The Rough Beginnings of Greek Morphology

This is the beginnings of the morphology work I’ve done in Language Explorer. Nouns are for the most part done, though I have some tidying up to do. I’m in the process of analyzing Adjectives & Adverbs derived from Adjectives and after that I’ll be moving to Verbs.

The Labels (P I-III) denote the Noun Declensions. The “P” stands for “Productive” – that is, these are productive classes. Thus (P I) is 1st Declension, (P II) is 2nd, and (P III) is 3rd, though the declension system is actually far more complicated than that. In a sense, the idea of “three declensions” is still true, there are however what we could term multiple subsystems that arise from the vowels that stems end in as well as the patterns of the accent system, but for our purposes, that is less relevant here (and what you see below and in the PDF is slightly simplified). The rest of the abbreviations should be self-explanatory. There are also some places where the there is apparent discrepancies with the suffix morphemes:


*δρέπανο ν
*δρέπαν ς
n (P II) neut n:Case/Num

But this isn’t really a discrepancy. The first line of the and word and its suffix marks the form in the word being parsed and the second is the form being used as a the head work in my lexicon. With regard to the lack of the omicron, I’m following Gerhard Mussies in identifying two different base forms of the stem to which the suffixes are attached (these stem differences are morphophonemic issues). In general, I’ve followed Mussies as much as I possibly could. His morphological work is the best and most up-to-date linguistically that I’ve seen for Koine Greek, though there are other high quality works for Homeric and Classical Greek. The changes I’ve made away from Mussies have been more for pragmatic (in a non-technical sense) reasons than theoretical ones–mainly driven by the realities of the FLEx parser.

Just remember this is a work in progress.

Anyway, here’s a link to the PDF:

Productive Noun Morphology

I also have XML files of the same data.