What [Hegel] calls dialectic and what Plato calls dialectic depends, in fact, on subordinating language to the “statement.” The concept of the statement, dialectically accentuated to the point of contradiction, however, is antithetical to the nature of hermeneutical experience and the verbal nature of human experience of the world. In fact, Hegel’s dialectic also follows the speculative spirit of language, but according to Hegel’s self-understanding he is trying to take a hint from the way language playfully determines thought and to raise it by the mediation of the dialectic in the totality of known knowledge, to the self-consciousness of the concept. In this respect his dialectic remains within the dimension of statements and does not attain the dimension of the linguistic experience of the world….
Language itself, however, has something speculative about it in a quite different sense–not only in the sense Hegel intends, as an instinctive prefiguring of logical reflection–but, rather, as the realization of meaning, as the event of speech, of mediation, of coming to an understanding. Such a realization is speculative in that the finite possibilities of the word are oriented toward the sense intended as toward the infinite. A person who has something to say seeks and finds the words to make himself intelligible to the other person. This does not mean that he makes “statements.”
Truth and Method, III.5.3B
Though he may not have intended it as such, I think Gadamer here pins down the gap between Hegel’s instrumental use of language and Wittgenstein’s privileging of it. Gadamer is talking about Hegel’s notoriously obscure Preface to the Phenomenology and the focus on “speculative propositions.” Hegel distinguishes speculative propositions from the Kantian model of subject-predicate (i.e., object-property) in that the predicate does not limit the subject, but instead explicates the concept inherent to the subject. I won’t say more about speculative propositions as such. Instead, focus on the role of language in the process, which is purely instrumental in generating conceptual and dialectical content. The concept, though it may be disguised, logically precedes the subject, which logically precedes any descriptions given to it. Language does not perform any role over and above the underlying concept, nor does it elaborate on it. It only shows the way back to a revealing of the concept.
What Gadamer says, in effect, is that this underestimates language and overestimates concepts. He mentions “intelligibility” as a task that language can serve from which concepts (and the “statements” in which they are expressed) are excluded. Ignore Gadamer’s double-use of the word “speculative,” and think of language’s role as one of negotiation quite independent of conceptual baggage: an autonomous meaning generator.
This is not an uncommon move in deconstruction, but it’s rarer in hermeneutics because one must still “close the circle,” as it were, and constitute some gestalt of meaning. Gadamer does this at great length, and I believe Wittgenstein does too, though far more obliquely, in his idea of rule-following. Hegel, however, never takes that first step. His intersubjectivity remains one of concepts and not one of language.