What did people do at night before electricity?

October 15, 2026

It is 11:48 at night and you are in bed with one small rectangle of light in your hand. One more video. The alarm is set for seven, and you are doing the sum. Go back far enough and your family had no rectangle and no light switch. After sunset the only light came from fire, candles and the moon. So what did people do with all that night?

One historian's answer: they slept twice

In 2001 the historian Roger Ekirch published "Sleep We Have Lost" in the American Historical Review. In old English court testimony he kept finding the words "first sleep": a first sleep, a stretch awake, then a second sleep until morning. He later reported more than 500 references of this kind, from ancient to early modern sources, and described them again in 2015: people left their beds and, less often, their houses.

It is a famous idea. It is also contested, and the disagreement is below.

The night started long before bed

In 2015 Gandhi Yetish, Jerome Siegel and colleagues put wrist sensors on 94 adults in three societies without electric light: the Hadza in Tanzania, the San in Namibia and the Tsimane in Bolivia (Current Biology). On average they fell asleep about 3 hours and 20 minutes after sunset, and slept 5.7 to 7.1 hours a night, near the low end of industrial societies. Darkness was not bedtime. It was the evening, and it was hours long.

What filled those hours

In 2014 the anthropologist Polly Wiessner compared 174 conversations among the Ju/'hoansi of southern Africa, noted down by day and by night. By day about a third of talk was complaints, criticism and gossip, and 6% was stories. At night around the fire, 81% was stories, often about people the listeners knew, along with singing, dancing and ceremony. The tensions of the day mostly stayed in the day.

The caveat is that the 174 conversations are notes from 1974 at two camps of one people. They are a window, not the whole picture.

The waking hour

If Ekirch is right, the waking hour between sleeps was ordinary. Quoted in a 2023 review, he writes that people would urinate, smoke tobacco and even visit close neighbours, and that they made love, prayed and reflected on the dreams that preceded waking. Ekirch says it was sometimes called "the watching". If you wake at three and start counting the hours left, the claim is that your ancestors might have called that simply the middle of the night.

Is it real?

The evidence points two ways.

For: in 1992 the psychiatrist Thomas Wehr kept healthy men in the dark for 14 hours a night (about eight by most accounts) for roughly a month. Their sleep slowly split into two bouts of several hours with one to three hours of quiet waking between them. A tiny sample, in a lab, with no fire and no neighbours.

Against: the Yetish team reported that sleep in those three societies was not interrupted by extended periods of waking. Ekirch has questioned their conclusions. And in 2023 the historian Niall Boyce reread the English sources and found segmented sleep was "not the sole or even necessarily the predominant model".

So hold the two sleeps loosely. Some people, in some places, on long nights, may have slept that way. Not everyone.

What changed

The part nobody much disputes is the electric light itself. In 2015 Horacio de la Iglesia and colleagues studied two Toba/Qom communities in northern Argentina with the same background and culture. One had free electricity and the other relied on natural light. In winter, the people without electricity slept about 56 minutes longer, mostly because the electrified village went to bed later. In summer both villages got up at about the same time.

Wiessner's one-line summary: artificial light turned potential social time into potential work time. For most of human history the night was firelit and shared. Now it is lit, and it often ends on a screen.

The one test of the light

In 2015 Anne-Marie Chang, Charles Czeisler and colleagues at Brigham and Women's Hospital brought 12 healthy adults into a lab (PNAS). Each read for four hours before bed in dim room light, for five evenings with a printed book and five with an iPad. After the iPad evenings, people took nearly ten minutes longer to fall asleep. They made less melatonin, their body clocks ran later, and the next morning they took longer to feel alert.

The limits: twelve people, a lab, four hours of reading (far more than most of us do), and it tested the light from a screen, not what was on it.

The fix

What was tested was a paper book in dim light instead of a glowing screen. It was not giving up electricity and not waking yourself at midnight.

In your evening, read a paper book in dim light instead of a screen.
Try it for five evenings, which is how long the lab difference took to show.
Judge it on how long it takes you to fall asleep and how alert you feel in the morning.
If you wake at three, remember that some historians would call that the middle of the night.

This reports what was measured in one small trial. It is not medical advice, and whether it changes your night is yours to find out.

Sources

Ekirch (2001), Sleep We Have Lost, American Historical Review. muse.jhu.edu
Yetish et al. (2015), Natural sleep and its seasonal variations in three pre-industrial societies, Current Biology. pmc
Wiessner (2014), Embers of society: firelight talk among the Ju/'hoansi Bushmen, PNAS. doi
Wehr (1992), In short photoperiods, human sleep is biphasic, Journal of Sleep Research. europepmc
Boyce (2023), Have we lost sleep?, Medical History. cambridge.org
de la Iglesia et al. (2015), Access to electric light is associated with shorter sleep duration in a traditionally hunter-gatherer community, Journal of Biological Rhythms. europepmc
Chang, Aeschbach, Duffy and Czeisler (2015), Evening use of light-emitting eReaders negatively affects sleep, circadian timing, and next-morning alertness, PNAS. europepmc

Film version, with the sources on screen: What Did People Do at Night Before Electricity? (6:44).