If it turns out that I'm Leverrier's wife
What would you say if it turns out that the watery stuff in our rivers and lakes doesn't actually consist of H2O, but of XYZ: would you say that water consists of H2O or XYZ?
What would you say if it turns out that you are Leverrier's wife living in 1845 and the heavenly body your husband calls "Neptune" is not a planet, but a spaceship: would you say that Neptune is a spaceship or a planet?
There's something odd about the second question.
I am disposed to assent to certain sentences under certain conditions, to "it's raining" if it's raining, etc. For each sentence, this determines a function from conditions -- sets of centered worlds -- to truth values. (If I am disposed to assent to S under condition C, that doesn't mean I assent to S in all C-worlds. I need only do so in the closest C-worlds. I am not disposed to assent to "it's raining" under the condition that it's raining and I am halluzinating that it doesn't rain.)
I'm not disposed to do anything under the condition that I'm Leverrier's wife because (arguably) it's impossible that I'm Leverrier's wife. But those dispositions and the corresponding function aren't very useful anyway. One other problem is that it isn't a priori (for me) what I would say under given conditions, because that depends on the actual situation: if it actually rains and I'm just halluzinating that it doesn't, the closest world where it's raining is the actual world, and since I don't assent to "it's raining", it follows that I'm not disposed to assent to "it's raining" if it's raining.
What I would say under certain conditions mostly depends on what I would believe under these conditions. So we can define another, more interesting function for each sentence: a function from conditions to the truth value I'm disposed to assign to the sentence under the condition that I believe that the conditions obtain.
Related functions are determined by my conditional attitudes: I assign relatively high credence to the proposition that "it's raining" is true conditional on the proposition that it's raining; I intent to assent to "it's raining" conditional on the proposition that it's raining. Those three functions are related, but not identical: Perhaps in the closest world where I believe that I'm killed if I assent to anything I would not assent to "I'm killed if I assent to anything", and I don't have the corresponding conditional intention, though I do have the conditional belief. And if it is possible to believe that 7+5=10, I might assent to various sentences under the condition that I have this belief, yet I don't have the conditional attitudes, for I assign zero credence to the proposition that 7+5=10.
Anyway, what happens with conditions to which I assign very low but positive credence, like that I'm Leverrier's wife in 1848? This is not something I can absolutely rule out. It could turn out that I'm just halluzinating right now and that all my current memories are false, because I really am Leverrier's wife.
Part of the problem why I find it hard to say what I would say if that turned out is that it depends on exactly what is supposed to turn out: do I have to assume that I am in exactly the same position in which Leverrier's wife was in 1848, including living in a world without computers, cars and phones, and including not even believing or seemingly recalling that there are computers, cars and phones? That's very hard to assume. I'm almost inclined to say it can't possibly turn out that I'm wrong not only about almost everything I believe about the world but also about having and having had those beliefs about the world. Nevertheless, there are possible worlds where I have exactly the same mental states that Leverrier's wife had in 1845, so at least the what-I'm-disposed-to-assent-to function might be defined. But it seems to me that this part of the function is not very interesting. It certainly tells us little about the meaning of my words. (Consider a speaker of Inverted English, where every English sentence has the opposite truth value: I'm disposed to utter all kinds of sentences on the condition that I'm a mental duplicate of such a speaker, e.g. "7+5=10", but that doesn't tell us much about what I actually mean by my sentences.)
For many conditions and sentences, my conditional beliefs and intentions are similarly undetermined, and the dispositionally defined function similarly uninteresting. What would you say if it turns out that you're a rock? that you're a brain in a vat? that there are no people?
Interestingly, this seems to depend not only on the conditions but also on the sentences: If I suppose I now wake up and find that all the time I've been a brain in a vat and that the actual world is rather different from how I thought it is: there are human people, but with a quite different history; there are things that look like houses and water, but their physical properties and constitution differs in many ways from what we believe of ordinary houses and water, what would I say (to myself, or to a friend who just woke up with me) when talking about this surprising reality? I would be pretty confident in applying terms like "round", "liquid", "people", "watery", "space"; but I would avoid "water", "the 19th century", "Great Britain", "arthritis".
I think it is unfortunate that two-dimensionalists often focus on how we (would) describe certain counter-actual situations, with ourselves at the center. It seems to me that an equally important component at least of primary intensions for public language is contributed by our expectations towards other people's utterances. I have no clear opinion about what I would say if it turns out that I am the bus driver in front of me, but I have quite clear expectations about what he will (and should, and won't and shouldn't) say given that he believes such-and-such.