tory." As to objects of sight," says Dr. Reid, "I "understand what is meant by an image of their figure in "the brain: but how shall we conceive an image of their "colour, where there is absolute darkness? And, as to "all other objects of sense, except figure and colour, I

am unable to conceive what is meant by an image of "them. Let any man say, what he means by an image "of heat and cold, an image of hardness or softness, an "image of sound, or smell, or taste. The word image, "when applied to these objects of sense, has absolutely no meaning.”—This palpable imperfection in the ideal theory, has plainly taken rise from the natural order in which the phenomena of perception present themselves to the curiosity.

86

The mistakes, which have been so long current in the world, about this part of the human constitution, will, I hope, justify me for prosecuting the subject a little farther; in particular, for illustrating, at some length, the first of the two general remarks already referred to. This speculation I enter upon the more willingly, that it affords me an opportunity of stating some important principles with respect to the object and the limits of philosophical inquiry, to which I shall frequently have occasion to refer in the course of the following disquisitions.

SECTION II.

Of certain natural prejudices, which seem to have given rise to the common Theories of Perception.

It seems now to be pretty generally agreed among philosophers, that there is no instance in which we are able to perceive a necessary connection between two successive events, or to comprehend in what manner the one proceeds from the other, as its cause. From experience,

indeed, we learn, that there are many events, which are constantly conjoined, so that the one invariably follows the other but it is possible, for any thing we know to the contrary, that this connection, though a constant one, as far as our observation has reached, may not be a necessary connection; nay, it is possible, that there may be no necessary connections among any of the phenomena we see and if there are any such connections existing, we may rest assured that we shall never be able to discover them.*

I shall endeavour to shew, in another part of this work, that the doctrine I have now stated does not lead to those sceptical conclusions, concerning the existence of a First Cause, which an author of great ingenuity has attempted to deduce from it.-At present, it is sufficient for my purpose to remark, that the word cause is used, both by philosophers and the vulgar, in two senses, which are widely different.-When it is said, that every change in nature indicates the operation of a cause, the word cause expresses something which is supposed to be necessarily connected with the change, and without which it could not have happened. This may be called the metaphysical meaning of the word; and such causes may be called metaphysical or efficient causes.-In natural philosophy, however, when we speak of one thing being the cause of another, all that we mean is, that the two are constantly conjoined, so that, when we see the one, we may expect the other. These conjunctions we learn from experience alone, and without an acquaintance with them, we could not accommodate our conduct to the established course of nature. The causes which are the objects of our investigation in natural philosophy, may, for the sake of distinction, be called physical causes.

*See Note (C.)

I am very ready to acknowledge, that this doctrine, concerning the object of natural philosophy, is not altogether agreeable to popular prejudices. When a man, unaccustomed to metaphysical speculations, is told, for the first time, that the science of physics gives us no information concerning the efficient causes of the phenomena about which it is employed, he feels some degree of surprise and mortification. The natural bias of the mind is surely to conceive physical events as somehow linked together, and material substances as possessed of certain powers and virtues, which fit them to produce particular effects. That we have no reason to believe this to be the case, has been shewn in a very satisfactory manner by Mr. Hume, and by other writers, and must, indeed, appear evident to every person, on a moment's reflection. It is a curious question, what gives rise to the prejudice ?

In stating the argument for the existence of the Deity, several modern philosophers have been at pains to illustrate that law of our nature, which leads us to refer every change we perceive in the universe, to the operation of an efficient cause.*-This reference is not the result of reasoning, but necessarily accompanies the perception, so as to render it impossible for us to see the change, without feeling a conviction of the operation of some cause by which it was produced; much in the same manner in which we find it to be impossible to conceive a sensation, without being impressed with a belief of the existence of a sentient being. Hence, I apprehend, it is, that when we see two events constantly conjoined, we are led to associate the idea of causation, or efficiency, with the former, and to refer to it that power or energy by which the change was produced; in consequence of which association, we come to consider philosophy as the knowledge of

*See, in particular, Dr. Reid's Essays on the latellectual Powers of Man.

efficient causes, and lose sight of the operation of mind in producing the phenomena of nature. It is by an association somewhat similar, that we connect our sensations of colour with the primary qualities of body. A moment's reflection must satisfy any one, that the sensation of colour can only reside in a mind; and yet our natural bias is surely to connect colour with extension and figure, and to conceive white, blue, and yellow, as something spread over the surfaces of bodies. In the same way, we are led to associate with inanimate matter the ideas of power, force, energy, and causation, which are all attributes of mind, and can exist in mind only.

This bias of our nature is strengthened by another association. Our language, with respect to cause and effect, is borrowed by analogy from material objects. Some of these we see scattered about us, without any connection between them, so that one of them may be removed from its place, without disturbing the rest. We can, however, by means of some material vinculum, connect two or more objects together; so that whenever the one is moved, the others shall follow. In like manner, we see some events, which occasionally follow one another, and which are occasionally disjoined: we see others, where the succession is constant and invariable. The former we conceive to be analogous to objects which are loose, and unconnected with each other, and whose contiguity in place, is owing merely to accidental position; the others to objects, which are tied together by a material vinculum. Hence we transfer to such events, the same language which we apply to connected objects. We speak of a connection between two events, and of a chain of causes and effects.* That this language is merely analogical, and that we know nothing of physical events, but the laws which regu*See Note (D.)

late their succession, must, I think, appear very obvious to every person who takes the trouble to reflect on the subject; and yet it is certain, that it has misled the greater part of philosophers, and has had a surprising influence on the systems, which they have formed in very different departments of science.

A few remarks on some of the mistaken conclusions, to which the vulgar notions concerning the connections among physical events have given rise, in natural philosophy, will illustrate clearly the origin of the common theories of perception; and will, at the same time, satisfy the reader, with respect to the train of thought which suggested the foregoing observations.

The maxim, that nothing can act but where it is, and when it is, has always been admitted, with respect to metaphysical or efficient causes. "Whatever objects," says Mr. Hume, "are considered as causes or effects, are contiguous; and nothing can operate in a time or place, "which is ever so little removed from those of its exist❝ence.' "We may therefore (he adds) consider the "relation of contiguity as essential to that of causation.”— But although this maxim should be admitted, with respect to causes which are efficient, and which, as such, are necessarily connected with their effects, there is surely no good reason for extending it to physical causes, of which we know nothing, but that they are the constant forerunners and signs of certain natural events. It may, indeed, be improper, according to this doctrine, to retain the expressions, cause and effect, in natural philosophy; but, as long as the present language upon the subject continues in use, the propriety of its application, in any particular instance, does not depend on the contiguity of the two events in place or time, but solely on this question, whether the one event be the constant and invariable fore