How did people sleep in the Middle Ages?

It seems normal that people go to sleep for seven to nine hours, straight from evening to morning, but was that always the case? Here is a look at the theory that medieval people slept in two periods during the night.

The theory comes from Roger Ekirch, a historian who specializes in historical sleeping patterns. His book, At Day’s Close: Night in Times Past, argues that until modern times, when artificial lighting allowed us to stay awake longer, most people would go to bed around sunset. The actual time spent sleeping was split into two phases – known as first sleep and second sleep.

A sleeping man surrounded by birds. ,Image taken from L’Estoire del Saint Graal. – British Library MS Add. 10292, f.29

Ekirch explains:

Both phases of sleep lasted roughly the same length of time, with individuals waking sometime after midnight before returning to rest. Not everyone, of course, slept according to the same timetable. The later at night that persons went to bed, the later they stirred after their initial sleep; or, if they retired past midnight, they might not awaken at all until dawn. Thus, it ‘The Squire’s Tale’ in The Canterbury Tales, Canacee slept “soon after evening fell” and subsequently awakened in the early morning following “her first sleep”; in turn, her companions, staying up much later, “lay asleep till it was fully prime” (daylight).

In between the first and second sleep the person would be awake about an hour – enough to say prayers during Matins, which would typically fall between 2 am and 3 am, study or even have sex. The French physician Laurent Joubert (1529-1581) even advised that couples have intercourse during this period, because “they have more enjoyment” and “do it better.”


Ekirch adds:

Although in some descriptions a neighbor’s quarrel or a barking dog woke people prematurely from their initial sleep, the vast weight of surviving evidence indicates that awakening naturally was routine not the consequence of disturbed or fitful slumber. Medical books, in fact, from the fifteenth to the eighteenth centuries frequently advised sleepers, for better digestion and more tranquil repose, to lie on their right side during “the fyrste slepe” and “after the fyrste slepe turne on the lefte side.”

Not everyone slept in two periods – Ekirch cites some people from the pre-industrial period who note that they would sleep throughout the night. But he also found evidence that similar sleep patterns could be seen outside Europe:

The French priest André Thevet, on traveling to Rio de Janeiro, Brazil in 1555, reported that the Tupinamba Indians ate whenever they had an appetite, “even at night after their first sleep they get up to eat and then return to sleep.” In the early 19th century, residents of Muscat, the capital of Oman, were said to retire early, lying “down before 10 o’clock,” so that “before midnight their first sleep” was “usually over.”

In this interview on The Agenda, Ekirch reveals more about the history of sleeping:

Not all historians are convinced by Ekirch’s arguments. A recent re-evaluation by Niall Boyce notes that you can interpret the phrase first and second sleep in different ways. It might be just different phases of the same sleep, and if there was any period of waking, it was very short. For example, another historian told Niall:


I was able to find some matching references in the Middle High German texts I analyzed: There are several records of a ‘first sleep’ (though none of a ‘second’ sleep). However, I used to interpret these passages with regard to the ‘common knowledge’ (which, by the way, still seems to be an issue in our days) that sleep before midnight is deeper and ‘better’ than sleep after midnight. In medieval German literature I found no references to actual periods of wakefulness that would have separated first and second sleep (apart from examples related to religious practice, i.e. if a person rises to celebrate the Liturgy of the Hour).

Meanwhile, Jean Verdon, author of Night in the Middle Ages, notes that some medieval people had different sleeping patterns. Children, for instance, were advised to sleep the entire night, for nine or ten consecutive hours. However, for the very young, this task might be tricky. The fifteenth-century story La Farce du Cuvier, offers this verse on the troubles of getting one’s child to sleep – something that every parent nowadays can relate to:

At night, if the child awakes
As they do in many places,
You must take the trouble
To get up to rock him,
To walk, carry, and feed him
In the bedroom, even at midnight.


Medieval monks were also required to sleep differently – according to the Rule of St. Benedict, they would go to bed at about 7:00 pm, and then wake up for Matins around 2:00 in the morning. While other monastic rules allowed for a second sleep, the Benedictine monks would continue to stay awake (they might be allowed to have a nap during the day). Some monks were tempted not to get out of bed – Raoul Glaber, who lived during the 11th century, wrote that he was plagued by a ‘demon’, who whispered to him:

I wonder why you are so eager to jump so quickly out of bed, as soon as you’ve heard the signal, and to interrupt the sweet rest of sleep, while you could give yourself up to rest until the third signal.

Verdon adds that medieval people could have the same problems related to sleeping that we do, including insomnia, sleeping too much, and even sleepwalking. The chronicler Jean Froissart heard the story of a noble named Pierre de Béarn who had a traumatic experience when he killed an exceptionally large bear in hand-to-hand combat. Afterwards, during his sleep he would rise, grab a sword and swing it around in the air. If he could not find his weapon, Pierre “created such noise and clamor that it seemed like all the demons of hell were there with him.” Eventually, his wife and children would leave him over the problem.

See also Sleepwalking and Murder in the Middle Ages

See also: The Medieval Sleeping Beauty

Top Image: A sleeping man in a medieval manuscript – from British Library Royal 19 D III f. 458