The alarm goes off at 6:40. The app on your phone reports 6 hours 52 minutes of sleep. And somewhere at the edge of your mind a familiar thought surfaces: "short again". Because everyone knows you need eight hours.
How do we know that? And is eight really the line below which harm begins?
Where the eight came from
The oldest trace of the number is not medical but social. The slogan "eight hours' labour, eight hours' recreation, eight hours' rest" is usually attributed to Robert Owen, the British reformer, who is said to have formulated it in 1817. It was about workers' rights: limiting the working day, not measuring out anyone's sleep. But the idea of dividing the day into three equal parts proved remarkably durable.
Science later gave the number a different footing — and, interestingly, partly confirmed it. We'll come back to that.
What experts recommend
Look at the official recommendations and you won't find a single number. You'll find a range.
In 2015 the American Academy of Sleep Medicine and the Sleep Research Society published a joint statement. Fifteen experts put it this way: adults should regularly sleep seven hours or more. Regularly sleeping less than seven hours is associated with worse health outcomes.
More than nine hours may be appropriate for young adults, for people recovering from sleep debt, or for those who are ill. For everyone else, the authors write, it is uncertain whether long sleep is associated with risk.
The same year the US National Sleep Foundation published a broader table:
| Age | Recommended | May be appropriate |
|---|---|---|
| 18–25 | 7–9 h | 6 or 10–11 h |
| 26–64 | 7–9 h | 6 or 10 h |
| 65+ | 7–8 h | 5–6 or 9 h |
The "may be appropriate" column was added precisely because people differ. Here, eight hours is the middle of a range, not a required norm.
What large studies show
The most convincing data on sleep and health come from large, long observational studies. People are asked how much they sleep and then followed for years to see what happens to them.
In 2010 a team led by Francesco Cappuccio pooled 16 such studies: almost 1.4 million participants, followed for between 4 and 25 years. The shortest and longest sleepers (cut-offs varied between studies) were more likely to die during follow-up than those who slept "in the middle". The risk was about 12% higher for short sleep and about 30% higher for long sleep.
It is easy to draw the wrong conclusion here: "long sleep is more harmful than short sleep". But these are observational data. They show an association, not a cause.
Long sleep is often a consequence of illness, depression or an inactive lifestyle rather than their cause. The authors themselves call sleep duration a "predictor", that is, a marker associated with risk.
A 2022 study in the journal Nature Aging gives a similar picture. Almost 500,000 people in Britain aged 38 to 73 reported how much they slept and took tests of memory and attention. The best results — both on the tests and in well-being — were among those who slept about seven hours.
Both less and more were linked to worse scores. The authors are equally careful: this is an association, not proof that sleep itself causes the difference.
"Eight hours in bed" and "eight hours of sleep" are not the same
There is another subtlety that is easy to miss: we are not very good at knowing how much we actually sleep.
In a 2008 study, 669 middle-aged people wore an actigraph on their wrist — a motion sensor used to estimate sleep. The device recorded about six hours of sleep on average. The participants themselves believed they slept 6.8 hours. In other words, someone who "sleeps seven hours" might actually be sleeping noticeably less.
An interesting perspective comes from a 2015 study of three communities that still live without electricity: the Hadza in Tanzania, the San in Namibia and the Tsimané in Bolivia. Researchers observed 94 people. Their sleep lasted on average between 5.7 and 7.1 hours.
The time from falling asleep to waking up was longer — between 6.9 and 8.5 hours. In winter they slept about an hour longer than in summer.
So even among people without phones or night shifts, sleep was closer to seven hours than to eight. And "eight" more often describes time in bed than sleep itself.
When less really is less
Does all this mean that eight is simply a fiction and six will do? No. And here is the most important study on the subject.
In 2003 Hans Van Dongen and colleagues housed 48 healthy people aged 21 to 38 in a laboratory for two weeks. Some were allowed eight hours in bed a night, others six, others four. Another group went without sleep entirely for three days.
After two weeks of six hours in bed, attention and working memory had declined about as much as after one entire sleepless night. After four hours, as much as after two sleepless nights.
The most interesting part is something else. Participants barely noticed they were getting worse. Sleepiness rose in the first few days and then hardly changed, even though their test results kept falling. The authors write that people were largely unaware of their mounting deficits.
The data also produced an estimate of sleep need: for this group of young people, about 8.2 hours. So eight is not entirely plucked from the air. For some people it is close to the truth.
Where this story falls short
People differ, and this shows even at the level of genes. In 2009 geneticists found a rare mutation in the DEC2 gene in a mother and her adult daughter. Both slept about 6.25 hours a night on average, while other family members slept around eight. The women went to bed at the same time as everyone else but woke up an hour and a half to two hours earlier.
But such people are very rare. Far more common are those who think six hours is enough for them — and, as Van Dongen's study showed, that feeling is not the best judge.
There are other limits too. Sleep need changes with age. Much of the data comes from the United States and Britain. And most large studies rely on people's own reports of their sleep, even though we know how inaccurate those can be.
If sleep problems persist for weeks — difficulty falling asleep, frequent waking, constant daytime sleepiness — that is a reason to talk to a doctor rather than to pick a number for yourself.
What to take from this
Eight hours is neither a law nor a myth. It is the middle of a range that is easy to remember.
The main things the data say:
- for most adults, a sensible guide is seven hours of sleep or more — of actual sleep, not time in bed;
- regularity matters too: in a large UK study, regular sleep timing predicted mortality risk better than sleep duration (again an association, not proof of cause);
- the feeling that "this is enough for me" can deceive you if short sleep has become a habit.
A small experiment, with no promises: next time you have several free days in a row, try not setting an alarm and simply notice when you wake up. You may catch up on sleep for the first few days, so look at the last ones. It isn't a measurement, but it is a hint about how much sleep you in particular need.
Questions for discussion
- How much do you think you actually sleep — and how much time do you spend in bed?
- Have you noticed getting so used to short sleep that you stop feeling it?
- Why do you think the number eight is so persistent, even though experts talk about a range?
Log in or Sign up to join the discussion.