Page 501 - Encyclopedia of Philosophy of Language

have: imperative and subjunctive. All 10 voice-aspect systems have an optative, an infinitive, a participle, and at least one indicative: the future has no forms with augment (i.e., a prefixed vowel) and past tense endings, the aorist (perfective) none without augment and with present endings, but the other two have both. Later grammarians (e.g., Dionysius Thrax and Apollonius Dyscolus) detached the participles entirely from the verb (because they satisfied both the [+tense] and the [+case] requirements which otherwise dis- tinguish verbs from nouns). Two other forms are now- adays associated with verbs, the verbal adjectives or 'verbals' in -tos, -ts, -ton (like lekton, aisthston, noSton, favorite Stoic terms) which more or less resemble adjectives in -able (sayable, perceivable, thinkable), and -teos, -tea, -teon (more or less like the Latin pas- sive periphrastic, 'which must be said, perceived, thought'). All of the ancients, if they mention these at all, treat them as derivatives, not part of the conju- gation. In each of the regular moods and tenses, most verbs have eight forms (first, second, and third person singular and plural, second, and third person dual), but imperatives lack a first person (for which the sub- junctive is used). Some verbs are impersonal, and have only a third person singular (and it cannot have a nominative subject), meaning things like 'it is necess- ary,' 'it is best,' 'it is of interest,' etc.
Now can be considered which of these things the Stoics dealt with, and how. The treatment of predi- cates (ellipS lekta, 'incomplete sayables' or kat- SgorSmata (instead of katSgoriai) differs sharply from Aristotle's, most of whose predicates required a verb 'is,' in that they count verbs only, separated into four classes by the features [+ nom] and [± oblique]. Verbs which are personal and intransitive (requiring no oblique case) are called sumbama. If they need an object of some kind, they are hsttonIelattonS sumbama (less than sumbama). If they are impersonal, i.e., do not allow a nominative case (i.e., subject) or more than one oblique case (usually dative), they are par- asumbama. And if they occur with two obliques but no nominative, they are elattonjhstton Sparasumbama ('less than p'). But if they occur with two nominatives, but require no obliques (like 'is,' 'becomes,' etc.), the Stoics had no known name for them. These types continued to be distinguished, but new names were adopted for them.
The Stoics also introduced names for the voices: sumbamata in the active (governing accusative, dative, or genitive) are ortha (upright), a term which is some- times mentioned by later grammarians, but is nor- mally replaced (by energetika or drastika 'active'); the same verbs in the passive, capable of taking hupo (by) with a genitive ('of agent') are called huptia (prone), again normally replaced later by another word (pathstika, 'passive'); and verbs which take neither of those two constructions, being intransitive, whether active or middle in form, are called oudetera ('neither).
Dionysius Thrax includes both epoiSsamSn (aorist middle for us) and pepoitha (second perfect active) in his category mesotSs (middleness), and Apollonius also treats second perfect actives as middle, but appar- ently no other intransitive actives are called that; so the later 'middle' is not the same as Stoic oudetera. Nor is it equal to the Stoic antipeponthota freflexively influenced), which are middle verbs with some ref- erence back to the subject, as keiretai 'he gets a haircut' (not 'he cut his own hair') from keirei 'he cuts (some- one's hair).'
Turning to the treatment of tense by the Stoics, there is a curious passage in the Scholia to Dionysius Thrax (on a passage where Dionysius speaks of the three kinships—sungeneiai—present to imperfect [par- atatikos], future to aorist, and perfect to pluperfect):
The Stoics say the present should be called 'present imper- fect' (paratatikos, 'extended' which in Dionysius and later means simply the imperfect tense) for when you say poid 'I am doing,' you imply that you were doing it before and will be doing it after the moment of speaking; and the imperfect should be 'past imperfect' for similar reasons... And the perfect tense (parakeimenos) is called [sc. 'by the Stoics'?] 'present suntelikon,' and its past is the pluperfect.
But suntelikon cannot be translated here as 'perfect' (to give a neat match 'present perfect' and 'past per- fect' corresponding to 'present imperfect' and 'past imperfect') because everywherethat suntelikosorsun- teleia is used to refer to an aspect (four or five times, in all) it refers to the aorist, not the perfect (some of these are infinitives, others imperatives, and one, hixon, is indicative). A possible conclusion is that the perfect is 'present perfective,' the aorist is 'indefinite perfective' [aoristos means 'indefinite'] and the plu- perfect is 'past perfective,' while suntelikos by itself includes all forms of the aorist system; but it is not known what the nonindicative forms of the perfect system were called by the Stoics. As for the generic term for aspect, it is clearly diathesis (which also serves for mood and voice). The effects of this analysis are evident in Apollonius Dyscolus as well as Quintilian and Varro (for Latin, which has no distinction like the Greek one between aorist and perfect).
It is difficult to summarize the contribution of the Stoics to linguistic theory, but it appears enormous, bothinthewaytheyraisenewgrammatical questions and in providing solutions for older ones.
Bibliography
Arens H 1984 Aristotle's Theory of Language and its Tra- dition: Textsfrom 500 to 1750,Studies in the History of Linguistics 29. Benjamins, Amsterdam
Arnim H von 1902 Stoicorum vetenon fragmenta. Vol. II: Chrysippi fragmenta logica et physica. Teubner, Leipzig. Repr. 1923
Ax W 1986 Quadripartita Ratio: Bemerkungen zur Ge- 479
Aristotle and the Stoics