Perception

Howard Robinson

• Publisher: Routledge
• Year: 2003
• eBook ISBN: 9781134533015
• Topic: Philosophy
• Subtopic: Philosophy History & Theory

Questions about perception remain some of the most difficult and insoluble in both epistemology and in the philosophy of mind. This controversial but highly accessible introduction to the area explores the philosophical importance of those questions by re-examining what had until recent times been the most popular theory of perception - the sense-datum theory. Howard Robinson surveys the history of the arguments for and against the theory from Descartes to Husserl. He then shows that the objections to the theory, particularly Wittgenstein's attack on privacy and those of the physicalists, have been unsuccessful. He argues that we should return to the theory sense-data in order to understand perception. In doing so he seeks to overturn a consensus that has dominated the philosophy of perception for nearly half a century.

CHAPTER I

The Classical Empiricist Conception of the Content of Perceptual Experience

1 The traditional empiricist conception of sensory content

The majority of modern philosophers—that is, the majority of philosophers writing since the seventeenth century—have believed that in perception one is aware of some item other than the physical object one takes oneself to be perceiving. So if I see a tree, for example, what I am really or directly aware of is something which can roughly be thought of as a tree-image, in or before my mind, rather than a mindindependent physical tree in external physical space. The ideas of Locke and Berkeley, Hume’s impressions and the qualia, sensa and sense-data of twentieth-century philosophers are all generally supposed to be of this type. On the other hand, the majority of strictly contemporary philosophers—that is, the majority of philosophers active in the analytic tradition since the Second World War—have denied that one need postulate such entities and affirm that we are, normally, directly aware of the external world itself.

The earlier modern conception I shall call the classical or empiricist conception of sense-contents, and this book is largely about the arguments that are brought for and against this conception. Contrary to the general contemporary opinion, I shall be arguing that this classical empiricist conception is essentially correct—that there are things more or less of the kind which I shall, for convenience, call ‘sense-data’. The empiricist conception of sense-contents—sense-data—can be roughly characterised as follows.¹

A sense-datum, as I shall understand it, meets five conditions:

1. It is something of which we are aware.
2. It is non-physical.
3. Its occurrence is logically private to a single subject.
4. It actually possesses standard sensible qualities, for example, shape, colour, loudness, ‘feel’ of various sorts.
5. It possesses no intrinsic intentionality; that is, though it may suggest to the mind through habit other things ‘beyond’ it, in itself it possesses only sensible qualities which do not refer beyond themselves.

The concept of intentionality plays such a central role in anti-sensedatum theories that a brief introduction to it at the outset is essential. Intentionality is the property of states—usually mental states—of being about things. An ordinary physical state is not about anything, it is just there, unless it is interpreted as being about or signifying something. But thoughts, beliefs, desires and most mental attitudes are essentially about things—they refer beyond themselves; they are directed upon objects. These objects need not be actual things, for one can think about, believe in, desire, etc., things that do not exist. The objects of intentional verbs are called ‘intentional objects’, and, because they need not exist for the activity to be directed to them, they are described as ‘intentionally inexistent’. If perceptual states are intentional then they are essentially of things, as are thoughts. In this case the question will not arise of how perceptual states can put us in touch with the external world, for they are essentially of things in the external world. Furthermore, as the objects of intentional activities can be things that do not exist, we would not need to worry about the status of things that we seem to perceive that are not really there. On the other hand, intentional states are mysterious things. If they are intrinsically about other things, what properties, if any, do they possess intrinsically? Sense-—data are, in some ways, less mysterious than intentional states, for the properties sense-data are supposed to possess are straightforward sensible qualities. These issues will recur persistently in this book.

The outright opponent of sense-data denies that there are any private objects of awareness involved in normal perception. A more moderate opponent allows that there are private objects of awareness, but claims that their content is purely intentional; that is, that all the content is essentially—if only seemingly—of external physical features and objects, the contents themselves not actually possessing or instantiating any sensible properties. A still more moderate opponent would allow that there are private objects of awareness that instantiate certain qualities—at least including thesecondary qualities—but also possessing intentional properties, so that a patch of brown could be essentially as of a table-shaped public object, quite independently of any inclination of the subject so to interpret it. In my conclusions I shall tend to favour the full sense-datum theory, but will remain somewhat agnostic about the presence of a moderate intentional element.

There are two different views of the relationship between sense-data and the physical world. According to idealists and phenomenalists, such as Berkeley, Hume and Mill, the data are actually part of physical objects, for objects consist only of actual or actual and possible sensedata; according to representative realists, such as Locke is generally supposed to be, they are caused by physical objects, which they resemble, though the resemblance may only be a rather abstract and structural one. I shall discuss representationalism and phenomenalism in Chapter IX.

Many further questions arise concerning sense-data even for their proponents; for example, whether they are mental or neither mental nor physical; whether in the visual case they are two or three dimensional; whether knowledge of them is incorrigible and, in general, what their relation is to concept-involving activities such as recognition; whether they are particulars or a sort of sensory universal; whether it is coherent, within the scope of the classical conception, to regard them as modes of sensory activity rather than objects of it. Some of these issues will be dealt with in the following pages of this book. In Chapters II and III, my main purpose is to consider the arguments that have traditionally been brought for the classical conception minimally conceived in accordance with (l)–(5) above, and the objections raised against these arguments. The arguments take the form of attempted refutations of naive or direct realist theories of perception, the thought being that if one is not directly aware of external objects, then one must instead be aware of sense-data as defined by our five conditions. There is obviously some slack here, because one could reject naive realism without adopting exactly the classical conception in one’s replacement theory: the sense-datum theory is merely the commonest candidate. One could, for example, hold, with Russell, that we are aware of states of our own brain; or, with certain phenomenologists, that the private data do possess intrinsic intentionality.² These and other variations will come up in the course of the discussion, which is centred on the classical conception.

In the remainder of this chapter I shall look, not at the arguments for sense-data, but, briefly, at the role of the empiricist conception in the history of philosophy. I do not try to be definitive in this, but to raise some of the questions, largely about intentionality and perception, that will recur throughout the book.

2 The ancient and medieval background to the empiricist conception of perception

It is natural to make a connection between the empiricist conception and earlier philosophies via the notion of phantasm. In the Aristotelian tradition, phantasms are images set up in the senses and, among the scholastics at least, they look very much like a kind of sense-datum.³ The interest of classical theories in a study like this, however, is not to provide primitive cases of empiricism, but to bring out certain tensions that exist in the project of developing an adequate theory of perception. In particular they can be used to illustrate the problems that arise when one tries to harmonise cognition with sensation. I shall not, therefore, begin the discussion of classical theories with phantasms, but start with what looks like a bizarre attempt to reconcile the human activity of perception with the passive reception of sensations.

Certain early Greek philosophers claimed that vision occurs when a stream of particles coming from an object meets a physical emission coming from the eye.⁴ As what we know of these theories is rather limited, we cannot be sure why they opted for such a seemingly odd theory, but it can be interpreted in a way that helps to set up one of the basic problems of perception. This problem can be expressed as follows. On the one hand, a little reflection—that is, thought that does not resort to any science that goes beyond common experience—shows that perception involves some sort of physical influence running from the external object to the sense organ of the perceiver. On the other hand, the essential nature of experience seems to be that the subject mentally reaches out to, and makes conscious contact with, the external object. The directionality of the physical process and that of the lived experience seem to be in direct conflict. How can a process in which the subject is the passive recipient of a stimulus be the physical aspect or realisation of a process in which the subject reaches actively and consciously out into the world? The way that perception both reaches out fromsubjects and reaches into them expresses itself in three tensions or dichotomies. These are between the physical process and the conscious or phenomenological aspects of perception; between the fact that experience and its immediate causes are internal—‘in the head’—on the one hand, and that its content is apparently external, on the other; and the tension between the passive receptivity and interpretative activity of the subject in perception. Perhaps it is helpful to think of the bizarre theory of effluents as an unconscious attempt to resolve these tensions. The active mental contribution is represented as a physical outflow, and the experience is somehow caused ‘out there’—if not where the object perceived is located, at least, as a compromise, halfway there and not in the eye or the brain. Early Greek philosophers are often said to be scientists rather than philosophers, and this dual causal process can be seen as an attempt to integrate in a purely physical story the active role of consciousness with the physical action of the stimulus. This way of reconciling the three tensions is, of course, blatantly inadequate. Empirically it fails because there is no evidence of a flow out from the senses; conceptually it fails because there is nothing to explain why such a collision of particles should constitute an experience. It seems to present a solution to the problem only by failing to grasp the more subtle way in which the causal and the phenomenological levels are both interdependent yet different in kind.

The usual way of reconciling these dilemmas has been to say that the physical process causes a representation or picture of the world within the subject and that the subject interprets this as being a feature of, or projects it onto, the external world. This is, roughly, the representative theory of perception which is found in Locke and is adopted by most moderate empiricists. As a way of solving the problem it has two features which might be thought of as shortcomings. First, it seems to be an irredeemably dualistic account, for, although the process whereby the object influences the subject is, in principle, an unproblematic physical process, the projecting and interpreting of images are thoroughly mentalistic: so there is an uncomfortable change of gear in the middle of the process of perception. Second, and more important, the projection or interpretation of an internal image is not really a reaching out to the external world, only the illusion of doing so.

The most important ancient and medieval discussions of perception take place under the influence of Aristotle, and Aristotle’stheory can be seen as an attempt to reconcile the apparently conflicting factors without moving to a simple representative theory. Considered as a physical process, perception involves the perturbation of the sense organ so as to produce an image or phantasm which is—or is more or less—physical. Considered philosophically—that is, in the light of the need to give an account of what it is to perceive, as opposed to an account of how perception works—perception is the reception of the form of an object without its matter. So if I see Socrates, his form is present in a special way in the sense. The image set up in the organ is the vehicle for the form of the object perceived, so that through the physical process the very thing perceived (and not just a replica of it) is present to the perceiving subject.⁵ In this way the inward flow of the causal process and the outward reaching of experience are reconciled, and we see how the latter supervenes on the former.

Two basic questions arise about this theory. The first question is historical and scholarly: is it a correct account of Aristotle? A distinction could be made here between Aristotle and his followers, especially St Thomas Aquinas and the scholastic tradition. Some interpreters (as we shall see) think that Aristotle himself had a straightforwardly materialistic account of perception, but that Aquinas had developed it in the direction of the theory just sketched. The second question is whether such a theory can succeed in reconciling the causal facts and the common-sense phenomenology.

The simplest and most reductive interpretation of Aristotle says that (1) the image involved in perception is a purely physical object, so that when someone sees yellow the eye literally becomes yellow and that is the image that constitutes the sensation; (2) the form without the matter is the same thing as the image, because ‘without matter’ means no more than that the external object itself (or bits of it) does not enter the organ. So perception is a purely physical process in which a straightforwardly physical quality is established in the sense organ. This is transmitted from thence into the heart, where the senses are united, and is retained in the bloodstream, thereby constituting memory. Whether or not this is what Aristotle meant, he was usually interpreted by his commentators and successors in a less materialistic way.⁶ Most importantly, ‘form without matter’ was given a more mysterious interpretation, so that it signified a special kind of presence in the subject of the thing perceived. The scholastics expressed this by saying that the form ofthe object perceived possessed esse intentionale, not esse materiale—intentional existence, not material existence—in the sense. Brentano interpreted this as meaning that the form is an intentional object and exists in the soul, as mental, and not in the body, as physical.⁷ It is the possibility of this interpretation that makes the Aristotelian theory relevant to a general treatment of perception, for the notion of an intentional object will keep recurring throughout this book. It is a mysterious notion in certain fundamental respects, and the Aristotelian-cum-scholastic theories provide material for an interesting interpretation of it. The intentional state is actually constituted by a mode of existence of its object; so there is no mystery about how the mental content relates to the external reality, for it is that reality—or the form of that reality—taking on a psychological mode of existence.

Unfortunately, Brentano’s theory is not merely disputed as an interpretation of Aristotle, for there are also those who think it too metaphysical even as a reading of Aquinas. Such sceptics think that the notion of intentional existence in Aquinas has a less interesting sense than this interpretation gives to it. Sheldon Cohen has argued that, according to Aquinas, there is nothing particularly psychological about intentional existence.⁸ A colour exists intentionally in a mirror or in the air, because in neither case is the thing coloured in the ordinary, material sense. So if one looks at a red object in a mirror, the mirror is not literally red, in the same sense as that in which the object reflected is red, although red is, in some sense, in the mirror. And when the form of red passes from an object to the eye, the air in between does not become red. Intentional existence is not necessarily, therefore, mental. The jelly of the eye becomes red in just the sense in which a mirror does; the image in the eye has the same peculiar ontological status as a reflection, if one tries to take reflections realistically.

The example of the mirror does not, however, naturalise or physicalise the notion of intentional existence in the way Cohen thinks. The case of the mirror is not incompatible with the principal thrust of Brentano’s theory. In what sense is the form of red in the mirror, if not literally? Only, I think, in the sense that there is something in the mirror which communicates, transmits or makes available, the red in the object. This is exactly what the intentional object does in the psychological version of the theory. The only difference is that, in the case of the mirror, there need not be anythingto transmit it to, for no one might be looking, whereas, in the case of a sense, the subject is automatically present. It is true that Brentano thinks of the intentional object as mental, whereas, on this theory, intentionality can be embodied; but its logical features remain the same, namely those of communicating, transmitting or making available its object, without possessing the quality of that object in a literal material manner, as would a simple pictorial representation. It would be fair to say that the whole problem for ‘intentional object’ theories of perception is whether sense can be made of the idea that there is an entity that can do this; that is, represent its content transparently, not constituting the kind of ‘veil of perception’ that the empiricists’ contents are supposed to constitute. Whether such ‘transparent’ vehicles of representation are immaterial or physical is an extra—though important—point.

A concept is, of course, something that is essentially of its object, and so represents it transparently; and Aristotle talks of the intellect— which is wholly immaterial—as receiving form without matter, for there is no matter in the intellect.⁹ This might seem to make thought and perception very similar. But even if one is prepared to accept concepts as things that represent their content transparently, this does not solve the problem for perception. Some philosophers wish to treat perception simply as the acquisition of beliefs, so that perceptual content is not essentially different from purely conceptual content. But, in fact, something extra is needed to give the presentational nature of perception. This is something that will be discussed later, but Aristotle clearly believes that there is a crucial difference: that is why perception involves images in a physical sense organ, whilst intellect cannot be embodied. The image in the mirror may be just like the form in the eye or the faculty of sight, but it is not just like the form in the intellect. And the difference explains the experiential quality of perception. The analogy with a mirror image may seem to solve the problem of reconciling intentionality and sensory content, because such a thing possesses both the sensory element—by being an image—and the intentionality, because the image is essentially of its object and does not actually possess the property of the object, as would a literal picture. For this to work one would have to accept the scholastic account of mirror images, as the intentional yet physical manifestation of the form of the object. There has to be, that is, some special vehicle involved, corresponding to the intentional form in the mirror. If one adopts a simple causal story of how perception works,with light rays from the object to the eye being reflected in the mirror and causing an experience in the subject, then no particular illumination is given about the nature of experience by the example of the mirror; for, according to the causal account, red is not present in any sense—intentional or otherwise—in the mirror: the mirror simply reflects the light. If...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Preface
• CHAPTER I
• CHAPTER II
• CHAPTER III
• CHAPTER IV
• CHAPTER V
• CHAPTER VI
• CHAPTER VII
• CHAPTER VIII
• CHAPTER IX
• Notes
• Bibliography