A little while ago Tim Clarkson of Senchus brought an Andrew Breeze paper about the history and derivation of the name Hefenfeld*, the Old English version of Heavenfield, to my attention. Its taken me a while to get to it but here is what I think.
It is clear to anyone who has looked at the history of this place-name or even just the place-names that surround it, that versions of hefenfeld have spread over a wide landscape. S Oswaldes Asche is mentioned in several late medieval accounts presumably referring to the cross or a version of it. The entire valley was called halydene (holy valley) by Leland. I know I’ve read of more heavenfield related place-names than Breeze lists; suffice it to say that the holy site left a big footprint in local place-names and lore.
There is also nothing new about the annoying tendency of historians and antiquarians to confuse the camp site of Hefenfeld (modern Heavenfield) with the nearby battlefield site of Denisesburna. (Many otherwise good historians have made real hash out of the places and dates for Oswald’s camp site and battlefield!) This confusion reaches well back into the Middle Ages and may be a reflection of the vague notion about where both were located from the very beginning (though most modern mistakes are just careless reading of Bede). Breeze reviews all of this in considerable detail, although it is only important to his argument to show that the name for the site was never very fixed. [He says that he has shown the date to be 633 but I don't think he has shown that at all.]
Breeze then gets down to his main argument on the relationship and derivation of the names hefenfeld and caelestis campus. First he rules out the Old English name Hefa as a source for hefenfeld, though his reasons don’t seem very sound. Hefa’s becoming ‘hefan’ as in modern Hevingham in Norfolk doesn’t seem that far from Hefenfeld to me. Breeze opts to take Bede at his word, that Hefenfeld is derived from caelestis campus. Fair enough. The English would have been new enough settlers in that area that English place names like X’s field are unlikely to be completely supplanted by alternative place name lore by Bede’s time (though there may have been some intentional renaming of landmarks in English from their British names).
Breeze then turns to the “curious expression” of Caelestis campus. He points to two parallel constructions elsewhere in Bede’s History: campus roborum (‘plain of oaks’, Durrow) and in the Moore Bede campus Cyil, the plain of Kyle in Galloway. Equally he finds more similar constructions in Welsh-Latin texts including Campus Gaii, the plain of Gaius, the name for Bede’s Winwead in the Angles Cambriae and the Historia Brittonum (HB). The HB also includes campus Elleti, where the boy Ambrosius Aurelianus is found my Vortigern’s men. Looking to hagiography Breeze finds campus Heli in the Life of Padarn and Campus Malochu in a charter linked with St Dyfrig. Ok, so we have campus being a common Latin word for plain in Welsh-Latin and apparently taken up for at least place names in early English Latin. This wasn’t really in doubt but its good to see them all collected together. At this point I would like to point out that three of these plains are named for people (Gaius, Elleti, and Malochu) and two are descriptive, plain of oaks in Ireland and plain of brine/salt water (heli) in Brittany. Not surprising for its date and topic, Breeze zeroes in on campus Gaii for comparison.
Since Welsh-Latin used campus Gaii ‘plain of Gaius’ for the battlefield of Uinued, where the Roman road from York to Donchester crosses the river Gwent, Caelestis campus may be explained not as ‘heavenly plain’ but as ‘plain of Caelestis’. It would be a similar place-name survival from Roman times. There is no difficulty about Caelestis as a personal name in Celtic Britain. An inscription of about the year 500 at Barmouth in Gwynedd reads CAELEXTI MONEDORIGI ‘(monument of) Caelestis Mondorix (‘mountain king”). So the evidence suggests that, just as the flood-plain of Gwent was known in British-Latin tradition as campus Gaii, so also the defensive site used by Oswald was known as Caelestis campus, presumably after a local British chieftain or lord, a namesake** of the fifth-century Caelestis of North Wales. (Breeze, p. 196)
First, Caelestis is a late Roman name rendered in modern English as Celestine. It was not uncommon in late antiquity. Pope Celestine I had a tenure from 422 to 432. There is no problem with it being a name in Roman Britain or post-Roman Britain. I don’t think its helpful to think of 7th century Britain as Celtic Britain. To me, Celtic Britain was pre-Roman or areas never under Roman control. I don’t have a problem with caelestis campus referring to a Roman or Romano-British person. It makes more sense than there being a pagan shrine or sacred tree at the site.
Second, Breeze stresses that it was a plain not a field, a plain being much larger. I just want to say that there is another language issue here between US English and UK English. In US English a plain is a very large, flat stretch of land. When a friend and I visited England several years ago we went to Stonehenge. Apart from our impression that it is much smaller than all of the pictures make it out to be, we both agreed that we would never consider it to be sitting on a plain. We also visited Vindolanda on Hadrian’s Wall and Hexham; believe me, I didn’t see any plains. There were some large open fields of rolling but very large hills with lots of valleys. Most of what I saw I would consider hillsides. If you all want to see some plains come the US Midwest. I don’t have a problem with the translation of campus to field. Besides just because the name wasn’t fixed to a specific spot in the later medieval period doesn’t mean that it wasn’t originally more localized. When looking at this picture of Heavenfield to the right keep in mind that we don’t know how wooded the area was in the seventh century. Even so the slope in the land is visible even in this open field picture.
Northumbrian settlers, failing to recognize the personal name in the genitive case here, and taking caelestis as a masculine adjective, seemingly mistranslated the toponym as Hefenfeld. Thereafter Bede could exercise sacred wit on the form, even though in origin it had not more to do with Christian heaven than, say, Anguli in the anonymous Whitby life of Gregory had to do with angeli, Æelli with Alleluia, or Deire with de ira Dei. (Breeze, p. 197)
I’ve never really bought mistranslation explanations. It takes some knowledge of Latin to make this conversion. The average Northumbrian settler would not know that caelestis meant heavenly (as in the heavens, the sky). Knowledge of Latin means churchmen, and churchmen of presumably Hexham would have a motive to use word play to rename the site a fitting name for their shrine. Remember that Bede gives his explanation of the name in an episode that he credits directly to a source at Hexham. It is possible that visiting churchmen or churchmen stationed at the royal estate of Hexham (before it was given for a monastery) renamed the site using word play. It seems to me that the word play translation makes sense and may have been close enough to a translation of the original name (whose namesake would probably have been long dead) to be acceptable to local Britons.
Andrew Breeze. (2007). Bede’s Hefenfeld and the Campaign of 633. Northern History, XLIV: 2, p. 193-197.
*Hefenfeld is also sometimes written as hefenfelth.
**Namesake means different things in US English and UK English. In UK English namesake just means sharing the same name. In US English namesake usually means that one is named directly after the other, ie. John Jr is the namesake of John Sr but not of unrelated Johns.