[114] No description of their system can dispense with this fundamental enquiry, which had been already instituted by Zeno. It appears, however, to have been treated by several writers as a branch of dialectic. Diog. 43 says that the branch of dialectic which treats of σημαινόμενα may be divided εἴς τε τὸν περὶ τῶν φαντασιῶν τόπον καὶ τῶν ἐκ τούτων ὑφισταμένων λεκτῶν. (See Nicolai p. 23.) Compare with this the words of Diocles, in Diog. 49: ἀρέσκει τοῖς Στωϊκοῖς περὶ φαντασίας καὶ αἰσθήσεως προτάττειν λόγον, καθότι τὸ κριτήριον ᾧ ἡ ἀλήθεια τῶν πραγμάτων γινώσκεται, κατὰ γένος φαντασία ἐστὶ καὶ καθότι ὁ περὶ συγκαταθέσεως καὶ ὁ περὶ καταλήψεως καὶ νοήσεως λόγος προάγων τῶν ἄλλων οὐκ ἄνευ φαντασίας συνίσταται. According to this passage, the branch of dialectic which treated of φαντασία included the theory of knowledge. Diog. 53, Petersen’s conjecture is singular (Phil. Chrys. Fund. p. 25) that the theory of knowledge may have been understood by Chrysippus under the name rhetoric. [115] Our information on this head is very small. In the words: ῥητορικὴ verba curat et sensus et ordinem, a division of rhetoric is implied by Seneca, which differs in little, except in the position of the chief parts, from that of Aristotle. A fourth part is added to the three others by Diog. 43—on Delivery—εἶναι δ’ αὐτῆς τὴν διαίρεσιν εἴς τε τὴν εὕρεσιν καὶ εἰς τὴν φράσιν, καὶ εἰς τάξιν καὶ εἰς τὴν ὑπόκρισιν. Diogenes also claims for the Stoics the Aristotelian distinction between three ways of speaking—συμβουλευτικὸς, δικανικὸς, ἐγκωμιαστικός—and four parts in a speech: προοίμιον, διήγησις, τὰ πρὸς τοὺς ἀντιδίκους, ἐπίλογος. Definitions of διήγησις and παράδειγμα are given from Zeno by the anonymous author in Spengel, Rhet. Gr. i. 434, 23; 447, 11. The same author (Ibid. 454, 4) says that, according to Chrysippus, the ἐπίλογος must be μονομερής. The Stoic definition of rhetoric has been already given, p. 70, 1. Another—τέχνη περὶ κόσμου καὶ εἰρημένου λόγου τάξιν—is attributed to Chrysippus by Plut. Sto. Rep. 28, 1. Cic. Fin. iv. 3, 7, observes, in reference to the Stoic rhetoric, and in particular to that of Chrysippus, that such was its nature that si quis obmutescere concupierit, nihil aliud legere debeat—that it dealt in nothing but words, being withal scanty in expressions, and confined to subtleties. This neglect of the truly rhetorical element appears already in the quotations from Plut. Sto. Rep. 28, 2. We have not the slightest reason to complain, as Prantl does, p. 413, of the purely rhetorical value of dialectic with the Stoics.