ABSTRACT: Berkeley introduces his water experiment in order to demonstrate that in perception the perceiver does not reach the world itself but is confined to a realm of representations or sense data. We will attempt to demonstrate that Berkeley's description of our experience at the end of the water experiment is inauthentic, that it is not so much a description of an experience as a reconstruction of what we would experience if the receptor organs (the left and right hands) were objects existing in a space partes extra partes. Our argument is that there is nothing in our experience of the illusion to suggest that under normal conditions perception does not reach the world itself.

Traditionally Empiricists claim that all knowledge and all basic concepts are derived from experience. At the same time they argue that all experience is reducible to private entities, the so-called 'sense data'. Phenomenologists claim that there is nothing in experience itself to suggest that it is reducible to sense data, and that this doctrine is derived from metaphysical prejudices, the so-called 'assumptions of the natural attitude'. They argue that if we could in some way 'bracket' these assumptions and reflect only on our experience of perceiving and on the results of scientific measurements of our perceptual powers, we would discover that perception, rather than presenting us with private entities or 'data', 'opens up' to the world itself. (1)

In A New Theory of Vision, Berkeley attempts to show that all experience is reducible to sense data by exploiting two types of argument. At times he exploits a scientific account of perception and of the functioning of the perceptual organs, while at other times he uses the argument from illusions.

For example, he argues, that the experience of temperature can be understood with the analogy of the experience of pain, and just as the pain is not 'in the needle', so the warmth I feel is not in the fire. (2) He then argues in a similar vein that visual experience is reducible to collections of colour sensations because light passes into the eye ball and strikes the retina, in much the same way that a sharp object striking the skin produces a sensation of pain, such as a sensation of blue or red. (3) The sensation being the effect of the physical and chemical properties of the world on the sense organs and is as distinct from the world as photographic images are from the objects which cause them.

When approached from such an external point of view perception indeed appears unable to grant any direct 'access' to the world itself, yet when described from within, from the so-called 'first' person point of view, it appears to 'open up' to the world itself. Berkeley's argument assumes that because from an external point of view perception cannot 'reach' the world, any direct experience that I may have of such a 'reaching' or 'opening up' from within, must necessarily be illusory or subjective, and illusory and subjective not merely from some or other point of view, but illusory and subjective in itself, and therefore could present no challenge to the external view.

This exploitation of the scientific account of perception therefore assumes that the external view has a bearing on the internal view, that in some way it 'prevails over' or 'encompasses' the internal view. and therefore, that they belong to the same universe. But there is no attempt to show how these two points of view measure up against each other. There is nothing in the external view itself to indicate that it encompasses and hence prevails over or discounts what can be revealed about perception when described from the internal point of view, or even that the two belong to the same universe. This conviction is derived from metaphysical assumptions about the relationship between the internal and the external views. What is needed, if philosophy is not to be stranded on the irreducibility of these two points of view, is another, which would embrace the two and from which they could be grasped as belonging to the same universe and their accounts of perception revealed as abstract descriptions of one indivisible process. Empiricism however offers no such point of view. Until we can reveal how the external and internal points of view gear into each other, we cannot exploit the external or scientific point of view in an argument meant to demonstrate that perception is limited to the intuition of private data. (4)

The argument from illusions is generally held to corroborate the assumptions of the natural attitude. At one moment I see cracks in the wall, as I get closer I no longer see cracks. All I have in front of me, are shadows. If there never were any cracks, what did I see? I must have seen something which I mistakenly took to be cracks. The phenomenon of illusions would therefore appear to offer evidence that the internal view is 'subjective' in itself, and so demonstrate that the external view prevails over it. What we are claiming however is that the assumptions of the natural attitude, far from being corroborated by the phenomenon of illusions, are taken for granted in the definition or the description of the phenomenon itself. The illusory experience for example is generally described as if it were an event in the external world, and hence subject to the temporal and spatial structures of the external world. We have argued elsewhere that if we could bracket this concept of time, we would be able to give an account of the phenomenon of illusions which would not undermine our conviction that in perception we reach the things themselves (Wait;1997).

In this paper we investigate another type of perceptual illusion, one which has been misrepresented through the surreptitious introduction of a scientific concept of the spatiality of the body.


Berkeley's argument, based on the results of his water experiment, is meant to show that perception does not reach the world itself, and that in perception we are confined to a realm of private or subject dependent 'data'. If I keep the left and the right hands in bowls of hot and cold water respectively, then plunge both into the same bowl of lukewarm water, Berkeley claims that the left hand would experience the water as cold while the right would experience it as warm. (5) He argues that since the same water cannot be both warm and cold at the same time, what I feel with my hands cannot be an attribute of the water itself, what I feel must in some way be determined by the past experience of the hand. The conclusion of his argument is that in order to make sense of the experience at the end of the experiment, we are obliged to adopt a certain theory of perception, in which the perceiver is conceived as being restricted to the intuition of 'sense data', while the world itself is conceived as lying beyond all direct experience. (6) This argument would therefore appear to corroborate the empiricist account of perception purely from the internal point of view.

We will attempt to demonstrate that Berkeley's description of our experience at the end of the water experiment is misleading, that it is not so much a description of an experience, as a 'reconstruction' of what we would experience, if the receptor organs, the left and the right hands, were objects existing in a space 'partes extra partes'. We will try to show that this conception of the spatiality of the body is not derivable from experience, but from the scientific or 'external' account of the world and of the body. In other words, in the way in which Berkeley describes the experience, it is clear that he has already assumed that the world of experience is a 'sub world' contained within the external world. If Berkeley's argument is to be convincing, it would require an unbiased description of the experiences we have throughout the water experiment. What we are claiming is that there is nothing in these experiences to suggest that the body and the receptor organs are in such a space partes extra partes. Once this conception of the spatiality of the body is bracketed, we can give a description of the experience at the end of the experiment which would not require the concept of a sensation and which would not undermine our faith that under normal conditions we perceive the warmth of the water itself.

If I reflect on my experience of testing the temperature of something with both hands, I find that I am not given two experiences of temperature any more than using one hand gives me a collection of separate experiences of temperature for each finger, for the palm, for the upper hand etc. In general the receptor organs appear to combine to form one indivisible 'channel' opening up to one temperature in the world. If there are actually two temperatures in the world, as there often are in the sea, where certain lower currents are colder than those above, I recognise this not through a judgement in which two private sensations arising from two receptors are compared, but because the hands or the body acting as a single integrated receptor organ, 'opens up' to two masses of water each with its own temperature. I do not have two experiences of temperature, but one experience of two temperatures. My ability to experience two distinct temperatures at the same time is not due to my having two organs or receptors of temperature, each functioning on its own, it is due to my ability to synthesize a multitude of receptor organs so as to constitute one opening onto a world where different objects can have different temperatures.

This phenomenon of 'synthesis' is unmistakable in visual perception. Here there is nothing to suggest that specific parts of the sensory field are 'driven' by specific groups of retinal cells. Patients with optic nerve lesions, whose visual fields are actually restricted, are able to overcome their "tubular vision" by active tracking and saccadic movements of the eyes, such that they experience their visual fields as normal (Luria, 1963;220). The organization of the visual field reflects the spatial distribution of objects in the world, rather than the distribution of stimulated retinal cells. There is nothing in our experience to suggest that the visual field is itself a product, or something synthesized out of sense data, each datum produced by a localized sense organ in the retina. It is well known in psychological research that it is easier for subjects to identify shapes of objects in the world, because of their actual resemblance to a known shape, than because of the resemblance of the shapes of their retinal images. (7)

Similarly the organization of the sensory field as I experience the warm and the cold layers of water in the sea, is related to the localities of the two masses of water, rather than to the location of the receptor organs. I do not know that the cold water currents are below because I know that the sensory organs providing the sensations of cold are below those providing the sensations of warmth. In fact my impression of the localities of the two currents is more concrete and accurate the faster I move my limbs, and hence the more difficult it would be to know where any receptor organ is at a particular moment. The organization of my sensory field expresses the spatial distribution of the masses of water rather than the anatomical localities of the receptor organs in the body. This is made possible through the 'synthesis' of the receptor organs to constitute a single organ of temperature.

Using a multitude of receptor organs does not mean that I am offered a multitude of sensations. The multiplicity of sense organs simply makes possible a more articulated and unambiguous 'grip' on the same reality, namely the two masses of water.

For the same reason my experience at the end of Berkeley's experiment is not reducible to two experiences, an experience of cold for the left hand and warmth for the right hand. It is neither the experience of two different temperatures, as in the sea, nor is it the experience of the same temperature. It is rather an experience of the disintegration of the organization of the sensory field, an experience of being unable to synthesize the hands in a joint exploration of the temperature of the water and hence an experience of 'not being able to reach the temperature of the water itself'. Berkeley's description, that the one hand feels the temperature as warm while the other feels it as cold, is already what we may call a 'reification' of a much more ambiguous, lived experience.

The experience of the disintegration of the perceptual field and the alienation from reality which we are describing is more familiar in the modality of the visual perception of depth. The synthesis and disruption of the perceptual powers of the two hands is comparable to the synthesis and disruption of the functioning of the two eyes in the binocular vision of an object. In our experience of 3-D perception for example, we are not simply aware that both eyes perceive the same object. The two eyes work together and make possible a single but more concrete grasp of the object.

"On passing from double to normal vision, I am not simply aware of seeing with my two eyes the same object, I am aware of progressing towards the object itself and finally enjoying its concrete presence." (Merleau-Ponty, 1962:233)

Simply because we are using two eyes does not mean that there must be two images, or that our experience of the concrete 3-D object must be 'built up' of two images. In binocular vision each eye ceases to function on its own account and the two are "used as a single organ by one single gaze" (Merleau-Ponty,1962:232), making of my two eyes "the channels of one sole Cyclopean vision" (Merleau-Ponty, 1968:141). If in some way this binocular co-operation were to break down or become impossible, the experience might run in reverse. For example, if while looking at a table, I push against my left eye-ball making binocular focussing impossible, I will not be inclined to say that I now see two tables, one above the other. It is not for me as if a real table has moved up into the sky, but only a thin film on the surface of things, while at the same time I experience the loss of the concrete presence of the thing. (8) If 3-D focusing can be "progress towards reality", then the breakdown of 3-D perception can be the loss of that concrete presence, or a 'regression from reality'. It is not as if in double vision I am confronted by two tables, one straight in front of me and another above my head, and that I am therefore lead to conclude that since the table cannot be in two places at once, visual perception does not make contact with the object in the world, only with images of those objects. Double vision is not simply the perception of two objects and an inference that the eyes are not focussed. There is a world of difference between the perception of two similar objects and double vision of a single object. The disruption of focussing and the loss of the concrete presence of the object, or the regression from reality are experienced 'in the flesh'. There is in this unstable experience a 'demand for a resolution', which will once again bring me into contact with the object itself.

Similarly at the end of the water experiment I am not simply left with two temperatures, as I was at the beginning of the experiment. There is something unreal or insubstantial about the 'temperatures' I feel. They are like the 'thin film', on the surface of things. The only authentic answer to the question as to what I feel at the end of the experiment, would be that I don't feel anything distinctly, and this would not be a judgement I make on the basis of an experience of two distinct temperatures, it would be a description of an experience of failing to achieve an unambiguous grip on the world. The 'insubstantiality' of the temperatures, their lack of concrete presence, is inseparable from the failure to grasp the world itself. It is not as if I feel two temperatures and infer from this that I cannot be feeling the temperature of the water. It is easier for me to tell that I am no longer reaching the actual temperature of the water than it is for me to be sure about what sensations I am having. (9)

Nor is it possible to withdraw from this 'not being able to reach', or this 'regression from reality', in order to describe it as itself a subjective experience, something which would itself presuppose subject dependent entities. The experience is not reducible to 'contents of consciousness' for it is not so much that I am conscious of the disruption of my perceptual power, it is that consciousness is itself disrupted, or rather that I am disruptedly aware of the disruption. Similarly I am not conscious of a regression from reality, it is consciousness itself that has lost its grip on the world. Any attempt to reflect on my experience of this disruption or this 'regression from reality' will be as if 'rent by an inverse movement' (Merleau-Ponty,1964(b):161), throwing me beyond the realm of experience, such that I 'live' my regression from reality, as opposed to being confined to signs of it.

We have in this phenomenon of 'disruption' and 'regression from reality', then, an example of an experience which cannot be fully described from the 'inside', i.e. from the first person point of view. Any description will presuppose a description, from an external point of view, namely of myself relating to the world, reaching or failing to reach it.

"As soon as I see, it is necessary that the vision (as is so well indicated by the double meaning of the word) be doubled with a complementary vision or with another vision: myself seen from without, such as another would see me, installed in the midst of the visible, occupied in considering it from a certain spot." (Merleau-Ponty, 1968:134)

It is because of this "complementary vision" that the experience I have at the end of the water experiment is doubled with a complementary vision, myself failing to reach through to the temperature of the water itself. Any attempt to reflect on my experience leads me beyond any 'contents of consciousness' to my relationship with the world. (10) There is therefore nothing in the experience of being unable to synthesize the two hands in a joint exploration of the water to suggest that it is itself an inference made from certain sense data, or that there is a need for a representative conception of consciousness.

To a large extent the flaws in Berkeley's description of the experience are not as obvious as they would have been had the argument been conducted on the level of visual perception. This is because the sense of temperature is not as developed or articulated as the sense of sight, and partly because we are not accustomed to using both hands to feel the temperature of water. (Perhaps the sense of disruption, and regression from the world would have been more difficult to overlook had Berkeley used the upper hand and the palm of the same hand, rather than two different hands) Consequently it is difficult to be sure about what it is that is experienced, and even more difficult to provide an authentic description of the experience. What we are claiming however is that the disruption of the joint exploration of temperature with two hands is analogous to the disruption of binocular depth perception as I press against the eyeball. Placing the two hands in bowls of water at different temperatures has the effect of making it impossible for them to work together, to constitute 'one' sense organ for the experience of temperature-in-the-world. In both cases the experience of being unable to synthesize the perceptual organs in a joint exploration of the water and the failure to reach the world itself, is an irreducible phenomenon. An authentic description of this phenomenon in no way suggests that in normal perception we are confined to the experience of subjective entities such as sensations. On the contrary the experience of not being able to reach, is remarkable precisely because it contrasts with our more usual experience of reaching the world itself, remarkable because it is a demand for a re-synthesis of the perceptual body in order to re-establish openness to the world itself.


(1) "The alleged self evidence of sensation is not based on any testimony of consciousness, but on widely held prejudice" (Merleau-Ponty, 1962:5)

One essential difference between phenomenology and empiricism, is the former's claim that in perception we 'reach' the things themselves. Husserl, for example, has insisted that "the spatial thing which we see is, despite all its transcendence, perceived"... that, "we are not given an image or a sign in its place" and that "we must not substitute the consciousness of a sign or an image for a perception", and that we should not assume that "perception does not come into contact with the thing itself". (1931:136)

(2) "Since therefor you neither judge the sensation itself occasioned by the pin, nor anything like it to be in the pin; you should not...judge the sensation occasioned by the fire, or anything like it to be in the fire" (Berkeley;1960:209)

(3) "...external light is nothing but a thin fluid substance, whose minute particles being agitated with a brisk motion, and in various manners reflected from the different surfaces of outward objects to the eyes, communicate different motions to the optic nerves; which being propagated to the brain, cause therein various impressions: and these are attended with the sensations of red, blue, yellow etc." (Berkeley;1960:217)

(4) "...will not the water, seem cold to one hand, and warm to the other?" (Berkeley, 1960:208)

(5) Berkeley eventually goes on to argue against the existence of such a transcendent world. His argument should therefore undermine his original exploitation of the external point of view.

(6) "...changing a form's orientation from its normally upright environmental position (with retinal orientation remaining normal) makes for greater difficulty in recognition than changing its orientation on the retina (with environmental orientation remaining normal)" (Rock, 1957:493) It is thus easier to recognise the shape of an object even when looking at that object from an inverted position, resulting in an inverted image, than it is to recognise the shape of an inverted object, even when looking from an inverted position, resulting in an upright i.e. normal retinal image.

(7) "... when I press on my eyeball, I do not perceive a true movement, the things themselves are not moving, but only a thin film on their surface." (Merleau-Ponty:1962;279)

(8) "It is through my relation to 'things' that I know myself; inner perception follows afterwards..." (Merleau-Ponty, 1962:383).

(9) "One of its 'results' [Husserl's inquiry] is the realization that the movement of return to ourselves - of 're-entering ourselves', St. Augustine said - is as if rent by an inverse movement which it elicits. Husserl rediscovers that identity of 're-entering self' and 'going-outside self' which, for Hegel, defined the absolute." (Merleau-Ponty, 1964:161)


Berkeley, 1960. A New Theory of Vision. J.M.Dent and Sons Ltd London.

Luria, A.R., Pravdina-Vinarskaya, E.N. and Yarbuss A.L. 1963 "Disorders of Ocular Movement in a Case of Simultanagnosia" Brain Vol 86:219-228

Merleau-Ponty, M. 1962. Phenomenology of Perception. C. Smith (Trans).. London : Routledge and Kegan Paul Ltd.

Merleau-Ponty, M. 1964. Signs. R.C. McCleary (Trans). Evanston, Ill. : North-Western University Press.

Merleau-Ponty, M. 1968. The Visible and the Invisible. A. Lingis (Trans).. Evanston, Ill. : North-Western University Press.

Rock, I. 1957. "The Effect of Retinal and Phenomenal Orientation on the Perception of Form". The American Journal Of Psychology. LXX(4):493-511

Wait, E.C. 1995 "A Phenomenological Rejection of the Empiricist Argument from Illusions". South African Journal of Philosophy 14(3):83-89

Wait, E.C. 1997 "Dissipating Illusions" Human Studies 20(2)221-242