One of the most common concerns people bring to sleep therapy is not difficulty falling asleep. It is waking in the middle of the night and being unable to get back to sleep. The experience is almost universally accompanied by a specific kind of dread: something is wrong with me. I am broken. This is going to ruin tomorrow.
That dread, it turns out, may be a more modern invention than the waking itself.
For most of recorded human history — from ancient Greece through the Middle Ages and into the early modern period — people did not sleep in a single consolidated block. They slept in two distinct phases, separated by an hour or two of wakefulness. This was not considered a disorder. It was simply how the night worked.
The History: First Sleep, Second Sleep
Historian A. Roger Ekirch of Virginia Tech spent years researching the history of nighttime before stumbling onto something he had not expected. Buried in court documents, diaries, letters, medical texts, and literary sources spanning from the early Middle Ages to the Industrial Revolution, he found more than 500 references to a segmented sleep pattern — what people of the time called "first sleep" and "second sleep."
The pattern was consistent across sources. People went to bed not long after dark, typically between eight and ten in the evening. They slept for three to five hours. Then they woke. For an hour or two, sometimes longer, they would lie quietly, pray, talk with a bed partner, tend to household tasks by candlelight, or simply rest in a state of calm wakefulness that contemporaries described as peaceful rather than distressing. Then they returned to sleep — their second sleep — until dawn.
The interval between the two sleeps was not something to be endured. Historical records are filled with references to this practice — Chaucer mentioned it in the Canterbury Tales, and it appears across cultures and centuries as an unremarkable feature of normal life.
What Ekirch concluded was striking: the assumption that human beings are supposed to sleep in one uninterrupted block is not a biological given. It is a historical artifact — a pattern that emerged, and rapidly displaced all others, only after the Industrial Revolution and the widespread adoption of artificial lighting in the late nineteenth century.
The Science: What Happens When You Remove Artificial Light
Ekirch's historical findings might have remained a curiosity were it not for the work of psychiatrist Thomas Wehr at the National Institute of Mental Health. In 1992, Wehr published findings from an experiment in which he divided the experimental day into ten hours of daylight and fourteen hours of darkness, simulating preindustrial winter conditions.
After a period of adaptation, all participants began sleeping like people did in the past — four hours of sleep, followed by a short wakeful break, and then another four hours of rest. The break between the first and second sleep was experienced as relaxation rather than fatigue. During the interval, prolactin levels were elevated — a hormonal profile associated with deep rest even while awake. Wehr's conclusion was careful but significant: the biphasic pattern appeared to emerge naturally when people were removed from the influence of artificial light and allowed to follow their biology without external scheduling pressure. This was not a sleep disorder. It was, in his assessment, a default mode that modern life had overwritten.
What This Means for People Who Wake at Night
It is estimated that nearly one in three American adults regularly wakes in the middle of the night. For most of them, this experience is accompanied by anxiety — the worry that the waking is pathological, that the lost sleep will accumulate into impairment, that something is fundamentally wrong.
That anxiety is clinically significant in its own right. One of the most well-documented maintenance mechanisms of insomnia is not the waking itself but the response to waking: the alarm, the clock-checking, the mental calculation of hours remaining, the escalating distress that activates the sympathetic nervous system and makes returning to sleep considerably harder. Worry about sleep disrupts sleep. Disrupted sleep generates more worry. The cycle tightens.
Ekirch has argued that knowledge of biphasic sleep history can help people with insomnia by "easing their anxiety" about middle-of-the-night waking — that reframing the experience from pathological to biological can interrupt the catastrophizing that drives the cycle.
This does not mean nocturnal waking is always benign or never warrants clinical attention. Waking driven by sleep apnea, pain, or mood disorder is different from waking that reflects a biological rhythm. But for people whose primary distress is the anxiety about waking — the conviction that something is wrong and the consequences will be severe — the history of biphasic sleep offers something genuinely therapeutic: the possibility that what they are experiencing is not a malfunction but a memory.
The Dream Life We Lost
One of the more intriguing aspects of Wehr's experiment was that participants consistently woke from REM sleep during the interval — the stage most associated with dreaming. Ekirch notes that the historical evidence bears this out: waking directly after dreaming "afforded people a pathway to their subconscious." "With morning dreams we don't have the opportunity to let our dreams settle," he observed. "The light goes on and we get out of bed immediately."
The biphasic sleep pattern, with its built-in interval of quiet wakefulness after a period of dreaming, created a natural moment for that material to be held, noticed, and integrated before the demands of the day arrived. Modern consolidated sleep, with its abrupt transition from sleep to alarm to daylight, forecloses that space almost entirely.
The Clinical Takeaway
None of this is a prescription to restructure your sleep into two deliberate phases. The evidence does not support abandoning consolidated sleep for most people, and the practical demands of modern life make genuine biphasic sleep difficult to implement. As Ekirch himself acknowledged: "There's no going back because conditions have changed."
What the history and biology of biphasic sleep offers, clinically, is a reframe. And reframes, in sleep therapy, are not merely cognitive exercises. They can directly alter the physiological response that maintains insomnia.
If you wake at three in the morning and immediately experience fear — if you lie there calculating lost sleep, catastrophizing about the day ahead, fighting the waking as though it were an attack — you are activating a stress response that makes returning to sleep harder and that, over time, conditions your nervous system to associate the bed with threat. That is the cycle CBT-I is designed to interrupt. Part of what CBT-I works with is precisely the belief that sleep must be continuous, that waking is pathological, that anything less than unbroken consolidation is failure.
Before electricity, before industrial work schedules, people lay awake in the middle of the night and considered it unremarkable. They prayed. They talked. They thought. Then they went back to sleep.
That possibility is still available. What has to change, for many people, is not the sleep itself. It is the story they tell about waking up.
APA Citations:
Ekirch, A. R. (2005). At day's close: Night in times past. W. W. Norton & Company.
Ekirch, A. R. (2015). The modernization of western sleep: Or, does insomnia have a history? Past and Present, 226(1), 149–192. https://doi.org/10.1093/pastj/gtu040
Wehr, T. A. (1992). In short photoperiods, human sleep is biphasic. Journal of Sleep Research, 1(2), 103–107. https://doi.org/10.1111/j.1365-2869.1992.tb00019.