Did people really sleep in two segments before industrialization?

Many early texts mention a 'first sleep' and a 'second sleep', and the historian A. Roger Ekirch argues that this segmented pattern was the predominant one in pre-industrial Western Europe. How common it was, how regular, and whether it was a universal human habit are all contested. The claim that it was the norm in the West rests on historical texts, not on measurements of sleep. [1]

Ekirch published 'Sleep We Have Lost' in The American Historical Review in 2001 and later wrote the book At Day's Close. He describes a pattern in which most people retired between 9 and 10 p.m., woke around midnight or later, stayed awake for up to an hour or so, and then took a second sleep until dawn. He bases this on some two thousand textual references of varying quality. Mentions of first and second sleep appear in early modern texts, in medieval literature and, less often, in classical works such as Livy, Plutarch and Virgil's Aeneid. [1] [2]

Ekirch also points to a 1990s National Institute of Mental Health experiment led by Thomas Wehr. After four weeks of 10-hour days, participants no longer slept in one stretch but in two roughly equal halves, separated by a period of one to three hours of wakefulness. [3] [4]

Niall Boyce, in Medical History in 2023, criticised Ekirch's work and reinterpreted the early modern English evidence. Ekirch replies that Boyce's reading is erroneous or selective and holds that segmented sleep predominated. Ekirch also says he never described the waking interval as well-defined: people's timetables varied, and a French essayist in 1752 noted that the time of first sleep is not the same for everyone. For Ekirch, variation does not undo predominance, which does not mean universality. [1]

A separate challenge came from a team led by Jerome Siegel, who studied three pre-industrial societies without electric light. The members of all three did not regularly wake for extended periods in the middle of the night. The authors inferred that the biphasic pattern was probably not present before humans moved into Western Europe, and suggested it may have resulted from longer winter nights at higher latitudes. [4]

Ekirch calls the latitude inference highly questionable. He says segmented sleep occurred across Europe all year, including in southern countries, and that summer shifted sleep onset by at most about an hour. He also cites accounts of first and second sleep from the Tupinamba of Brazil, Oman, Surinamese Maroons, the Asante and Fante, the Tiv and the Sinhalese, among others. He concedes that segmented sleep was not predominant among all pre-industrial peoples, and calls the Siegel study a welcome but singular counterpoint. [4]

Ekirch connects the shift to consolidated sleep with artificial lighting and changing attitudes during the Industrial Revolution. The search results do not show that a consensus among specialists has been reached. [5]