David Embick: The Latin Conjugation

[Martha McGinnis]

We would like to thank Andrew Carstairs-McCarthy (henceforth ACM) for
his detailed comments on the synopsis of our work on Latin that Morris
Halle posted last week.  Our comments here will focus on part 1 of
ACM's message, which deals specifically with topics touched upon by
Morris.  Part 2 of the message will have to be dealt with on another
occasion.

We are grateful to ACM for reminding us of Lieber's attempt to treat
the behavior of these verbs phonologically.  However, we do not
believe that there is an additional question of "...whether or not
such a rule can be related to some phonological universal".  We
believe that the operation of this synchronic vowel-deletion rule
allows for a cleaner analysis of the Latin verbal system.  The
significance of deletion in understanding morphological systems has
been evident at least since Jakobson's (1948) analysis of the Russian
conjugation.  It has been shown that this deletion rule plays a major
role in the phonology of other Slavic languages as well
(cf. e.g. Rubach's book on Slovak, and Czaykowska-Higgins' (1988) MIT
Thesis).

ACM then raises two further questions about:

(a) coverage of relevant facts,
(b)  compatibility with independent facts about inflectional systems

Point (a) covers the behavior of _capio:_ and _du:co:_ type verbs
phonologically, and their relationship to Conj. IV verbs like
_audio:_.

The apparent vacillation throughout the history of Latin which
confused _capio:_ (`mixed') and Conj. IV verbs does speak to certain
issues in the development of Romance, but is not an essential
ingredient of an analysis of the Latin Conjugation viewed
synchronically.  However, it should be noted that on our account,
mixed verbs and Conj. IV verbs differ only in the length of the theme
vowel, whereas _du:cere_ type verbs are hypothesized to have a
distinct vowel underlyingly.  The confusion noted by ACM thus arises
among the verbs whose theme vowels are identical in terms of features
and is thus phonetically very plausible.

ACM's observations about the relation between the phonetic (syllabic)
nature of the verb stem and its assignment to `mixed' vs. 4th
conjugation are well-taken.  It is to be noted, however, that they say
nothing about the contrast between mixed vs. 3rd conjugation verbs
(_capio:_ vs. _du:co:_), which was at issue in Halle's original note.

David Embick and Morris Halle