Sleeping Through the Night Is a Relatively New Invention

Photo: Vintage Images/Getty Images

On a sultry night in New Orleans in late August 2005, I found myself unwillingly enrolled in a crash course on both the tenacity and the fragility of the rules that govern sleep. On that night, my wife, Devora, and I packed up our car with suitcases, some books, snacks, CDs, and toys for the kids, then caught a few hours of sleep before heading out at four in the morning. Hurricane Katrina was bearing down, and suddenly we were rudely cast outside the gates of normal sleep, trying to find our way back in.

Yet we were fortunate. Our two children were young enough not to be overly anxious about the storm — for them it was an adventure. We had a car and some money in the bank. We had devoted family and friends far from the storm’s path who were ready to help in whatever ways they could. I had professional connections to people who could help me get back on my feet if disaster struck. But what we thought most about in the hours leading up to our evacuation was where we would sleep.

Shelter is an obvious human need, one that grows most intense during slumber. The need to find safe sleeping accommodations made our trek understandable, even inevitable. But much of our effort went toward something else, toward fulfilling a “need” that was culturally conditioned rather than biologically dictated: Devora and I were intent on finding a hotel spread out of harm’s way that would allow the kids their own sleeping spaces. Part of this desire was rational: We wanted a space large enough that we could shelter the kids from our own anxious conversations. But on another, semiconscious level we simply wanted to re-create, on the fly, the aspects of “normal” sleep that bear most directly on children. Like so many other parents, we had spent countless painful nights trying first one method and then another to teach our children to sleep on their own, in one straight shot through the night, at regularly scheduled times. We had achieved the sleep schedule and configuration of a typical middle-class American family, and we didn’t want the storm to put all that effort to waste.

What’s strange, though, is that the sleep schedule we were trying so hard to re-create is a relatively new invention: of all the elements that make up what we consider the natural or normal way to sleep, not one of them seems to have been in force at any time anywhere before around 1800 in Europe and North America. This is worth reiterating: virtually nothing about our standard model of sleep existed as we know it two centuries ago, including the ideal of eight hours of unbroken sleep.

* * *

The first scholar to put consolidated sleep—today’s standard “one straight shot throughout the night”—under the microscope was historian Roger Ekirch. In his fascinating 2001 essay “Sleep We Have Lost: Pre-Industrial Slumber in the British Isles,” Ekirch revealed that across a wide range of nationalities and social classes in early modern Europe and North America, the standard pattern for nighttime sleep was to do it in two shifts of “segmented sleep.” These two sleeps—sometimes called first and second sleep, sometimes “dead sleep” and “morning sleep”—bridged an interval of “quiet wakefulness” that lasted an hour or more. (The interval itself was sometimes called “the watching.”)

Ekirch’s subsequent work offered evidence that a segmented nighttime pattern persisted well into the twentieth century in many non-Western locales, including among indigenous cultures in Nigeria, Central America, and Brazil. During the period of nighttime wakefulness, Ekirch showed, different cultures elaborated rituals—of prayer, lovemaking, dream interpretation, or security checks—and while the rituals varied, the pattern itself was so pervasive as to suggest an evolutionary basis that somehow became disrupted in the modern West.

So why did this mode of sleeping fall by the wayside, in favor of the eight-hour, lie-down-and-die model that has become an unquestioned norm? According to Ekirch, the main culprit was the spread of powerful artificial lighting in the nineteenth century in Europe and North America, and later in other locales. As activities that were previously nearly impossible to conduct under cover of darkness became fashionable under an ever-widening penumbra of powerful light, Europeans and Americans gradually shifted their bedtimes later. And as the available space between first and second sleeps shrank, the pattern of two nocturnal sleeps—and the enchanted space between them—became untenable. So complete was the transition to consolidated sleep that an American newspaper advice column in 1911 counseled readers who couldn’t sleep well to take their sleep in two shifts—as if this were a novel suggestion! Ekirch argues that the reason so many of us experience middle-of-the-night insomnia (the kind that comes after a few hours of sleep), is that ever since electric lights reordered our sense of time, we’ve disrupted our ancestral—perhaps our evolutionary—rhythms. And while Ekirch eventually came to view the reasons for the shift from segmented to consolidated sleep as more complicated than just exposure to light—including shifts in technology, changing cultural attitudes toward work and rest, and the economic pressure to manage time more efficiently under industrial capitalism—powerful artificial lighting, he wrote, still “exerted the broadest and most enduring impact upon sleep’s consolidation.”

Ekirch’s thesis has taken surprising hold in some medical and scientific circles. In 1993, at about the same time that Ekirch was doing his historical research into the erosion of segmented sleep patterns by the advent of electric lighting, psychiatrist Thomas Wehr of the National Institute of Health was conducting clinical experiments in which subjects were deprived of artificial lighting for several weeks. Wehr found that under these circumstances, the subjects began to gravitate toward a common pattern of waking up for approximately an hour after midnight. During this interval, the brains of Wehr’s subjects showed higher levels of prolactin, a hormone that reduces stress and that is also released during orgasm. Struck by the congruence with his own historical findings, Ekirch contacted Wehr and the two exchanged notes. Perhaps this hormonal activity, they speculated, was the biological basis for the fertility rituals that were so common during the interval between first and second sleep and that seemed to have vanished in the modern world. (The 16th-century physician Thomas Cogan, for instance, advised that intercourse occur not “before sleepe, but after the meate is digested, a little before morning, and afterward to sleepe a while.”)

Sleep specialists in the United States and Europe have begun to take these findings seriously, reevaluating the common wisdom that healthy sleep means uninterrupted nocturnal slumber. Russell Foster, a professor of circadian neuroscience at Oxford University, saw a therapeutic value in this new view of what constitutes normal sleep: “Many people wake up at night and panic,” he said in an interview. “I tell them that what they are experiencing is a throwback to the bi-modal sleep pattern.”

So does that mean, as Ekirch’s and Wehr’s work suggested, that humans are evolutionarily adapted to sleep in two shifts at night? Not all scholars agree. The historian Sasha Handley, for example, questioned whether Ekirch’s sources were representative enough to indicate a universal model of sleep across millennia of human history. A recent scientific study by Jerome Siegel of the University of California at Los Angeles and Gandhi Yetish (now also at UCLA) presents a very different evolutionary scenario. Studying sleep patterns in three contemporary hunter-gatherer societies in Tanzania, Namibia, and Bolivia that lacked electricity, Siegel, Yetish, and their group of researchers found little evidence of segmented sleep at night, but some evidence of daytime napping, especially during the summer months. Surprisingly, the average sleep time among these societies was approximately six hours per night, but the lower number, compared to the eight hours recommended in contemporary Western medicine, had none of the adverse health effects—including obesity, diabetes, and mood disorders—that authorities so often link to sleep deprivation.

Even more surprisingly, this supposedly ancient sleep pattern more closely resembles the contemporary Western predilection for consolidated sleep than the preindustrial segmented variety that Ekirch documented, except that six hours a night is usually deemed unhealthful and often blamed on overexposure to artificial light, computer screens, and the like. In a coauthored article, the Siegel/Yetish group claimed that because the tribes they studied shared environments similar to those in which the human species evolved, their sleep patterns represented the truly natural way to sleep: they were the “core human sleep patterns … characteristic of pre-modern Homo Sapiens.”

Yet this claim, too, may be too sweeping. Ekirch, in a response, allowed that segmented sleep may not have been the pattern for “all preindustrial peoples in the non-Western world,” but he pointed to dozens of examples provided by anthropologists to show that it was a predominant one. In a commentary on the Siegel/Yetish group’s article, another group of prominent sleep researchers rejected the conclusions as an “over-interpretation” of data, arguing that without a control group, their study simply could not yield “normative values” for an evolutionary default pattern for human sleep. The Siegel/Yetish group defended their conclusions and in a sense doubled down, suggesting that the evolutionary pattern they had discovered should make sleep clinicians question the notion that sleeping less than seven hours a night is detrimental to the health of adults.

Meanwhile, several anthropologists who study sleep patterns weighed in with their own doubts about the Siegel/Yetish team’s conclusions. No culture, Kristen Knutson argued, is a “living fossil,” and so extrapolating from current practices to a universal evolutionary basis for sleep is problematic. Matthew Wolf-Meyer went further, pointing out that the very societies that Siegel and Yetish studied were far from premodern hunter-gatherers: all three groups had centuries of experience in dealing with colonial administrations and state governments; one had a burgeoning tourist industry; another had members working for big logging companies; a third had extensive trade networks with other communities—and all of these historical factors likely had some effect on their patterns of sleeping and waking. The arguments are still spinning out as I write this.

So is it “natural” to sleep through the night, or instead to break sleep up into segments? The dispute between the Siegel/Yetish camp and the others raises profound questions not only about what might be the natural way to sleep, but about whether any particular sleep pattern is more natural for Homo sapiens than others. The argument seems irresolvable, at least by me; but it does indicate the depth to which our society—including its academic researchers—feels dissatisfied with sleep: we are looking to the ancestral past as well as to medical experts for solutions. Because sleep is always governed by society’s rules and environmental pressures as well as by physiological needs, though the search for one unchanging “natural” way to sleep seems unlikely to solve our current collective sleep frustrations. The efforts are noble, and they yield fascinating accounts of sleep’s mutability; but as the experts present conflicting visions of the best, most naturally human way to sleep, they may only feed sleep anxiety rather than conquer it.

Perhaps we can view these conflicting visions of sleep’s evolutionary forms the way we view different consumer products, picking the one that suits our particular sleep quandary the best. Each of the positions staked out in this academic battle might be psychologically comforting to contemporary troubled sleepers in search of some historical or evolutionary perspective on their troubled passage through the night and their drowsiness during the day. The Ekirch position suggests that if you can’t stay asleep through the night, you’re not an insomniac, but simply more in touch with ancestral rhythms than your culture wants you to be. The Siegel/Yetish position says, in a sense, that we should stop worrying about sleep loss, since, for all our distractions, we don’t need as much sleep as the experts tell us we do. And those who argue that there is no single way to sleep naturally or correctly give us license to be more forgiving of our own sleep patterns, to stop thinking that there is a “right” way that we’re failing to achieve.

Adapted excerpt from WILD NIGHTS: How Taming Sleep Created Our Restless World by Benjamin Reiss. Copyright © 2017. Available from Basic Books, an imprint of Perseus Books, LLC, a subsidiary of Hachette Book Group, Inc.

Sleeping Through the Night Is a Relatively New Invention