I still have a problem with the pragmatist solution to the rule-following paradox, and I think I can reduce my problem to one question. Consider two points:
a. There are an infinite number of functions consistent with any finite set of argument-argument-value trios.
b. If we are, as Rorty says, “good Darwinians”, then we will posit that rule-following ability is a kind of complex mechanical process, different in degree but not in kind from the vending machine’s ability to discriminate between a quarter and a counterfeit coin.
So here’s the question: If both a and b are justified, how can agreement and communication be possible? It would be easy if everyone had the exact same brain, so everyone had the exact same dispositions. But each person’s brain is different, both in the hardware and the software. My neurobiological process for adding will be very different from your neurobiological process for adding. Yet the odds are that if we both do the same addition problem, we will arrive at the same answer. Given that there are an infinite number of functions that we could have been computing, and that there is no such thing as “intrinsic meaning”, how is this not a giant freaky coincidence?