Aristotle · a new plain-English translation from the original language
1| for of the things predicated in common, the genus would be the one most predicated in the 'what it is.' Likewise, the study of likeness is also useful for definitions in the case of things far apart, for example, that calm is the same thing in the sea as stillness is in the air (for each is a kind of rest), and that a point is in a line what a unit is in a number, for each is a starting point. So by rendering the common item as the genus in all such cases, we will seem to be defining not inappropriately. And indeed those who give definitions generally tend to render them this way: they say that the unit is the starting point of number, and the point the starting point of a line. It is clear, then, that they are placing both under their common genus. These, then, are the instruments through which the syllogisms are constructed; and the following are the topics for which what has been said is useful. B.
2| Of problems, some are universal and some are particular. Universal, for instance, that every pleasure is good, and that no pleasure is good; particular, for instance, that some pleasure is good, and that some pleasure is not good. Common to both kinds of problems are the universal constructive and destructive arguments: for by showing that something holds of every case, we will also have shown that it holds of some case; and likewise, if we show that it holds of no case, we will also have shown that it does not hold of every case. We must speak first, then, about the universal destructive arguments, both because such arguments are common to universal and particular problems, and because those who argue tend more to put forward theses about whether something holds or not, and it is destroying such theses that dialecticians engage in. Now the most difficult thing is to convert the name proper to an accident, for holding 'in a certain respect' and not universally is possible only in the case of accidents. For from a definition, a property, or a genus, conversion is necessary: for example, if it holds of some animal to be footed and two-footed, then converting, it will be true to say that it is a footed two-footed animal. Likewise also from the genus:
3| for if it holds of something to be an animal, it is an animal. The same holds also for a property: for if it holds of something to be receptive of grammar, it will be receptive of grammar. For none of these admits of holding, or not holding, in a certain respect, but only of holding, or not holding, without qualification. But in the case of accidents, nothing prevents something from holding in a certain respect, for example whiteness or justice, so that it is not enough to show that whiteness or justice holds, in order to show that a thing is white or just; for it is open to dispute whether it is white or just in a certain respect only. So conversion is not necessary in the case of accidents. We must also distinguish the errors that occur in problems, that they are of two kinds: either by stating something false, or by transgressing the established usage of language. For both those who state falsehoods, saying that what does not hold of something holds of it, err; and also those who address things by names not their own, for example calling the plane tree a man, transgress the established naming.
4| One topic, then, is to look and see whether one has rendered as an accident what in fact holds in some other way. This error occurs most of all with regard to genera, for example if someone were to say that it is an accident of white that it is a color: for it is not an accident of white that it is a color, but color is its genus. Now it is possible for the one who states this to make the distinction in terms of the naming itself, for example that it is an accident of justice to be a virtue; but often even without such a distinction it is evident that he has rendered the genus as an accident, for example if someone should say that whiteness is colored, or that walking is being moved. For predication from any genus is never applied to the species paronymously, but all genera are predicated of their species synonymously; for the species admit both the name and the account of their genera. So the one who called white 'colored' has rendered it neither as a genus, since he has spoken paronymously, nor as a property, nor as a definition; for a definition and a property belong to nothing else, whereas many other things besides white are colored, for example wood, stone, man, horse. It is clear, then, that he is rendering it as an accident. Another topic is to look and see whether it has been said to hold of all things, or of none.
5| One must examine by species and not among an unlimited number of cases; for the inquiry proceeds more methodically and among fewer instances. One must examine and begin from the first cases, then proceed in order down to the indivisible ones, for example, if someone said that the same knowledge is of opposites, one should examine whether the same knowledge is of relatives, of contraries, of terms spoken of by privation and possession, and of terms spoken by way of contradiction; and if it is not yet clear from these, one must again divide these further down to the indivisible cases, for example whether the same knowledge is of the just and the unjust, or of the double and the half, or of blindness and sight, or of being and not being. For if it is shown in some one case that the knowledge is not the same, we will have destroyed the problem; and likewise if it holds of none. This topic converts for both destroying and constructing: for if the division, when carried out, appears to hold in all cases, or in many, one may claim the right to lay it down as universal too, or else to bring an objection as to some case where it does not so hold; for if one does neither of these, one will appear absurd in not laying it down. Another topic is to construct accounts of the accident and of that to which it belongs, either of both, in respect of each, or of one of the two, and then to examine whether something not true has been assumed as true in the accounts.
6| For example, if the question is whether it is possible to wrong a god, what is 'to wrong'? For if it is to harm someone willingly, it is clear that a god cannot be wronged, for it is not possible for a god to be harmed. And if the excellent man is envious, what is the envious man, and what is envy? For if envy is pain at the apparent good fortune of some person of decent character, it is clear that the excellent man is not envious; for he would then be base. And if the indignant man is envious, what is each of the two? For in this way it will become evident whether what has been said is true or false, for example, if the envious man is one who feels pain at the good fortune of good people, while the indignant man is one who feels pain at the good fortune of bad people, it is clear that the indignant man would not be envious. One should also take, in place of the names used in the accounts, further accounts, and not give up until one arrives at something familiar; for often, even when the whole account has been rendered, what is being sought is not yet clear, but once some account has been given in place of one of the names within the account, it becomes evident. Further, one should turn the problem into a premise for oneself and raise an objection to it; for the objection will serve as an argument against the thesis. This topic is nearly the same as looking to see whether it has been said to hold of all cases, or of none;
7| It differs in the manner. Further, one must distinguish which things ought to be called what the many call them and which not. For this is useful both for establishing and for demolishing — namely, that things should be addressed by their names as the many do, but which of the things are of such-and-such a character or not, one should no longer defer to the many about that. For instance, one should say that the healthy is what produces health, as the many say; but whether the thing proposed does produce health or not, that is no longer to be judged as the many judge it but as the doctor does.
8| Further, if a term is said in several ways, and it has been laid down that it either applies or does not apply, one must show one or the other of the several senses meant, if it is not possible to show both. One should use this where the multiple senses go unnoticed; for if it is noticed that the term is said in several ways, the opponent will object that what he himself was in difficulty about has not been discussed, but rather the other sense. This topic is convertible for both establishing and demolishing. For if we wish to establish, we shall show that one of the senses applies, if we cannot show both apply; and if we wish to demolish, we shall show that one of the senses does not apply, if we cannot show both fail to apply. Except that in demolishing there is no need to argue from an agreed concession, whether it has been said to apply to everything or to nothing; for if we show that it does not apply to some one thing, we shall have destroyed the claim that it applies to everything, and likewise if we show it applies to some one thing, we shall destroy the claim that it applies to nothing. But in establishing, one must first get it agreed that if it applies to any one thing whatever it applies to everything, provided the proposition is plausible. For it is not enough, to show that it applies to everything, to argue the case for one instance — for example that the soul of a human being is immortal, therefore every soul is immortal — but rather one must first get it agreed that if any soul whatever is immortal, every soul is immortal. And this should not always be done, but only when we are not well supplied with a single account common to all cases, as the geometer can say that the triangle has its angles equal to two right angles.
9| But if it is not unnoticed that a term is said in several ways, one should divide it into as many senses as it is said in, and both demolish and establish accordingly. For instance, if the obligatory is either the advantageous or the fine, one should try to establish or demolish both with respect to the thing proposed — namely that it is both fine and advantageous, or that it is neither fine nor advantageous. But if it is not possible to do both, one should show one of the two, indicating explicitly that it holds in one sense but not the other. The same account applies even if there are more senses into which the term divides. Again, take cases where a term is said in several ways not by homonymy but in some other manner — for instance, one knowledge is knowledge of several things either as of an end and of what conduces to the end (as medicine is knowledge of producing health and of prescribing a regimen), or as of both ends alike (as the same knowledge is said to be of contraries, since neither one is any more an end than the other), or as of what holds in its own right and of what holds accidentally (as it holds of the triangle in its own right that it has its angles equal to two right angles, but accidentally of the equilateral triangle — for it is because it is accidental to the equilateral triangle to be a triangle that we recognize it has its angles equal to two right angles). If, then, it is in no way possible for the same knowledge to be of several things, clearly it is not possible at all; or if it is possible in some way, clearly it is possible.
10| One should divide a term into as many senses as is useful. For instance, if we wish to establish, we should bring forward such senses as are possible for the thing, and divide it into only those senses that are useful for establishing; but if we wish to demolish, only those senses that are not possible, leaving the rest aside. And one should do this also in cases where it goes unnoticed in how many senses a term is used. And one must establish that this belongs to that, or does not, by the same topics — for instance, that this knowledge is of that thing either as an end, or as among the things conducing to the end, or as an accidental object; or again, that it is of it in none of the ways mentioned. The same account applies also concerning desire, and whatever else is said of several things. For desire is of this thing either as an end, for example of health, or as among the things conducing to the end, for example of being given medicine, or as an accidental object — just as with wine, the lover of sweet things desires it not because it is wine but because it is sweet. For in its own right he desires the sweet, but the wine accidentally; for if the wine is dry he no longer desires it. So he desires it accidentally. This topic is useful in dealing with relatives; for practically all such cases are cases of relatives.
11| Further, there is the substitution of a more familiar name — for instance, in place of 'exact' in an opinion, 'clear,' and in place of 'meddlesomeness,' 'a liking for activity.' For when what has been said becomes more familiar, the thesis is easier to attack. This topic too is common to both purposes, both establishing and demolishing. To show that contrary attributes belong to the same thing, one should look at the genus — for instance, if we wish to show that there is correctness and error concerning perception, and that perceiving is a kind of judging, and judging can be done correctly or incorrectly, then there would be correctness and error concerning perception as well. Here, then, the demonstration proceeds from the genus to the species, since judging is the genus of perceiving — for one who perceives judges in some way. Again, one can argue from the species to the genus; for whatever belongs to the species belongs also to the genus — for instance, if there is knowledge that is base and knowledge that is excellent, then there is also a state that is base and a state that is excellent, since a state is the genus of knowledge. Now the former topic is false for the purpose of establishing, but the latter is true. For it is not necessary that whatever belongs to the genus belongs also to the species — an animal is winged or four-footed, but a human being is neither.
12| But whatever belongs to the species necessarily belongs also to the genus; for if a human being is excellent, then an animal is also excellent. For the purpose of demolishing, the former topic is true and the latter is false; for whatever does not belong to the genus does not belong to the species either, but whatever does not belong to the species is not necessarily absent from the genus. And since it is necessary that, of those things of which the genus is predicated, some one of the species is also predicated, and whatever has the genus, or is named paronymously from the genus, must also have some one of the species or be named paronymously from some one of the species — for instance, if knowledge is predicated of something, then grammar or music or some other one of the branches of knowledge will be predicated of it, and if something has knowledge, or is named paronymously from knowledge, it will have grammar or music or some other branch of knowledge, or be named paronymously from one of them, for instance 'grammatical' or 'musical.' If, then, something is proposed as said in some way from the genus — for instance that the soul moves — one should examine whether it is possible for the soul to move according to some one of the species of motion, for instance to grow, or to perish, or to come to be, or whatever other species motion has.
13| For if it is according to none of them, clearly it does not move. This topic is common to both purposes, both demolishing and establishing; for if it moves according to some one of the species, clearly it moves, and if it moves according to none of the species, clearly it does not move. If one is not well supplied with an argument bearing on the thesis, one should examine the definitions, or the true or accepted facts, of the matter proposed, and not from one alone but from several. For it will be easier to attack once things have been defined; for the attack goes more easily against definitions. And one should examine, concerning the matter proposed, given what its existence follows, or what necessarily follows if the matter proposed exists — for one wishing to establish, on what the matter proposed will follow from being the case (for if that is shown to hold, the matter proposed will also have been shown); for one wishing to demolish, what follows if the matter proposed holds — for if we show that what follows from the matter proposed does not hold, we shall have destroyed the matter proposed. Further, one should keep an eye on the factor of time, to see whether it is anywhere out of tune — for instance, if someone asserted that what is nourished necessarily grows; for animals are always being nourished, but they do not always grow. Likewise too if someone asserted that knowing is remembering —
14| for the one belongs to past time, while the other belongs also to the present and the future. For we are said to know present and future things, for instance that there will be an eclipse, but it is not possible to remember anything except what is past.
15| Further, there is the sophistical method: leading the argument into a position for which we will have a good supply of arguments available. This will sometimes be necessary, sometimes only apparently necessary, and sometimes neither apparent nor necessary. It is necessary when, once the answerer has denied one of the points useful for the thesis, one directs the arguments to this point, and this happens to be the kind of thing for which arguments are readily available. Similarly too when, having produced an induction toward some point through what has been laid down, one tries to demolish it; for once this is demolished, the point originally proposed is demolished along with it. It is apparently necessary when it appears useful and proper to the thesis, but is not really the point toward which the arguments are directed—whether because the one sustaining the argument has denied it, or because an induction carrying general acceptance is produced through the thesis toward it, and one tries to demolish that. The remaining case is when the point toward which the arguments are directed is neither necessary nor apparently so, but the answerer happens to be refuted in some other, incidental way. One should be wary of the last of the methods mentioned; it seems to be wholly detached from and foreign to dialectic. That is why the answerer, too, should not be difficult, but should grant what is not useful to the thesis, noting whatever he grants that he does not actually accept. For questioners generally find themselves more at a loss when all such points are granted them, if they still fail to bring the argument to a conclusion. Further, anyone who has said anything at all has in a sense said many things, since a number of consequences necessarily follow from each thing; for example, one who has said that something is a man has also said that it is an animal, that it is ensouled, that it is two-footed, and that it is receptive of intellect and knowledge, so that if any one of the consequences is demolished, the original point is demolished along with it. One must also be wary of taking on the more difficult exchange; for sometimes it is easier to demolish the consequence, sometimes the original point itself.
16| For things such that necessarily only one of a pair belongs—for example, to a man either disease or health—if we are well supplied with arguments for one of the two, that it belongs or does not belong, we shall also be well supplied for the remaining one. For this converts to cover both: if we show that one of the two belongs, we shall thereby have shown that the remaining one does not belong; and if we show that it does not belong, we shall thereby have shown that the remaining one does belong. It is clear, then, that the commonplace is useful for both. Further, there is the method of taking the word over into the argument by transferring it, taking it in whatever sense best suits one's purpose rather than as the word is actually used—for example, taking "good-souled" not as meaning "courageous," as it is now used, but as meaning one who has his soul in a good state, just as "of good hope" means one who hopes for good things. Similarly too "happy," as meaning one whose guardian spirit is good, just as Xenocrates says that the happy man is the one who has a good soul; for this, he says, is each person's guardian spirit. Now since, of things, some are of necessity, some hold for the most part, and some are however it may chance, if what holds of necessity is set down as holding for the most part, or what holds for the most part as holding of necessity—either the thing itself or its opposite to what holds for the most part—this always gives an opening for an argument.
17| For if what holds of necessity is set down as holding for the most part, it is clear that he is saying it does not belong to every case, though it does belong to every case, so that he has erred; and likewise if he has said that what holds for the most part holds of necessity—for then he is saying it belongs to every case, when it does not belong to every case. Similarly too if he has said that the opposite of what holds for the most part holds of necessity; for the opposite of what holds for the most part is always said to hold in fewer cases—for example, if for the most part men are bad, then good men are fewer—so that he has erred even more if he has said that good men exist of necessity. Likewise too if he has said that what is however it may chance holds of necessity or for the most part; for what is however it may chance holds neither of necessity nor for the most part. And it is possible, even if he has not distinguished whether he has said it holds for the most part or of necessity, but the fact of the matter is that it holds for the most part, to argue against him as though he had said it holds of necessity—for example, if he said, without qualification, that men chosen by lot are bad, to argue against him as though he had said this of necessity. Further, one should also look for cases where he has set something down as accidental to itself under the guise of being something different, merely because it has a different name—just as Prodicus divided pleasures into joy, delight, and good cheer.
18| for all these are names for the same thing, pleasure. So if someone says that rejoicing is accidental to feeling cheerful, he would in effect be saying that the thing is accidental to itself. Since opposites are combined with one another in six ways, and combined in four of these they produce a contrariety, one must take up the opposites in whichever way is useful, both for demolishing and for establishing a position. That they are combined in six ways is clear: either each of the two opposites will be combined with each of the other two—and this in two ways, for example, doing well by one's friends and doing badly by one's enemies, or conversely doing badly by one's friends and doing well by one's enemies. Or both may be combined with respect to the same one thing—and this too in two ways, for example, doing well by one's friends and doing badly by one's friends, or doing well by one's enemies and doing badly by one's enemies. Or the same one thing may be combined with both, and this too in two ways, for example, doing well by one's friends and doing well by one's enemies, or doing badly by one's friends and doing badly by one's enemies. Now the first two combinations mentioned do not produce a contrariety; for doing well by one's friends is not the opposite of doing badly by one's enemies—both are choiceworthy and belong to the same character. Nor is doing badly by one's friends the opposite of doing well by one's enemies; for these too are both to be avoided and belong to the same character. And it does not seem that one thing to be avoided is the opposite of another thing to be avoided, unless one is spoken of as an excess and the other as a deficiency;
19| for excess seems to be among the things to be avoided, and likewise deficiency. But all the remaining four combinations produce a contrariety. For doing well by one's friends is the opposite of doing badly by one's friends; it proceeds from an opposite character, and one is choiceworthy while the other is to be avoided. The same holds for the others as well; for in each pairing one member is choiceworthy and the other to be avoided, and one belongs to a decent character and the other to a base one. It is clear from what has been said, then, that more than one opposite happens to belong to the same thing. For to doing well by one's friends both doing well by one's enemies and doing badly by one's friends are opposite. Likewise, if one examines each of the other cases in the same way, two opposites will turn out to belong to it. One should therefore take whichever of the opposites is useful for the thesis. Further, if there is something opposite to the accidental attribute, one should examine whether it belongs to the very thing to which the accidental attribute has been said to belong; for if this belongs, that other thing could not belong—since it is impossible for opposites to belong to the same thing at the same time. Or one should examine whether something of the following sort has been said of some subject, such that, given its existence, the opposites necessarily belong to it. For example, if someone said that the forms exist in us;
20| it will follow that they are both in motion and at rest, and further that they are both perceptible and intelligible. For the forms are held to be at rest and intelligible by those who posit that forms exist, but if they exist in us it is impossible for them to be unmoved; for when we are in motion, everything in us must necessarily move along with us. And it is clear that they would also be perceptible, if indeed they exist in us; for it is through the perception involved in sight that we come to know the shape present in each thing. Again, if an accidental attribute has been posited that has some opposite, one should examine whether the subject is also receptive of the opposite, which is what the accidental attribute is receptive of; for the same thing is receptive of opposites. For example, if someone said that hatred follows upon anger, hatred would then be in the spirited part; for that is where anger is. One should therefore examine whether its opposite, friendship, is also in the spirited part; for if it is not, but friendship is rather in the appetitive part, then hatred would not follow upon anger. Similarly too if someone said that the appetitive part is ignorant. For it would then also have to be receptive of knowledge, if indeed it is receptive of ignorance; but this does not seem to be so—that the appetitive part is receptive of knowledge. In demolishing a position, then, one should make use of this as described; but in establishing one, the commonplace is not useful for showing that the accidental attribute belongs, though it is useful for showing that it may possibly belong.
21| For if we show that the subject is not receptive of the contrary, we will have shown that the accident neither belongs nor can possibly belong; but if we show that the contrary belongs, or that the subject is receptive of the contrary, we will not yet have shown that the accident belongs, but only that it can possibly belong — that much alone will have been shown. Since oppositions are of four kinds, one should look, in the case of contradictories, at the consequence taken in reverse, both for demolishing and for establishing, and take it by induction — for instance, if man is an animal, what is not an animal is not a man. And likewise in other cases too. For here the consequence runs in reverse: animal follows on man, but not-animal does not follow on not-man; rather, in reverse, not-man follows on not-animal. In every such case, then, this is the principle to claim: for instance, if the fine is pleasant, then what is not pleasant is not fine; and if this is not so, neither is that. Likewise too, if what is not pleasant is not fine, then the fine is pleasant. It is clear, then, that the consequence based on contradiction converts both ways, running in reverse. In the case of contraries, one should look at whether the contrary follows on the contrary, either in the same direction or in reverse, both for demolishing and for establishing; and such things too should be taken by induction, so far as it is useful. In some cases the consequence runs in the same direction, as with courage and cowardice: excellence follows on the one, badness on the other, and what is choiceworthy follows on the one, what is to be avoided on the other.
22| So the consequence of these too runs in the same direction, since what is choiceworthy is contrary to what is to be avoided. And likewise in other cases. But in other cases the consequence runs in reverse — for instance, health follows on good condition, but disease does not follow on bad condition; rather, bad condition follows on disease. It is clear, then, that in these cases the consequence runs in reverse. It is rare, though, for the reverse to hold among contraries; in most cases the consequence runs in the same direction. If, then, the contrary does not follow on the contrary either in the same direction or in reverse, it is clear that in the cases under discussion too the one does not follow on the other. But if it does hold among the contraries, then in the cases under discussion too the one must necessarily follow on the other. One should examine states and privations in the same way as contraries, except that with privations the reverse case does not occur — the consequence must always run in the same direction, as sight follows on perception and blindness on lack of perception. For perception is opposed to lack of perception as state to privation: the one is a state of them, the other a privation. And one should treat relatives in the same way as states and privations, for in their case too the consequence runs in the same direction.
23| For instance, if the triple is a multiple, the third part is a fraction of many; for the triple is said in relation to the third part, and the multiple in relation to the fraction. Again, if knowledge is a supposition, the knowable is a supposable; and if sight is a perception, the visible is a perceptible. Objection: the consequence need not hold among relatives, as has been said — for the perceptible is knowable, yet perception is not knowledge. This objection, however, does not seem to be true; for many say that there is no knowledge of perceptibles. Further, what has been said is no less useful for the opposite case — for instance, that the perceptible is not knowable, since perception is not knowledge either.
24| Again, one should look at cognate words and inflections, both for demolishing and for establishing. Cognates are things of this kind: just acts and the just man are cognate with justice, and brave acts and the brave man with courage. Likewise, things productive or preservative of something are cognate with that of which they are productive or preservative — for instance, healthy things with health, and things productive of good condition with good condition. And the same holds in other cases too. Such, then, are what are usually called cognates; inflections are things like 'justly,' 'bravely,' 'healthily,' and whatever is said in that way. Words formed by inflection are also thought to be cognate — for instance, 'justly' is cognate with justice, 'bravely' with courage. All the words belonging to the same cognate group are called cognates — for instance, justice, just man, just act, justly. It is clear, then, that once any one whatever of the members of the same cognate group has been shown to be good or praiseworthy, all the rest will have been shown too — for instance, if justice is among the praiseworthy things, then the just man, the just act, and the justly done act are also among the praiseworthy things. And 'justly' will be said to stand to 'praiseworthily' by the same inflection as 'praiseworthily' stands to 'praiseworthy,' just as 'justly' stands to justice.
25| One should look not only at the term itself as stated, but also at the contrary in relation to the contrary — for instance, that the good is not necessarily pleasant, since the bad is not necessarily painful either; or, if the latter holds, so does the former. And if justice is knowledge, injustice is ignorance; and if to act justly is to act with knowledge and with experience, to act unjustly is to act in ignorance and without experience. But if these latter do not hold, neither do the former — as in the case just mentioned, for acting unjustly would seem more likely to occur with experience than without it. This topic has already been stated among the consequences of contraries; for we are now claiming nothing other than that the contrary follows on the contrary. Further, one should look at comings-to-be and destructions, and at things productive and destructive, both for demolishing and for establishing. For things whose comings-to-be are goods are themselves goods, and if they themselves are goods, so are their comings-to-be; and if the comings-to-be are of bad things, the things themselves are among the bad. With destructions it is the reverse: if the destructions are of good things, the things themselves are among the bad; if the destructions are of bad things, the things themselves are among the good. The same account holds for things productive and destructive: things whose productive causes are good are themselves among the good, and things whose destructive causes are good are themselves among the bad.
26| Again, one should look at similar cases, to see whether they hold similarly — for instance, if one branch of knowledge covers several things, so does one opinion; and if having sight is seeing, having hearing is hearing. And likewise in other cases too, both among things that are so and things that are thought to be so. This topic is useful both ways: if the matter stands thus for one of a set of similar cases, it does so for the other similar cases too; but if it does not hold for one, it does not hold for the others either. One should also look at whether it holds similarly for the one case and for the many, since in some places they diverge. For instance, if knowing is thinking, then knowing many things is thinking many things. But this is not true: for it is possible to know many things, but not to think many things at once. If, then, this does not hold, neither does the single-case claim that knowing is thinking. Further, one should argue from the more and the less. There are four topics of the more and the less: one is whether the more follows on the more — for instance, if pleasure is a good, then more pleasure is a greater good, and if doing wrong is bad, doing more wrong is worse. This topic, then, is useful for both purposes —
27| For if the increase of the accident follows the increase of the subject, as has been said, clearly it is an accident; but if it does not follow, it is not an accident. This must be grasped by induction. Another topic: when one thing is said of two things, if it does not belong to the thing to which it is more likely to belong, neither does it belong to the thing to which it is less likely; and if it belongs to the thing to which it is less likely to belong, it also belongs to the thing to which it is more likely. Again, when two things are said of one thing, if the one that seems more likely to belong does not belong, neither does the one that seems less likely; or if the one that seems less likely to belong does belong, so does the one that seems more likely. Further, when two things are said of two things, if the one that seems more likely to belong to the one does not belong, neither does the remaining one belong to the remaining one; or if the one that seems less likely to belong to the one does belong, the remaining one also belongs to the remaining one. Further, arguments can be made in three ways from belonging, or seeming to belong, equally — just as was said about the 'more', in the three topics stated above. For whether one thing belongs, or seems to belong, equally to two things — if it does not belong to the one, neither does it belong to the other, and if it belongs to the one, it also belongs to the remaining one; or whether two things belong equally to the same thing — if the one does not belong, neither does the remaining one, and if the one does, so does the remaining one.
28| In the same way also if two things belong equally to two things: for if the one does not belong to the one, neither does the remaining one belong to the remaining one, and if the one does belong to the one, the remaining one also belongs to the remaining one. From the 'more' and the 'less' and the 'equally', then, it is possible to argue in this many ways; further, from addition. If one thing added to another makes it good or white, when it was not white or good before, the thing added will be white or good, of just the sort it makes the whole. Further, if something added to what already belongs makes it more of the sort it already was, that thing too will be of that sort. The same holds in other cases too. But this topic is not useful in all cases, only in those where the excess of the 'more' happens to occur. This topic does not convert for demolishing an argument. For if what is added does not make the whole good, it is not yet clear that it itself is not good; for the good, when added to bad, does not necessarily make the whole good, nor does white added to black make the whole white. Again, if something is said to be more and less, it also belongs without qualification; for what is not good or white will not be said to be more or less good or white either; for the bad is not said to be more or less good than anything, but rather more or less bad. Nor does this topic convert for demolishing an argument either; for many things not said to admit of 'more' still belong without qualification —
29| for man is not said to admit of more and less, but he is not, on that account, not a man. The same way must be examined also in the case of 'in a certain respect', 'at some time', and 'somewhere'; for if a thing is possible in a certain respect, it is also possible without qualification. Likewise for 'at some time' or 'somewhere' too; for what is impossible without qualification is not possible either in a certain respect, or somewhere, or at some time. An objection: some people are excellent in a certain respect by nature — for instance liberal or temperate by disposition — but they are not excellent by nature without qualification; for no one is practically wise by nature. Likewise too, it is possible at some time for something perishable not to perish, but it is not possible without qualification for it not to perish. In the same way, too, it is advantageous somewhere to follow such a regimen — for instance, in unhealthy regions — but it is not advantageous without qualification. Further, it is possible somewhere for only one to exist, but it is not possible without qualification for only one to exist. In the same way too, it is fine somewhere to sacrifice one's father — for instance among the Triballians — but it is not fine without qualification. Or rather, this does not signify 'somewhere' but 'to certain people'; for it makes no difference where they are — it will be fine for them everywhere, being Triballians. Again, it is advantageous at some time to take medicine — for instance, when one is ill — but not without qualification.