[69] 39. [70] xii. 44. [71] _sapientiam._ xv. 56. [72] 57. [73] xvi. 61. [74] xvii-xxiii. [75] xxiv-xxvii. [76] xxviii-xxxv. [77] xiv. 53. [78] xxxii. 126. [79] Sandys notes that the avowed object is “criticism, and not direct instruction.” This, however, is part of Cicero’s literary method and of his habit of scorning the manuals. As to his main topic, _elocutio_, he writes _doctrina_ as definite as that of _De oratore_ on the other parts; and though his headings are not all conventional, his outline and order are thoroughly systematic. [80] 37, seq. [81] 40, _verba iunxisse_; cf. 77, _vinculis numerorum_; 208. [82] 44, seq. [83] 50, seq. [84] 61-236. [85] 20-23. [86] See the articles by Hendrickson cited in the first foot-note to this chapter. [87] 69. [88] 100. [89] One could wish that Cicero had been content with his twofold division in _Brutus_, xxiii. 89: _cum duæ summæ sint in oratore laudes, una subtiliter disputandi ad docendum, altera graviter agendi ad animos audientium permovendos_. [90] The digest of the whole _Orator_ at pages lxxiv-lxxvi of the edition of Sandys need be neither repeated nor revised. Assuming this, I have added here certain significant rhetorical details, translation of some important passages, and the connection of the topics. [91] See above. [92] _Mensionem_, 177. The word in a similar passage at 67 is _mensura_. [93] Because, says Sandys, their style is unperiodic, and there can hardly be rhythm without periods. He cites the famous passage from Aristotle discussed above at page 27, and notes Quintilian’s demur as to Herodotus. This is a fair inference from Cicero’s context; and, indeed, the ancients generally considered prose rhythm as oratorical rhythm. The narrative rhythms of imaginative prose were naturally not much discussed separately in a time when prose fiction was undeveloped. The nearest approach to these in oratory was in panegyric. But Dionysius with more discernment praises the _compositio_ of Herodotus. (See below, Chapter v.) [94] The translation is closer to that of Colin than to that of Sandys. The point—and if it is obvious, it is often forgotten—seems to be that variety in prose depends on rhythm. [95] _concinnitas_ (201). Cicero does not say explicitly what I have summarized in the last sentence above; but I think he implies it. He does not hint what Stevenson brings out in _Some Technical Elements of Style in Literature_, that subconscious rhythmical predilection may be a cause, or a determining factor, in adaptation. [96] Dionysius of Halicarnassus exhibits this specifically with telling effect in the first part of _De compositione verborum_. See below, Chapter v.