Page 1 of 1

On Goodness and Normativity (Judith Jarvis Thomson)

Posted: Sat Sep 02, 2017 4:00 am
by NonZeroSum
_________

Excerpts from Normativity by Judith Jarvis Thomson:
  • Contents
  • Preface
  • Chapter 1
Earlier versions of various parts of the material were presented at several seminars at the Massachusetts Institute of Technology, and at a seminar I gave at the University of California at Los Angeles in the winter of 2003
Normativity with Judith Jarvis Thomson
- https://youtu.be/NZ4hjuQMGUo


_________

Contents

Preface
I. Goodness
II. Goodness Properties
III. Expressivism
IV. Betterness Relations
V . Virtue/Kind Properties
VI. Correctness Properties (Acts)
VII. Correctness Properties (Mental States)
VIII. Reasons-For (Mental States)
IX. Reasons-For (Acts)
X. On Some Views about “Ought”: Relativism,
Dilemmas, Means-Ends
XI. On Some Views about “Ought”: Belief,
Outcomes, Epistemic Ought
XII. Directives


_________

Preface

The following material is an expanded version of the Carus Lectures that I gave in April, 2003. I thank the American Philosophical Association’s Board of Officers for inviting me to give those lectures that year.

I had originally intended that this material should form the first of two connected volumes on normativity, this one on metaethics, the second on moral theory. In light of the amount of time it took me to think through what seemed to me to be the most important issues in metaethics, I decided not to wait until the second volume is completed, and instead to submit what I have in hand to readers for consideration now.

Earlier versions of various parts of the material were presented at several seminars at the Massachusetts Institute of Technology, and at a seminar I gave at the University of California at Los Angeles in the winter of 2003. Versions of parts were also presented as the Shearman Lectures (2004) at University College, London, as the Howison Lecture (2005) at the University of California at Berkeley, and as a keynote address at the Second Annual Metaethics Workshop at Madison, Wisconsin (2005). I am very grateful for the participants’ comments and criticism on all of those occasions.

I thank Richard Kraut for written comments on an early draft of parts of the material. I also thank Mahrad Almotahari, Selim Berker, Tom Dougherty, Adam Hosein, and Seth Yalcin for helpful discus-sion, and for suggestions for revision of parts of more recent drafts.


______________

Chapter 1

Goodness

1.



Our thinking is rich in what is often called normativity. We think that A ought to be kind to his little brother, that B ought to move his rook, and that C ought to get a hair cut. These are normative judg-ments. Intuitively, they differ starkly from such nonnormative judg-ments as that A kicked his little brother, that B is playing chess, and that C has brown hair.

Moreover, we think that D is a good person, E is a good tennis player, and F is a good toaster. These too are normative judgments. Intuitively, they differ starkly from such nonnormative judgments as that D is a human being, E is a tennis player, and F is a toaster.

Our aim will be to try to come to an understanding of what makes our normative judgments true when they are.

As is plain, we make a great variety of normative judgments. Our judgment that A ought to be kind to his little brother is presumably a moral judgment; our judgments that B ought to move his rook and that C ought to get a hair cut are presumably not moral judgments. Our judgment that D is a good person is presumably a moral judg-ment; our judgments that E is a good tennis player and that F is a good toaster are presumably not moral judgments. We can think of normativity as the subject matter of ethics, but if we do, we need to remember that moral philosophy is not the whole of ethics, but only part of it. So the enterprise we will be engaged in is not restricted to moral philosophy.

I suggest that we should focus on a different difference among our normative judgments. I will call our judgments that A ought to be kind to his little brother, that B ought to move his rook, and that

C ought to get a hair cut, directives. Intuitively, they differ from our judgments that D is a good person, that E is a good tennis player, and that
F is a good toaster, which I will call evaluatives.1

We will want to attend to both kinds of normative judgment. I begin with the evaluatives.


2.

Which judgments are the evaluatives? I gave three examples; what have they got in common? One thing we can say straightway is that they are judgments to the effect that a certain thing is good in a cer-tain respect. Let us have a closer look at such judgments.

Twentieth-century ethics opened with the appearance in 1903 of G. E. Moore’s Principia Ethica, in which Moore made some claims that we do well to begin with.

Moore said at the outset of Principia Ethica:

E
thics is undoubtedly concerned with the question what good conduct is; but, being concerned with this, it obviously does not start at the beginning, unless it is prepared to tell us what is good as well as what is conduct. For ‘good conduct’ is a complex notion: all conduct is not good; for some is certainly bad and some may be indifferent. And on the other hand, other things, beside conduct, may be good; and if they are so, then, ‘good’ denotes some property, that is common to them and conduct; and if we examine good conduct alone of all good things, then we shall be in danger of mistaking for this property, some property which is not shared by those other things: and thus we shall have made a mistake about Ethics even in this limited sense; for we shall not know what good conduct really is.2
A number of things seem to be suggested in this passage. One is perfectly clear: Moore claims that there is such a property as being good, or, as I will sometimes put it, goodness—it is the property that all good things have in common. I will call that the Goodness Thesis.

But more seems to be suggested. One hypothesis is suggested by his example of ‘good conduct’. Moore seems to be suggesting that for a thing to be an instance of good conduct is for it to have the following two properties: being good and being an instance of con-duct. So one might take him to be suggesting, similarly, that for a thing to be a good person is for it to possess the properties being good and being a person, for a thing to be a good tennis player is for it to possess the properties being good and being a tennis player, and for a thing to be a good toaster is for it to possess the proper-ties being good and being a toaster. More generally, for a thing to be good in a respect is for it to possess the properties being good and being of the relevant kind. I will call this thesis the Rationale.

My reason for giving the second thesis that name lies in its rela-tion to the Goodness Thesis. Suppose we accept the Rationale. There are things that are good people, good tennis players, and good toasters. From the Rationale, it follows that there are things that possess the property goodness. So there is such a property—and the Goodness Thesis is true.

Did Moore really accept the thesis I am calling the Rationale? I suspect that he did. But whether or not he did, it won’t do.

Many philosophers have drawn attention in recent years to the fact that it won’t do.3 Peter Geach said that the adjective “good” is not a predicative adjective: it is an attributive adjective. We can express what he had in mind as follows.

Let K1 and K2 be any pair of kinds of things, and let “adj” be any adjective. “Adj” is a predicative adjective just in case the conjunction of the propositions

and

A is a K2

entails the proposition

A is an adj K2.

“Red”, he said, is a predicative adjective, since the conjunction of

A is a red car

and

A is a Mercedes

entails

A is a red Mercedes.4

By contrast, “big” is an attributive adjective, since the conjunction of

(1) A is a big mouse

and

(2) A is an animal

does not entail

(3) A is a big animal.

“Good” is also an attributive adjective, since the conjunction of

(4) A is a good tennis player

and

(5) A is a chess player does not entail

(6) A is a good chess player.

Other attributive adjectives are “tall”, “slow”, and “heavy”.

A consequence of the fact that “big” is an attributive adjective is this: we cannot say that for a thing to be a big K—mouse, animal, teapot, chair—is for it to have the following two properties: being big and being a K. For suppose that being a big K was having the following two properties: being big and being a K. Then (1) would be equivalent to


(1*) A is big and A is a mouse,

and (3) would be equivalent to

(3*) A is big and A is an animal.

It is plain, however, that the conjunction of (1*) and (2) entails (3*). So if it were the case that being a big K was being big and being a K, then the conjunction of (1) and (2) would entail (3). Which it doesn’t.

Similarly for “good”. We cannot say that for a thing to be a good K—tennis player, chess player, toaster—is for it to have the follow-ing two properties: being good and being a K. For suppose that being a good K was having the following two properties: being good and being a K. Then (4) would be equivalent to

(4*) A is good and A is a tennis player,

and (6) would be equivalent to

(6*) A is good and A is a chess player.

It is plain, however, that the conjunction of (4*) and (5) entails (6*). So if it were the case that being a good K was being good and being

a K, then the conjunction of (4) and (5) would entail (6). Which it doesn’t.

According to the Rationale, for a thing to be good in a respect is for it to possess the properties being good and being of the relevant kind. That is false, since being a good K isn’t being good and being a K.

We should notice a second reason for rejecting the Rationale, which I’ll express as follows: there isn’t always such a thing as ‘the relevant kind’. What I have in mind is this. I take it that a thing can be a big K for very many K—mouse, animal, teapot, chair, and so on. Similarly, a thing can be a good K for very many K—tennis player, chess player, toaster, and so on. But a thing can be good in a respect without there being any kind K such that for a thing to be good in that respect is for it to be a good K. As we may put it: “good” has a much broader range of occurrences than “big” does. For we may say, not only for very many K that a thing is a good K, we may also say, for example:


A is good at doing crossword puzzles

A is good for England

A is good for use in making cheesecake

A is good to use in teaching elementary logic

A is good to look at

A is good in Hamlet

A is good as Ophelia in Hamlet

A is good with children.5

To say any of these things is to say that A is good in a respect; but there are no kinds K such that to say these things is to say, for some K, that A is a good K.

So we must plainly reject the Rationale.


3.

Perhaps we should have focused on something different in that pas-sage I quoted from Moore, namely:

. . . other things, beside conduct, may be good; and if they are so, then, ‘good’ denotes some property, that is common to them and conduct. . . .

We very often say sentences such as “A is a good tennis player,” “A is a good toaster,” and “A is good at doing crossword puzzles.” But we also very often say the sentence “A is good.” Perhaps what Moore had in mind was really this: when we say “A is good,” we are ascribing to A the property goodness—and the property goodness just is the property had in common by all those things such that if you said the sentence “That is good” about them, you would be speaking truly. I will call this thesis the Alternate Rationale.

My reason for giving this idea that name lies in its supplying an alternate rationale for the Goodness Thesis. For suppose we accept the Alternate Rationale. There are things such that you would be speaking truly if you said “That is good” about them. From the Alternate Rationale, it follows that there are things that possess the property goodness. So there is such a property—and the Goodness Thesis is true.

Did Moore really accept the thesis I am calling the Alternate Rationale? I suspect that he did. But whether or not he did, it won’t do.

It pays to notice first that anyone who opts for the Alternate Rationale has a job ahead of him. Suppose we were to grant that when we say “That is good” about a thing we are ascribing the prop-erty goodness to it. Now when we say “That is a good toaster” of a thing, we are presumably ascribing the property being a good toaster to it. But as we just saw, the property being a good toaster is not the conjunction of the two properties being good and being a toaster. So how does what we ascribe when we say “That is a good toaster” connect with what we ascribe when we say “That is good”? There ought to be some answer to that question. After all, it is one and the same word “good” that appears in both sentences. But it is not in the least clear how you are to answer it if you accept the Alternate Rationale while rejecting the Rationale.

Let us bypass that difficulty. Is it true that when we say “That is good” about a thing we are ascribing the property goodness to it?

And that the property goodness just is the property had in common by all those things such that if you said the sentence “That is good” about them, you would be speaking truly?

We very often say such sentences as “A is a big mouse,” “A is a big animal,” and “A is a big teapot.” But we also very often say the sentence “A is big.” Should we similarly say that when we say “A is big,” we are ascribing to A the property bigness—and the property bigness just is the property had in common by all those things such that if you said the sentence “That is big” about them, you would be speaking truly?

It’s a terrible idea! Suppose that on Wednesday, I show Smith my new pets: six mice. Five are small mice, one is a big mouse. “Remarkable,” Smith says about my sixth, “That’s big!” Does he speak truly? Suppose he does. Then the idea we are looking at yields that my sixth has the property bigness. On Thursday, Jones tells me about his trip to the Museum of Natural History, which currently has a display of animals that describes the sources of the immense vari-ety in animal size. “That,” he says, by way of example, pointing to my sixth, “is small.” Does he speak truly? Suppose he does. If there is a property bigness, then there surely also is a property smallness, so the analogue of the idea we are looking at yields that my sixth has the property smallness. Did my sixth have bigness on Wednesday and smallness on Thursday? Well, it didn’t shrink over night. So did it, throughout both days, have both bigness and small-ness? If so, what on earth can those properties be ?

We could, of course, insist that when Smith said on Wednesday “That’s big,” he didn’t speak truly, that one doesn’t speak truly if one says “That’s big” of any mouse. That one doesn’t speak truly when one says “That’s big” unless the thing is as big as an elephant. Or better, as big as the Empire State Building. Or better, as big as our galaxy—though even that is dubious, for our galaxy is a small galaxy. To say this is to float free of reality. When people say, in amazement, “That’s big” of a certain mouse or cat or tree or house, they may very well be speaking truly.

Similarly for goodness. When we were choosing tennis players for our tennis team, Smith said of Alfred “Choose him. He’s good.” When we were choosing chess players for our chess team, Jones said of Alfred “Don’t choose him. He’s bad.” Did Smith and Jones both speak truly? If so, then Alfred had both the property goodness and the property badness; and if so, what on earth can those properties be ? Or should we insist that Smith didn’t speak truly? Even if Alfred really is a good tennis player? What’s missing in Alfred that he would have to have or be if Smith was to have been speaking truly of him when he said “He’s good”?

It’s all a mistake. When people say “That is big” of a thing, they are not ascribing a property bigness to it: they are ascribing differ-ent properties to the things. One person may be ascribing the prop-erty ‘being a big mouse’. Another may be ascribing the property ‘being a big animal’. Yet another may be ascribing the property ‘being a big teapot’. More generally, a person who says “That is big” is, for some kind K, ascribing the property ‘being a big K’. The con-text in which a person says “That is big” tells us which kind K it is such that he is saying that the thing is a big K, for it tells us which kind is under discussion—thus perhaps mice, or animals, or teapots. Where a speaker can assume that his hearers know which kind is under discussion, he need not say “That is a big K”; he can say, more briefly, “That is big.” But if we don’t know which kind is under dis-cussion when we overhear a person say “That is big,” then we don’t know what property the speaker is ascribing to the thing.

We should be clear that there is a difference between its being the case (i) that a person is, for some kind K, ascribing to a thing the property ‘being a big K,’ and its being the case (ii) that a person is ascribing to the thing the property ‘being a big K for some kind K’. There is such a property as ‘being a big K for some kind K’, but it is a boring property, a property that just about every physical object has. (Perhaps it is lacked only by the smallest of the small strings that physicists tell us everything is composed of.) So I doubt that any-body ever ascribes that property to a thing in saying “That is big”. What a person ascribes to a thing in saying that sentence is instead, for some K, the property of being a big K.

Similarly for people who say “That is good” of a thing. They are not ascribing a property goodness to it: they are ascribing different properties to the things. One person may be ascribing the property ‘being a good tennis player’. Another may be ascribing the property ‘being a good chess player’. Yet another may be ascribing the prop-erty ‘being good at doing crossword puzzles’. More generally, a per-son who says “That is good” is, for some respect R, ascribing the property ‘being good in respect R’. The context in which a person says “That is good” tells us which respect R it is such that he is saying that the thing is good in respect R, for it tells us which respect is relevant—thus perhaps tennis playing, chess playing, or doing crossword puzzles. Where a speaker can assume that his hearers know which respect is relevant, he need not say “That is good in respect R”; he can say, more briefly, “That is good.” But if we don’t know which respect is relevant when we overhear a person say “That is good,” then we don’t know what property the speaker is ascribing to the thing.

And again, we should be clear that there is a difference between its being the case (i) that a person is, for some respect R, ascribing to a thing the property ‘being good in respect R’, and its being the case

(ii) that a person is ascribing to the thing the property ‘being good in respect R for some respect R’. There is such a property as ‘being good in respect R for some respect R’, but it is a boring property, a prop-erty that just about everything has. Indeed, I should think that every-thing has it. If you think that you have fastened on a thing that is not good in any respect at all, then you should notice that it is good in at least this respect: it is good for use in a discussion of the question whether everything is good in at least some respect. So I doubt that anybody ever ascribes that property to a thing in saying “That is good.” What a person ascribes to a thing in saying that sentence is instead, for some respect R, the property of being good in respect R.

I asked earlier: how does what we ascribe when we say “That is a good toaster” connect with what we ascribe when we say “That is good”? And I said that there ought to be some answer to that question—after all, it is one and the same word “good” that appears in both sentences. The answer to it is easy. When Smith says “That is good,” the property he is ascribing to the thing may just be the property being a good toaster. Though of course it may instead be the property being good for use in burning secret messages from one’s spies.



4.

Let us return to Moore. The Goodness Thesis says: there is such a property as being good, or, alternatively, goodness—it is the prop-erty that all good things have in common. I said that I suspect that he believed both the Rationale and the Alternate Rationale, and I suspect that that is why he believed the Goodness Thesis. Thus he simply overlooked the fact that “good” is like “big” in being an attributive adjective.

In that “good” is like “big” in being an attributive adjective, there is no such property as goodness just as there is no such property as bigness.

Moore was not the first philosopher to overlook the fact that “good” is an attributive adjective, but in light of the impact Principia Ethica had on twentieth-century moral philosophy, he made a major contribution to the two bad effects of that omission. Let us have a look at them.

First, the omission had a bad effect on twentieth-century metaethics. Suppose we think that there is such a property as goodness, and that it is the property that people ascribe to the things they refer to when they say “That is good,” thus the property that all and only good things have in common. Then the property goodness that we are committed to is epistemologically dark. It is fairly easy to find out whether a thing is a good tennis player or a good ham sandwich; how is one to find out whether the thing is (simply) a good thing? The other side of this coin is that the property goodness that we are committed to is metaphysically dark. For what could it consist in? It is clear what it is for a thing to be a good tennis player or a good ham sandwich; what could it be thought to come to for a thing to be (simply) a good thing?

Moore himself concluded that goodness is a nonnatural property. And what is that? Moore’s characterization of the notion ‘nonnatural property’ in Principia was grossly unsatisfactory, as he himself later recognized.6

Others resisted. They concluded (rightly) that there is no such thing as the property goodness, which a person who says “A is good” is ascribing to A. But they thought that the only available alter-native was to say that a person who says “A is good” ascribes no property at all to A. Then what does the speaker do? They said he merely displays a favorable attitude toward A. Those who chose this option came to be called Emotivists—in later years, they came to be called Expressivists—and twentieth-century metaethics was dominated by debate over whether they were right.

What was overlooked was the markedly better third alternative, namely that “good” is an attributive adjective. Opting for this alter-native allows us to avoid the epistemological and metaphysical darkness consequent on taking it that there is such a thing as the property goodness which a person who says “A is good” is ascribing to A. (Compare the epistemological and metaphysical darkness that would be consequent on taking it that there is such a thing as the property bigness which a person who says “A is big” is ascribing to A. It is fairly easy to find out whether a thing is a big house or a big football player; how is one to find out whether the thing is [simply] a big thing? It is clear what it is for a thing to be a big house or a big football player; what could it be thought to come to for a thing to be [simply] a big thing?) And it also allows us to have that when a person says of a thing “That’s good,” he does ascribe a property to the thing, and that the assertion he makes about the thing is true or false, and that there is room for asking or wondering which it is, and for arguing with those who disagree.

I said that overlooking the fact that “good” is an attributive adjective had two bad effects. I have mentioned one, namely its effect on metaethics. The other was its effect on moral theory.

Consequentialism is the thesis that what we ought to do is (very roughly) to maximize the amount of goodness in the world. So if of all the courses of action open to you at a time you would produce most good by doing such and such, then, according to the Consequentialist, you ought to do the such and such. What is good-ness? Well, following Moore, it is the property which a person who says “A is good” is ascribing to A, thus the property that all and only good things have in common. Then it is intuitively plausible to think that those who resist Consequentialism have the burden of proof— they are under pressure to explain how it could be perfectly all right for a person to choose to produce less of what has that property than he could produce. Twentieth-century moral theory was dominated by debate about efforts to carry that burden—a debate that issued from a mistake.


5.

How could Moore have overlooked the fact that “good” is an attribu-tive adjective? Wasn’t he, after all, the patron saint of what later came to be called Ordinary Language Philosophy? Well, his attention was fixed on philosophy, and not on ordinary language.

What I have been concerned with is our actual, ordinary, com-mon or garden use of the word “good”. You are standing in front of the array of melons at your grocer’s, feeling helpless. Your grocer notices. He points to one in particular, and says, “That one’s good.” It is quite certain that he is not ascribing Moore’s (putative) property goodness to the melon he points to; what he is ascribing to it is probably the property of being a good melon. It would be utterly astonishing if when you asked, “Do you mean that that’s a good melon?” he replied, “Oh dear me no, I haven’t the faintest idea whether it’s a good melon, I meant only that it’s a good thing.”

It is worth stress that the fact that our actual, ordinary, common or garden use of the word “good” is such as to mark it as an attribu-tive adjective is not itself a philosophically deep fact that calls for argument: it is an empirical fact about ordinary English usage.

However, there are places where one finds “good” used differ-ently: many books by philosophers. W. D. Ross drew attention to the attributive use of “good”, and allowed that in ordinary life it is much the most common. But he said that there is another use of “good”, a philosopher’s use, “as when it is said that knowledge is good or that pleasure is good.” A philosopher who says “Knowledge is good” or “Pleasure is good” does not mean, for some respect R, that knowl-edge, or pleasure, is good in respect R. He means that knowledge, or pleasure, is (simply) good. And Ross said that it is this—its use to ascribe a property—that is the use of “good” that is most important for philosophy.7

It is certainly not impossible for an adjective to have two uses, only one of which is attributive. Here is an example: “famous”. We might at first thought have taken it to be only an attributive adjec-tive, for there are pairs of kinds of things K1 and K2 for which the conjunction of

A is a famous K1

and

A is a K2

does not entail

A is a famous K2.

For example, the conjunction of the propositions that A is a famous novelist, and that A is a physicist, does not entail the proposition that A is a famous physicist. (A may be famous as a novelist, that is, well known as a novelist, but not well known as a physicist.) And a per-son who says “A is famous” meaning, for some K, that A is a famous K, is using “famous” as an attributive adjective.

But on second thought we remember that “famous” has a second use, for a person may say “A is famous,” meaning to ascribe to A the property of being (simply) famous—that is, the property of being (simply) well known. So there is nothing in the mere fact of an adjective F’s having an attributive use that prevents it from also hav-ing a use to ascribe the property F-ness.8

What assures us that “famous” does have this second use is that we know what the property of being (simply) famous is—it is the property of being (simply) well known. What is the property that Ross claims is ascribed to knowledge, or pleasure, by a philosopher who says “That is good” of it?

When philosophers ask, as many have done, “Is knowledge, or pleasure, good?” what exactly do they mean? Geach said there is no good reason to think they mean anything at all.

Geach was too hasty, I think. Here is a hypothesis. What those philosophers want to know is what people ought to do—for exam-ple, whether we ought to act in such a way as to conduce to or pro-mote the existence of A, more briefly, whether

(1) We ought to promote A.

And when they ask whether

(2) A is good,

what they are asking is whether A has the property such that A’s having it would make (1) true. Alternatively put, when such a philosopher asks whether (2) is true, what he is asking is whether A has the property that is ‘conclusively ought-making’.

Now that hypothesis is over-simple, for most (all?) of the philoso-phers I have in mind accept that it is possible that we ought to pro-mote A, but only where promoting A does not conflict with promoting B, which takes pride of place. (As Ross said, some philosophers say that knowledge is good; and some say both that knowledge is good and that pleasure is good, one taking pride of place.) When such a philosopher asks whether (2) is true, then, what he is asking is not whether A has the property such that A’s having it would make (1) true, but instead whether A has the property such that A’s having it would count in favor of (1)’s being true, and, if other things are or were equal, would make (1) true. Alternatively put, when such a philosopher asks whether (2) is true, what he is asking is whether A has the property that is ‘prima facie ought-mak-ing’ or (as some put it) ‘pro tanto ought-making.’ 9

So if my hypothesis is right, then the philosophers we are con-cerned with do mean something when they ask, “Is pleasure good?” What they are asking is whether pleasure has the property that is prima facie ought-making.

Alas, we have to ask: what property is that? Three answers sug-gest themselves.

(i) The property they have in mind is just, simply, prima facie ought-makingness. Thus when they ask, “Is pleasure good?” what they are asking is just, simply, whether other things being equal, we ought to promote pleasure. But if that is what a philosopher means when he asks, “Is pleasure good?” we might well wonder why he asks his question in those words. Why does he help himself to the word “good”? Why doesn’t he just, simply, ask whether other things being equal, we ought to promote pleasure?

We can surely suppose that it is not for nothing that he asks his question in those words. Yes, he does want to know whether other things being equal, we ought to promote pleasure; but he thinks that that is true only if pleasure has the property that would make it true—and it is because he takes “good” to stand for that property that he asks his question in the words “Is pleasure good?”

The property, then, is the property whose possession by pleas-ure would make it the case that other things being equal, we ought to promote pleasure, that being the property that he takes “good” to stand for.

But what property is it? He can’t really take “good” to stand for

(ii) the property being good in some respect or other. Everything is good in some respect, so whatever A may be—whether pleasure or anything else—the question “Is A good in some respect?” is all too easily answered yes.

At the bottom of the barrel is the remaining alternative, that the property he takes “good” to stand for is (iii), our old friend, the prop-erty goodness. Unfortunately there is no such property, so there is no room for thinking that A’s possessing that property would make it the case that other things being equal, we ought to promote A.10

In the end, then, Geach was right: the philosopher who asks, “Is knowledge, or pleasure, good?” is not asking an intelligible ques-tion—it is no more intelligible than the question whether the melon your grocer points to is (not a good melon, but all simply) a good thing.

But only in the end. It is worth taking note of what it is (if my hypothesis is right) that motivates such philosophers, for there unquestionably is a link between a thing’s being good in thus and such respects, and its being the case that other things being equal, we ought to promote it. Arguably, it is never the case that a person ought to do a thing unless his doing it would be good in certain respects; and arguably, there are respects in which a person’s doing a thing would be good such that it follows that other things being equal, he ought to do it. What goes wrong in the literature I refer to is only the idea that there is something simple at work here—the idea that the link between “good” and “ought” is like that between a hook and an eye.


6.

I asked earlier: which judgments are the evaluatives? I gave three examples, namely that D is a good person, E is a good tennis player, and F is a good toaster. They are obviously judgments to the effect that a certain thing is good in a certain respect.

We also took note of the existence of such judgments as that G is good at doing crossword puzzles, H is good for England, and I is good for use in making cheesecake. These too are judgments to the effect that a certain thing is good in a certain respect.

We can surely say that all judgments to the effect that a certain thing is good in a certain respect are evaluative judgments. Let us look at a way of organizing them.



Notes

1 That there is a difference between what I call directives and evaluatives is familiar enough. I borrow the names from David Wiggins, “Truth, Invention, and the Meaning of Life,” Needs, Values, Truth (Oxford: Oxford University Press, 1998), 95, with an emendation: I substitute “evaluatives” for his “evaluations”.

2 G. E. Moore, Principia Ethica, rev. ed., ed. Thomas Baldwin (Cambridge: Cambridge University Press, 1993), 54.

3 See Peter Geach, “Good and Evil,” Analysis 17 (1956); Paul Ziff, Semantic Analysis (Ithaca: Cornell University Press, 1960); and G. H. von Wright, The Varieties of Goodness (London: Routledge & Kegan Paul, 1963). I take it that Philippa Foot’s “Utilitarianism and the Virtues,” Mind 94 (1985), can be interpreted as expressing the same objection. I have discussed these matters in a number of places in recent years, for example, in “The Right and the Good,” Journal of Philosophy 94 (1997), and in “The Legacy of Principia,” Spindell Conference 2002, Southern Journal of Philosophy, Supplement (2003).

J. L. Austin’s discussion of the adjective “real” is worth drawing attention to in this connection. He said that “real” is “substantive-hungry”: “whereas we can just say of something ‘This is pink,’ we can’t just say of something ‘This is real’. . . . An object looking rather like a duck may be a real decoy duck (not just a toy) but not a real duck.” And he drew attention to the fact that “real” is in this respect like “good”. See his Sense and Sensibilia (Oxford: Oxford University Press, 1962), 69–70.

4 For more on “red” and other issues that I discuss in this chapter, see Addendum 1 on “Red” and “Good”.

5 Paul Ziff drew attention to still other uses of “good”, as in saying “Good morn-ing!” (I might say this to you even though I am well aware that it’s a terrible morn-ing, rainy and windy.) Again, consider the butler who replies “Very good, Sir,” when his master has asked that dinner be served at eight. (The butler is not telling his mas-ter that serving dinner at eight is a very good idea.) See Ziff, Semantic Analysis, 207–10. I throughout ignore these nonsentential uses of “good”, as they might be called.

6 Moore wrote: “Can we imagine ‘good’ as existing by itself in time, and not merely as a property of some natural object? For myself, I cannot so imagine it, whereas with the greater number of properties of objects—those which I call the natural properties—their existence does seem to me to be independent of the exis-tence of those objects. They are, in fact, rather parts of which the object is made up than mere predicates which attach to it. If they were all taken away, no object would be left, not even a bare substance: for they are in themselves substantial and give to the object all the substance that it has. But this is not so with good” (Principia, 93).

I think it possible that what lay behind, and issued in, these dark words was Moore’s having intuitively felt that “good” is an attributive adjective.

7 W. D. Ross, The Right and the Good (Oxford: Oxford University Press, 1930),

8 I thank Matthew Hanser for “famous”.

9 Shelly Kagan says: “on one level, to say that there is a pro tanto reason to pro-mote the good is actually to make a trivial claim. Everyone has a standing reason to promote the objectively best outcome, because the existence of such a reason is in part just what it is for something to have objective value.” The Limits of Morality (Oxford: Oxford University Press, 1989), 61. If we ignore that unexplained “in part,” we can take him to think that for (2) to be true is for A to be prima facie (he calls it “pro tanto”) ought-making. (See also his pp. 65–80.)

10 I am sure that some readers will say that there is another, and better, alterna-tive at the bottom of the barrel, namely that the property our philosopher takes “good” to stand for is (iv) intrinsic goodness. But what is that? We might well sup-pose it to be nonderivative goodness: that is, the goodness a thing has, but not because it stands in a suitable relation to (as, for example, being conducive to) some-thing else that is good. This does seem to be what Moore had in mind when he used the term in Principia. So understood, however, there is no such property as intrin-sic goodness if there is no such property as goodness.

In a later work, Moore pointed to the possibility of an account of intrinsic good-ness according to which it consists in oughtmakingness, though he himself rejected it. See his “Reply to My Critics,” The Philosophy of G. E. Moore, ed. P. A. Schilpp, 3rd ed. (La Salle: Open Court, 1968).

A recent oughtmakingness account of intrinsic goodness may be found in Michael J. Zimmerman, The Nature of Intrinsic Value (Oxford: Rowman & Littlefield, 2001): for A, B, C, and so on, to possess intrinsic goodness is for it to be the case that “there is a moral requirement to favor them (welcome them, admire them, take satisfaction in them, and so on) for their own sakes” (24, Zimmerman’s own italics). (That “for their own sakes” is presumably intended to ensure that intrinsic goodness be nonderivatively ought-making.) This account of intrinsic goodness should remind us of alternative (i) in the text above. And if offered as an account of what the philosophers I discuss in the text above have in mind, it would be subject to the same objection, namely that what they have in mind by “good” is not just, simply, being ought-making, but rather the property that makes its possessors be ought-making.