An Essay Towards A New Theory Of Vision

In the present edition the text of this last edition is adopted, after collation with those preceding. The Appendix has been restored, and also the Dedication to Sir John Percival, which appeared only in the first edition.

A due appreciation of Berkeley's theory of seeing, and his conception of the visible world, involves a study, not merely of this tentative juvenile Essay, but also of its fuller development and application in his more matured works. This has been commonly forgotten by his critics.

Various circumstances contribute to perplex and even repel the reader of the Essay, making it less fit to be an easy avenue of approach to Berkeley's Principles.

Its occasion and design, and its connexion with his spiritual conception of the material world, are suggested in Sections 43 and 44 of the Principles. Those sections are a key to the Essay. They inform us that in the Essay the author intentionally uses language which seems to attribute a reality independent of all percipient spirit to the ideas or phenomena presented in Touch; it being beside his purpose, he says, to “examine and refute” that “vulgar error” in “a work on Vision.” This studied reticence of a verbally paradoxical conception of Matter, in reasonings about vision which are fully intelligible only under that conception, is one cause of a want of philosophical lucidity in the Essay.

The Essay moreover abounds in repetitions, and interpolations of antiquated optics and physiology, so that its logical structure and even its supreme generalisation are not easily apprehended. I will try to disentangle them.

From the second section onwards the Essay naturally falls into six Parts, devoted successively to the proof of the six following theses regarding the relation of Sight to finite spaces and to things extended:—

I. (Sect. 2-51.) Distance, or outness from the eye in the line of vision, is not seen: it is only suggested to the mind by visible phenomena and by sensations felt in the eye, all which are somehow its arbitrarily constituted and non-resembling Signs.

II. (Sect. 52-87.) Magnitude, or the amount of space that objects of sense occupy, is really invisible: we only see a greater or less quantity of colour, and colour depends upon percipient mind: our supposed visual perceptions of real magnitude are only our own interpretations of the tactual meaning of the colours we see, and of sensations felt in the eye, which are its Signs.

III. (Sect. 88-120.) Situation of objects of sense, or their real relation to one another in ambient space, is invisible: what we see is variety in the relations of colours to one another: our supposed vision of real tangible locality is only our interpretation of its visual non-resembling Signs.

IV. (Sect. 121-46.) There is no object that is presented in common to Sight and Touch: space or extension, which has the best claim to be their common object, is specifically as well as numerically different in Sight and in Touch.

VI. (Sect. 149-60.) The true object studied in Geometry is the kind of Extension given in Touch, not that given in Sight: real Extension in all its phases is tangible, not visible: colour is the only immediate object of Sight, and colour being mind-dependent sensation, cannot be realised without percipient mind. These concluding sections are supplementary to the main argument.

The fact that distance or outness is invisible is sometimes regarded as Berkeley's contribution to the theory of seeing. It is rather the assumption on which the Essay proceeds (sect. 2). The Essay does not prove this invisibility, but seeks to shew how, notwithstanding, we learn to find outness through seeing. That the relation between the visual signs of outness, on the one hand, and the real distance which they signify, on the other, is in all cases arbitrary, and discovered through experience, is the burden of sect. 2-40. The previously recognised signs of “considerably remote” distances, are mentioned (sect. 3). But near distance was supposed to be inferred by a visual geometry—and to be “suggested,” not signified by arbitrary signs. The determination of the visual signs which suggest outness, near and remote, is Berkeley's professed discovery regarding vision.

The sections in which Touch is thus introduced are among the most important in the Essay. They represent the outness given in hearing as wholly sensuous, ideal, or mind-dependent: they recognise as more truly real that got by contact and locomotion. But if this is all that man can see, it follows that his visible world, at any rate, becomes real only in and through percipient mind. The problem of an Essay on Vision is thus, to explain how the visible world of extended colour can inform us of tangible realities, which it does not in the least resemble, and with which it has no necessary connexion. That visible phenomena, or else certain organic sensations involved in seeing (sect. 3, 16, 21, 27), gradually suggest the real or tangible outness with which they are connected in the divinely constituted system of nature, is the explanation which now begins to dawn upon us.

The way in which visual ideas or phenomena “suggest” the outness or distance of things from the organ of sight having been thus explained, in what I call the First Part of the Essay, the Second and Third Parts (sect. 52-120) argue for the invisibility of real extension in two other relations, viz. magnitude and locality or situation. An induction of the visual signs of tangible size and situation is given in those sections. The result is applied to solve two problems then notable in optics, viz. (1) the reason for the greater visible size of the horizontal moon than of the moon in its meridian (sect. 67-87); and (2) the fact that objects are placed erect in vision only on condition that their images on the retina are inverted (sect. 88-120). Here the antithesis between the ideal world of coloured extension, and the real world of resistant extension is pressed with vigour. The “high” and “low” of the visible world is not the “high” and “low” of the tangible world (sect. 91-106). There is no resemblance and no necessary relation, between those two so-called extensions; not even when the number of visible objects happen to coincide with the number of tangible objects of which they are the visual signs, e.g. the visible and tangible fingers on the hand: for the born-blind, on first receiving sight, could not parcel out the visible phenomena in correspondence with the tangible.

The next Part of the Essay (sect. 121-45) argues for a specific as well as a numerical difference between the original data of sight and the data of touch and locomotion. Sight and touch perceive nothing in common. Extension in its various relations differs in sight from extension in touch. Coloured extension, which alone is visible, is found to be different in kind from resistant extension, which alone is tangible. And if actually perceived or concrete extensions differ thus, the question is determined. For all extension with which man can be concerned must be concrete (sect. 23). Extension in the abstract is meaningless (sect. 124-25). What remains is to marshal the scattered evidence, and to guard the foregoing conclusions against objections. This is attempted in sections 128-46.

The enunciation of the summary generalisation, which forms the “New Theory of Vision” (sect. 147-8), may be taken as the Fifth and culminating Part of the Essay.

The closing sections (149-60), as I have said, are supplementary, and profess to determine the sort of extension—visible or tangible—with which Geometry is concerned. In concluding that it is tangible, he tries to picture the mental state of Idominians, or unbodied spirits, endowed with visual perceptions only, and asks what their conception of outness and solid extension must be. Here further refinements in the interpretation of visual perception, and its organic conditions, which have not escaped the attention of latter psychologists and biologists, are hinted at.

That the Recherche of Malebranche, published more than thirty years before the Essay, was familiar to Berkeley before the publication of his New Theory, is proved by internal evidence, and by his juvenile Commonplace Book. I am not able to discover signs of a similar connexion between the New Theory and the chapter on the mystery of sensation in Glanvill's Scepsis Scientifica (ch. 5), published some years before the Recherche of Malebranche, where Glanvill refers to “a secret deduction,” through which—from motions, &c., of which we are immediately percipient—we “spell out” figures, distances, magnitudes, and colours, which have no resemblance to them.

An approach to the New Theory is found in a passage which first appeared in the second edition of Locke's Essay, published in 1694, to which Berkeley refers in his own Essay (sect. 132-35), and which, on account of its relative importance, I shall here transcribe at length:—

“To which purpose I shall here insert a problem of that very ingenious and studious promoter of real knowledge, the learned and worthy Mr. Molyneux, which he was pleased to send me in a letter some months since, and it is this:—Suppose a man born blind, and now adult, and taught by his touch to distinguish between a cube and a sphere of the same metal, and nighly of the same bigness, so as to tell, when he felt the one and the other, which is the cube and which the sphere. Suppose then the cube and the sphere placed on a table, and the blind man be made to see: quere, whether, by his sight, before he touched them, he could not distinguish and tell, which is the globe and which the cube? To which the acute and judicious proposer answers: ‘Not.’ For, though he has obtained the experience of how a globe, how a cube affects his touch; yet he has not obtained the experience that what affects his touch so and so, must affect his sight so and so; so that a protuberant angle in the cube, that pressed his hand unequally, shall appear to his eye as it does in the cube.—I agree with this thinking gentleman, whom I am proud to call my friend, in his answer to this his problem, and am of opinion that the blind man, at first sight, would not be able to say with certainty which was the globe and which the cube, whilst he only saw them; though he would unerringly name them by his touch, and certainly distinguish them by the difference in their figures felt.

“But this is not I think usual in any of our ideas but those received by sight: because sight, the most comprehensive of the senses, conveying to our minds the ideas of light and colours, which are peculiar only to that sense; and also the far different ideas of space, figure, and motion, the several varieties of which change the appearance of its proper object, i.e. light and colours; we bring ourselves by use to judge of the one by the other. This, in many cases, by a settled habit, in things whereof we have frequent experience, is performed so constantly and so quick, that we take that for the perception of our sensation, which is an idea formed by our judgment; so that one, i.e. that of sensation, serves only to excite the other, and is scarce taken notice of itself; as a man who reads or hears with attention and understanding takes little notice of the character or sounds, but of the ideas that are excited in him by them.

This remarkable passage anticipates by implication the view of an interpretation of materials originally given in the visual sense, which, under the name of “suggestion,” is the ruling factor in the New Theory of Vision.

The following sentences relative to the invisibility of distances, contained in the Treatise of Dioptrics (published in 1690) of Locke's friend and correspondent William Molyneux, whose son was Berkeley's pupil, illustrate Locke's statements, and may be compared with the opening sections of the Essay on Vision:—

“Wherefore, distance being only a line and not of itself perceivable, if an object were conveyed to the eye by one single ray only, there were no other means of judging of its distance but by some of those hinted before. Therefore when we estimate the distance of nigh objects, either we take the help of both eyes; or else we consider the pupil of one eye as having breadth, and receiving a parcel of rays from each radiating point. And, according to the various inclinations of the rays from one point on the various parts of the pupil, we make our estimate of the distance of the object. And therefore (as is said before), by one single eye we can only judge of the distance of such objects to whose distance the breadth of the pupil has a sensible proportion.... For, it is observed before (prop. 29, sec. 2, see also Gregorii Opt. Promot. prop. 29) that for viewing objects remote and nigh, there are requisite various conformations of the eye—the rays from nigh objects that fall on the eye diverging more than those from more remote objects.” (Treatise of Dioptrics, Part I. prop. 31.)

Occasional discussions of the New Theory may be found in the Gentleman's Magazine, from 1732 till Berkeley's death in 1753. Some criticisms may also be found in Smith's Optics, published in 1738.

Essential parts of Berkeley's analysis are explained by Voltaire, in his Élémens de la Philosophie de Newton. The following from that work is here given on its own account, and also as a prominent recognition of the new doctrine in France, within thirty years from its first promulgation:—

Condillac, in his Essais sur l'Origine des Connaissances Humaines (Part I. sect. 6), published in 1746, combats Berkeley's New Theory, and maintains that an extension exterior to the eye is immediately discernible by sight; the eye being naturally capable of judging at once of figures, magnitudes, situations, and distances. His reasonings in support of this “prejudice,” as he afterwards allowed it to be, may be found in the section entitled “De quelques jugemens qu'on a attribués à l'âme sans fondement, ou solution d'un problème de métaphysique.” Here Locke, Molyneux, Berkeley, and Voltaire are criticised, and Cheselden's experiment is referred to. Condillac's subsequent recantation is contained in his Traité des Sensations, published in 1754, and in his L'Art de Penser. In the Traité des Sensations (Troisième Partie, ch. 3, 4, 5, 6, 7, 8, &c.) the whole question is discussed at length, and Condillac vindicates what he allows must appear a marvellous paradox to the uninitiated—that we only gradually learn to see, hear, smell, taste, and touch. He argues in particular that the eye cannot originally perceive an extension that is beyond itself, and that perception of trinal space is due to what we experience in touch.

Its influence is apparent in British philosophy. The following passages in Hartley's Observations on Man, published in 1749, illustrate the extent to which some of the distinctive parts of the new doctrine were at that time received by an eminent English psychologist:—

“Distance is judged of by the quantity of motion, and figure by the relative quantity of distance.... And, as the sense of sight is much more extensive and expedite than feeling, we judge of tangible qualities chiefly by sight, which therefore may be considered, agreeably to Bishop Berkeley's remark, as a philosophical language for the ideas of feeling; being, for the most part, an adequate representative of them, and a language common to all mankind, and in which they all agree very nearly, after a moderate degree of experience.

In other parts of Hartley's book (e.g. Prop. 58) the relation of our visual judgments of magnitude, figure, motion, distance, and position to the laws of association is explained, and the associating circumstances by which these judgments are formed are enumerated in detail.