to Smith's making and keeping the promise, when that is most clearly worthless. Focusing on this sort of case of promise breaking, it does not seem worse for Smith to break the promise than never to have made it.
Smith's final reason for denying moral value to promise keeping per se employs a timeless perspective on an episode in two lives. Pete promises $25 to a worthy cause, which however does not rely on the promise. Pete keeps this promise. Tom donates $25 to the same cause without promising to do so. Smith contends that from a timeless point of view we can see that what Pete does is no better than what Tom does (184-85).
In order to be sure that we are considering the value of promise keeping itself, we should again stipulate that the case is one in which Pete simply does as promised. He does not act out of an allegiance to his commitment, or out of respect for the value of the worthy cause. These are morally valuable things that might mislead us. Setting them aside, the example seems clearly to work as Smith intends.
Again the analogous cases of promise breaking do not seem morally bad. We can suppose that Pete promises Ralph that he will not donate $25 to a particular worthy cause. Ralph wants this at the time, though he does not rely on the promise. Pete soon innocently forgets it for all time to come, as does Ralph. This ensures that the potentially distracting evils of trouble for Ralph and duplicity by Pete are not present. Pete later donates the $25 to that cause. Tom donates $25 to the cause without having promised not to do so. Here, on reflection Pete's act seems no worse than Tom's, in spite of Pete's having broken his promise.
Thus, each of Smith's reasons for thinking that promise keeping per se is not morally valuable has a counterpart arguing that promise breaking is not intrinsically bad.
We have been evaluating the breaking of a promise on a strict understanding of what that is. Strictly understood, a promise is broken by anyone who fails to do as promised without having been released from the promise. This can be an unforeseen and unintended result of innocent error. In more typical cases, the promisor knows that a promise is being broken. There may be a sense in which "breaking a promise," in contrast to failing to keep a promise, implies having this knowledge or a culpable ignorance of it. Presumably in Nepotism if Baker were to break her promise to Classics, she would know that she was doing so. This raises the question, Does knowingly breaking a promise always have negative moral value?(4)
No. Suppose that at noon Alice promises Zoe to tell her in the evening an elaborate silly joke. In the afternoon Alice learns that a tragedy has befallen Zoe. Alice realizes that it would be horrible to tell Zoe the joke as promised, and comparably bad to ask Zoe in her grief to release Alice from the promise. Alice knowingly breaks the promise. This is not morally bad. The circumstances have rendered it worthless to keep. The promise has become a commitment to a senselessly insensitive act. Nothing is bad about by breaking such a commitment.
Knowingly breaking a promise thus does not always have negative moral value. The most general evaluations that stand scrutiny are that there is usually something bad in the breaking of a promise and there is virtually always something bad in knowingly doing so. This shows that a different account is needed of the origin of the negative value that usually attaches to breaking a promise apart from its effects.
4. Re-evaluating Nepotism
It is best to accept both Smith's arguments against any intrinsic goodness of promise keeping itself and the counterpart arguments against any intrinsic badness of promise breaking itself. This gives us resources to evaluate Nepotism in a satisfactory way using Principle I. We can now deny that Baker ought to keep her promise to Classics. Rather, Baker ought to renounce this wrongful commitment and assign the adjunct position to History, the neediest department. The evaluative basis for this is that there is nothing intrinsically morally bad about breaking a cynical promise made with nepotistic intent, if it is broken for the sake of making a responsible choice. Assigning the position to History has the greatest underlying value. The moral value resulting from breaking the promise by giving History the position is not morally outweighed by any value that derives from the existence of the promise, because in this sort of case keeping the promise would not include anything morally good and breaking the promise would not include anything morally bad.
What about Smith's explanatory grounds for attributing negative moral value to all promise breaking? It does appear that there is more reason to act so as to keep a promise than to perform the same act without having promised. If we do not account for this by attributing intrinsic moral value either to keeping promises or to breaking them, how is this difference in reasons to be explained?
There is no completely general phenomenon here to account for. In typical cases of promising that spring most readily to mind the promise is relied on. Keeping it would exemplify integrity while breaking it would exemplify some vice such as selfishness. There is thus plenty of contingent reason to keep typical promises. But promising does not automatically give the promisor any moral reason to keep it. One sort of example where all moral reason is lacking is a promise to carry out some evil plan, which would be kept just to serve a morally bad end. There would be nothing morally good in keeping such a promise. The mere fact that an alternative would do so is not a moral reason for the agent to take that alternative. Unless the contingent effects of keeping such a promise happen somehow morally to justify its fulfillment, there is no good reason to keep the promise (and of course there is good reason to break it). Thus, denying that there is automatically good reason to act as promised is no liability.
5. Evaluating Promise Making
Sheer Nepotism poses another problem, given the present view that there is nothing intrinsically bad about Baker's breaking her promise to Classics. It seems that in Sheer Nepotism Baker's promising Classics and then breaking the promise by assigning the position to History is morally permissible according to Principle I. More precisely, this conduct seems permissible according to a natural extension of Principle I which adds that an agent is morally permitted as of a time to carry out any sequence of acts that is tied for morally best among those available to the agent at that time. As breaking the promise is now being evaluated, it has no moral cost of its own and produces no other moral cost in the circumstances of Sheer Nepotism. So Baker's making and then breaking the promise seems to be ascribed the same moral value as that of her not making the promise before assigning the position to History. Yet in Sheer Nepotism Baker's making the promise in the first place seems clearly not to have been permissible.
We should take into account the intrinsic values that promise making can involve. When done on a virtuous basis, making a promise involves something intrinsically good. When a promise is made to pursue a bad end, doing this is itself something intrinsically bad because it exemplifies a vice. By recognizing these sorts of value we are in a position to say that Baker's promising the job to Classics includes something intrinsically bad. It was done merely to secure the job that she wants for her cousin and at the expense of academic interests that it was her duty to serve. This is a vicious shunning of duty. Since making the promise is bad in this way, the course of action of making and then breaking the promise is worse than that of never making it and simply giving the position to History. With this added evaluation, Principle I implies that it is not the case that the promise ought to be made. Another natural extension of Principle I asserts that acts that are included only in suboptimal sequences ought not to be done. This extension of the view implies that the promise ought not to be made.(5)
6. Justifying The Values
The evaluations asserted here, in combination with Principle I, imply the intuitively correct moral classifications in our examples. But the values might seem ad hoc.
The evaluations have a unified and credible basis. A choice to keep or break a promise presents a strong appearance of having moral significance beyond the effects of keeping or breaking the promise. This significance is best accounted for within a consequentialist moral view by holding that there are intrinsic moral values involved. It is plausible that these values at least include the instances of virtues or vices that would be involved. More generally, it is plausible that wherever virtue is exercised there is something intrinsically good, and wherever vice is exercised there is something intrinsically bad.(6) G. E. Moore's isolation test for intrinsic value bears this out. If we bring to mind the exercise of a virtue, and we contemplate this in isolation from any contingently valuable accompaniments, we find that we value it nonetheless. Likewise, contemplating the exercise of a vice, while ignoring any contingently associated value, ,yields a negative assessment. This evinces to us that these things have value on their own.
These evaluations enable us to account for the pattern of moral values in promises. The outline is clear. Effects aside, making a promise is typically morally neutral, but not always. This is because making a promise typically involves neither virtues nor vices, but it can spring from a virtue such as compassion or exemplify a vice such as duplicity. Keeping a promise usually includes the exercise of virtues such as integrity and respect for others, though it need not. Knowingly breaking a promise ordinarily involves vices such as selfishness and disrespect for others, though in rare circumstances no vice is exemplified. A promise that is broken because it is forgotten is usually negligently forgotten. If the negligence is not itself a vice, then the selfishness or disrespect from which it usually arises surely is. Broken promises that have been innocently forgotten need not involve any of this and therefore can be morally benign.
Furthermore, there is good reason to think that the intrinsic moral values of promising, apart from contingent effects, are exhausted by the exercise of virtues or vices. Promising is a relation of at least three terms: promisor, promisee, and what is promised. The idea of a promise that is a promise to no one, not even to oneself, seems incoherent. Since this is so, we need to account for the value in what we can call "phantom promises": attempts to promise under a justified but false impression that there is someone to whom the promise is made. For instance, Mark may attempt to promise his dying Uncle Bob that he will take good care of Bob's collection of police whistles, reasonably believing that Bob is alive but unable to respond. The fact may be that Bob has died. In such cases there is no genuine promise, since there is no one to whom the promise is made.(7) Yet for Mark conscientiously to do as he thought he promised seems as morally good as if he had kept the counterpart genuine promise. For Mark irresponsibly to neglect to do as he thought he promised seems as morally bad as if he had irresponsibly broken the counterpart promise.
The virtues and vices in this sort of phantom promise are the same as in the counterpart genuine promise. So we would expect the same moral values, if virtue and vice were their whole source. Since the values do seem to be the same, such cases confirm the view that exercise of virtues or vices are what give promise makings, keepings, and breakings their morally relevant value.
If instead we hold that keeping a promise is a bearer of intrinsic moral goodness, or we adopt the Moral Costs proposal that breaking a promise is morally bad per se, then we have yet to account for the moral value in phantom promise makings, keepings, and breakings. We might allege that it is virtues and vices that give rise to the value in such cases, though not in actual promises. This would be a specially contrived complication. It is simpler and in no way worse to hold that the moral values have identical origins in both genuine and phantom cases in the exercise of virtues or vices.
Thus, the proposed evaluations derive from a coherent and defensible conception of a source of morally relevant value.
(1) Holly M. Smith, "A Paradox of Promising," Philosophical Review 106 (1997): 153-96. Page references in the text are to this work.
(2) Smith's description of the case has been slightly altered, and built into this initial description of the case is the assumption Smith eventually makes that the needier department cannot be promised the position (171 n. 26, 187-88 n. 41).
(3) Smith's Principle I is more elaborate, but not relevantly different.
(4) I am grateful to a reader for the Philosophical Review for raising this question.
(5) Though we have located intrinsic moral value included in some acts of promise making, we can concur with Smith's view that opportunities to promise do not invariably give rise to moral reasons. Returning to Smith's example of promising her secretary, we need not say that it morally enhances her conduct for her first to promise to do something that she intends to do anyway to someone who has no reason to want the promise, and then to keep the promise. This sort of promise is neither virtuously nor viciously made or kept. So the present view of the intrinsic values involved in promising allows that it is without moral value to make or to keep.
(6) Thomas Hurka argues that virtues are intrinsically good, in "How Great a Good is Virtue," Journal of Philosophy 95 (1998): 181-03, and in other writings. Hurka identifies virtues as certain suitable attitudes toward certain bearers of intrinsic value. For present purposes, this view of virtues as attitudes is not the best one available. Conscientiousness, integrity, fidelity, and the other virtues that can be exemplified in promising might rather be dispositions toward action founded in appropriate attitudes. This other view makes the exercise of a virtue a more straightforward matter. It should also be noted that as the notion of virtue is being employed here, a virtue need not be a stable trait. An act might be virtuous because the decision to perform the act is based on a morally good reason. Engaging in some such virtuous action does not imply having any particular enduring disposition or attitude.
(7) If it is thought that the promise was made to the departed Uncle Bob, we can modify the case so that the entire existence of "Uncle Bob" is an elaborate hoax.
University of Rochester
|Printer friendly Cite/link Email Feedback|
|Publication:||The Philosophical Review|
|Date:||Jul 1, 2000|
|Previous Article:||The Lottery Paradox, Knowledge, and Rationality.|
|Next Article:||THE ART OF LIVING: SOCRATIC REFLECTIONS FROM PLATO TO FOUCAULT.|