Chiattelli, Edoardo
In ancient Greek, past tenses of verbs starting with a consonant are prefixed with the vowel epsilon,
but some verbs in Homer show an eta instead. These forms were linked to Vedic instances of long
augments by part of the relevant literature. Other scholars preferred to explain the Greek instances as
inner developments, while others denied any philological validity for the long augment as an actual
morpheme, and rather explained its presence in some Homeric verbs as the product of analogical
processes.
My thesis focusses on specific Homeric forms within the heterogeneous set of evidence offered by
ancient Greek. I take these long-augmented verbs to be artificial creations of the Homeric language.
This line of interpretation is supported not only by considerations of historical phonology and
morphology, but also through an innovative formula-based method. Its aim is to describe, through an
analysis of the Homeric traditional language, the possible reasons and dynamics for the creation in
the Kunstsprache of artificial long-augmented forms. More specifically, it might be possible to
explain them as part of analogical modifications of pre-existent formulaic patterns, so as to provide
an accurate description of how (and why) their artificial structure was used as a metrical alternative
to their counterparts in the spoken language.
The focus of my thesis is on ἠείδη ‘she/he knew’ (and ἠείδης ‘you knew’), ἤϊκτο ‘she/he resembled’,
ἀπηύρα ‘she/he took away’, and the trisyllabic forms of the imperfect of εἶμι, i.e. ἤϊα ‘I went’, ‘ἤϊε
‘she/he went’, ἤϊσαν/ἤϊον ‘they went’. A detailed analysis of historical morphology shows that none
of them can be assumed to be the result of linguistic processes in the ordinary language. This is also
confirmed by their attestations limited to Homeric diction or later poetry imitating it. Furthermore, I
show through the formula-based method how most of these long-augmented forms are part of
analogical modifications derived from formulaic patterns, which contain forms of the same paradigm
but without long augment (i.e. ᾔδεε ‘she/he knew’, ἐϊκυῖα/ἔϊκτο ‘resembling’/‘she/he resembled’,
ἴσαν ‘they went’). This suggests that long-augmented forms are used primarily as metrically
functional alternatives, which is a feature typical of artificial creations in Homer.
Since the results of the method confirm the artificial nature of these forms, they cannot be compared
with the Vedic data, nor can their long augment be deemed a genuine morpheme of ancient Greek. It
is rather the product of analogical processes within the Homeric Kunstsprache. In particular, I provide
a possible narrative for the origin and analogical use of a Homeric long augment in the pluperfect of
οἶδα and imperfect of εἶμι, while explaining the initial long vowels of ἠειδ- and ἤϊκτο as analogical
temporal augments used by the Homeric poets for metrical purposes. As for ἀπηύρᾱ, this
morphologically controversial form is best explained as another analogical use of temporal
augmentation, applied by the Ionian bards to original *ἀπεύρᾱ despite its metrical irrelevance. My
method shows how forms like ἀπηύρᾱ, which had become extraneous to the Ionian bards, could
nonetheless undergo analogical reshaping through the influence of structural connections among
formulaic patterns.
Advisors
Meissner, Torsten
Clackson, James
Date
2022-04-22
Awarding Institution
University of Cambridge
Qualification
Doctor of Philosophy (PhD)
Posts
Stumble It!
