Valérī vs. Válerī

On pg. 221 I mention that according to Nigidius Figulus (apud. Aul. Gell. 13.26.1) the vocative of Valerius was Válerī, but the genitive was Valérī. Should we take this testimony seriously and if so how is it to be interpreted?

First, it is interesting to note that Gellius himself clearly did not accent the vocative of Valerius in this way since he says that anybody who pronounced the vocative as Válerī would be laughed at (Sed si quis nunc Valerium appellans, in casu vocandi, secundum id praeceptum Nigidii, acuerit primam, non aberit quin rideatur). Presumably this means that Gellius said Valérī for both the genitive and the vocative.

From the historical point of view, however, the vocative Válerī could make sense: since the change of *-ije to -ī (the fīlī rule, Weiss 2009:122) was probably Proto-Italic, this word would have been a trisyllable at the time of the emergence of the Classical Latin stress rules.

On the other hand, explaining the position of the stress in the genitive as phonologically regular is difficult: if in the genitive *-ijī contracted to -ī before the emergence of the CL stress rules then we should expect Válerī just like the vocative; on the other hand, if *-ijī contracted to -ī after the emergence of the CL stress rules we would still expect Válerī since *Valerijī would have been stressed on the the first syllable by the facilius rule still operative in Plautus' day. Thus it seems that Valérī can only be the result of a columnar accent.

This still leaves open the possibility that the vocative Válerī simply escaped the analogy until sometime after Nigidius and before Gellius. But there is perhaps another possible explanation. Nigidius says: in casu vocandi summo tonost prima, deinde gradatim descendunt. He seems to be describing a gradual (gradatim) tonal downglide. Could Nigidius be describing the so-called vocative chant? In English and other languages there is a special tonal contour for calling people or animals, e.g. to supper. In English this is typically realized as a H(igh) tone associated with the primary stress of the word followed by a M(id) tone on the following unstressed syllables. See Hammond 1999:156.

See Hammond, Michael. 1999. The phonology of English. A prosodic optimality-theoretic approach. New York: Oxford University Press.