10 November, 2011

Cousins at war: Ecgfrith of Northumbria and Bridei son of Beli

In 685, Ecgfrith of Northumbria and Bridei son of Beli, King of the Picts, fought a decisive battle which resulted in Ecgfrith’s defeat and death and an end to Northumbrian ambitions in Pictland. Historia Brittonum adds the intriguing detail that Ecgfrith and Bridei were cousins. How might that be so?


Historia Brittonum

Egfrid is he who made war against his cousin Brudei, king of the Picts, and he fell therein with all the strength of his army and the Picts with their king gained the victory; and the Saxons never again reduced the Picts so as to exact tribute from them. Since the time of this war it is called Gueithlin Garan
--Historia Brittonum ch. 57, available online

Brudei, Brude, Bruide, Bride, Bridei are all alternative spellings; Egfrid is an alternative spelling of Ecgfrith.

The Latin text is:
echfrid ipse est qui fecit bellum contra fratruelem suum, qui erat rex pictorum nomine birdei et ibi corruit cum omni robore exercitus sui et picti cum rege suo uictores extiterunt et numquam addiderunt saxones ambronum ut a pictis uectigal exigerunt. a tempore istius belli uocatur gueith lin garan.
--Historia Brittonum ch. 57, Latin text available online

In the original Latin, the term translated into English as ‘cousin’ is ‘fratruelem’. Wikipedia says that this is a specific term meaning ‘maternal first cousin’, i.e. indicating that Ecgfrith and Bridei were the sons of two sisters. I have also seen definitions saying that it can mean that they were the sons of two brothers; it’s unclear to me whether the term can also extend to sons of two siblings, i.e. sons of a sister and a brother.

Pictish Chronicle
Bride filius File .xx. annis regnauit.
Tolorcan filius Enfret .iiii. annis regnauit.
--Pictish Chronicle, available online

File is an alternative spelling of Beli or Bile. Enfret is an alternative spelling of Eanferth.

Annals of Ulster
642 Afterwards Domnall Brec was slain at the end of the year, in December, in the battle of Srath Caruin, by Hoan, king of the Britons

686. The battle of Dún Nechtain was fought on Saturday, May 20th, and Egfrid son of Oswy, king of the Saxons, who had completed the 15th year of his reign, was slain therein with a great body of his soldiers
722 Mael Corgais from Druim Ing, and Bile son of Eilphín, king of Ail Cluaithe, die.
693. Bruide son of Bile, king of Foirtriu, dies
--Annals of Ulster available online

Strathclyde genealogy
Run map arthgal map Dumnagual map Riderch map Eugein map Dumnagual map Teudebur map Beli map Elfin map Eugein map Beli map Neithon map Guipno map Dumngual hen map Cinuit map Ceritic guletic …
--Harleian genealogies, available online

In an Irish Life of St Adamnan, Bridei is described as “son of the king of Dumbarton” (according to Tim Clarkson’s website Senchus). Dumbarton, also known as Alt Clud (“Rock of Clyde”), was an important centre for the kingdom of Strathclyde.

There are two Belis to choose from in the Strathclyde genealogy. Beli map Neithon appears in the middle of the list. His son Eugein (a variant spelling of Owain) may be the ‘Hoan King of the Britons’ recorded as having won the battle of Strath Carron in 642 in the Annals of Ulster. This date is consistent with Beli having lived in the early-to-mid seventh century (since he had an adult son in 642). If correct, it is in turn possible that Beli could also have fathered a son who was adult in 685.

Another Beli, Beli map Elfin, appears three generations later, but he died in 722 according to the Annals of Ulster and so cannot be the Beli who was the father of Bridei.

For in the following year [685], King Egfrid [...] rashly led an army to ravage the province of the Picts. The enemy pretended to retreat and lured the king into narrow mountain passes, where he was killed with the greater part of his forces on the twentieth of May in his fortieth year and the fifteenth of his reign.
--Bede, Ecclesiastical History, Book IV Ch.26

Bede does not say who Ecgfrith’s mother was. As Ecgfrith was around 40 in 685, he was born in around 645. His father Oswy married Eanflaed, daughter of Eadwine of Deira/Northumbria some time before 651, because Bede recounts a story about a miracle performed by Bishop Aidan (who died in August 651) about Eanflaed’s voyage to Northumbria (Bede, Ecclesiastical History Book III ch. 15). It seems likely (although not certain) that Ecgfrith was the son of Oswy and Eanflaed, since Ecgfrith appears to have succeeded Oswy without opposition, and a son of Oswy (of the royal house of Bernicia) and Eanflaed (of the royal house of Deira) would have had a strong claim.

If Ecgfrith was the son of Oswy of Bernicia and Eanflaed of Deira, how could he have been a cousin to Bridei, son of Beli of Strathclyde and king of the Picts?

Scenario (a): through an unrecorded daughter of Eadwine

Click to enlarge

This scenario postulates that Eadwine had another daughter, unrecorded, who married Beli ap Neithon of Strathclyde and became the mother of Bridei. This hypothetical daughter would be the sister or half-sister of Eanflaed, making Bridei and Ecgfrith the sons of two sisters or maternal first cousins. In its favour, this scenario fits the use of the term ‘fratruelem’.

It has several disadvantages. First, if Bridei had two non-Pictish parents (father king of Strathclyde, mother a Northumbrian princess), where did his claim to be king of the Picts come from? One possible resolution to this problem is to suggest that Beli of Strathclyde may have had Pictish ancestry and that this was the source of Bridei’s claim. Another is to suggest that Bridei’s mother had Pictish ancestry and Bridei’s claim came through her. A maternal claim would fit with the hypothesis that the Pictish royal succession had at least some matrilineal component (discussed in an earlier post). If one postulates that Eadwine married or had a liaison with a lady of the Pictish female royal line (‘X’ in hypothetical family tree (a) above), then matrilineal succession would mean that the sons of this union were eligible for the Pictish kingship via their mother. What about the daughters? If the daughters were eligible to be mothers of future kings of the Picts, then a daughter of Eadwine and X would be able to pass a claim to the Pictish kingship to her son (the grandson of X through the female line). By this mechanism, a daughter of Eadwine and a Pictish royal lady could bear a son (Bridei) who would be eligible to be considered as king of the Picts. This relies on matriliny operating over two generations, so that as well as the sons of a Pictish royal lady being eligible for the Pictish kingship, the sons of her daughters were also eligible. This doesn’t sound implausible, but as far as I know there is absolutely no evidence for it.

A second disadvantage is that as far as I know there is no evidence to suggest that Eadwine had any dealings with the Picts, either friendly or hostile.

Scenario (b): through a marriage between Oswald and an unrecorded sister of Beli

Click to enlarge

This scenario postulates that Oswy’s brother Oswald married an unrecorded sister of Beli of Strathclyde. In its favour, it introduces no difficulty with Bridei’s claim to the Pictish kingship, as it makes no assumptions about Bridei’s mother and therefore she could have been a lady of the Pictish royal line whose sons were eligible for the Pictish kingship. Oswald was in exile among the Scots of Dal Riada (roughly modern Argyll) from 617 to 633 or 634. A marriage with the neighbouring kingdom of Strathclyde during this period of exile would make reasonable sense, either as an alliance between Oswald’s hosts in Dal Riada and their neighbours across the Clyde, or as an alliance between Oswald in his own right and an ally who he may have seen as a potential source of support for his own claims to Northumbria, or a bit of both.

Against it, this scenario makes Ecgfrith and Bridei cousins only by marriage, with Ecgfrith’s uncle Oswald marrying Bridei’s (hypothetical) aunt. This may not have counted as ‘fratruelem’, depending on how the author of Historia Brittonum used the term.

Scenario (c): through an unrecorded sister of Oswy

Click to enlarge

This scenario postulates that Oswy had an unrecorded sister or half-sister, daughter of Aetheferth of Bernicia and his wife Bebba (or another lady), and that this unrecorded sister married Beli of Strathclyde and gave birth to Bridei.

In its favour, this scenario would make Ecgfrith and Bridei the sons of a brother and a sister (or half-sister), which would make them first cousins and could be consistent with the term ‘fratruelem’ if it extended to include children of siblings of either sex. Eanflaed and Oswy’s (hypothetical) unrecorded sister would have been sisters-in-law. Depending on how in-laws were viewed, the writer of Historia Brittonum may have considered them sisters and thus their sons as ‘fratruelem’ even if the term was meant specifically to mean sons of two sisters.

It has the same disadvantage as the first one mentioned for scenario (a) above: the source of Bridei’s claim to the Pictish kingship if he had two non-Pictish parents. As above, a possible resolution to this problem is to postulate that Bebba was a lady of the Pictish royal family, and that Pictish matriliny extended for two generations so that the sons of her daughter were eligible for the Pictish kingship. There is a slight straw of evidence that might support this, as Bede says that one of Aethelferth’s sons, Eanferth, lived in exile among the Picts. This would be consistent with a connection between Eanferth and the Pictish royal family, which would fit with Eanferth being the son of a Pictish lady.

Scenario (d): through Eanferth’s Pictish marriage

Click to enlarge

Aethelferth’s son Eanferth, brother or half-brother of Oswy and Oswald, appears in the Pictish king-list as the father of a Pictish king, Talorcan. If the Picts followed a form of matrilineal succession to the kingship, the logical implication is that Eanferth married a lady of the Pictish royal family while he was in exile among the Picts. This scenario postulates that Eanferth and his Pictish wife also had an unnamed daughter, sister of the Pictish king Talorcan, and that she married Beli of Strathclyde and was the mother of Bridei.

In its favour, this hypothesis fits easily with Bridei’s claim to be king of the Picts. In this scenario, Bridei would be the maternal nephew of a previous king of the Picts, Talorcan, a likely candidate for the kingship.

Against it, under this scenario Bridei and Ecgfrith would be second cousins a generation apart, which may not have counted as ‘fratruelem’, depending on how the author of Historia Brittonum used the term.


Any of these scenarios is possible, and they all have advantages and disadvantages. No doubt there are other possibilities as well.

Scenario (a) relies on Eadwine having married into the Pictish female royal line, on an unrecorded daughter from such a marriage who then married Beli of Strathclyde, and Pictish matriliny extending for two generations. It is possible that Eadwine’s wanderings in exile “through all the kingdoms of Britain” extended to the Pictish lands and a romantic entanglement and/or dynastic marriage. However, despite the obvious romantic appeal of such a notion, there is nothing in Bede or Historia Brittonum to support any involvement of Eadwine in Pictish affairs. One might imagine that if Eadwine had had ties with Pictish royalty close enough to involve marriage and/or children, he would have had dealings with the Picts and at least some would have been recorded (though the sources are so patchy that this does not necessarily follow).

Scenario (b) relies on a distant connection by marriage being sufficient for the writer of Historia Brittonum to consider Ecgfrith and Bridei cousins. This is possible, if the relationships had become obscured by then (Historia Brittonum was written well over a century after the events, although it may have drawn on earlier sources), or if the writer was using the term ‘fratruelem’ loosely. However, the writer presumably chose the term for a reason, and could equally well have chosen a term for a distant relationship if that was what was meant.

I prefer (c) or (d), as these two scenarios both rely on connections between the Pictish and Northumbrian royal families for which there is some evidence. Bede is clear that Eanferth lived in exile among the Picts and the Pictish king-list is clear that he was the father of a Pictish king. Scenario (d) relies on Eanferth’s Pictish marriage having also produced an unrecorded daughter who then married Beli of Strathclyde, and on the writer of Historia Brittonum using the term ‘fratruelem’ loosely to include second cousins; (c) relies on Eanferth’s mother having been a Pictish royal female, on an unrecorded daughter (sister of Eanferth) who married Beli of Strathclyde, and Pictish matriliny extending for two female generations. Of the two I have a slight preference for (c), because it makes Ecgfrith and Bridei first rather than second cousins, and because if Eanferth had Pictish ancestry it provides a context for his Pictish exile (if he already had family connections there through his mother) and his Pictish marriage.


Annals of Ulster available online

Bede. Ecclesiastical History of the English People. Translated by Leo Sherley-Price, Penguin Classics, 1968, ISBN 0-14-044565-X

Harleian genealogies, available online

--Historia Brittonum ch. 57, translation available online; Latin text available online

Pictish Chronicle, available online


Nicola Griffith said...

Isn't it fun drawing family trees? I've just finished wrestling with the first part of Hild's. We made some similar choices and some different ones:


Gabriele Campbell said...

A pest on those Christian chroniclers who mention women only if they did something really good (found a monastery) or bad (commit adultery) but otherwise ignore them. ;)

Carla said...

Nicola - indeed, part of the challenge and the fun of filling in the gaps with plausible imaginings. Why a niece of Beli for Oswald's wife, incidentally? A marriage into Strathclyde/ Alt Clut seems entirely reasonable, just wondering why you moved on a generation when Beli's son and Oswald's nephew were contemporaries?

Gabriele - Well, most of the chroniclers were churchmen and therefore interested in church history, so founding monasteries was high on their priority list. Hey ho.

Nicola Griffith said...

@Carla, I don't know! I keep asking myself that and simply can't figure it out. I might have to just change it to a younger sister or something. But I'd rather try recreate my thinking. (Just having one of those autumns where there's Too Much to Do, y'know?)

It's a headscratcher :)

@Gabriele, the first thing I'd do with a time machine is go back and get all the women's names...

Carla said...

Is Oswald's wife much younger than him in your story? Or is Beli very much older than Oswald? Or is Beli the youngest of a large family with much older siblings? I can see that any of those could account for placing Oswald's wife in the next generation. It doesn't matter; generations get mixed up all the time, particularly if second marriages are involved.

Well, keep an eye on the neutrinos at CERN :-) Personally I think it'll turn out to be a measurement glitch, but if it isn't, start drawing up your list of questions ...

Annis said...

There you go -I hadn’t realised that Bridei was an actual historical figure! He’s the focus of Juliet Marillier’s fantasy series the Bridei Chronicles.

Carla said...

Is it the same Bridei, Annis? The Bridei Chronicles website says mid sixth century, which sounds a bit early for Bridei son of Beli, who fought Nechtansmere in 685. Could it be Bridei son of Maelchon, who was the king who met St Columba in the late sixth C? Or as its a fantasy series, a bit of both? They were both major historical figures.

Doug said...

It would be picky, I think, to allow the children of two sisters or two brothers but not one of each. Also, assuming that royal descent was passed through daughters because everybody knows who the mother is, but not necessarily the father, there seems no reason why descent could not be through the daughter of a daughter . . .

Annis said...

Sorry Carla, you're quite right -it is Bridei I who is the main character in Marillier's trilogy - it is a while since I read them, so getting a bit confused there. I do remember the St Columba figure - I think he was called Colmcille in the story.

Carla said...

Doug - depends on exactly what the term meant and how it was used. If there were different terms to indicate different types of cousin relationship, then picking the correct one would just be a matter of factual accuracy for the scribe. Or there may have been only one term for cousin, as there is in modern English, or even if there were several they may have been used loosely and interchangeably for all we know.

Yes, it seems likely to me that if descent is reckoned through the female line, daughters' daughters would have some status. Pity that none of the surviving sources provides a detailed explanation of how Pictish matriliny (if it existed) worked in practice.

Annis - the Pictish kings can be confusing, partly because of the variety of spellings. Columcille of Colmcille (sometimes one word, sometimes two) are alternative names for St Columba; if I remember rightly Columba is the Latin form and the others are variants of Irish and/or Gaelic, but don't quote me on that.

Rick said...

I'll also bet on a measurement glitch at CERN.

I don't know any Latin, but it strikes me odd that a Latin word beginning with fratr- would refer specifically to sisters, or relationships derived from sisters.

Beyond that, genealogy is surprisingly good at making my head explode, so I don't have much insightful to add!

Rick said...

Forgot to click the notification button!

Carla said...

This is where you need a linguist who could tell you the origin of the Latin word. Maybe the sons of two sisters were at one time considered to be a special sort of 'brothers' and the word originated from some sort of concept like that? I couldn't say.