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

1. The Picture

Wittgenstein wants to say that language has its historical origin, and logical foundation, in the primitive reactions that early human beings first had to certain regularities in nature (Wittgenstein 1980, 31e; Wittgenstein 1967, §391, §540-§541, §545; Wittgenstein 1975, §474-§475, §401-§402, Wittgenstein 1958, 230, 56, §142).¹ He believes that this view expresses an obvious historical truism that characterizes our world picture (Wittgenstein 1975, §617-§618; Wittgenstein 1975a, I-§5, Wittgenstein 1958, 230; cf., ibid., §415), but is uncertain of how to interpret this truism. He thinks it a faulty realism to construe the origins of thought as extralinguistic, and a faulty idealism to construe them as intralinguistic. I offer an account of this apparent dilemma and suggest two ways it can be navigated.

2. Taking the Picture Literally

We see the inclination toward realism in Wittgenstein’s concern that the origins of language are logically indescribable. It seems to him that, if it is indeed obvious that “the possibility of a language-game is conditioned by certain facts”, these facts would have be shown, rather then said, to exist (Wittgenstein 1975, §617-§618). Intolerant of the saying / showing distinction, he adds that “that’s not how it is” (ibid.), but he remains uncertain of how this false appearance is to be dissolved.

Struggling to make sense of this illusion, Wittgenstein deviates from his temptation to picture the foundations of language as clearly specifiable primitive behaviours. He now considers picturing them as mysterious somethings in an indescribable “bustle of life” (Wittgenstein 1980a, II-§624-§625; cited in Stern 1995, 191). How are we to imagine the evolution of language from this ethereal background? “The picture is something like this: Though the ether is filled with vibrations the world is dark. But one day man opens his seeing eye, and there is light” (Wittgenstein 1958, 184). Still unsatisfied, Wittgenstein tells us that this picture is obscure, and that it somehow “takes us in” (Ibid.).

This concern is illuminated when we compare this picture with one Wittgenstein considered using to represent the foundations of language circa 1930, but which he recognized as senseless. At this time he believed these foundations were sense impressions of a specious present experiential flux (Wittgenstein 1975a, V-§54, VII-§68, XIV-§168; cf. ibid., XIV-§166; Wittgenstein 1980, 8e, 16e). He realized that this given could not be significantly imagined even as an imprecise array of vibrations in the phenomenal field, for this picture locates the given in the grammatical space of ‘precision’ and ‘imprecision’ and, therefore, fails to station it at its supposed post logically prior to language (Wittgenstein 1975a, XX-§217, §213).²

This realization resounds in the ambivalence about picturing the foundation of thought as an array of vibrations in the ether of a dark pre-linguistic world. No less than when we try to imagine an extralinguistic background as an order of specifiable primitive behaviors, when we try to imagine one as an indefinite blur, the descriptive content of the notion evaporates under scrutiny and leaves us with an inarticulate sound (see, Wittgenstein 1958, §257, §261). The notion, then, ‘take us in’ in the same way it took in Wittgenstein during his middle period; it gives us the impression that language is derived from some shown reality beyond the bounds of sense (Wittgenstein 1975a, XIV-§168). Though it prescinds from the early phenomanlism, the later attempt to imagine the foundations of language still tries to ‘begin before the beginning’ (Wittgenstein 1975, §471; cf., Wittgenstein 1975a, VII-§68).

Despite its apparent senselessness, the view that language is made possible by a reality independent of internal relations is not abandoned (Wittgenstein 1958, §374, 184). This is odd, for when the logical atomist indulges this view, Wittgenstein counters with the observation that “[w]hat looks as if it had to exist, is part of the language” (ibid., §50; cf., Wittgenstein 1974, VII-§95). Why does Wittgenstein not heed this same corrective himself? Why not recognize language as the domain of possibility within which facts exist, and, in line with the view that our world picture might be a kind of mythology (Wittgenstein 1975, §95), recognize the notion of a natural order that makes this domain possible as a useful cosmological metaphor? On this view the foundations of language would be no more logically prior to language than our private mental states are logically private so the problem of sensibly describing those foundations would dissolve.

In fact, the troubles with this tempting realism motivate a countervailing temptation toward just this alternative view.

3. Taking the Picture Metaphorically

After struggling to describe any genuine preconditions of language (Wittgenstein 1956, III-§24) Wittgenstein desists in that effort and submits: “The limit of the empirical—is concept formation” (Wittgenstein 1956, III-§29; cf., Ibid., V-§18; Wittgenstein 1980, 9e-10e). Indeed, in these moods, when he is not “putting the cart before the horse” and construing those founding primitive behaviors as literally prior to language, he describes those behaviors clearly, and without concern that doing so tries to ‘begin before the beginning’ (Wittgenstein 1967, §540-§542, §545; cf., Wittgenstein 1958, 226). Here the background is thought to be fully expressed in hinge propositions (Wittgenstein 1975, §401-§402; cf., ibid., §87, §167; Wittgenstein 1980, 8e), and since the being of propositions presupposes the being of a whole linguistic practice, “[w]hat has to be accepted, the given, is – so one could say – forms of life” (Wittgenstein 1958, 226; cf., ibid., §241).

Though picturing the world as internal to language allows us to speak about the preconditions of thought, Wittgenstein worries that he lacks the conceptual contrivances necessary to support the picture (Wittgenstein 1956, III-§54; cf., Wittgenstein 1980, 9e-10e). Bernard Williams and Jonathan Lear propose that these contrivances are notions of alternative languages.

The view that the body of fact falls within the limits of language is meant to capture the idea that those limits are contingent; they are distinctly our limits as opposed to ones we might have had if the interests of human life had engendered different concepts. However, as Williams and Lear contend (Williams 1981, 158, 160; Lear 1984, 232) it seems to Wittgenstein that the notion of ‘different concepts’ is incoherent (Wittgenstein 1975a, V-§47, VI-§58; Wittgenstein 1974, VI-§71; Wittgenstein 1958, 147). Struggling with this impression he writes:

If [people] really have a different concept than I do, this must be shown by the fact that I can’t quite figure out their use of words. But I have kept on saying that it’s conceivable for our concepts to be different than they are. Was all that nonsense? (Wittgenstein 1978, III-§123-§124, quoted in Forster 2004, 173)

Given our quietist fidelity to the actual grammar of ‘language’ (or ‘concept’), for any use of words to count as ‘language’, we must be able to learn the meaning of those words (Wittgenstein 1958, §124, §207). Sharing Donald Davidson’s premise that an alternative language would not meet this criterion, Wittgenstein also shares his impression that the notion ‘alternative language’ is confused (Davidson 2001, 190, 185). This same impression is expressed as a concern that we misuse the modal operator when we say that these unimaginably different concepts are ‘possible’ (Wittgenstein 1974, VI-§82; Wittgenstein 1958, §497 cf. Wittgenstein 1967, §253-§254; Wittgenstein 1956, II-§84)

Williams and Lear plausibly submit that this Davidsonian observation makes it appear to Wittgenstein that the notion that facts are internal to language amounts to an unsustainable idealism that collapses into the Kantian view that facts fall within the modal boundaries of the world, which could not be other than they are (Lear 1984, 238; Williams 1981, 161, 163). Less plausibly, they suggest that Wittgenstein entertains this idealism, despite its senselessness, because he thinks its truth can be shown but not said (Lear 1984, 242; Williams 1981, 163). This is unlikely not only because the saying / showing distinction is denounced in the later work, but because we can account for why this idealism is tolerated, without disregarding this denunciation, once we consider the later-day temptation to say that grammar is arbitrary (see, Wittgenstein 1974, §133-§134).

Wittgenstein retains an inclination toward idealism because he is reluctant to trust his impression that radically alternative grammars are impossible. The fact that concepts are not rationally necessitated by corresponding metaphysical essences appears to undermine the idea that there are any concepts we couldn’t rightfully countenance or discard (Wittgenstein 1958, §372-§374, §520, §230; Wittgenstein 1956, I-§74; Wittgenstein 1974, VII-§95). Though the view seems to violate his quietist commitment to the grammar we have, Wittgenstein is tempted to say that this failure of essentialism leaves open the ‘possibility’ of even ‘languages' unimaginable to the language we currently have (Wittgenstein 1967, §387-§388, §390, §339; Wittgenstein 1974, VI-§73; cf., Wittgenstein 1956, II-§84). Ultimately, he intimates that grammar can be described in contrast to these possible alternatives, but he is uneasy about this belief having not yet determined how these apparent possibilities can be expressed (Wittgenstein 1975, §501). Correspondingly, he uneasily entertains the idealism that presupposes the possibility of this description, unwilling to endorse that view before he can offer the description with confidence.

We’re told that this puzzlement over the origins of concepts is connected with the apparent impossibility of alternative languages (Wittgenstein 1974, VI-§71). We can now see this connection, and answer our question of why that awkward realism about those origins is not dismissed. Uncertain of the alternative grammars necessary for picturing the preconditions of concepts as internal to a transcendental language, Wittgenstein returns to considering language as an empirical phenomenon that evolves from a genuinely pre-linguistic reality. Reflections on the awakening of consciousness leave him equivocating between an idealism that seems to presuppose shown but ineffable grammars and a realism that seems to presuppose shown but ineffable pre-linguistic facts.

4. Dissolving the Dilemma

How should Wittgenstein navigate this dilemma? Barry Stroud and Michael Forster claim that he need not be able to describe radically different languages in order to countenance the possibility of such things (Stroud 1984, 255; Forster 2004, 181-182). If this were so, he could sensibly specify language as the idealistic limit of the world and dissolve his ineffable realism about the origins of language.

But what if Davidson is correct? If radically alternative concepts are chimerical, idealism really will collapse into a kind of post-Kantian pure realism. However, this result would leave room for us to navigate our dilemma.

To stay true to his quietism, at this juncture Wittgenstein should not follow Davidson and abandon the notion of alternative languages, but part with him and abandon the problematic premise that alternative languages would have to be unimaginable from the perspective of the language we have. The notion of these alternatives would remain important and intelligible; it would simply need to be sensibly cashed out.

To do this, Wittgenstein could side with Quine and say that different conceptual schemes are individuated by different ontologies (Quine 2004, 184). Here, ‘our language’ might be given by the ontology of everyday objects that human beings have in common, and could be contrasted with languages given by the other actual ontologies, and the infinitely more possible ontologies, that we can easily comprehend. This countless plurality of languages would thus replace the monolithic ‘language’ of idealism as the landscape of all that we can imagine.

The general notion of ‘language’ would here amount to a shorthand way of speaking about these particular languages that carries no idealistic implication that these particulars compose a larger language when taken together. The realism of this view recognizes that these languages only compose the necessary modal order of the world. Accordingly, the slogan ‘facts are internal to language’ would paraphrase the idea that every fact is internal to one or another of these various grammars.

Like that idealism that would be left open by the Stroud-Forster view, this realism allows for a picture of concept formation that rests upon neither inexpressible languages, nor upon inexpressible pre-linguistic facts. Our talk about the evolution of concepts would amount to one language’s mythology about what had to exist as a precondition for the possibilities of mind and world that we find in the space of grammars.

Whether or not quietism undermines the arbitrariness of grammar and precludes the possibility of radically different concepts, Wittgenstein can conclude that talk about the origins of language specifies facts internal to language and, so, amounts to a kind of cosmological metaphor.