Topics Language

Question I'm grateful for Allen Stairs' response to question 5821, but he, like Richard Heck and Stephen Maitzen when answering question 5792, ASSUMES that words like "all" have the same meaning in everyday English as they have when used by logicians. That's what seems very strange to me. At least, everyday "all" is ambiguous. Professors Stairs, Heck and Maitzen believe that "all the strawberries he has" always means "all the strawberries he may have", and never "all the strawberries he does have". But look at the latter example ("does have"): you're still using the word "all", but here it is clearly said that he has some strawberries. Why can't that happen (in the right context) with "all the strawberries he has"? By the way, in several Romance languages, there is a difference between (e.g., in Portuguese) "todos os morangos que tem" (indicative) and "todos os morangos que tenha" (subjunctive). Both can be translated as "all the strawberries s/he has", but the first sentence indicates that he (or she) does have some strawberries, and the second sentence says nothing about that. If you need to make the difference clear in English, you'll say "all the strawberries he does have" vs. "all the strawberries he may have". You have to have a special reason to use the subjunctive form (the one that does not imply that he has some strawberries: "all the strawberries he may have") when you're talking about some¬thing you know (whether he has strawberries or not), because in that case you're explicitly refusing to give some relevant information. In English, too, I suppose, when the speaker knows what he's talking about, there must be a special reason for someone hearing him to interpret "all the strawberries he has" as "all the strawberries he may have". Without such reason, it means "all the strawberries he does have". At least I think you should agree that "all" is, in English, ambiguous.

Accepted:
April 6, 2015

Comments

Thanks for your thanks. I'm not sure whether we really disagree. The point of my post is that we can go different ways here, but there are costs and benefits. To repeat my last paragraph,

"There are approaches to logic that find ways around this sort of thing. But the carpet will have to bulge somewhere. Either the rules of inference will be a bit more complicated or we'll have to give up principles that seem appealing or we'll end up with some cases of "correct" inferences that seem peculiar. Different people will see the costs and benefits differently. My own view, which would not win me friends in certain circles, is that there's nothing deeply deep here. But not everyone agrees."

Your concern is about what words like "all" really mean in English, and in particular about whether "all the strawberries he has" actually entails that he has at least one strawberry. Perhaps it does, but I'm not sure this is a question that has a uniquely correct answer. One linguistic approach is to distinguish between semantics—in this case, roughly, what the words really mean— and pragmatics, which deal with how context, background knowledge, etc. enter into determining what someone intends to communicate by their words.

An example might be helpful here. Suppose Alice and Bob were among the candidates to be members of a team. I happen to know that Alice, was picked but Bob wasn't. Dave asks me what I know about the team. I say "At least one of Alice or Bob is a member." On most understandings, what I said is literally true. However, on most understandings, my answer to Bob's question was a bit misleading. Given the usual "rules" of conversation, he will assume that I don't know whether Bob was picked. If I wanted to keep him thinking that Bob might be on the team, this would have been a way to do it without actually saying anything false.

The case of "all" is different in various ways, but the point is this: one way to treat "all the Xs he has" is the way you seem to prefer: it entails that he has at least one X. Another way is to say that this isn't strictly so, but to use the words "all the Xs he has" when I know he has none would be a violation of the usual pragmatic conventions of conversation. I'm skeptical that there is a uniquely right answer to the question, though I'm happy to be persuaded otherwise. There will be theoretical costs and benefits no matter which choice we make, or so I suspect, and the question is how we do the cost-benefit analysis. But in matters like these, I'm not sure what sort of deeper fact would definitively settle the question.

I'm not convinced that your

I'm not convinced that your expression "all the strawberries he does have" is a recognized way of disambiguating the expression that you say is ambiguous: "all the strawberries he has." When would we use the expression "all the strawberries he does have"? As far as I can see, only in special contexts such as this one: "He doesn't have all the strawberries in the county. But all the strawberries he does have are organic." In that example, "does" isn't used to signal the indicative mood; instead it's used merely to emphasize a contrast.

Nor am I convinced that "does" + infinitive always carries existential import (i.e., implies the existence of at least one thing satisfying the verb phrase). Consider:

(P) "All the intelligent extraterrestrials our galaxy does contain are extraterrestrials."

Again, P will sound awkward except in a context such as this:

(Q) "Our galaxy may not contain any intelligent extraterrestrials. But all the intelligent extraterrestrials our galaxy does contain are extraterrestrials."

Whether or not you believe our galaxy contains intelligent extraterrestrials, it would be wrong to deny the second sentence in Q, wouldn't it?