For example, suppose I enter "πράσσουσα" and it outputs πραάͳοντσα or even, πρααͳ-ο-ντ-σα (root, ablaut, participle marker, feminine). Or I put in πᾶς and it outputs πάαντ-ς (root, 3rdNS), or εἰσώμεθα outputs ἐϝεϳδσαόμεθα or even ἐ-ϝεϳδ-σ-α-ο-μεθα (augment, root, imperfect, past, subjunctive, 1st-person-plural.)? I ask this with the intentions to abuse the language, etymologically spell some words to eliminate ambuguity, reconstruct inflections that shouldn't exist, and enable myself to create new words, and to possibly use in writing fiction. For example, I might try to hastily generate a fake dialect in Greek, specifically marking places that assimilate dental stops not completely dissolving into sigma (τς, δς, θς to ϡ or ϻ, instead of lone ς, to complete the trilogy with ψ and ξ. I would need to know rather quickly whether a word in third-declension nominative singular (& dative plural) that I find with a final-s is made from this assimilation or not.) Another possibility would be a fake dialect that has different rules for contraction and lengthening, so I would need to know quickly and easily where I am encountering an invisible silent yot or waw, and have a better grip on spirious and genuine diphthongs. Can such a device be found, which etymologically and inflectionally dissects words in ancient Greek?

  • 4
    Given that there are a lot of words whose Proto-Greek reconstructions are at best ambiguous, at worst completely unknown, probably not. Plus it’s not really clear what ‘Proto-Greek’ means – it’s not like it’s a single moment in time. For example, while the form you write as the PG form of πράσσουσα is definitely not right for any period (there was never an /aa/ in the word), it could equally well be claimed to be prāččontsa, prākjontja or pr̥a̯kjontia̯ – all those are Proto-Greek forms at one stage or another. Apr 7, 2022 at 23:31
  • Okay. I wrote it like that assuming that the construction r̥+h₂ became ra+a became rā. But the h probably fixed itself a vowel before the r
    – Oron61
    Apr 8, 2022 at 3:10
  • Are you aware of Constructed Languages? Your question may be a better fit over there. Apr 8, 2022 at 8:22
  • 2
    There are errors in your reconstructions -- the long vowel of πᾶς results from compensatory lengthening, so wouldn't have cooccurred with the nasal; εἰσώμεθα isn't a possible form of any verb (except εἰσίημι), and subjunctive forms have no augment. Anyway there's no way of automating this reliably -- you just need to learn more Greek historical linguistics.
    – TKR
    Apr 8, 2022 at 11:40

1 Answer 1


Find? Probably not. There hasn't been much desire for such things.

Make? Presumably. There are systems like Morpheus that can analyze the morphology of a word, and from there it's feasible to break it down into morphemes in whichever way you like. Then you just need to pick a reconstruction you like of Proto-Greek (ideally one used by a solid etymological dictionary), and code in the Proto-Greek form of each morpheme.

Or, alternately, you could come up with a table of Proto-Greek forms of all the affixes you care about, then look up roots in the aforementioned etymological dictionary. This is probably more feasible.

For either of these options, you'll need to deal with the fact that many Greek roots have uncertain origins. Beekes is (in)famous for ascribing all sorts of words to "Pre-Greek", but it's true that there often isn't a solid link back to PIE. So you'll have to decide how far you want to try to extrapolate these back.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.