XVII
[17arg] Marcus Cicero's observations on the nature of certain prepositions; to which is added a discussion of the particular matter which Cicero had observed.AFTER careful observation Marcus Tullius noted that the prepositions in and con, when prefixed to nouns and verbs, are lengthened and prolonged when they are followed by the initial letters of sapiens and felix; but that in all other instances they are pronounced short. Cicero's words are: 1 “Indeed, what can be more elegant than this, which does not come about from a natural law, but in accordance with a kind of usage? we pronounce the first vowel in indoctus short, in insanus long; in immanis short, in infelix long; in brief, in compound words in which the first letters are those which begin sapiens and felix the prefix is pronounced long, in all others short; thus we have conposuit but cōnsuevit, cŏncrepuit [p. 169] but cōnficit. Consult the rules of grammar and they will censure your usage; refer the matter to your ears and they will approve. Ask why it is so; they will say that it pleases them. And language ought to gratify the pleasure of the ear.” In these words of which Cicero spoke it is clear that the principle is one of euphony, but what are we to say of the preposition pro? For although it is often shortened or lengthened, yet it does not conform to this rule of Marcus Tullius. For it is not always lengthened when it is followed by the first letter of the word fecit, which Cicero says has the effect of lengthening the prepositions in and con. For we pronounce prŏficisci, prŏfugere, prŏfundere, prŏfannu and prŏfestumn with the first vowel short, but prōferre, prōfligare and prōficere with that syllable long. Why is it then that this letter, which, according to Cicero's observation, has the effect of lengthening, does not have the same effect by reason of rule or of euphony in all words of the same kind, 2 but lengthens the vowel in one word and shortens it in another. Nor, as a matter of fact, is the particle con lengthened only when followed by that letter which Cicero mentioned: for both Cato and Sallust say faenoribus copertus est." 3 Moreover cōligatus and cōnexus are pronounced long. But after all, in these cases which I have cited one can see that this particle is lengthened because the letter n is dropped; for the loss of a letter is compensated by the lengthening of the syllable. This principle is observed also in the word cōgo; and it is no contradiction that we pronounce cŏegi [p. 171] short; for this form cannot be derived from cōgo without violation of the principle of analogy. 4