(no subject)
So, I've started a sleep journal: http://capn002.livejournal.com/82689.html Based on my reading of Ekirch (A. Roger Ekirch, 2005, At Day's Close: Night In Times Past, ISBN 0-393-05089-0), I'm trying a no-artificial light / bimodal sleep. So far, not quite the pre-industrial polyphasic trip I was expecting, but fun regardless. The time between sleeps is known as dorveille, aernemergen(?), the Watch, etc. I spent last dorveille reading Bede's De Temporibus (trans. The Reckoning of Time by Faith Wallis). Bede divided night into seven units, as seen here:
Ælfric provided their Anglo-Saxon counterparts and I've provided their rough QE-English* translations:
ǽfen-glóma :: crepusculum :: evening twilight
ǽfen :: vesperum :: evening
swí(g-)tíma :: conticinium :: the time of silence
midniht :: intempestum :: mid-night
han-créd :: gallicinium :: cock's crow
dæg-réd :: matutīnum :: daybreak
ǽrnemergen :: diluculum :: early morning
Now, if Ekirch is right, then anyone born after 1700 or so wouldn't have the cultural knowledge to have translated the middle chunk of these correctly. However, Thomas Wright in Biographia Britannica Literaria (Thomas Wright, 1842) mentions (p.86-87) "some scholar of the 10th century [in MS Cotton, Titus D xxvii]" gives us the following translation, which, while not mentioning "second sleep" directly, certainly:
Assuming Bede and Ælfric knew what they was talking about *and* that they had a natural bimodal sleep pattern, what can we make of this? If we take a hypothetical day where sunset is at 6pm and sunrise is at 6am (a 12 hour night), then we can figure that (very roughly; based on my own bimodal sleep)...
(1) ~18:00-19:00 = ǽfen-glóma :: crepusculum :: evening twilight // I would suggest that "twilight" is misleading here and that this referes specifically to the beginning of twilight, just after sunset, when the sky is still ruddy. This, then, would correspond to the gloaming, a common enough term for the same; the AngloSaxon, Latin, and modern forms would then all seem to reference the same period.
(2) ~19:00-21:00 = ǽfen :: vesperum :: evening // The translation to evening seems fine... so long as we recognize that this is the part of twilight when everything turns blue (Fr. l'heure bleu). A better understanding may be dusk. Alternatively, this may refer to first sleep (see below).
(3) ~21:00-00:00 = swí(g-)tíma :: conticinium :: the time of silence // This is where our modern translations really begin to fail. The Latin conticinium comes down to Spanish as conticinio which is generally translated as "dead of night". It seems, then, that this would refer to the first sleep of a bimodal pattern. Alternatively, however, as a "time of silence" is could be that this was the first waking and my previous comments on the "ǽfen" times is wrong.
(4) 00:00-03:00 = midniht :: intempestum :: mid-night // Although I'm estimating it at beginning around 00:00 ("midnight"), for me it begins closer to 1 or 2am (and my first sleep begins closer to 21:00). This is almost certainly first waking. In this reading, then, the folklore of "midnight" to do "witches' work" makes more sense. If the first waking was a hour of quiet reflection, prayer, sex, visiting neighbors, etc. then of course this would be when a coven gathers to work spells (often tied to an inversion of prayer, sex, and small gatherings of people). The problem, though, is that Wallis gives this time as "Dead of night... when, in the deep sleep of peace, there is not time for activity" (quoting someone, I think, but still). Based on this reading, is seems that midniht is really the second sleep, when there is no time, no activity, and fewer dreams (per Erkich).
(5) ~03:00 = han-créd :: gallicinium :: cock's crow // This is where our trouble really begins. There's simply no way around the clash between the folklore of cock's crowing at dawn, the fact that this is only the fifth part of night, and the apparent timing.
The 3am timing is supported by a discussion of Ælfric in the (1879) article "Yorkshire Dials" by Rev. Daniel Henry Haigh, as excerpted below:
If the timing and my conjecture about the previous parts is correct, then this is second sleep. If so, we've misunderstood the folklore of the rooster's crow... it apparently doesn't signal a time to wake up so much as it signals a time to return to bed (which isn't wholly impossible... if the crowing signals a time, and the time is synechdoche for an event, and modern electricity flips that event on its head... you see where I'm going...).
The second option, however, is that this is final waking. This reading goes along with the folklore, and fits well enough with the division of night. The *massive* problem here is that this reading requires us, as a species, to end our day around 7pm and begin it around 3am. That doesn't seem right.
The third option, that this is first waking is clearly impossible. There's not enough time to wake up at 3am, stay awake for an hour or so, go back to sleep, then wake up again with the rising sun. Also, if that were the case, we'd expect Bede and Ælfric to have put this more toward the middle when dividing the night.
(6) ~04:00(?) = dæg-réd :: matutīnum :: daybreak // It seems that we can agree that this was the time of first light; this was probably a very short period; the beginning of "morning twilight". The translation to "daybreak" or "dawn" are problematic. First, see the diluculum probelm, below. Also, in the Catholic Church, matins traditionally ended at dawn, so is may make more sense if this is the period just before dawn. Maybe around 4am?
(7) ~05:00-06:30 = ǽrnemergen :: diluculum :: early morning // True "morning twilight", just before the sun has risen over the horizon. The Latin diluculum is also given the translation of "dawn" or "daybreak". Although final waking could have occured at dæg-réd, because "ǽrnemergen" is listed as the last hour of the night rather than as the beginning of day, it makes more sense to assume that it occured here. We know actual sunrise hasn't occured yet because Ælfric tells us that "ǽrnemergen" is followed by sunnan upgang (sunrise).
Alternatively, it could be that "sunnan upgang" refers to the sun coming fully into the sky; this would make "ǽrnemergen" the time when the sun is in the process of rising from sliver to full disk, "dæg-réd" the time of first true daybreak, and "han-créd" the morning twilight. This would jibe more easily with the notion of a rooster crowing when day begins, though, again, it's difficult to reconcile this with a 3am hour.... Though, if this is true, then "midniht" becomes the second sleep and the folklore of witches making havoc while everyone else is asleep is possible. (Though, as per Ekirch, how a time when only witches were awake would become a time when parents are encouraged to keep children awake doesn't make sense. And, again, the 3am part is hard to reconcile.)
SO... what can our new appreciation of bimodal sleep tell us? First, someone might want to double check the time of the 734AD lunar eclipse. If it occured at 4:45 rather than 2:45, then all the red readings are suddenly just fine (and, for the record, I can't find it anywhere in NASA's lunar eclipse database!). Second, we are on the way toward a deeper appreciation of the folklore of midnight: this was either the time when we all first woke up, or the time when we all went back to sleep. Either way, it changes our view of how this became a time for spells and whatnot.
So, yeah. I'm trying it. Polyphasic, natural, bimodal sleep. We'll see how it goes.
-dsb
*QE-English: my term for the English used from the reign of Queen Elizabeth I to Queen Elizabeth II. This is more satisfying than "Modern English" as "modern" is itself relative. It's also a nod to Stephenson's The Diamond Age.
Ælfric provided their Anglo-Saxon counterparts and I've provided their rough QE-English* translations:
ǽfen-glóma :: crepusculum :: evening twilight
ǽfen :: vesperum :: evening
swí(g-)tíma :: conticinium :: the time of silence
midniht :: intempestum :: mid-night
han-créd :: gallicinium :: cock's crow
dæg-réd :: matutīnum :: daybreak
ǽrnemergen :: diluculum :: early morning
Now, if Ekirch is right, then anyone born after 1700 or so wouldn't have the cultural knowledge to have translated the middle chunk of these correctly. However, Thomas Wright in Biographia Britannica Literaria (Thomas Wright, 1842) mentions (p.86-87) "some scholar of the 10th century [in MS Cotton, Titus D xxvii]" gives us the following translation, which, while not mentioning "second sleep" directly, certainly:
Assuming Bede and Ælfric knew what they was talking about *and* that they had a natural bimodal sleep pattern, what can we make of this? If we take a hypothetical day where sunset is at 6pm and sunrise is at 6am (a 12 hour night), then we can figure that (very roughly; based on my own bimodal sleep)...
(1) ~18:00-19:00 = ǽfen-glóma :: crepusculum :: evening twilight // I would suggest that "twilight" is misleading here and that this referes specifically to the beginning of twilight, just after sunset, when the sky is still ruddy. This, then, would correspond to the gloaming, a common enough term for the same; the AngloSaxon, Latin, and modern forms would then all seem to reference the same period.
(2) ~19:00-21:00 = ǽfen :: vesperum :: evening // The translation to evening seems fine... so long as we recognize that this is the part of twilight when everything turns blue (Fr. l'heure bleu). A better understanding may be dusk. Alternatively, this may refer to first sleep (see below).
(3) ~21:00-00:00 = swí(g-)tíma :: conticinium :: the time of silence // This is where our modern translations really begin to fail. The Latin conticinium comes down to Spanish as conticinio which is generally translated as "dead of night". It seems, then, that this would refer to the first sleep of a bimodal pattern. Alternatively, however, as a "time of silence" is could be that this was the first waking and my previous comments on the "ǽfen" times is wrong.
(4) 00:00-03:00 = midniht :: intempestum :: mid-night // Although I'm estimating it at beginning around 00:00 ("midnight"), for me it begins closer to 1 or 2am (and my first sleep begins closer to 21:00). This is almost certainly first waking. In this reading, then, the folklore of "midnight" to do "witches' work" makes more sense. If the first waking was a hour of quiet reflection, prayer, sex, visiting neighbors, etc. then of course this would be when a coven gathers to work spells (often tied to an inversion of prayer, sex, and small gatherings of people). The problem, though, is that Wallis gives this time as "Dead of night... when, in the deep sleep of peace, there is not time for activity" (quoting someone, I think, but still). Based on this reading, is seems that midniht is really the second sleep, when there is no time, no activity, and fewer dreams (per Erkich).
(5) ~03:00 = han-créd :: gallicinium :: cock's crow // This is where our trouble really begins. There's simply no way around the clash between the folklore of cock's crowing at dawn, the fact that this is only the fifth part of night, and the apparent timing.
The 3am timing is supported by a discussion of Ælfric in the (1879) article "Yorkshire Dials" by Rev. Daniel Henry Haigh, as excerpted below:
If the timing and my conjecture about the previous parts is correct, then this is second sleep. If so, we've misunderstood the folklore of the rooster's crow... it apparently doesn't signal a time to wake up so much as it signals a time to return to bed (which isn't wholly impossible... if the crowing signals a time, and the time is synechdoche for an event, and modern electricity flips that event on its head... you see where I'm going...).
The second option, however, is that this is final waking. This reading goes along with the folklore, and fits well enough with the division of night. The *massive* problem here is that this reading requires us, as a species, to end our day around 7pm and begin it around 3am. That doesn't seem right.
The third option, that this is first waking is clearly impossible. There's not enough time to wake up at 3am, stay awake for an hour or so, go back to sleep, then wake up again with the rising sun. Also, if that were the case, we'd expect Bede and Ælfric to have put this more toward the middle when dividing the night.
(6) ~04:00(?) = dæg-réd :: matutīnum :: daybreak // It seems that we can agree that this was the time of first light; this was probably a very short period; the beginning of "morning twilight". The translation to "daybreak" or "dawn" are problematic. First, see the diluculum probelm, below. Also, in the Catholic Church, matins traditionally ended at dawn, so is may make more sense if this is the period just before dawn. Maybe around 4am?
(7) ~05:00-06:30 = ǽrnemergen :: diluculum :: early morning // True "morning twilight", just before the sun has risen over the horizon. The Latin diluculum is also given the translation of "dawn" or "daybreak". Although final waking could have occured at dæg-réd, because "ǽrnemergen" is listed as the last hour of the night rather than as the beginning of day, it makes more sense to assume that it occured here. We know actual sunrise hasn't occured yet because Ælfric tells us that "ǽrnemergen" is followed by sunnan upgang (sunrise).
Alternatively, it could be that "sunnan upgang" refers to the sun coming fully into the sky; this would make "ǽrnemergen" the time when the sun is in the process of rising from sliver to full disk, "dæg-réd" the time of first true daybreak, and "han-créd" the morning twilight. This would jibe more easily with the notion of a rooster crowing when day begins, though, again, it's difficult to reconcile this with a 3am hour.... Though, if this is true, then "midniht" becomes the second sleep and the folklore of witches making havoc while everyone else is asleep is possible. (Though, as per Ekirch, how a time when only witches were awake would become a time when parents are encouraged to keep children awake doesn't make sense. And, again, the 3am part is hard to reconcile.)
SO... what can our new appreciation of bimodal sleep tell us? First, someone might want to double check the time of the 734AD lunar eclipse. If it occured at 4:45 rather than 2:45, then all the red readings are suddenly just fine (and, for the record, I can't find it anywhere in NASA's lunar eclipse database!). Second, we are on the way toward a deeper appreciation of the folklore of midnight: this was either the time when we all first woke up, or the time when we all went back to sleep. Either way, it changes our view of how this became a time for spells and whatnot.
So, yeah. I'm trying it. Polyphasic, natural, bimodal sleep. We'll see how it goes.
-dsb
*QE-English: my term for the English used from the reign of Queen Elizabeth I to Queen Elizabeth II. This is more satisfying than "Modern English" as "modern" is itself relative. It's also a nod to Stephenson's The Diamond Age.