20th World Congress of Philosophy Logo

Persons and Personal Identity

Hume's Ontology of Personhood

Stanley Riukas

bluered.gif (1041 bytes)

ABSTRACT: The paper critically analyzes Hume’s view that human persons are "nothing but a bundle of different perceptions" in order to find out which one of the two possible interpretations of this view, the mentalistic or the physicalistic, is the more probable and free from serious difficulties. First, I examine Hume’s view of personhood from the mentalistic perspective only to discover that his all-important distinction between ideas and impressions is logically untenable. If ideas indeed resemble impressions, as Hume claims, then ideas should be present to our mind at the same time as impressions so we could compare them in order to find out whether there is any resemblance between the two kinds of perceptions and whether the impressions are indeed more forceful or vivacious than the ideas corresponding to them. But this is logically impossible because by the time we have an idea, its impression is gone, and if we think that we are comparing an idea with its impression, we are in fact comparing an idea only with a memory of its impression. But a memory of an impression is, in Hume’s view, already an idea. So we are comparing only two ideas, not an idea with an impression. Second, since ideas and concepts have no logical standing, we are forced to interpret the realm of ideas as an extension of the realm of impressions, coping with various problems arising from this interpretation as best we can.

bluered.gif (1041 bytes)

1. Introduction

Hume distinguishes, though somewhat informally, between the human being and the human person. The human being is composed for him of both the body and the mind, whereas the human person is the same as the self, the mind, or the soul. While the body may be viewed as something relatively static, the human person or mind is for him something essentially dynamic, an ongoing process. After pointing out that we do not have any idea of "the self" or person since we do not have any "particular impression" of the self that would correspond to the "idea" of self and thus it would be its "copy," Hume declares in a famous passage:

For my part, when I enter most intimately into what I call myself, I always stumble on some particular perception or other, of heat or cold, light of shade, love or hatred, pain or pleasure. I never can catch myself at any time without a perception, and never can observe anything but perception...setting aside some metaphysicians...,I may venture to affirm of the rest of mankind, that they are nothing but a bundle or collection of different perceptions, which succeed each other with an inconceivable rapidity, and are in a perpetual flux and movement Treatise,I,IV,6).

The exact meaning of this passage, i.e., Hume’s view of the ontology of personhood, is somewhat problematic. It would seem, at first sight, that Hume understands personhood in mentalistic and, by implication, dualistic terms. For he is describing the self or person here, and in many other passages, as the sum total of perceptions, emotions, and sensations — as something made up of these mentalistic contents. At the same time, he asserts that these contents "are in a perpetual flux and movement," which would be a false assertion if that "flux" is constantly interrupted for long periods by sleep, and the physicalistic interpretation would be the correct one. We begin by examining the mentalistic interpretation.

2. The Mentalistic View of Personhood

In describing the human self or person, Hume routinely uses a dual terminology. He speaks of the self, the mind, the soul, or the person as an agent or an author of manifold cognitive and emotive acts, while, at the same time, in other numerous passages, he explicitly treats the self as nothing more than the sum total of these same activities interrelated among themselves in many complex ways but remaining independent of the dualistic substantial self and its alleged authorship. This conflicting terminology is no doubt due to the fact that Hume, by his own admission, feels necessitated to employ mentalistic terms to describe physical data — the brain processes. But this poses a serious problem for him. Having rejected the Cartesian concept of the soul as a mental substance as well as the agent or author of mental acts, he is forced to leave mental acts, i.e., ideas, impressions, perceptions, suspended, so to speak, in the air as acts without an agent.

No matter how serious at the first sight, this difficulty, arising from his dual terminology, need not perhaps be fatal to his conception of the self as the sum total, "a bundle or collection of different perceptions." Whenever he is using expressions suggestive of a mental agent such as "mind," "soul," or "self," he can perhaps be always interpreted as accommodating himself to the popular linguistic usage.

The really serious difficulties arise when we undertake to examine the components themselves that make up the human self or person according to Hume, i.e., his "ideas" and "impressions." These components prove themselves to be problematic to the degree where we may justly wonder whether the Humean self is a logically viable entity.

To begin with, "ideas" are said to "resemble" "impressions" from which they are supposed to arise, to be "copies" of impressions, to be pretty much the same as impressions, only less vivacious, less forceful. If we now inquire into how we discover these crucial differences between ideas and impressions, we find that Hume proposes introspection as the method of this discovery. In a word, we are invited to "observe" our "ideas" and compare them with their corresponding "impressions" and to see for ourselves that impressions are almost always more vivid and forceful than their corresponding ideas. The basic difficulty with this introspective method is that, in order to discover whether ideas indeed "resemble" impressions, are "copies" of them, and are less vivid than they, the impressions, we should be able to compare the two among themselves by means of a perception, which, by definition, can come only after the first two, the impression and the idea, have occurred. But such a comparison is plainly impossible because by the time we have formed that third comparing perception or idea, the impression from which the idea to be compared arose has ceased to exist, leaving that idea by itself with nothing to compare. Should we, however, suppose that the impression continues to exist along with the idea, then the two, the impression and the idea, would after all be only one entity and the comparison, which, by definition, presupposes here two entities would be out of the question. If, however, we attempted to compare the idea with its impression before this impression has ceased to exist, we would have only that impression alone but no idea yet. In either case, the alleged or recommended comparison would be a sheer impossibility.

Another serious difficulty concerns the concept of "resemblance" itself as employed by Hume. "Resemblance," you see, is ordinarily used to compare colors, shapes, sounds, or tastes of some external objects with those of other external objects, but not with those of ideas. We are probably joking when we say that ideas smell, taste, or have a color. For instance, we may call a painting or a photo a "copy" or "resemblance" of some particular landscape or maybe of another painting, but we are using the terms "copy" or "resemblance" purely metaphorically when we apply them to the ideas or impressions of these objects. Hume seems to confuse the merely metaphorical resemblance with the real one, which alone is required here. When he claims that ideas "resemble" their impressions, he is speaking metaphorically. But he is also speaking metaphorically when he claims that impressions resemble external objects. Impressions may indeed resemble other impressions as regards their strength or their constituent qualities just as apples may resemble other apples. But to assert that impressions resemble external objects makes just as little sense as to claim that, as a saying goes, apples resemble oranges. The comparison is purely metaphorical. Again, although "resemblance" might make some sense when applied to visual phenomena since we can perhaps meaningfully speak of "images" of visual things, it loses any meaning when we try to apply it to smells, tastes, touches, or sounds. "Resemblance" comes, after all, from "idea," i.e., "eidos" or "image" of visible things. So while an idea may perhaps be said, at least according to Hume, to resemble its visual impression, to be an "image" of it, it hardly makes any sense to speak of an "image," "eidos," or "idea" of the taste, smell, touch, or sound impressions, and still less sense to speak of these impressions as "resemblances" or "images" of external objects as their alleged causes.

The final difficulty with the distinction, so crucial for Hume’s epistemology and metaphysics, between impressions and ideas consists in the fact that, as he himself recognizes, ideas can sometimes be more vivid and forceful than impressions. But if in some cases the higher degree of vividness is not a reliable criterion of distinction between ideas and impressions, why is it to be considered as reliable in all others? To suggest that a greater vividness is abnormal for ideas is to beg the question by assuming that a lesser vividness is normal for them and that the real or imagined statistical majority is always to be regarded as the criterion of normalcy. Can we not assume that, in an overwhelming majority of cases, our ideas come in a variety of degrees of forcefulness and vividness just as in case of impressions, without us being aware when these ideas are more and when less vivid than their corresponding impressions, especially since the ideas as memories and their corresponding impressions are often separated by long intervals of time? The greatest difficulty with the comparison of ideas and impressions in order to determine their respective vividness is that, as we saw earlier, no comparison of the two kinds of perceptions is logically possible. For the comparing perception presupposes that both the idea and the impression are available for comparison, that they exist at the same time. But, as we saw, only one of them is available at any given time. Should we say that we can compare an idea with a memory of its impression, we would be merely comparing two ideas, and not an idea with an impression, since a memory, for Hume, is an idea.

The above considerations force us to conclude that the constituents that make up our minds or persons, the "ideas" and "impressions," cannot be logically conceived in mentalistic terms. For what Hume calls ideas, may be nothing else than continuing impressions or continuations of impressions with alternating degrees of vividness.

3. The Physicalistic View of Personhood

Now since the mentalistic interpretation of Hume’s fundamental constituents of selfhood or personhood is highly problematic, if not logically impossible, we should attempt a physicalistic one, especially since these constituents seem to be eminently susceptible to it. After all, Hume is conceiving the self or person as a dynamic process rather than as a static agent that occasionally erupts into activities by positing some thoughts or emotions. Since the preceding analysis of Hume’s "perceptions" has shown to us that his distinction between "impressions" and "ideas" is gratuitous, we are going entirely to dispense with it in this part of our paper.

What Hume treats as an "impression" that allegedly gives origin to an "idea", its "copy", logically should be treated merely as the point of inception of the perceptual process of longer or shorter duration, and what he calls an "idea" should be viewed as nothing more than the "impression" itself whose external (or internal) stimulus, after producing it, has ceased to operate. The contemporary research into physico-chemical electrical processes of neurons makes it at least difficult, if not impossible, to conceive ideas on the metaphor of a Lockean jar of pickles or even on that of a Cartesian agent with ideas as it products.

If we are to use any metaphor at all, it should be one that denotes a dynamic process. We should compare ideas and impressions to the sound waves arising, for instance, from a piano when one of its keys is struck or from a bell, and spreading perhaps long distances before gradually subsiding completely at the end. In this way, ideas would not be viewed as "copies" of impressions. They would be viewed — and correctly — as these impressions themselves, i.e., as continuations of these impressions minus the stimuli that gave rise to them. What Hume calls "ideas," these entities dangling in the air detached from their moorings, their "impressions", "would be viewed as relatively stable systems of electrical intra-neuronal and inter-neuronal currents that function as indispensable parts or factors within the physico-chemical processes of individual neurons and of their specialized clusters. Individual neurons and their specialized functional clusters, the "seats" of "perceptions," could also be compared to boiling cauldrons, each containing within itself a large system of smaller "pots" each of them filled with a variety of chemicals interacting with each other in countless ways. Every new electrical wave viewed by Hume as an "idea" or a cluster of new waves or "ideas" would be incorporated more or less successfully into this highly complex and dynamic system for longer or shorter periods of time in direct proportion to the strength of the original stimuli as well as of additional reinforcements.

We see at once the crucial role which the strength or weakness of the original stimuli and their subsequent reinforcements plays in determining the duration of the individual components of the perceptual system, when we consider that out of thousands, maybe millions of all kinds and "sizes" of perceptions that we may have in the course of one single day we are hardly able to remember a few dozens or at most hundreds. And if we go back years or decades in our lives, the number of our memories or "ideas" eventually may shrink to a zero. What is happening to all other countless billions of our memories or "ideas?" The same thing which happens to sounds of bells and movements of hurricanes: they cease to exist. Also the brain capacity to absorb and to retain new contents or incorporate new processes, like that of any other organ in its functions, is strictly limited. It may be like the capacity of a container filled to the brim: new liquid can perhaps be forced into it only by dislodging some of the old one.

It would seem then that the life expectancy of neuronal processes, i.e., "ideas" or memories, is in direct proportion to the brain’s being a "blank tablet" in early years or to the force of some particular experiences or "impressions," or stimuli in later years, or to both factors combined. The less the neurons or their clusters are cluttered up with large numbers of electrical currents and chemical processes of often trivial consequence, the easier it must be for a few major currents to establish themselves, to get "rooted" more permanently.

The difficulty facing our conception of "ideas" on the metaphor of vibrating chords or ringing bells, i.e., as processes, is that, while an interruption of the vibrating chord brings to an end that process, the process called "idea" seems to be capable of numerous interruptions without being brought to an end. Maybe "ideas" are after all somewhat like pickles that can be lifted out of a jar and dropped back into it as often as we please without any change in their life expectancy. This difficulty can be easily answered by pointing out that available evidence strongly suggests that physiological processes called "ideas" are incapable of an interruption, that they merely alternate between being peripheral or unperceived, or "unconscious" in mentalistic terminology, and emerging into the primary focus as "memories" from among numerous other perhaps momentarily less forceful competitors — past and present "perceptions." It seems quite obvious that it is the survival needs of an organism that determine which neuronal processes or "memories" are going to move into focus and at what frequency rate while the rest, the overwhelming majority of them, are doomed to faster or slower decay on the periphery.

Should someone dispute our contention that "ideas" or neuronal currents are "incapable of an interruption" by pointing out that sleep notoriously interrupts them regularly, we could validly argue that it is precisely the REM stages of sleep — that is, if such stages as described exist at all — that prove, at least in part, the continuity of our "ideas" as dreams. Our critic could further urge that deep sleep dreamless stages certainly prove that "ideas" conceived as dreams or neuronal processes cease to exist during these stages, and that, therefore, after we wake up we should have no "ideas" or memories of what has preceded our sleep. And since this is notoriously not the case, are we not entitled, our critic might ask, to assume that we are closer to truth when we compare "ideas" to pickles rather than to processes. After all, we can repeatedly pick them up while they remain the same or maybe leave at least a certain sediment which helps our memories to arise after we wake up, or our dreams to occur when we reach the REM stage.

Our answer to this potentially fatal objection would be that the rapid eye movements of the REM stages do not prove at all, that, while we are in this stage, we are actually dreaming, if dreams can be explained, as they certainly can and as some researchers are presently doing, as pseudo-memories, or clusters of them, that occur at those times or in those persons in which the physiological closure between sleeping and waking is maybe too slow or otherwise imperfect, i.e., when being only partially awake, we are actually "spinning" our dreams while at the same time unconsciously imagining and believing that we are remembering them. This theory would also explain why, upon being awakened from a deep sleep stage, the subjects cannot report any dreams. For even during the REM stage rapid eye movements are, as we just saw, no reliable proof of dreaming during that stage since they can plausibly be interpreted as a part of many other signs or symptoms, commonly observed, of general restlessness before an organism’s gradual awaking. We must conclude perhaps that apparently no presently known criteria, not even the dreamer’s own memory, prove beyond a doubt that any dreams at all occur during sleep at any of its stages. This obviously does not mean that dreams do not occur during all stages of sleep — only that their occurrence during any stage cannot be reliably proven. But it certainly can and must be postulated as highly plausible, i.e., it must be assumed as real since otherwise we would not be able to interpret "ideas," memories etc. as continuing processes each with a life expectancy of its own, and would be forced to fall back on the self-contradictory mentalistic interpretation.

Should we, however, accept the above physicalistic interpretation of "ideas", memories, and dreams, many puzzling "psychological" phenomena would stop being puzzling. The regular or "normal" phenomena of schizophrenia, "multiple" personality, somnambulism, hallucination, and religious visions of the saints, not to mention occasional occurrence of dreams parallel with or within the normal waking consciousness, would turn out to be, on a closer examination, just so many different forms, or perhaps degrees, of dreaming.

But if dreaming is an ongoing process both in sleep and at least sometimes in waking state, that is, if dreaming and waking can sometimes coexist, how does the dream differ from waking consciousness? Apparently the dreaming as a neuronal process does not stop just because another neural process triggered by a new set of stimuli, called waking consciousness comes onto the scene. The dreaming process only becomes unconscious, with the waking consciousness dislodging the dreaming consciousness from the primary focus, in other words, depriving dreaming, either entirely or at least in part, of its own consciousness. The waking consciousness is capable of doing this simply because, being a "bundle" or a system of reactions to external stimuli of various kinds, it is more powerful than the dream system that consists of only decaying perceptions or long past reactions to the stimuli, which, like so many echoes, are slowly vanishing into distance. We may add that by consciousness we do not mean "the second order" thoughts supervening upon "the first order" thoughts, as some mentalists contend. Consciousness is apparently nothing else than perception or the neuronal process itself insofar as it, at some points of its duration, due to its interaction with other processes, happens to acquire more strength than at other points.

The constraints of time do not permit us to analyze much of Hume’s mentalistically conceived philosophy in physicalistic terms. A couple of examples will suffice. Hume’s theory of thinking as association of ideas possessing a merely self-contradictory "reality" would have to be conceived in terms of complex neuronal processes interacting with each other according to certain quite predictable patterns determined by the stimuli operating over longer or shorter periods of time. Hume’s theory of causality as association of ideas in terms of the priority in time as well as the constant conjunction could be interpeted in terms of the association of neuronal processes. Two neuronal processes that have occurred in constant conjunction a certain number of times could be supposed to have established some sort of affinity, even a kind of unity, so that when, in the future, one process occurs the other one is triggered so the two occur as inseparable parts of one and the same process. The same principle of constant conjunction of neuronal process could be used to interpret in physicalistic terms also other parts of Hume’s philosophy such as his theory of truth, both conceptual and factual, his theory of morality, and perhaps even that of religion.

4. Conclusion

The preceding considerations make it plausible that, since the mentalistic interpretation of the Humean selfhood or personhood runs us into unsolvable difficulties, the physicalistic interpretation is the correct one and even in harmony with Hume’s thought. They strongly suggest that, from the ontological point of view, the human person is essentially the sum total of neuronal processes that occur as reactions, immediate or remote, to stimuli. These reactions can be viewed either as haphazard or, perhaps more correctly, as constituting some form of system. One of the advantages of this view of personhood is that it provides to human beings solid moorings within the realm of biology rather than leaving them suspended in the realm of mythology. It provides a firm basis for animals’ rights advocates: non-human animals must or at least can be viewed as possessing rights since they are persons. For their selves or persons are made up of the same neuronal processes, instead of being mere "machines," as Descartes imagined. However, these remarks do not intend in the least to dismiss the "realm of mythology" as worthless. Human beings might very well possess spiritual souls as claimed by Descartes and a long tradition. Only we have no evidence, at least no rational evidence, to support our belief in such souls.

bluered.gif (1041 bytes)


Beck, Lewis White, Essays on Kant and Hume, New Haven and London: Yale University Press, 1978.

Churchland, Paul M., Matter and Consciousness, revised edition, Cambridge, Mass,: MIT Press, a Bradford Book, 1993.

Hume, David A Treatise of Human Nature, Garden City, New York: Doubleday & Co., Inc., Dolphin Books 1961.

Hill, Christopher S., Sensations: A Defense of Type Materialism, New York: Cambridge University Press, 1991.

Livingston, Donald W. & King, James T., eds., Hume: A Re-Evaluation, New York: Fordham University Press, 1976.

Nelson, R.J., The Logic of Mind, Boston: Kluwer Academic Publishers, 1989.

Norton, David Fate, ed., The Cambridge Companion to Hume, Cambridge: Cambridge University Press, 1993.

Rosenthal, David M., ed., The Nature of Mind, New York: Oxford University Press, 1991.

bluered.gif (1041 bytes)


Back to the Top

20th World Congress of Philosophy Logo

Paideia logo design by Janet L. Olson.
All Rights Reserved


Back to the WCP Homepage