This episode of the “Sleep Review Conversations” podcast features A. Roger Ekirch, PhD, and Siobhan Banks, PhD, discussing biphasic, or segmented, sleep patterns. Ekirch brings a historian’s perspective and Banks brings the modern-day perspective of a sleep researcher.

Ekirch’s book on the history of nighttime and sleep is At Day’s Close: Night in Times Past. Banks’ essay on biphasic sleeping can be found on The Conversation.

Hosted and produced by Rose Rimler, associate editor of Sleep Review

Run time: 48:43

Transcript Below

Rose Rimler, Sleep Review associate editor (RR): Hello and welcome. You’re listening to “Sleep Review Conversations,” a podcast by healthcare media company Allied 360. I’m Rose Rimler, associate editor of Sleep Review magazine and and I’m here with two guests today.

A. Roger Ekirch is a professor of history at Virginia Tech and is the author of five books, the most recent of which is American Sanctuary: Mutiny, Martyrdom, and National Identity in the Age of Revolution. His 2005 book, At Day’s Close: Night in Times Past, explored nighttime and sleeping in the pre-industrial age. Professor Ekirch holds a PhD and MA from Johns Hopkins University.

We also have Siobhan Banks with us today. Professor Banks is an associate professor in the Centre for Sleep Research at the University of South Australia. She holds a PhD from Flinders University in Australia and her research focuses on, among other things, the impacts of sleep deprivation and shift work. Prof Banks co-wrote an article about biphasic sleeping for the online magazine The Conversation last summer.

Professor Ekirch and Professor Banks, thanks so much for being here.

A. Roger Ekirch: Thank you.

Siobhan Banks: No worries.

RR: Professor Ekirch, you’re credited with making the discovery that it was once common for people to sleep in two shifts. Can you talk a little bit about what pre-industrial sleeping patterns were like and how you made this discovery?

Ekirch: Absolutely. Ironically, in undertaking the book you mentioned, At Day’s Close, many years ago, the subject that I most dreaded having to address was sleep, which I wrongly thought was not only a universal necessity but a universal constant. So I was perplexed as to how to make sleep fresh, new and as interesting as possible. So you can imagine my thrill, and it was one of the most exhilarating moments of my life, when after finding references in legal documents, literary correspondence…From 1500 to 1800, first sleep, second sleep, casually referred to as if these terms needed no explanation, punctuated by an interval of wakefulness of up to an hour or so in which people did just about anything and everything imaginable.

Absolutely critical to connecting those dots was reading in the New York Times about the research of Doctor Thomas Weir at the National Institute of Mental Health outside Washington, DC and a team of researchers who, really out of curiosity to discover how prehistoric people slept, subjected I think roughly 15 male subjects to a nighttime devoid of artificial light and after three weeks, their sleep became biphasic and, appropriately enough, I made this discovery surfing the internet at midnight and I knew immediately that what he was describing dovetailed with my own research and that’s truly how things got underway.

RR: What do we know about the sleeping habits of cultures outside of Europe and even in ancient times?

Ekirch: Well, that’s a matter of dispute actually at the moment. Siobhan, please correct me if I’m mistaken, if I have my dates wrong but roughly a year and a half ago a team led by Jerry Siegel at UCLA, having looked at three hunter-gatherer tribes, cultures, in Africa and South America, concluded first of all that they slept for I think around six hours or a little bit more, so that was the big discovery but then parenthetically, they also argued in the journal Current Biology that these individuals did not experience a segmented pattern of sleep which caught, of course, my attention and I then responded in a briefer piece in the journal Sleep several months later in which I disputed the notion that segmented sleep in pre-industrial Europe was the product hundreds, thousands of years ago of Africans migrating to the Northern latitudes of Europe and experiencing biphasic sleep during the winter. That was fundamentally incorrect on several points. Segmented sleep was common to all parts of Europe.

In fact, hundreds of references in the siesta cultures of the Mediterranean alone and nor was it seasonal, this pattern. And then I use this as an opportunity to broach the possibility that indeed this was not just a Western pattern of sleep but that it occurred internationally in pre-industrial societies. I didn’t claim that it was as dominant outside Europe as it was, I’m convinced, within Europe but I proceeded in this article, Sleep, to introduce about ten examples drawn mostly from anthropological literature of more than 100 years ago as well as descriptions that incontestably demonstrated that pre-industrial hunter-gatherer cultures, one virtually next door to one of the ones that the team from UCLA studied experienced biphasic sleep.

RR: Professor Banks, I wanted to throw this one to you, although either of you can feel free to jump in for any question. My question is about whether biphasic sleeping can be, can you consider that an ancestor of what today we might call sleep maintenance insomnia? So for some people, maybe sleep maintenance insomnia is a throwback to an old pattern of sleeping. Do you think that there’s any truth in that?

Banks: Well there are biological changes that do occur, for example, as we age and certainly sleep maintenance insomnia becomes more prevalent as we age. We find it easy to fall asleep but then wake up, have a period of wakefulness and often people then become very anxious about being awake during the night and so I think it’s very useful, even though there are these natural changes that may or may not have driven the previous biphasic sleep, I think it’s very good for people to understand that there are different kinds of sleep and that sleep doesn’t have to mean that you close your eyes and are comatose for 7 or 8 hours but in actual fact, it’s okay to wake up, that it’s okay to be awake during the night and that if you stay calm and relax, you will be able to go off to sleep in most cases.

So I think it’s this kind of discussion around what could be normal sleep is actually really, really helpful for a lot of people. We see that sleep has changed a lot in our current society. We squeeze it into the smallest slot in our day as possible, or if you’re a shift worker, for example, you’re sleeping at all different times of the day. So I just think it’s really great to have a discussion about sleep in general and that it doesn’t have to be one particular way, that we can still get health benefits and remain cognitively healthy from different types of sleep as long as it’s long enough. As long as we get enough sleep in a 24-hour period, we can structure it many different ways.

biphasic sleep

A. Roger Ekirch, PhD

RR: Professor Ekirch, in your research did you find reference to different ages of people sleeping in different patterns? Different classes of people sleeping in different patterns? Different types of sleeping habits?

Ekirch: No, with one significant exception and that was those who were so wealthy that they had access to candles and other more primitive forms of artificial illumination preceding the introduction of gas and then, of course, electricity and coupled with that, they did not have to wake up to go to work and they lorded this over their so-called social inferiors, so they went to bed much later. The interesting thing is their sleep was still biphasic. It was just that their timetable was altered so that they might wake after their first sleep around dawn and then, after a period of consciousness, return to slumber and not wake up for good until noon but I just want to say, I agree with virtually everything that Siobhan has mentioned and, bear in mind, I’m a PhD not an MD, so I’m certainly not going to contest what I think is a pretty voluminous body of literature for current sleep patterns that indeed as we age our sleep becomes more broken and we’re apt to take naps more often during the day.

The one thing that I would add to your original question as to whether middle of the night insomnia or sleep maintenance insomnia, the most common form, at least in the United States, might represent a persistent echo or remnant of this long, dominant, segmented pattern of sleep. I have argued using purely historical evidence in an article that appeared a couple of years ago in the Oxford historical journal Past & Present that in fact there’s strong evidence to believe that, not in all cases certainly but that in many cases middle of the night insomnia, judged from a historical perspective, might actually be more natural than the compressed, consolidated sleep to which we aspire but notably are not always successful in achieving. Let me just cite one example. Before 1900, which really represents the tail end of this biphasic pattern of sleep, transformed for both cultural and technological reasons, technological reasons being artificial illumination primarily, over the course of the 19th century and the Industrial Revolution. Only then does waking up in the middle of the night for no explicable reason, only then is that regarded as abnormal.

It becomes conflated with sleep onset insomnia and is referred more generally to insomnia increasingly in the early 20th century but before the late 19th, people regarded this as utterly normal, whereas they did not consider normal sleep onset insomnia for which they prescribed medical literature, prescribed from the 16th century on up to the 19th century any number of potions and prayers by which to fall asleep when you first went to bed and I would just reinforce, underscore Siobhan’s, Professor Banks’ point that if people recognize that waking up in the middle of the night is not necessarily abnormal, that they are not freaks, a word that is commonly used in people who have emailed me, their anxiety is lessened and according to any number of patients and psychiatrists and sleep scientists to whom I’ve spoken, a fair number of them are able to return to sleep more readily as a result of this knowledge of segmented slumber.

RR: It’s so interesting that, as Professor Banks mentioned, there is a lot of anxiety associated today with waking up in the middle of the night when that, in pre-modern times, was a pleasant thing. It was associated with a lot of fraternal bonding with your bed partner, doing things in the middle of the night, of course other activities that we can kind of imagine people probably did in the middle of the night. So it wasn’t an anxiety-ridden thing prior to, as you were saying, around the 19th century.

Ekirch: Precisely. Medical literature before the Industrial Revolution referred to the interval between first and second sleep generally for three reasons, none of which were phrased in a maligned or abnormal fashion. First of all, it was a good time to take certain types of medicine, thereby anticipating the science of chronopharmacology today, taking medicine for anything from sour stomachs to smallpox. Secondly, it was thought an ideal time to turn from one side of your body to the other side to aid your digestion given the inaccurate notions that people had three, four hundred years ago of how in fact food was digested in the stomach. Third and finally, more than a few physicians endorsed this interval of wakefulness as a prime time to conceive children. As a French physician in the 16th century, Laurent Joubert, put it and I quote virtually verbatim, they enjoy it more and do it better because they’re rested and presumably their fertility was heightened, though again I emphasize I’m not a physician but that would seem to be a plausible inference.

RR: Professor Banks, do we know anything about hormone production in the middle of the night and how that might differ in people who sleep in two phases and are awake in the middle of the night to experience some of these nighttime hormones while awake versus people that sleep all in one go? I know there’s some anecdotal ideas that maybe creative people like to get up and work in the middle of the night because they’ve got some sort of magical juice flowing that could be a hormonal component. I don’t know if there’s any truth to that or not.

Banks: Well not that I think that’s been particularly quantified. Certainly we do know that there are a number of important hormones that are released during deep sleep and we tend to get a bulk of our deep sleep, our slow-wave sleep, EEG in the first half of the night. We tend to have more dreaming sleep and lighter sleep in the second half of the night. We sort of cycle in and out of these different sleep stages and we have them through the entire night, both deep sleep and dreaming sleep but we tend to have more deep sleep in the first half of the night and more dreaming sleep in the second, so there are slightly different rhythms to the hormones based on the structure of your sleep. So growth hormone, for example, is released during deep sleep and we see a big spike in it in the early part of the night. There are other hormones that are suppressed during sleep and then rise to help us wake up in the morning like cortisol and so there’s a whole orchestra of things that are going on during sleep biologically within the body and we assume that this is part of the restoration and rest process, the sort of recovery and in the brain, the forming of memories and the consolidation of what we’ve learnt through the day.

So there are certain processes that seem to need to happen during sleep and certainly when we sleep deprive somebody of their sleep below four hours, so if we were to take somebody and only let them sleep 4 hours or give them a 4-hour sleep opportunity, we see all kinds of problems. So we know that there is at least a core amount of sleep that is essential and in the sense that we know we need more sleep than four hours to keep our cognitive performance going, our health and a whole range of proper metabolic functions. Now whether we need to have an eight hour period of sleep to maintain all those metabolic functions, and that’s due to hormone regulation at night. All we really know is when we truncate sleep, we disrupt all of those regular metabolic functions. What we seem to see though is if you are able to maintain a regular amount of sleep over the whole 24-hour period, so as we were talking before about siestas, so longish naps in the afternoon, if we have a slightly shorter main sleep at night but make that up in the afternoon, or indeed as we did with our research and split the sleep opportunity into two 4- or two 5-hour sleep opportunities, what we actually see is that cognitive performance is maintained and people still feel quite good.

So we seem to be able to manipulate sleep quite a lot as long as the overall amount of sleep that you’re getting in a 24-hour period is a good amount, so there’s sort of 7 or 8 hours and as long as you don’t restrict any one sleep period, the main sleep period down below 4 hours then you’re okay. So you’ve got to make sure that you’re, within a 24-hour period, getting that good amount of sleep, so that 7 or 8 hours and that you’re not reducing your sleep down to much smaller amounts and this is where we are always a little skeptical of people who say they can do one of those power sleep type regimes where they’re only sleeping 20 minutes in a 2-hour period and try to maintain their sleep over a long period of time. We know that that’s just impossible, that you do need at least a minimum amount of sleep initially.

So what I’m imagining was happening in older times was people were going to sleep, having that initial 4-hour sleep, waking up, doing some things and then when they fell back to sleep, engaging perhaps in more dreaming sleep and lighter stages of sleep, so I think it flowed naturally with the biology and I think people were allowing more time in general for sleep so that there wasn’t the anxiety about waking up. I think this is where we’ve become very stressed about our sleep, ironically, because that makes us sleep less, is that if we allow a decent period of time then waking up for an hour or so in the middle of it is not a bad thing. But of course, as we were saying before, we squeeze sleep down into the smallest portion possible and then there is great stress associated with being awake and potentially not achieving that seven hours of sleep a night.

biphasic sleep

Siobhan Banks, PhD

RR: And people today when they wake up in the middle of the night, now we have the option of looking at our smartphone or watching TV or turning on a light. Is that something that … We want to encourage people not to feel anxious about simply waking up in the middle of the night but is that something that should be avoided?

Banks: Yes. Yeah, very disruptive for your sleep. So light affects a normal … A hormone called melatonin and melatonin is very important for our sleep, getting off to sleep and our regular sleep patterns through the night and as soon as you’re exposed to light, it suppresses that hormone and so it makes it much more difficult to get back off to sleep because this hormone helps you get to sleep. So it’s very, very important that if you do wake in the night and we do recommend for people that are very anxious in bed and then they wake up and they toss and turn, what we would recommend is that perhaps they get out of bed but don’t expose themselves to very bright light.

But you can go and sit in another room but have a very dimly lit environment and I’m sure this is also what Roger would agree many, many years ago, there wasn’t the availability of these bright lights. If anything, it was a very dim candle. So when people did engage in doing things in the middle of the night, they weren’t exposing themselves to this quite damaging bright light and also the very arousing effect of being connected to social media which is pretty much what people tend to go on if they wake up in the middle of the night, or television, so these things can have very psychologically arousing effects which are not good for sleep either.

Ekirch: Absolutely. There are teenagers, one of whom was our eldest daughter several years ago, who would sleep with her smartphone and she and her friends would text during the night as if they didn’t want to miss anything no matter how trivial the information. If I could just add a further irony to what Siobhan said, here I’m really not drawing on history, just my own observation and that is that on the one hand we, as Siobhan has said, we treat sleep very badly. It’s the one period of time over the course of 24 hours that we think we can cheat on and get away with.

The irony for me is that if we then reduce our sleep, let’s say to 5, 6, or even fewer hours, we expect it to be absolutely perfect so that we invest in…some people at least, not myself I assure you, in $3,000 mattresses. We run to the medicine cabinet for pharmaceutical products and don’t get me wrong, I have nothing against taking pills but I think it can be overdone. And then to get through the day, we rely upon not just coffee but power drinks, highly caffeinated and, if we’re fortunate given the nature of our employment, perhaps a power nap. But we have only ourselves to blame it seems to me by not respecting sleep in the first place and having these unrealistic expectations then of what sleep should be in the course of a night.

RR: Professor Ekirch, was it the invention of the electric light that changed everything or were there other factors too?

Ekirch: There were other factors and not just…I would broaden your point to include gas lighting as well. Artificial illumination that was much more powerful beginning in the early 19th century. A single gas light, it’s been estimated, was 10 times as powerful as a candle. Light from an electric bulb, toward the end of the 19th century, about 100 times more powerful than a single candle and as Siobhan says, light is absolutely critical in resetting our circadian pacemaker. As the eminent sleep scientist Charles Czeisler at Harvard put it, every time we turn on a light switch, it’s as if we’re taking a dose of medicine that can potentially affect our sleep. But having said that, though the most important factor in my opinion born largely of the Industrial Revolution that transformed this older biphasic pattern into the consolidated sleep to which we aspire, also important were cultural influences, again, born of the Industrial Revolution.

Efficacy, practicality, materialism, cutting back on sleep, taking a single sleep rather than two sleeps, all of these became increasingly important over the course of the 1800s. People became increasingly time-conscious so that they were much more open to sleeping in one consolidated segment if at all possible. And then once, by the early 20th century, that was increasingly regarded, at least in urban areas, it took a little bit longer in the countryside, at least in urban areas, not just cities but towns, is once that was regarded as being the norm, there was no looking back as people raced at that time in history to embrace modernity. Since the early 20th century, knowledge of this older pattern of biphasic sleep has virtually been lost in our collective memory.

RR: Professor Banks, You wrote “To successfully maintain a split sleep schedule, you have to get the timing right. That is, commencing sleep when there is a strong drive for sleep and during a low circadian point in order to fall asleep quickly and maintain sleep.” I think you’ve really hit the nail on the head there for a lot of people who struggle with sleep. Do you think that our society is too regimented and inflexible for people to get their best sleep if it falls outside the confines of a typical schedule? And then the other side of that coin is do you think there might be any advancement in more flexible work schedules or permissiveness for people to sleep in the way that works for them? Do you see any progress in that arena?

Banks: I’m not sure about progress but certainly I think people are starting to understand that sleeping … One period, one consolidated sleep period isn’t necessarily the only way they can get a good sleep, a good, healthy sleep and certainly in our study, what we wanted to stress was that…And what we found by placing sleep at different times, was that obviously people slept the most and the best if they had a sleep period in the early hours of the morning and one in the afternoon and so really what you’re taking advantage of in the afternoon is that siesta time, that after lunch period where people feel quite sleepy and in the time in the early hours of the morning, the 12 am to 6 am period which is our absolute biological low when we’re biologically programmed to sleep.

And so if we are to take advantage…if we are to restructure our sleep then you would want to take advantage of those periods in your day and I think really the idea of being able to have a nap is just one that is the simplest form of slightly restructuring your sleep and we all know that glorious feeling of having the time to be able to lay down and have a half an hour or an hour on the couch to have a little nap and I think just, even though sometimes people describe feeling a bit groggy when they wake up, that’s really the body just saying that it needs more sleep really. I think that is just more of a reason then why to take advantage of any period of time that you have for sleep to catch up. Certainly there are difficulties always with sleeping in. It can cause circadian misalignment.

So for example, the person who stays up late on a Friday and Saturday night and sleeps in will find it very difficult come Monday morning to get up and go to work, so we always advocate that you try and go to bed at the same times every day but if you feel like you need to catch up and have more, have a little nap because it can certainly make a big difference and I think that’s part of that feeling of that it’s okay, having that permission. We certainly live in a society where it’s seen as lazy to get those extra hours of sleep and indeed it’s some bizarre kind of prize, somehow you’ve won a prize if you’re able to sleep the least and work the most and get by on caffeine.

So I think we need a reversal of that and have that kind of appreciation for sleep, make time for it, make sure that you are going to bed at a good time every night so that you’re allowing yourself, if you do happen to wake up, to be able to get back to sleep and certainly we know that there are cases where you’re disrupted for all sorts of reasons in the night but as long as you’re allowing a good enough time for sleep then there is often the opportunity to be able to go back to sleep and get that overall good amount of sleep.

RR: I know you’ve written about implications here for shift workers. Can you talk a little bit about that?

Banks: Yeah, I suppose that we initially were inspired to conduct some of this research about looking at splitting sleep opportunities and having … Placing sleep at different times of the day because we know there are various different industries that require workers to be on shift at all different times. There are some of the more traditional shift work industries like you might imagine where workers work through the night and then have to sleep during the day, or in some industries it’s rotating, so they might have to do a morning shift then an afternoon shift and then a night shift and then they have a period of time off. But there are other industries or other sectors where being able to keep going with shorter amount of sleep opportunities is really important.

So for example, in emergency services where, for example, if you have to fight or be involved in some kind of … Fight a fire or be involved in some kind of disaster relief then you will need to be on-site or with … Your work is for a continuous period of time, for several days and so we were interested in looking at what if we split sleep and so shortened the day effectively and gave people shorter periods on shift but also shorter sleeps and so enabling them to do some work and then get back and be able to sleep and then get back to work reasonably quickly. And we found that as long as the amount of sleep, the time available for sleep was adequate over the 24-hour period, so as I was saying before, at least a 4- or 5-hour opportunity at each time for sleep, then people were able to maintain their performance and their well-being.

But this all works under the proviso that the working and sleeping environment are close together and so this is also … This kind of work practice is often seen on oil rigs or other kind of facilities where the sleeping environment is only a couple of steps away from the working environment. If you add commute time into any of these kinds of scenarios, it doesn’t work but I think what the interesting piece to take from all of this is that we can structure sleep in different ways, we can use nap opportunities in the afternoon to supplement a nighttime sleep and this can maintain our cognitive performance and our safety and so that there are … I would advocate that there are … We should look at alternate ways to structure our sleep and work practices for optimal benefit.

RR: My final question is for both of you. This is a topic that really interests people. Why do you think people are so fascinated with how our ancestors slept?

Ekirch: I think they’re interested to begin with in sleep generally and especially as they’ve become increasingly anxious about the quality of their sleep. As to how their ancestors slept, I’m perhaps not the best person to answer this question but I would venture the hypothesis that it’s come as quite a surprise as it did to me that there was this very different pattern of sleep and that contrary to perhaps nostalgic notions of how prior generations slept in … Without artificial illumination, noise, that this older sleep was perhaps in some way idyllic.

In point of fact, nothing could be further from the truth, not because people woke up in the middle of the night but because their sleep was frequently disrupted by illness in this pre-penicillin, pre-analgesic age, disrupted by the trinity of early modern entomology, lice, bedbugs and fleas, urban areas but also rural villages could be very, very noisy and during the era from the Late Middle Ages to the 19th century, never was there a period since the birth of Christ in which nighttime was more widely feared for both real and imaginary reasons, one of the greatest perils being fire.

So there were all sorts of explicable reasons for people of those earlier ages to have … There were all sorts of reasons, again, both real and a product of … There were all sorts of reasons both real and imaginary for people to have their sleep disrupted apart from the very natural period of wakefulness that occurred between first and second sleep. I think all of that combined with sleep having been biphasic comes as quite a surprise to people today.

Banks: Yeah, I agree. I think that there’s … We have such a technologically driven lifestyle now that’s ever-changing that we tend to look back in the hope that we might find something that is more natural and more basic and think that there’s going to be a solution to today’s ills by turning back in time and thinking that what we experienced many years ago was more true biologically or more true to our natural self but I’m absolutely in agreement that there’s many, many benefits about living today, purely from comfort but also from health and so I think if we can embrace both pieces, both see that what we have today is a much more healthy lifestyle and we’re not disrupted in our sleep by all those other things and that potentially we can fix small problems and get a better sleep, it’s just really relaxing, I think, relaxing about not being so anxious about the idea of getting one consolidated sleep and that if we can feel that it’s okay if every now and again we lose a little bit of sleep and that you’ll try and make it up on another night, indeed just going to bed a little bit earlier sometimes can mean a world of difference. So I think it’s just about making … Giving sleep a priority, making sure it’s an important piece of your life and not squeezing it into the back corner somewhere and expecting then to keep feeling good.

Ekirch: Exactly. I’m sometimes asked whether it’s possible or desirable to revert to this older pattern of segmented sleep and I think there is no turning back unless you want to go live in a cabin in the Canadian Yukon devoid of electricity or any form of artificial illumination. The important thing, as Siobhan has said, is that waking up in the middle of the night does not mean that you suffer from chronic insomnia necessarily. It’s not to say that but it’s very important that when people do wake up in the middle of the night that they reassure themselves that they need not be victims necessarily of chronic insomnia and that in fact there are many other people even today who share a common experience and that this experience was the dominant form of sleep for literally thousands of years and that the consolidated, compressed sleep to which we aspire is indeed very youthful, very young. It’s, in the Western world, no older than two centuries. So in my view, it’s perfectly natural that it would take longer than a couple hundred years for this older pattern of sleep to have fully run its course.

RR: Well, thank you so much to both of you, we really appreciate you coming on Sleep Review Conversations.

Ekirch: Thank you.

Banks: It’s been fantastic, thank you.

Ekirch: I’ve learned a lot listening to Siobhan.

Banks: And I’ve thoroughly enjoyed listening to you, Roger.

Ekirch: I’m very grateful. I didn’t know that four hours was … Not that I try to cut my sleep that close but I will definitely make certain that I’m above the four hour mark from here on.

RR: Thanks for listening to this episode of “Sleep Review Conversations.” Visit and click on “Resources” for a transcript of this audio file as well as links to Prof Ekirch’s book and Prof Banks’ article in The Conversation.