Perception by H. H. Price
"Perception" by H. H. Price is a philosophical exploration of the nature of perception, focusing on how we experience and interpret sensory data. Price critiques existing theories, particularly naïve realism, which posits that sensory experiences directly correspond to external objects. He introduces key questions regarding the nature of perceptual consciousness and the concept of 'belonging to' in relation to sense-data—individual sensory experiences that inform our understanding of material objects. Price argues that sense-data are particulars rather than universals and emphasizes the importance of perceptual acceptance and assurance as we navigate our sensory experiences.
Central to his theory is the idea that sense-data arise from a complex interaction between physical conditions, our sensory faculties, and the material world, leading to the notion of a "nuclear solid" formed from perfect sensory experiences. He distinguishes between the existence of material things, which have causal characteristics, and the families of sense-data, which do not have the same status as the physical realities they represent. Ultimately, Price’s work contributes to ongoing discussions about the relationship between perception and reality, offering insights that encourage deeper contemplation of how we engage with the world through our senses.
Perception by H. H. Price
First published: 1932
Type of Philosophy: Epistemology
Context
In Perception, H. H. Price undertakes an examination of existing theories of perception, rejecting what is bad, retaining what is good, and adding original reflections to construct a new, more adequate theory avoiding the difficulties of the old. Price phrases the problem of perception in two separate questions:
1. What is perceptual consciousness and how is it related to sensing?
2. What is the relation of “belonging to” when we say a sense-datum “belongs to” a thing?
Consciousness contains givens—color expanses, pressures, noises, smells. These givens are sense-data, and the act of apprehending them intuitively is sensing. There are other data of consciousness, such as of introspection or memory. Sense-data differ from these solely in that they lead us to conceive of and believe in the existence of material things (whether or not such things actually do exist). By accepting sense-data as given, we do not commit ourselves to believing (1) that they persist when not being sensed (but only that they exist when sensed), (2) that the same sense-datum may be a datum of more than one mind, (3) that sense-data have some particular status in the universe, or (4) that they originate in any particular way.
Naïve Realism
Price examines and rejects naïve realism and its offshoots, the selective theory and the causal theory. He makes some use of the theory of phenomenalism and gathers the apparatus with which to exclude it only after stating most of his own theory. He never loses sight of his two basic questions, and all his very thorough and detailed arguments and rebuttals keep constant contact with them.
Naïve realism asserts that (1) perceptual consciousness is knowing that there exists an object to which a sense-datum presently sensed belongs and (2) for a visual or tactual sense-datum to belong to an object is for that sense-datum to be part of the surface of a three-dimensional object. Those who oppose the naïve realist usually regard hallucinations and perceptual illusions as ample refutation, but they commonly assume in their premises the theory they profess to upset. Price argues that the argument that actually disposes of naïve realism is simply that the many surfaces as seen, if they were really surfaces, simply would not fit together to construct the sort of three-dimensional solid in which the naïve realist believes. From the controversy, we can rescue the facts that the totum datum (the sum of all the data of all the senses at a particular time) has main parts, the somatic sense-data and the environmental, and that in certain respects these vary concomitantly. They are always copresent and covariant; that is, the totum datum is somato-centric. Similarly, from the remains of the effort to establish an improved form of realism that he names the selective theory, Price draws the lessons that we must account for abnormal and illusory sense-data as well as normal ones, that to various persons the same material things may be present to the sense in various ways, and that provision must be made for obtainable as well as actual sense-data.
The causal theory of perception states, as answers to Price’s initial questions, that “belonging to” means “being caused by” a material thing, and that perceptual consciousness is fundamentally an inference from effect to cause. Although Price expects to refute the theory readily, he notes that it seems to be the official foundation for the natural sciences. The only plausible version of the causal theory is, if every event has a cause and if sense-data are events, something other than sense-data must exist: the causes of sense-data. However, we can know nothing of the character of the causes and cannot prove them to be material. The fact is that from the first, we are “on the lookout” for sets of sense-data of the sort that we already expect to adumbrate solids. We already have a notion of the material thing before being capable of formulating ideas of causality. This notion seems not only innate but a priori, a necessary condition of certain kinds of experience. This must be a whole complex notion of thinghood, including the factor of causality.
Price then progresses toward his own theory. Sense-data are not universals but particulars; not redness, but instances of red. They are not facts, but the bases for judging facts. Visual and tactual sense-data are the primary ones for establishing the existence of material things. A visual sense-datum can be called a colored expanse, but to call it a colored surface assumes too much. Sense-data are not substances; when suitable bodily and external conditions are present, they are created out of nothing and, when these conditions disappear, vanish into nothing. They have a finite, usually very small, duration. They take up no space and do not have causal characteristics, such as inertia or impenetrability. They are not so much like mechanical processes as vital processes. They are generated in neither the brain nor the mind alone, but in the substantial compound of the two, having certain characteristics that neither would have by itself.
Perceptual Consciousness
The primary form of perceptual consciousness of our sense-data is perceptual acceptance. As a sense-datum arises, a state of taking for granted the material thing to which the sense-datum belongs also arises. The question whether to believe simply does not occur with the first sense-datum, though it may be introduced later. Other than the datum itself, perceptual acceptance has no content. Although I take for granted that the front surface of what I see has a back, I leave it until later to determine the nature of the unseen part. Moreover, the first sense-datum does not completely specify the details of the accessible side, but rather it simply limits the possibilities to some extent. Further observation adds greater detail. I therefore do not take my first datum as identical with the front surface of the object, but simply take it that a thing now exists that has a certain general character.
Unlike the transitory sense-datum, a material thing persists through a period before and after the sense-datum. It is spatially complete in three dimensions, whereas a sense-datum is spatially incomplete. It is public and accessible to many minds, whereas a sense-datum, being somato-centric, is private. “Belonging to” it are sense-data of many, usually all, of the different senses. Finally, it has causal relations, whereas the relation between a sense-datum and the awareness of that sense-datum is not causal.
Price accounts for error, illusion, and even mere peculiarities of interpretation by referring to the difference in nature between sensing and perceptual consciousness. Sensing is undoubted and perfectly intuitive reception of the given. Perceptual consciousness is a mode of taking as existing a material thing, although that thing is not necessarily present; it seems intuitive only because it is instantaneous, not discursive, and is yet the raw material of judgment, not the product. However, the claim of any ostensibly sensed object to exist and have a certain character may be tested and corrected.
The further development of perceptual consciousness is perceptual assurance. Through additional perceptual acts, our primary acceptance is led to a settled conviction of the existence and nature of the material thing that first gave rise to sense-data. With succeeding acts of perception, the mode of reception is not mere acceptance, as it was with the first. Conditioned by the first sensing, it is now a progressive confirmation of the thinghood of what is sensed, a continual further specification of previously unspecified detail and addition of other parts not at first sensed. For we take the initial sense-datum as confirmable by other obtainable sense-data, which will fit together in a unified, enduring something that is spatially complete and has causal characteristics. As further acts of perception provide such confirmations, the existence of the thing and the specifications of its nature become settled rational beliefs. (Those cases with some confirmation but in which the way to adequate confirmation is blocked may be said to bring perceptual confidence, our ordinary state regarding most things.)
How Sense-Data Interrelate
Price carries out a subtle and complex analysis of the relation of sense-data to one another. From the separate sense-data of sense-fields, as both change, we gain our beliefs in the existence of individual things. The theory that sense-data that resemble one another compose a class, and that the thing is simply the class of resembling data, is inadequate because many of the data of an individual thing do not resemble one another in any sensible way. A better theory is that of gradual transition, which accounts not only for the changing of sense-data with motion of the observer but also for distorted data (as in perspective viewing) as well as for changing qualities (while, it is assumed, the thing itself does not change). According to this view, the whole group of changing data gathered as we move about a thing to get further specification of it “belong to” the particular thing. However, shapes, sense-data apprehended as spatial, must be related not only geometrically but also locally. We need an account of the manner in which sense-data “fit together” to form a solid. Price calls data that thus fit together constructible.
Price cites certain empirical facts. Although optical theory has not recognized it, vision has two types of stereoscopy, perfect and imperfect. The familiar sort associated with perspective seeing is the imperfect; it allows us to construct solid objects from its data but incompletely. Immediately before the eyes there is a range of depth in which the usual perspective effects are reversed—the parallel edges of a matchbox when held against the nose seem closer together at the nearer rather than farther end. Between this nearest range and the outer one, there is a range of perfect stereoscopy, in which the visual sense-data of any object small enough to lie within it actually do coincide, approximately, with the surfaces of the particular thing. In this range, things are seen without distortion. The rectangular sides of the matchbox are seen as rectangular, not trapezoidal, even though the three sides seen at once are facing in quite different directions.
These facts allow Price to conclude that the thing seen within the range of perfect stereoscopy is seen virtually as it is. The solid perceptually constructed in this range is constructed from perfectly constructible sense-data. He calls it a nuclear solid. Around it, we can arrange all constructible sense-data in order from the least deviation to the greatest. The variations in perspective order will form a perspectival distortion series, and those varying concomitantly with that series from greatest to least specification will form a differentiation series. The members limiting these series at the position of the nuclear solid are their nuclear members. The nuclear solid provides the ground for uniting both continua, and indeed the continua of all sense-data both of spatial and nonspatial senses. The nuclear members with respect to a single sense are standard forms—the nuclear visual sense-datum, for example, gives the standard figure and standard color. The nuclear solid constructed from the perspectival distortion series, amended by the best presentation (that with most specific detail) of the differentiation series, gives us the standard solid. The collection of all sense-data unified by a standard solid is a family of sense-data.
To establish the validity of the construction of material things from sense-data, Price cites three propositions that seem obviously true, although he cannot imagine a way to prove them.
(1) Some sense-fields are not momentary but have a finite duration.
(2) Some two sense-fields must be continuous rather than discrete in time and quality.
(3) Successive sense-fields sometimes overlap in time—have a part of their durations in common.
These are the logical requirements of three-dimensional construction. If they actually occur, then, given two sense-fields, one containing a pair of sense-data AB and the other containing a pair BC, each pair sensibly adjoining and also arranged in a spatial relation R (such as “to the right of”), we can know that A and C are also related by relation R. This method of progressive adjunction is our validation of the “beyond”—the existence of unseen surfaces.
It is easily seen that the nonnuclear data do not exist in space. The tabletop that is narrower at the far end exists in sense only, as a member of the family of sense-data of the table, not as a constituent of the standard solid of the table. The nuclear sense-data are not parts or surfaces of the standard solid but may be said to coincide with it. Position in physical space is not a characteristic of a single sense-datum but rather a collective characteristic of a whole group of sense-data constituting a standard solid.
A family of sense-data perhaps should not be said to exist in time and undergo change but to prolong itself through time and differ through time. The point of view of the observer, and its motion, are definable in terms of the space of standard solids. Obtainable sense-data are definable in terms of changes of the observer’s point of view.
Causality
So far, it might seem that the theory of perception Price has offered is an elaborated phenomenalism, identifying a material thing with the family of sense-data. He shows, however, that this is not his theory. A material thing physically, as well as sensibly, occupies a space. By this we mean that it manifests in that space certain causal characteristics, most notably that of impenetrability. When a chestnut drops on a stone and bounces away, it is a causal characteristic of the stone, not a sense-characteristic, that has acted on the chestnut. An obtainable sense-datum, on the other hand, is not an existent particular, but a fact of the form “If any observer were at such and such a point of view, such and such a sense-datum would exist.” Because an individual observer can occupy only one point of view at a time, he or she can realize only one of the alternatively obtainable sense-data at a time.
The family of sense-data thus has a peculiar mode of existence, as a collection of actual sense-data, plus an infinite number of obtainable sense-data existing as contemporary alternatives, centering on a standard solid. This is not the kind of entity to which causal characteristics may belong. Further, in many instances, a family of sense-data is manifested in only one part of a region when a causal characteristic is being manifested in many parts of that region. Again, causes continue to be manifested during times when no sense-data are obtained. There is no necessary concomitance of caused events with sense-data, though there is concomitance among sense-data themselves. Accordingly, the causative physical occupant and the family of sense-data are not identical, and phenomenalism is false.
Collective Delimitation Theory
Now we see that for a sense-datum s to belong to a certain material thing M, first s must be a member of a family of sense-data prolonging itself through time and centering in a standard solid having a certain place in the system of standard solids; and second, the place must be physically occupied—causal characteristics must be manifested there. This is Price’s theory, the collective delimitation theory. It stresses that the primary relation of “belonging to” is between an entire family and a material thing, the relation of a single sense-datum to the thing being derivative; moreover, that the family is related to the material thing by coinciding with the physically occupative portion of the thing.
Probably we get our first hint of the relation of causation to sense-data by observing other people’s changed behavior in the face of certain evident changes in their environment. More important is the influence of the fact of our having several senses. By reflection on the data of one, we discern how the data of another sense originate. Concomitant changes in our tactual sense-data, for example, with the introduction of objects of the visual field to the skin, are a source of knowledge. The complete thing is the physically occupative thing with causal properties plus the family of sense-data. A changing complete thing is not the cause of changes in our sense-data, for the data are part of the thing. However, a change in the thing as physical occupant of space may entail changes in the sense-data that depend upon the presence of both it and its observer. The laws of sense-data are of a different order from ordinary causal laws. We can define a given type of physical occupant only by reference to the kind of family of sense-data it is coincident with and to all those foreign ones whose mode of prolongation it influences. However, of the intrinsic qualities of physical occupants, we have no knowledge at all.
Beyond Phenomenalism
Price’s Perception perhaps represents our farthest point of advance in its area. Whether his position is adopted, his close study of the problems and his careful solutions have much that is useful. For the introspective part of his procedure, Price asks of us only the accordance with our own experience. In addition, he seems to begin with the perhaps intuitive preconceptions that have made us suspicious of some of the stranger theories of perception and to exhibit and justify these conceptions. Finally, Price’s convincing exposition of the differences between the separately certifiable orders, sense-data, and the manifestations of causality seems, by correcting the “emptiness” of phenomenalism, to be a considerable contribution.
Principal Ideas Advanced
•Sense-data are the given elements in our sense experience, the colors, odors, noises, and pressures of things as they are met in consciousness; such sense-data are particulars, not universals.
•Material objects are spatially complete, three-dimensional objects; but sense-data are spatially incomplete and private, and they are free from the causal relations that characterize material objects.
•Sense-data that fit together to form solids are said to be constructible, and solids perceptually constructed within a mediate range of vision in which perfect stereoscopy is possible are said to be constructed from perfectly constructible sense-data.
•The material object, known by reference to constructible sense-data, has causal properties that could not belong to any family of sense-data; consequently, phenomenalism is false.
Bibliography
Davis, Stephen T., ed. Death and Afterlife. New York: St. Martin’s Press, 1989. Discussions of the afterlife, from secular and a variety of religious perspectives with some references to H. H. Price; not every religious tradition teaches personal survival.
Davis, Stephen T. “Survival of Death.” In A Companion to the Philosophy of Religion, edited by Philip L. Quinn and Charles Taliaferro. Oxford: Blackwell, 1997. Discussion of the notion of a person surviving the death of her body, with reference to Price’s classic article, “Survival and the Idea of ‘Another World.’”
Flew, Antony, and Alasdair McIntyre, eds. New Essays in Philosophical Theology. London: SCM Press, 1955. Essays by D. M. Mackinnon and Antony Flew regarding survival.
Lewis, H. D. Persons and Life After Death. New York: Barnes and Noble, 1978. Essays on the title topics by Lewis, with contributions from and discussions with Anthony Quinton, Bernard Williams, Antony Flew, and Sydney Shoemaker with numerous references to Price; Lewis favors the claim that persons survive the death of their bodies, and the others are critics of this claim.
Yandell, Keith. The Philosophy of Religion. London: Routledge, 1999. Contains discussions of views of persons and the afterlife as held in various religious traditions.