WinterReadingGroup < Discussion

Truth, ze Past Winter Reading Group

Intro
Links
Discussion

Intro

This is to be a space for the winter philosophy reading group. The text will be Dummett's Truth and the Past. If you are interested in joining, please email dsidi [*at*] u [*dot*] arizona [*dot*] edu.

To add: (1) register with the wiki, (2) click the "edit" link above (3) read the "help" link next to "release edit lock" and "minor changes...", (4) choose a legible color for your posts (the last person who adds will automatically append a signature with his/her name and the date). Please type %ENDCOLOR% %[whatever color] before your addition, and %ENDCOLOR immediately following it.

Here is a list of TWiki TextFormattingRules. If you have trouble doing something, you can also look at the preexisting text in the "edit" box as an example. Its pretty easy, just play around a bit.

Discussion

Chapter 1: The Concept of Truth

(I'll be brown)

Dummett begins with an articulation of truth-conditional semantics. I think he does so because his (still unstated) general understanding of Realism is as a semantic thesis. More specifically, if I remember right, he requires:
- A language with a truth-conditional semantics
- bivalence
- the referents of the terms in the language to be mind-independent entities
Two criteria for a genuine theory of meaning are that it must explain:
- the propositions or concepts expressed by the linguistic entities, and
- in virtue of what this expression takes place
  - I think this is a useful generalization from Frege's initial distinction between sense and reference.
What do you think of Dummett's identified tension between deflationist theories of truth and truth conditional theories of meaning? I found his argument a little quick.
Anscombe's example of a conditional that fails the principle of semantic ascent:

If Henry can eat any fish, then the statement "Henry can eat any fish" is true.

- Dummett points out that Henry's eating trout makes the antecedent true, but that the consequent then doesn't follow. Hmm, is the problem that the scope of "any" in the antecedent is more limited in the antecedent than the consequent, and we can't recognize this difference in scopes, since we are allowing ourselves only to notice that the same form of words is used in each clause?
  - p. 22 Does only noticing the same form of words in each clause work if we consider only the Tarskian truth schema? In that case, the language in which the definition is expressed could be logic, thus avoiding circularity by supposing that the language of the definition conincides with the object-language.
Dummett espouses (with Frege, Russell, and the logical positivists) a tenet of "ideal language philosophy" by pointing to the failure in natural languages of "harmony" between grounds and consequences (which are features constituting "use"), and leveraging this failure to support reform of natural language.
- p. 26: "If the best fully coherent theory of meaning for a language fails to fit completely with the conventional practices of its speakers, the nlanguage is in need of reform; and the theory will show in which respects it needs to be reformed."
  - Though Dummett appeals to Witt., Witt. did not endorse the program of reform, even early on. His early writing was rather concerned with revealing the deep structure of nat. language. Davidson accepts something closer to this program than to Dummett and Frege's.
It is pointed out that the notion of truth is somewhat weak that comes from theories which have use as central to meaning. All we get for truth is warranted assertability.
- Putnam at times accepts truth as something like that. From "Why Reason Can't Be Naturalized": "Truth, in the only sense in which we have a vital and working notion of it, is rational acceptability ...."
p. 22 - One stated requirement for the truth-conditional theory is that we bring our intuitive understanding of "truth" - that "truth" is not a purely theoretical term.
- Dummett later (p. 26-27) argues that explaining "truth" (making it a theoretical term) requires getting from it to the features of use.
  - But will not any such explanation still presuppose an understanding of "truth", since we must evaluate what counts as the conditions for truth prior to showing how to get from these conditions to (say) the grounds for asserting a statement?
If use cannot be explained without prior appeal to truth and truth-conditions, the truth-conditional theory is necessary to explain meaning in terms of use.
- this could go wrong several ways, I think:
  - if there is something else we "implicitly grasp" (speaker intention?)
  - if a grasp of truth and conception of truth-conditions would require supplementation (e.g., in cases of vagueness). A failure to know the truth-conditions of a statement in vague cases doesn't make our understanding of the meaning of the statement incomplete, but it might make the meaning itself incomplete in some sense.

Chapter 2: The Indispensability of the Concept of Truth

pp.35,39: Dummett's relative placement of three big topics: truth, meaning, and metaphysics
- metaphysics is concerned with what truths hold good in general
- a theory of meaning must include a theoretical notion of truth (which I think must be primitive)
  - truth is presupposed to judge whether the principles of semantic shift hold
- a conception of truth is constrained by the theory of meaning with which we understand our language.
  - different theories of meaning accept different central notions (proof, possible worlds) and different conceptions of truth
two reasons truth is indispensible:
1. compositionality: how the fundamental semantic property of a sentence is determined by its composition.
  - deduction is vital to this broader project of compositionality. Deduction requires a notion of truth, even if you think other kinds of compositionality don't. Truth must be included as what is preserved from premise to conclusion.
2. assertoric content: assertions require judgement, which generates a notion of truth as whatever would justify the assertion of a statement.
The examples of divergent assertoric content and ingredient sense (what a sentence contributes to a more complex sentence of which it is a part) are faulty, I think:

1. It is raining here.
2. It is raining where I am.
Becomes
1a. It is always raining here.
2a. It is always raining where I am.

Isn't 2a ambiguous, though? Dummett favors taking it as temporally flexible, but consider

2b. where I am, it is always raining.

Doesn't this seem more rigid, and the ingredient sense the same in 1a and 2a?
Next example:

1. I shall give you a D.
2. I intend to give you a D.
Becomes
1a. If I give you a D, you will forfeit your grant.
2a. If I intend to give you a D, you will forfeit your grant.

Do 1 and 2 even have the same assertoric content? 1, asserted "on its own on any occasion," is not a claim about intentions. Am I being dense?
Last example:

1. Clare's husband is Egyptian
2. Clare is married to an Egyptian.
Becomes (though it is left unwritten by Dummett)
1a. Clare does not have a husband who is Egyptian.
2a. Clare is not married to an Egyptian.

the problem is supposed to be presuppositions here, but as I have formulated the negations, I don't see it. I could see it if the negation of 1a had a narrow scope, e.g.

1b. Clare's husband is not Egyptian.

but why would we accept that?

lastly, the assertoric contents of 1 and 2 aren't the same, since Clare may be married to another woman.

Chapter 3: Statements about the Past

why is it that justificationist and pragmatist theories of meaning come to the same in the end, as he says on p. 40? Because of the requirement of "harmony" (p. 25-26), which yields one of the two notions when the other is taken as central.

In the previous chapter, Dummett broadens arguments for the necessity of a notion of truth based on deductive inferencefor deduction by distinguishing between ingredient sense and assertoric content. Now he takes things a bit further: truth conditions for mathematical statements are permanently available, whereas those for empirical statements may be lost in the past.

but this permanence/changeability division applies only to a particularly exacting, direct sort of proof. Indirect proof fixes things, so it is taken on board as the notion of truth to be utilized in a truth-conditional theory of meaning.
- p. 46 - Ultimately, the point of the chapter is that truth conditions are sufficient but not necessary for a theory of meaning: the possibility of verification is the direct form of proof. What a statement is understood to say is the indirect form of proof, and it is acceptable too (and preferable, much of the time).

Dummett qualifies his at times positivist-sounding claims by pointing out that his notion of justification goes toward warranted assertability, which is weaker than what verification goes toward.

p. 46 - some programmatic statements call for some clarification: Dummett wants to show how "if meaning is to be explaineed in accordance with a justificationist theory, a speaker may be supposed to acquire a tacit grasp of truth so conceived," so we're aiming for an explanation of why any justificationist theory must take either the grounds or the consequences of a statement as central, and implicitly grasped. Davidson, on the other hand, just took truth to be implicitly grasped.

Chapter 4: The Semantics of the Past Tense

A distinction is made among mathematical proofs between those that can potentially be known and those that potentially are, independent of our knowing. This distinction is interesting when applied to Dummett's discussion of empirical proof.
- Proof in the empirical sense, whether direct or indirect, is the potential for knowledge transmissable through the common channel of language. In those instances in which there is no observer (so a proof cannot potentially be known) but there could potentially have been one, we are left with a conditional whose antecedent is unfulfilled.
  - This gives us the excluded middle for the observation ( $A \lor \neg A$ ), but with neither disjunct determinately holding: a "gap in reality," and a maintained source of Dummett's justificationist's anti-realism.