Eeli Kaikkonen wrote:So, your referring to semantic roles. Do you mean that the subjective/objective (or agentive/patientive?) are possible explanations for genitive only if the phrase or its paraphrase has an agent in the strict sense of the word - that e.g. experiencer isn't enough? Why? As far as I can see this wouldn't explain the examples used by Smyth.
Well, a couple points...
1) First note that Smyth's description independent of his examples fits precisely with what I'm claiming: "The Subjective Genitive is active in sense" = agentative; "The Objective Genitive is passive in sense" = patientive.
2) I think Smyth's examples are pretty terrible, at least the subjective genitive ones. Smyth uses a φόβος example for the subjective genitive, but goes on to say later that it is actually the objective genitive that "is very common with substantives denoting a frame of mind or an emotion." In my view, the so-called subjective genitive τῶν βαρβάρων φόβος the fear of the barbarians is more easily conceptualized as involving a possessive schema. That is, it isn't parallel to "the barbarians were afraid." The phrase makes far more sense as an extension of the normal possessive use of the genitive, where emotions or experiences are conceived as possessions. The same goes for τὸ ὀργιζόμενον τῆς γνώμης their angry feelings.
The fact of the matter is that is that Smyth's examples don't match his description (with the exception of βασιλέως ἐπιορκία the perjury of the king). Either his examples are wrong or his description is wrong. They don't really match.