a sure cable and anchorage in this life and securing to thyself against the day of thy departure hence, with a good hope, a kindly God to be thy guide.'

Had Julian succeeded in establishing the State Church of his dreams, with an ecclesiastical hierarchy modelled on that of Christianity, a debased Neo-Platonism as its philosophy, and Sallustius' treatise On the Gods and the World' as its catechism, the vocabulary of devotion would have differed but little from that of the Christian. But the project perished with its author.

Sixteen years after the death of Julian an Emperor ascended the throne who earned his title to the epithet of 'Great' by the ruthless and decisive blows which he dealt against heterodoxy and heathenism. The reign of Theodosius put an end to the public practice of pagan rites, which lingered on amongst the obscure tribes of remote Alpine valleys and in the private chapels of that section of the Roman aristocracy which made the Senatehouse the home of lost causes. The coterie led by Symmachus and Prætextatus clung fast to the old creeds; the latter, in the last dateable Mithraic inscription, recounts how, to make assurance of salvation doubly sure, he had not only been initiated into the mysteries of Eleusis but had undergone the blood-bath of the 'taurobolium' and attained the grade of pater patrum in a Mithraic confraternity. We may smile at the pathetic mingling of pedantry and pietism in this clique of reactionary grands seigneurs; but much may be forgiven to the orator who, in pleading for the restoration of the altar of victory in the Senate-house, reminded the Emperor that not by one path alone can the Great Mystery be approached'; nor can we wholly refuse our sympathy to those who, ere the once-resplendent figure of Mithras vanished for ever into the dim twilight of the Gods of Paganism, 'caught the serene surprises' of that setting sun.

H. STUART JONES.

Art. 6.

-

- THE LOGIC OF THOUGHT AND THE LOGIC OF SCIENCE.

1. Logic, or the Morphology of Knowledge. By Bernard Bosanquet. Second edition. Two vols. Oxford: Clarendon Press, 1911.

2. The Scope of Formal Logic. By A. T. Shearman. London: University Press, 1911.

3. A New Law of Thought and its Logical Bearings. By E. E. Constance Jones. Cambridge: University Press,

1911.

4. A New Logic. By Charles Mercier, M.D. London: Heinemann, 1912.

5. Formal Logic. A Scientific and Social Problem. By F. C. S. Schiller. London: Macmillan, 1912.

6. Scientific Method; Its Philosophy and its Practice. By F. W. Westaway. Glasgow and London: Blackie, 1912. 7. The Science of Logic. An Enquiry into the principles of accurate thought and scientific method. By P. Coffey. Two vols. London: Longmans, 1912.

8. Encyclopædia of the Philosophical Sciences. Vol. I: Logic. By Arnold Ruge and others. London: Macmillan, 1913.

And other works.

'ALL these books on logic, and not text-books either! How strange! Who reads them?' Such will probably be the natural comment of the average man who may have yawned over some elementary text-book at college. The subject was so very uninteresting, so painfully fixed and certain. There was a dry and wearisome discussion of the laws of thought and terms and propositions. Then we passed on to the syllogism with its figures and moods. And there were the familiar lines which remain when their meaning is forgotten :

'Barbara, Celarent, Darii, Ferioque prioris ;

Cesare, Camestres, Festino, Baroco secundis ;'

and so on.

Aristotle wrote a book on logic many centuries ago, and critics and commentators have discussed it ever since. We look at Dr Coffey's volumes and find it all there, what so many generations have learnt. The Oxford man meets it in his 'Greats' course, and occasionally a

Cambridge man may be foolish enough to take it in the 'Little Go' instead of Paley; Paley is so much easier, as all the coaches have known since the days of Tom Brown. A London man, reading alone, may find it a useful substitute for an experimental science. There is no practical work, and you can do so much more in the vacation if you reduce the number of experimental subjects. But what is all the bother about? And why should anyone wish to discuss it?

One disturbing cause is due to the old use of the term science as a synonym for logic and allied studies. As Dr Schiller aptly points out, his Oxford degree of Doctor of Science does not imply competence in natural science, merely in logic and other branches of philosophy. It is the old scholastic use of the term. Froude's famous essay on the relation between natural science and general science illustrates the meaning. What we now term science corresponds to Froude's natural science. But the paradox of the situation lay in the fact that logic, which was formerly termed science, though it professed to be a study of the methods of accurate thinking, could tell us little or nothing of the methods by which science progressed and scientific discoveries were made. Mill endeavoured to fill the gap by a treatise dealing with what he termed inductive logic, under which heading he included the formulation of generalisations and the inferring of scientific truths from the particular facts of observation and experiment. The extension has invaded the text-books and has now received the highsounding name of methodology. To this branch, which I prefer to term the logic of science, I shall refer later. But, apart from the new development, there has, during the past few years, been a vigorous onslaught on formal logic, even in its present somewhat restricted sphere.

The two most notable opponents of the old scholastic and Aristotelian logic are Dr Schiller and Dr Mercier. Dr Schiller is the more powerful controversialist. He is himself a teacher of philosophy, thoroughly acquainted with the history of the subject, its current controversies, its place in the university curriculum, the methods of teaching, and the relation of the subject to other branches of philosophy. In a volume of more than 400 pages, he gives vent to a profound dissatisfaction. What can be Vol. 221.-No. 440.

K

said against formal logic, its validity or its usefulness, we shall find clearly and sometimes bitterly expressed by Dr Schiller. He thinks the subject fundamentally unsound, doubtful mental training, dull and wearisome even to philosophers. It finds no support from the newer study of psychology, it has only failed to block scientific progress because science has ignored it; it has had a malign influence on religion, and is a potent cause of intolerance and persecution; its social effects, so far as it is operative, are without a redeeming feature. Such are some of the charges he brings against formal logic. All this sounds somewhat exaggerated, but it deserves serious notice. The average man who has forgotten his logic will agree that it certainly was dull, but, otherwise, will be somewhat surprised. He will be interested, therefore, to know that not only has Dr Schiller collected all these faults of formal logic, but has diagnosed the malady from which they arise. We will state his diagnosis in his own words:

'It is NOT possible to abstract from the actual use of the logical material and to consider "forms of thought" in themselves, without incurring thereby a total loss, not only of truth but also of meaning.'

To deal fully with this attack, and with its fundamental basis, would require a consideration, not merely of logic, but also of psychology, of philosophy and especially of pragmatism, and would take us far beyond the limits of the present short essay. Dr Schiller is the leading exponent of the pragmatist school of thought; and one of the characteristics of his philosophy is a depreciation, not only of logic, but also of reason. Hypothetically, therefore, we are bound to agree with him. The latter implies the former. In so far as reason is of small importance, a fortiori, so is logic. Dr Schiller seeks to supplement, or, it would be more correct to say, to displace logic by a science as yet unformulated, which he terms psychologic. Logic, according to his exposition, deals with thinking as the academic logician supposes it to work; psychologic is to deal with real human thinking as it actually occurs. But, if we make any enquiry concerning the content of psychologic, we can obtain no information. We

cannot displace an old logic by a old logic by a new psychologic unless and until we can gather some idea of what psychologic may be. As no one would be more ready to admit than Dr Schiller, the value of any proposal of the kind depends entirely on its practical working. If that is non-existent, we cannot assign to the vague idea any considerable significance. Dr Schiller's remarks, or many of them, can be admitted by the most formal of the formalists and the most scholastic of the schoolmen, without detriment to logic as commonly understood. In actual practice, we are bound to admit that, in many of the inferences we do make, psychology, will, and purpose enter very largely. They are not formally valid. We do take the risk that they may or may not be true; and their degree of truth can only be shown by their agreement with subsequent experience. Any enquiry, though it would not be within the limits of logic as commonly understood, concerning how, when and why, and under what conditions such actual, practical' risking' inferences are likely to contain more or less of truth would be of the utmost value. But this no pragmatist has ever given us; and moreover, to engage in such an enquiry, it is not necessary to accept the pragmatist position.

Granting this to Dr Schiller, we must disagree with his fundamental diagnosis. The ground of disagreement lies in philosophy rather than in logic. The process to which Dr Schiller objects is the necessary condition of all reasoning whatever. Any process of inference implies the abstraction of form from matter to a greater or a less extent. If I state that two and two make four, I have abstracted from the properties of material things a formal concept to which it is not easy to attach an intelligible meaning apart from the material things from which it is abstracted. The things which may be said to possess the property number have disappeared. Whether or no we have sacrificed meaning is a verbal question hardly worth discussion. Certainly we have not sacrificed truth. On reasoning of this kind all mathematics is based. There is thus a large field for the study of absolutely valid and necessary formal reasoning.

It would hardly be profitable, even if it were possible, to enter into the question of the teaching of logic in the