Written by the Nuvirox Research Team
Key points
- Historian A. Roger Ekirch documented over 500 references to a segmented "first sleep, second sleep" pattern in pre-industrial Western texts.
- A 1992 National Institute of Mental Health lab study found human sleep spontaneously became biphasic under extended, artificial-light-free dark periods.
- This is a historical and biological curiosity, not a green light to treat all modern middle-of-the-night waking as harmless.
Short answer: historically and biologically, waking in the middle of the night has real precedent — but that doesn’t mean today’s insomnia is the same phenomenon. Historian A. Roger Ekirch's research into pre-industrial texts, and a striking 1992 laboratory study, both suggest segmented sleep may be closer to a natural human pattern than the consolidated eight-hour block most of us aim for today. That's a genuinely interesting finding — and also one that needs some careful caveats before you apply it to your own 3am wake-ups.
What is biphasic or segmented sleep, historically?
Ekirch's research, drawing on diaries, court records, and literature spanning centuries, identified a common pattern in pre-industrial Western societies: people went to bed shortly after sunset for a "first sleep," woke naturally around midnight for an hour or so of quiet activity (praying, reading, socializing, or sex), then returned to bed for a "second sleep" until morning. Ekirch found more than 500 references to this pattern across ancient, medieval, and early modern sources, and other historians have found supporting evidence in cultures well beyond Western Europe, including in Africa, South America, and Asia.
Is there actual lab evidence for this, or is it just historical anecdote?
The Wehr 1992 study. Researchers at the National Institute of Mental Health placed participants in extended 14-hour dark periods (mimicking long winter nights, without artificial light) for several weeks. Under these conditions, participants' sleep spontaneously reorganized into two distinct bouts separated by one to three hours of quiet wakefulness — closely resembling Ekirch's historical descriptions, and suggesting the pattern isn't purely cultural but may reflect an underlying biological tendency that modern artificial lighting has largely erased.
Was it universal, though? This is where the honest counterweight belongs. A separate research team (Yetish et al., 2015) studied natural sleep patterns in three pre-industrial societies still living without electricity today (in Tanzania, Namibia, and Bolivia) and found their sleep was largely monophasic — a single consolidated block — challenging the idea that segmented sleep is a universal default for humans without artificial light. Naps did occur in about 22% of summer days in that population, which is itself a mild form of biphasic sleep, but it wasn't the dominant nighttime pattern Ekirch described for pre-industrial Europe. Historians have also directly debated Ekirch's interpretation of the written record, with one 2023 paper in Medical History arguing the evidence for segmented sleep as the dominant pattern is less clear-cut than commonly presented — a disagreement Ekirch himself has since responded to in the same journal, defending his original analysis.
The honest summary: there's real historical and some experimental evidence that segmented sleep can occur naturally, particularly under long, artificially-unlit nights, but it wasn't necessarily the single universal human sleep pattern across all pre-industrial societies, and the historical record itself remains actively debated among historians.
Does this mean my 3am waking is nothing to worry about?
Not automatically, and this distinction matters. The historical and Wehr-study segmented sleep pattern involved a person who fell asleep easily both times, felt little distress about the wakeful interval, and functioned well during the day. Modern insomnia — waking at 3am with racing thoughts, difficulty returning to sleep, and daytime impairment — is a different experience even if the clock-time looks similar. If your middle-of-the-night waking is calm, brief, and you fall back asleep easily, the historical framing may be genuinely reassuring. If it involves real distress or an inability to get back to sleep, that's worth treating as insomnia, not dismissing as "natural."
Frequently asked questions
Is it normal to wake up in the middle of the night?
Brief awakenings are a normal part of sleep architecture for most people, and historical research suggests a segmented two-part sleep pattern may have real biological precedent under certain conditions. But if the waking involves real distress or trouble getting back to sleep, it's still worth addressing as insomnia rather than dismissing it.
Did people really used to sleep in two shifts every night?
Historical texts document this pattern extensively in pre-industrial Western Europe, and a 1992 lab study reproduced something similar under extended dark conditions. However, research on modern pre-industrial societies without electricity found largely single-block (monophasic) sleep, so it likely wasn't a universal human default.
Should I try to sleep in two segments on purpose?
There's no strong trial evidence recommending this as a treatment approach for modern insomnia, and forcing an artificial wake period could plausibly backfire for people already struggling with sleep anxiety. This is interesting historical and biological context, not a prescribed sleep protocol.
From Nuvirox
Why we formulated Sleep+ Restore
Whether your sleep is naturally consolidated or you notice a middle-of-the-night stretch of wakefulness, the goal is falling back asleep calmly rather than lying awake anxious about it. Sleep+ Restore’s 10 mg melatonin and calming botanical blend is formulated to support sleep onset, whether that's at the start of the night or after a natural wakeful interval.
Learn more about Sleep+ Restore →The bottom line
The historical case for biphasic, segmented sleep is genuinely well-documented, and a 1992 lab study reproduced something similar under artificial-light-free conditions. But it wasn't necessarily universal even in pre-industrial societies, and it's a different phenomenon from modern, distressing insomnia. If your own middle-of-the-night waking is calm and brief, this research may be reassuring context; if it's not, it's still worth treating as something to address.
What did people do during the middle-of-the-night wakeful period historically?
According to Ekirch's research, the wakeful interval — sometimes called "the watch" — was used for a wide range of quiet activities: prayer and reflection, interpreting the dreams from the first sleep (which were reportedly considered more vivid and meaningful at that hour), reading or writing for those who were literate, visiting neighbors, or intimacy between couples. It wasn't treated as a problem to be solved; it was simply an expected, unremarkable part of the night.
Why did segmented sleep disappear in modern societies?
Ekirch attributes the shift primarily to the spread of artificial lighting, which extended evening activity later into the night and compressed the available sleep window, alongside broader cultural changes during and after the Industrial Revolution that increasingly treated a single consolidated block of sleep as the ideal. He's noted the transition wasn't instant or uniform — some individuals, even relatively late into the early 1900s, still showed sleep patterns resembling the older biphasic norm, and the pace of change varied by social class and how urbanized an area was.
Do modern hunter-gatherer societies still show any segmented sleep pattern?
The Yetish et al. research on three pre-industrial societies found predominantly single-block sleep overall, but also documented regular daytime napping in a meaningful share of days, which is itself a mild form of biphasic sleep even if it doesn't match the specific midnight "first sleep, second sleep" pattern Ekirch described for historical Europe.
References
- Ekirch AR. Segmented Sleep in Preindustrial Societies. Sleep. PMCID: PMC4763365.
- Wehr TA. In short photoperiods, human sleep is biphasic. J Sleep Res. 1992;1:103–107. doi:10.1111/j.1365-2869.1992.tb00019.x.
- Yetish G, Kaplan H, Gurven M, et al. Natural sleep and its seasonal variations in three pre-industrial societies. Curr Biol. 2015;25:2862–2868. doi:10.1016/j.cub.2015.09.046.
- Biphasic sleep and human health: A theoretical paradigm for personalized sleep. ScienceDirect, 2025.
- Ekirch AR. Reflections on 'Have we lost sleep?' Med Hist. 2024. PMCID: PMC11655156.
Related reading
*These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease. This article is for informational purposes only and is not a substitute for professional medical advice.