I’m guessing we disagree less than you think we do. Likely, it would help anyone interested in understanding this that the object of these unusual verbs is a noun whose location is its important feature. This is, as you like to point out, a semantic issue, not a grammatical one.

I don't think a semantic versus syntax argument is really the issue here.

We understand that the object of {ghoS} and its ilk is a location. We don’t have to mark it grammatically with {-Daq}, and if we do mark it with {-Daq}, we potentially give the noun a role that is not the object of the verb.

Not quite. We don't have to mark it grammatically with -Daq, and if we do mark it with -Daq we potentially confuse the reader or listener as to whether we're talking about an object or a noun in the pre-object position.

If I say DujDaq ghoStaH, I know whether I mean that the subject is approaching the ship or on the ship and approaching, but the meaning is ambiguous to anyone else. The role of DujDaq doesn't change; it's just ambiguous.

Such ambiguities happen all the time, though, and shouldn't cause us worry. Context will make the correct meaning clear. And often probability will play a role. It's possible to interpret DujDaq vIghoStaH as I am on the ship, approaching it (an elided 'oH it as object), but it's unlikely I'm going to elide the object in this way if the chance of misunderstanding is high. DujDaq 'oH vIghoStaH.

-- 
SuStel
http://trimboli.name