A

to construct a scientific mode of accomplishing what had been done by practical rules; and I suspect that the enunciation of axioms and some of the more elementary demonstrations, came at a later date than practical rules, or even than certain of the more advanced propositions. We find, in like manner, that systematized and connected Ethics, proceeding from original principles, and going on to applications, came later in the history of moral philosophy than the injunctions of parents, or the moral codes of legislators and the laws of religion. There was reasoning, and there were even rules of reasoning, before a regular Logic appeared. Metaphysics have arisen out of the contests of sects, or have been interposed as a breakwater against a tide of scepticism.

In all times and circumstances, the most effectual means of testing logical, ethical, and metaphysical principle, is by the application of it to actual cases, which should be as numerous and varied as possible. It is when appropriate examples are before it that the mind is able to appreciate the meaning of the general formulæ. It is only when it has considered them in their application to a number of diversified instances that the mind is in circumstances to pronounce them to be probably, or approximately, or altogether correct. Without observational testing such processes as definition, division, arrangement, and deduction may have rather a tempting and misleading influence. A power of dissection and inference can do as little in metaphysical as in physical investigation, that is, it is of no value at all, or may be positively injurious unless it proceed on a previous collation of facts. Minds of great logical and critical discernment are apt to go further wrong than others who are no philosophers at all, by seizing on some

1 Kant has laid down a very different maxim, declaring that examples only injure the understanding in respect of the correctness and precision of the appre hension. Speaking of examples: "Denn was die Richtigkeit und Präcision der Verstandeseinsicht betrifft, so thun sie derselben vielmehr gemeiniglich einigen Abbruch, weil sie nur selten die Bedingung der Regel adäquat erftillen (als casus in terminis), und überdies diejenige Anstrengung des Verstandes oftmals schwächen, Regeln im Allgemeinen, und unabhängig von den besonderen Umständen der Erfahrung, nach ihrer Zulänglichkeit, einzusehen, und sie daher zuletzt mehr wie Formeln, als Grundsätze, zu gebrauchen angewöhnen" (Krit. d. r. V. Trans. Log. p. 119; Rosen). This shows that Kant had no correct idea of the way in which the general rule is reached. The same view is evidently taken by many of the formal logicians of our day.

mutilated or imperfectly expressed principle, and carrying it out fearlessly, according to the rules of a rigid deduction. Of all men, those who live in the region of high abstractions, which they never bring down to realities, are most apt to go astray as in snow-drift; and when they do wander, they go faster and further wrong than other men.

At the same time, it is to be observed that the abstraction, or generalization, is not got from an outward object or event which may fall under ocular inspection or instrumental experiment, but from the operations of a mental law, which may be altogether missed by those who are exclusively engrossed with the object at which the mind is looking when the regulative principle is working. Of all men, the ardent sense-observer, or the lively picturer of external scenes, is the most inclined to shrink from reflex inspection, and is the worst fitted to propound or to judge of abstract mental principles.

3. The expression of the abstract or general truth is more or less easy, and is likely to be more or less correct, according to the simplicity of the objects to which the spontaneous conviction is directed. It is evident that some of the intuitive principles of the mind are more difficult to detect and formalize than others. Those which are directed to sensible objects, and simple objects, will be found out more easily, and at an earlier date, than those which look to more complex or spiritual objects. Thus the intuitions. regarding space-seen by the eye, and readily pictured in the imagination—were abstracted, and generalized into geometrical definitions and axioms, at an early stage of intellectual culture. It is a vastly more difficult task to express accurately, and in their ultimate form, the intuitive convictions regarding such objects as substance, and quality, and the laws involved in thought and moral perception. Still the war of contending sects, and the assaults of the sceptic, and the insidious underminings of the sophist, would compel men at an early date, to evolve some sort of logic, and we have the nature of genera and species and Jefinition, chalked out by Socrates, the principle of contradiction employed by Plato, and the formula of reasoning determined, at least approximately, by Aristotle, and, in a looser form, even in India,

more than two thousand years ago. The practical interest collecting round moral questions would also lead to an early enunciation of ethical principle, which, however, owing to the innumerable relations involved in the discharge of duty, would not, at an early stage, take a thoroughly fundamental or rigidly exact form. The crude nature of the classification embodied in the cardinal virtues, is a proof of the difficulty of expressing the ultimate laws of morality, or the supreme rule of right and wrong. A similar complexity presents itself in all inquiries in which substance and force enter as elements, and hence, while attempts have been made from the commencement of speculation to express first principles in regard to such objects, the rule announced has commonly combined intuitive and experiential elements, has been able to serve only a provisional purpose, has seldom been more than approximately correct, and ever requires to be rectified by much subsequent examination and comparison with concrete cases.

4. In their spontaneous action the intuitions never err, properly speaking; but there may be manifold mistakes lurking in their reflex form and application. I have used the qualified language that properly speaking they do not err in their original impulses ; for even here they may carry error with them. They look to a representation given them, and this representation may be erroneous, and error will appear in the result. The mind intuitively declares that on a real quality presenting itself, it must imply a substance; but what is not truly a quality may be represented as a quality, and then it is declared that this quality implies a substance. Thus Sir Isaac Newton and Dr. S. Clarke represented time and space as qualities (which I regard as a mistake), and then represented reason as guaranteeing that these qualities implied a substance in which they inhere, which is God. But the error in such cases cannot legitimately be charged on the intuition, which is exercised simply in regard to the presentation or representation made to it.

But there is room for innumerable errors creeping into the abstract or general enunciation, and the scientific application of it. For we may have made a most defective, or exaggerated, or totally inaccurate abstraction or generalization of the formula out of the

individual exercises, or we may employ it in cases to which it has no legitimate reference. From such causes as these have sprung those oversights, exaggerations, and not unfrequently glaring and pernicious errors, which have appeared in every form of metaphysical speculation. This is a topic which will fall to be resumed in next section. 5. The tests of intuitive convictions admit of an application to the abstract and general principle, only so far as the abstraction and generalization have been properly performed. It is only as applied to singulars, that our perceptions can be regarded as intuitive. The tests of intuitions, viz., self-evidence, necessity, and catholicity, apply directly only to individual convictions. To the formalized expression of them, the tests apply only mediately, and on the supposition and condition that the formulæ are the proper expression of the spontaneous perceptions.

It is always possible that the abstraction and the generalization may not have been correctly executed. In some cases, this is no more than barely possible. Whenever the object is a very simple one, presenting itself very much apart from all other circumstances, there is scarcely the possibility of error creeping in. Hence the assurance which the mind feels in regard to mathematical axioms, and the propositions founded on them by steps every one of which is intuitive. Even in regard to mathematics there may be doubts and contests, but it is only in more recondite topics, such for instance as those into which the idea of infinity enters. But in regard to intuitions which refer to objects which are more complicated, that is, which are mixed up with divers other matters in our comprehension, there may be difficulties in exactly seizing and expressing the principle, and there may therefore be doubts and disputes as to whether any given account of them is correct and adequate. It is self-evident as to this particular quality, that it implies a substance, but there is much obscurity about the general relation of substance and quality. The mind at once declares of this given effect that it must have a cause, but there may be doubts and difficulties as to the proper form in which to express the law of causation. Every man is convinced that he is the same person to-day as he was yesterday, but how few have had consciously before them the general principle of self and of personality.

SECT. II.-SOURCES OF ERROR IN METAPHYSICAL SPECULATION.

All proposed metaphysical principles are attempted expressions of the intuitions in the form of a general law. Now error may at times spring from the assumption of a principle which has no existence whatever in the human mind. I am persuaded however that the errors thus originated are comparatively few, and are seldom followed by serious consequences. In regard to the assumption of totally imaginary principles, I am convinced that there have been fewer mistakes in metaphysical than in physical science. As the intuitions of the mind are working in every man's bosom, it will seldom happen that the speculator can set out with a principle which has no existence whatever; and should he so venture, he would certainly meet with little response. It is possible also for error to arise from a chain of erroneous deduction from principles which are genuine in themselves and soundly interpreted. The mistakes springing from this quarter are likewise, I believe, few and trifling, the more so that those who draw such inferences are generally men of powerful logical mind, and not likely to commit errors in reasoning; and if they did, those who have ability to follow them would be sure to detect them. By far the most copious source of aberration in philosophic speculation is to be found in the imperfect, or exaggerated, or mutilated expression of principles which really have a place in our constitution. In such cases the presence of the real metal gives currency to the dross which is mixed with it.

In regard to many of our intuitions, the gathering of the common quality, out of the concrete and individual manifestations, is as subtle a work as the human understanding can be engaged in. This arises from the recondite, the complicated, and fugitive nature of the mental states, from which they must be drawn. But from the very commencement of speculation and the breaking out of discussion, attempts have been made to give a body and a form to the native convictions. It is seldom that the account is altogether illusory; most commonly there is a basis of fact to set off the fiction. But the principle is seen and represented only under one aspect, while others are left out of sight. It often happens that