Metaphysical Realism and Moral Relativism

Citation preview

Metaphysical Realism and Moral Relativism: Reflections on Hilary Putnam's Reason, Truth and History Gilbert Harman The Journal of Philosophy, Vol. 79, No. 10, Seventy-Ninth Annual Meeting of the American Philosophical Association Eastern Division. (Oct., 1982), pp. 568-575. Stable URL: http://links.jstor.org/sici?sici=0022-362X%28198210%2979%3A10%3C568%3AMRAMRR%3E2.0.CO%3B2-L The Journal of Philosophy is currently published by Journal of Philosophy, Inc..

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/jphil.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is an independent not-for-profit organization dedicated to and preserving a digital archive of scholarly journals. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Sat May 12 00:34:56 2007

568

THE JOURNAL OF PHILOSOPHY

METAPHYSICAL REALISM AND MORAL RELATIVISM:

REFLECTIONS O N HILARY PUTNAM'S

REASON. T R U T H AND HISTORY*

P

UTNAM advocates a middle way between what he calls "metaphysical realism" a n d what he calls "total relativism." Metaphysical realism is the view that "the world" is a single complex thing, a connected causal or explanatory order into which all facts must fit. So, for example, there is for metaphysical realism a mind/body problem, and a fact/value problem. Metaphysical realism would have us try to locate the place of mind in physical nature and, similarly, would have us try to specify the place of value i n the world of fact. But Putnam objects to this attempt to locate all facts i n a single world. H e argues that it leads to what he takes to be highly implausible consequences, such as the theory that sensory experiences can be identified with certain physical events, or the "alienated" view that there are n o nonrelative facts about value and morality. Now, it may seem that the only alternative to metaphysical realism in this sense is what Putnam calls "total relativism," a view he finds in the writings of Paul Feuerabend and (sometimes) Thomas Kuhn. T h i s view supposes there is not just one world and one truth, but many worlds and truths, corresponding to all the many different ways of looking at a n d thinking about things-where nothing makes one of these ways "truer" than any of the others. But Putnam argues that total relativism is incoherent, since it allows n o way to contrast truth with falsehood. According to Putnam, the correct alternative to metaphysical realism is not total relativism but rather a view he calls "internal relativism," which he attributes to Nelson Goodman. In this view, there are indeed many worlds a n d many truths, but there are falsehoods too. T r u t h is not identified with correspondence with reality; rather "the only criterion for what is a fact is what it is rational to accept." But this is not to take truth to be merely relative truth, since we are to retain the idea of nonrelative truth as the ideal outcome of rational inquiry. We are to suppose that there are standards of rational inquiry; we are not to think that "anything goes." According to Putnam, we can believe in truth, i n the sense of the * T o be presented in an APA symposium on Hilary Putnam's Reasorz, Truth and Hzstory (New York: Cambridge, 1981), on December 29, 1982. Hartry Field will be co-symposiast, and Hilarv Putnam will comment; see this J O I ~ R N A I . , this issue, 553-567 and 575-577, respectively, for their contributions. 0022-362X/82/7910/0568.$00.80

%

1982 The Journal of Philosophy, lnc.

REASON, TRUTH A N D HISTORY

569

ideal outcome of rational inquiry, without believing in "the world" of metaphysical realism, a single, unified world to which our beliefs must correspond if they are to be true. T h e only world we need to believe in, i n this view, is "internal" to our inquiry, not external to it. Nor is there just one such world, nor only one such truth. There are many distinguishable inquiries, many different ways of looking at things, each with its own ideal limit, each with its own contrast between truth a n d falsehood. Putnam argues, furthermore, that rationality is not confined to science. In his view, moral inquiry is as rational as scientific inquiry and, indeed, he says o u r conception of rationality is part of our conception of h u m a n flourishing, and so truth in science, conceived as the ideal outcome of rational scientific inquiry, is dependent on values. Since we can and must suppose there is a n ideal outcome to rational moral inquiry, we can a n d must suppose there is such a thing as nonrelative moral truth. I n this way, then, Putnam argues for "a non-alienated view." I a m intrigued by this, although I am not convinced. I confess I still feel the attraction of metaphysical realism. I can't help thinking that, yes, there is a single world out there, a single causal or explanatory order, a world which is as it is quite apart from what anyone says or thinks about it. It does seem to me that truth is not just a n idealization of rational acceptability and, indeed, that truth involves a relation between a remark or thought a n d the way things are i n the world. I think Putnam is absolutely right i n identifying the crucial point with a commitment to locating all facts i n one world. Because I believe there is a single causal physical order, I face the questions: "What is the place of mind i n the physical world?" a n d "What is the place of value in the world of fact?" Indeed, I believe it is a serious philosophical error to suggest we can avoid these questions. As I have indicated, Putnam argues that a n attitude like mine will lead to two conclusions he takes to be implausible: the mindbody identity theory and moral relativism. I have different reactions to these two cases. As for the first, I agree that the mind-body identity theory is implausible, but I a m not convinced that belief in a single causal or explanatory order leads inevitably to the mindbody identity theory, rather than to the sort of functionalism Putnam himself first suggested some years ago. As for Putnam's second claim, I agree that a hard look at the place of value i n the world of facts does lead to moral relativism. But I don't see that there is anything problematic about moral relativism.

570

THE JOURNAL OF PHILOSOPHY

Since I d o not have the space to take u p both of these points, I will concentrate on fact/value issue. T h a t is the issue to which Putnam himself devotes the most attention. (As regards the mind/ body problem, Putnam argues that a functionalist will have to resort to a n identity theory i n order to account for inverted-spectrum cases. I have elsewhere suggested a different approach.') Now, the most basic philosophical issue about fact and value arises from the observation that h u m a n morality and evaluation s e e m to depend o n one or another set of standards i n a way that rationality does not. Let us consider this more closely. We normally assume there are basic principles of rationality that apply to all normal human beings. Not that I or anyone else knows what these principles are. T h e relevant principles are principles of reasoning in the sense of the principles i n accordance with which one changes one's view both by adding to it a n d by subtracting from it. They are not principles of logic, which are really principles of implication and not principles of reasoning in the relevant sense. T h e relevant principles need not be a n d in fact are not explicitly accepted by anyone. Instead, these principles are reflected i n or constituted by the way people reason. They are the rules of operation of reason, rather than the initial beliefs and goals from which people reason at any moment. It is such principles that we normally suppose apply to all normal people; we suppose that at bottom everyone reasons i n the same way. Of course, we realize that people differ in their powers of reasoning a n d i n the beliefs and goals they start with. Some people can reason further a n d faster than others. Sometimes something "goes wrongH-considerations are overlooked, or given too much or too little weight. But we suppose that i n the end the same basic principles underlie everyone's reasoning, i n the way that the same grammar may underlie the speech of different speakers who have different vocabularies and different skills at speaking. It is the existence of these common principles that allows us to understand the reasoning of other people a n d to appreciate our own a n d others' mistakes. We come to understand someone else by coming to appreciate that person's reasons for his or her beliefs and actions, or by seeing how that person has made a mistake. Someone w h o reasoned in a fundamentally different way from the way in which we reason would really a n d truly be unintelligible to us. His or her reasons for doing what he or she does would not make sense to us. T h i s is because, being ignorant of the relevant principles of reason-

'

"Conceptual Role Semantics," Notre Dame Journal of Formal Logic, xxrrr, 2 (April 1982): 242-256.

REASON, T R U T H AND H I S T O R Y

57 1

i n g ( i n the sense of change in view), we cannot understand the actions of others by seeing them as instances of explicitly formulated principles. T h e only thing we can do is imaginatively project ourselves into their situation with their beliefs and goals to see how we might reason to their conclusions. In assuming, as we normally do, that we can make sense of other people, given sufficient information about them, we presuppose that everyone else operates in accordance with the same basic principles as we do. There d o not seem in the same way to be moral principles that apply to all normal people. Basic moral principles d o not seem to be among the principles of rationality. (Kant seems to be wrong about that.) Although we cannot state any of the principles of reasoning that govern us, we can state various moral principles. T h i s indicates that the relevant moral principles function as initial assumptions rather than principles of the operation of practical reasoning. However, there d o not seem to be any substantive basic moral principles that everyone either follows or has reason to follow i n the way that there seem to be basic principles of rationality that everyone follows. For any moral principle that might be suggested, there would seem to be any number of nearby people who in fact d o not act in accordance with that principle, where this does not seem due to weakness of will, or ignorance, or irrationality, or stupidity, or a failure to consider or appreciate certain reasoning. So it would seem that such people d o not have any reason to follow that principle. Now there are three possible responses to this situation. First there is my own response, which is to accept this result a n d to conclude that there is moral relativism i n a way there is not what we might call "rationality relativism." Second, there is the response of rationality relativists like Kuhn (as he is sometimes interpreted) a n d Feuerabend who argue that, contrary to appearances, there are n o basic standards of rationality that everyone follows or has reasons to follow. T h i r d there is Putnam's response, which is the response of moral absolutism, arguing that, contrary to appearances, there are substantive basic moral principles that everyone follows or has a reason to follow. l'utnam rejects the first response, which sees a n important difference between morality and rationality, on the grounds that this response must presuppose either or both of two unacceptable views, namely, metaphysical realism a n d the instrumentalist theory of reason. H e rejects the second response, with its commitment to rationality relativism, o n the grounds that it implies total relativism which he holds is incoherent. I a m going to discuss only his rea-

572

THE JOURNAL OF PHILOSOPHY

sons for rejecting the first response, which advocates moral relativism but not rationality relativism. Moral absolutists hold that there are certain basic moral principles that apply to everyone i n the sense that everyone either follows them or has reasons to follow them. These might include such principles as, "Do not lie!" "Do not steal!" "Help those who need help!" "Do not injure others!" a n d so on. Now it is obvious that each of these principles is often violated a n d that there are many people within a twenty-mile radius of here w h o are unmoved by the consideration that a certain action would be lying, or stealing, or failing to help someone, or injuring someone. Putnam considers the problem for moral absolutism posed by dedicated Nazis or by the possibility of Benthamite utilitarians' taking over Australia. I think it is useful to notice that you can find the same problem i n your own back yard. Moral relativism begins at home. Some of the people I have i n mind are more or less professional criminals, others are simple egoists who think only of themselves, others are ordinary people who d o have a n d act o n certain moral principles but w h o see nothing wrong with stealing from their employers, say, or lying to a n insurance company or to their customers, or ignoring accident victims they pass o n the road. Now, it certainly seems quite unlikely that whenever people act in this way it is always because of weakness of will, ignorance, irrationality, stupidity, or a failure to consider or appreciate certain reasoning. In particular, it does not seem that there will always be a n argument or piece of reasoning that could be given to such people that would lead them to appreciate their error if they were fully to understand this reasoning. But then it is unclear what sort of reason they might have not to lie or steal or whatever. T o this, a moral absolutist might well reply that these people d o indeed have a reason not to lie and steal, namely, the fact that lying a n d stealing are wrong. But I d o not find that a satisfactory response. What does the absolutist mean in calling lying and stealing wrong? I see a dilemma here. If the absolutist means merely to express his or her negative attitude toward lying and stealing, then, although I can see that this negative attitude may give the absolutist a reason not to lie a n d steal, I d o not see how the appropriateness of expressing that attitude by saying that lying and stealing are wrong shows in any way that someone else who lacks that negative attitude has any reason not to lie and steal. O n the other hand, if i n saying that lying a n d stealing are wrong the absolutist means to say something that implies that everyone has a reason not to lie a n d steal, then I d o not see what justification there is for such a claim.

REASON, T R U T H AND HISTORY

573

It seems to me that, if the people under discussion d o have a reason not to lie or steal, then it must be possible to give a neutral statement of this reason which specifies the reason i n nonnormative terms while making it intelligible why the specified consideration is a reason for that person. It also seems to me this cannot always be done, and the absolutist will have to fall back on a normative specification, saying the reason in question is that it is wrong for the person in question to lie, for example, or that the person ought not to lie. But, to repeat, I d o not find that to be a satisfactory response. As Putnam insightfully observes, this is connected with metaphysical realism. Since I believe there is a single unified causal and explanatory order, I believe any facts that give someone a reason to d o something must be locatable i n the natural order. I a m skeptical of alleged facts that can be specified only normatively. An "internal realist"" like Putnam, who does not believe in a single unified causal a n d explanatory order, can allow for normative facts that are not locatable in the natural order a n d can therefore be a moral absolutist. But I cannot. It seems to me that reasons are considerations that could influence someone i n the relevant way, considerations that would have such a n influence if fully appreciated and if there were n o (nonmoral) defects i n the person's reasoning. Furthermore, I believe that the basic principles of reasoning are coherence a n d conservatism; one tends to make minimal changes in one's beliefs, desires, plans, intentions, values, etc. which might make them more coherent with each other. In particular, I believe that motivational attitudes, like intentions, plans, values, desires, or goals, result from reasoning only because practical reasoning seeks coherence with antecedent motivational attitudes. I n other words, I accept something like a Humean account of practical reasons. Except for qualifications that are irrelevant in the present context, I believe all practical reasons are "hypothetical" in Kant's sense a n d never "categorical." They all presuppose some antecedent motivational attitude. It may be that I a m therefore committed to what Putnam calls "the instrumentalist conception of reasons." If so, I d o not see what is objectionable in that conception, when it has been properly qualified. Let me now consider what Putnam says against this way of looking at things. As I have already indicated, his main objection is to the metaphysical realism that plays such a n important part in the case for moral relativism. Putnam observes, first, that metaphysical realism involves a theory of truth, since it takes a proposition to be

574

T H E JOURNAL OF PHILOSOPHY

true if a n d only if the proposition describes or represents some locatable part of the single causal explanatory order. Against this Putnam argues we must accept a different view of truth as the ideal limit of rational inquiry. His argument is this: consider the result that would be the ideal limit of rational inquiry. Supposing there are at least infinitely many things, there will always be a n interpretation ( i n the logician's sense) of the language i n which the ideal result of rational inquiry is expressed which counts the result true a n d which satisfies all reasonable "operational constraints" o n such a n interpretation. But then one can deny that the result is true only by invoking an absurd "magical theory of reference." I d o not see that this is conclusive. First, a n d perhaps most importantly, I see n o obvious incompatibility between metaphysical realism a n d the view that truth is the ideal outcome of rational inquiry. O n e might believe both that there is a single causal explanatory order a n d that every aspect of this order is knowable through rational inquiry. Such a belief is not obviously inconsistent. Second, it is perfectly obvious that truth cannot be identified with a n idealization of rational acceptance, since there are many trivial truths it would be irrational to accept. It is irrational to clutter one's mind with useless information, a point Putnam himself almost acknowledges when he discusses the important of relevance in reasoning. So, even if the ideal result of rational inquiry must be counted as true, there will be other truths as well that are not included i n the ideal result of rational inquiry. And it is hard to see how to accommodate these other truths without supposing that what determines which these other truths are is at least in part a matter of how things are i n the world. T h i r d , Putnam's basic argument seems to be fallacious. A key move i n his argument is to observe that any legitimate constraint o n a n interpretation of a language is itself going to be expressed as part of the view that is the ideal result of rational inquiry, so that the interpretation of that result is i n part a n interpretation of the constraints on interpretation. H e concludes that any interpretation of the language counting the result true satisfies those constraints, since it counts them true. But clearly there is a difference between saying ( 1 ) that the ideal result of rational inquiry comes o u t true i n some interpretation that satisfies all reasonable constraints on interpretation a n d saying (2) that it comes o u t true in some interpretation, where the statement of the result includes certain constraints o n interpretation that are themselves interpreted as true in that interpretation. T h e second claim (2) is utterly trivial, since we might simply interpret the statements of constraints to say "any-

575

RE,-\SON, T R U T H A N D H I S T O R Y

thing goes." In particular, given a n y consistent view, be it the ideal result of rational inquiry or not, there will be a way to interpret the language of the view so that the view comes out true, including all statements of constraints o n interpretation contained in the view. I take it Putnam would not suppose this is a good argument for the conclusion that truth is the same thing as consistency! But then Putnam's argument does not show that truth is a n idealization of rational acceptance, either. So, I d o not see that Putnam here provides any serious objection to a view that accepts moral relativism while rejecting rationality relativism, although I may not fully understand his argument. L I L B k R T HARXIAN

Princeton University

REPLY T O T W O REALISTSX Hartry Field's criticisms seem to me, for the most part, to be based on confusions. For example, suppose metaphysical realism is true. T h e n there is a definite set I of individuals of which the world consists (in the world-picture of Field's Science w i t h o u t N u m b e r s , these would be just the space-time points). And there is a definite set P of all properties a n d relations of these individuals. Consider a "language" with a name for each member of I a n d a predicate for each mernber of P. Such a language is not a denumerable language; true. But there is n o reason for a metaphysical realist to think that the true a n d complete theory of the world s h o u l d be expressible i n a d e n u m e r a b l e language. T h i s language is unique ( u p to isomorphism), and the theory of the world-the set of true sentences-is likewise unique. So there is nothing wrong with the assertion (which Field spends pages i n criticizing) that, on a metaphysical realist view, there is "one true theory." Metaphysical realists have always thought i n terms of a n ideal language, not a natural one, with its vagueness, its finite vocabulary, etc. T h e most substantive-but also the most amazing-idea i n Field's paper is the suggestion that a realist can also be a disquotationalist. I shall argue that a disquotational theory can have n o explanatory *.ihstract of a paper to he read in .AP.A symposium o n Hilary Putnam's R r a s o n , T r u t h and Hzstory, commenting o n papers by Hartry Field a n d Gilbert Harman; see this J O I . R K A I . , this issue, 553-567 and 568-575, respectivel\. 0022-362X 82 7910 0575SOO 50

1982 T h e Journnl of Philosophb, Inc