gc: “how can you teach mathematics if you believe that 1 = 1 has the same truth value as 1 = 2?”
I don’t know, but Derrida didn’t believe this or argue for it. Nowhere in Derrida’s writing does he challenge the notion of truth as applied within discursive domains—among them mathematics. What he takes himself to be challenging is an uncritical metaphysical account of what our traffic in truth, plain meanings of words, etc. amounts to. As Derrida himself puts it in Limited Inc.:
“[T]he value of truth … is never contested or destroyed in my writings, but only reinscribed in more powerful, larger, more stratified contexts. … [W]ithin interpretive contexts … that are relatively stable, sometimes apparently almost unshakeable, it should be possible to invoke rules of competance, criteria of discussion and of consensus. … I take into account and believe that it is necessary to account for this stability [of interpretive contexts], as well as for all the norms, rules, contractual possibilities, that depend upon it. But … to account for a certain stability … is precisely not to speak of eternity or of absolute solidity; it is to take into account a historicity, a nonnaturalness, of ethics, of politics, of institutionality, etc. … I say that there is no stability that is absolute, eternal, intangible, natural, etc. But that is implied in the very concept of stability. A stability is not an immutability; it is by definition always destabilizable.”
Derrida is not a truth-nihilist. There are deep, fundamental problems with his philosophy and his method, but this is not among them. If lit crit popularizers, or unsympathetic critics, have tried to make Derrida into one, that is their problem, not his. For an excellent overview of what Derrida is (and isn’t) doing, and some of the interpretive and philosophical problems involved with reference to contemporary Analytic philosophy, I recommend Martin Stone’s “Wittgenstein on Deconstruction,” anthologized in THE NEW WITTGENSTEIN (eds. Alice Crary and Rupert Read).