I made a bold claim earlier:
If I am right, the patois of possible worlds is a dispensable manner of speaking: we can make every [modal] point we want to make without engaging in possible worlds talk. What I just said is not perfectly obvious and there may be counterexamples.
Here is a candidate counterexample that I borrow from Barbara Vetter, 'Can' Without Possible Worlds, 22:
(CC) Someone can see us.
This is an alethically or as Vetter would say "dynamically" modal statement. It is modal but not expressive of either epistemic or deontic modality. Interestingly, (CC) is susceptible of being read either de re or de dicto:
(DR) There is a person who can see us.
(DD) It is possible that someone see us.
(DR) commits us to an actual person who is able to see us. (DD) does not. The first entails the second, but the second does not entail the first. So the two readings are non-equivalent. Suppose that no actual person can see us. Suppose, that is, that no actual person has both the ability to see us and is positioned in such a way that he can exercise his ability. Even so, it is 'surely' possible that there be such a person. There could have been a person, distinct from every actual person, who sees us.
So (DD) is a true alethically modal statement whose truth is not grounded in, or made true by, a power or ability of any actual item.
(DD) would thus appear to be a counterexample to Vetter's "potentiality semantics" according to which "all dynamic modality is de re . . . ." (22) It seems that (DD) expresses a 'free-floating' possibility, one not grounded in any actual concrete thing's power or potentiality. If so, then 'possible worlds' talk might not be wholly dispensable.
One response to the putative counterexample is rejectionist: Vetter toys with simply rejecting (DD) as a statement of dynamic modality by suggesting that it is really an example of epistemic possibility (23).
I fail to see, however, how (DD) could be construed as epistemic. The idea is not that, for all we know, someone can see us, but that it is really possible, apart from our knowledge and ignorance, that there be someone who can see us. In the actual world, no one can see us now. But 'surely' there is a possible world, very much like the actual world, in which someone can see us now. If there is this possibility, it is real, not epistemic.
But there is another line of rejection that Vetter does not clearly distinguish from the first. And that is simply to say that her topic is dynamic modality, the sort of real modality that we encounter in actual changing things, and not real modality in general. 'Dynamic' is from the Greek dynamis which in Latin is potentia, whence our 'potency' and 'potentiality.' The second way of rejection, then, is to dismiss (DD) as simply off-topic.
But then her thesis is less interesting: it is not the thesis that all alethic modality is de re, but that only the modality of actual concrete things subject to change is de re. If this is her thesis, then it seems we need possible worlds to accommodate such de dicto possibilities as (DD).