Sleep

High on any list of futurological fantasies is the control of sleep.

We spend a third of our lives asleep. Wasted time: let’s do better. So far, we’ve failed. And there is reason to think we will continue to do so, not only in the near future but for ever – that innovations to reduce or eliminate sleep are not possible.

Activity levels in single-celled organisms often fluctuate predictably but they do not really sleep since they cannot properly be said to be awake. Pretty much every other form of life does and one does not need to rise very far up the tree of life before one can delete this sentence’s opening two words. Some worms and flies sleep, and like everything else that needs sleep they suffer without it. Most fish sleep; it may be that all of them do, even those that appear not to and need to keep moving in order to drive water over their gills and survive. Birds, certainly, are able to sleep with half their brain at a time, leaving one eye open and alert. The small number of apparently sleepless fish may be doing something similar.

There has to be a good explanation for sleep. Its costs are too high for there not to be. An organism that sleeps is not resting and conserving energy. It is doing both of those things, but it is doing something more since it is possible to rest and conserve energy without being asleep. When you sleep you withdraw from the world. Animals sleep not only when there is nothing much to do, like a lion after a big meal, but also when there is a very great deal to do, from searching for food to ensuring one does not become a meal. The lack of activity and awareness comes at a price of vulnerability and missed opportunity. The presence of sleep is different from the absence of activity. Something vital is going on.

Apes, ourselves included, sleep more, and sleep better, than monkeys. This may have come about simply because we got bigger while still living among the trees. Being large and arboreal entails the need to create a bed to stop yourself waking up as you fall. Creating those beds put us in a position, literally, to sleep better. Sleeping a lot is not particularly human; other creatures sleep more. Those that graze generally sleep less while those that hunt and gorge take a longer break. But it may still be that evolving to get a better night’s sleep was followed the next day by being able to live more brightly. ‘Our results’, wrote two physical anthropologists of their research, ‘suggest that relaxed sleeping postures may have been enabled by sleeping platforms as a behavioural facilitator to sleep, which could have allowed for greater sleep depth and next-day cognitive capacities in both great apes and hominins.’1

Odd that a period in which we seem lifeless should be so absolutely vital. Yet it is. A disease called fatal familial insomnia (FFI) illustrates that. Steadily destroying the brain, it destroys sleep, and with sleep gone, life follows. The torture of sleep deprivation makes it an unnerving way to die.

How do we know that the disease kills people by depriving them of sleep, rather than depriving them of sleep as it eats fatally away at their brain? ‘It is for the doctors to decide whether sleep is such a necessity that our very life depends on it’, wrote Michel Montaigne, ‘for we are certainly told that King Perseus of Macedonia, when a prisoner in Rome, was done to death by being prevented from sleeping.’2

Neither doctors nor Romans are needed: science stands ready. In 1995 two Chicago researchers took rats and put them on a half-metre disc suspended in a cage. There was as much food and water as the rats wanted. The cage was divided in two with the disc in the middle a couple of centimetres above water. In each experiment, one rat served as the control and the other as the subject. When the subject fell asleep, the disc was rotated: in order to avoid being pushed into the water, the subject rat needed to wake up and walk along the disc. When that happened the control rat, on the other side of the divide, needed to do the same. The difference was that when the subject was awake, the control could sleep. The object was to compare the fate of the two rats. ‘All rats subjected to unrelenting total sleep deprivation died, usually after 2–3 weeks’, ran the report. ‘Evidently sleep and its substages serve vital functions.’3

What is so essential? It cannot be rest because one can rest without sleeping. The subject rat was quite at liberty to rest. It might be that we are so used to sleep – meaning we have evolved in its presence for so long – that our rest cannot physiologically now happen without it. But that would not explain why sleep developed, with its enormous costs to evolutionary fitness, or why it persists. If it can be reduced to a need to rest, it would have been. Sleep could gradually be transferred to a state of lazing. It hasn’t happened and it doesn’t happen, not in us and not in any creature like us. Something other than the need for rest must have driven sleep and must be driving it still.

In FFI one aspect of sleep, when sleep becomes impossible, breaks through into waking life:

this condition of persistent subwakefulness or drowsiness was interspersed with episodes of unresponsiveness during which the patient would be animated by massive twitches and perform complex purposeful gestures that they subsequently referred to dreams. These episodes, which we termed oneiric stupor or enacted dreams, are one of the most characteristic features of FFI.4

The aspect of sleep that forced its way through was not restfulness but dreams. People being deprived of sleep to the extent it was killing them, people quite able to rest, experienced breakthrough convulsions of dreams. Might dreams be the essential quality that differentiates sleep from rest, without which we die?

For that to be the case, it would have to be true that the situation was not unique to humans, just as sleep is not. The researchers who killed the rats did not comment on whether their bodies made any last efforts to dream. But we can prove that other species, too, do dream. The term ‘oneiric stupor’ comes from the Greek for dreams, oneiros. Oneiric behaviour was a term used by the late Michel Jouvet, the neurophysiologist who helped explore the phenomenon of rapid eye movement – REM – sleep. He showed that dreaming appeared to be as much a part of the most evolutionarily ancient portions of our brain as it was of the later additions. He also showed that if you destroyed certain small parts of a cat’s brain, you eliminated the physical stillness of sleep. During the periods when an observer might expect it to be dreaming, during periods when it was otherwise asleep, the brain damage meant the cat was not still. It moved. It did what we might imagine a cat would do when acting out its dreams. There was no way of asking the cat, Jouvet wrote, but any sensible interpretation ‘would lead us to believe that cats do dream’.5

Brain imaging with electroencephalographs (EEGs) and functional magnetic resonance imaging (fMRI) scanners can tell you if someone is in REM sleep. We know from experience that people in REM sleep are often dreaming, and so with scanners and electrical recordings we can predict that they are. But the only way of knowing they have been dreaming, the only way in which the electrical and MRI appearances of dreaming were discovered, remains the gold standard for determining if dreaming has been taking place: you poke the person and when they wake up you ask them. Against that standard, the power of brain scans has nothing extra to offer. It is notable that dolphins and whales have no REM sleep. The likely implication is not that, uniquely among mammals, they do not dream. It is more likely that dreaming is not unique to REM sleep, only that REM sleep is the period when we remember our dreams if we are woken. The implication of all of this is that we sleep in order to dream.

It is impressive to say that we can see the activity of dreaming with an fMRI scanner but the impression is overdone if we don’t point out how little this tells us, or how limited it is. In a play Molière has a medical student being asked why opium makes people sleep. He is being examined to see if he is worthy of becoming a doctor. He answers that opium makes people sleep because it contains a sleep-inducing virtue.* His examiners applaud. When it comes to talking about sleep, science and medicine have a habit of saying things that do not amount to much, like highlighting the awesome power of fMRI to reflect what someone tells us when we ask them. Dreaming is often explained on the basis of being the random firing of neurons when the brain is at rest. It’s an explanation that doesn’t fit with what happens when the firing of neurons is randomly stimulated: dreams are not created. Nor does invoking randomness pass the test of Occam’s razor. It introduces a new mystery – how random neural firing can result in dreams – without explaining the existing mystery of the importance of sleep.

We sleep less with age and do not know why. It might be that we have less need of sleep and dreams, or that we are unable to sleep and dream with the youthful vitality we once had. It is likely to be some mix of the two, and likely also to be nothing that medicine, in the form of drugs or surgery or gene alteration, can do anything about. No single pathways or targets seem addressable. Nor should we expect them to be. If easy options with only benefits existed, evolution would have found them. The conclusion has to be that sleep, by virtue of the dreams that come, is an essential part of biological life and cannot be removed.

*

We complain about lack of sleep and are right to do so. We are right not because our lack of sleep is unique but because it is commonplace. Lives immemorial have been lived on the edge of the line between as much wakefulness as we can get and as little sleep as we need. There is a tension between the two and it is proper to feel the strain of it.

More evidence for the unavoidable position in our lives of sleep, and of futurological fantasies of overcoming it being poor predictors of the future, comes from our history of using drugs. Alcohol and opium are long familiar to us. Temporary sedatives and stimulants have their uses, but they are not the same as interventions that alter the nature of sleep. The host of modern drugs to influence sleep, drugs that have been searched for with all the energy of an industry quite aware of where the billions are, are largely failures. The benzodiazepines – Valium being a successful trade name – were sold and swallowed as drugs that enhanced sleep. Like alcohol and morphine, they are sedatives and in the long term their benefits are small or swallowed up by their harms. If you have long-term problems getting enough sleep, avoid benzodiazepines. Drugs to make us wakeful share similar properties. From coca, cocaine and amphetamines through to caffeine, their temporary effects can be useful to many. None, though, provides profound, lasting and helpful changes to our wakefulness or our need to sleep.

We are built to sleep. People vary in how much they need, but no single biochemical route exists to turn a nine-hour-a-night slugabed into a four-hour powerhouse of wakeful industry. We should not expect one to be discovered. Managing the tension between what sleep brings to life, and needing to be awake to live, has always lain beyond the province of medical technology. It will continue to do so.

* ‘Mihi a docto doctore / Demandatur causam et rationem quare / Opium facit dormire. / A quoi respondeo, Quia est in eo / Vertus dormitiva.’ Most English translations give the question as ‘Why does opium induce sleep?’ and the answer, ‘Because it has a dormitive virtue’. But the candidate is not even replacing sleep with a synonym: dormire merely becomes dormitiva. The Works of Molière, vol. 10, ‘Le Malade Imaginaire’, John Watts (London, 1748), p. 362.