( I )
The two formal or meta-ethical criticisms I shall deal with in this chapter and the next are both concerned with the relation of fact and value.
Although there are many terms in which human beings express their evaluation of things, “good” and “bad” are the archetypical value terms. All other evaluative adjectives are variations on or derivatives from these two. For brevity and simplicity, I will confine our attention to the meaning of these two basic value terms as they relate to matters of fact.
What I have just said does not apply to the words “ought” or “should.” These are not adjectives; they are the operative verbs in prescriptions or what are called “normative” as opposed to “descriptive” judgments — “ought-statements” as opposed to “is-statements.” “Ought” and “ought not” are the other two fundamental or archetypical terms we shall have to deal with. But it will be easier to deal with them after we have clarified the meanings of “good” and “bad,” for then we shall see that ought and ought not are strictly correlative with only one of the two basic meanings of good and bad and totally irrelevant to the other.
Of the two formal theses that have a critical impact on the commonsense view, the first is associated with the naturalistic or empiricist approach to moral philosophy and the second with the repudiation of naturalism and empiricism. The first thesis is that all evaluations are reducible to describable facts — that, for example, whenever we call something good or bad, these words serve as short-hand for a state of facts that can be reported or described. Although this is the position of the philosophical school that goes by the name of “ethical naturalism” or “ethical empiricism,” it might be more appropriately called “ethical reductionism,” since it holds that whatever can be validly said in discourse that uses terms of value or makes moral judgments can be reductively equated to what we know about the nature of things through our empirical sciences or through other empirical knowledge. It denies that moral philosophy is even a relatively autonomous discipline that has principles of its own.
The second formal thesis is expressed in the proposition that the good is indefinable. This not only (I) excludes the possibility of defining good and bad by reference to the describable properties of observable natural things or processes; it also (2) excludes the possibility of defining good and bad by reference to unobservable or trans-empirical entities. Of the two foregoing points, only the first involves the repudiation of naturalism and empiricism in ethics; neglect of the second point has resulted in the adoption of a misleading name for the logical mistake made by those who attempt to define an indefinable predicate, such as good, whether they attempt to do so in naturalistic or in non-naturalistic terms. Although G. E. Moore, who called this logical mistake the “naturalistic fallacy,” explicitly noted the inaccuracy of that name for the error he is often credited with discovering, the name has, unfortunately, stuck and has been the source of misunderstandings and embarrassments in the controversy he provoked.
If statements of value and normative judgments can all be reduced to statements of fact and descriptive judgments, as the naturalists and empiricists in ethics claim, then it follows, as we have seen, that moral philosophy has no principles of its own and so has no autonomy whatsoever as a discipline. The opposite view — that ethics does have independence as a discipline, either an autonomy that is relative or one that is absolute — is advanced by those who maintain, as Moore does, that the good is indefinable, and also by those who go beyond Moore in maintaining that the fundamental principles of ethics (concerning right and wrong conduct and our unconditional duties or obligations) are known to be true not only without any dependence on empirical evidence, but also without rational demonstration. It is for this reason that they are often grouped together as “intuitionists,” though it must be noted, in fairness to Moore, that this grouping overlooks the fact that such writers as Carritt, Ross, and Pritchard take the more extreme view that ethics is absolutely autonomous, whereas Moore thinks it has only a relative autonomy — not all its propositions are independent of empirical evidence or unsusceptible to proof, but just those that involve the indefinable predicate good.
In this chapter, we shall be concerned only with the reductionism of the naturalists or empiricists in ethics; in the next chapter, we will deal with the indefinability of the good and with the misnamed naturalistic fallacy. Being incompatible with one another, the position of the naturalists and that of the antinaturalists cannot both be true; but since they are not contradictory and do not exhaust the alternatives, they can both be false. I am going to try to show that both are wrong, each for a different reason, because each goes to an opposite extreme, extremes that can be avoided and were avoided in earlier centuries. No moral philosopher of note prior to the eighteenth century claimed absolute autonomy for ethics; none prior to the nineteenth century attempted to reduce all statements of value to statements of fact, or normative judgments to descriptive propositions.
( 2 )Can we defend the distinction between statements of fact and statements of value, and are the latter irreducible to the former?
Waiving for the moment the point at issue (whether statements of value are completely reducible to statements of fact), let me first explain how these two kinds of statements appear to differ. On the one hand, a statement of fact is one that asserts that something is or is not, or asserts that it has certain observable properties, that it behaves in certain observable ways, that it stands in certain observable relations to other things; and it may even take the form of an explanation of the facts described by positing the existence and operation of non-observable entities. On the other hand, a statement of value is one that asserts that something that exists, or some property that it has, or something that it does, or something that has happened or will happen, is good or bad. And the apparent difference of such statements from statements of fact is that the words “good” and “bad” do not designate observable properties or attributes of existent things or processes. Goodness and badness are not matters of observable fact. When we say that something looks or sounds good to us, we do not mean that we see or hear its goodness with our eyes or ears, but rather that what we do see or hear is something we appraise as good.
What has just been said about statements of fact and statements of value can be expanded to include descriptive and normative statements — is-statements and ought-statements. These certainly appear to be formally different: To describe the way things are is one thing; to prescribe how they ought to be is quite another. Applied to human conduct, it appears to make all the difference in the world whether one says how men do in fact behave or how they should or ought to behave. This apparent difference is made much of by those who deny that normative judgments or ought-statements can have objective truth; such truth, in their opinion, can be found only in descriptive propositions or is-statements. I will return to this point in Chapter 13. I mention it here only to clarify the apparent difference between fact and value, between descriptive statements and normative statements.
The naturalist in ethics unhesitatingly concedes the apparent differences that have just been pointed out, including the one concerning the attribution of objective truth or falsity exclusively to statements of fact or descriptive statements; but he contends — and this is his central contention — that the differences are only apparent and without significance, for upon examination it can be shown that every statement of value or normative judgment can be reduced to a statement of fact or descriptive judgment. What truth there is in statements of value or normative judgments is ultimately descriptive truth — truth about matters of fact.
How is this reduction accomplished? I submit that it can be accomplished only in the following two ways.
(1) The first consists in equating the good with the useful, and the better with the more useful. A thing is called good if it serves as a means to some desired end; one thing is better than another if it is a more efficient means. The relation of one thing to another as a means to an end is reducible to a cause-and-effect relationship, or at least to an observable sequence in which the first is seen as leading to the second or resulting in it. When we say that X is good, we are saying something that is subject to empirical observation and testing; namely, that in fact X serves as a means to or results in Y — a state of affairs that we think is good. One further step must be taken by those who claim all values can be reduced to matters of fact. They must argue that everything that is called “good” is so called because it is a means to something else or useful in reaching some result beyond itself. Nothing can ever be called good in itself, good simply as an end and not as a means to anything else, good without being in any way useful. Thus, when we say that X is good because it results in Y and then describe Y as a state of affairs that we also think is good, we must be calling Y good in the same way that we called X good, because it in turn is a means to Z, as X was a means to Y. And if any question is asked about the goodness of Z, we must answer it in the same way that we previously answered the question about the goodness of Y and of X, and so on ad infinitum, for there is nothing that can be called good in itself or good simply as an end.
(2) What we have just seen is how the naturalist reduces statements of value to statements of fact. We must now see how he reduces normative judgments or ought-statements to descriptive judgments or is-statements. The principle of reduction is very much the same. The first mode of reduction equated the good with the useful and asserted that nothing can be called good except as a means. The second mode of reduction converts all oughtstatements into hypothetical statements. just as the first mode of reduction rested on the denial that anything can validly be called good simply as an end, so the second mode of reduction rests on the denial that there are any valid categorical ought statements.
A hypothetical ought-statement always takes the following generic form: “If you want Y, then you ought to do X.” There are many species of this generic form: the hypothetical penal ought: if you want to avoid the sanctions imposed by the law, then you ought to behave in conformity with it; the hypothetical approbative ought: if you want the approval of your fellowmen or of your community, then you ought not to behave in a certain fashion; the hypothetical technological or artistic ought: if you want to produce a certain result, then you ought to take the following measures; the hypothetical pragmatic ought: if you want to make a good life for yourself, then you ought to do this or that in order to achieve it.
I have mentioned these sub-forms of the hypothetical oughtstatement to indicate that unlike is-statements which express the kind of knowledge that can be called “know-that,” these hypothetical ought-statements express a kind of knowledge that can be called “know-how.” We can see at once that know-how is purely factual knowledge, as much as descriptive know-that is. Know-how consists in knowing what steps to take, what means to employ, in order to achieve or avoid a particular result. Thus we see that the second mode of reduction also turns all ought-statements into statements of fact by making all of them hypothetical. If the condition precedent that constitutes the hypothetical part of the statement is questioned, the reduction is simply carried one step further. You may in fact want Y, but should you? Ought you to want Y, which is the condition given for asserting that you ought to want X as a way of getting it? The only answer admitted by the naturalists is that, if in fact you do want Z, and if in fact Y is a way of getting Z, then and only then ought you to want Y.
( 3 )If the foregoing contentions were tenable, the ethical naturalist would be right in maintaining that all statements of value can be reduced to statements of fact and all normative or ought-judgments to descriptive or is-statements. But that is not the case, for two reasons.
First, because there is at least one end that is not a means to anything beyond itself. That is the end posited in the question that common sense has tried to answer. How can we make a good life for ourselves? Whether or not the common-sense answer is true or adequate, the question remains a thoroughly intelligible question, and it asks about something that is good as an end, not as a means. When we say of a whole life that it is good, we cannot be saying that it is good as a means to anything else. Unless the naturalist dismisses the question about a good life as meaningless — which in fact he does not and which, without begging the question, he cannot — the meaning of “good” in the phrase “good life as a whole” cannot be reduced to a matter of fact because a whole life cannot be called “good” as a means to be used for anything beyond itself. But the naturalist may still say: “If you want to make a good life for yourself, then everything you do in the course of it can be regarded as useful to that end and, therefore, good only as a means.” Whether it really is or not is always a question of fact.
This brings us to the second reason for saying that the reductionism of the naturalist is untenable. It would be tenable only if it could not be said, categorically, that I ought to make a good life for myself. Were that the case, then either an individual does or does not in fact want to make a good life for himself, and only if in fact he does, do hypothetical oughts follow. The only residual difficulty the naturalist might then have to slough off is the inexplicable meaning of “good” when it is said of a whole life. That residual difficulty would, in my judgment, be a serious stumbling block for him. In addition, I think I shall be able to show that it can be said categorically that one ought to seek that which is really good for himself. We will see that this is self-evidently true when, in Chapter 10, we understand the distinction between the real and the apparent good. With that clear, I hope to show, in Chapter 11, that the meaning of “good” when said of a whole life is not inexplicable, as it must always remain for the naturalists.
( 4 )Before I leave ethical naturalism, let me explain the misunderstanding that lies at its root, a misunderstanding that need not have occurred if the work of earlier philosophers had been more carefully studied.
This misunderstanding is most explicit in the writings of John Dewey, in the chapter on moral conceptions in Reconstruction in Philosophy and in his major work, Human Nature and Conduct, the very title of which bespeaks the essence of naturalism. That book, by the way, is concerned with the problem of leading a good life, and its attempt to solve the problem by reference to the facts about human nature is quite sound in general and in many of its details. But throughout that book, as well as in his earlier writings, Dewey wages an unremitting attack on the notion of what he calls “fixed ends” or “ultimate ends” — ends that are not themselves means to further ends.
If, by an end, one must always mean a terminal end, an end that can be fully attained at a given moment in time, and an end in which the person seeking it can come fully to rest and say of that moment, as Mephistopheles promised Faust he would be able to say, “Stay, thou art so fair!” — if this were the case, then Dewey would be completely right, because there are no terminal ends in this life. This life’s only termination is in death.
However, long before Dewey struggled with this matter, philosophers had distinguished between two senses in which an end can be called “ultimate” — terminally, in the sense just described, and normatively. An end is a final or ultimate end in a purely normative sense of ultimate if (a) it is a whole good toward the achievement of which all other partial goods serve only as means, and if (b) that whole good is never attained at any moment in time. The only good that satisfies these two conditions is a good life as a whole. It is the only good that is not a means to anything else and so is always an end; and as an ultimate end, it is purely normative and in no sense terminal. Not understanding this is Dewey’s fatal error.
Just as Dewey and other naturalists are right in thinking that, with one exception, all goods are good as means (that one exception being a good life as a whole), so they are also right in thinking that, with one exception, ought-statements are all hypothetical (that one exception being, once again, in discourse about the good life as an end to be sought and about the means to be employed in achieving it). This last point will, I hope, become clearer in the next chapter.
Excerpted from Dr. Adler’s book The Time of Our Lives: The Ethics of Common Sense.