§17#1Πρῶτον μὲν οὖν, ὥσπερ συλλογίζεσθαί φαμεν ἐνδόξως ποτὲ μᾶλλον ἢ ἀληθῶς προαιρεῖσθαι δεῖν, οὕτω καὶ λυτέον ποτὲ μᾶλλον ἐνδόξως ἢ κατὰ τἀληθές.
First, then, just as we say that we must sometimes choose to reason plausibly rather than truly, so too we must sometimes resolve [arguments] plausibly rather than according to the truth.
Ὅλως γὰρ πρὸς ποὺς ἐριστικοὺς μαχετέον οὐχ ὡς ἐλέγχοντας ἀλλ᾿ ὡς φαινομένους· οὐ γάρ φαμεν συλλογίζεσθαί γε αὐτούς, ὥστε πρὸς τὸ μὴ δοκεῖν διορθωτέον.
For, in general, against contentious people we must fight not as if they were refuting truly, but as if they appeared to do so; for we do not say that they actually reason, so that we must make corrections so as not to seem [to be refuted].
Εἰ γάρ ἐστιν ὁ ἔλεγχος ἀντίφασις μὴ ὁμώνυμος ἔκ τινων, οὐδὲν ἂν δέοι διαιρεῖσθαι πρὸς τἀμφίβολα καὶ τὴν ὁμωνυμίαν· οὐ γὰρ ποιεῖ συλλογισμόν.
For if a refutation is a contradiction that is not homonymous, derived from certain premises, there would be no need to make distinctions against ambiguous terms and homonymy; for they do not produce a reasoning.
Ἀλλ᾿ οὐδενὸς ἄλλου χάριν προσδιαιρετέον ἀλλ᾿ ἢ ὅτι τὸ συμπέρασμα φαίνεται ἐλεγχοειδές.
But for no other reason must we make further distinctions than because the conclusion appears like a refutation.
Οὔκουν τὸ ἐλεγχθῆναι ἀλλὰ τὸ δοκεῖν εὐλαβητέον, ἐπεὶ τό γ᾿ ἐρωτᾶν ἀμφίβολα καὶ τὰ παρὰ τὴν ὁμωνυμίαν, ὅσαι τ᾿ ἄλλαι τοιαῦται παρακρούσεις, καὶ τὸν ἀληθινὸν ἔλεγχον ἀφανίζει καὶ τὸν ἐλεγχόμενον καὶ μὴ ἐλεγχόμενον ἄδηλον ποιεῖ.
Therefore, we must beware not of being refuted, but of seeming to be, since to question ambiguous things and those depending on homonymy, and all other such deceptions, obscures even a true refutation and makes it unclear who is refuted and who is not.
Ἐπεὶ γὰρ ἔξεστιν ἐπὶ τέλει συμπεραναμένου μὴ ὅπερ ἔφησεν ἀποφῆσαι λέγειν, ἀλλ᾿ ὁμωνύμως, εἰ καὶ ὅτι μάλιστ᾿ ἔτυχεν ἐπὶ ταὐτὸν φέρων, ἄδηλον εἰ ἐλήλεγκται· ἄδηλον γὰρ εἰ ἀληθῆ λέγει νῦν.
For since it is possible, when the questioner has drawn the conclusion at the end, for the respondent to say that he is denying not what he affirmed, but only homonymously—even if the questioner happened to direct it as much as possible to the same point—it is unclear whether he has been refuted; for it is unclear whether he is speaking the truth now.
Εἰ δὲ διελὼν ἤρετο τὸ ὁμώνυμον ἢ τὸ ἀμφίβολον, οὐκ ἂν ἄδηλος ἦν ὁ ἔλεγχος.
But if the questioner had made a distinction and then asked about the homonymous or ambiguous term, the refutation would not have been unclear.
Ὅ τ᾿ ἐπιζητοῦσι νῦν μὲν ἧττον πρότερον δὲ μᾶλλον οἱ ἐριστικοί, τὸ ἢ ναί ἢ οὔ ἀποκρίνεσθαι τὸν ἐρωτώμενον, ἐγίνετ᾿ ἄν.
And what the contentious seek, less so now but more in the past, namely that the person questioned should answer either "yes" or "no", would have been realized.
Νῦν δὲ διὰ τὸ μὴ καλῶς ἐρωτᾶν τοὺς πυνθανομένους ἀνάγκη προσαποκρίνεσθαί τι τὸν ἐρωτώμενον, διορθοῦντα τὴν μοχθηρίαν τῆς προτάσεως, ἐπεὶ διελομένου γε ἱκανῶς ἢ ναί ἢ οὔ ἀνάγκη λέγειν τὸν ἀποκρινομενον.
But now, because those who inquire do not question well, it is necessary for the person questioned to add something in his answer, correcting the badness of the premise, since if the questioner has distinguished it sufficiently, the respondent must say either "yes" or "no".
Εἰ δέ τις ὑπολήψεται τὸν κατὰ ὁμωνυμίαν ἔλεγχον εἶναι, τρόπον τινὰ οὐκ ἔσται διαφυγεῖν τὸ ἐλέγχεσθαι τὸν ἀποκρινόμενον· ἐπὶ γὰρ τῶν ὁρατῶν ἀναγκαῖον ὃ ἔφησεν ἀποφῆσαι ὄνομα, καὶ ὃ ἀπέφησε φῆσαι.
But if someone supposes that a refutation based on homonymy is [a true refutation], in a way it will not be possible for the respondent to escape being refuted; for in the case of visible things, it is necessary to deny the name which he affirmed, and to affirm what he denied.
Ὡς γὰρ διορθοῦνταί τινες, οὐδὲν ὄφελος.
For the way some people make corrections is of no use.
Οὐ γὰρ Κορίσκον φασὶν εἶναι μουσικὸν καὶ ἄμουσον, ἀλλὰ τοῦτον τὸν Κορίσκον μουσικὸν καὶ τοῦτον τὸν Κορίσκον ἄμουσον.
For they say not that Coriscus is musical and unmusical, but that *this* Coriscus is musical and *this* Coriscus is unmusical.
Ὁ γὰρ αὐτὸς ἔσται λόγος τὸ τὸν Κορίσκον τῷ τοῦτον τὸν Κορίσκον ἄμουσον εἶναι ἢ μουσικόν· ὅπερ ἅμα φησί τε καὶ ἀπόφησιν.
For "Coriscus is unmusical or musical" will be the same statement as "this Coriscus is unmusical or musical"; which he affirms and denies at the same time.
Ἀλλ᾿ ἴσως οὐ ταὐτὸ σημαίνει· οὐδὲ γὰρ ἐκεῖ τοὔνομα.
But perhaps they do not signify the same thing, since even in the other case the name is not the same.
Ὥστε τι διαφέρει.
So there is some difference.
Εἰ δὲ τῷ μὲν τὸ ἁπλῶς λέγειν Κορίσκον ἀποδώσει, τῷ δὲ προσθήσει τὸ τινὰ ἢ τόνδε, ἄτοπον· οὐδὲν γὰρ μαλλον θατέρῳ· ὁποτέρῳ γὰρ ἂν οὐδὲν διαφέρει.
But if he assigns to the one the speaking of "Coriscus" simply, and adds to the other "a certain" or "this", it is absurd; for there is no more reason to do so for the one than for the other; for whichever of the two he adds it to, it makes no difference.
Οὐ μὴν ἀλλ᾿ ἐπειδὴ ἄδηλος μέν ἐστιν ὁ μὴ διορισάμενος τὴν ἀμφιβολίαν πότερον ἐλήλεγκται ἢ οὐκ ἐλήλεγκται, δέδοται δ᾿ ἐν τοῖς λόγοις τὸ διελεῖν, φανερὸν ὅτι τὸ μὴ διορίσαντα δοῦναι τὴν ἐρώτησιν ἀλλ᾿ ἁπλῶς ἁμάρτημά ἐστιν, ὥστε κἂν εἰ μὴ αὐτός, ἀλλ᾿ ὅ γε λόγος ἐληλεγμένῳ ὅμοιός ἐστιν.
Nevertheless, since he who does not define the ambiguity is unclear as to whether he has been refuted or not, and since in discussions to make a distinction is permitted, it is clear that to grant the question simply without defining is an error, so that, even if he himself is not, at least the argument is like one that has been refuted.
Συμβαίνει μέντοι πολλάκις ὁρῶντας τὴν ἀμφιβολίαν ὀκνεῖν διαιρεῖσθαι διὰ τὴν πυκνότητα τῶν τὰ τοιαῦτα προτεινόντων, ὅπως μὴ πρὸς ἅπαν δοκῶσι δυσκολαίνειν· εἶτ᾿ οὐκ ἂν οἰηθέντων παρὰ τοῦτο γενέσθαι τὸν λόγον, πολλάκις ἀπήντησε παράδοξον.
Yet it often happens that, though we see the ambiguity, we hesitate to make a distinction because of the frequency of those who propose such things, so that we may not seem to be making difficulties at every point; then, although we did not think that the argument would turn upon this, we often meet with something unexpected.
Ὥστ᾿ ἐπειδὴ δέδοται διαιρεῖν, οὐκ ὀκνητέον, καθάπερ ἐλέχθη πρότερον.
Since, therefore, it is permitted to make a distinction, we must not hesitate, just as was said before.
Εἰ δὲ τὰ δύο ἐρωτήματα μὴ ἓν ποιεῖ τις ἐρώτημα, οὐδ᾿ ἂν ὁ παρὰ τὴν ὁμωνυμίαν καὶ τὴν ἀμφιβολίαν ἐγίνετο παραλογισμός, ἀλλ᾿ ἢ ἔλεγχος ἢ οὔ.
Again, if one does not make two questions into one question, neither would the fallacy depending on homonymy and ambiguity occur, but either a refutation or not.
Τί γὰρ διαφέρει ἐρωτῆσαι εἰ Καλλίας καὶ Θεμιστοκλῆς μουσικοί εἰσιν ἢ εἰ ἀμφοτέροις ἓν ὄνομα ἦν ἑτέροις οὖσιν;
For what difference does it make to ask whether Callias and Themistocles are musical, or whether there was one name for both of them, being different people?
εἰ γὰρ πλείω δηλοῖ ἑνός, πλείω ἠρώτησεν.
For if it signifies more than one, he has asked more than one question.
Εἰ οὖν μὴ ὀρθὸν πρὸς δύο ἐρωτήσεις μίαν ἀπόκρισιν ἀξιοῦν λαμβάνειν ἁπλῶς, φανερὸν ὅτι οὐδενὶ προσήκει τῶν ὁμωνύμων ἀποκρίνεσθαι ἁπλῶς, οὐδ᾿ εἰ κατὰ πάντων ἀληθές, ὥσπερ ἀξιοῦσί τινες.
If, then, it is not right to demand to receive simply a single answer to two questions, it is clear that it is not fitting to answer simply to any homonymous question, not even if it is true of all [senses], as some people demand.
Οὐδὲν γὰρ τοῦτο διαφέρει ἢ εἰ ἤρετο, Κορίσκος καὶ Καλλίας πότερον οἴκοι εἰσὶν ἢ οὐκ οἴκοι, εἴτε παρόντων ἀμφοῖν εἴτε μὴ παρόντων· ἀμφοτέρως γὰρ πλείους αἱ προτάσεις· οὐ γὰρ εἰ ἀληθὲς εἰπεῖν, διὰ τοῦτο μία ἡ ἐρώτησις.
For this is no different from asking whether Coriscus and Callias are at home or not at home, whether both are present or both are not present; for in both cases the premises are multiple; for even if it is true to say [the answer], that does not make the question one.
Ἐγχωρεῖ γὰρ καὶ μυρία ἕτερα ἐρωτηθέντα ἐρωτήματα ἅπαντα ἢ ναὶ ἢ οὒ ἀληθὲς εἶναι λέγειν· ἀλλ᾿ ὅμως οὐκ ἀποκριτέον μιᾷ ἀποκρίσει· ἀναιρεῖται γὰρ τὸ διαλέγεσθαι.
For it is possible that even ten thousand other questions having been asked, it is true to say "yes" or "no" to all of them; but nevertheless, one must not answer with a single answer; for [otherwise] discussion is destroyed.
Τοῦτο δ᾿ ὅμοιον ὡς εἰ καὶ τὸ αὐτὸ ὄνομα τεθείη τοῖς ἑτέροις.
And this is similar to if the same name were given to different things.
Εἰ οὖν μὴ δεῖ πρὸς δύο ἐρωτήσεις μίαν ἀπόκρισιν διδόναι, φανερὸν ὅτι οὐδ᾿ ἐπὶ τῶν ὁμωνύμων τὸ ναί ἢ οὔ λεκτέον.
If, therefore, one must not give a single answer to two questions, it is clear that neither in the case of homonymous terms must one say "yes" or "no".
Οὐδὲ γὰρ ὁ εἰπὼν ἀποκέκριται ἀλλ᾿ εἴρηκεν.
For he who says so has not answered, but has merely spoken.
Ἀλλ᾿ ἀξιοῦνταί πως ἐν τοῖς διαλεγομένοις διὰ τὸ λανθάνειν τὸ συμβαῖνον.
But they are somehow demanded in discussions because what happens escapes notice.