The "Purpose" of Dreams

by G. William Domhoff

Dreams are so compelling, and they often seem so weird and strange -- surely they must have a "purpose"; that is, an "adaptive role" in the maintenance of our bodily or psychological health. Furthermore, all the famous theorists who talk about dreams claim that dreams do have one or another purpose (although the famous theorists disagree on just what those functions are), but the best current evidence suggests otherwise. Dreams probably have no purpose!

So let's review the arguments and the evidence. We'll start with the claims made by psychoanalysts and clinical psychologists in the first 50 years of the century based on their work with patients, then turn to more recent claims, some of which are based on work in sleep and dream laboratories that flourished in the 1950's and 1960's. The views presented here are those of research psychologists who have studied dreams inside and outside the laboratory, especially David Foulkes and Calvin Hall. References to Foulkes' work are provided at the end of this document.

The first and most famous dream theorist of the modern era, Sigmund Freud, said that the function of dreams was to preserve sleep, but that theory from the year 1900 is contradicted by the fact that dreams happen very regularly at least five or six times per night in an active stage of sleep called REM sleep (after the rapid eye movements that are part of it, along with many other neurological and physiological changes). In other words, dreams don't just happen as we are about to wake up due to hunger pangs, sexual urges, or the need to go to the bathroom, as Freud thought way back when, before REM sleep was discovered in 1953.

The other famous dream theorist of the modern era, Carl Jung, an early follower of Freud who broke away to develop a very different theory, claimed that the function of dreams is to compensate for those parts of the psyche (total personality) that are underdeveloped in waking life, but Calvin Hall's studies of two-week dream series from students and longer dream journals from adults of all ages strongly suggest that dream content is continuous with waking thought and behavior. That is, if we are outgoing and active in our waking life, and not very introspective and reflective, then so too in our dream life, which contradicts Jung's view.

Still other dream theorists say that dreams have a problem-solving function. Dreams supposedly deal with problems we can't solve in waking life and offer solutions. But a variety of systematic studies find precious little support for this view. However, this is one of those places where we have developed "uses" for our dreams as part of our cultural lore. Looking at them in the light of waking day, and believing that they may be full of insight, we may sometimes come up with new ideas or insights while studying them. That is, we have invented a "use" for dreams, but that doesn't mean that problem solving is a psychological function of dreams built into us over evolutionary time.

So much for the claims by clinical theorists. Now we look at claims that have emerged in recent years, but are tied to no particular theory or famous theorist. They are the new "common sense" of our day, based on a reverence for physiological findings and the awesome capabilities of computers.

When REM sleep was first discovered, it was thought that dreams only occurred during that stage of sleep. This led to many functional theories about dreaming that were based on alleged functions for REM sleep. But we now have reason to believe that plenty of dreams happen in non-REM (NREM) sleep, especially late in the sleep period.

Furthermore, awakenings of children under age 5 in the sleep laboratory reveal that they only report dreams from REM sleep awakenings 20-25% of the time, so REM sleep does not automatically equate to dreaming. In addition, REM sleep can be found in all mammals, and it is unlikely that they are dreaming, i.e., imagining a world or story in which they are taking part and interacting with others. Dreams, as the pre-eminent American psychologist on dreams, David Foulkes, likes to say, are a "cognitive achievement." We only gradually develop the ability to dream. What all this adds up to is that REM sleep and dreaming are not the same thing, so whatever functions REM sleep may have cannot be taken as functions for dreaming and dreams.

The fact that we remember so few of our dreams -- a few percent at best -- also argues against any function for dreams. If they are so important, why don't we remember more of them? Furthermore, the people who remember a great many dreams don't seem to be any different from those who remember few or none, at least on the standard personality tests that have been used in many studies to date. If dreams are important, why aren't the recallers of them better off in some way?

With the advent of computers, it became fashionable to say that dreams are "clearing out the software" from a busy day, or that they are a form of "off-line" processing to save the good stuff and get rid of the useless. Aside from the fact that such theories show how susceptible our supposedly highest thinking is to the dominant technology of any given era, the problem with this theory of dream function is that very, very little in dreams deals with the events of the day. Often there is some little leftover from the day, first noticed by Freud and named "day residue," but the rest of the dream is a story that does not deal with actual events. The story is usually plausible and even mundane, and it often contains the most important people and concerns of our lives, but it is nonetheless a story.

We are thinking creatures because thinking is a valuable adaptation, but that doesn't mean that all forms of thinking have a function. Dreams at this moment in the collective findings of dream researchers seem to be a "throw-away" production, an off-hand story to while the night away. That judgment could be changed tomorrow by new and original studies by a new generation of young dream researchers, but right now the preponderance of the evidence weights against any physiological or psychological function for dreaming and dreams.

(Click here for a detailed refutation of the "problem-solving" theory of dreams.)

But Dreams Have Meaning

This doesn't mean that dreams have no "meaning," that they make no sense. To the contrary, dreams correlate with age, gender, culture, and personal preoccupations, as evidence on this site and in many research studies suggests.

"Meaning" has to do with coherence and with systematic relations to other variables, and in that regard dreams do have meaning. Furthermore, they are very "revealing" of what is on our minds. We have shown that 75 to 100 dreams from a person give us a very good psychological portrait of that individual. Give us 1000 dreams over a couple of decades and we can give you a profile of the person's mind that is almost as individualized and accurate as her or his fingerprints.

And, Yes, Dreams Have Their "Uses"

Even if dreams have no physiological or psychological functions, human beings gradually invented uses for them. In more technical terms, dreams have an "emergent" function that develops through culture. For example:

  • In a great many societies, dreams are used by shamans to diagnose illness (often thought to be caused by evil or angry spirits) and to enter the spiritual world. In that sense, shamans were the first psychoanalysts, and Freud and Jung are modern-day shamans.

  • In some societies, dreams are used to find game, predict the weather, or prophesy about the future. In our society, at least since about 1900, they have been used in psychotherapy, although not as much in recent years when the emphasis is on short-term therapy and on thinking sensible thoughts. Dreams can be an "occasion" for a reticent patient to talk more personally, especially when we note that people do not take as much personal responsibility for their dreams as they do most of their other thoughts, making dreams easier to talk about.

  • In our society, dreams are also an excuse to say something intimate to someone, maybe a tentative way to see if a deeper relationship is possible, as in "I had this nice dream about you last night."

  • Finally, the phrase "I had this dream last night..." is a platform to say whatever nonsense, lie, or fantasy someone might have on his or her mind, because there's no way to determine if the claim is true or not. Now, we have every reason to believe that people are honest when they are reporting their dreams for academic studies, as we explain in our section on representative samples and the quality of our data. But, when the popular dream hustlers tell you of their amazing dreams and promise that you can have similarly amazing dreams if you buy their book or attend their workshops, then hold on to your hat -- and your wallet.

Don't Like Your Dreams? Forget Them!

Most dream researchers think it is worthwhile to remember your dreams, and they have tips for improving your recall. But the evidence we have presented here suggests something else: they are not important, so perhaps not worth remembering. So, unless you find your dreams fun, intellectually interesting, or artistically inpiring, then feel free to forget your dreams. If they just upset you or leave you puzzled, then why bother with them?

But how does one forget his or her dreams? Well, since we've found that thinking of dreams as useful or important is the best predictor of high dream recall, then maybe telling yourself that they are not useful or important will lower your recall. It also helps to turn your attention to the external world and events of the forthcoming day when you wake up -- don't lie there and daydream and let your mind drift because that might lead to dream recall. And if you do happen to remember a fragment of a dream, don't try to recall the rest, just ignore it.

Of course, if you enjoy your dreams and they don't bother you, dream away, and have fun!

Suggested Reading

  • David Foulkes (1985). Dreaming: A cognitive-psychological analysis. Hillsdale, NJ: Lawrence Erlbaum Associates.

  • Corrado Cavallero and David Foulkes (1993). Dreaming as Cognition. New York: Harvester/Weatsheaf. [Especially chapters 1, 2, 6, and 7]

  • Allan Moffitt, Milton Kramer, and Robert Hoffman (1993). The Functions of Dreaming. Albany, NY: SUNY Press. [Especially the chapters by Foulkes and Antrobus.]

Go back to the Dream Library index.
dreamresearch.net home page dreamresearch.net contact info