What is Value?

An Essay in Philosophical Analysis

Everett W. Hall

• Publisher: Routledge
• Year: 2014
• Print ISBN: 9780415225397
• eBook ISBN: 9781317829607
• Topic: Philosophy
• Subtopic: Philosophy History & Theory

First published in 2000. This is Volume IV of six in the Nineteenth and Twentieth Century Anglo-American Philosophy series and focuses on value with an essay in Philosophical Analysis.

1

IN WHAT SENSE CAN VALUE BE ANALYSED?

THE present essay is meant to be an instance of philosophical analysis. Just what is meant by ‘philosophical analysis’ can best be indicated by actual examples. The author hopes the following essay furnishes such an example. But a few preliminary remarks may help to indicate the sort of problem being raised.

It is just foolish to attempt complete analysis, at least in any one essay. By this I mean that analysis must be carried on with conceptual tools that, in the given undertaking, are not themselves analysed.¹ What these tools are must be gathered by alert attention to their actual use. I leave this, in the present instance, to the reader.² But I do wish to point out at the very start that I am assuming throughout that in some sense there are values. Unless one accepted this, there would be nothing, in the area here chosen for investigation, to be analysed. Thus the present essay is based on the assumption of the objectivity of values in one very broad yet not wholly trivial sense. This sense of ‘objectivity of values’ should not be confused with a species of it: that there are values outside human experience.¹ I shall not raise at all the question of the locus of values. An analogy may help to clarify this distinction. One may hold that there are colours without committing oneself to the view that there are colours outside human experience; without raising this latter question, one may seek to answer various questions about the nature of colours, quite apart from the question where colours are and what sorts of things are coloured. I am attempting to do something like this for values.

Let me be even more explicit. There is a weak and a strong sense of the proposition, ‘there are values.’ Compare, ‘there is a property, redsquare.’ Suppose by ‘redsquare’ in this last sentence is meant a complex property exemplified by patches that are both square in shape and red in colour. Then one might reasonably say that one accepts ‘there is redsquare’ but that redsquare is not ultimate. One could even say that redsquare is analysable into two components (and a form of union). Likewise one could say that red is not so analysable into components, that it is simple. Holding this, one might say ‘there is redsquare’ is weak whereas ‘there is red’ is strong. Now I accept ‘there is value’ (for somevalue) in this strong sense. That is, there is some value which is not analysable into components, such as, for example, being pleasant and being desired. Or again, there is for me at least one value-term which is indefinable. It goes without saying that I am speaking of what has traditionally been referred to as “intrinsic value.” The phrase, ‘extrinsic value’ has, quite generally, been used to characterize something as being a cause or part-cause of something else that is valuable. Thus it is a complex notion one of whose elements is precisely value in the non-complex form which is the subject of the present inquiry.

This, however, does not commit me to the position that value is a quality, or that there is a value-quality. Beyond the simplicity of some value, it is still an open question what such a simple value is—a question for analysis. But did I not just say that value is unanalysable? True, but ‘analysis’ is used in different senses. One cannot analyse something simple by pointing out component elements, for it has no component elements. But one can characterize it. One can, for example, say that purple is simple, that it is a quality, not a relation, that it is more like red than like green. And this sort of characterization can be called, without too much conflict with common usage, “an analysis of purple.” Such a view raises many serious problems for analysis—do qualities have properties? etc.—without ever going outside the assumption that purple is simple in the sense of not being composed of elements or constituent parts (such as red and blue).

I propose to give an analysis of value in a sense of ‘analysis’ closer to “‘yellow is a colour not an odour’ or ‘yellow is a quality not a relation’ are analyses of yellow” than to “‘redsquare is simply the property of being red in colour and square in shape’ is an analysis of redsquare.” In asking, What sort of thing is value?, I shall to some degree have to ask, What is a property? What sorts of properties are there? What sorts of things can properties characterize? If there is something not a property which can in some sense be displayed or exhibited by other things, what is it and how can it be described or referred to? But in the main these questions will be left to the reader to deal with as best he may, for they divert me from my analysis to a consideration of the tools with which it is carried on, and so must be reserved for other essays in analysis.

¹ The refusal to admit this bit of common sense leads to the involutions of the “analysis” of the neo-Wittgensteinians, a symptom of which is the inability to stop the analysis once the initial inhibition to undertaking it, which is usually great, is overcome. A case in point is John Wisdom’s “Other Minds,” Mind, vols. xlix and 1, 1940 and 1941.

² It will be noted that I use a mixture of linguistic and metaphysical terms, shifting to suit the context. I think I could have kept consistently to linguistic terminology, avoiding the many pitfalls awaiting the user of what Carnap called “the material mode of speech.” The cost however would have been not only an in sufferably affected style, but worse, the illusion that language constituted the subject-matter of analysis, not merely, as is the case, the tool thereof. Moreover, I shall shift back and forth, no doubt disconcertingly, between correct and incorrect linguistic formulations. This is partly the result of indolence—actual on my part and assumed on the reader’s. But it also seems to be involved in the very nature of the case that we must set up our problem and state criteria for its solution in incorrect, common-sensical language. Then we must come to see this incorrectness and try to attain a more nearly correct formulation. I have a duty, in this connection, of explaining my use of quotation marks. On the one hand, they are used in accordance with the ordinary rules of punctuation for direct quotation, although with a slight extension. Not only are actual quotations put within pairs of double inverted commas, but also expressions that are used in a way which is peculiar to or indicative of a specific point of view. Along with this, however, is a technical semantical usage. When a linguistic expression is “mentioned,” i.e., when it is used to refer to itself, I put it within a pair of single inverted commas. This gets a little complicated when I am mentioning an expression which itself contains the mention of an expression. Here the use of single quotation marks is continued for each such occurrence (as contrasted with an alternation of single and double marks as in ordinary quotation). By adhering strictly to this programme I think some confusions may be avoided; moreover, the reader is warned whenever there is a shift from the “formal” to the “material mode of speech” or vice versa (to recur to Carnap’s terminology). By this punctuational commitment, however, I am forced to use expressions that may seem improper. For example, speaking linguistically I can use the expression, ‘‘Red’ is a predicate,’ but to put the same thought objectively I must use the sentence, ‘Red is a property’; i.e., I am prohibited from using such expressions as, ‘‘Red’ is a property’ or ‘“Red” is a property.’

¹ James Ward Smith, in “Senses of Subjectivism in Value Theory” (Journal of Philosophy, vol. xlv, 1948), distinguishes three senses of ‘subjectivism,’ the first two of which are the contradictories of the senses of ‘objectivism’ I differentiate in the text. This distinction likewise correlates roughly with A. C. Ewing’s first and third meanings of ‘objectivity of ethical judgments’ on the first two pages of his book, The Definition of Good. Although the objectivism of this essay does not involve the assertion that there are values outside human experience, it does involve a double commitment. This may perhaps be best indicated by stating that it is opposed to both of two forms of naturalism, viz., the naturalism which would refuse to allow to value-expressions any extra-linguistic reference, and the naturalism which requires that the extra-linguistic reference of value-expressions be limited to matters of fact or sensibly observable properties. It is opposed, that is, both to naturalisms that deny outright that there are, strictly speaking, value-expressions and to those that make all value-expressions redundant, i.e., properly replaceable by descriptive expressions. There is value and it is unique.

2

IS VALUE A FIRST-ORDER PROPERTY?

BY ‘first-order property’ I mean a property of particulars or individuals, thus not a property of properties. By ‘property’ I mean any entity that is capable of being exemplified by something, a test of which is that the name of such an entity can properly occur in the predicate-place in a sentence. There are, I assume, two general sorts of first-order properties, namely qualities (such as sour) and relations¹ (such as simultaneous with). A quality can be exemplified by a single particular. A relation requires for its exemplification two or more particulars each of which must perform its special exemplificational function.²

2a. Is value a quality?

It would seem that value is not a quality. For one thing, it does not appear to be a sensory quality, like bitter or magenta. This strikes one as so obvious upon any thoughtful inspection of experience that the point probably does not need arguing. However, a very elementary clarification may be inserted. The above denial does not rest on any physiological considerations, such as the absence of any appropriate physiological receptors. The position that there are sensory qualities and that among them is no value is epistemologically prior to any statement about physiological mechanisms, not in the sense that requires a phenomenalism, but in the sense that such a position is categorial rather than empirical (in the ordinary sense). Various colour-qualities have an affinity to one another, marking them off as a class from taste-qualities. So, in general, sensory qualities have an affinity to one another such that it seems appropriate to treat them as a class. And in this class is found no simple quality of value.

No one, to my knowledge, has maintained that value is a colour or an odour. But may it not be a simple quality not falling within any sense modality, a unique quality? Again, it would seem not. Speaking for myself, I do not so find it in my own experience. Nor, apparently, do others, to go by what they say.¹ Indeed, the outstanding proponent of the view that there is a simple value-quality, namely G. E. Moore, qualified the doctrine by characterizing this quality as “non-natural.” Although it is difficult to make out just what this qualification meant, it at least indicates that Moore himself was aware that he used ‘quality’ in a rather special sense when he said that good is a quality.²

Moreover, if good (positive value) be a quality, what of bad (negative value)? It would seem only reasonable to suppose that it, too, is a quality, unless one wished to hold the view that strikes all but theologically inclined neo-platonists as bizarre, namely, that badness is just absence of goodness. But if good and bad are both qualities, how are they related? Like sweet and sour or other Aristotelian contraries; that is, sufficiently like one another to be classed together, but being the most dissimilar in their class? This would require other value-qualities. And in any case we would seem to need a sort of contrariety or opposition here not furnished by positive qualities however dissimilar.

Furthermore, good, or at least some form of positive value, seems to include intentionally, i.e., as a non-asserted feature of its meaning, the element of occurrence, as is felt in such phrases as ‘ought to be’ or ‘it is good that such and such is the case,’ and contrariwise for negative value.¹ This reference to existence is part of the very nature of positive value and thus makes it quite unlike a quality.

For these reasons, I reject the view that there is a (first-order) value-quality.

2b. Is value a (first-order) relation?

The parenthetical qualification in the question above is important. It is meant to rule out, in the present stage of our analysis, a certain type of view that could be and has been characterized as a relational theory of value.¹ I refer to the view, suggested by C. D. Broad and developed by A. C. Ewing, that the basic value-predicate is ‘fitting’ or ‘appropriate.’ They themselves characterize these terms as “standing for a relation.” Sometimes this relation is described as a relation that may hold between an action and a situation.² But again, it is also described as a relation between an attitude or desire and its object.³ Now it seems to me that a “relation” of appropriateness between a desire and its object is not a first-order relation. Though a desire and its object may both be particulars and in fact be related by first-order relations (for example, a desire may occur before its object becomes actual), yet appropriateness or fittingness is not such a relation. If fittingness holds between a desire and its object, it holds between them not in that they occur as particulars, but in that the desire is directed toward that object and that that object is the object of just that desire. The fittingness is in short a property of the direction or intention of the desire. It is a semantical or quasi-semantical property. Whether value is such a property will be considered below;¹ the question is irrelevant here.

However, it might seem that when fittingness is treated as a relation between an act and its environment we really have in it a first-order relation. I think that this is not the case. We sometimes say that an act fits the situation in which it occurs. And ‘fits’ in this usage may seem to be the same sort of word as ‘fits’ in the sentence ‘This peg fits that hole.’ I doubt whether this is so, or, if it ever is so, whether it is in any such usage that ‘fittingness’ is urged as the basic value-term. ‘This peg fits that hole’ does not imply nor need it suggest that it would be a good thing, that it would be appropriate, to put this peg in that hole. But ‘fittingness,’ as used, e.g., by Ewing, does have such a value-significance as attached to the occurrence of an act that is appropriate to its situation.² This is revealed in the more usual expression, not, ‘the act fits the situation’ but ‘it is fitting that that act occur in that situation.’ To make the grammatical structure of this last expression clear, let us put it, “That that act occur in that situation is fitting.” Thus we see that ‘fitting’ is used adjectivally to modify a substantival clause. If then it refers to a property, it must be a property of what is referred to by a clause; i.e., it must be a property not of a particular or of a set of particulars, but of a (possible) fact or state of affairs. Essentially the same remarks apply when Ewing speaks as though ‘fitting’ designated a property of the occurrence of the object of a “pro-attitude.”³ That this object exists would then be that which is, or could be, fitting. If there be such a property as ‘fitting’ used in this manner would serve to designate, it is not a first-order relation and its discussion must hence be postponed.⁴

There is a view, however, that may be properly described as holding tha...

Table of contents

• Cover
• Title Page
• Copyright Page
• Table of Contents
• PREFATORY REMARKS
• ACKNOWLEDGEMENTS
• 1. IN WHAT SENSE CAN VALUE BE ANALYSED?
• 2. IS VALUE A FIRST-ORDER PROPERTY?
• 3. IS VALUE A SECOND-ORDER PROPERTY?
• 4. IS VALUE A PROPERTY OF STATES OF AFFAIRS?
• 5. IS VALUE THE REFERENT OF A SEMANTICAL PREDICATE?
• 6. WHAT IS THE PROPER SYNTAX OF VALUE-SENTENCES?
• 7. WHAT IS THE SIGNIFICANCE OF LINGUISTIC ANALYSIS FOR VALUE-THEORY?
• CONCLUDING REMARKS
• INDEX OF PROPER NAMES
• SUBJECT-MATTER INDEX