The root is ἑλ-, and, per Smyth (par. 435), augmented ε is η (cf., e.g., ἤλασα < ἐλάσαι); so, why is it ει here?
It didn’t begin with a vowel originally, as you can see from the spiritus asper. This is why the augmentum temporale is different from the “normal”. We see the same elsewhere, of course, as e.g. in εἶχον (initial aspiration lost in accordance with Grassmann’s law).
Smyth sec. 431 is consistent with Timothée:
- Some verbs beginning with a vowel take the syllabic augment because they formerly began with a consonant. Thus,
ἄγνυ_μι break (Ϝάγνυ_μι), ἔα_ξα, aor. pass. ἐά_γην.
ἁλίσκομαι am captured (Ϝαλίσκομαι), imperf. ἡλισκόμην, aor. ἑά_λων (with temporal augment) or ἥλων.
ἁνδάνω please (Ϝανδάνω), aor. ἕαδον (Ionic).
ἀν-οίγω open (Ϝοίγνυ_μι), imperf. ἀν-έῳγον.
ἐάω permit (σεϜαω), εἴων, εἴα_σα, εἰά_θην.
ἑζόμαι sit (for σεδιομαι), εἱσάμην.
ἐθίζω accustom (σϜεθίζω, cp. 123), εἴθιζον, εἴθισα, εἰθίσθην
ἐλίττω roll (Ϝελίττω), εἴλιττον, εἴλιξα, εἰλίχθην.
ἕλκω or ἑλκύω draw (σελκω), εἷλκον, εἵλκυσα, εἱλκύσθην.
ἕπομαι follow (σεπομαι), εἱπόμην.
ἐργάζομαι work (Ϝεργάζομαι), εἰργασάμην.
ἕρπω creep (σερπω), εἷρπον.
ἑστιάω entertain (Ϝεστίαω), εἱστίων, εἱστία_σα, εἱστιά_θην.
ἔχω hold (σεχω), εἶχον.
ἵημι send (σισημι), aor. du. εἷτον for ἐ-ἑ-τον, εἵθην for ἐ-ἑ-θην.
ἵστημι put (σιστημι), plup. εἱστήκη for ἐ-σε-στηκη.
ὁράω see (Ϝοράω), ἑώρων, ἑώρα_κα or ἑόρα_κα.
ὠθέω push (Ϝωθέω), ἐώθουν, ἔωσα, ἐώσθην.
ὠνέομαι buy (Ϝωνέομαι), ἐωνούμην, ἐωνήθην.
εἶδον saw, 2 aor. of ὁράω (for ἐ-Ϝιδον).
εἷλον took, 2 aor. of αἱρέω (for ἐ-ἑλον).
http://www.perseus.tufts.edu/hopper/text?doc=Smyth+grammar+431&fromdoc=Perseus%3Atext%3A1999.04.0007
The etymologies of both αἱρέω and the suppletive aorist εἷλον are apparently uncertain, though a few theories have been proposed. Timo, do you have any thoughts about this?
So, as concerns augmentation, does aspiration at the beginning of a root always counts as a consonant? (I understand that this does not cover cases like εἰργασάμην but would like to extend the rule as much as possible).
as concerns augmentation, does aspiration at the beginning of a root always counts as a consonant?
I would not say “always”, but as a synchronic matter (looking at ancient Greek as it is attested in the texts), that’s true in many cases.
As a diachronic matter (looking at the reconstructed history of Greek), aspiration
of the initial vowel in many cases reflects a consonant (generally, σ or digamma or both) that has been weakend. As a result, what was originally ε- augment (“syllabic augment”) either remains in place or is contracted with the initial vowel resulting from the weakening of the original consonant(s).
Thanks a lot!
As I see it, ἑλεῖν has a good, plausible Indo-European etymology, cognates seen particularly in Germanic (to sell, Sw. sälja) but also in Celtic. More uncertain is the possible connexion of these with Latin consulere.
But αἱρεῖν is truly a conundrum.