rentia, as being generic, must be classed with Genus. But since some Properties signify the essence, whilst

ence.

die nicht wird, sondern zeitlos ist, war vor dem concreten Einzelding, das in der Zeit wird; the idea (Idee) of the house,' he continues, existed before the particular single house came into existWhy is this given material a house? Because it is that ĥv oikią elvaι.' Aristotle was not the first to employ the imperfect in this way, for we read (Diog. L. VI. 1, 3) πрŵtos d 'Avtiσ0évns épíσato λόγον, εἰπών· λόγος ἐστὶν ὁ τὸ τί ἦν ἤ ἐστι δηλῶν.” This τί ἦν εἶναι, or essence, is closely equivalent to Plato's idéa, and is that something in anything which causes it at once to be what it is, and to be called what it is called. The fundamental error of realism lies in assuming that these two are identical. The corresponding phrase in Mill's language is connotation,' though it only covers half the meaning of essence, i.e. that which causes a thing to be called what it is called, and the absence of which would cause it to be called something else. The other and more prominent half of its meaning, i.e. the primary cause of all its properties, he, like all other nominalists, regards as non-existent, or, at any rate, unknowable. The schoolmen, on the other hand, regarding essence in both its aspects, consider the distinctions of the predicables as equally applicable to the things existing in nature, and to language. In Aristotle himself the realism is not so apparent, and the predicables are rather a classification of predicates considered in relation to their subjects, than of things and qualities; the idea of essence being, however, one of the main fundamenta divisionis.

Aristotle arrives at his predicables by a process of dichotomy, (vid. (112).) Every predicate must be either convertible with its subject or not; if it is convertible, it is an idiov (in the wider sense); if it is not convertible, it is either genus (which includes difference as being generic) or else accident: genus, if it expresses a part of the essence: accident, if it does not. Then otov is divided into opos (definition), and totov proper, according as it does or does not express the essence. Genus in like manner is divided into genus and difference, which are the two parts of a definition; the genus being that which is predicated of many things differing in species, in answer to the question what is it?' (109), and the differences the properties peculiar to those species, by which they are distinguished from one another. All this we may exhibit in a table :—

106.

others do not, let Property be divided into both the aforesaid classes, and let those which express the essence be called Definition, the rest, according to the common name given to them, Property.

Topics I. 4, § 3.

Now the problem1 and the premiss differ in their form. For when it is thus expressed, 'Is two-footed land-animal the definition of man?' and 'Is animal the genus of man?' a premiss is produced. But if it be asked whether two-footed land animal is a definition of man or no? and, whether animal is the genus (of man or no)? a problem is produced.

It will be observed that species is excluded from the Aristotelian list of predicables; and not without good reason, since it cannot properly appear in the predicate, except κarà σνμßeßηкós. It should express the relation of subject to predicate, not of predicate to subject. 'The so-called Species prædicabilis is in the manner of its predication in no way distinguished from genus. Man when predicated of philosopher expresses a part only of the essence of its subject, i.e. a portion of the attributes which the subject notion comprehends; precisely as does animal when predicated of man.'— Mansel, appendix, note A. The schoolmen, however, who regarded species as natural classes, were obliged to invent a species prædicabilis to predicate of individuals, whilst all the other heads of predicables were predicated of the species subjicibilis; e.g.

Alexander was a Macedonian (species prædicabilis).
Macedonians are Greeks (yévos).

The problem was a question proposed for dialectical discussion in an alternative form, as in the examples given in the text.

Topics I. 5, § 1.

We must say what Definition is, what Property, 107. Genus, Accident. Definition is a sentence which signi

fies the essence.1

Either a sentence is rendered in place of a noun, or a sentence in place of a sentence-for it is possible to define even some of the things signified by a sentence. But as many as make their answer (Tv åπódoσι, i.e. their definition) in any way by means of a noun, clearly do not render the definition of the thing, seeing that every definition is a kind of sentence. However, even such expressions as these must be considered to be of the nature of definition, as that 'The becoming is beautiful.' .

Topics I. 5, § 4.

Property is that which does not indeed declare the 108. essence, but belongs to the thing alone, and is convertible with it; for instance, it is a property of man to be capable of learning grammar; for if he be a man, he is capable of learning grammar, and if (a creature) be capable of learning grammar, he is a man.

Topics I. 5, § 6.

Genus is that which is predicated of several things, 109. differing in species,2 in answer to the question, 'What

Vide note 4, page 38.

2 That is differing in part of the connotation. Supposing you take two homogeneous species, and ask, What is a man? What is a horse? The answer is the common genus, i.e. an animal.

110

111.

is it?' Now we must understand such things to be predicated in answer to the question, What is it?1 as it is fitting to give in answer, when asked, What is the subject? as, in the case of man, it is fitting when asked, What is the subject? to say He is an animal.'

Topics I. 5, § 8.

6

6

Accident is that which is none of these, neither Definition nor Property nor Genus, and yet belongs to the thing, and that which can belong and not belong to one and the same individual; as, for instance, sitting may belong and not belong to the same individual. The same, too, is the case with 'whiteness'; for nothing hinders the same thing from being sometimes white and sometimes not white.

Topics I. 5, § 10.

It is clear from them2 that nothing hinders accident from becoming at a certain time and in a certain rela

Tí ẻσTI expecting for answer a substantive, is here opposed to ποῖόν ἐστι, expecting an adjective : e.g. τί ἐστι τοῦτο ; answer, ἄνθρωπος —ποῖόν ἐστι τοῦτο ; answer, λευκόν.

This passage is otherwise taken to distinguish between (1) Inseparable and (2) Separable Accidents. The former, though in: separable from the individual, are neither part of the essence, nor constant properties of the class: the latter are variable in the individual. Aristotle, however, does not seem to intend any such distinction, for he goes on to compare his two definitions of accident, in a way which would hardly be possible if he had distinguished between two kinds: (ἐστὶ δὲ τῶν τοῦ συμβεβηκότος ὁρισμῶν ὁ δεύτερος βελτίων, K.T.λ. § 9.)

2 i.e. Such questions as, Is honour or expediency preferable? Is the life of virtue or of enjoyment the sweeter? In which cases the accident 'preferable' or 'sweet' is virtually a property. Top. I. 5, § 9.

6

tion property; for instance, sitting,' which is an accident, will, when one is the only person who is sitting, be a property; and if one be not the only person sitting, it will be a property with respect to those who are not sitting. So that nothing hinders accident from becoming in a certain relation, and at a certain time, property. But simply (ie. without such limitation) it will not be property.

Topics I. 8, §§ 2, 3.

Whatever is predicated of anything must be either convertible with the thing or not.1 If it be convertible, it would be definition or property; if it signify the essence, definition-if not, property; for property was defined as that which is convertible, but does not express the essence. If it be not convertible with the thing, it is either one of the parts of the definition of the subject, or not. If one of the parts of the definition, it would be genus or differentia, since definition consists of genus and differentiæ; if not one of the parts of the definition, it is clear that it would be accident.

1 Vide note, page 39.

112.