consider that derived properties are not connoted by the name that it is only "all those attributes that cannot be derived (or inferred) from others" that "are to be regarded provisionally as an assemblage of primary qualities or, in logical language, the connotation of the term ".

Surely there is something incongruous in saying that, if we define a triangle by the number of its sides, the word triangle does not connote having three angles: or, if we define it by the number of its angles, then it does not connote having three sides. Yet either of these can be inferred from the other, and they cannot, therefore, both be primary in the above sense. Again, we might define a "plane triangle as "a rectilineal figure, the sum of whose internal angles is equal to two right angles,"1 in which case having three sides and having three angles would both be derived qualities of the triangle, "requiring to be proved of it". These properties are all mutually dependent, and surely it cannot make any difference in the connotation of the term whether we choose to look on the one or the other as the primary. Does "circle" not connote "of the second degree," and "having a greater area in proportion to its circumference than any other plane figure"? Does "angle of 30°" not connote "having a sine = "? Surely they do to the mathematician! These are constituents of the notions," which he cannot by any means get rid of.2

It may, perhaps, be worth while here to remind the reader that the very possibility of thus deriving or inferring one property from another requires a considerable amount of assumption or knowledge as to the constitution of things. Thus we cannot, for example, prove that the sum of the internal angles of a rectilineal triangle is equal to two right angles without the help of the axioms of geometry. Mill's requirement of a good definition, that we should be able to evolve from it "without the aid of any other premiss" all the essential propositions that can be framed with the name defined for their subject, would therefore oblige us either to take the words "essential proposition" in a very narrow sense or else to include in most definitions many of the laws of nature. Is it or is it not essential to a

1 This Prof. Fowler mentions as a case of an attribute which "is not connoted by the term plane triangle,' but requires to be proved of it" (p. 38). Of course a definition by the number of sides or angles is much more convenient, and therefore, generally, far better than one by the sum of the angles.

2 Prof. Bain appears to go to the opposite extreme when he says (Logic, Vol. I., p. 70), that the predication of newly discovered properties is real predication only on their first announcement and is merely verbal ever after. True, the word oxygen now connotes being magnetic to him who knows of this property; but surely that it is magnetic is just as real a fact of the thing called oxygen now as it was when it was first discovered, and no amount of knowledge of the fact can make it less so. So mortality is connoted by "man but "man is mortal" is more than a mere verbal proposition, as every man finds in due time.

rectilineal triangle to have the sum of its internal angles equal to two right angles? Inasmuch as, provided that the axioms of geometry are true, no rectilineal triangle can have the sum of its internal angles equal to anything else, and no rectilineal figure that has the sum of its internal angles equal to two right angles can be anything but a triangle, I do not see how it is possible to answer this question otherwise than in the affirmative.

66

The other question as to my having apparently overlooked this distinction between Definition and Description, has already been partly answered in the foregoing. As I do not admit any fundamental distinction between primary and derived qualities, I cannot, of course, admit as fundamental the difference between Definitions and Descriptions so far as founded on that distinction. On the contrary, I look upon every real determination of the denotation of a term by means of any part of its connotation, as a real definition— although of course not all such definitions will be equally good or serviceable. On the other hand, we often do distinguish things, or classes of things, from others by means of qualities or accidents which form no part of the connotation of their names: and such propositions should undoubtedly be distinguished from definitions. If, for example, when asked, "Who is the owner of this house?" I reply, "The man in a white coat sitting on that bench; and a very disagreeable fellow he is," I have distinguished the individual in question from all others, and probably much more to the purpose than if, guided by Prof. Bain's definition of "Property," I "define" him as the man who, having either acquired the house by his own labour, or obtained it by free gift or by fair agreement from those that have so acquired it, has the right of disposing of it". But in the former case I have done it without reference to connotation. It certainly forms no part of the meaning of the term "owner of this house" to be occupying a certain seat or to be personally distasteful to me, and the man in question will not forfeit his right to the name by rising from the bench or by proving to me that I wronged him and that he is exceptionally pleasant. Between such an account and a definition the difference is fundamental, and the former may very well be called a description. In common language, however, I believe the word description" is quite as often used of cases in which the object described is hardly, if at all, distinguished from others. Thus, if I say of a man that he is "a blustering fellow with a red face and brown whiskers, who generally wears a shooting coat and grey trousers," I shall be held to have described him, although the description will suit countless others quite as well as the man it is meant for.

66

2 Prof. Bain apparently considers a description by means of distinctive accidents to be a definition, although an imperfect one: e.g., of "diamond," that "it is, quantity for quantity, the most precious substance in nature" (Log., Vol. I., p. 72). This is clearly not connoted by "diamond," for no fall in their value would affect their right to the name.

Throughout the foregoing pages I have spoken of the connotation of a name as comprising the whole of the properties common to the class named, and this is, indeed, one of the cardinal points on which the whole of the above turns; but, although I am persuaded that this is the correct view, I must not close these remarks without adverting to a difficulty out of which I must admit that I do not, at present, see my way. I mean the difficulty of finding a criterion for distinguishing between the property which is connoted and the accident which is not connoted by the name. It often happens that, so far as our experience reaches, an accident always accompanies the properties of an object, and in many cases some quite trivial accident is precisely what rises most prominently before the mind when we think of the object. If, for example, anybody always shudders when he sees a black beetle, the thought of a black beetle will suggest the feeling of disgust and the involuntary shudder as strongly as, if not more so than, the elements involved in his idea of the creature. Moreover, this effect is produced by any individual of the whole species. Every black beetle will make him shudder if he sees it; yet this fact is not connoted, even to him, by the name, and he knows that a black beetle would still be a black beetle if he were to get over his weakness. Why is this?

This difficulty is not likely to trouble us in practice. There are cases in which it may be difficult to decide whether what appears to be a property is so in reality, but generally there is little room for doubt. Gold is an object of desire to men; diamond is "quantity for quantity the most valuable substance in nature"; Switzerland is a very favourite resort for the English tourist yet everybody knows that gold and diamonds would still be gold and diamonds though they were to become so common that nobody cared to have them, and that Switzerland would still be Switzerland though never another Englishman were to climb one of its mountains. What is really the distinction?

The property, quality, or attribute is, of course, of the essence of the thing the accident is not. But this does not help us, for it is only saying again, in other words, that the one is and the other is not connoted by the name.

One feels tempted to say that by the attribute of a thing we mean, not the way in which it affects certain men and under special circumstances, but the way in which it affects all men at all times: or that we mean those modes of affecting us in which the greater part is furnished, not by the men perceiving or using it, but by the object itself. But such explanations are only very transparent cloaks to our ignorance or attempts to explain ignotum per ignotius. How can we know how a thing affects all men at all times? How can we even know how it affects anybody but ourselves? Moreover, many

1 Prof. Bain certainly cannot mean that this circumstance is connoted by the word "diamond"; but his language is not quite clear on this point. See his Logic, Vol. I., pp. 72-3.

undoubted properties or attributes can be perceived only by certain men under special circumstances, e.g., by trained observers with all the appliances of their laboratories. And to determine how much of an act of consciousness is contributed by the objective and how much by the subjective elements involved would require the knowledge of the transcendental philosopher, who has got behind phenomena to the things-in-themselves.

No doubt, satisfactory definitions of Attribute and Accident may be given. At present I can only say of them as Dr. Johnson said of light: "We all know what light is, but it is not easy to tell what it is". ERNEST CHARLES BENECKE.

"A NEW DEPARTURE IN METAPHYSICS."

THE REV. E. R. Conder in his Basis of Faith, the "Congregational Union Lecture" for 1877, dealt with the "Nature and Validity of Knowledge" in a chapter which now is removed to stand as an appendix in a second edition (Hodder and Stoughton, 1881). "I attach (says Mr. Conder) great importance to the views which I have endeavoured to establish, and which appear to me, if just, to furnish a new departure in Metaphysics." "It is not easy (he adds) to overrate the importance of dealing with the metaphysical root of scepticism;" and with a just sense of the moment of what he has essayed, he complains in his new preface that "no attempt has been made to discuss any of the main characteristic points". I therefore propose to make brief inquiry into the truth of his metaphysic and to examine its claim to novelty. He sums up his essay in twelve theses which he holds to have made good. For how many of them he claims originality he does not say. They have been already printed in MIND XIV., 297-8.

66

Theses (1) and (2) state that "human knowledge is collective," implying an interacting society of minds, and so no criticism of knowledge can be valid that proceeds from "the standpoint of a single isolated mind". That all this is abundantly evident, and that no one has ever attempted to dispute it in word or deed, we see at once so soon as Mr. Conder explains that by "knowledge he means a wellascertained and provisionally permanent consensus, such as is what we call " common sense," or when further purified, compacted, and organised "science "—e.g., Chemistry," to take his own instance. Of course these are social products. "Knowledge as we possess it could never have come into existence without the interaction of at least two minds." That goes without saying. And when has there been a critique of knowledge "from the standpoint of a single isolated mind"? It is quite true that Kant, Fichte, and Hegel base their "Kritik," "Science of Knowledge," and "Phenomenology" on the reflective analysis of their own minds. But they were not "single isolated minds," "absolutely self-educated," as Mr. Conder tells us

fishes are; nor was any one of them the impossible infant cast on a desert island without even inheritance in his brain, and suckled by some amiable wolf. The existence of other selves may be "one of the foundations of knowledge" as we possess it; but it does not follow that their existence is "not one of the problems" of metaphysic. The question "How do I come to know of them?" is quite legitimate; and we should be begging it, if we put them, as known existents, amongst our data, as Mr. Conder seems to propose.

The next four theses form an interdependent and correlevant group. (3) Knowledge involves (a) correlativity of human minds and nature, (b) correspondence and communion of human minds inter se, (c) correlativity of "the parts and elements of nature ". This perfect correlativity of all minds and things forms one harmonious universe, suggesting a "First Cause" and central intelligence. (4) It follows that an adequate doctrine of "the relativity of knowledge" concerns very much more than the mere correlation of subject and object; (5) that the possibility, truth and value of knowledge depend on the universal correlation; and (6) that there are no things-in-themselves, i.e., out of all relation. If there is anything anywhere enjoining the otium cum dignitate of a thing-in-itself, it is certainly not in this absolutely correlated universe. Most that is said in these four theses is perhaps beyond question; but it has all been said many times before. Renouvier and others have repudiated things-in-themselves. Kant only coquetted with them in a half-hearted way. Hegel may well be called the prophet of absolute correlativity; and Spinoza was his forerunner. Besides, his "causa sui et mundi" and his infinite attribute of thought or consciousness that subsumes and comprehends all the others, anticipate Mr. Conder's "first cause" and "intelligent To go on merely naming Mr. Conder's predecessors in the line of these four theses would be an almost endless task; and any discussion of their speculations lies beyond the compass of this brief paper. When (p. 381) Mr. Conder tells us that "universal knowledge oozes forth from every pore of nature," and at the same time that "all nature is built on" this ooze, one feels that his speculation at least is "passing strange". On the other hand, when he charges "the Philosophy of the Conditioned" and to some extent the Critical Philosophy with representing " as our disability that which constitutes in fact our ability, namely, the mutual relation of the human mind and the universe," one who had been indoctrinated by Hegel would be disposed to agree with him. "Metaphysicians have always been trying to get at the back of knowledge impossible quest!" See Ferrier anywhere: that sentence is the soul of his metaphysic. Again, "Properties are relations;" "that which has no properties is nothing". Exactly so. Seyn Nichts. "Absolute existence is a phrase absolutely without meaning." So have said Renouvier, Hodgson, and others.

=

Thesis (7) is to the intent that knowledge is or ought to be a body of true and certain judgments. Everybody says that. It is every

one's ideal of Truth. After difference, it is pleasant to find oneself