Tlic RJictorica of Pliilodemus. 373 



or the shrewd statesmanship of a Themistocles were as much 

 the products of "philosophy" as the mathematical subtleties of 

 Pythagoras. Cicero is drawing- from a Greek source which 

 ultimately goes back to Isocrates. From this source he derives 

 a triad of Greeks who combined deep speculations with power 

 of speech. They are Lycurg'us, Pittacus, Solon. (De Orat. Ill, 

 15. 56.) He parallels these by a list of Romans, Coruncanius, 

 Fabricius, Cato, Scipio, thus clothing the thought in Roman 

 dress. This same method he has followed in the passage in the 

 first book, by putting into the mouth of Charmadas a Roman 

 example, Antonius, instead of the Greek examples, Demades and 

 Aeschines which Charmadas really used. 



Cicero thus distinctly points to Charmadas as the source of 

 this thought ; Sextus may be interpreted to favor either Char- 

 madas or Critolaus ; Pliilodemus is slightly in favor of Critolavis, 

 but not enough so to outweigh the definite statement of Cicero. 

 The possibility must be considered, however, that both Critolaus 

 and Charmadas may have used the same line of thought and the 

 same illustrations ; Quintilian implies that the kernel of the 

 thought "rhetoricen . . . ohscrvationcm quandam esse, non 

 artem" is as old as Lysias, hence it would be common property 

 by the time of Critolaus and Charmadas. Moreover the rapid 

 rise of Demades from a common seaman to a position of political 

 leadership was well enough known to pass into a proverb. At 

 least this is the most probable origin of the expression, 'Atto kwttt^s 

 €7ri Prifjua, From the oar to the rostrum, found in Apostolius III, 

 65. I am inclined to believe, however, that whatever may have 

 been the ultimate origin of the idea, the form in which it appears 

 is due to its use in an attack on rhetoric published by Charmadas. 

 It is quite likely that this was in dialogue form, and that the chief 

 persons were Critolaus, Diogenes, and Carneades, the master of 

 Charmadas. Such a debate would be highly natural, suggested 

 perhaps by their association on the embassy to Rome. More- 

 over the assumption of such a dialogue removes any difificulty 

 that may be felt in regard to the references to Critolaus in our 

 Greek authorities, wdiich have given some color to the claim 

 that some work of his was the source of this argument. Cicero 

 refers to Charmadas, because he was known to be the author of 

 the dialogue, and the ideas might be assumed to be his also ; 

 Sextus and Philodemus mention Critolaus because these argfu- 



