&c.) suggest to us the real or tangible extensions with which they are arbitrarily connected in the system of nature, is the Berkeleian solution of this problem.

It is at this point, too, that the principal ambiguity of the Essay appears. It is argued, that visible objects and their accompanying organic sensations cannot exist without the mind,' and that they have no resemblance to what is external: at the same time it seems to be taken for granted by Berkeley that we are percipient of absolute externality in touch. We appear to be told that we cannot see, but that we can and do touch, a distant world that exists independently of our intelligence. The reader is ready to say that we are as much (or as little) conscious of th's perception when we see as when we touch. This difficulty is relieved in the Principles of Human Knowledge, where the objects of sight and touch are put on the same level, and the possibility of the absolute or purely external existence of either is denied. The term 'external' may still be applied to our tactual and locomotive sensations or ideas, as contrasted with our visual; but that happens, according to Berkeley, not because of the ideal nature of what we see, and the absolute externality of what we touch, but only because our tactual and locomotive sensations are more steady and practically important, while 'externality' is a term appropriate to what is found to have these characteristics. In the developed philosophy of Berkeley, the tangible world itself, as well as its visible signs, consists of objects which cannot be independent of a mind; their esse is percipi. And, the distinction between organic and extra-organic existence is merged in the more general truth, that the human body and its organs, along with the entire extra-organic world, is itself dependent on mind. The organised body is contained in mind, and not mind in the organised body. Externality, in the deeper meaning of an absolute independence of all mind, is argued to be impossible, because contradictory; while 'externality,' either as significant of what is extraorganic, as contrasted with what is within our vital organism, or of what is tangible, as distinguished from what is visible, is simplified and rationalised. But Berkeley preferred, as he tells us, to insinuate his metaphysical system by degrees, and he has in consequence to some extent exposed his Essay to the charge of incoherence.

He seems in some passages (e. g. sect. 41) to take for granted that the visibility of distance in the line of sight would imply externality -i.e. independence of sensation- on the part of the objects we see, as if this independence could be determined at all by what we see or do not see. Even if distance, in the sense of extra-organic externality, could be seen, it would not follow that it was absolutely

independent of mind: rather the contrary, on his own principles. That we are not conscious of seeing it only proves that an 'ambient space' is not an immediate object of sight-not that coloured extension, in its other and visible dimensions, is purely ideal*.

The way in which visual phenomena suggest extension in its third dimension, and objects as actually at a distance from the organ of vision, having been thus explained in the first part of the Essay, the second and third parts (sect. 52-120) profess to prove the invisibility of real extension in two other relations-magnitude and locality.

The proof consists of a reference to consciousness, for the purpose of purifying the meaning of the term 'real,' as applicable to sizes and situations; and an induction of the actual signs of real magnitude and situation. The only real sizes, shapes, and situations must, it is argued, be tangible ones, dependent on tactual and locomotive experience. The reason given for this is, that they only are steady, uniform, and practically connected with our physical comfort in this life of sense (sect. 55; 59). The signs (which are all arbitrary) of these tangible realities are partly visible, and partly muscular sensations in the organ of sight (sect. 56).

This theory of antithetical phenomena-visual and tactual-connected by arbitrary signs, is, in these two parts of the Essay, applied to solve two celebrated difficulties in optics

1. The fact that the size of the horizontal moon is greater to the eye than that of the moon in its meridian (sect. 67-87). This problem of the visible appearance of the horizontal moon gives rise to some curious incidental discussions-in particular one regarding minima visibilia, or units of colour, pluralities of which constitute the various degrees of visible extension (sect. 79, &c.).

2. The physical fact that we see objects erect only on condition of the formation of inverted images of them on the retina (sect. 88-120). The explanation of this is sought for in a thorough-going application of the principle, that what we see and what we touch are extensions numerically different while nominally the same. In this part of the Essay

The reasoning against the visibility of distance is not to be confounded with that against the absolute externality or non-ideal character of what we see. The proof of each thesis may be put thus

1. The visibility of any distance implies the visibility of an interval between two visible points; but we are conscious of seeing only one point in the line of sight:

therefore distance in the line of sight is invisible.

2. The absolute externality of what we see implies its independence of sensation; but seeing is simply being conscious of colours, which are sensations: therefore what we see is not independent of sensation-is not without the mind' or absolutely external.

the antithesis between the visible and the tangible world—the world of coloured extension and the world of resistant extension-is pushed to the utmost. The 'high and low' of the visible world is not the 'high and low' of the tangible world (sect. 91-106). In no case is there any resemblance between these two extensions; not even when the number of visible objects happens to coincide with the number of tangible objects of which they are signs, e. g. of visible and tangible fingers on the human hand. The mind constitutes its concrete units in an arbitrary way. At first we should not distinguish our feet as two, nor our fingers as five we should not spontaneously recognise either a foot or a finger as one thing. A man born blind would not, on first seeing, parcel out phenomena as those do who have learned by experience what those phenomena are which may be most conveniently or most scientifically conjoined in thought, and afterwards spoken about as qualities of particular substances (sect. 107-110). On the whole, it is argued, that the famous difficulty about erect vision is one which illegitimately takes for granted a relation between the visible and tangible worlds can be shown, by analysis of our sense-consciousness, not to exist. The visible pictures on the retina no more resemble real things than what we see resembles what we touch. A purification of visual consciousness, by reflecting upon it, is assumed to be all that is required to remove the difficulty, and to unfold the true theory of vision-which however cannot become familiar to us without habitual thoughtfulness, and a precise use of the part of language which refers to seeing (sect. 111-120).

The fourth part of the Essay (sect. 121-145) seeks to prove not only the numerical but also the specific or generic difference of the objects of sight and touch. Its thesis is, that sight and touch have no sensible object, nor any abstract notion even, that is common to the two senses. Extension or space has the best claim to these characteristics. If its claim can be disposed of, the a fortiori argument may be applied to all other claimants. Now, coloured extension, which alone is visible, bears no resemblance to resistant extension, which alone is tangible.

And, if these concrete or sensible extensions have nothing in common, the dispute, it is maintained, is at an end. The one possible extension is a sensible extension (sect. 123). Extension in the abstract is incomprehensible and absurd, and cannot therefore be an object either of sight or touch, still less the common object of both (sect. 124-126). The only question is, whether the (so-called) extension we see is in any respect similar to the extension we touch. That they are

absolutely heterogeneous has been proved, it is alleged, in the preceding parts of the Essay.

What remains is to marshal this scattered evidence, and to guard the foregoing conclusions against objections. This is attempted in sections 128-146.

The enunciation of the summary generalization, constructive principle, or New Theory proper (sect. 147, 148), forms what may be regarded as the fifth part of the Essay.

The last part of the Essay (sect. 149-160) relates to some deductions from the Theory which, with one exception, Berkeley declines to pursue. He restricts himself here to an inquiry as to the sort of extension with which geometry has to do—an investigation in harmony with the favourite mathematical studies of his youth. In concluding that tactual or real, and not visible or symbolical, extension is the only object reasoned about in geometry, he imagines the condition of Idomenians or unembodied spirits, endowed with only a visual sense-experience, and endeavours to determine what their conception of solid and plane extension should be (sect. 153, &c.). In arguing that planes of colour are as invisible as solids of colour (sect. 154-158), he opens the way to further refinements in the interpretation of our visual consciousness which have not escaped the attention of recent psychologists.

Whether we can see more than non-resembling, arbitrary signs of the real distances, sizes, shapes, and situations of things, is a question of fact. It may seem accordingly that experiments upon the visual and tactual consciousness of persons who are in circumstances to test the conclusions of the foregoing reflective analysis, is a suitable means for verifying its conclusions. Of this sort is the experience of persons who have been relieved from born blindness; the notion of extension and externality possessed by individuals originally destitute of the visual sense, who have not been so relieved; the experience (if it can be got) of those absolutely void of all tactual and locomotive experience, but endowed with visual; and even the phenomena of sight in infants of the human species, and in the lower animals.

Berkeley apparently made no observations of his own in this field. In various parts of the Essay (sect. 41, 42, 79, 92-99, 103, 106, 110, 128, 132-137), he infers, from the contrast of our visual and tactual consciousness, what the first experience of those rescued from born-blindness must be; he speculates about the experience of

'unbodied spirits,' able to see, but originally destitute of the sense of touch (sect. 153-159); and in the Appendix he refers, in corroboration of his theory, to a newly reported case of one born blind who had obtained sight. But he everywhere treats his Theory independently of these extremely delicate and difficult investigations. His testing facts were sought elsewhere. Nor is it difficult to understand his comparative indifference to empirical experiments. An appeal to consciousness on behalf of the principle, that we cannot see tangible objects, nor touch visible ones-with inductive evidence that our tendency to unite certain tangible and visible qualities in the same substance is a result of the constant association of the latter with the former-appeared to him to fulfil all the conditions of proof. Any apparently original fusion of tactual with visual experience was then dismissed as illusory. And indeed those physiologists and mental philosophers who have since tried to determine what vision in its purity is, by observation and experiment, in cases either of communicated sight, or of continued born-blindness, have illustrated the truth of Diderot's remark-"préparer et interroger un aveugle-né n'eût point été une occupation indigne des talens réunis de Newton, Des Cartes, Locke, et Leibnitz*."

Berkeley's New Theory has been quoted as an almost solitary example of a discovery in metaphysics. It has, on the whole, kept its ground, as the acknowledged modern theory of vision, amid the proverbial fluctuations of thought on such subjects. I venture to append to the preceding analysis of the Essay a few notices of the state in which it found philosophical opinion on the questions to which it relates, and of the change which it has inaugurated.

The subtle analysis which distinguishes seeing strictly so called from judgments about extended things, which are only suggested, after a sufficient tactual and locomotive experience, by what we see, appears to have been unknown to the ancient philosophers, who, it may be added, were ignorant of much now known with regard to the organic conditions of this sense. Aristotle, indeed, speaks of colour as the proper object of sight; but, in the passages of the De Animat where he names properties peculiar to particular senses, he enumerates others, such as motion, rest, number, figure, and magnitude, which belong to all the senses in common.

* In Diderot's Lettre sur les aveugles, a l'usage de ceux qui voient, where Berkeley, Molyneux, Condillac, and others are referred

to.

Cf. also Appendix, pp. 111, 112; and Theory of Vision Vindicated, sect. 71, with the note, in which some recorded experi

ments are alluded to.

This