From the ALWS archives: A selection of papers from the International Wittgenstein Symposia in Kirchberg am Wechsel

What Wittgenstein says

Wittgenstein uses the terms ‘sekundäre Verwendung’ (secondary use) and ‘sekundärer Bedeutung’ (secondary sense/meaning) in Philosophical Investigations (PI 282, p. 216) and in LWPP I (797-798) in relation to the following cases:

• using the concept of pain or pity when playing with dolls (PI 282). In the same remark Wittgenstein discusses applying words like see, hear, talk to inanimate objects in fairy tales or when children play at trains. Playing trains is also discussed in LWPP I 800.
• saying that Wednesday is fat and Tuesday lean and that the vowel e is yellow (PI p. 216, LWPP I 795-799).
• calculating in the head (PI p. 216, LWPP I 801, 802, 804) and reading silently (LWPP I 803)¹

Wittgenstein distinguishes between primary and secondary sense and says that words, like fat and lean have a secondary sense when applied to days of the week and a primary sense when applied, for instance, to human beings. The word yellow has a secondary sense when applied to sounds and a primary one when applied to ordinary perceptions of coloured objects. The same distinction holds of psychological words in their use for human beings (primary use) and inanimate objects (secondary use) as well as of words which refer to activities publicly visible (primary) or inner (secondary).

While it may seem natural to extend the uses of words which we have learned in relation to human beings and overt behaviour to inanimate objects and inner processes, it sounds unintelligible to call days of the week fat and lean or vowels yellow. It seems as if people using these words are talking nonsense. Cato Wittusen, without differentiating between cases, interprets Wittgenstein as saying that secondary uses are, in general, devoid of meaning. “We actually fail to say anything when using words in a secondary mode” (2001, p. 387)².

Now, Wittgenstein, pace Wittusen, does not maintain that secondary sense is nonsense (a proposition which is itself nonsensical) for any of the aforementioned cases. About secondary sense he says the following things:

• 1. that words used in a secondary mode do not have a different meaning from the usual one (the primary one), but a different use. They have their familiar meanings.
• 2. that this different use is not like the different use we find in ambiguous words such as bank and, so one cannot employ different words for the two different uses. One needs to use the same word in primary and secondary use.
• 3. that this different use is not metaphorical.

If meaning is understood as use, then a tension already emerges: How could a word have a different use and yet the same meaning (cf., PI, p. 215)? And if meaning is the same in primary and secondary use, in what sense is the secondary meaning different from the primary to deserve a different label, i.e., “secondary meaning”?

Some problems of Diamond’s interpretation

Cora Diamond, who is one of the very few scholars who have addressed the issue of secondary sense³, eases the above tension by saying that “when we talk about meaning, we do not always mean use” (1991, p. 240). She appeals to Wittgenstein’s remarks that we cannot always explain meaning as use (PI 43), and that, in relation to “is”, he would say that the word has two different meanings (as copula and as sign for identity) but not two different uses (PI 561). The exceptions Wittgenstein is alluding to in PI 43 most probably relate to using the word “mean” in expressions such as “I mean what I say” (cf. Wittgenstein 1988, p. 182). If this is correct, then what he says in PI 43 is not really relevant to resolving the tension mentioned above. In PI 561 Wittgenstein is discussing ambiguous words where the same term stands, as a matter of coincidence, for two different uses and two different meanings. Here Wittgenstein does make a distinction between use and meaning but only in relation to a particular occasion, namely, whether we would prefer to say that the term “is” has two different meanings rather than two different kinds of use. But he does not really say that meaning is independent of use. Consequently, the passage is again irrelevant to our original problem, i.e., the clash between explaining meaning as use and understanding secondary sense as involving the same meaning but a different use.

The rest of Diamond’s article, which applies Wittgenstein’s considerations about secondary sense to ethical language, does not really clarify what secondary sense is and the role it has in Wittgenstein’s work. Her thesis is that ethical uses of expressions involve secondary uses of words. For instance, speaking of “absolute ought” is a secondary use compared to the relative and primary use of ought (e.g., in relation to some task). We first learn the primary use of words and then learn to master the secondary. She concentrates on Wittgenstein’s distinction between secondary use and metaphor stressing that secondary uses, unlike metaphors, cannot be paraphrased. She is interested to show that ethical language cannot be reduced to descriptions of some special sort of facts using words in the primary mode. Ethical language, according to Diamond, is “forced on us”, we are “impelled” to use it, we do not choose (pp.233, 235 237). But there are several problems and some inconsistencies in this account, at least, if it is taken to be exegetical of Wittgenstein’s view:

• Diamond takes the amenability of metaphor to paraphrase in words in their primary sense to mark the difference between secondary use and metaphor when, according to Wittgenstein, the difference is that figurative use, but not secondary use, is up to us. Wittgenstein insists that in secondary use words have their primary meaning which means that paraphrasing, as Diamond understands it, is not even an option.
• Although Diamond underlines the distinction between secondary use and metaphor she says that “[t]here is no harm in saying that a secondary use is in a sense metaphorical, provided we are aware of the differences” (emphasis in the original, Diamond 1991, p. 227). As noted, the differences concern paraphrasing, but it is not clear what this sense is which allows secondary use to be considered metaphorical⁴. She also, indirectly, takes secondary use to be figurative (1991, pp. 236-237), contrary to what Wittgenstein explicitly says (PI, p. 216; LWPP I, 799, 800)⁵.
• As long as Diamond does not explain why Wittgenstein notes the spontaneity of the secondary use of words (PI pp. 197, 204, 215; cf., Wittgenstein 1988, p. 148), it may be taken that secondary uses are somehow mysterious⁶. This is reinforced by the following passage (Diamond 1991, p. 229): “to recognize that expressions may be used in a secondary sense is to see that they are not meaningless in these secondary uses even if we cannot give an account of what they mean in words used in their primary sense.” According to Diamond, words in secondary mode are not meaningless, they have a sense, but we cannot give an account of it in primary terms or in any other way. Earlier in the text, however, she says that in the case of secondary use, “there is no question of giving you an explanation of how I meant the words, different from the ‘perfectly ordinary one’” (ibid., p. 228).
• Diamond (1991, p. 228) is not again clear as regards the meaning of words in secondary use. First she says that in secondary use, as opposed to metaphor, there is no shift of meaning. But later she notes that “if I think of the shift from the usual range, I may be inclined to say that the meaning must be different; while if I recognize that there is no question of giving you an explanation of how I meant the words, different from the ‘perfectly ordinary one’, I may say that the words mean what they always mean.”

How Wittgenstein understands secondary sense

It is the contention of this paper that all cases discussed by Wittgenstein in relation to secondary sense present an increasing difficulty to his statement that meaning is use and this is the reason he considers them. They all involve a reference to some experience or inner process which, purportedly, is connected to meaning. Wittgenstein rejects this view.

Saying that vowels are light, dark or coloured is a clear case of synaesthesia, that is, involuntary cross modal sensation, and in particular grapheme-colour and sound-colour synaesthesia. In cases of synaesthesia, one may be inclined to say that synaesthetes are describing their peculiar perceptions and it is these idiosyncratic experiences that give their reports meaning. Wittgenstein says that even in these extreme cases, where one can indeed say that queer experiences are being spontaneously described, the meaning of linguistic expressions is given by the primary use of words, i.e., use which has been learned prior to the current experience. Meaning is still use and secondary meaning is parasitic upon ordinary primary use. The difference from metaphor is important in this respect. For one, synaesthetic experience is involuntary, so it is not up to us to use certain expressions as it happens with metaphor. Certain concepts are forced on us (PI, p. 204). Secondly, if reports of synaesthetic experiences were assimilated to metaphor, then there wouldn’t be anything really special about this phenomenon, nothing that would connect the use of words to particular experiences. So, it wouldn’t be an extreme case for Wittgenstein to consider in relation to his view that meaning is use.

The examples of other cases, i.e., calculating in the head and reading silently, are considered by Wittgenstein in order to make the same point, namely, that even when our words are supposedly referring to inner processes, their meaning is not given by some inner facts but by the past uses of words in ordinary overt behaviour.

So, if the meaning in secondary use is the same as the primary, how does secondary use differ from primary use and how can we say, at the same time, that meaning is use? Every particular application of words is a new use which is assimilated or not to previous applications establishing the words’ meaning. Wittgenstein says that some new uses of words, in new surroundings, may, or may not, start a new game (cf. BB, pp. 139-140). There is no principled way to determine in advance when this will happen. Wittgenstein speaks of secondary sense, not in order to mark a particular kind of use with certain characteristics, but to point out that certain uses of words, which seem idiosyncratic and dependent upon some inherently hidden entity, are actually parasitic upon primary use. “It is most enormously important that first we learn ‘is red’ and then ‘seems red’. This is fundamental. There is a tendency and temptation to say that all we have are our impressions” (Wittgenstein 1988, p. 151; cf. p. 61; Z 182). Secondary use is dependent upon primary use the way, one might say, secondary qualities are dependent upon primary qualities or secondary signs upon primary signs⁷ (Wittgenstein 2005, pp. 40-47). The secondary simply presupposes the primary.