[221] Sext. Math. viii. 112; κοινῶς μὲν γάρ φασιν ἅπαντες οἱ Διαλεκτικοὶ ὑγιὲς εἶναι συνημμένον, ὅταν ἀκολουθῇ τῷ ἐν αὐτῷ ἡγουμένῳ τὸ ἐν αὐτῷ λῆγον. περὶ δὲ τοῦ πότε ἀκολουθεῖ καὶ πῶς, στασιάζουσι πρὸς ἀλλήλους καὶ μαχόμενα τῆς ἀκολουθίας ἐκτίθενται κριτήρια. Cic. Acad. ii. 47, 143: In hoc ipso, quod in elementis dialectici docent, quomodo judicare oporteat, rerum falsumne sit, si quid ita connexum est, ut hoc: Si dies est, lucet; quanta contentio est! aliter Diodoro aliter Philoni, Chrysippo aliter placet. (The further remarks on the points of difference between Chrysippus and Cleanthes have no reference to hypothetical judgments.) The Philo here alluded to—the same Philo against whom Chrysippus wrote his treatises (Diog. vii. 191 and 194)—is the well-known dialectician and pupil of Diodorus, who declared all conditional sentences to be right in which a false inferential clause is not drawn from a true leading clause. According to this view, conditional sentences would be right, with both clauses true, or both false, or with a false leading clause and true inferential clause (Sext. l.c. viii. 245 and 449; Pyrrh. ii. 110). According to Sext. Pyrrh. ii. 104, the view of Philo appears to have gained acceptance among the Stoics, perhaps through Zeno, for whose connection with Philo see Diog. vii. 16. But, in any case, the meaning appears to have been (Diog. vii. 81), that, in the cases mentioned, conditional sentences may be right, not that they must be right. Others more appropriately judged of the correctness of conditional sentences by the connection of the clauses, and either required, for a conditional sentence to be right, that the contradictory opposite (ἀντικείμενον) of the inferential clause should be irreconcileable with the leading clause, or that the inferential clause should be potentially (δυνάμει) contained in the leading clause (Sext. Pyrrh. ii. 111). The first of these requirements, which is mentioned by Diog. 73 as the only criterion of the Stoic School, was due to Chrysippus, who accordingly refused to allow sentences in which this was not the case to be expressed hypothetically (Cic. De Fato, 6, 12; 8, 15): it was not right to say, Si quis natus est oriente canicula, is in mari non morietur; but, Non et natus est quis oriente canicula et is in mari morietur. It may be observed, in connection with the enquiry into the accuracy of conditional sentences, that a true conditional sentence may become false in time. The sentence, If Dion is alive now, he will continue to live, is true at the present moment; but in the last moment of Dion’s life it will cease to be true. Such sentences were called ἀπεριγράφως μεταπίπτοντα, because the time could not be previously fixed when they would become false (Simpl. Phys. 305, a). Chrysippus also wrote on the μεταπίπτοντα, according to Dionys. Comp. Verb. p. 72 Schäfer. Diog. vii. 105, mentions two treatises of his on the subject, but characterises them as spurious.