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

1. Introduction

How is it that the words come to have the meanings they have? Wittgenstein’s Tractatus Logico-Philosophicus has been often interpreted as giving a psychologistic answer to this question. Especially 3.11 is taken to suggest strongly such an interpretation:

We use the perceptible sign of a proposition (spoken or written, etc.) as a projection of a possible situation.

The method of projection is to think of the sense of the proposition. (TLP 3.11)

This remark has been often interpreted as saying that it is the “thinking” independent itself of language that gives meanings to propositions. Such a psychologistic reading seems to be natural in a sense. In fact, not few interpreters adopt such a reading¹and regard the view of the Tractatus concerning language and thought as psychologistic one.

In what follows, I will deal with the Norman Malcolm’s book, Nothing Is Hidden, as a typical example of such a reading and point out the problems that it gives rise to and like to present a possible way of understanding the Tractatus’ framework of views concerning language and thought in a non- psychologistic way. One of the most important points in doing so is how to read 3.11.

2. A psychologistic reading

Malcolm’s psychologistic interpretation is clear in what he calls “the ‘hidden’ philosophy of mind and of language of the Tractatus” (NIH, p.76). Here let M designate a mental element, S a simple sign (name), and O a reality-element (object). Then it can be formulated in the following three theses (ibid.).

• i. M intrinsically means O.
• ii. S is correlated with M by stipulation or convention.
• iii. As a consequence of 1 and 2, S comes to mean O.

A simple sign S comes to mean a reality-element O by means of its being correlated (by stipulation or convention) with a mental element M that means O intrinsically, i.e., by its inherent nature. Thoughts are composed of psychical elements and propositions are composed of simple signs. Thus propositions has senses by means of their being correlated (by stipulation) with thoughts that are in turn intrinsically significant.

How does Malcolm come to attribute such a psychologistic picture to the Tractatus? Firstly, he cites a Wittgenstein’s letter to Russell in which he answers to Russell’s question of whether thoughts are composed of words or not (NIH, p.65).

No! But of psychical constituents that have the same sort of relation to reality as words. What those constituents are I don’t know. (NB, p.131)

On the ground of this remark, Malcolm comes to separate (ontologically) thoughts composed of mental elements from propositions composed of physical elements (signs). Then he points out that in the Tractatus a relation is supposed to obtain between a thought and a proposition, i.e., the relation of the latter’s expressing the former (NIH, p.66).

In a proposition a thought finds an expression that can be perceived by the senses. (TLP 3.1)

I call the sign with which we express a thought a propositional sign. – And a proposition is a propositional sign in its projective relation to the world. (TLP 3.12)

In a proposition a thought can be expressed in such a way that elements of the propositional sign correspond to the objects of the thoughts. (TLP 3.2)

Malcolm draws from this the conclusion that “in the Tractatus thoughts (Gedanken) are more basic than word-propositions (Saetze)” (NIH, p.66). I think that for validity of this conclusion the “psychical constituents” should be established in advance as being independent of language. But he gives no argument supporting such a view. He rather presupposes it and simply says that “a thought does not have to be expressed in a physical sentence” (ibid.).

After in this way Malcolm attributes to thoughts a character of being independent of propositions, he proceeds to interpret 3.11 (“The method of projection is to think of the sense of the proposition”). 3.11 can be just what he wants, because he ascribes priority to thoughts over propositions (NIH, p.67) and in this remark Wittgenstein seems to say that propositions come to have their senses only by means of thoughts (or an activity of thinking). And it is just the way he interprets it (NIH, p.73, cf., Kenny 1981, p.142, Hacker 1996, p.31).

… A thought already has sense; it is already a picture of a possible situation. The question now is: how does a physical sentence become a picture? The answer of 3.11 is that a sentence becomes a picture (a projection) by virtue of a thought’s transmitting its own sense to the physical sentence. A sentence, by itself, is ‘dead’: it is without sense, it says nothing. … But when a sentence is filled with a thought it takes on life: it becomes a visible or audible thought. (NIH, p.73)

3. How to read it non-psychologistically

Wittgenstgein’s views upon which Malcolm’s interpretation is based are (1) that thoughts are composed of psychical constituents (his letter to Russell) and (2) that thoughts are expressed by propositions (TLP 3.1, 3.12, 3.2) and (3) that to think (Denken) plays a certain role for a proposition to have its sense (TLP 3.11). Malcolm elicits the alleged “‘hidden’ philosophy of the Tractatus” only from these. But it should be noted that at least (1) and (2) are not theses that imply commitments to a particular philosophical position concerning thought and language.

Both theses are compatible especially with what I regard as a basic line of thought of the Tractatus, i.e., what may be called a loose identity theory of thought and language:

Now it is becoming clear why I thought that thinking and language were the same. For thinking is a kind of language. For a thought too is, of course, a logical picture of the proposition, and therefore it just is a kind of proposition. (NB 12/9/16)

A propositional sign, applied and thought out, is a thought. (TLP 3.5)

A thought is a proposition with a sense. (TLP 4)

I think we can understand this view more properly from the view point of inner utterances of propositions (as physical signs). For example, when we, at a restaurant, “think” about what kind of dish we will order, is it that we, representing a dish of meat and of fish, innerly utter the proposition “I got a dish of meat last night, so I’ll get fish tonight” or the like?

Or think about a case where we are required to speak in an unfamiliar language. In such a case, are we forced in turn to be aware that our everyday thinking is operated in our own mother tongues? An activity that we ordinarily call “thinking” seems at least to me to have a highly linguistic character, though of course it is not perceptible with senses.

Though we regard thinking as inner utterances of propositions, at least (i) and (ii) give rise to no problem. Propositions, as far as they are innerly uttered, can be said to being composed of psychical constituents (i.e., signs, as far as innerly uttered). And propositions innerly uttered can be transformed easily into physical propositions composed of visible or audible signs. Physical propositions thus obtained can be said to “express” the content of the original thinking (i.e., the original thoughts) (for example, the remark 1 of the Tractatus that “the world is all that is the case” can be said not only to mean that the world is all that is the case, but also to express the Wittgenstein’s thought that the world is all that is the case).

How about the third datum of which Malcolm makes use for his interpretation, i.e., the remark 3.11?

We use the perceptible sign of a proposition (spoken or written, etc.) as a projection of a possible situation. The method of projection is to think of the sense of the proposition. (TLP 3.11)

In the Tractatus for a proposition to be “a projection of a possible situation” is just for it to have a particular truth-condition, i.e., a particular meaning (“Sinn”). Therefore we can rephrase “the method of projection” as “the method of giving a meaning”. Then the above remark can be interpreted as saying that the method of giving a meaning to the proposition is to think of the meaning of the proposition. If we assume a mental activity of “thinking” that is independent of language, then this remark is supposed to evoke a psychologistic interpretation (a la Malcolm) that it is such a “thinking” that gives a meaning to a proposition.

But here we should note that the “meaning” has at least two aspects. I like to make a distinction between an “expression meaning” and an “utterer’s meaning” (cf., Searle 1979, chapter 4, Kripke 1977). Roughly speaking, the former is known by everyone who masters the language to which the expression belongs and is a meaning that the expression itself has, and, in many cases, it is its lexical meaning. We can say of the latter, roughly speaking too, that an utterer’s meaning of an expression is a function of its expression meaning and the context in which it is uttered (i.e., who utters it and in what circumstance it is uttered, etc.).

By the way, the Tractatus’ method of philosophy is the analysis of our everyday language. It means that the alleged “completely analyzed” language (TLP 3.201, 3.25) cannot but take over its meaningfulness from our everyday language. That our everyday language is presupposed as a datum means that the conception of meaning in everyday language, i.e., the conception of meaning that cannot be reduced into individual “thinking”, in short, the “expression meaning” distinguished above is presupposed. This is the matter of course in the loose identity theory of thought and language, or, it amounts to the same, in the conception of thought as inner utterances of propositions. For a proposition that is innerly uttered in our thinking must have a particular (expression) meaning in advance before it is entertained in the thinking.

3.11 was paraphrased above as “the method of giving a meaning to the proposition is to think of the meaning of the proposition”. Here are two occurrences of “meaning”. If these are identical, the remark becomes merely circular. For what we are supposed to have to “think” in order to give a meaning to a proposition can be nothing but the meaning of the proposition itself. Then the following paraphrase will be suggested.

The method of giving an utterer’s meaning to the proposition is to think of the expression meaning of the proposition.

Uttering innerly a proposition that has in advance a particular expression meaning (in short, thinking) constitutes an utterer’s meaning of the proposition.

But here a question arises immediately. Why thinking? Why an inner utterance? An “utterance” in an ordinary sense is to utter something in voice (externally). In fact, an utterer’s meaning of a proposition is constituted not only by an inner utterance of it (thinking) but also by an ordinary (external) utterance of it.

Here it is worthwhile to refer to the corresponding remarks of the Prototractatus that is (presumably the last) draft of the Tractatus:

A propositional sign is a projection of a thought. (PTLP 3.11)

It is a projection of the possibility of a situation. (PTLP 3.111)

The method of projection is the manner of applying the propositional sign. (PTLP 3.12)

Applying the propositional sign is [the same as] thinking of its sense. (PTLP 3.13)

As was seen above, the Tractatus connects “the method of projection” and “to think of the sense of the proposition” immediately. But the Prototractatus mediates them through “the manner of applying the propositional sign”. I think this is highly suggestive. For “applying the propositional sign” can be thought as accommodating not only an inner utterance of the sign but also an external utterance of it. The picture that to apply a proposition that has a particular expression meaning constitutes its utterer’s meaning can be said even to formulate succinctly a conceptual relation between these two aspects of meaning.

Now the German original of the Prototractatus 3.13 above reads, “Die Anwendung des Satzzeichens ist das Denken seines Sinnes”. If this remark is read as defining or explaining “Anwendung” (applying) by means of “Denken” (thinking), this remark is interpreted as closing the sight again that is opened by introducing the concept of “Anwendung”. For, read in such a way, “applying the propositional sign” as “the method of projection” is supposed to be restricted to an inner utterance of the proposition (in short, Denken or thinking).

This remark rather should be read simply as an identity statement in just the way the English translators of the Prototractatus read it. It reads, “applying the propositional sign is [the same as] thinking of its sense”. If it is read in this way, it turns out to be almost another version of the cited remark of the Notebooks 1914-1916 (NB 12//9/16) where the loose identity theory of thought and language is suggeted. In fact there it is said that “thinking and language were the same” and that “thinking is a kind of language” and that “a thought … just is a kind of proposition”.

In virtue of the above consideration, we can read the Tractatus 3.11 from the view point of the loose identity theory of thought and language. Namely we can regard what is meant by “to think of the sense of the proposition” as including not only an inner utterance of the proposition but also an external utterance of it, i.e., to apply it in general. Thus the following reading will be suggested.

The method of giving an utterer’s meaning to the proposition is to think of the expression meaning of the proposition, in other words, to apply the proposition having a particular expression meaning .

Compare this with Malcolm’s psychologistic reading: the method of giving a (expression) meaning to the proposition is to think of the (expression) meaning of the proposition. We can point out at least two problems. (1) This reading seems not to be able to avoid the vicious circle mentioned above. Even if it could avoid, (2) this reading presupposes a mental activity of thinking that is independent of language, but that is incompatible with the Tractatus’ basic line of thought, i.e., the loose identity theory of thought and language².

4. Summary

I have presented a possible way of reading 3.11 non-psychologistically that is a stronghold of a psychologistic interpretation of the Tractatus. The point is (1) to regard the loose identity theory of thought and language as the basic line of thought of the Tractatus and (2) to distinguish between an expression meaning and an utterer’s meaning (or the like) and (3) to interpret the phrase “to think of the sense of the proposition” by which what is said is a main source of a psychologistic interpretation as meaning “an inner utterance of the proposition” or “to apply the proposition”, either of which has no psychologistic implication.

The above interpretation based upon the loose identity theory of thought and language can secure the consistency with the aim of the Tractatus in a way a psychologistic reading cannot. Namely the aim of the Tractatus is “the logical clarification of thoughts” through an analysis of (our everyday) language (TLP 4.112).