A Limit on Intuitionistic Methods of Moral Reasoning

A Limit on Intuitionistic Methods of Moral Reasoning
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  463 A   LIMIT   ON   INTUITIONISTIC   METHODS   OF   MORAL   REASONING   The Journal of Value Inquiry 37 : 463–470, 2003.© 2004  Kluwer Academic Publishers. Printed in the Netherlands. A Limit on Intuitionistic Methods of Moral Reasoning SANFORD S. LEVY  Department of History and Philosophy, Montana State University, Bozeman, MT 59718,USA; e-mail: slevy@montana.edu Philosophers working in applied ethics often appeal to moral intuition to testmoral theories and to resolve moral controversies. Intuitions are understoodin a variety of ways. Some philosophers, following Thomas Reid and G.E.Moore, give them an objectivist interpretation. Others, such as John Rawls,forsake these objectivist pretensions, or bypass that issue altogether. Theymight characterize moral intuition, our moral sense, as analogous to, say, oursense of grammar. 1  The use of moral intuition has come under a number of attacks over the years. For example, when associated with objectivist viewslike those of Reid or Moore, it has been rejected by John Mackie because of its metaphysical and epistemological presuppositions. 2  When divorced fromthe objectivist pretensions, intuition has been attacked by Richard Brandt asnothing but socially inculcated bias having no real epistemic force. 3 There is a different problem with appeals to intuition. Even if we acceptthat such appeals have epistemic value and can be used to resolve moral contro-versies, on certain assumptions, there are limits to what intuitionistic methodscan show and to the kinds of disputes intuition can resolve. The assumptionsare widely shared among moral philosophers, even among those with verydifferent views as to the nature of moral intuition. They are about the properobjects of moral intuition and about when intuition is most likely to yield trust-worthy results. For our purpose, there are four key assumptions.First, moral intuition has as its main object particular cases. The particular judgments are then used to determine the truth or falsity of other moral state-ments, including general moral principles. For example, one of the classicattacks on act-utilitarianism is that it would have us say that certain acts areright which are intuitively wrong.Second, intuitions about simpler cases are more trustworthy and useful thanintuitions about complex cases. Complexity leads to confusion for at least threereasons. Few people are able to keep a large number of features accurately inmind. Indeed, different people often give different judgments about a casespecifically because they focus on different features of the case. Also, it isnot always easy to tell how to weigh competing considerations we do keep  464 SANFORD   S . LEVY clearly in mind. As well, even when intuition is clear in a complex case, it isnot always easy to tell just what in the complex case makes the act right orwrong. It is therefore unclear what conclusions to draw from the case.Third, intuitions about ordinary cases are less trustworthy than intuitionsabout unusual cases. We often already have opinions about the sorts of caseswe come across in ordinary life. The opinions can be strong even thoughformed under less than ideal conditions. We are less likely to have alreadyformed opinions about highly unusual cases and can, hopefully, come to formthem in an unbiased fashion under more or less appropriate conditions.Fourth, when we are engaged in a moral controversy about a topic, we areengaged in the controversy because we have confused, uncertain, and con-flicting intuitions about that topic. We should try to resolve the issue by elic-iting intuitions about a different, but analogous topic. Hopefully, we can findground that is not particularly controversial where we can come to agreement.The areas of agreement can then be used to try to resolve the controversy weare interested in.Putting all this together, work in applied ethics is filled with examples whichare extremely simple, abstracting from as many factors as possible, whichrange from the unusual to the bizarre, and which are often distant from theissue under consideration. As an illustration of all this, we have only to look to Judith Jarvis Thomson’s well-known examples which she used to try toresolve the abortion debate. 4 Given these assumptions, there are important moral controversies that in-tuition cannot resolve. Many moral debates turn on the significance of vari-ous distinctions. For example, opponents of active euthanasia often dependon the distinction between killing and letting die. They argue that even thoughit is sometimes right to withdraw medical treatment and allow patients to die,it is morally different to give them a lethal injection. Proponents of liberal-ized euthanasia, in contrast, often argue that the distinction between killingand letting die is not morally significant. When it is permissible to let some-one die, it is also permissible to kill, other things being equal.On the assumptions about intuition just stated, intuitionists often lack theresources to resolve this sort of debate because they lack the resources to es-tablish that distinctions, such as the distinction between killing and letting die,are not morally significant. The reason is not that the distinctions in questionreally are morally significant. Even if the distinctions are not, intuition can-not provide the kinds of data needed to refute their significance. Suppose weare interested in whether a distinction is morally significant. How might anintuitionist go about arguing that it is not? Let us start with a natural, but faultyapproach. We could begin by embedding the distinction in a moral principle.For example, the distinction between direct and indirect consequences mightbe embedded in the principle of double effect that it is never permissible todo evil directly, but it is permissible to do evil indirectly, as side effect, if there  465 A   LIMIT   ON   INTUITIONISTIC   METHODS   OF   MORAL   REASONING is proportionate reason. An intuitionist could then try to find consequencesof the principle that are false by the test of intuition.This approach for proving a distinction is not morally significant has aserious problem. A single false consequence proves a principle false. But prov-ing a principle false which makes use of a distinction does not refute the moralsignificance of the distinction. There might be other principles that embodythe distinction which avoid that false consequence. A counterexample to theprinciple of double effect refutes that principle, but it does not refute the moralsignificance of the distinction between direct and indirect consequences. Forthe distinction might be morally relevant even if not in the way the proponentof the principle of double effect believes.James Rachels develops another approach to showing that a distinction isnot morally significant when he argues that the distinction between killingand letting die is, in itself, not morally significant. 5  A similar argument wasgiven by Michael Tooley. 6  Though Rachels does not call himself an intuitionistin his essay, he does seem to employ intuitionistic methods. In any event, hisargument embodies a general strategy which may be the best hope of an in-tuitionist for an approach to refuting the moral significance of distinctions. Itwill therefore be useful to view his argument as intuitionistic.There are two problems with Rachels’s argument. The first is not really aproblem with the strategy itself, but with Rachels’s application of it. The sec-ond problem is more serious. It is not just a problem for Rachels’s own ar-gument, but also for the general approach he employs. It is a problem thatintuitionistic methods cannot deal with, given the assumptions mentionedabove. If Rachels’s approach is the best the intuitionist has to offer, this is aserious limit on the power of intuitionism.Rachels attacks the traditional view on euthanasia according to which ac-tive euthanasia is never permissible though passive euthanasia sometimes is.One of the reasons people believe the traditional view is that they assume thatkilling is morally worse in itself than letting die. Rachels rejects this.Is killing, in itself, worse than letting die? To investigate this issue, twocases may be considered that are exactly alike except that one involves kill-ing whereas the other involves letting someone die. Then, it can be askedwhether this difference makes any difference to the moral assessments. Itis important that the cases be exactly alike, except for this one difference,since otherwise one cannot be confident that it is this difference and notsome other that accounts for any variation in the assessments of the twocases. 7 Rachels sets out two cases. In the first, an individual named “Smith” will gaina large inheritance if anything happens to his young cousin. One evening, whilethe child is bathing, Smith sneaks in and drowns him, making it look like an  466 SANFORD   S . LEVY accident. In the second, an individual named “Jones” is in the same positionas Smith. Like Smith, he sneaks in preparing to drown his cousin in the bath.But as he enters, the child slips, hits his head, and drowns. Jones is ready togive an extra push if necessary, but all is quiet. Rachels continues,Now Smith killed the child, whereas Jones “merely” let the child die. Thatis the only difference between them. Did either man behave better, from amoral point of view? If the difference between killing and letting die werein itself a morally important matter, one should say that Jones’s behaviorwas less reprehensible than Smith’s. But does one really want to say that?I think not. 8 Rachels concludes that the “bare difference between killing and letting diedoes not, in itself, make a moral difference.” 9  Of course, Rachels admits, mostactual cases of killing that we hear about really are worse than most actualcases of letting die that we hear about. But the reason for this is not due to adifference between killing and letting die. Other factors tend to be correlatedwith killing and letting die. For example, cases of letting die in which doctorsallow the hopelessly ill to pass on are usually motivated by compassion. Mostkillings are not. It is, perhaps, the difference in motives, as well as other things,and not the distinction between killing and letting die, that determines themorality of the act.At first Rachels’s argument seems cut and dry. If we accept intuitionisticarguments, and if we share Rachels’s intuitions about the Smith and Jonescases, then we seem committed to rejecting the moral significance of the dis-tinction between killing and letting die. But things are not so simple. To un-derstand this, it is useful to look at a commonly drawn analogy between hisreasoning and causal reasoning. Suppose we want to determine whether thereis a causal link between smoking and lung cancer. The first step is to findwhether there is a significant correlation between them. But correlation doesnot itself prove causation because of the possibility of confounding factors.For example, there is a correlation between carrying matches and lung can-cer though carrying matches does not cause lung cancer. In testing for causalrelations, we can sometimes identify and control for likely confounding fac-tors. For example, in medical research, age, gender, and socioeconomic sta-tus are often confounding factors and can be controlled for. One way to controlfor them is by selecting study participants with little or no variability withrespect to them. When we cannot identify all the confounding factors, we cando our best to eliminate their influence by selecting study participants whoare as similar as possible to one another all around.The approach Rachels takes is of this kind. There are correlations betweenacts of killing and wrongness on the one hand, and between lettings die andrightness on the other. To determine whether it is the distinction between kill-  467 A   LIMIT   ON   INTUITIONISTIC   METHODS   OF   MORAL   REASONING ing and letting die itself that makes the difference, or whether there is a con-founding factor, Rachels selects two cases that are as alike as possible, saveone is a killing and the other is a letting die. He reasons as follows. The actsof Smith and Jones are both wrong. Their acts are identical in all morally rel-evant respects save that one is a killing and the other a letting die. Hence thedistinction between killing and letting die is not relevant to whether or not anact is wrong.This argument is invalid. Suppose we are wondering whether having gasis relevant to whether or not a car runs. We find two cars, one owned by Samand the other by Jane, neither of which runs. The cars are identical in all re-spects relevant to their running save that Sam’s lacks gas while Jane’s has gas.If we accept Rachels’s reasoning, whether or not a car has gas is not relevantto whether cars run. This conclusion is obviously unwarranted.Arguments in the form that Rachels advances can show that neither side of a distinction is necessary for a feature, and in that respect, that the distinctionis not significant for whether something has the feature. The examples of Smithand Jones show that neither killing nor letting die is necessary for an act to bewrong. In that respect, the distinction is not significant. The examples of Samand Jane show that neither having nor lacking gas is necessary for a car not torun. In that respect, the distinction is not significant. But this leaves open thepossibility that one side of the distinction is sufficient for the feature whilethe other side is not. If that turns out to be so, then the distinction is signifi-cant after all for whether something has the feature. This is what the car caseillustrates. Lacking gas is sufficient for the car not to run. Having gas is not.Thus, even if we accept Rachels’s judgments about the Smith and Jones cases,it is possible that the distinction between killing and letting die is morallysignificant. It is possible, as many people believe, that killing is sufficient forwrongness in this sort of case while letting die is not.Though the sorts of data Rachels presents, our intuitive judgments aboutthe Smith and Jones cases, cannot prove the conclusion that he wants, anintuitionistic approach does have the resources to deal with this problem. Toshow that a distinction is not morally significant, we need to use intuition notonly to test necessity, but also sufficiency. To show that the distinction be-tween killing and letting die is not morally significant, we need to show notonly that neither is necessary for wrongness, but also that neither is sufficientfor wrongness. This is easy to show in the case of killing and letting die, sincethere are acts that are right by the test of intuition and which are as alike aspossible save that one is a killing and the other a letting die. First, supposethat the life of an innocent Sally is threatened by a killer and that she shootsand kills him. Second, suppose that the life of another innocent person, Sue,is threatened by a killer and that unbeknownst to the killer, he is about to stepon an Australian Death Adder hidden in the leaves. Though normally placid,these snakes have an aversion to being squished. Sue sees the snake, but lets
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks