This work offers an important and critical analysis of Moore's conception of good and right. Preface G.E. Moore's ethical theory is usually remembered and studied these days because of Moore's denunciation of ethical naturalism on the ground that it involves a fallacy, 'the naturalistic fallacy,' which, according to Moore, is inherent in all positions of this kind. But Michela Marzano invites us to turn away from this over-familiar theme and attend to a different element of Moore's theory, his theory of intrinsic value. By doing so, she shows, we gain a much better perspective on Moore's ethics and its relationship to his metaphysics. We can put his criticisms of ethical naturalism into the broader context which is provided by his account of the relationship between 'intrinsic nature' and 'intrinsic value'. But the main gain is that we obtain a perspective from which we can appreciate his positive theory of value. I In ordinary thought we contrast the terms 'intrinsic' and 'instrumental'. There are things such as friendship and beauty which we value for their own sake; these are things which we judge to have 'intrinsic' value.
And there are other things such as wealth and fame which we value because of the value of the things which they enable us to obtain; these are things we judge to have 'instrumental' value. These classifications are not, of course, exclusive: whereas wealth has instrumental value only, good food has both intrinsic and instrumental value; indeed anything that has intrinsic value also has instrumental value. Furthermore, although, as I have indicated, instrumental value contrasts in principle with intrinsic value, it is not an altogether new kind of value; for the value of the things which something with instrumental value enables us to obtain must be ultimately definable in terms of intrinsic value. So, as far as this contrast goes, intrinsic value is the fundamental type of value and instrumental value is definable in terms of it. But for Moore intrinsic value is not merely non-instrumental value. For he holds that as well as considering consequences we also need to consider context, in that the context in which something occurs is liable to affect its contribution to the overall value of the broader situation which provides the context.
This line of thought is less familiar to us than that which underpins the concept of instrumental value but he illustrates his point by inviting us to consider the value of knowledge. Moore suggests that knowledge has little intrinsic value when we consider it just by itself; but he also suggests that the value of the appreciation of beauty (something which is, according to Moore, potentially of great value) is greatly enhanced by knowledge about the object one is appreciating (such as knowledge concerning the artist's intentions and the broader culture to which the artwork belongs). So he holds that knowledge of this kind can have 'value as a part', as something which in a particular case makes a difference to the overall value of the situation in which it occurs, which is distinct from the contribution it makes to this overall value merely in virtue of the intrinsic value which it contributes to the value of the situation. This aspect of Moore's position is of course linked to his 'principle of organic unities', his claim that the overall intrinsic value of a complex situation is not simply the overall sum of the intrinsic values of each 'part' (or aspect) of this situation.
Moore's thought is that the way in which things are inter-related in a complex situation itself makes a difference to the value of the resulting situation without altering the intrinsic value of each part. So, in the case above, the contribution knowledge about the artist and artwork makes to the value of informed aesthetic appreciation greatly exceeds its limited intrinsic value. As Marzano acknowledges Moore's principle of organic unities is not uncontentious; it obstructs the straightforward consequentialist reasoning which his ethics seems at first to commend as a way of determining one's duty. But it is its implications for his conception of intrinsic value that concerns me here. For what the contrast between value as a part and intrinsic value entails is that intrinsic value is precisely value not as a part. Indeed because something's intrinsic value is conceived by Moore to be independent of any context in which it occurs, it can be thought of as its value when it occurs in the 'null' context, i.e. altogether by itself (insofar as this makes sense).
This may make it sound as though it is only relatively simple things ('simple parts') that have intrinsic value; but this is not so. For situations of any degree of complexity, such as the aesthetic appreciation of a work of art or a benevolent action undertaken to help someone in distress, will themselves have an intrinsic value which is to be taken into account in determining how one ought to act. Equally, such situations will also have some degree of instrumental value and may also have value as a part insofar as they occur within even broader contexts. So the distinction between intrinsic value, instrumental value and value as a part applies across the board. Nonetheless it remains the case that intrinsic value is, from a theoretical perspective, the fundamental type. For just as instrumental value is defined in terms of the intrinsic value of the consequences of some state of affairs, the 'value as a part' of a state of affairs is defined in terms of the difference the existence of this state makes to the intrinsic value of a complex situation of which it is a part.
The combined effect of the contrasts with instrumental value and value as a part show that in assessing something's intrinsic value we have to consider it without reference to any potential consequences or any broader context in which it might occur. This is of course Moore's method of 'reflective isolation' whereby we are enjoined to form an intuitive judgment of the (intrinsic) value of a state of affairs by reflectively isolating it and trying to assess the value of a world wholly comprised of states of this kind. This may well not strike us, on further reflection, as a sensible approach to ethical judgment; it is not clear, for example, how one is to apply it to judgments concerning the value of knowledge. The obvious alternative would appear to be to consider what difference the presence or absence of a state of the relevant kind makes to the intrinsic value of a situation that remains the same in other respects - thus, for example, what difference the acquisition of knowledge makes to the intrinsic value of one's general situation.
But Moore's principle of organic unities makes this approach problematic: for the principle implies that there is no reliable inference from the difference in value which the presence or absence of knowledge makes to its intrinsic value. To suppose otherwise is precisely to conflate its value as a part with its intrinsic value. This point suggests that although the principle of organic unities captures an important insight concerning the context-dependence of value, it may be that there is a better way of capturing this insight, and I shall briefly return to this below. But, staying within Moore's approach for the moment, we can see, I think, why Moore holds that judgments of intrinsic value are 'universal' (i.e. necessary) judgments concerning the value of things of some general kind (e.g. pains, knowledge etc.). For once one has stripped away the consequences of a state of affairs and the broader context in which it occurs, one is simply left with the fact that it is a state of affairs of some kind (e.g. a pain), and the value such a state possesses must apply equally to all states of the same kind; for there is nothing left to provide a basis for different evaluations of them.
This last line of thought has, however, an important presupposition - that something's value is dependent upon the kind of thing it is. Without this presupposition, it could not be ruled out that, say, exactly similar pains should have different degrees of value even though there would be no basis for this difference in value; indeed there would be no reason for demanding that there be such a basis. The fact that this conclusion strikes us as irrational shows our attachment to the presupposition in question, which Moore expressed as the thesis that something's intrinsic value depends only upon its 'intrinsic nature'. As Marzano shows, Moore's conception of 'intrinsic nature' is not straightforward: it will be clear from the preceding discussion why it needs to exclude consequences and context, and thus all extrinsic factors, since these are relevant to instrumental value and to value as a part but not to intrinsic value. Marzano further suggests that intrinsic nature approximates to the traditional conception of essence, though since Moore says that size (which is not an essential property) is an intrinsic property it appears to me that intrinsic nature embraces more than essence.
But the details here are not, I think, important; what matters more are the connections, or lack of them, between Moore's conception of intrinsic value and his critique of ethical naturalism. Before discussing this further, however, I want to return briefly to the question that came up earlier as to whether Moore's principle of organic unities, and the associated conception of 'value as a part', provides the right way to handle the phenomenon of the context-dependence of value. A good case to focus on here is that of friendship. As Moore rightly says, friendship is one of the most valuable aspects of human life. And yet there are contexts in which one has to set aside the claims of friendship - for example in courts of law or when making an appointment to a post. For Moore this phenomenon is to be handled by supposing that in these contexts, friendship has a negative value as a part which overrides its great positive intrinsic value.
But that seems the wrong way to think about the matter: rather, one wants to say, in a legal context the claims of friendship have no place at all; they are simply not a legitimate consideration and are instead to be placed behind the blindfold of ignorance that is placed around the eyes of justice. Hence even though friendship does have great value, it is wrong to think of this as an absolutely universal value which it brings to all contexts. Instead it is a value which can be set aside in certain contexts, just as when one is assessing the epistemic value of evidence, eye-witness reports which would normally be strong evidence have to be set aside if there is reason to doubt the reliability of their source. Just how this context-dependence of ethical value is then to be conceptualised and applied is a complicated matter not to be settled here. But if I am right about this, it follows that one important aspect of Moore's conception of intrinsic value, namely its unqualified universality, needs qualification in a way which is likely to affect the structure of his theory.
II As Marzano notes, one of Moore's reasons for rejecting ethical naturalism is that he takes it that the intrinsic value of a state of affairs, unlike its natural properties, depends on the totality of its intrinsic nature. If goodness were a property like yellowness, then it could vary without reference to other properties. As Moore himself noted later, there are complications here, in that there are essentially derivative natural properties such as being coloured. But setting that issue to one side for the moment, one should ask why it is that value is a dependent, or derivative, property in the way that Moore maintains; for, on the face of it, there is some tension between this thesis and the conception of intrinsic value as the value of a state considered in complete isolation. The greater the isolation of intrinsic value, the less basis there would seem to be for insisting that it is derivative. Moore (and Marzano) will of course respond that the isolation simply serves to uncover intrinsic nature as the basis of intrinsic value; and that is fair enough as far as it goes - but it does not altogether explain why value is derivative in the way that Moore says it is.
One way to press this point is by considering judgments concerning 'relative' goods such as what is 'good for someone', or what someone is 'good at' or 'good as'. For in cases of this kind the dependence thesis is easy to substantiate, since the relative term in question provides criteria for evaluating the thing whose goodness is in question. For example, it is just obvious when one is judging whether someone is good at cooking that the value judgment is dependent upon their skills as a cook. The suggestion that such a judgment might not be dependent upon the relevant skills of the person being evaluated is nonsensical once the relative term has been explicitly introduced. But this line of thought cannot be carried across to Moore's absolute, non-relative, judgments of intrinsic value: precisely because value is here being considered entirely without reference to anything extrinsic to the situation being evaluated which might provide criteria for its evaluation, such as the interests of the person making the judgment, it is less clear why it is still right to insist that the value is dependent on the intrinsic natural properties of the situation being evaluated.
Having set up this challenge, however, I do not want to insist that it cannot be met. What is required is some further consideration concerning judgments of intrinsic value which explains why intrinsic value is essentially derivative and I shall make one suggestion to this effect below. But I want now to pursue a different aspect of Moore's critique of naturalism in the context of his theory of intrinsic value, namely his thesis that although intrinsic value is dependent upon intrinsic nature, it is not itself an aspect of a thing's intrinsic nature, since it is not a natural property at all. This is a difficult issue to discuss briefly, since, as Marzano well shows, Moore's treatment of this issue is permeated by his metaphysics. Nonetheless the basic structure of the dialectic here can, I think, be set out briefly. A good starting point is the distinction between the thesis that goodness is not a natural property and the thesis that it is not definable in terms of natural properties or reducible to them. One needs some argument to rule out the possibility that goodness is an irreducible natural property; but Moore, in effect, provides such an argument.
For if the previous point, that goodness is essentially dependent upon natural properties, is assumed, then goodness could not be a natural property without being reducible to the natural properties on which it depends. It may seem that there is a way of managing this, by invoking the idea of the 'supervenience' of goodness as a form of non-reductive dependence. Marzano discusses and supports this hypothesis (and I think she is right to think that Moore would agree with her); but I am afraid that I am not persuaded by her. Without going into details, it seems to me that supervenience inescapably induces a disjunctive reduction of the supervenient property (goodness in this case) to the disjunction of all the various subvenient circumstances (complex natural properties in this case) on which the supervenient property depends. Admittedly, since the disjunction is openended, the reduction serves no epistemological purpose: one cannot use it to decide questions of value. But it serves the metaphysical purpose of implying that if goodness is dependent on natural properties, then it is not an irreducible natural property.
The trick is now to reverse this argument and infer, on the same assumption, that if goodness is not reducible to natural properties, or definable in terms of them, then it is not itself a natural property. Thus this line of thought, so far from subverting Moore's aims, actually helps to support his critique of ethical naturalism. For it links the 'open question' argument for the indefinability of goodness to the apparently separate thesis that goodness is not a natural property. At this point, then, attention switches to Moore's defence of the indefinability of goodness, and thus to the 'open question' argument. Marzano shows how this argument fits into Moore's treatment of intrinsic value and its relationship to natural properties, but, standing back a bit further from the details of Moore's discussion, the issue that reappears here is that of the role of the value judgments Moore attends to. For although Moore undoubtedly captures a truth about the phenomenology of value judgment by his open question argument, more needs to be said to vindicate this phenomenology if it is to be relied upon to sustain the indefinability of intrinsic value.
One familiar approach here is to suggest that what underpins this phenomenology is the normative significance of these judgments insofar as they imply that the nature of certain things constitutes by itself a reason for us to want them, to care about them, or, generally, to promote their existence. This normative dimension of value judgment is picked up by the open question argument because describing something by saying what kind of thing it is not thereby saying or implying that its being of that kind is a reason for anything; so the indefinability of goodness, on this view, arises from the fact that value judgments concern what there is reason to think, feel or do, and such reasons are distinct from descriptive claims concerning the way things are, including descriptions of the ways in which we think, feel and act.
More needs to be said to reinforce this line of argument and its presumption that claims of reason are not reducible to descriptions of natural facts; but, returning now to the earlier point about the need for some deeper account of the value judgment to explain why goodness is essentially derivative, it looks as though the suggestion made here about the normative dimension of such judgments will also provide such an explanation. For reasons are inherently general: so, if value-judgments express practical reasons, Moore's thesis that goodness, intrinsic value, is essentially derivative will follow at once, without the need to add the kind of extra term which I suggested earlier is characteristic of judgments concerning 'relative' goodness, such as 'good at cooking'. Hence the conclusion which all this discussion points to is that the way to bring together Moore's conception of intrinsic value and his critique of naturalism is to make explicit the normative significance of judgments concerning intrinsic value. So far as I can see this suggestion does not conflict with Marzano's admirable discussion of intrinsic value.
But there is a further step which one might take, which would, I think, introduce conflict: namely the suggestion that judgments of intrinsic value just are judgments to the effect that the kinds of thing in question have an intrinsic nature which constitutes by itself a reason for thinking, feeling or doing something. This is, in effect, Scanlan's recent 'buck-passing' account of value, and it is in a way reductive.