Turning to details we must point out several technical deficiencies and mistakes, which considerably detract from the value of the book as a manual of Formal Logic. The Proposition is not distinguished from-it is in fact apparently identified with(pp. 12, 21) the Complex Term: the terms Quantity, Quality, Distributed, though frequently used, are nowhere defined: the rules for determining the distribution of the predicate, though required for application of the syllogistic rules, are not given; and Immediate Inference is nowhere treated. Certainly Conversion is dealt with, but the student is not informed that the converse is intended to be inferred from the convertend. Conversion being defined (p. 29) "the transposition of the terms of a proposition," all we learn is (on p. 30) that "in simple conversion the same quantity is retained as well as the same quality," and here "the two judgments are identical"; while in conversion per accidens the quantity is changed," and here "if the convertend is true, the converse must be true, but not vice versa". Everything said on the subject leads the student to conclude that "All P is S" is identical with "All S is P" and may be inferred from "Some S is P". For, in the first case have we not retained, and in the second have we not changed, the quantity of the judgments?

Again, the Syllogism having been defined so as to include every kind of inference (especially inductive), such rules of syllogism, as apply only to demonstrative arguments containing three terms and involving the elimination of one term are at once given. The apparent object of so defining syllogism is that the author may introduce a flagrant petitio principii on p. 130, where he proves that Induction must be syllogistic, because "syllogism is the form of Logic". Besides the error of applying the ordinary syllogistic rules indiscriminately to every kind of formal reasoning, his explanations of these rules suffer from several very serious defects. Thus (on p. 39) an illustration of Ambiguous Middle is explained thus: "The middle term is applied in the major premiss" to a certain class of things, and "in the minor to an individual". The student is bound to infer that "All beasts are dumb, my dog is a beast, therefore my dog is dumb" involves the fallacy of ambiguous middle. On p. 40, a case of Illicit Process is explained as involving the fallacy of four terms, two of these terms being a certain term used as "a particular term," and the same term used as "a universal term". This leads the student to infer that if the middle term is distributed in one premiss and not in the other, or if a term is distributed in the premiss and not in the conclusion, we commit the fallacy of four terms. On p. 42, "Some books are hard to understand, but Robinson Crusoe is not hard to understand," are given as two particular premisses from which no conclusion can be drawn, whereas, of course, we may rightly conclude "Some book is not Robinson Crusoe".

The treatment of Conjunctive and Disjunctive Syllogisms is

strangely confused. On p. 52, both are said to be based on the "principle of the excluded third".

"For instance (p. 53), the conjunctive proposition, 'No one can at the same time sing and play the flute,' leaves no middle course open between affirmation and denial; neither does the disjunctive proposition into which it can be resolved. He either sings or plays the flute, which may be further simplified into,--He either sings or he does not sing; he either plays the flute or he does not play the flute."

The punctuation here is peculiar, but apparently we are to understand that the conjunctive and the disjunctive propositions given above are equivalent! But what connexion has either of these propositions with the "principle of the excluded third"? Are we to understand that Mr. Johnstone has discovered no alternative amusement besides the practice of singing and that of playing the flute? May he not, for instance, employ himself in writing an Introduction to Logic? Even if the statement, "He either sings or plays the flute," were really equivalent (as the author says) to the two statements, "He either sings or he does not sing," and "He either plays the flute or he does not play the flute," nevertheless the exemplification of the principle of the excluded third would be not the original disjunction between singing and playing the flute, but between singing and not singing, and between playing and not playing. However, Mr. Johnstone here errs in good company. Again, on p. 53, we are told that "no member of a disjunctive should include another member," yet, on p. 55, we learn that, though two alternatives are necessarily exclusive, yet three or more are not necessarily so: "e.g., a magistrate is either a justice of the peace, or a mayor, or a stipendiary magistrate; but it does not follow that one who is a justice of the peace is not a mayor". But why should we be prevented from saying, 'A non-stipendiary magistrate is either a justice of the peace or a mayor,' and thus using two non-exclusive alternatives? author's distinction seems quite arbitrary.

The

The treatment of the Predicables and of the terms Essence and Accident, which form the basis of the exposition of Induction, is very loose and inadequate. The author begins with a circle in defining species-not the only circular definition to be foundthus (p. 19): "The application of a universal term to things is

specific, when it indicates the species of the thing to which it is applied". On p. 74, "The essence is that which makes the thing what it is"; and on p. 75, "The essence of the thing is comprehended in the idea or concept representative of it". Equipped with these two explanations of essence, which he has to reconcile as best he can, the student passes to Induction, and learns (p. 129) that one of the rules for its correct performance is, "not to mistake merely accidental qualities for those that are essential," and yet (on p. 133) it is by induction that "we come to know the essential characteristics of the species". Thus, first, the examination of our concepts will give us the knowledge of

the essence of things; then, using this knowledge, we are enabled to generalise; and, finally, the result of our generalisation discloses to us what is essential! The circular character of the I whole procedure is best shown by quoting the illustration on p. 132:

"Because the quality of having warm red blood belongs to all known birds, it must be part of their specific nature; but unknown birds have the same specific nature as known birds; therefore (according to the axiom, like reasons have like results), the quality of having warm red blood must belong to the unknown as well as to the known birds; i.e., be a universal and essential property of the species."

Circle within circle is here disclosed. The induction is made in the major premiss, for nothing more nor less is to be understood by the expression "specific nature," than "that which belongs to every member of the species". After having made this inference, it is difficult to see what need we have of the axiom, "like reasons have like results," since we have only to conclude what is syllogistically involved in the premisses.

Part Third, on Material Logic, contains in brief the author's system of philosophy, which is simple, if not profound. The key-note is Faith-faith in Reason, in Intuition, in Common Sense, in Sense-Perception, in Authority, in the Church, and in Revelation. Still (p. 192),

"To assert the existence of a subjective criterion, by which the individual intellect unfailingly assures itself of the truth of each and every one of its cognitions, would be to vindicate for the human race the gift of infallibility. . . . But a criterion there is . . . and this criterion must be Objective Evidence, i.e., the intelligibility of the thing; or-in more intelligible phraseology-the object of thought itself, which is either mediately or immediately so obviously presented to the understanding, that we necessarily judge the thing must be as it is, and cannot be other than it is."

This sentence apparently expresses the author's deepest thought. After many further elaborate divisions and subdivisions, he propounds rules to be observed for ensuring the material truth of our cognitions. The patient student will, I fancy, with much labour be able from this complex mass of advice to extract the valuable truth that, in order to be quite certain that what he is quite certain of is really true, he must be quite certain that he is quite certain.

W. E. JOHNSON.

Essai sur le Libre Arbitre, sa Théorie et son Historie. Par GEORGE L. FONSEGRIVE, Professeur agrégé de Philosophie au Lycée de Bordeaux. Ouvrage couronné par l'Académie des Sciences Morales et Politiques. Paris: F. Alcan, 1887. Pp. 592. Die Willensfreiheit des Menschen. Von FR. J. MACH, k. k. Professor am Staats-Obergymnasium in Saaz. Paderborn u.

Münster F. Schöningh, 1887. Pp. ix., 274.

A pretty full summary of M. Fonsegrive's Essay on Free-will

has already been given in MIND Xii. 621. From the mere statement of the author's results, it must have been obvious that the book was one of no ordinary merit. The historical part of the Essay is, indeed, not only a good history of the question of freewill so far as the facts are concerned, but at several points throws new light on the development of doctrines; while its critical and positive part contains one of the best recent statements of the indeterminist position. On all these grounds, some more detailed consideration seems desirable both of the historical and positive conclusions arrived at or suggested by it than was at first possible.

It must be recalled that M. Fonsegrive's indeterminism is founded on the spiritualistic conception of the soul, and that his definitions and distinctions are attached historically to those of the accepted representatives of Catholic theology. In both these respects his doctrine differs from the indeterminism of M. Renouvier, with its phenomenist basis and its divergence-as regards, in particular, the doctrine of divine prescience-from theological indeterminism. It is, however, from M. Renouvier that M. Fonsegrive has adopted his method of proof by appeal to the "practical reason"; and, partly in consequence of this, he states with exceptional clearness the modification that indeterminism makes in the conception of scientific law.

The extent of M. Renouvier's influence may be seen by comparing M. Fonsegrive's Essay with Dr. Mach's, published almost simultaneously with it, and briefly noticed in MIND Xii. 631. The essential points of Dr. Mach's indeterminist theory are exactly the same as those of M. Fonsegrive's. It is distinguished, as "relative indeterminism," at once from "absolute determinism" and from "absolute indeterminism". It is made dependent on the doctrine of an immaterial soul of which not all the operations are correlated with processes in the organism, and it is attached in the same way to the orthodox Catholic doctrine. Just as M. Fonsegrive finds it necessary above all to fortify his position against the psychological determinism of the English empirical school and its French representatives, so Dr. Mach is especially concerned to oppose the psychological determinism of the Herbartians. Both writers make a decided advance in the psychological statement of the indeterminist position; rejecting the term "liberty of indifference" and much of the old phraseology about "the will" as a power independent of "motives". Where the difference appears is in the mode of proof of their thesis. There is, indeed, an almost verbal resemblance between Dr. Mach's argument that, when "absolute indeterminism" and "absolute determinism" have been refuted, "relative indeterminism may be regarded as proved, and the same argument as stated by M. Fonsegrive. M. Fonsegrive, like Dr. Mach, contends that "relative indeterminism" is not inconsistent with the "principle of sufficient reason" and the law of causality derived from it; since

a "free" cause satisfies the principle just as much as one that acts by necessity. Nor, on the other hand, does Dr. Mach omit all appeal to the "practical reason". The difference is that with Dr. Mach the appeal to the practical reason is subsidiary when it is not surreptitious, while with M. Fonsegrive it is the argument that is regarded as finally conclusive. Dr. Mach undertakes to prove the indeterminist position by positive psychological arguments. M. Fonsegrive, in spite of an occasional inconsistency, such as the assumption that "absolute determinism" has been positively disproved, simply maintains that scientifically there is no absolute proof of determinism; that the metaphysical proof fails; that, consequently, since neither the thesis nor the antithesis is capable of proof on theoretical grounds, the practical reason must be called in to decide the antinomy in favour of freewill as an ethical postulate. The importance of this difference will become plain when the leading points of the history of the question have been briefly reviewed.

What M. Fonsegrive first of all brings out is that the whole controversy had its beginning in the efforts of the earliest humanistic thinkers to distinguish "that which depends on ourselves," Tò è' nμiv, from what depends on causes external to us. The speculative theory that was at first dominant both in Greek religion and philosophy was the conception of an external fate, beneath which man is powerless. The formation of the humanistic sciences required that this conception should be modified by a definition of the limits within which man has power, and of the nature of this power. The clear demarcation of that which depends on man was first achieved in Aristotle's definition of man's power over things as consisting in liberty of choice between equally possible alternatives. Man, according to Aristotle, not only has the power of choosing between the higher and the lower, but it is impossible to predict his choice; the future is really contingent. Aristotle, therefore, has a doctrine of free-will.

In interpreting Aristotle's formula, M. Fonsegrive makes clear, though without dwelling on the distinction, two aspects-the marking off of that which depends on man, and the ascription of a certain contingency to human action. Aristotle's doctrine, according to M. Fonsegrive's interpretation, which regards the "contingency of futures" affirmed by Aristotle as real, becomes a free-will doctrine in the modern sense. For the historical course of the free-will controversy has brought it about that indetermination of human actions has become identified in philosophy with free-will. If the controversy had taken another course, however, the Aristotelian formula in its other aspect, as the distinction of that which depends on us from that which depends on things, might equally have been called a free-will doctrine. This may be shown historically from an incidental use of the term "free-will " that occurs at the beginning of the modern period.

When humanistic speculation of a scientific kind began again in