خوب به یک اعتبار

The Journal of Ethical Reflections

Vol.2, Issue 3, No. 7, Autumn, 2021, pp. 7-19

GOOD IN A RESPECT

Jonathan Dancy [1]

Professor of Philosophy, University of Texas, Austin, USA.

jdancy@austin.utexas.edu

 Abstract

In this paper I consider all the ways known to me of trying to make sense of the idea of something’s being good in a respect, though perhaps not good overall. My conclusion is that none of them is at all successful, so that the idea of something’s being good in a respect remains a mystery.

Keywords: Good, Respect, Ross, Pro tanto good.

Introduction                 

To my mind there are at least three unsolved mysteries in the theory of the good. One of these mysteries is unsolved because it has hardly been noticed; that is the one that this paper is about. The other two are the notion of a benefit and the notion of welfare. There are at least some discussions of these notions. As far as benefit goes, it seems clear that when someone is benefited something is made better or improved, but it is not at all clear what that thing is. Promising answers include: (1) that person’s life and (2) that person’s situation. There are difficulties with both these answers, though they are not my present concern. And if there are difficulties with both of them, there are probably difficulties with similar approaches to the notion of welfare.

This short paper, however, is about the third mystery.

1

What is it for an object to be good in a respect? In addressing this question, one might be tempted to appeal to existing accounts of what it is for an action to be right in a respect. And fortunately there is a going theory of that, since it seems to be exactly what Ross was getting at with his idea of ‘prima facie duty’. Ewing said of this idea that it was ‘one of the most important discoveries of the century in moral philosophy’ (1959: 126)[2]. An action that is a prima facie duty is right in a certain respect even if it is not in fact right overall – that is, not our ‘duty proper’, as Ross would put it. Let us, perhaps unwarily, suppose that we understand such a remark. One could then imagine trying to say similar things in the theory of value. And sure enough, Pekka Vayrynen has tried this line. He writes:

To say of something that it is pro tanto good is to say that it is good as far as some particular characteristic or respect is concerned and that it is genuinely good in that respect even if it is sufficiently bad in other respects as to make it bad all things considered (Vayrynen, 2013, pp. 28-9).

I imagine that this notion of the pro tanto is the same as Ross’s notion of the prima facie. (This is because of the general agreement among theorists that the term ‘prima facie’ is not very helpful, since the idea we are after is not the idea of first appearances – in fact it is nothing to do with appearances at all.) So let us try the same approach for pro tanto duty. The problem is that what is pro tanto good is genuinely good in that respect while what is a pro tanto duty is not a genuine duty of any sort (yet). Ross knew this perfectly well.  But he did make some use of the notion of the prima facie in the theory of value. He wrote: ‘There is no self-evident connexion between the attributes ‘right’ and ‘optimific’. The theory we are examining [utilitarianism, which says that there is such a connexion - JD] has a certain attractiveness when applied to our decision that a particular act is our duty … But it is not even plausible when applied to our recognition of prima facie duty. For if it were self-evident that the right coincides with the optimific, it should be self-evident that what is prima facie right is prima facie optimific. But whereas we are certain that keeping a promise is prima facie right, we are not certain that it is prima facie optimific (though we are perhaps certain that it is prima facie bonific’ (Ross, 1930, p. 36).

What would it be, then, for keeping a promise to be prima facie bonific? Is that any different from being actually bonific to a certain extent? This looks like a puzzle in Ross scholarship. But we should notice already that ‘prima facie bonific’ is not quite the same as ‘prima facie good’, since something can be intrinsically good without being bonific; or at least we might suppose that to be a possibility, since ‘bonific’ seems to mean ‘having good results/consequences’ as opposed to being intrinsically good. Taking it in that light, we could consider other passages where Ross makes use of a notion of prima facie value. Here is one:

But I am inclined to think that there is involved the further thought that a state of affairs in virtue of being painful is prima facie (i.e. where other circumstances do not enter into the case) one that a rational spectator would not approve, i.e. is bad; and that similarly our attitude towards kindness involves the attitude that pleasure is good (Ross, 1930, p. 135).

Ross’s gloss on ‘prima facie’ here is peculiar. One would have thought that he would have given an analogue of his usual isolation test for the prima facie: an action is a prima facie duty in virtue of having a certain property if that property would make it a duty proper in any case where the action has no other such property. The reason why he does things this way is no doubt to do with the notion of approval. If the good is that of which a rational spectator would approve, the prima facie good is that of which the rational spectator would approve in certain circumstances. This is because there is no notion of prima facie approval – even though there is (we may optimistically assume) a notion of thinking something prima facie good.

Here is another passage from Ross:

Pleasure seems, indeed, to have a property analogous to that which we have previously recognized under the name of conditional or prima facie rightness. An act of promise-keeping has the property, not necessarily of being right but of being something that is right if the act has no other morally significant characteristic (such as that of causing pain to another person). And similarly a state of pleasure has the property, not necessarily of being good, but of being something that is good if the state has no other characteristic that prevents it from being good (Ross, 1930, p. 138).

The main difficulty in making sense of these remarks is that a prima facie duty is not merely something that would be a duty in certain circumstances. Prima facie duty is a sort of normative force which succeeds in making genuine duty (duty proper) in all cases where it is not overwhelmed or even equalled by a similar force set against it. This can happen either where the action itself is also prima facie wrong, and is more prima facie wrong than it is prima facie right, or in cases where some other action, alternative to this one, is more prima facie right than this one is.

Can we make any sense of the idea of a similar arrangement in the theory of value? One problem is that actual rightness is not identical with the property of being most prima facie right. In alternative language: a duty proper is the action that has most of a distinct property, that of being a prima facie duty. But what we are summing in order to determine which action is best is not which action has most of a distinct property which we are calling prima facie goodness, but which action has most actual goodness.

My conclusion so far is that it is not going to be easy to apply Ross’s technical notion of the prima facie in the theory of value. Something that is good in a respect has a definite value already, whereas something that is a duty in a certain respect may not be a duty at all.

But perhaps in all this we have failed to profit from the fact that Ross offers us two accounts of prima facie duty. The first, in terms of which I have been working so far, involves a sort of isolation test. An action is a prima facie duty in virtue of having a certain feature if, were it to have no other such feature, it would be a duty proper. I have often suggested in print that such an isolation test is very peculiar. But there is a second account of the prima facie which is not subject to this complaint. This second account is run in terms of tendencies. An action is a prima facie duty in virtue of having a certain feature iff actions with that feature tend to be duties proper.

On these lines, Ross writes: “We have to distinguish from the characteristic of being our duty that of tending to be our duty. Any act that we do contains various elements in virtue of which it falls under various categories. In virtue of being the breaking of a promise, for instance, it tends to be wrong; in virtue of being an instance of relieving distress it tends to be right. Tendency to be one’s duty may be called a parti-resultant attribute, i.e. one which belongs to an act in virtue of some one component in its nature. Being one’s duty is a toti-resultant attribute, one which belongs to an act in virtue of its whole nature and of nothing less than this. This distinction between parti-resultant and toti-resultant attributes is one which we shall meet in another context also. [In fact, in a later chapter on the nature of goodness: JD]

Can we make better sense of this suggestion? It seems to amount to the idea that for an action to be good in a certain respect, which Ross is calling ‘being prima facie good’, is to have a feature that tends to make the action good overall. But I wonder whether this makes any sense at all. There seem to be two possible ways of running the idea. The first is in terms of frequency: an action is good in a certain respect if it has a feature which usually or normally makes it good overall. But this is obviously hopeless. The second is in terms of a sort of value-force: for an action to be good in a respect is for it to have a feature which pushes it towards being good overall.

In trying to assess this second suggestion we need to distinguish between two thoughts. The first is that if an action is good in a certain respect it does have a feature which pushes it towards being good overall. The second is that if an action is good in a certain respect, it’s being good in that respect consists in a tendency to be good overall. And the first of these is much more plausible than the second. We might agree that if an action is good in a certain respect, it is overall better than it would have been without that feature – though we should be wary even of this, because it might be that, though the action is good in respect of being F, it might have been even better for being not-F (a sort of win-win situation). For instance, it might be good to be alone and even better to be accompanied. So even the more plausible version is probably false. The less plausible version cannot be right, because it seems impossible to understand being good in a respect as a tendency to be good overall. What if there were two features which necessarily go together, but such that though one is good (or good-making) the other is bad (or bad-making) and worse than the first one is good. It seems very difficult to understand the good-making feature as a tendency to be good overall in such a case.

In a way the main source of difficulty here lies in Ross’s attempt to take certain aspects of his theory of the right – his moral theory – and apply them in the theory of the good. An act that is a prima facie duty is an act that has a feature in virtue of which it would be a duty proper if it had no other such features. But value doesn’t work that way (and maybe duty doesn’t either). There is no need to understand value as a contribution to something called ‘good overall’; so the notion of contributory value looks likely itself to be a mistake. Of course we want to allow that something that is good in a respect is the better for it; but better than what? Better than it would have been without it – not necessarily, alas. For, as I have already suggested, it might be good to be alone and even better to be accompanied.

Another serious issue concerns the fact that the contributory is itself a matter of degree. (As we might put it, some reasons are stronger than others.) Talk in terms of tendencies seems well suited to capture this aspect, since some tendencies are stronger than others. But accounts in terms of subjunctive conditionals are less well equipped. One cannot have more of ‘being an act that would be a duty if it had no other relevant feature’; such a thing does not come in degrees.

It is always worth keeping an eye on Prichard in connection with such issues. In the theory of the right, he came to the conclusion that all we need is a contributory notion of being in some respect a duty. The thought here is that this would retain the normativity of the contribution and allow for degrees. But one should, I suggest, be wary of this latter suggestion, since there is a distinction between being in more respects a duty and having more of ‘being in some respect a duty’.

Returning to Ross: he makes a further and quite different suggestion which we have not so far considered: ‘Some may think, no doubt, that the mere thought that a state of affairs would be painful for another person is enough to account for our conviction that the desire to produce it is bad. But I am inclined to think that there is involved the further thought that a state of affairs in virtue of being painful is prima facie (i.e. where other circumstances do not enter into the case) one that a rational spectator would not approve, i.e. is bad; and that similarly our attitude towards kindness involves the thought that pleasure is good’ (Ross, 1930, p. 135). Again, can one have more or less of this? Does it even involve the notion of prima facie value? Perhaps it only involves the notion of ‘being prima facie one that a rational spectator would not approve’. But is this last a matter of degree? Approval might be a matter of degree - indeed it probably is, since I seem to be able to approve of two things while approving of one more than of the other. But it does not follow from this that something can have more or less of ‘being such that a rational spectator would approve of it’. But the question seems rather to be whether approval is itself a deontic or an evaluative stance. If to approve is to think good, we are being told that to be good is to be what a rational spectator would think good – which is no advance. If to approve is to think right, Ross has introduced what I would think of as a very improbable direct link between the deontic and the evaluative.

Finally, in our consideration of Ross, I should mention another remark of his:

Another respect in which value differs from mathematical properties is that while mathematical (i.e. spatial, temporal, and numerical) properties follow from part of the intrinsic nature of their possessors, value follows from the whole intrinsic nature of its possessors. … Value is a toti-resultant property, based on the whole nature of its possessors. And this is true not only of ‘good’, but also of ‘right’ and ‘beautiful’ (Ross, 1930, p. 122).

My view about this contrast is that it looks like an exaggeration, at least as far as beauty is concerned. Is it true that the beauty of an object results from (or ‘follows from’) the whole intrinsic nature of the object? Every painting has a back side whose nature is irrelevant to the beauty of the painting. Perhaps the beauty belongs to the front side only, not to the painting conceived as a 3-dimensional object. And we might think that the beauty of a piece of music results only from the intrinsic properties of the music. But what about the beauty of a performance of that piece of music? Suppose, for instance, the beauty partly derives from the fact that the string players used no vibrato. Is this an intrinsic property of the performance? As for the value of an action, presumably the identity of the agent is an intrinsic feature, but it doesn’t seem that the value of the action is always partly dependent on who did it.

As a final historical comment, I should mention Broad’s later distinction between what he called ‘right-inclining’ and ‘good-inclining’ characteristics.[3] It depends, of course, on how such a distinction is drawn, but the suggestion that there is something called ‘inclining’ which operates in the same sort of way in the deontic and the evaluative realms is something we should be wary of. The basic problem is that while an action (or any other object) can be good in some respects and bad in others, it cannot in the same way be right in some respects and wrong in others. There can perfectly well be reasons to do it and reasons not to do it, but an action is either right or not right; it cannot be partly right and partly wrong. In this sense ‘right’ and ‘wrong’ are verdictive concepts (to use Philippa Foot’s technical term) while ‘good’ and ‘bad’ are not.

2

So what sense can we make of the notion of something’s being good in some respects and bad in others?

If the buck-passing view of the evaluative were sound, there would be no more difficulty about ‘good in a respect than there is about ‘right in a respect’. So long as we start from the idea that both rightness and wrongness and goodness and badness are to be understood in terms of reasons (to put the matter rather vaguely), we seem well equipped to capture the relevant distinctions.

Now my own view is that the buck-passing conception of value is in the end unsustainable. But since I am hardly going to argue the matter here, let us just take our main question to be what sense we can make of the notion of something’s being good in some respects and bad in others, if we do not appeal to the metaphysical simplifications of the buck-passing view.

In addressing that question, we confront two difficulties. The first difficulty will be the distinction between a ground and an enabler.  A ground is a feature that plays a certain distinctive role: as we say, it makes the action wrong. An enabler is a feature in whose absence something that in fact goes to make the action wrong is prevented from doing so. So a feature whose presence is necessary for (enables) the presence of some value may not in fact be (part of) the ground for that value. This means, as we will see, that appeal to certain subjunctive conditionals is not going to cut much ice.

The second difficulty derives from the notion of an attenuator. An attenuator is a feature in whose absence the whole would be more valuable, though the explanation of this is not that it contributes disvalue itself; as an attenuator, its absence would not itself contribute value, but would allow some other feature to contribute value (something of which otherwise it would be incapable) or to increase the value it already contributes. (This is of course a matter of degree.)

The effects of these two simple points will emerge as we go along. So how are we to understand the notion of something’s being good in a respect? Let us take this question to be the question ‘what is it for an object X to be good in respect R?’.

Obviously to be good in respect R is not the same as being good and having R. we need more than that. In what follows I consider 16 possible suggestions.

1. The model of the curate’s egg, which was good in parts, is clearly not the one we are after. A book which is good in some respects and not in others is not necessarily one which has good parts and bad parts. It may be good in some respects all the way through, and bad in other respects all the way through.
2. ‘X is good in respect R iff X has R and having R is good.’ Here I only want to say that even though having R may be good, X may have R and not be good in that respect. X may have R and be the worse for it.
3. ‘X has R and having R is good in this case’. But we don’t want having R to be good, we want the thing that has R to be good, or at least the better for it. Being good in a respect is not the same as being in a good respect.
4. ‘X has R and to that extent/in that respect is good’. This is Vayrynen’s version, which I have already discussed.
5. ‘Insofar as X has R, X is good’.[4] This phrase seems to be practically unintelligible.
6. ‘To the extent that X has R, X is good’. If this is better than its predecessor, it might mean that if X has R to a considerable extent, X is considerably good. But even though X may be good in respect R, it may not be the better the more R it is; it may not be improved by becoming more R.
7. ‘X has R and is better than it would be without R’. This may perhaps mean the same as ‘X is improved by having R’. This suggestion runs up against the distinction between an enabler and a ground.
8. ‘X. has R and is the better for it’. This suggestion too fails to recognise the distinction between enabler and ground.
9. ‘X has R and is better than it would be without R’ – which may perhaps be the same as ‘X is improved by having R’. This suggestion fails to accommodate the distinction between ground and intensifier.
10. ‘X has R and is made better by having R than it would be if it did not have R’. This suggestion doesn’t work because having R may itself enable something else to make X good without being a good-maker itself.
11. ‘That X has R is part of the explanation of X’s goodness’. This suggestion fails to distinguish between the absence of an attenuator, an enabler and a contributor.
12. ‘Given only that X has R, X is good’. Here I just want to say that this does not make any sense at all.
13. ’If X has R, X is so far forth good’. I found this suggestion in a paper I was refereeing for the journal Ethics. It is plainly meaningless. The expression ‘X has R and is good as far as that goes’ is not much better.
14. ‘X has R and we should approve of it in that respect.’ This I think does not improve on suggestions already dismissed. There is perhaps a distinction between approving of X in respect R and approving of X’s being R; if there is, it is the former we are after. But what is the sense of ‘approve in a respect’ if it is not ‘approve of a respect’?
15. ‘If all you knew about X was that it has R, you should approve of X’. This runs up against the distinction between approving overall and approving in a respect. Further, it is an attempt to use an epistemological concept to make sense of a metaphysical one, which I take to be a distraction.
16. ‘Having R is the ground of some goodness in X’. This I think is the notion we are trying to understand, and it does not itself provide that understanding.

Final Remark

The failure of all these attempts causes me to wonder whether we might not need three distinct value-concepts: good, benefit (good for) and improve (make good), where no one of these can be explained in terms of the others. But even if we allowed that, we would still not be in a position to explain what it is to be good in a respect. Perhaps we should just accept the situation. In doing so, we might be encouraged by the similar expression ‘similar in a respect’; but I think we should not be. It is easy to say what it is for two objects to be similar in a respect: it is for them both to have a same property. Goodness in a respect is a much more complex notion. I agree with Broad’s remark that ‘We may agree that when ‘good’ is used in the sense of ‘benefic’ or of ‘contributively good’, it stands for a characteristic which is complex.’[5] But this alone does not tell us what the complex characteristic is.

[2]. In fairness, it should be said that C. D. Broad introduced effectively the same notion in the final pages of his Five Types of Ethical Theory, though with much less detail; Broad used the term pro tanto in preference to Ross’s prima facie, a preference which recent thinkers have all endorsed. See Broad (1930), 282 and also 222 (for talk of tendencies).

[3]. This is from his Lectures on Ethics.

[4]. I found this in D. Davidson (2000), p. 21.

[5]. C. D. Broad, ‘Is Goodness the Name of a Simple Non-natural Quality?’ in D. Cheney (2013), p. 100.