Unformatted text preview:

Brandom10/28/09Sellars Week 9 Notes “Being and Being Known” (1960)“The Lever of Archimedes” (1981—First Carus Lecture)1. TLA is a late, on the whole disappointing piece. (Certainly it was a disappointment as the opening one of the much-anticipated, much-delayed Carus lectures.) It is an intricate discussion of a 1949 essay of Roderick Firth, and of WS’s debate with Roderick Chisholm (the two ‘Roderick’s). It is a very careful, subtle, and sensitive working out of a dialectic, in which he shows great sensitivity to the possibilities of the positions with which he is disagreeing.2. The 1960 “Being and Being Known” (BBK)—given the same year as “Phenomenalism”, just after the two 1959 GE lectures, and just before NS—tries to say what is right and what is wrong about the Thomistic neo-Aristotelian “doctrine of the mental word.” It is by far the most sophisticated discussion of this topic I have ever seen. I am an amateur with that scheme, needing to be guided by his account. He also has an intertwined discussion of Descartes, as insisting that “direct knowledge” of our sensations must be “non-analogical” knowledge of them. This is largely orthogonal to the main discussion (it sets up a final point functionalist point at [228] passage (37)), and I won’t say much about it. I’ll discuss two issues about this essay:a) A fascination passage on two kinds of logical “words”.b) The main discussion, which is summarized in the passage from [227] in (7) below, also at. I shall argue that a confusion between signifying and picturing is the root of the idea that the intellect as signifying the world is the intellect as informed in a unique (or immaterial) way by the natures of things in the real order. [218-9].[218-9] at (22) in the Passages. As that passage makes clear, everything turns on the distinction between two kinds of isomorphism between words and world: picturing and signifying. So we must get clear about that.c) Throughout he works in a framework that distinguishes “the real order” from “the logical order”, or, apparently equivalently, “the intentional order.” Words regarded one way are in the real order and stand in picturing relations; regarded another way, they are in the logical-intentional order, and stand in signification relations.1 tmp18ks2itzBrandom3. Here is the focal passage on logic: [T]he intellect in first act has logical words in its vocabulary. Some of these logical words are, in the contemporary phrase, ‘truth-functional connectives’ (e.g. and and not), the most significant feature of which is that if a sentence or group of sentences is about the real order, the sentence which is formed from them by the use of these connectives is also about the real order. Thus Socrates is not wise is as much about the real order as Socrates is wise. Other logical words, e.g. implies in thesense of logical implication, are such that sentences involving them are about the logical order, as is shown by the fact that these sentences require abstract singular terms (e.g. Triangularity implies trilaterality). [BBK-217]a) On the face of it, WS is making an astonishing distinction between bits of logical vocabulary: some produce statements sabout the real orders and some produce statements that are sabout the logical orders (which later on he identifies with—or maybe treats as species of the same genus as—the intentional order). b) Presumably the two-valued conditional, i) as a truth-functional connective and ii) as definable from negation and either conjunction or disjunction, is also “about the real order.”c) But what about non-truth-functional conditionals? Modal conditionals, or just those that are counterfactually robust, are, he tells us in CDCM, covertly metalinguistic, conveying inferential commitments involving expressions related (in a dot-quote way) to the sentences appearing as antecedent and consequent of the subjunctive or counterfactual conditionals. d) ‘Implies’ is often used in an explicitly metalinguistic way, as in: “‘It is raining,’ implies ‘The streets will be wet.’” And WS thinks it is covertly metalinguistic in “That it is raining implies that the streets will be wet.” For he thinks ‘that’-clauses express (and purport to name) propositions, a kind of abstract entity, and that their proper analysis is by using both DSTs and dot-quotes: that-p = the -p-. He would not approve of un-quoted, un-thated sentences flanking ‘implies’: *”It is raining implies the streets will be wet.” e) Read this way, ‘implies’ is not really a logical connective, since it is not a kind of conditional in the object-language.f) What about quantifiers? There are two reasons to think that they also count as forming statements sabout the real orders: i) They do figure in perspicuous (Bradleyan-regress-avoiding) Jumblese, alongside names. Notice that in the definitive “Naming and Saying”version of Jumblese, which includes rules for translating PM-ese into Jumblese, nothing is said about how Jumblese handles logical sentential connectives. Presumably, logically compound truth-functions of Jumblese sentences are standard displays of those sentences 2 tmp18ks2itzBrandomrelative to one another. So if “a is red” is a, and “a is square” is a, one might write “a is red or a is square” as aa. Conjunction might be: aa. ii) If quantification is handled substitutionally, then universal quantification is a special kind of conjunction, and particular quantification is a kind of disjunction. (We know from Belnap’s and Kripke’s work that, contra Quine’s claims, the expressive power of substitutional quantification is not less than that of ‘objectual’ quantification. The problem is quantifying over uncountable collections, e.g. the real numbers, in a language that only has a countable number of singular terms. The solution is, in effect, in the metalanguage to quantify over arbitrary extensions of the countable language. In any case, Quine must think there is some way to invoke such collections in the metalanguage in which the domain of his models is specified.) As far as I know, WS never says how he thinks about quantifiers.4. Here is the next passage:The idea that acts of sense are informed by not as well as by white, by implies as well as triangular is rooted in the fact that whiteness is what it


View Full Document

Pitt PHIL 2245 - Sellars Week Notes

Download Sellars Week Notes
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Sellars Week Notes and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Sellars Week Notes 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?