III. Figures of speech have always been liable to change and are continually in process of change in accordance with the variations of usage. Consequently when we compare the language of our ancestors with our own, we find that practically everything we say nowadays is figurative. For example, we say invidere hac re for to grudge a thing, instead of hanc rem, which was the idiom of all the ancients, more especially Cicero, and incumbere illi (to lean upon him) for incumbere in ilium, plenum vino (full of wine) for plenum vini, and huic adulari (to flatter him) for hunc adulari. I might quote a thousand other examples, and only wish I could say that the changes were not often changes for the worse.

But to proceed, figures of speech fall into two main classes. One is defined as the form of language, while the other is mainly to be sought in the arrangement of words. Both are equally applicable in oratory, but we may style the former rather more grammatical and the latter more rhetorical. These grammatical figures would not be styled figures of speech in English. Figures of language would perhaps be more comprehensive, but figures of speech is the translation and direct descendant of the original Greek σχήματα λέξεως and has therefore been used throughout. The former originates from the same sources as errors of language. For every figure of this kind would be an error, if it were accidental and not deliberate.

But as a rule such figures are defended by authority, age and usage, and not infrequently by some reason as well. Consequently, although they involve a divergence from direct and simple language, they are to be regarded as excellences, provided always that they have some praiseworthy precedent to follow. They have one special merit, that they relieve the tedium of everyday stereotyped speech and save us from commonplace language.

If a speaker use them sparingly and only as occasion demands, they will serve as a seasoning to his style and increase its attractions. If, on the other hand, he strains after them overmuch, he will lose that very charm of variety which they confer. Some figures, however, are so generally accepted that they have almost ceased to be regarded as figures: consequently however frequently they may be used, they will make less impression on the ear, just because it has become habituated to them.

For abnormal figures lying outside the range of common speech, while they are for that very reason more striking, and stimulate the ear by their novelty, prove cloying if used too lavishly, and make it quite clear that they did not present themselves naturally to the speaker, but were hunted out by him, dragged from obscure corners and artificially piled together. Figures, then,

may be found in connexion with the gender of nouns; for we find oculis capti talpale Georg. i. 183. (blind moles) and timidi damae Ecl. viii. 28. (timid deer) in Virgil; but there is good reason for this, since in these cases both sexes are covered by a word of one gender, and there is no doubt that there are male moles and deer as well as female. Figures may also affect verbs: for example, we find such phrases as fabricatus est glatdium Cic. pro Rab Post. iii. 7. He made a sword. or inimicum poenitus es. pro Mil. xiii. 33. You punished an enemy.

This is the less surprising, since the nature of verbs is such that we often express the active by the passive form, as in the case of arbitror (think) and suspicor (suspect), and the passive by the active, as in the case of vapulo (am beaten). Consequently the interchange of the two forms is of common occurrence, and in many cases either form can be used: for example, we may say luxuriatur or luxuriat (luxuriate), fluctuatur or fluctuat (fluctuate), adsentior or adsentio (agree). Figures also occur in connexion with number,