This is a follow-up to my earlier post on whether Keith DeRose’s bank cases provide prima facie support for contextualism.
I thought I had an argument that contextualists couldn’t explain some of the data about the bank cases. Then I realised that given enough tools the contextualists could explain anything the invariantists could explain. Whether such a contextualism is plausible is another matter.
Here’s the hard question, in the context of my Hannah/Susanna case below.
Q: Which of the women know that the bank is open Saturday morning?
A: (If you’re having the intuitions the contextualist needs) Susanna.
How, I thought, could a contextualist explain this? After all, there’s only one token usage of ‘know’ here, so only one contextually relevant standard can be in play. And by whatever standard we use, if Susanna knows by that standard so does Hannah. Ah ha, I thought, contextualism is sunk now.
Not so fast! Contextualism can say that what context fixes is not a standard, but rather a (hidden) clause in the sentence that in turn determines the standard. And that clause may well include an anaphoric element. So the question may well turn out to be
Which of the women know (by the standard appropriate to their situation) that the bank is open Saturday morning?
If context fixes the parenthetical material, then it can explain the answer to that question. But now we should start to worry about contextualism from the other direction. If context can set not just a standard, but a function from speakers to standards, we should expect to see much more contextual variability than we actually do. (That’s not an argument yet, but it is a schema for what may be a future argument.)
On the other hand, it might be the case that the kinds of contextualism we accept, in particular contextualism about comparative adjectives, needs just this kind of anaphoric standard-setting. Assume that Jill has two daughters, Margaret and Hillary. Margaret is 5 years old and 5 foot 1 inch tall. Hillary is 21 years old and 5 foot 2 inches tall.
Q: Which of Jill’s daughters is tall?
A: Margaret.
You might wonder how this could be. If ‘tall’ sets a standard for height, then if Margaret is above that height, so is Hillary. The answer here is reasonably simple. The context determines that the content of the question is Which of Jill’s daughers is tall (for her age)? And the answer to that is Margaret.
It’s interesting, and I think instructive, to ask whether we can defend something like subject-sensitive invariantism for comparative adjectives. Can we say that x is tall invariably means x is tall for the kind of thing that it is? I suspect that some, but not all, of the arguments for contextualism about comparative adjectives will work against this kind of invariantism, but I still think it will be worthwhile working out the details.