A Tour of Greek Morphology: Part 11
In part 10, we looked at some new active forms. Now it’s time to look at the corresponding middle forms.
In the middle forms, there is no change in the vowel and so it doesn’t need to be included in the distinguisher. In this sense, we really only have one distinguisher paradigm for all these forms in the middle.
However, if we were contrasting against the active forms as well, we could identify a PM-6, PM-7, PM-8, and PM-9 paired up with PA-6, PA-7, PA-8, PA-9:
But the common endings for the μι verbs are very clear. Here they are alongside our previously reconstructed endings for the previous middle paradigms:
|2SG||-σαι||ε σαι > ῃ|
This not only provides clear support for the ε+σαι reconstruction of the ῃ MID 2SG form but also makes clear how the ω verbs (both barytone and circumflex) use the same endings as the μι verbs but with the ε/ο vowel (the so-called thematic vowel) attached to the stem before the ending. In the middle, this is the only difference (slightly obscured when ῃ is used in the 2SG).
As mentioned in part 9, there are some tantalising patterns here: the αι in 5 out of 7 cells; the μ/σ/τ in the 1st/2nd/3rd person.
The appearance of μι in the ACT 1SG is particular interesting because we now have a μι/μαι contrast in the 1SG between active and middle which exactly mirrors the οντι/ονται contrast in the 3PL.
One might well question why the active 2SG and 3SG forms don’t end in σι and τι to mirror σαι and ται. Or why the active infinitive isn’t σθι. Or why the 1PL and 2PL have only a vague relationship between the active and middle. And we still have the question of where the alpha in the ACT 3PL ασι(ν) ending comes from. We’ll touch on some of these questions in the next post and we will reveal some more historical and dialectal patterns.
But it is again worth reiterating that the primary role of a distinguisher is not to be decomposable but merely to discriminate meaning. That there are patterns between the distinguishers at all is not a fundamental requirement of the role they play in conveying information. There may be historical reasons for the patterns (as we’ve already seen) and learnability pressures that favour them (or even conspire to introduce them over time) but we should not expect them and therefore view their absence as any kind of defect or irregularity.