Skip to main content



Almost all of us have had dreams, yet few could say with confidence what they are, beyond agreeing that they occur during sleep and have some likeness to waking experience. Yet most people would in all probability accept the kind of definition given by philosophers, for example Plato's "visions within us, which are remembered by us when we are awake and in the external world" (Timaeus, 46a) or Aristotle's "the dream is a kind of imagination, and, more particularly, one which occurs in sleep" (De Somniis, 462a). Indeed, such notions seem to be summarized in the Oxford Dictionary 's definition: "A train of thoughts, images, or fancies passing through the mind during sleep; a vision during sleep." Dreams are striking phenomena, and the more superstitious see in them signs and portents of what is to happen; even today divination by dreams has not lost its popularity. A more sophisticated way of looking at dreams is to regard them as revealing something about the sleeper, either about his physical condition or about his mental state. An example of the former can be seen in the diagnostic technique used in the temple of Aesculapius; patients seeking a cure had to sleep all night in the temple precincts and would experience a "vision" that would indicate the disease or its cure. Many writers had suggested that mental states were revealed by dreams, but there was little serious study of the idea until the work of Sigmund Freud and his followers. Freud's doctrine of the unconscious, and the way in which it is revealed in dreams and other less rational activities, is important for psychiatry; but he had little to say about the nature of dreams that is of interest to the philosopher, though the fact that they had been found worthy of study may have resulted in an increase in philosophic concern about the problems they raise.

While we are having them, dreams often appear to be as real as waking experience; children have to be told that the object of their terror "was only a dream," hence not part of the world. William James expressed this well in his Principles of Psychology : "The world of dreams is our real world whilst we are sleeping, because our attention then lapses from the sensible world. Conversely, when we wake the attention usually lapses from the dream-world and that becomes unreal." This similarity has led philosophers to pose the question, "How can you prove whether at this moment we are sleeping, and all our thoughts are a dream; or whether we are awake, and talking to one another in the waking state?" (Plato, Theaetetus, 158). In perhaps the most famous example of the difficulty of distinguishing dreams from reality, René Descartes introduced his method of universal doubt. He concluded, "I see so manifestly that there are no certain indications by which we may clearly distinguish wakefulness from sleep that I am lost in astonishment" (First Meditation ). Descartes finally resolved his doubts in this respect by appealing to a criterion of consistency: "For at present I find a very notable difference between the two, inasmuch as our memory can never connect our dreams with one another, or with the whole course of our lives, as it unites events which happen to us while we are awake" (Sixth Meditation ). Such a consistency criterion has been adopted by several more recent writers on the topic. Unfortunately, this will not do the task required, for consistency can only be used as a test of a particular experience by waiting to see what happens in the future. It would enable me to tell that I had been dreaming, not that I am now dreaming; for however confident I am of the reality of my surroundings, something may happen in the future that will reveal them to be part of a dream. Further, the problem remains whether any consistency discovered is a real or a dreamed one.

The failure of consistency to provide a test need not be worrying, for the times in which genuine doubt arises are normally those involving memoryI am not sure if this event actually happened or whether I dreamed it. In such a case I would normally try to remember some part of the event that would have left a mark in the physical world, and then see if there is such a trace of the event; if there is nothing, I conclude that I had dreamed the occurrence. In spite of Descartes's remark, it is rare that we are in doubt about whether we are dreaming. The expression "I must be dreaming" is normally used in circumstances when I am quite sure that I am not dreaming, to express surprise at some pleasant occurrence, for example the arrival of a friend whom I thought to be somewhere distant. There are times when we are aware that we are dreaming, though normally a dream presents itself as real and no questions about its genuineness arise. It seems that the conviction that one is dreaming does not come from a previous doubt within the dream about the status of the experience; it just occurs, though sometimes accompanied with a feeling of relief. But in most cases the dream convinces us that it is reality, in that no doubt or questioning arises during its course. The difference between dreams and hallucinations lies in the fact that there is nothing external to dreams with which they can be compared, no tests that can be applied. For if we did apply a test in a dream, the result would be to confirm its reality. Philosophers have sought for some mark or test that would solve this problem, but there is none available. Any suggested sign of reality could be duplicated in the dream, and if all dreams bore marks of unreality, then there could not even be confusion over the remembering of them.

It has been generally agreed that dreams are due to the workings of the imagination no longer under the control of the intellect or the senses, as can be seen from the quotations at the beginning of this article; but it would seem that in such contexts the meaning of the word "imagination" had been left vague, serving rather as an indication of puzzlement than as a solution to a problem. Some recent work by physiologists has led to the suggestion (by W. Dement and N. Kleitman) that dreaming is correlated with rapid eye movements during sleep. Such a suggestion would seem to confirm Aristotle's remark that "dreaming is an activity of the sensitive faculty, but of it as being imaginative" (459a). The use of a physiological criterion for dreams has been challenged by Norman Malcolm in his book Dreaming (1959), which is clearly the most important contemporary discussion of the whole topic. In the course of it he challenges virtually all the assumptions made by previous philosophers. In criticism of the physiological work, he asserts that waking testimony is the sole criterion of dreaming (p. 81). The obvious difficulties that arise from the common belief that external stimuli can cause or influence the course of a dream, or that observers can sometimes tell from bodily movements that a sleeper is having a violent dream, he dismisses by means of a definition that dreams can take place only when the subject is sound asleep and that a person who is sleeping cannot respond to external stimuli (pp. 2526). It might be thought that Malcolm was here doing the same thing for which he criticizes the physiologists, namely introducing a new concept of dreaming, for surely the ordinary unsophisticated notion includes the possibility of our recognizing that someone asleep is having a dream, in some cases at least, as well as the possibility of the dreamer being aware that he is dreaming. If both of these beliefs are ruled out by a philosophical argument, then it would appear that the concept of dreaming held by most people has been changed in important ways. Most of the points made in the earlier part of this article would be understood by those with an unsophisticated notion of dreaming.

Malcolm's arguments are, however, powerful and subtle, and his critics, of whom A. J. Ayer is perhaps the most eminent, have found it not at all easy to refute them. Malcolm bases his reasoning on Ludwig Wittgenstein's Philosophical Investigations, in particular on the dictum that "an 'inner process' stands in need of outward criteria" (I. § 580). Malcolm argues that we can come by the concept of dreaming only by learning it from descriptions of dreams, "from the familiar phenomenon that we call 'telling a dream'" (II, p. 55). To talk of "remembering a dream" is to use the word remember in a sense different from the normal, for there is no external criterion by which we can check our memory, as there is in the paradigm cases of remembering, that of remembering an event in the public world, which can be checked by ourselves and others. What is told sincerely on waking is the dream, because there is no other way of finding out what, if anything, occurred while the teller slept. (This can be compared with Freud's reliance on the narration of the dream, but this was essential for its use in diagnosis. Nevertheless, Freud was willing to evaluate critically the veracity of actual dream accounts on the basis of his theory or as a result of previous analysis of its dreamer. For most purposes, it made no difference whether the dream account or the dream itself was being considered; Freud's concern was with different problems.)

Yet Malcolm rejects Ayer's suggestion that this theory amounts to saying that "we do not dream, but only wake with delusive memories of experiences we have never had." Malcolm is clearly correct in stressing the importance of the report of a dream and its difference from reports of public events; what the dreamer says on waking is final. Though we must learn the use of the word dream in the way Malcolm indicates, this does not rule out the possibility of its use being extended by further experience, for instance, correlating dream reports with observations of the dreamer, as Dement and Kleitman have done. The trouble is Malcolm's use of the term criterion, which is never clearly explained, and which seems to lead him into a crude verificationism; he even talks of "the senselessness, in the sense of the impossibility of verification, of the notion of a dream as an occurrence" (p. 83). A further consequence of Malcolm's use of the dream report as a criterion for dreaming is that it becomes impossible to talk of children having dreams before they have learned to speak (p. 59). If, as Malcolm apparently wishes to maintain, words can be used only if their application can be strictly verified, then many ordinary uses will be cut out. That we now have a particular concept of some mental activity does not make it impossible that further experience will lead us to introduce a modification of it, in which case the way in which we first learned it may have no bearing on the criterion of its use. For example, many words used in the sciences are first learned in an approximate way and their criteria of application refined in the course of education. Malcolm claims that his argument applies only to words that refer to "inner" processes. What he seems to do, however, is extend Wittgenstein's argument, valid in the area Wittgenstein intended it for, beyond its legitimate sphere. The primary use of the word dreaming depends upon the notion of telling a dream, but this does not prevent an extended use. Peter Geach remarks that Wittgenstein mentioned in a lecture Lytton Strachey's description of Queen Victoria's dying thoughts: "He expressly repudiated the view that such a description is meaningless because 'unverifiable'; it has meaning, he said, but only through its connexion with a wider, public, 'language-game' of describing people's thoughts" (Mental Acts, p. 3). In fact it is only because we know what it is to dream that we can understand the difficulties raised by talk of "verifying" reports of dreams.

Ayer also criticizes Malcolm's denial that one can make assertions while asleep, but in this case with less effect. It does seem clear that the words "I am asleep" cannot be used to make a genuine assertion, because such an utterance would contradict what was asserted, just as the only possible truthful reply to the question, "Are you asleep?" is "No." An absence of reply is what would lead the questioner to assert that the man was really asleep.

In spite of Malcolm's statement (p. 66) that there is no place for an implication or assumption that a man is aware of anything at all while asleep, many would claim, and understand others' claims, that they had become aware that they were dreaming. This also implies that they were aware that they were asleep. As part of a dream narrative, such awareness could be reported by the words, "I suddenly realized that it was all a dream." Clearly, such an assertion could not be taught by ostensive means. However, there seems no reason why, having learned how to use the ordinary concept of dreaming and expressions such as "I suddenly realized that," we should not combine the two into an assertion that would be commonly understood to apply to a possible experience. Malcolm's claim that a person must be partially awake to be aware that he is dreaming (pp. 3844) seems, as suggested above, a redefinition of the term for which no adequate reason is advanced.

Malcolm wishes to say that the problem of what dreams are is a pseudo problem; he refuses to allow that they can be called experiences, illusions, workings of the imagination, or anything else they have been thought to be by previous philosophers. Ayer concludes his criticism of Dreaming by maintaining that dreams are experiences and mostly illusions, and "are found to be so by the same criteria that apply to illusions in general." This remark is difficult to understand; here Malcolm's stress on the report of the dream comes into its own; in recounting it I am not claiming that these things happened. Because while dreaming there is no possibility of making assertions about my experiences to other people, to describe dreams as illusions makes no sense. Malcolm has clearly made out his case in this respect. On the other hand, it seems difficult to deny that dreams are experiences, if only because the description is sufficiently vague to cover almost any "mental" phenomena. The same may be said of talking of dreams as being composed of images; here dreaming is being used as one of the examples of mental imagery, a vague concept. In spite of Malcolm's work, the problem of the nature of dreaming is still open for philosophic discussion, but any future examination of the problem will have to take his book fully into account. Many philosophers would still wish to assert that dreams occur, that they take place during sleep, while admitting that the meaning and justification of such claims is by no means clear.

See also Aristotle; Ayer, Alfred Jules; Descartes, René; Freud, Sigmund; Imagery, Mental; James, William; Malcolm, Norman; Plato; Psychoanalysis; Unconscious; Wittgenstein, Ludwig Josef Johann.


historical works

Aristotle. De Divinatione per Somnum (On Prophesying by Dreams ). Translated by J. I. Beare, in Works of Aristotle, Vol. III. Oxford: Clarendon Press, 1931.

Aristotle. De Somniis (On Dreams ). Ibid.

Bradley, F. H. "On My Real World." In Essays on Truth and Reality. Oxford: Clarendon Press, 1914.

Descartes, René. Meditationes de Prima Philosophia. In The Philosophical Works of Descartes, Vol. I, translated by E. S. Haldane and G. R. T. Ross. Cambridge, U.K., 1934.

Freud, Sigmund. The Interpretation of Dreams. Edited and translated by James Strachey. New York: Basic, 1955.

Freud, Sigmund. Introductory Lectures on Psycho-analysis. Translated by J. Riviere. London, 1949.

James, William. The Principles of Psychology. New York: Dover, 1950.

Malcolm, Norman. Dreaming. London: Routledge, 1959.

Plato. Theaetetus (158a). Translated by B. Jowett. Oxford, 1871.

Plato. Timaeus (46a). Ibid.

Russell, Bertrand. Our Knowledge of the External World. London, 1949.

Sartre, Jean-Paul. L'imaginaire. Psychologie phénoménologique de l'imagination. Paris: Gallimard, 1940. Translated by B. Frechtman as The Psychology of the Imagination. London, 1949.

Wittgenstein, L. Philosophical Investigations. Translated by E. Anscombe. Oxford: Blackwell, 1953. Especially pp. 184, 222223. Also relevant is Malcolm's review of this work in Philosophical Review (October 1954): 530559.


Ayer, A. J. "Professor Malcolm on Dreams." Journal of Philosophy 57 (1960): 517535. Malcolm's reply and Ayer's rejoinder in ibid. 58 (1961): 294299.

Chappell, V. C. "The Concept of Dreaming." Philosophical Quarterly 13 (July 1963): 193213.

Dement, W., and N. Kleitman. "The Relation of Eye Movements during Sleep to Dream Activity: An Objective Method for the Study of Dreaming." Journal of Experimental Psychology 53 (1957): 339346.

MacDonald, M. "Sleeping and Waking." Mind 62 (April 1953): 202215.

Manser, A. R., and L. E. Thomas. "Dreams." Proceedings of the Aristotelian Society Suppl. Vol. 30 (1956): 197228.

Putnam, H. "Dreaming and 'Depth Grammar.'" In Analytical Philosophy, edited by R. J. Butler. Oxford, 1962.

A. R. Manser (1967)

Cite this article
Pick a style below, and copy the text for your bibliography.

  • MLA
  • Chicago
  • APA

"Dreams." Encyclopedia of Philosophy. . 14 Aug. 2018 <>.

"Dreams." Encyclopedia of Philosophy. . (August 14, 2018).

"Dreams." Encyclopedia of Philosophy. . Retrieved August 14, 2018 from

Learn more about citation styles

Citation styles gives you the ability to cite reference entries and articles according to common styles from the Modern Language Association (MLA), The Chicago Manual of Style, and the American Psychological Association (APA).

Within the “Cite this article” tool, pick a style to see how all available information looks when formatted according to that style. Then, copy and paste the text into your bibliography or works cited list.

Because each style has its own formatting nuances that evolve over time and not all information is available for every reference entry or article, cannot guarantee each citation it generates. Therefore, it’s best to use citations as a starting point before checking the style against your school or publication’s requirements and the most-recent information available at these sites:

Modern Language Association

The Chicago Manual of Style

American Psychological Association

  • Most online reference entries and articles do not have page numbers. Therefore, that information is unavailable for most content. However, the date of retrieval is often important. Refer to each style’s convention regarding the best way to format page numbers and retrieval dates.
  • In addition to the MLA, Chicago, and APA styles, your school, university, publication, or institution may have its own requirements for citations. Therefore, be sure to refer to those guidelines when editing your bibliography or works cited list.