Did people used to sleep in two shifts?

First sleep and second sleep

Historian A. Roger Ekirch found references to first sleep and second sleep in diaries, literature, court records, and other documents. He argues that divided sleep was common in the preindustrial societies he studied, particularly Western Europe. ¹

How common it was, and what individual references mean, remain disputed. The evidence is worth examining before treating an older sleep pattern as advice for tonight.

Why historians disagree about the records

References to first sleep do not always tell us exactly how a night unfolded. In a 2023 paper, historian Niall Boyce reexamined material used to support segmented sleep. Some passages, he argues, could describe an early, deep portion of an otherwise continuous night, an occasional awakening, or sleep disturbed by illness. Boyce does not claim that segmented sleep never happened; he questions how confidently particular texts establish a widespread, uniform schedule. ²

The meaning of a phrase depends on its context. A letter mentioning work while waiting for sleep to return can preserve a personal difficulty as readily as a cultural norm. Establishing a cultural norm requires evidence of how often the practice occurred, among whom, and under which conditions.

Wehr’s experiment with longer nights

There is experimental evidence that people can develop divided sleep under particular conditions. In Thomas Wehr’s 1992 report, extending the daily dark period by moving participants from 16 hours of light and eight hours of darkness to ten hours of light and fourteen hours of darkness expanded their sleep, which usually separated into two bouts with a waking interval of one to three hours. ³

The experiment demonstrates that sleep can divide when people have a long daily period of darkness. It does not establish how common this was in earlier societies, or whether intentionally splitting sleep benefits health.

Sleep in communities without electric light

A 2015 field study followed members of the Hadza, San, and Tsimane communities using activity-based sleep measurements. Their nights did not regularly contain an extended waking interval. Sleep also began well after sunset, and its duration varied with season. ⁴

These are living communities with their own histories, environments, and cultures, not preserved samples of the remote past. Their sleep cannot prove what every ancestor did. Equally, the absence of a routine midnight interval in the groups studied complicates the claim that electric light displaced a single universal human schedule. A study in one environment and a historical archive from another need not tell identical stories.

What nighttime waking means today

History can give a nighttime awakening a less alarming interpretation without supplying a new timetable. Ask how the whole pattern works: whether there is enough opportunity to sleep, whether waking is distressing, and whether you can function during the day. There is little reason to manufacture a midnight break simply to imitate an account of the past.

Persistent trouble falling asleep or staying asleep deserves attention in its present context. NHLBI describes insomnia as a problem that can occur despite adequate time and conditions for sleep, and that can interfere with daily activities. Historical reassurance should not erase that distinction. ⁵

If waking is brief and you feel well during the day, an occasional interval need not mean something is wrong. Seek help when waking becomes persistent, distressing, or disruptive; the historical debate does not determine whether you need care.