Friday 9 April 2021

Easter, Yule, and the old English calendar

‘Easter’ and ‘Yule’ started out in English as month-names in the early mediaeval English calendar. They’ve taken on a life of their own, of course, and have become identified with Christian festivals: I’ve talked about Easter and Yule as festivals previously, looking at how much modern customs are linked to anything ancient (they aren’t). Here we’re only looking at the names.

The opening of Bede’s De temporum ratione in Brit. Lib., Royal MS 13 A xi fol. 32v (11th–12th cent.)

We have three main sources for month names in the Old English calendar: Bede’s Reckoning of time, written around 730 CE; the Old English Martyrology, a 9th century calendar of important days throughout the year; and the Menologium, a calendar poem dating perhaps to ca. 1000 CE (Karasawa 2015: 70–72). Bede wrote in Latin, the others are in Old English.

Bede Martyrology Menologium Meaning given by Bede
giuli æftera geola giuli = ‘midwinter’
solmonath solmonað cake month
rhedmonath hredmonað hlyda month of goddess Rheda
eosturmonath eastermonað eastermonað month of goddess Eostre
thrimylchi þrymylce þrymilce (emendation) three milkings
lida ærra liða ærra liða lida = ‘gentle, good for sailing’
lida æftera lyða lida = ‘gentle, good for sailing’
weodmonath weodmonað weodmonað weed month
halegmonath haligmonað haligmonð month of holy rites (sacrifices, acc. to Martyrology)
winterfylleth winterfylleð winterfylleð winter (moon) waxes, full moon that begins the winter half of the year
blodmonath blodmonað blotmonað month of cattle sacrifices
giuli ærra geola ærra iula giuli = ‘midwinter’
Sources in detail. Bede: De temporum ratione §15. Martyrology: entry for 1 Jan., and headings for Mar., Apr., May, June, July, Aug., Sep., Oct., Nov., and Dec. Menologium: lines 16, 37, 72, 78, 108, 138, 184, 195, and 221. For a reliable translation of the Menologium, see Kazutomo Karasawa’s superb edition (2015).

Some of the month names appear piecemeal in other sources. For example Hlyda — apparently unrelated to Bede’s rhedmonath — continued to be used for ‘March’ into the 17th century (OED s.v. ‘Lide’). And in Bede, the Latin spellings vary from manuscript to manuscript. Rhedmonath also appears as hred-, red-, redh-, and reth-; blodmonath can be blot- (just like in the Menologium); and so on. These variants are simply the result of scribes converting Old English to Latin. For simplicity’s sake we’ll just take one set of spellings as standard.

Months, solstices, and equinoxes

Bede tells us the calendar originally began on the winter solstice.

But they began the year from the eighth day before the Kalends of January, when we now celebrate the Lord’s birthday. And that night, which is sacred to us, they would call by the local name of modranicht, that is, ‘night of mothers’ ... and whenever it was a ‘common’ year, they would assign three lunar months to each season of the year.
Note. Mothers’ night, not mother’s night as it’s often mistranslated (including by Wallis 1999). Shaw and others link Bede’s modranecht to the ‘cult of matrons’ attested by ancient votive inscriptions, addressed to matronae, matres, or matrae, found sparsely around England and copiously on the mainland in the area around Cologne. See Shaw 2011: 41–47; 44–45 on modranecht.

In the Roman calendar, the solstice was traditionally reckoned as occurring on 25 December, eight days before the start of the Roman calendar. As a result, months in the two calendars did not originally line up. At the start of each month there would be an overlap. However, all three sources are happy to equate the Old English names with the Roman names Ianuarius, Februarius, and so on.

Note. For the traditional assignment of the solstice to 25 December in the Roman calendar, see Columella, De re rustica 9.14.12; Pliny, Natural history 18.221; and many later sources (including Bede himself at Reckoning of time §30). On the incorrectness of this date and possible sources, see this piece of mine from 2015, about halfway down.

Each solstice had two months surrounding it going by the same name: December and January were ‘Former Yule’ and ‘Latter Yule’ (Ærra Geola and Æftera Geola in the Martyrology), and June and July were ‘Former Liða’ and ‘Latter Liða’ (Ærra Liða, Æftera Liða). In years with an intercalatary month — an embolismus, in Bede’s terminology — there were three months of Liða in a row.

Bede carries on:

Similarly they originally separated the year into two seasons, winter and summer, by assigning the six months with days longer than the nights to summer, and the other six to winter. As a result the month in which the winter period began was called Winterfylleth, the name made up of ‘winter’ and ‘full moon’, since the beginning of winter was marked from the full moon of that month.

The months were lunar, so we cannot assume that all of the solstices and the equinoxes were considered to fall on the first day of a month. Still, it is clear from Bede that the two solstices and one of the equinoxes were linked to particular months: Liða, Geola, and Winterfylleð.

Old English month names in a manuscript of Bede (Brit. Lib., Royal MS 13 A xi fol. 49r, 11th–12th cent.).


‘Yule’, as a name for the season around the winter solstice, is common to several mediaeval Germanic languages in a variety of local forms. The oldest is in a 5th century liturgical calendar in Gothic, as a month name: fruma jiuleis ‘Former Yule’, corresponding in meaning directly to Ærra Geola. Only a single palimpsested leaf survives of the Gothic calendar, but it is the last leaf, so there is no doubt that it, like Ærra Geola in English, was the last month of the year.

Gothic is an East Germanic language, while English is West Germanic, so this could well be a pan-Germanic name. The etymology is uncertain, but some later forms show it had a medial velar fricative: Old English geohhol, geochol, and Finnish (loanword) juhla. Koivulehto (2000) points out that this phoneme suggests an Indo-European origin.

Note. The full Gothic phrase is fruma jiuleis ·l· ‘Former Yule 30’, where ‘30’ could refer to the number of days in the month, or to a regnal year in Theoderic’s reign (i.e. December 522 CE). The palimpsested leaf is bound in a codex in Milan, Biblioteca Ambrosiana MS S 36 sup., known to scholars of Gothic as ‘codex Ambrosianus A’. Gothic jiuleis has often been misread as ‘July’ or something else; for many decades it was reported that the same line also mentioned November, using its Roman name, until Landau (2006) showed that that reading is completely imaginary.

On the etymology: Koivulehto’s point about the velar fricative rules out Landau’s suggestion of a derivation from Greek Ἰωβηλαῖος, the biblical Jubilee.

The Old English sources and the Gothic calendar show that Yule started out as a season, not a festival, religious or otherwise. Evidence of customary celebrations linked to the name ‘Yule’ only starts to appear in 9th century Norse sources. I’ve discussed this more fully elsewhere. The seasonal name began to be used as a periphrasis for ‘Christmas’ in England as early as the 9th century.

Note. ‘Yule’ = ‘Christmas’ in 9th–10th cent. sources: the law-code of King Alfred, §5.5 (‘he who steals on Sunday, or at Yule, or at Easter ...’), §43 (‘twelve days [rest] at Yule, ... and seven days before Easter ...’); the Old English version of Bede’s History, 4.19 (318,17–18 ed. Miller: ‘she would seldom bathe in hot water, except at the highest festivals and seasons, as Easter and Pentecost and the 12th day after Yule’).
The Old English month names as they appear in the manuscript of the Menologium (Brit. Lib. Cotton MS Tiberius B I, fols. 112r–114v).

Easter and the goddess Eostre

My suggestion is this: Eosturmonað the month was named for the equinox, but that has no necessary bearing on Eostre the goddess.

Let’s look at the goddess first.

Bede’s testimony about Eostre the goddess has often been doubted, starting with Karl Weinhold. In the 19th century he called her ‘an invention of Bede’s’, and many scholars have followed suit. (Weinhold 1869: 52; see further Shaw 2011: 50.)

This scepticism arose in the first place because Bede is tangled in a long-running debate over Jacob Grimm’s reconstruction of a pan-Germanic past. When Weinhold described Eostre as an invention, he was arguing against Grimm, not Bede.

Grimm had proposed that Eostre was the English name for a pan-Germanic goddess which he named ‘Ostara’. In south-eastern Old High German, ostarun meant ‘Easter’ (the Christian festival); the corresponding month in Charlemagne’s calendar was called ostarmanoth; and then there’s Eostre in England. On this slender evidence, Grimm invented a goddess. (Grimm 1835: 180–182; English translation; see further Shaw 2011: 51–52.)

And it is slender. There’s no evidence that Old High German ostar- ever had anything to do with anything pre-Christian.

No evidence, except for its etymology. Linguists normally derive ‘Eostre’ from a Proto-Indo-European root *h2eusṓs ‘east, dawn’. ‘Dawn’ goddesses in several other pantheons have names derived from the same root: Vedic Uṣas, Greek Eos, Latin Aurora, and Lithuanian Aušrinė (see West 2007: 217–227). With that Indo-European backdrop, it’s much easier to treat Grimm’s ‘Ostara’ conjecture as feasible.

But that assumes the goddesses are derived from a pan-Indo-European goddess. It’s much more likely that only their names come from a Proto-Indo-European word. The goddesses have nothing in common except their etymology. Uṣas has no similarity to Eos; Eos has no similarity to Aušrinė; we know nothing at all about Eostre and Aurora. Divine names in different pantheons don’t typically have names that are linguistically cognate — Zeus/Jupiter and Hestia/Vesta are exceptions, not the rule.

Shaw makes the compelling point that it’s especially difficult to treat Eostre as a ‘dawn’ goddess when Old English uses *ēast exclusively as an adverb, never as a noun (2011: 57). That is: Old English *ēast- didn’t mean ‘dawn’, it meant ‘easterly, eastward’.

Based on this, and based on a review of East- place names and personal names in Old English and Middle English, Shaw proposes (2011: 49–71) that Eostre was a local goddess, linked especially to Kent. Bede calls her Eostre, not Eastre, because Kentish sometimes used eo- spellings. Outside England, the only links Shaw finds plausible are to the matronae Austriahenae that appear in 2nd-3rd century votive inscriptions found near Cologne.

I find Shaw’s approach compelling, and well grounded in real evidence. I also wholeheartedly support his prioritising localised evidence ahead of Grimm’s reckless speculations about pan-Germanism.

Easter and the ‘eastward’ equinox

Shaw dismisses the idea of Eostre as a ‘spring goddess’ (2011: 55). I agree — in regard to Eostre as a goddess. But Eosturmonað as a month is a different matter.

Remember Bede’s outline of the calendar: two months of Geola surround the winter solstice; the months of Liða are midsummer; and Winterfylleð gets its name from the dividing line between summer and winter, that is, the autumn equinox.

That outline leaves a conspicuous hole at the spring equinox.

The months of the Old English calendar in relation to the solstices and equinoxes. The upper half of the diagram is Bede’s six-month summer season, the lower half is his winter season; Geola and Liða stand at the solstices.

Just to repeat, I’m not suggesting Eostre was a ‘spring goddess’. Only that, if Winterfylleð was effectively named after the autumn equinox, then there’s good reason to suspect that Eosturmonað as a month name had a comparable relationship to the spring equinox.

Shaw’s linguistic evidence shows that eostur- was a directional adverb: it meant ‘eastward’, not ‘dawn’. (Sunrise happens every day anyway, so it’d be strange to think dawn is specific to a time of year.)

But the direction of sunrise is linked to the time of year. If you’re in the northern hemisphere, from April to September the sun rises to the north; from October to March it’s to the south. But at the equinoxes, no matter where in the world you are, sunrise occurs exactly due east.

Shaw’s own linguistic evidence indicates that the most natural meaning of ‘Eosturmonað’ is ‘eastward month’. We know it marked a part of the year when sunrise is exactly eastward. It’s hard to imagine that’s a coincidence — especially when we have Bede telling us that the calendar was framed around the solstices, and that the other equinox has a month named after it.

There’s a cost to this interpretation: the link to the goddess Eostre. If the equinox fully explains the month name, and I think it does, then where does the goddess fit in? Maybe Bede’s right, and Eostre did have a festival in April. Or maybe she didn’t: maybe Bede knew of the goddess, and he knew the month name, and he assumed that one was caused by the other. Shaw has made a very compelling argument in favour of Eostre the goddess, but I think a name like ‘eastward month’ raises real questions about its relationship to the goddess.

The main consideration in favour of an Eostre festival in ‘eastward month’ is that Bede says the previous month, Hredmonað, was also named after a goddess. I won’t pretend to have a full answer for the relationship between the month and the goddess. But the meaning ‘eastward month’, and its position opposite Winterfylleð in the calendar, can’t be a coincidence. I suspect Eosturmonað was named more for the equinox than for the goddess.


  • Grimm, J. 1835. Deutsche Mythologie, 1st edition. Dieterichsche Buchhandlung (Göttingen). [Internet Archive link]
  • Karasawa, K. 2015. The Old English metrical calendar (Menologium). D. S. Brewer (Cambridge).
  • Landau, D. 2006. ‘On the reading and interpretation of the month‐line in the Gothic calendar.’ Transactions of the Philological Socety 104.1: 3-12. [Wiley link]
  • Miller, T. 1890–1898. The Old English version of Bede's Ecclesiastical history of the English people. N. Trübner and Co. (London).
  • Sermon, R. 2008. ‘From Easter to Ostara: the reinvention of a pagan goddess?’ Time and mind 1: 331–343. [Taylor & Francis link]
  • Shaw, P. A. 2011. Pagan goddesses in the early Germanic world. Eostre, Hreda and the cult of matrons. Bristol Classical Press.
  • Wallis, F. 1999. Bede: the Reckoning of time. Liverpool University Press.
  • Weinhold, K. 1869. Die deutsche Monatnamen. Verlag der Buchhandlung des Waisenhauses (Halle). [Internet Archive link]
  • West, M. L. 2007. Indo-European poetry and myth. Oxford University Press.