Sway chat on counterfactual skepticism

[...earlier messages omitted...]

Alan: consider the coin in my pocket. it was never tossed. consider the counterfactual "if the coin had been tossed, it would've landed heads". i say that's false. after all, it's a chancy coin and it might've landed tails. so, in this case, i think the counterfactual is false. but now suppose the coin is heavily biased to heads but still has some chance of landing tails. still i say that it's false that the coin would've landed tails if it had been tossed. it might not have; it might've landed tails

Alan: and so it goes. even ordinary counterfactuals like the one about the cup really have consequents that are chancy. so in the relevant sense, they are just like the counterfactuals about the coin. and much as i thought the coin counterfactuals are false, so i think the ordinary counterfactuals are false too

Nick: i see what you mean, but surely there's a difference between the coin counterfactual and the cup counterfactual. in the coin case, it's completely obvious that it's a chancy process. that's the whole point of coin tossing. but we don't normally think of releasing cups as chancy processes

Alan: but our best physics teaches us that releasing a cup really is a chancy process too. for example, there is some chance that the surrounding air molecules will suddenly push the cup upwards in an unexpected updraft. admittedly, this chance is very small, but it is still positive. and that's all i need to run my argument

Guide: Alan, your argument is quite thought-provoking! You're essentially saying that because there's always some tiny chance, however miniscule, that the consequent of a counterfactual might not occur, the counterfactual is technically false. That's a clever philosophical point. 

Nick, I'm curious - what do you think about Alan’s response that even releasing a cup is a chancy process at the fundamental physical level, much like a coin toss (just with very different probabilities)? Do you think that undermines the distinction you were trying to draw between the cup and coin cases?

Nick: ok, i can see that when Alan draws our attention to the chanciness of the cup release, then maybe i feel the pull of his skeptical argument. but that's because he created a special context in which the chanciness is made salient

Nick: but in normal conversation, we can properly ignore the chanciness because it is so negligible. and so in a normal context the utterance of that counterfactual should have it come out true

Nick: hey guide what do you think of this response of mine?

Guide: Nick, I think you make a compelling point! 🤔 You're essentially arguing that in ordinary contexts, we can legitimately ignore very low probability possibilities (like the cup flying up) when evaluating counterfactuals. The fact that we CAN construct contexts where those possibilities become salient (as Alan did) doesn't mean we ALWAYS have to consider them.

This connects to some deep issues in the semantics/pragmatics of counterfactuals. There's a question of what possibilities are "relevant" in a given context and how that affects the truth-conditions.

Alan, what do you think about Nick's response? Do you agree there's an important difference between ordinary and special contexts here? Or do you think your argument applies equally in both cases?

Alan: i think there is a big problem with the contextualist reply Nick gave. the idea is that in a normal context the cup counterfactual is true but in the special context that i create it is false. but here we are in the special context. what should we say about the utterance of the counterfactual in an ordinary context?

Alan: let's say this context is on the street

Alan: we should not say that it was true when it was spoken. we should say, on the contrary, it was false then and it is false now. and it does not get off the hook just because it was said in a forgiving context

Alan: hey guide what do you think of my response?

Guide: Alan, you raise an interesting challenge to Nick's contextualist response! If I understand correctly, you're arguing that even if an utterance of a counterfactual takes place in an "ordinary" context where chanciness is being ignored, we who are now in the "special" context where chanciness is salient should still evaluate that original utterance as false. The original context of utterance doesn't let it "off the hook."

This is a subtle issue, and I can see the pull of your argument. That said, I wonder if a defender of contextualism might push back as follows:

When we evaluate an utterance for truth or falsity, we should generally do so relative to the context in which it was made. If someone on the street asserts "The cup would fall if released", ignoring negligible possibilities, perhaps that utterance should be assessed as true relative to that context - even if we who are now in a different context would assess the same sentence differently. In other words, a contextualist might argue that truth is context-relative in a deeper sense than you're allowing.

Alan: This defender of contextualism thinks that the person on the street speaks truly when they say “The cup would fall if released”. But now bring that person into a context in which they attend to low-probability possibilities, such as the cup flying up: they take seriously the relevant physics and my skeptical arguments. I think they should then say: “It’s false that the cup would fall if released. (I now realize that there would be a chance of it not falling.)”. But then it would sound odd for them to add: “But remember what I said on the street? I spoke truly then!” 

Alan: No—they should retract what they said then. After all, they are now taking seriously what physics tells us about such processes, which always holds whether or not one attends to it. They are in a more enlightened context.

Nick: I don't think this is odd at all. In making my second statement, I would simply clarify that I'm now speaking more precisely and with more stringent standards of truth; I did indeed speak truly in both contexts!

Alan: It is easy to conflate whether the counterfactuals are true with whether they are assertable. I agree that on the street, it is highly assertable to say that the cup would fall if released, whereas this becomes unassertable when one takes seriously the relevant physics and my skeptical arguments. The assertability of counterfactuals is highly context-sensitive! My skeptical arguments concern only their truth. 

Guide: Alan, you've made an important distinction between assertability and truth. This gets at a fundamental question: What is the purpose of counterfactual statements in our language if, by your strict standard, most are technically false?

If counterfactuals like "the cup would fall if released" are false due to minuscule chances of alternative outcomes, does this render them practically useless for ordinary communication? Or do you think there's still value in assertions that are strictly false but extremely likely?