No abstract

The wide range of interpretations of aoristic and imperfective aspect in Ancient Greek cannot be attributed to unambiguous aspectual operators but suggest an analysis in terms of coercion in the spirit of de Swart (Nat Lang Linguist Theory 16:347-385, 1998). But since such an analysis cannot explain the Ancient Greek data, we combine Klein's (Time in language, 1994) theory of tense and aspect with Egg's (Flexible semantics for reinterpretation phenomena, 2005) aspectual coercion approach. Following Klein, (grammatical) aspect relates the runtime of an eventuality and the current time of reference (topic time). We claim that these relations can trigger aspectual selection restrictions (and subsequent aspectual coercions) just like e.g. aspectually relevant temporal adverbials, and are furthermore susceptible to the Duration Principle of Egg (Flexible semantics for reinterpretation phenomena, 2005): Properties of eventualities must be compatible with respect to the duration they specify for an eventuality. The Duration Principle guides the selection between different feasible coercion operators in cases of aspectual coercion but can also trigger coercions of its own. We analyse the interpretations of aorist and imperfective as cases of coercion that avoid impending violations of aspectual selection restrictions or of the Duration Principle, which covers cases that are problematic for de Swart's (Nat Lang Linguist Theory 16:347-385, 1998) analysis.

In this paper we present the lemmatizer that we developed for Ancient Greek: GLEM. As far as we know, GLEM is the rst publicly available lemmatizer for Ancient Greek that uses POS information to disambiguate and that also assigns output to unseen words, words that are not yet in the lexicon. As the basis for the lemmatizer we used an existing memorybased learning tool, Frog, that was originally developed for Dutch and that we converted to work for Ancient Greek. As the results of Frog on Ancient Greek were rather modest, we used Frog to create a smarter lemmatizer, GLEM, that uses a lexicon look up in addition to the memory-based tool Frog. We evaluate and compare the performance of GLEM against the Frog lemmatizer and the already existing CLTK lemmatizer and observe that GLEM achieves the highest accuracy of 93% on an unseen test corpus sample. GLEM's look up component overcomes the di culty of a relative small training set in combination with a morphologically rich language, while the memory-based learning component enables GLEM to handle unknown words.

The paper offers a formal account of the discourse behaviour of participles, which to some extent behave like main clauses in having semantically undetermined relations to their matrix clause, but which should nevertheless be integrated into the compositional semantics of complex sentences. The theory is developed on the basis of Ancient Greek participles and offers an account of their syntax, semantics and discourse behaviour (focusing on the temporal dimension of discourse), integrating Lexical-Functional Grammar, Compositional DRT and Segmented DRT using Glue semantics.

No abstract