Semantic externalism, or more broadly externalism about mental content, is the idea that propositions, intentions, attitudes, etc. derive their meaning from the external world as opposed to internally from the mind itself. Some of the most famous arguments for this view are Putnam's twin Earth thought experiment and Davidson's swampman thought experiment.
The most common, and widely accepted as being the most philosophically well founded, semantics of modal statements are possible worlds semantics which say that a modal statement, either something about necessity or possibility, derive their meaning from a set of possible worlds. If a proposition is true in all possible worlds then it is necessary, if it is true is some possible worlds then it is possible, and if it is true in no possible worlds then it is impossible.
David Lewis was a proponent of modal realism, the idea that possible worlds are all physically real. He argues that in a statement that begins with "In the real world," "real" in this context in just indexical and refers to our own world. Saul Kripke, who put down the foundation for modal logic with possible world semantics, rejected modal realism and argued that possible worlds are absolutely not physically realized but are instead just linguistic and philosophical tools.
If a philosopher accepts semantic externalism and accepts that possible worlds are the correct form of semantics for modal statements, is she then forced to commit to a belief in modal realism?
If those two assumptions aren't strong enough to lead directly to modal realism by themselves, what happens if she also accepts a general metaphysical realism?
My assumption is that there is no way to get around modal realism if you believe (1) external things are real, (2) external things are what give propositions meaning, and (3) possible world semantics are what give modal statements meaning. Most of all, I am interested in any papers that deal with this question.