The Committee of Sleep by Deirdre Barrett: An Honest Review

🕑 15 min read

What This Book Is

The Committee of Sleep, published in 2001, argues that dreaming does real cognitive work and that you can put it to use deliberately.

Barrett builds the case from three sources. She collects historical episodes in which a discovery or a work of art is said to have arrived in a dream. She interviews living artists, scientists, athletes and inventors about their own dream material. And she reports her own experimental work at Harvard on whether ordinary people can direct their dreams toward a chosen problem.

The subtitle names the audience precisely: how artists, scientists and athletes use their dreams, and how you can too. It is a popular science book with a practical payload, and it does not pretend otherwise.

Who Wrote It

Deirdre Barrett taught at Harvard Medical School and spent her career on dreams, hypnosis and mental imagery. She edited the academic journal in the field and has published peer reviewed research alongside books for general readers.

This matters more than it usually would. The dream shelf is dominated by people with no research background at all, and the difference shows in how Barrett handles her weakest material. She names the uncertainty. She reports the failures in her own studies. She does not claim the cases prove anything they do not prove.

An author who tells you where their argument is soft has bought some trust, and she spends it carefully.

Where the Title Comes From

It is Steinbeck's, from a remark that a problem difficult at night is resolved in the morning after the committee of sleep has worked on it.

The phrase is doing a lot of work and Barrett knows it. It suggests deliberation, a group of somebodies conferring on your behalf while you are unavailable, which is a much stronger claim than anything the book actually demonstrates.

As a description of the experience it is exact. You go to bed stuck and wake unstuck, and the sensation is of having been given something rather than of having worked it out. As a description of what happened it is a metaphor, and later in this review it becomes clear why the distinction matters.

The Famous Cases

The collection is the most entertaining part of the book and includes every case you have heard plus a good many you have not.

Dmitri Mendeleev, exhausted after days of arranging the elements, seeing the table laid out complete. August Kekule dozing and seeing a snake seize its own tail, giving him the ring structure of benzene. Otto Loewi waking twice in the night with the design for the experiment that demonstrated chemical transmission between nerves, and which later won him a Nobel Prize. Elias Howe and the needle with the hole at its point.

From the arts: Mary Shelley's waking dream that became Frankenstein. Robert Louis Stevenson, who described small helpers he called brownies supplying his plots, including the central scene of Jekyll and Hyde. Paul McCartney waking with the melody of Yesterday intact and spending weeks convinced he must have heard it somewhere. Keith Richards finding the Satisfaction riff on a tape recorder he did not remember switching on. Salvador Dali building an entire method around hypnagogic imagery.

She adds sportspeople who rehearsed technique in dreams and surgeons who worked through procedures, from her own interviews. That material is fresher than the historical anthology and gets less attention than it deserves.

How Well Documented Are They

This is where an honest review has to slow down, because the cases are not equal and the book presents them in a single stream.

Loewi's is the strongest. He described it close to the event, the experiment is on the record, the result was real and the account has held up.

Kekule's is the most famous and the shakiest. He told the snake story about twenty five years after the fact, at a banquet held in his honour, in a speech designed to be memorable. Historians of chemistry have pointed out that his path to the ring structure is traceable through his published work without needing the dream at all. That does not prove he invented it. It means the single most repeated case in the whole literature rests on an after dinner reminiscence.

Howe's sewing machine story appears to have entered circulation through family accounts well after his death and has no contemporary support.

Mendeleev's is contested in a similar way, with his own later telling being the main source.

The musical cases are better, because McCartney and Richards were alive to be asked and told their stories consistently over decades.

The Problem With Collecting Successes

Set aside the reliability of individual cases. There is a deeper issue with the method.

The book assembles the times a dream produced something. It cannot assemble the times a dream produced nothing, because nobody records those and there is no register of them to consult.

Everyone dreams, several times a night, for their whole life. The number of dreams involved is astronomically large. Against that denominator, a few dozen striking cases across the entire documented history of science and art is a very small number, and it is not obvious that it exceeds what coincidence alone would supply.

What the collection establishes is that dream solutions occur. What it cannot establish is that they occur at a rate that makes dreaming a meaningful part of how problems get solved.

Barrett is aware of this. It is the reason she ran an experiment.

The Harvard Study

Her incubation study asked students to select a real problem, personal or academic, and to focus on it before sleep for a week, recording dreams each morning.

The reported outcome was that a majority dreamed about their chosen problem at some point during the week, and that around a quarter had a dream they judged to contain a solution.

That is a genuinely interesting result and it needs its limits stated. The sample was small and made of undergraduates. Whether a dream was about the problem and whether it solved it were judged by the dreamer. There was no control condition of people who thought hard about the same problem before bed and did not sleep on it. And a person primed all week to notice a topic will notice it, in dreams and everywhere else.

None of that makes the finding worthless. It makes it a reason to try the technique rather than proof that it works.

The Incubation Method

Here is the practical core, and it is short enough to state completely.

The protocol is free, takes minutes, has no plausible downside, and produces a dream journal as a side effect whether or not it solves your problem. As a proposition that is hard to argue with.

Which Problems Suit Dreaming

The most valuable idea in the book gets less emphasis than the anecdotes, which is a shame.

Barrett distinguishes between problems dreaming might help with and problems it will not.

It suits problems needing a visual or spatial insight, since dreaming is strongly visual. It suits problems where you are stuck in a framing and need it broken, since the associative looseness of dreams is exactly a failure of the constraints that keep you in a rut. It suits problems where you already have all the pieces and cannot see the arrangement.

It does not suit problems requiring sustained logical steps, arithmetic, or careful verification. Those depend on the faculties that dreaming suspends. Nobody proves a theorem in a dream, and the famous cases are all restructurings of material the person already had rather than new information arriving.

Notice what follows from that. Every case in this book was a specialist deeply immersed in the problem. Mendeleev had been arranging elements for days. Loewi had held the question for years. The dream did not supply expertise; it rearranged expertise that was already there.

Whether There Is a Mechanism

Barrett gestures at neuroscience and the gesture holds up reasonably well against what came after.

During dreaming sleep, activity in the prefrontal regions that handle planning, self-monitoring and reality checking is reduced, while associative and visual areas remain active. That combination is a decent account of why dreams connect things waking thought keeps apart, and why they feel unquestionable while they are happening.

Sleep also has a well-supported role in memory consolidation, including the reorganisation of material rather than its simple storage, and some experimental work has found improved insight on restructuring tasks after sleep.

What none of this establishes is deliberate problem solving. The picture is compatible with dreams being a byproduct of consolidation that occasionally throws up something useful, which is a much weaker claim than the book's title implies and is not ruled out by anything in it.

The Trouble With the Committee

The metaphor is charming and it smuggles something in.

A committee has members. It has intentions. It works on your behalf and hands you a result. Every one of those implications goes beyond the evidence, and the book's framing keeps them quietly in play.

The material supports something more modest: that a brain in a particular state, primed with a particular problem, sometimes produces a rearrangement the waking version had not reached, and that you can improve your odds of catching it.

That is still worth having. It is not a committee, and readers arriving from the more mystical end of the dream literature should notice that Barrett is not claiming a source outside the dreamer. Whatever is doing this is you.

How It Relates to Lucid Dreaming

The two are often confused and they are separate projects.

Lucid dreaming aims at awareness inside the dream, so the dreamer knows they are dreaming and can direct what happens. It takes considerable training and the induction techniques are a discipline in themselves.

Incubation requires none of that. You aim the dream beforehand and then stay out of it. There is nothing to control and nothing to become aware of.

Some people do combine them, tackling problems inside a lucid dream deliberately, and Barrett mentions the possibility. Her method is deliberately the easier one, available to anyone who can remember a dream at all, which is a large part of why the book is useful.

Barrett and Jung

Worth setting side by side because the practices look identical and the philosophies are not.

For Barrett a dream is cognition, continuous with waking thought, running on the same material under different conditions. Set it a task and it may perform it.

For Jung a dream is communication from a part of the psyche with its own view of your situation, frequently one you would rather not hear. You do not set it tasks. You listen to what it raises.

Both keep a journal. Both take the material seriously. But Barrett asks whether the dream answered your question, and Jung would ask what the dream is telling you that you did not think to ask, and the difference in what you do next is total.

The honest position is that neither has demonstrated its account. Barrett's has the advantage of proposing something testable.

Who Should Read It

Read it if you are stuck on something specific and want a method to try that costs nothing.

Read it if you like the history of discovery, because the case collection is excellent company even where the cases are shaky.

Read it if you want an example of popular science written honestly, with the author's own null results left in.

Do not read it expecting the thesis to be proved, because the evidence assembled cannot do that and Barrett does not claim it does. And do not read it for meaning, symbolism or what your dreams say about you. That is a different bookshelf entirely.

Frequently Asked Questions