Yeah, I realize I didn't quite explain that as well as I could (and frankly I'm still not entirely sure how to best do it). I'm not looking for incorporation, but was using "absorption" (a term I actually have seen elsewhere) to refer to how a participle or another part of speech essentially subsumes within itself one of its core arguments implicitly, not explicitly as is the case with incorporation.
For instance, your standard active participles "absorb" their subject by allowing them to drop out of the relative clause the participle is replacing. The active participle then ends up directly modifying its subject in the main clause. Its object, which is not absorbed, remains in place modifying the participle.
Let's look at an example. Hopefully it'll make more sense this time.
"The man who [he] sees the dog"
Here we have a finite transitive verb in a relative clause with both subject and object. (Yes, in English the verb in the relative clause technically agrees with the relative pronoun, but bear with me)
If we convert that verb to an active participle, its subject "he", or "who" as may be the case in English, is dropped. The participle then attaches itself to the antecedent of that pronoun in the relative clause.
"The man seeing the dog".
This participle retains many of the qualities of the original finite verb, such as the free ability to either mark or not mark its direct object (the man seeing the dog, or the seeing man), although with a change in meaning.
"The man who [he] was seen (by the dog)"
Same deal, with a passive verb. The relative clause can be reworded using a passive participle; the verb's subject drops out and the verb attaches itself to its antecedent in the main clause. And just as the "by the dog" phrase is optional in the relative clause, it remains so when dealing with a passive participle.
"The man seen (by the dog)"
Now, what I'm looking for is a participle that attaches itself to its object. And, since the subject in a Nom/Acc language is mandatory, presumably the subject will be a mandatory argument for the participle as well.
"The man whom the dog saw [him]"
"The the-dog-saw man"
which would be glossed as something like
dog-NOM see-PARTICIPLE.PAST.ACTIVE.OBJECT man-NOM
Notice that the noun the participle is modifying is actually its direct object.
(I would have said "The man the dog saw", except that would be a little confusing, given how English can simply drop the "that" in relative clauses).
THC's examples aren't quite what I'm after. Show me an active participle that can incorporate its agent while modifying its patient, then I'll be impressed
Zompist, your first example does seem to work. You're subordinating "he likes" (an active verb and its agent) to "things" (the verb's patient). The second is more like a typical participial construction, where you're subordinating "love drinking wine" (an active verb and its patient) to "people" (the verb's agent). This certainly does seem to be an example, even though they're not true participles.