On Wednesday, November 9, 2016 at 12:11:01 AM UTC-8, taf wrote:
> In my continuing series of posts on Anglo-Saxon genealogical sources online,
> Next up is what Dumville called the West Saxon Genealogical Regnal List.
Looking back into these, I see something that conflicts with Dumville's explanation. Seeing that versions P & Q (associated with ASC A and ASC G) have a pedigree that runs Cynric, Cerdic, while the others that have this portion, N, R (x2), T, and V, all have Cynric, Creoda, Cerdic, he concludes that the original originally had Creoda, but that the version associated with ASC A purged this generation to conform with the descents in the ASC annals. There is a problem with this, though. The West Saxon GRL begins with an account of the conquest of Wessex by "Cerdic and Cynric, his son". Thus P & Q are internally consistent, while R, T and V are internally inconsistent.
Setting aside Q, since it is known to be a direct copy of P, we are left with one version without Creoda and all of the others with, including the oldest, that appears to be very close to the original. P is clearly the one that was modified. The fact that it is associated with the ASC need not be relevant in the least - it may have been corrected simply to be internally consistent.
Be that as it may, this suggests that when the tract first achieved its coherent form, it was inconsistent. This indicates that it was probably not composed by a single compiler, de novo, but instead constructed from preexisting components.
It can be viewed as consisting of the following parts:
1. The conquest of Wessex by Cerdic and Cynric (without Creoda)
2. The pedigree from Cerdic to Woden
3. A regnal list of the kings of Wessex
4. The pedigree from AEthelweard to Cerdic (with Creoda)
5. The succession of AEthelweard's sons
While the two pedigrees take the same format, each of the others have a different flavor to them. It looks to me like this tract was constructed in pieces:
A. They started by creating a non-genealogical chronology. They began with the conquest, for which they used a proto-ASC 495 annal. The next text also has the flavor of annals, compressed to a bare minimum, but soon it became little more than a regnal list in sentence format. I think they either mined the annals of a chronicle for succession and death records to come up with years ruled, or alternatively they took a regnal list and expanded it out a little, relating each king to the previous and repeatedly stating the right by descent from Cerdic.
B. They took a pedigree from AEthelweard to Woden. As Dumville pointed out (talking about just the portion above Cerdic), this pedigree is part way along the path Sisam described He suggested that the Anglian collection pedigrees are the oldest survivors, and represent a stage just after the hijacking of the Bernician royal pedigree. He then sees the Wig-Freawine pair being added, Frithogar and Esla are added alliteration (at some point Aluca having become Elesa). The GRL pedigree fits nicely within this progression.
Bernicia: Woden-Baeldaeg-Brand-Benoc-Aloc-Ingui
(transfer from Bernicia to Wessex)
WessexAC: Woden-Baeldaeg-Brond-Gewis-Aluca-Cerdic
(addition of Wig-Freawine, Aluca becomes Elesa)
Sisam
Hypothetical
Intermed: Woden-Baeldaeg-Brond-Freawine-Wig-Gewis-Elesa-Cerdic
(addition of Frithogar to establish alliteration with Freawine, perhaps as part of the same step)
WessexGRL: Woden-Baeldaeg-Brond-Frithogar-Freawine-Wig-Gewis-Elesa-Cerdic
(addition of Esla to establish alliteration with Elesa)
WessexASC: Woden-Baeldaeg-Brond-Frithogar-Freawine-Wig-Gewis-Esla-Elesa-Cerdic
Thus, a pedigree older than that in the ASC 855 annal, but more derived than that of the Anglian collection would have been used. This pedigree, like that of the Anglian collection, had yet to lose Creoda. It was then broken into two parts, the first inserted right after Cerdic is first mentioned, the second at the end to connect the (then current) king back to Cerdic. In combining a pedigree that had Creoda into a text based on an alternative tradition that lacked Creoda, they created the internal inconsistency. I see this happening during the reign of AEthelwulf, else they would have used a pedigree that built on one of the later kings.
This produced the proto-GRL.
C. During the reign of Alfred, text was added to the end to bring it down to Alfred. This resulted in GRL, as was passed down to all the surviving versions, except for S and U. S was inspired by a copy of the GRL, but did not use any of its original text, just the continuation of T. The situation with U is different Dumville presents it as a regnal table derived from the GRL, but it could have gone the other way - this could be the regnal list from which the original regnal document (step A) was expanded.
In summary, I think the conflict that must have been within the original GRL tells us that it did not arise, fully formed, from the head of its compiler, but resulted from the combination of several distinct packets of information.
taf