雜記（八卷） 第 1 章 第 41/42 段

For another place and another crowd await turbulent men and their marketplace hair-splitting; but it is fitting that the lover of truth, who is also her intimate, should be peaceable even in his inquiries, approaching comprehensible knowledge through scientific demonstration, without self-love and with love of truth.

What better or clearer method, then, could there be for the beginning of such teaching than to go through in speech the term proposed, so clearly that all who share the same language may follow along? Is, then, the term 'demonstration' something like the nonsense-word 'blityri' -- a mere sound signifying nothing at all? And how is it that neither the philosopher nor the orator, nor even the judge, utters 'demonstration' as a meaningless term, nor does any litigant fail to know what it signifies -- that it exists? At any rate, the philosophers provide 'demonstration' as something that has real existence, each in his own way.

...to those who share the same language, it is agreed to signify the same thing by its designation.

Then, setting out from this point, it is necessary to inquire whether that which is signified -- the thing concerning which the discourse is -- actually exists or not;

and next, if it should be shown to exist, its nature must be inquired into precisely -- what sort of thing it is, and whether it ever oversteps the order that has been assigned to it.

But if it is not enough merely to state one's opinion simply about the thing sought (for it is equally possible for one's opponent to counter-assert whatever he wishes), but rather what has been said must be confirmed -- then, if the judgment concerning it is referred back to something equally disputed, and that again likewise to yet another disputed thing, it will proceed to infinity and remain undemonstrated; but if the confirmation of what is not agreed upon is referred to something agreed upon by all, that must be made the starting point of the teaching.

Every term, then, that is put forward must be translated into a definition that is agreed upon and clear to those who share in the investigation, which will serve as the starting point of the teaching and will explain the discovery of the things sought.

Come then, let the name 'sun' be put forward as an example.

The Stoics, then, say that it is an intelligent fire kindled from the waters of the sea.

Is not this account more obscure than the name itself, since it requires further proof to establish whether it is true? It is better, then, to say, in plain and generally intelligible terms, that 'sun' is the name given to the brightest of the bodies that move across the heaven.

For this account, I think, is more credible and clearer, and is acknowledged alike by all.

In the same way, all men would agree that demonstration is a reasoning process which, from things already agreed upon, supplies conviction for the matters disputed.

Not only demonstration, conviction, and knowledge, but also foreknowledge, are spoken of in two senses: the one scientific and certain, the other merely a matter of expectation.

Demonstration, then, in its strictest sense, is that which implants scientific conviction in the souls of learners, while the other kind is merely a matter of opinion — just as, in the case of 'man,' the one who truly is a man is he who possesses a rational mind held in common with others, while the other, though also called 'man,' is savage and brutish.

So indeed the comic poet also said:

'How charming a thing man is, so long as he is truly a man!'

And the same reasoning applies in the case of the ox, the horse, and the dog, according to the excellence or defect of the animal.

For it is by looking to the perfection proper to the kind that we arrive at the strictest sense of the terms signified.

For instance, we understand by 'physician' one who lacks nothing of the physician's skill, and by 'the Gnostic' one who lacks nothing of scientific knowledge.

And 'indication' differs from 'syllogism' in that what is indicated, being itself one thing, is indicative of one thing — as we say that being pregnant is an indication that a woman is no longer a virgin — whereas what is inferred by syllogism, though itself one thing, follows from several things, as, for example, that Python betrayed the Byzantines, if that is indeed the case; here not one but several indicative facts are taken together.

And to draw a conclusion from things agreed upon is to reason syllogistically, while to draw a conclusion from things true is to demonstrate; so that the benefit of demonstration is a composite one, arising both from taking the premises relevant to the question as true and from drawing a conclusion that follows consistently from them.

But if either the former condition does not hold, or the latter does not follow from it, then one has not demonstrated, though one has reasoned syllogistically.

For to draw a conclusion appropriate to the premises is merely to reason syllogistically, but to have each of the premises also be true is not merely to have reasoned, but to have demonstrated.

'To conclude,' as is clear even from the word itself, is to bring to an end.

And surely, in every argument, the 'end' is that which is being sought, which is also called the 'conclusion.'

No simple and single statement is called a syllogism, even if it is true; rather, a syllogism is composed of at least three such statements — two as premises and one as conclusion.

Either everything requires demonstration, or else some things are credible in and of themselves.

But if the former is the case, then by demanding a demonstration for every demonstration we shall proceed to infinity, and thus demonstration itself will be overturned.

But if the latter is the case, then these very things that are credible in themselves will become the starting points of demonstrations.

For instance, the philosophers acknowledge that the first principles of all things are indemonstrable.

So then, if demonstration exists at all, it is altogether necessary that something be credible in itself prior to it, which is called 'first' and 'indemonstrable.'

All demonstration, therefore, is traced back to an indemonstrable conviction.

There would also be other starting points for demonstrations, besides the source that arises from conviction — namely, the things that appear clearly evident to sense-perception and to the intellect.

For the things that occur with respect to sense-perception are simple and indissoluble, while the things pertaining to the intellect are simple, rational, and primary; and the things generated from these are composite, yet no less evident and credible — indeed even more rational than the primary things.

It is, then, of what is consistent and what is contradictory that we all by nature possess an innate faculty of judgment proper to reason itself.

If, then, an argument is found to be such that it is able to procure conviction for what is not yet believed, from what is already believed, we shall say that this very thing is the essence of demonstration.

It has been said that both the genus of conviction and that of demonstration are twofold: the one merely producing persuasion in the souls of the hearers, the other producing scientific knowledge.

If, then, one begins from things evident to sense-perception and to the intellect, and then draws the appropriate conclusion, one truly demonstrates; but if one begins only from things reputable, yet not primary — that is, neither evident to sense-perception nor to the intellect — then if one draws an appropriate conclusion, one will indeed reason syllogistically, but will not thereby produce a scientific demonstration; and if the conclusion is not appropriate, one will not even reason syllogistically at all.

Demonstration differs from analysis:

For each of the things demonstrated is demonstrated through certain other things which are themselves demonstrated, and these in turn were previously demonstrated by yet others, until we run back up to things credible in themselves, or to things evident to sense-perception and to the intellect — and this process is called 'analysis.'

Demonstration, on the other hand, is when one proceeds from the first principles to the thing sought, passing through all the intermediate steps.

The man skilled in demonstration, then, ought to take great care regarding the truth of his premises, but be unconcerned about their names — whether one wishes to call them axioms, propositions, or premises. Likewise, he should take equally great care regarding what is concluded from what underlying premises, but ought to care least of all whether one wishes to call the argument 'conclusive,' 'concluding,' or 'syllogistic.'

For I say that there are these two things which the man of demonstration must observe in every case: to take his premises as true, and to draw from them a conclusion that follows consistently — which some also call the 'inference,' that which is drawn upon the premises.

Concerning everything that is sought, in each problem, various premises are needed, but ones appropriate to the problem proposed; and it is necessary to convert the very thing proposed into a statement, and it is fitting that this statement be one acknowledged by all.

But if the premises taken are not appropriate to the problem, it is impossible to discover anything properly regarding it, since the whole nature of the problem — which is also called the 'question' — remains unknown.

In all matters of inquiry, therefore, there is something known beforehand (which, being altogether credible in itself, is believed without demonstration), which must be made the starting point of the inquiry into them, and the criterion for what seems to have been discovered.

For every inquiry proceeds to its discovery from pre-existing knowledge.

It is possible that the knowledge which pre-exists concerning anything sought is, in some cases, knowledge merely of its essence, while its activities remain unknown — as with stones, plants, and animals, whose operations we do not know — or knowledge of its affections, or of its powers, or, simply put, of one of the properties belonging to existing things.

And at other times, some of these powers or affections may be known, as, for instance, the desires and passions of the soul, while the essence remains unknown and is being sought.

For it is by taking hold in our understanding of the concepts of both essence and activity that we thus proceed to inquiry.

And there are cases in which, even while knowing both the activities and the essences of things, we remain ignorant of their affections.

The method of discovery, then, is as follows:

For one must begin from the very act of understanding the problems themselves.

Often, indeed, the form of the expression deceives and confuses and disturbs the understanding, so that it is not easy to discover from what kind of distinction the difficulty arises — as, for example, the question whether the unborn fetus is not an animal.

For, having some concept both of 'animal' and of 'the unborn,' we inquire whether it belongs to the unborn to be an animal — that is, whether the capacity for self-movement, and further, for sense-perception, belongs to the essence that is being carried in the womb.

So that the inquiry concerns activities and affections, based upon an essence already known beforehand.

Straightaway, then, one must ask the questioner in turn what exactly he means by 'animal.'

For this above all must be done, whenever we observe that a name has been extended to various uses.

And one must investigate whether the meaning signified by the appellation is disputed or is agreed upon by all.

For if he says that he calls 'animal' whatever grows and is nourished, we shall further ask him in turn whether he also considers plants to be animals; and then, if he answers in this way, it is now necessary to show that the unborn fetus grows and is nourished.

For Plato calls even plants 'animals,' since they partake of the third kind of soul, the appetitive; but Aristotle thinks that plants partake of the vegetative and nutritive soul, yet does not think it right to call them 'animals' as well.

For he holds that only that which partakes of the other kind of soul, the sensitive one, deserves to be called 'animal.'

The Stoics, however, do not call the vegetative faculty 'soul' at all.

But if the one who proposed the question denies that plants are animals, we shall show that he is contradicting himself.

For having defined 'animal' by being nourished and growing, yet also declaring that a plant is not an animal, he seems to be saying nothing other than that what is nourished and grows both is and is not an animal.

Let him say, then, what he actually wants to learn: whether the fetus in the womb grows and is nourished, or whether it has a share in some kind of sense-perception, or even in movement by impulse.

For according to Plato, the plant is both ensouled and an animal; but according to Aristotle, it is not yet an animal — for it lacks the sensitive faculty — though it is already ensouled.

For him, then, 'animal' is a sentient, ensouled substance.

But according to the Stoics, a plant is neither ensouled nor an animal.

For 'animal' is an ensouled substance.

If, then, the animal is ensouled, and the soul is a sentient nature, it is clear that what is ensouled is already sentient.

If, then, the one who raised the question, when asked in turn whether the fetus in the womb is an animal, should say that he applies the name 'animal' to what is nourished and grows, he has his answer.

But if he should say that what he is actually inquiring into is whether the fetus already has sense-perception, or whether it is even capable of self-movement by some impulse —

— then the investigation of the matter at once becomes clear, since the deception concerning the name no longer persists.

But if he does not answer the counter-question, nor is willing to say what he actually has in mind, or in reference to what thing he applied the name 'animal' when he made his proposition, and instead bids us to make the distinction ourselves, he would thereby be recognized as merely a contentious wrangler.

But if, there being two possible ways of proceeding — the one by question and answer, the other by continuous exposition — he has refused the one, then let him listen while we go through everything pertaining to the problem in the other way.

Then, once we have finished, it will be open to him at that point to examine each point individually, one by one.

But if he should attempt to interrupt the inquiry by asking questions, it is clear that he does not even wish to listen.

But if he should choose to answer, first of all he must be asked to what thing the name 'animal' applies; and when he answers this, he must again be asked what he means by 'that which is conceived' or 'that which is in the womb' -- whether he wishes it to signify the not-yet-formed things and the animals and the very seed already deposited in the womb, or only the articulated things already formed, which are called embryos.

One must now bring the matter proposed to its conclusion in due order, and give instruction.

But if, without answering himself, he should wish us to speak, let him hear:

Since you are unwilling to say according to which signification you assert what you have proposed (for in that case I would not be making my argument about significations, but would already be examining the things themselves), know that you have done something like this: as if you had proposed the question, whether the dog is an animal.

Which dog? For I could go on speaking in turn about the land-dog, the sea-dog, and the star in the heavens, and also about Diogenes, and the other dogs one after another.

For I could not divine whether you are asking about all of them or about some one in particular.

Therefore, what you would in any case do afterward -- namely, let it be learned about which one you are inquiring -- say clearly now.

But if you are quibbling about names, then this very name, 'that which is conceived,' is evidently to everyone neither an animal nor a plant.

But it is a name, and a sound, and a body, and a being, and a something, and anything else rather than an animal.

And if this is indeed what you proposed, you have your answer.

Nor, moreover, is that which is signified by the name 'that which is conceived' an animal; rather, that is incorporeal, and a sayable, and a matter of fact, and a thought, and anything else rather than an animal.

But the nature of the animal itself would be something else again.

For it was clearly being shown concerning the very thing under inquiry -- I mean, what sort of nature the embryo has.

The problem concerning what is signified by the name 'animal' is a different problem.

I say, then, that if by 'animal' you mean that which is capable of sensation and of movement by impulse, then that which is simply moving by impulse and perceiving is not thereby an animal --

for it is also possible to be asleep, or not to perceive when the objects of sense are not present; but what is by nature capable of either impulse or sensation is the mark of an animal.

For something of this sort is signified by these words: first, whether that which is conceived is already able to perceive or to move by impulse -- which is precisely what is proposed for consideration; and second, whether that which is conceived will at some point be able to perceive or to be moved by impulse -- a signification concerning which no one inquires, since it is obvious.

The question raised was whether the embryo is already an animal or still a plant; and then the name 'animal' was translated into a formula, so that the matter might be clear.

Having found, then, that it is by sensation and by movement according to impulse that it differs from that which is not an animal, we again distinguished this from things closely bordering upon it, saying that one thing is that which is such potentially -- which is not yet perceiving and moving, but will at some point be such -- and another thing is that which already exists as such in actuality; and of this latter, one part is already actually functioning, while the other is capable of functioning but is at rest or asleep.

And this is the very thing under inquiry.

For one must not say that the embryo is an animal on the ground that it is nourished -- this belongs to those who depart from the essence of the matter under inquiry, and turn their mind instead to accidental attributes unrelated to it.

Common to all things said to be discovered is demonstration, which is an argument that establishes one thing as true on the basis of other things.

And the things from which the matter under inquiry must be established need to be agreed upon and known by the one who is learning.

And the starting point of all these is that which is evident to sense-perception and to thought.

The primary demonstration, then, is composed out of all these; but the demonstration that concludes something further by means of things already previously demonstrated is no less trustworthy than the former, yet it cannot be called primary, because it is not concluded from primary premises.

The first kind, then, of the threefold distinction among matters of inquiry has been shown, namely: that, while the essence is known, something of its activities or affections is unknown. The second distinction of problems is that in which we all know the activities and affections, but are ignorant of the essence -- as, for instance, in which part of the body the ruling faculty of the soul resides.

The same method of demonstration applies also to the third problem.

Some, then, say that it is not possible for a single animal to have more than one principle.

Now it is indeed not possible for a single animal to have more than one principle of the same kind; but there is nothing absurd in its having several principles differing in kind.

It is clear that, beginning from itself, it will first of all invalidate itself.

Either, then, it concedes that something is true and that suspension of judgment need not be exercised about everything, or it insists in saying that nothing is true -- and it is clear that, in that case, not even it will previously have spoken truly.

For either it itself speaks truly, or it does not speak truly.

But if it speaks truly, it unwittingly concedes that something is true; while if it does not speak truly, it leaves standing as true those very things which it wished to abolish.

For at the very point where the abolishing suspension of judgment is shown to be false, at that same point the things being abolished are shown to be true -- just like the dream which asserts that all dreams are false.

For, being destructive of itself, it thereby becomes confirmatory of everything else.

And, in general, if it is true, it will make its beginning from itself -- being a suspension not of something else first, but of itself.

Furthermore, if anyone apprehends that he is a man, or that he is suspending judgment, he is evidently not suspending judgment.

For how, in the first place, could he even have entered into the dispute at all, if he were suspending judgment about everything? And how could he have answered the question put to him? For on this very point he is evidently not suspending judgment -- indeed, he even declares outright that he is suspending judgment. Hence, if, in obedience to them, we must suspend judgment about all things, we shall first suspend judgment about suspension of judgment itself -- whether or not it is to be trusted.

Furthermore, if this very thing -- namely, not knowing the truth -- is the truth, then nothing true is being granted by that man in the first place.

But if he shall say that even this -- the being ignorant of the truth -- is itself disputable, then by this very statement he grants that the truth is knowable, since with regard to it he plainly does not confirm suspension of judgment.

A sect is an inclination toward doctrines; or, as some say, an inclination toward many doctrines that have mutual consistency with one another and with the phenomena, and that tend together toward living well.

And a doctrine is a certain rational apprehension, while apprehension is a settled disposition and assent of the mind.

Not only the skeptics, but every dogmatist as well, is accustomed to suspend judgment on certain matters -- either on account of the weakness of his own judgment, or on account of the obscurity of the matters themselves, or on account of the equal force of the opposing arguments.

The methods and starting-points of inquiries concern, and consist in, these matters. Before definitions, demonstrations, and divisions, one must set forth in how many ways the thing under inquiry is spoken of; equivocal terms must be dealt with, and synonymous terms must be clearly arranged according to their meanings.

Next, one must inquire whether the matter proposed is one considered in relation to other things, or is taken in itself; and after this, whether it exists, what it is, what belongs to it -- or in this way too: whether it exists, what it is, and why it exists.

Toward the consideration of these matters, knowledge of particulars and of universals contributes together, as do prior things, and the divisions of these into their differences.

Induction, then, supplies the universal and the definition; divisions supply the species, the essence, and the individual; the consideration of the number of ways a term is used supplies the grasp of what is proper to it; while the raising of difficulties augments, along with the differences arising from them, both the demonstrations and, in another way, the theoretical consideration itself and the things attendant upon it.

And from all of these together, both knowledge and truth are established.

Again, that which is summed up from division becomes a definition;

for the determination is taken both prior to the division and posterior to it --

prior, when it is conceded or proposed; posterior, when it is demonstrated.

And also according to sense-perception, the universal is summed up out of particulars.

For the starting point of induction is sense-perception, and its end is the universal.

Induction, then, does not show what a thing is, but that it exists or does not exist; division, however, sets forth what a thing is; and determination, like division, teaches the essence and what a thing is, but not whether it exists; while demonstration clarifies all three -- whether it exists, what it is, and why it exists.

And there are also some definitions that include within themselves the cause as well.

Since we know a thing scientifically when we perceive its cause, and there are four causes -- matter, the moving cause, the form, and the end -- the determination of a thing will accordingly be fourfold.

One must, then, first take the genus -- that is, the nearest of the things above it -- and after this, the proximate differentia.

And the continuous series of differentiae, being cut and successively divided off, completes the essence of the thing.

But it is not necessary to state all the differentiae of each thing, only those that constitute the species.

By means of division we proceed back to the simpler and more primary things.

We divide the genus of the thing under inquiry, then, into the species contained within it -- for instance, in the case of man, we divide 'animal,' which is the genus, into the species that appear within it, mortal and immortal; and thus, ever cutting the composite things that seem to be genera into simpler species, we arrive at that which is not the thing sought, and no longer admits of further division.

For having divided 'animal' into mortal and immortal, and then in turn divided the mortal into land-dwelling and aquatic, and again the land-dwelling into winged and footed, and so on -- by continuing to divide the species proximate to the thing under inquiry, which itself also comprises the thing sought, we shall, by this cutting, arrive at the simplest species, which comprises nothing else but the thing sought alone.

For again we divide 'footed' into rational and irrational.

And then, selecting from among the species obtained by division those most proximate to man, and combining them into a single formula, we render the definition of man, which is: a mortal, land-dwelling, footed, rational animal.

Hence, division occupies, with respect to the definition, the place of matter, seeking out the simple elements of the name; while the definition holds the place of the craftsman and maker, combining and constructing them, and thereby presenting the knowledge of the thing that is.

Definitions are definitions neither of the things themselves nor of the Ideas; rather, we say that definitions are the interpretative formulas of the universal conceptions that we have of the things in question.

For it is of these conceptions, too, that the divisions are made.

Of divisions, one kind divides the thing divided into species, treating it as a genus; another divides it into parts, treating it as a whole; and another divides it into its accidents.

The division of a whole into its parts, then, is conceived for the most part according to magnitude; but the division into accidents can never be exhaustively completed, seeing that every existing thing must also, without exception, possess an essence.

Hence both these divisions are unsound, and only the division of a genus into species is approved, by which both the identity according to genus and the difference according to the specific differentiae are characterized.

The species is always considered as being in some part, yet not conversely: if something is a part of something, it does not follow that this part will also become a species.

For the hand is a part of man, but it is not a species.

And the genus is inherent in the species — for 'animal' is present both in man and in ox — but the whole is not inherent in its parts, for man is not present in his feet.

Therefore the species has more authority than the part, and whatever is predicated of the genus will also be predicated of the species.

It is best, then, to divide the genus into two species; failing that, into three.

The species, then, when divided in a more generic manner, are characterized by 'the same' and 'the other'; and afterward, when divided by what is signified in a generic way, they are characterized accordingly.

For each of the species is either a substance (as when we say:

'Of existing things, some are bodies, some are incorporeal') or a quantity, or a quality, or a relation, or a where, or a when, <or a position, or a having,> or a doing, or a being-affected.

Whoever has knowledge of any given thing whatsoever will also be able to render its definition, since one who is unable to comprehend and define something by reason could never have knowledge of it.

And it happens that from ignorance of the definition arise many disputes and deceptions.

For if he who knows a thing has the knowledge of it in his mind, and is also able to make clear in speech what he has in mind, and the interpretation of what is in the mind is the definition, then it is necessary that he who knows the thing be able also to render its definition.

Differentia, too, is included together with definitions, occupying the place of a sign for the definition.

For instance, when 'capable of laughter' is added to the definition of man, it makes the whole: 'a rational, mortal, land-dwelling, two-footed, laughter-capable animal.'

For the things added to a definition by way of differentia are signs proper to the things in question, but they do not disclose the nature of the things themselves.

For instance, they say that the differentia is the rendering of a property.

...of the thing in definitions, it is necessary to take the first genus as the originating and foundational one.

In longer definitions, then, there is the multitude of species discovered through the ten categories; but in the briefest ones, the preceding proximate species that are taken indicate the substance and nature of the thing; and the shortest definition consists of three elements — the genus and two of the most necessary species.

This is done for the sake of brevity.

We say, then, that man is 'the rational, laughter-capable animal.'

One must also take in addition whatever belongs peculiarly to the thing being defined — either its proper excellence, or its proper function, and certain other such things.

The definition, then, being explanatory of the substance of the thing, is unable to comprehend the nature of the thing with exactness, but it makes its disclosure of the substance through the most proper species, and the definition holds the substance almost entirely within a quality.

There are two supreme causes productive of suspension of judgment: the first is the manifold and unstable character of human opinion, which by its nature tends to generate discord — whether of people with one another, or of each with himself; the second is the discord found among existing things, which likewise, and reasonably, proves productive of suspension of judgment.

For being unable either to trust all impressions on account of their conflict, or to distrust all of them (since the very impression that declares all to be untrustworthy, being itself one among them all, would be included in that same condemnation), or to trust some and distrust others on account of their equal standing, we were brought down to a suspension of judgment.

Now, of these very principal causes of suspension of judgment, the unsteadiness of the mind is generative of discord, and discord is the proximate cause of suspension of judgment; whence life is full of law-courts and deliberative councils and assemblies, and in general of the pursuit and avoidance of what are called goods and evils — all of which are evidences of a mind that is perplexed and wavering before the equal weight of opposing things.

And the bookcases are full of the treatises and compositions and works of men who disagree in their doctrines and yet have persuaded themselves that they know the truth among existing things.