A claim you often see round about this time of year is that the word Easter comes from the name for a pre-Christian goddess. The aim of this blog post is to weigh up the evidence for this claim.
 |
| Depiction of “Eastre”, by Jacques Reich (1909) |
The oft-repeated story goes something like this:
“The English word Easter corresponds to German Ostern. Both forms derive from the name of the pagan Germanic goddess of spring, Eastre. Her name is probably related to the English word east (German Ost), which connects her to dawn and to the rising of the sun.”
This version’s taken from a recent column by Peter Trudgill, “How the pagans gave us Easter”. And it’s dotted about here and there in popular culture. The video game Assassin’s Creed: Valhalla, for instance, has a quest relating to Ostara, a goddess of fertility; this is another version of the same name (more on this later). She also shows up in the American Gods TV series. But how do we know about this purported “pagan Germanic goddess”?
Bede’s story
It comes down to a single source: Bede’s Reckoning of Time (De Temporum Ratione), written in 725. Here’s what he says:
“Eosturmonath [April—GW] has a name which is now translated as “Paschal month”, and which was once called after a goddess of theirs named Eostre, in whose honour feasts were celebrated in that month.” (trans. Wallis 2004: 54)
“They” here refers to “the English people”.
Bede, once thought of as a paragon of respectability and objectivity, is now viewed more sceptically by historians: whatever his merits as a scholar, he was not above twisting the facts of the matter in ways that suited him (see e.g. Goffart, Narrators of Barbarian History). Would he invent a goddess? Or perhaps two? Another goddess, Hretha, is also mentioned in this passage just beforehand, in the context of the name of Hrethmonath (March). Like Eostre, she’s only mentioned here, by Bede.
The Oxford English Dictionary (s.v. Easter) makes reference in its etymology to the goddess, and to scholars who suppose that Bede invented her. They respond that it “seems unlikely that Bede would have invented a fictitious pagan festival in order to account for a Christian one”. Perhaps; perhaps not. Careful readings of Bede can reveal many motivations that might not have been obvious on the surface, as Goffart has shown. One such might be simple parallelism: if March is named after a goddess, why shouldn’t April be, too? But there’s also a non sequitur in the OED’s reasoning: it’s the goddess that’s in question, not the festival, and we could in principle perfectly well have the latter without the former. In any case, Bede’s story receives no corroboration of any kind from any other sources, so we can either choose to take him at his word, or not.
Jacob Grimm certainly chose to, and more. In his Deutsche Mythologie – first published 1835 – he takes Bede’s story and runs with it, speculating wildly.
An Old High German lullaby?
Another purported source for this goddess is an Old High German lullaby. This was discovered in 1852 by Georg Zappert, and includes the line Ostârâ stellit chinde / honak egir suozziu “Ostara provides the children with honey and sweet eggs”. It doesn’t explicitly say she’s a goddess, but we’re clearly dealing with an agentive subject of some kind here.
Unfortunately, the text is almost certainly a forgery; see the German Wikipedia entry for discussion and references. One argument for this is the sheer convenience of it: after Grimm in his Deutsche Mythologie in 1835 speculates about the existence of an Easter goddess and bemoans the lack of evidence from beyond Old English, boom! Along comes a text that provides supporting evidence, within twenty years. So we’re back to square one.
An Old Saxon hymn?
Another source, almost as dodgy, is a supposed Old Saxon hymn. Hocker (1853: 224) mentions a hymn preserved in the monastery library at Corvey that starts Eostar Eostar, eordhan modor “Eostar, Eostar, mother of earth”. Montanus (1854: 28) has a fuller version.
The text is very close to an extract from an Old English text, the Æcerbot (“field remedy”) . In the better-known Old English version, the name Erce is found, not Eostar. Montanus describes the Eostar text as Old Saxon, but the very name Eostar makes this dubious: the digraph, presumably representing a diphthong, is not something we’d expect to see in Old Saxon, since the expected outcome of Proto-Germanic *au is a monophthongal long /o/ in this variety.
More problematically, it’s not clear that the Eostar text referenced by Hocker and Montanus was real. More recent compendia of early German texts don’t contain this one, and I’ve been unable to find any source from the twentieth century that doesn’t refer exclusively to Montanus or Hocker. This raises two possibilities. First, like Zappert’s lullaby, the manuscript could have been a contemporary forgery (and for potentially for the same reason: it very conveniently fits Grimm’s narrative). Secondly, the supposed manuscript may never have existed at all: Hocker or Montanus might have simply invented it. It’s not at all easy to trace what happened to manuscripts from Corvey, but until this one physically shows up somewhere, the corroborating value of this particular source is virtually nil.
Votive inscriptions?
A final potential source is a set of votive inscriptions to the matronae Austriahenae, discovered in 1958. Philip Shaw’s 2011 book Pagan Goddesses in the Early Germanic World considers these to be fairly conclusive evidence in favour of a goddess, but I’m not convinced. First, the matronae are plural, not singular. Secondly, while Shaw goes to some length to argue that Eostr- and Austr- could be cognate – and I agree – what’s the -iahenae doing there?
Interim summary
I’ve discussed four sources: Bede’s Reckoning of Time, an Old High German lullaby, an Old Saxon charm/hymn, and some votive inscriptions. Of these, the second probably isn’t real, the third may well also not be, and the fourth doesn’t tell us anything useful. So we’re left with Bede, realistically.
But let’s consider what conclusions we can draw.
A Germanic goddess?
The cross-disciplinary consensus these days is that the descriptor “Germanic” is really only useful for languages. We can use the term in a sort of awkward metonymic way for “things related to people who speak Germanic languages”, sure, if we want to. In that sense we can speak of “Germanic law” or “Germanic art”, etc. But we have to be careful: by that logic, my iPad is Germanic. If we’re going to use the term responsibly, it should at the very least denote more broadly, for instance something shared or inherited across people who speak Germanic languages.
Let’s consider the evidence for an Easter goddess in that light. Even if we assume all of the sources discussed were real, and provided solid evidence for a singular goddess with a cognate name, we would still not be entitled to conclude that Eostre/Ostara/Eostar/Austr… was a Germanic goddess. This is because all these sources come from people who spoke a single branch of the Germanic languages: West Germanic. Evidence from the other branches is entirely lacking. There are no relevant North Germanic sources for such a goddess, despite a literature on pre-Christian customs that is in general much richer than that of West Germanic: Eostre had ample opportunity to show herself, but she didn’t. (There is evidence for a cognate noun austr meaning “east”, but none at all for any goddess connection.) And as Grimm laments, Gothic – our earliest robustly attested Germanic language – doesn’t even use a cognate term for the festival, instead preferring paska, derived from Hebrew via Greek. So the broadest possible conclusion we could reasonably reach, under those assumptions, is that there was a pan-West Germanic goddess with a name related to Easter.
This is in stark contrast to other gods, e.g. Thor, Odin, Frigg etc., who Shaw (2011) terms the “great gods” – these are much better attested. Though even here we have to be careful: as Shaw argues in his 2002 PhD thesis Uses of Wodan, names like Odin did not reliably denote the same individual across different sources, even if formally cognate. Thus, even under the most “charitable” possible interpretation of the sources, and on an extended view of what it means to be “Germanic”, there’s nothing to suggest that Eostre was part of a pan-Germanic pantheon.
Conclusion
Acceptance of a goddess Eostre has been variable to say the least. Throughout the nineteenth century there were voices for and against. At the end of the twentieth century, opinion was still divided. At one end of the spectrum, Page (“Anglo-Saxon Paganism”, 1995: 125), describes her as “an etymological fancy on Bede’s part”, and Udolph (Ostern: Geschichte eines Wortes, 1999) dismisses the goddess etymology almost out-of-hand. The Dictionary of Old English also takes this stance. At the other extreme, Green (Language and History, 2000: 351–3) is happy to take Bede’s word as gospel with a philological shrug of the shoulders. And then there’s the Wikipedia page on Eostre, which treats her existence as having been conclusively demonstrated. Sadly, the latter is a horrendous dumpster fire of lousy argumentation, conflations of form and meaning, and ignorance of the basic principles of sound change, at least in its present form. (And in this blog post we’re not even going to talk about the purported association of Eostre with rabbits/hares or with the dawn, which fail to withstand even the most basic scrutiny.)
Shaw (Pagan Goddesses, 2011: ch: 4) tries to find a middle way. He argues that Eostre was a “local goddess”, venerated in Kent specifically, based mostly on place-name evidence. There’s still quite a lot of creative interpretation here, and in any case my aim with this blog post has not been to convince you that one view is correct. Rather, I aimed to show that the evidence is ropey even by early medieval standards, and how much one’s prior assumptions and understandings can skew one’s interpretation of it in one direction our another.
As for me, well, I’m no Christian, but I won’t be signing up to the cult of Eostre any time soon.
