3) The enigma of chapter 6

Near the end of the book, an enigmatic text passage draws our attention. Moore defines good as an objective predicate of value, then as an unanalyzable predicate of value: It appears, at first sight, to be a strange coincidence, that there should be two different objective predicates of value, good and beautiful, which are nevertheless so related to one another that whatever is beautiful is also good.
But, if our definition be correct, the strangeness disappears; since it leaves only one unanalyzable predicate of value, namely good, while beautiful, though not identical with, is to be defined by reference to this, being thus, at the same time, different from and necessarily connected with it ¹.

In this passage and for the first time, Moore explicitly focuses on the problem of the relation existing between the concepts of value and good. He simply says that one of these concepts is a predicate of the other but provides no further help. For being a predicate of something means being related to it by the verb “to be”. However, Moore does not clarify the nature of this connection. This relation can be analytic or synthetic. For example, in a tautology like “a cat is a cat”, the predicate “cat” is related to the subject “cat”.

In this case, Moore says nothing more than that there is a certain relation between value and good. Which type of relation? Are they synonyms? Kinds and species? Etc. Moore does not provide any clarification. However, for the first time, Moore explicitly raises the problem of the relations existing between the concepts of good and value. Nevertheless, we cannot be satisfied with this non-response.

4) A clarification in the preface of the second edition

The preface of the second edition is particularly interesting, because a few years after the publication of his book, Moore provided a critical review, and his remarkable intellectual honesty lead him to reject some of his own ideas.

In this case, he defines for the first time the relation existing between value and good: they are kinds and species: It may, I think, also be expressed by saying that G is an intrinsic kind of value. When people talk of a thing's possessing 'intrinsic value', part of what they mean is, I think, always that it possesses a kind of value which has this property ².

This means that value is the larger class, which includes good, such as a kind includes a species. This comes down to saying that good is for value what a Granny Smith is for an apple or what a waltz is for music.

Moore insists: G and some other predicates [...] share with it the peculiarity that to ascribe them to things is to ascribe a kind of value to these things [...]. In saying this of G, I am, therefore, I think, ascribing to it a very peculiar position among predicates ³.

What should we think about this idea?

We can see that value appears in full light and that Moore begins to examine it. However, his analysis stops there and this does not tell us what value is in itself, beyond its relationship with the concept of good. It is really extraordinary that Moore devotes hundreds of pages to the definition of good (to conclude that there is none) and not a single line to value. We know that good is a kind of value, but this gets us nowhere as we do not know what type of value Moore is referring to.

Above all, it is unclear how Moore can maintain that good is a simple notion, which contains, unlike complex notions, no other notion. For if good is a kind of value, it must at least contain the notion in question (value). We will continue this analysis with our example: What is a Granny Smith, if it does not itself contain the notion of apple? And what is a waltz if it has nothing to do with music?

Therefore, the main thesis developed by Moore, which is the subject of chapter 1 in Principia Ethica, is invalidated. Does he know this? I cannot say. However, in any case, he provides an argument which could be an interesting answer to our objection.

Finally, he concedes that there are synonyms for good, which can be used as substitutes. He even explicitly affirms that the concept of value is one of these synonyms (this increased awareness is a clear progress). However, he does not change his mind: we shall never be able to replace good by another concept containing an analysis thereof. Why? He makes a distinction between “to express the meaning of a term by other terms” and “to express the meaning of a term by other terms which include an analysis thereof”.

Only the first case is possible for good as in Moore’s opinion: It may possibly be true that G is unanalyzable, and therefore cannot be expressed by other words, which contain an analysis of it. But it is certainly not true that it cannot be expressed by other words, at all. [...] In English, we [use] other words or phrases as synonyms for 'good', in this sense. It is obvious, for instance, that the word 'desirable' is sometimes so used; and so also is the phrase 'intrinsically valuable', which I myself use for it later. It is therefore simply false that G is different from any predicate which we express by words or phrases other than 'good' ⁴.

1. Ibid., chap. 6, §121
2. Ibid., Preface to the second edition, 4
3. Ibid.
4. Ibid., 3