So far we have shown that there is a certain intuitive conception of what the content of the language is, that this is a plausible candidate for his criticisms of the Augustinian picture and that it bears similarity to his pre-Tractarian conception of how to use language with sense. This section will develop the argument by asking two further questions. How far can this conception be seen as influencing the arguments for the GFP? Secondly, in what way does Wittgenstein’s criticism of the Augustinian picture relate to his explicit critique of GFP in the Investigations?
The answer to this will be started in this section and continued in section 4. The criticisms will point out that phrases such as ‘This is how things are’ and ‘This is true’ are only informative in relation to a proposition, given a conception of what the understanding of that proposition entails. That the Tractatus thought that “This is how things are” was the GFP, meant that he had held a particular conception of what it is to understand a proposition that held regardless of a propositions specific content. It finds expression in the Tractatus in terms of us being able to say in advance what the final form of elucidation of any proposition will take.
There are many passages in the Investigations that are relevant to understanding why Wittgenstein later repudiated the GFP, but I will stick to three in which the GFP is specifically mentioned.
1. It ignores the diversity of language (§65, §66)
Wittgenstein believed that philosophical confusion often arose as a result of “an unbalanced diet: nourishing one’s thinking with only one kind of example” (PI 593). In §65 he criticises the GFP by casting his net over a rather wide net of examples. Perhaps it could be considered too wide given that he focuses on linguistic phenomena that prima facie have nothing to do with propositionhood. He refers back to the various sorts of language-games set up in §23. These not only include ones we recognise as propositions (e.g. describing an appearance), or borderline cases (e.g. requesting) but also making jokes and guessing riddles. In focusing on language-games in general he is implicitly arguing against his earlier claim that “The totality of propositions is language” (TLP 4.001). Discovering the essence of a proposition was designed to meet the challenge of finding something common to all language. In the hands of his interlocutor it is reneging on this task that makes him exclaim “You take the easy way out!” Yet, in parading such a variety of language-games in which our words make sense, he bids us “don’t think, but look!” (PI 66): can we see some common element to language? On the one hand, it is hard to see what this has to do with how a word contributes to the sense of a proposition; or whether they share a general form or not. On the other, it makes us question the relevance of finding a general form of propositions for how we use our words with sense.
2. ‘This is how things are’ sounds like a proposition (§114, §134)
“This is how things are”, §114 tells us “is the kind of proposition that one repeats to oneself countless times”. In doing so we come to believe that we come to grasp the essence of a proposition but “one is merely tracing the frame through which we look at it”. Given that it is a phrase that we can use in relation to any proposition, we are misled into thinking that it illustrates a feature they all share. We try to extract from it what a proposition is when abstracted from any particular content. By itself, however, it is just shorthand for the state of affairs in question or a placeholder where a proposition would be. The example Wittgenstein uses is “He explained his position to me, said that this were how things were, and that therefore he needed an advance” (PI 134). Indeed, this can be considered a propositional variable but it does not have any content of its own. It plays no more of a role as a propositional variable than ‘p’ or ‘This is how the cookie crumbles’ and no one would call those the general form of the proposition. The reason we are misled by “This is how things are” because it sounds like a proposition having a subject-predicate form.
3. ‘Truth’ belongs to but doesn’t fit a proposition (§136, §137)
PI 136 affirms certain basic continuities with the Tractatus and yet downplays their relevance for insisting there must be a GFP. The continuity is the idea of the bipolarity of the proposition: it must be capable of being true or false. PI 136 thinks it perfectly apt to say that truth belongs to our concept ‘proposition’. However, he seems to believe that his initial conception of truth was too inflated. Wittgenstein invokes the metaphor of truth as engaging with a proposition, in the way one cogwheel engages with another. If you add ‘is true’ after a sentence and it fits, then it is a proposition. It seems that he took this metaphor as being inherent in the Tractatus, and now renounced it as misleading. The only way ‘true/false’ could fit a proposition is in the way L fits the sequence H, I, J, K....(PI 137) It so happens that in our language L comes after K; likewise we happen to predicate true/false of propositions. Far from illuminating the nature of a proposition, it (as a relevant passage in Philosophical Grammar says) “merely enunciates a rule for signs” (PG p.161) in the following way:
‘p’ is true = p
‘p’ is false= not-p
It is correct to say that the way in which a proposition represents reality must leave it open whether what is asserted obtains or does not obtain. That is, the proposition must have a sense independently of whether it is true or false. However,
The use of ‘true or false’ has something misleading about it, because it is like saying, ‘It agrees with the facts or it doesn’t’, and the very thing in question is what ‘agreement’ is in here.
The criticism in the Investigations means to point out that, in merely being part of the concept of a proposition, it gives us no information as to how it represents reality. Nor does it help indicate that a proposition could only agree or fail to agree with reality in virtue of a single form.
The above criticisms seem to charge the early Wittgenstein both with some fundamental oversights and of building his positions on the basis of a few catchphrases, definitions and pseudo-propositions. The Investigations seems to suggest that had he followed the maxim “don’t think, but look!” (PI 66) he wouldn’t have made those mistakes. In seeing the diversity of language, he would have seen the many contexts in which language makes sense, the different ways propositions can be viewed, and the variety of propositions we can predicate truth and falsity of. Thinking about the questions “What is language?”; “How can we state something about the world?”; “What is the nature of the truth-relation?” may provoke the philosophers’ ‘craving for generality’. Looking should avert this craving and lead us to focus on the particular case. The object of investigation should be the temporal and spatial phenomenon of language and not some non-temporal, non-spatial phantasm.
The problem with the above invocation is that if we are to look, what are we looking for? Seeing the diversity of language does not automatically invalidate the view that there is a GFP. It depends on how it bears on the questions that the Tractatus was looking to answer and how it went about answering them. What is it about ordinary language that is relevant in rebutting the answers given? This is especially pertinent given that Wittgenstein was fully aware that language had diverse uses but decided this wasn’t relevant to the question at hand. He drew the line between meaning and use at a particular point. While we may indeed use language to make a joke, the words must already have a meaning for us to be able to do this. In the Tractatus, it is only a word’s contribution in a propositional context in which its meaning (what it is about) is determined. Even if this last point is incorrect, we cannot show that there is no ‘essence of language’ simply by drawing our attention to different uses without discussing its relevance to a word’s meaning.
The same issue arrives even if we restrict our focus to the variable forms that our propositions seem to take. What are the relevant features of our actual use? Wittgenstein, in the Tractatus, realises that propositions have various (external) forms and this may depend on the circumstances in which it is used. In 4.002 he says, “Language disguises thought... the outward form of the clothing is not designed to reveal the form of the body, but for entirely different purposes.” What matters is the thought expressed by the proposition and not the external form of clothing. Logical analysis will reveal the logical form of the thought- and so, how the proposition is used with sense. He may be- and as I will argue in this essay, is- wrong that analysis will reveal propositions to have the same form. However, the mistake is not that ordinary doesn’t seem to have the form he earlier proposed. He later made a similar distinction between ‘surface grammar’ and ‘depth grammar’ (PI 664). The surface grammar is how the word is used in the construction of the sentence (the part ‘that can be taken in by the ear’); the depth grammar is what, when revealed, makes the meaning clear. It is the purpose of philosophical elucidation to reveal this depth grammar. In both periods this meant removing or ignoring the accidental features of the proposition and only looking at the parts that contribute towards the sense.
 He doesn’t later reject the distinction between meaning and use e.g. PG p.189. However, a word’s meaning can only be ascertained in terms of its contribution to the wider context of life and language, and not just propositions.