Part thirty-one of a tour through Greek inflectional morphology to help get students thinking more systematically about the word forms they see (and maybe teach a bit of general linguistics along the way).
In the previous post we went through and made sure we had all our imperfect active indicative endings covered ready for counting. We still had some ambiguities, though, so we need to use rules based around the lemma to dismabiguate. We can then apply those rules to generate our data for counting.
Our disambiguation rules are:
| 2SG:-ης or 3SG:-η | is |
IA-5 if lemma ends in -ω IA-9 if lemma is ἵστημι IA-9b if lemma is φημί |
| 1PL:-αμεν or 2PL:-ατε or 3PL:-ασαν | is |
IA-9 if lemma is ἵστημι IA-9b if lemma is φημί |
| 2PL:-ετε | is |
IA-1 if lemma ends in -ω IA-7 if lemma ends in -μι |
| 2SG:-εις or 3SG:-ει | is |
IA-2 if lemma ends in -ω IA-7 if lemma ends in -μι |
| 1PL:-οῦμεν or 3PL:-ουν | is |
IA-2 if lemma ends in -έω IA-3 if lemma ends in -όω |
| 1SG:-ων or 1PL:-ῶμεν or 3PL:-ων | is |
IA-5 if lemma is ζάω (should really just lemmatise ζήω) IA-4 if lemma is other -άω |
| 2SG:-υς or 3SG:-υ | is |
IA-3 if lemma ends in -ω IA-6a if lemma ends in -υμι (or if form not -ους / -ου) IA-8 if lemma ends in -ωμι |
| 1SG:-υν | is |
IA-2 if lemma ends in -έω IA-3 if lemma ends in -όω IA-6a if lemma ends in -υμι (or if form not -ουν) IA-8 if lemma ends in -ωμι |
| 1PL:-ομεν | is |
IA-1 if lemma ends in -ω IA-8 if lemma ends in -μι |
| 1SG:-ειν or 3PL:-εσαν | is |
IA-7 if lemma is τίθημι or ἵημι (not an issue in SBLGNT) IA-11 if lemma is εἶμι IA-11-COMP if lemma ends in -ειμι |
| 2SG:-εις | is |
IA-2 if lemma ends in -ω (not an issue in SBLGNT) IA-7 if lemma is τίθημι or ἵημι (not an issue in SBLGNT) IA-11 if lemma is εἶμι IA-11-COMP if lemma ends in -ειμι |
| 1SG:-ην | is |
IA-7 if lemma is τίθημι or ἵημι IA-9 if lemma is ἵστημι IA-9b if lemma is φημί |
| 2PL:-ῆτε | is |
IA-5 if lemma ends in -ω IA-10-COMP if lemma ends in -μι (not an issue in SBLGNT) |
Encapsulating these rules into a Python script and running on our data, we now have an inflectional class for all 1,344 imperfect active indicative forms in the MorphGNT SBLGNT.
The output of my Python script looks like this:
010118 ἦν 3SG IA-10 εἰμί IA-10 ἦν _ ἦν
010125 ἐγίνωσκε(ν) 3SG IA-1 γινώσκω IA-1 Xε(ν) ἐγίνωσκ ε(ν)
010209 προῆγε(ν) 3SG IA-1 προάγω IA-1 Xε(ν) προῆγ ε(ν)
010209 ἦν 3SG IA-10 εἰμί IA-10 ἦν _ ἦν
010215 ἦν 3SG IA-10 εἰμί IA-10 ἦν _ ἦν
010218 ἤθελε(ν) 3SG IA-1 θέλω IA-1 Xε(ν) ἤθελ ε(ν)
010304 εἶχε(ν) 3SG IA-1 ἔχω IA-1 Xε(ν) εἶχ ε(ν)
010304 ἦν 3SG IA-10 εἰμί IA-10 ἦν _ ἦν
010314 διεκώλυε(ν) 3SG IA-1 διακωλύω IA-1 Xε(ν) διεκώλυ ε(ν)
010407 ἔφη 3SG IA-5/IA-9/IA-9b φημί IA-9b Xη ἔφ η
The columns are:
- the book/chapter/verse reference
- the normalized form
- the morphosyntactic properties
- the inflectional classes possible without disambiguation
- the lemma
- the disambiguated inflectional class
- the distinguisher pattern
- the theme (the value of X)
- the distinguisher
You can download the entire thing here.
We’ll use this to do our counts in the next post.
But before that, there are a couple of things we can check.
Firstly, are the disambiguated inflectional classes consistent for each lexeme?
There are five exceptions, all of which we raised in the previous post:
- ἔχω is variously IA-1 or IA-8 (the alternate εἴχοσαν for the 3PL)
- ἐρωτάω is variously IA-4 or IA-2 (the alternate ἠρώτουν for the 3PL)
- δίδωμι is variously IA-8 or IA-3 (the alternate ἐδίδουν for the 3PL)
- παραδίδωμι is variously IA-8 or IA-3 (the alternative παρεδίδουν for the 3PL)
- τίθημι is variously IA-7 or IA-2 (the alternative ἐτίθουν for the 3PL)
Notice they are all in the 3PL and, with the exception of the ἐρωτάω case are alternations between the thematic and athematic ending.
Secondly, is the value of X in our paradigm patterns consistent across a lexeme?
There seem to be four exceptions, three of which are to do with the augment:
- εὑρίσκω: ηὑρισκ or εὑρισκ
- μέλλω: ἠμελλ or ἐμελλ
- εὐκαιρέω: εὐκαιρ or ηὐκαιρ
So far we’ve glossed over the augment but we shall look at it in detail in a future post.
There is also
- συγχέω: συνεχε or συνεχυνν
which we previously brought up. This is not just an inflectional class difference but also a stem formation difference. We’ll talk a bit more about this in future posts, but for now it’s probably best though of as two distinct lemmas that are conventionally conflated under the single headword συγχέω, Notice also that συνέχεον is the one example of an uncontracted IA-2 in the SBLGNT.