Authors/Thomas Aquinas/posteriorum/L1/Lect8

Lecture 8 The second error is excluded by showing that circular demonstration is not acceptable

Latin	English
	Lecture 8 (72b-73a20) THE SECOND ERROR IS EXCLUDED BY SHOWING THAT CIRCULAR DEMONSTRATION IS NOT ACCEPTABLE
lib. 1 l. 8 n. 1 Postquam philosophus exclusit unam falsam radicem, ostendens quod non omnis scientia est per demonstrationem, hic excludit aliam, ostendens quod non contingit circulariter demonstrare.	After excluding one false basis by showing that not all science depends on demonstration, the Philosopher now excludes another by showing that it is not possible to demonstrate circularly.
lib. 1 l. 8 n. 2 Ad cuius evidentiam sciendum est quod circularis syllogismus dicitur, quando ex conclusione et altera praemissarum conversa concluditur reliqua. Sicut si fiat talis syllogismus: omne animal rationale mortale est risibile; omnis homo est animal rationale mortale; ergo omnis homo est risibilis: assumatur autem conclusio tanquam principium, et adiungatur ei minor conversa, hoc modo: omnis homo est risibilis; omne animal rationale mortale est homo; sequitur quod, omne animal rationale mortale sit risibile: quae erat maior primi syllogismi.	To understand this it should be noted that a demonstration is circular when the conclusion and one of the premises (in converted form) of a syllogism are used to prove the other premise. For example, we might form the following syllogism: Every rational mortal animal is risible; Every man is a rational mortal animal: Therefore, every man is risible. Now if the conclusion were to be used as one principle and the minor in converted form as the other, we would get: Every man is risible; Every rational mortal animal is a man: Therefore, every rational mortal animal is risible—which was the major of the first syllogism.
lib. 1 l. 8 n. 3 Ostendit autem Aristoteles per tres rationes quod non contingit circulariter demonstrare. Quarum prima est. In circulari syllogismo idem fit et principium et conclusio. Principium autem demonstrationis est prius et notius conclusione: ut supra ostensum est. Sequitur ergo quod idem sit prius et posterius respectu unius et eiusdem, et notius et minus notum. Hoc autem est impossibile. Ergo impossibile est circulariter demonstrare.	Accordingly, he. presents three arguments to show that it is not possible to demonstrate circularly. The first of these (72b25) is this: In a circular syllogism the same thing is at once a conclusion and a principle. But a principle of a demonstration is prior to and better known than the conclusion, as has been shown above. Therefore, it follows that a same thing is both prior to and subsequent to one same thing, and also more known and less known. But this is impossible. Therefore, it is impossible to demonstrate circularly.
lib. 1 l. 8 n. 4 Sed posset aliquis dicere quod idem potest esse prius et posterius alio et alio modo, scilicet ut hoc sit prius quoad nos, et illud sit prius simpliciter. Sicut singularia sunt priora quoad nos, et posteriora simpliciter; universalia vero e converso. Hoc autem modo inductio facit notum, scilicet altero modo a demonstratione. Nam demonstratio procedit ex prioribus simpliciter: inductio autem ex prioribus quoad nos.	But someone might say that a same thing can be both prior and subsequent, although not in the same way. For example, this might be prior in reference to us, but that prior absolutely. Thus singulars are prior in reference to us and subsequent absolutely: and conversely for universals. Again, induction makes something known in one way and demonstration in another way. For demonstration proceeds from things that are prior absolutely, but induction from things that are prior in reference to us.
Sed si sic fieret demonstratio circularis, ut scilicet primo concluderetur ex prioribus simpliciter, postea vero ex prioribus quoad nos; sequeretur quod non esset bene determinatum superius quid est scire. Dictum est enim quod scire est causam rei cognoscere. Et ideo ostensum est quod oportet demonstrationem, quae facit scire, ex prioribus simpliciter procedere. Si autem demonstratio, nunc ex prioribus simpliciter, nunc ex prioribus quoad nos procederet; oporteret etiam quod scire non solum esset causam rei cognoscere, sed dupliciter diceretur quia esset etiam quoddam scire per posteriora. Aut ergo oportebit sic dicere, aut oportebit dicere quod altera demonstratio, quae fit ex nobis notioribus, non sit simpliciter demonstratio.	Now if a circular demonstration were so constructed that something is first concluded from things that are absolutely prior, and then from things that are prior in reference to us, it would follow that our doctrine on scientific knowing was not well established. For we stated that to know scientifically is to know the cause of a thing. From this it followed that a demonstration which causes scientific knowledge must proceed from the absolutely prior. But if demonstration were at one time to proceed from the absolutely prior and at another time from things which are prior in reference to us, we would be forced to admit that scientific knowing is not confined to knowing the cause of a thing, but that there is another, namely, that form of knowing which proceeds from what is later. Therefore, one must either admit both or admit that the second form, namely, the demonstration which proceeds from what is better known to us is not a demonstration in the absolute sense.
lib. 1 l. 8 n. 5 Ex his autem apparet quare dialecticus syllogismus potest esse circularis. Procedit enim ex probabilibus. Probabilia autem dicuntur, quae sunt magis nota vel sapientibus vel pluribus. Et sic dialecticus syllogismus procedit ex his quae sunt magis nobis nota. Contingit autem idem esse magis et minus notum quoad diversos; et ideo nihil prohibet syllogismum dialecticum fieri circularem. Sed demonstratio fit ex notioribus simpliciter; et ideo, ut dictum est, non potest fieri demonstratio circularis.	The aforesaid also reveals why a dialectical syllogism can be circular. For it proceeds from things which are probable. But things are said to be probable if they are better known to the wise or to a great number of persons. Consequently, a dialectical syllogism proceeds from things that are better known to us. However, it happens that a same thing is better known to some and less known to others. Consequently, there is nothing to hinder a dialectical syllogism from being circular. But a demonstration is formed from things that are absolutely prior. Therefore, as we have already stated, there cannot be circular demonstration.
lib. 1 l. 8 n. 6 Secundam rationem ponit ibi: accidit autem etc.: quae talis est. Si demonstratio sit circularis, sequitur quod in demonstratione probetur idem per idem; ut si dicamus: si est hoc, est hoc. Sic autem facile est cuilibet demonstrare omnia. Hoc enim poterit facere quilibet, tam sciens quam ignorans. Et sic per demonstrationem non acquiritur scientia: quod est contra definitionem demonstrationis. Non ergo potest esse demonstratio circularis.	Then he sets forth the second argument (72b33) and it is this: If there were circular demonstration, it would follow that a same thing is demonstrated by the same thing, as if I were to say: If it is this, it is this. In this way it is easy for anyone to demonstrate everything, for anyone, wise or ignorant, will be able to do this. Accordingly, science is not acquired through demonstration. But this is against the definition of demonstration. Therefore, there cannot be circular demonstration.
Veritatem autem primae consequentiae hoc modo ostendit. Primo enim, dicit quod manifestum est quod accidit, circulari demonstratione facta, hoc quod prius dictum est, scilicet quod idem probetur per idem, si quis sumat tres terminos. Reflexionem autem fieri per multos, aut per paucos terminos nihil differt (nominat autem hic reflexionem processum, qui fit in demonstratione circulari a principio ad conclusionem, et iterum a conclusionem ad principium.) Huiusmodi ergo reflexio quantum ad vim argumentandi, sive fiat per multa, sive per pauca, non differt. Nec differt de paucis, aut de duobus. Eadem enim virtus arguendi est, si quis sic procedat: si est a, est b, et si est b, est c, et si est c, est d; et iterum reflectat dicens: si est d, est c, et si est c, est b, et si est b, est a; sicut si statim a principio reflecteret dicens: si est a, est b, et si est b, est a. Dicit autem per duos terminos, cum supra dixerit, tribus terminis positis, quia in deductione, quam faciet, utetur tertio termino, qui sit idem cum primo.	He proves the truth of the first consequence in the following way: It is obvious, first of all, that with a circular demonstration the same thing is proved by a same thing, as has been stated above, i.e., if only three terms are employed; although it makes no difference whether the reflexion be made with fewer terms or more. (By reflexion he means the process whereby one goes from principle to conclusion in a demonstration, and then from conclusion to principle). In such a reflexion it makes no difference, so far as the force of the argument is concerned, whether it involves several or fewer terms or even two. For an argument has the same force if one proceeds thus: “If it is A, it is B, and if it is B, it is C, and if it is C, it is D,” and then by reflecting continues, “If it is D, it is C, and if it is C, it is B, and if it is B, it is A”; or if he proceeds by reflecting at the very start, saying: “If it is A, it is B, and if it is B, it is A.” (Although he spoke above of three terms, he restricted himself to two terms in this example, because in the deduction he is about to make he will use a third term, which is the same as the first).
Deinde, cum dicit: cum enim etc., proponit formam argumentandi in tribus terminis, hoc modo: si sit a, est b; et si est b, est c; ergo si est a, ex necessitate est c. Deinde, cum dicit: si igitur cum sit a etc., per formam argumentandi praemissam ostendit quod in circulari demonstratione concluditur idem per idem, sumptis duobus terminis tantum. Erit enim dicere: si est a, est b; et reflectendo (quod est circulariter demonstrare): si est b, est a. Ex quibus duobus sequitur, secundum formam arguendi praemissam, si est a, est a.	Then (72b38) he gives the form of the argument in three terms, namely: “If it is A, it is B, and if it is B, it is C; therefore, if it is A, it is of necessity C.” Then (73a1) he shows by the aforesaid form of arguing that in a circular demonstration a same thing is proved by a same thing, using only two terms. For it consists in saying, “If it is A, it is B,” and then reflecting, “If it is B, it is A”—which is a circular demonstration. Now according to the above given form it follows from these two, that “if it is A, it is A.”
Quod sic patet. Sicut enim in prima deductione, quae fiebat per tres terminos, ad b sequebatur c; ita in deductione reflexa duorum terminorum, ad b sequitur a. Ponatur ergo quod idem significet a in secunda deductione reflexa, quod c significabat in prima directa, quae est per tres terminos. Igitur dicere in secunda deductione, si est b, est a, est hoc ipsum quod erat dicere in prima deductione, si est b, est c. Sed cum dicebatur in prima deductione, si est b, est c, sequebatur, si est a, est c. Ergo in deductione circulari sequitur, si est a, est a: quia c cum a idem ponitur. Et ita leve erit demonstrare omnia, ut dictum est.	That it does follow is obvious: for just as in the first deduction which involved three terms’ C followed from B, so in the reflex deduction of two terms, A followed from B. Let us suppose, then, that the A of the second deduction, i.e., the reflex, signifies the same thing that C signified in the first, i.e., in the direct deduction which was composed of three terms. Therefore, to state in the second deduction that “if it is B it is A” is to state the same thing as was stated in the first deduction, namely, that “if it is B, it is C.” But when it was stated in the first deduction that “if it is B, it is C,” it followed that “if it is A, it is C.” Therefore, in the circular deduction it follows that “if it is A, it is A,” since C is assumed to be the same as A. In this way, it will be easy to demonstrate all things, as has been said.
lib. 1 l. 8 n. 7 Tertiam rationem ponit; ibi: at vero nec hoc possibile est et cetera. Quae talis est. Ponentes omnia posse sciri per demonstrationem, quia demonstratio est circularis, necesse habent dicere quod omnia possunt demonstrari demonstratione circulari; et ita necesse habent dicere quod in demonstratione circulari ex conclusione possit concludi utraque praemissarum. Hoc autem non potest fieri nisi in his, quae ad se invicem convertuntur, idest convertibilia sunt, sicut propria. Sed non omnia sunt huiusmodi. Ergo vanum est dicere quod omnia possunt demonstrari, propter hoc quod demonstratio est circularis.	Then he presents the third argument (73a6) which is this: Those who suppose that everything can be known through demonstration on the ground that demonstration is circular, must grant that anything can be demonstrated by a circular demonstration and, as a consequence, grant that in a circular demonstration each of the premises can be concluded from the conclusion. However, the only cases in which this can be done are those in which mutual conversion is possible, i.e., in things that are convertible, as properties. But not all things are so related. Therefore, it is ridiculous to say that everything can be demonstrated on the ground that there are such things as circular demonstrations.
lib. 1 l. 8 n. 8 Quod autem oporteat in demonstratione circulari omnia esse convertibilia secundum positionem istorum, sic ostendit. Ostensum est in libro priorum quod, uno posito, non sequitur ex necessitate aliud; nec posito uno termino, nec posita una propositione tantum. Nam omnis syllogismus est ex tribus terminis et duabus propositionibus ad minus. Oportet ergo accipere tres terminos convertibiles in demonstratione circulari, scilicet a, b, c; ita quod a insit omni b et omni c, et haec scilicet b et c inhaereant sibi invicem, ita quod omne b sit c, et omne c sit b, et iterum haec insint ipsi a, ita quod omne a sit b, et omne a sit c. Et ita se habentibus terminis, contingit monstrare in prima figura ex alterutris, idest circulariter, omnia quaesita, idest conclusionem ex duabus praemissis et utramlibet praemissarum ex conclusione et altera conversa, sicut ostensum est in his, quae sunt de syllogismo, idest in libro priorum, in quo agitur de syllogismo simpliciter.	Now the reason is obvious why in a circular demonstration all the propositions must be convertible. For it has been shown in the book of Prior Analytics that if one thing is laid down, another does not follow of necessity, whether the thing laid down be one term or one proposition. For every syllogism must start with three terms and two propositions as a minimum. Therefore, in a circular demonstration three terms which are convertible must be taken, namely, A, B, C, such that A is in every B and in every C, and these, namely, B and C, must inhere in each other, so that every B is C and every C is B, and also inhere in A so that every A is B and every A is C. And so, the terms being thus related, it is possible, when using the first figure, to derive any one from any two circularly, i.e., the conclusion from two premises and each premise from the conclusion and the remaining premise, as we pointed out in the Prior Analytics, where we treated the syllogism formally.
Quod sic patet. Sumantur tres termini convertibiles, scilicet risibile, animal rationale mortale, et homo, et syllogizetur sic: omne animal rationale mortale est risibile; omnis homo est animal rationale mortale; ergo omnis homo est risibilis. Et ex hac conclusione potest iterum concludi tam maior quam minor. Maior sic: omnis homo est risibilis; omne animal rationale mortale est homo; ergo omne animal rationale mortale est risibile. Minor sic: omne risibile est animal rationale mortale; omnis homo est risibilis; ergo omnis homo est animal rationale mortale.	The way it is done is this: take the three convertible terms, “risible,” “rational mortal animal” and “man,” and form the syllogism: Every rational mortal animal is risible; Every man is a rational mortal animal: Therefore, every man is risible. Then from the conclusion it is possible to conclude both the major and the minor; the major thus: Every man is risible; But every rational mortal animal is a man: Therefore, every rational mortal animal is risible and the minor thus: Every risible is a rational mortal animal; But every man is risible: Therefore, every man is a rational mortal animal.
lib. 1 l. 8 n. 9 Ostensum est autem in libro priorum quod in aliis figuris, scilicet in secunda et tertia, aut non fit circularis syllogismus, scilicet per quem ex conclusione possit syllogizari utraque praemissarum: aut si fiat, non erit ex acceptis, sed ex aliis propositionibus, quae non sumuntur in primo syllogismo.	However, it has also been proved in the Prior Analytics that in figures other than the first, namely, in the second and third, one cannot form a circular syllogism, i.e., one through which each of the premises can be syllogized from the conclusion; or if one is formed, it is done not by using the premises already used but by using propositions other than those which appear in the first syllogism.
Quod sic patet. In secunda figura non est conclusio nisi negativa: oportet autem alteram praemissarum esse negativam, et alteram affirmativam. Non enim ex duabus negativis potest aliquid concludi, nec ex duabus affirmativis potest concludi negativa. Non est ergo possibile quod ex conclusione et praemissa negativa concludatur affirmativa. Si ergo debet affirmativa probari, oportet quod per alias propositiones probetur, quae non sunt sumptae. Similiter in tertia figura non est conclusio nisi particularis. Oportet autem alteram praemissarum ad minus esse universalem. Si autem in praemissis sit aliqua particularis, non potest concludi universalis. Unde non potest esse quod in tertia figura ex conclusione syllogizetur utralibet praemissarum.	That this is so is obvious. For the second figure always yields a negative conclusion. Consequently, one premise must be affirmative and the other negative. However, it is true that if both are negative, nothing can be concluded; and if both are affirmative, a negative conclusion cannot follow. Therefore, it is not possible to use the negative conclusion and the negative premise to obtain the affirmative premise as a conclusion. Hence, if this affirmative is to be proved, it must be proved through propositions other than the ones originally used. Again, in the third figure the only conclusion ever obtained is particular. However, at least one premise must be universal; furthermore, if either premise is particular, a universal cannot be concluded. Hence it cannot occur that in the third figure each of the premises can be syllogized from the conclusion and the remaining premise.
Et eadem ratione apparet quod nec in prima figura talis circularis syllogismus potest fieri, per quem utraque praemissarum concludatur, nisi in primo modo, in quo solo concluditur universalis affirmativa. Nec etiam in hoc modo potest fieri talis syllogismus circularis, per quem utraque praemissarum concludatur, nisi sumantur tres termini aequales, idest convertibiles. Quod ex hoc patet. Oportet enim ex conclusione, et altera praemissarum conversa, concludere reliquam: sicut dictum est. Non autem potest utraque praemissarum converti, cum utraque sit universalis affirmativa, nisi in terminis convertibilibus.	For the same reasons it is obvious that such a circular syllogism (through which each premise could be concluded) cannot be formed in the first figure except in the first mode, which is the only one that concludes to a universal affirmative. Furthermore, even in this mode the only case in which a circular syllogism could be formed such that each of the premises could be concluded, is when the three terms employed are equal, i.e., convertible. The proof is this: The premise must be concluded from the conclusion and the converse of the other premise, as has been stated. But such a conversion of each premise is impossible (for each is universal), except when the terms happen to be equal.