§2.2.3Ὅτι μὲν οὖν πάντα τὰ ζητούμενα μέσου ζήτησίς ἐστι, δῆλον·
That all inquiries, then, are inquiries of a middle term is clear.
πῶς δὲ τὸ τί ἐστι δείκνυται, καὶ τίς ὁ τρόπος τῆς ἀναγωγῆς, καὶ τί ἐστιν ὁρισμός καὶ τίνων, εἴπωμεν, διαπορήσαντες πρῶτον περὶ αὐτῶν.
But how 'what it is' is shown, and what is the mode of reduction, and what is a definition and of what things, let us speak, having first raised difficulties about them.
ἀρχὴ δʼ ἔστω τῶν μελλόντων ἥπερ ἐστὶν οἰκειοτάτη τῶν ἐχομένων λόγων.
And let the starting point of what is to come be that which is most appropriate to the following discussions.
ἀπορήσειε γὰρ ἄν τις, ἆρʼ ἔστι τὸ αὐτὸ καὶ κατὰ τὸ αὐτὸ ὁρισμῷ εἰδέναι καὶ ἀποδείξει, ἢ ἀδύνατον;
For someone might raise a difficulty: is it possible to know the same thing in the same respect both by definition and by demonstration, or is it impossible?
ὁ μὲν γὰρ ὁρισμός τοῦ τί ἐστιν εἶναι δοκεῖ, τὸ δὲ τί ἐστιν ἅπαν καθόλου καὶ κατηγορικόν· συλλογισμοὶ δʼ εἰσὶν οἱ μὲν στερητικοί, οἱ δʼ οὐ καθόλου, οἶον οἱ μὲν ἐν τῷ δευτέρῳ σχήματι στερητικοὶ πάντες, οἱ δʼ ἐν τῷ τρίτῳ οὐ καθόλου.
For definition seems to be of 'what it is', and 'what it is' is always universal and affirmative; but of syllogisms, some are negative and others are not universal—for instance, all those in the second figure are negative, and those in the third are not universal.
εἶτα οὐδὲ τῶν ἐν τῷ πρώτῳ σχή· ματι κατηγορικῶν ἁπάντων ἔστιν ὁρισμός, οἷον ὅτι πᾶν τρίγωνον δυσὶν ὀρθαῖς ἴσας ἔχει.
Next, there is not even a definition of all the affirmative syllogisms in the first figure, as for instance that every triangle has angles equal to two right angles.
τούτου δὲ λόγος, ὅτι τὸ ἐπίστασθαί ἐστι τὸ ἀποδεικτὸν τὸ ἀπόδειξιν ἔχειν, ὥστʼ ἐπεὶ τῶν τοιούτων ἀπόδειξις ἔστι, δῆλον ὅτι οὐκ ἄν εἴη αὐτῶν καὶ ὁρισμός· ἐπίσταιτο γὰρ ἄν τις καὶ κατὰ τὸν ὁρισμόν, οὐκ ἔχων τὴν ἀπόδειξιν· οὐδὲν γὰρ κωλύει μὴ ἅμα ἔχειν.
The reason for this is that to know a demonstrable thing is to have its demonstration, so that since there is a demonstration of such things, it is clear that there could not also be a definition of them; for otherwise one would know them by the definition also without having the demonstration, since nothing prevents one from not having them at the same time.
ίκανὴ δὲ πίστις καὶ ἐκ τῆς ἐπαγωγῆς· οὐδὲν γὰρ πώποτε ὁρισάμενοι ἔγναωμεν, οὐτε τῶν καθʼ αὑτὸ ὑπαρχόντων οὔτε τῶν συμβεβηκότων.
A sufficient assurance of this is also from induction; for we have never yet known anything by defining it, neither of essential properties nor of accidents.
ἔτι εἰ ὁ ὁρισμὸς οὐσίας τινὸς γνωρισμός, τά γε τοιαῦτα φανερὸν ὅτι οὐκ οὐσίαι.
Furthermore, if definition is a kind of recognition of essence, it is clear that such things as these are not essences.
Ὅτι μὲν οὖν οὐκ ἔστιν ὁρισμὸς ἅπόντος οὗπερ καὶ ἀπόδειξις, δῆλον.
That there is not, then, a definition of everything of which there is also a demonstration is clear.
τί δαί, οὗ ὁρισμός, ἄρα παντὸς ἀπόδειξις ἔστιν ἢ οὔ;
What then? Is there a demonstration of everything of which there is a definition, or not?
εἶς μὲν δὴ λόγος καὶ περὶ τούτου ὁ αὐτός.
One argument indeed concerning this is the same.
τοῦ γὰρ ἑνός, ἕν, μία ἐπιστήμη.
For of one thing, there is one science.
ὥστʼ εἴπερ τὸ ἐπίστασθαι τὸ ἀποδεικτόν ἐστι τὸ τὴν ἀπόδειξιν ἔχειν, συμβήσεταί τι ἀδύνατον· ὁ γὰρ τὸν ὁρισμόν ἔχων ἄνευ τῆς ἀποδείξεως ἐπιστήσεται.
So that if to know a demonstrable thing is to have its demonstration, something impossible will result: for the one who has the definition will know without the demonstration.
ἔτι αἱ ἀρχαὶ τῶν ἀποδείξεων ὁρισμοί, ὧν ὅτι οὐκ ἔσονται ἀποδείξεις δέδεικται πρότερον—ἢ ἔσονται αἱ ἀρχαὶ ἀποδεικταὶ καὶ τῶν ἀρχῶν ἀρχαί, καὶ τοῦτʼ εἰς ἄπειρον βαδιεῖται, ἢ τὰ πρῶτα ὁρισμοὶ ἔσονται ἀναπόδεικτοι.
Furthermore, the principles of demonstrations are definitions, of which that there will not be demonstrations has been shown before—either the principles will be demonstrable and there will be principles of principles, and this will go on to infinity, or the first things will be indemonstrable definitions.
Ἀλλʼ ἆρα, εἰ μὴ παντὸς τοῦ αὐτοῦ, ἀλλὰ τινὸς τοῦ αὐτοῦ ἔστιν ὁρισμὸς καὶ ἀπόδειξις;
But is it then possible that, if not of every same thing, yet of some same thing there is both definition and demonstration?
ἢ ἀδύνατον;
Or is it impossible?
οὐ γὰρ ἔστιν ἀπόδειξις οὐ ὁρισμός.
Because there is no demonstration of what there is a definition.
ὁρισμὸς μὲν γὰρ τοῦ τί ἐστι καὶ οὐσίας· αἱ δʼ ἀποδείξεις φαίνονται πᾶσαι ὑποτιθέμεναι καὶ λαμβάνουσαι τὸ τί ἐστιν, οἶον αἱ μαθηματικαὶ τί μονὰς καὶ τί τὸ περιττόν, καὶ αἱ ἄλλαι ὁμοίως.
For definition is of 'what it is' and of essence; but demonstrations all appear to hypothesize and assume 'what it is'—for instance, mathematical demonstrations assume what a unit is and what odd is, and the others likewise.
ἔτι πᾶσα ἀπόδειξις τὶ κατὰ τινὸς δείκνυσιν, οἷον ὅτι ἔστιν ἢ οὐκ ἔστιν· ἐν δὲ τῷ ὁρισμῷ οὐδὲν ἕτερον ἑτέρου κατηγορεῖται, οἷον οὔτε τὸ ζῷον κατὰ τοῦ δίποδος οὔτε τοῦτο κατὰ τοῦ ζῴου, οὐδὲ δὴ κατὰ τοῦ ἐπιπέδου τὸ σχῆμα· οὐ γάρ ἐστι τὸ ἐπίπεδον σχῆμα, οὐδὲ τὸ σχῆμα ἐπίπεδον.
Furthermore, every demonstration shows something of something, for instance that it is or is not; but in a definition nothing else is predicated of another, for instance neither animal of two-footed, nor this of animal, nor indeed figure of plane; for a plane is not a figure, nor is a figure a plane.
ἔτι ἕτερον τὸ τί ἐστι καὶ ὅτι ἔστι δεῖξαι.
Furthermore, to show what it is and that it is are different.
ὁ μὲν οὖν ὁρισμὸς τί ἐστι δηλοῖ, ἡ δὲ ἀπόδειξις ὅτι ἔστι τόδε κατὰ τοῦδε ἢ οὐκ ἔστιν.
Definition, then, makes clear what it is, but demonstration shows that this is of that, or is not.
ἑτέρου δὲ ἑτέρα ἀπόδειξις, ἐὰν μὴ ὡς μέρος ᾖ τι τῆς ὅλης.
And of a different thing there is a different demonstration, unless it is as a part of the whole.
τοῦτο δὲ λέγω, ὅτι δέδεικται τὸ ἰσοσκελὲς δύο ὀρθαί, εἰ τπᾶν τρίγωνον δέδεικται· μέρος γάρ, τὸ δʼ ὅλον.
And I mean by this that the isosceles is shown to have two right angles if every triangle is shown; for one is a part, the other the whole.
ταῦτα δὲ πρὸς ἄλληλα οὐκ ἔχει οὕτως, τὸ ὅτι ἔστι καὶ τί ἐστιν· οὐ γάρ ἐστι θατέρου θάτερον μέρος.
But these do not have this relation to each other, namely, 'that it is' and 'what it is'; for neither is a part of the other.
Φανερὸν ἄθρα ὅτι οὔτε οὗ ὁρισμός, τούτου παντὸς ἀπόδειξις, οὔτε οὗ ἀπόδειξις, τούτο παντὸς ὁρισμός, οὔτε ὅλως τοῦ αὐτοῦ οὐδενὸς ἐνδέχεται ἄμφω ἔχειν.
It is clear, then, that neither of everything of which there is a definition is there a demonstration, nor of everything of which there is a demonstration is there a definition, nor is it possible at all to have both of any same thing.
ὥστε δῆλον ὡς οὐδὲ ὁρισμὸς καὶ ἀπόδειξις οὐὔτε τὸ αὐτὸ ἄν εἴη οὔτε θάτερον ἐν τερῳ· καὶ γὰρ ἄν τὰ ὑποκείμενα ὁμοίως εἶχεν.
So that it is clear that neither definition and demonstration would be the same, nor one in the other; for otherwise their subjects would also be related in the same way.