The Old Icelandic Misseri Calendar
The Old Icelandic calendar is the best-documented calendar in any Germanic language and the one to lean on hardest, because it is concrete, computable, and described by the society that kept it.
The structure
Two seasons. The year divides into two misseri, half-years: sumar and vetr, summer and winter. There is no four-season year in the scheme 1.
Fifty-two weeks, which is 364 days. The year is built out of whole weeks, which is close to unique among world calendars and which is what makes the whole system tractable. Summer begins on a Thursday and winter on a Saturday, every year, without exception, which is what a week-based year buys 2.
A leap week. Because 364 falls short of the tropical year by more than a day, the calendar drifts, and the correction is a whole extra week, sumarauki, summer-increase, inserted into the summer half and producing a year of 371 days. The rule for when to insert it is the interesting part, and it is stated in the sources 2.
A dated reform. The introduction of the correction is credited to Þorsteinn surtr, at an assembly, in an account written by an Icelander within about two centuries of the event 3. A further adjustment relating the reckoning to the Julian year followed around 1200 2.
In the Julian reckoning of the medieval texts, summer began on a Thursday between 9 and 15 April and winter on a Saturday between 11 and 18 October. The first day of summer is still a public holiday in Iceland and still moves according to the same rule.
Why the two-season year matters
The same shape appears three times, in three languages, across six centuries and two thousand kilometres.
Tacitus, writing about peoples beyond the Rhine at the end of the first century, reports that they had names for winter, spring and summer and no word for autumn 4 ch. 26. Bede, writing in eighth-century Northumbria, says the English year had two seasons, and names a month Winterfilleth after the full moon that opened winter 5 ch. 15. And the Icelandic law texts give a year of two misseri outright.
Three independent witnesses is a great deal by the standards of this material. It does not establish a single Germanic calendar, and this site does not claim one: a shared two-part division of the year is a structural similarity only. What it does establish is that a two-season year was a real and durable way of dividing time in the Germanic-speaking north, which is a more useful conclusion than a reconstructed festival grid.
Correction
Commonly saidThe misseri calendar is the pre-Christian Norse calendar, so its dates give the authentic dates of the Norse festivals.
Two problems, and the second is the more interesting.
The texts are Christian-era. The law compilation and the computistical material are thirteenth-century Icelandic manuscripts, written centuries after the conversion, by people who also knew the Church’s own reckoning 1. The structure is very unlikely to have been invented by them and is plainly older; how much older cannot be established from the manuscripts themselves.
And the seven-day week, which the whole system is built on, arrived in the north with Christianity. A week-based year is therefore not straightforwardly a pre-Christian object, whatever the antiquity of its two-season shape. That is a real complication.
What stands in its place is better than what is withdrawn: a real calendar, with real rules, that can be computed exactly for any date, and whose reform has a name and an assembly attached to it. Very little else in the Germanic material can be run as an algorithm. A festival grid with invented dates would be worth less than this.
Computing it
The rules are complete enough to convert any Gregorian date into a misseri date: the season, the week within the season, the weekday, and whether the year carries a sumarauki. The standard technical treatment sets out the rules in a form that can be implemented directly and is freely available from the author’s repository 2.
That makes this the one indigenous northern European calendar this site can offer as a working converter without inventing anything, which is exactly what cannot be done for the Gaulish material: see The Coligny Calendar. For the Old English month names that share the two-season structure, see Bede and the Old English Months; for the midwinter season, see Yule.
Gap in the evidence
Whether the year began at the start of summer or the start of winter is argued both ways from the same texts, and the answer probably depends on what a beginning is taken to be in a two-season scheme where both halves open with a fixed weekday.
How the reckoning worked before the correction credited to Þorsteinn surtr is not described anywhere. A 364-day year drifts by more than a day annually, so something must have been done, and the sources record the fix rather than the problem it solved.
References
- Svante Janson. The Icelandic calendar. Scripta Islandica: Isländska Sällskapets Årsbok. Uppsala, 2011, 51-104. https://uu.diva-portal.org/smash/get/diva2:436604/FULLTEXT01.pdf recordpeer reviewed
Volume 62 (2011); ISSN 0582-3234. The rigorous open-access mathematical treatment of the misseri calendar, including the rule for the leap week and the correction relating the reckoning to the Julian year. Page range confirmed against the volume's own table of contents.
- trans. Andrew Dennis, Peter Foote, Richard Perkins. Laws of Early Iceland: Grágás I. The Codex Regius of Grágás with Material from Other Manuscripts. University of Manitoba Icelandic Studies. Winnipeg: University of Manitoba Press, 1980 (composed 13th c.). ISBN 9780887551154 recordprimary source
The Icelandic law compilation that describes the misseri reckoning: the two-season year, the 52-week year, and the rule for inserting the leap week. A second volume followed in 2000; the series volume number for that second volume was not confirmed and is therefore not given here.
- trans. Siân Grønlie. Íslendingabók, Kristni saga: The Book of the Icelanders, The Story of the Conversion. Viking Society Text Series XVIII. London: Viking Society for Northern Research, 2006 (composed c. 1122-1133). ISBN 9780903521710 recordprimary source
Ari Þorgilsson's account of early Iceland, including the calendar reform credited to Þorsteinn surtr at an Alþingi. Free from the publisher; the https form of the address fails, so the http form is given.
- Tacitus, trans. J. B. Rives. Germania. Clarendon Ancient History Series. Oxford: Clarendon Press, 1999 (composed c. 98 CE). ch. 26. ISBN 9780198150503 recordprimary source
Translated with introduction and commentary. Germania 11 for assembly at new and full moon and for reckoning by nights; Germania 26 for winter, spring and summer named with no word for autumn. Rives's commentary sets out the work's rhetorical purpose and its sources, which were not first-hand.
- Bede, trans. Faith Wallis. The Reckoning of Time. Translated Texts for Historians 29: Liverpool University Press, 1999 (composed 725). ch. 15. find a copy (catalogue search) recordprimary source
Chapter 15, De mensibus Anglorum, is the sole source for the Old English month names, including Eosturmonath and the doubled Litha.