A Couplet on King Penda of Mercia in the Old Welsh Englynion y Beddau (Stanzas of the Graves)
William Sayers
Cornell University
The life of Penda, king of Mercia in the first half of the seventh century and reputedly the last pagan king of Anglo-Saxon England, is well documented, primarily in Bede's Ecclesiastical History of the English People but also in The Anglo-Saxon Chronicle, and the Historia Britonum.1 Less sure reference is found in the seventh-century praise poem Mawrnad Cynddylan (Elegy for Cynddylan), in which a "son of Pyd" ("mab pyd") is mentioned as an ally of this Welsh leader. This last reference in turn provides a political and ethnic context in which to evaluate the presence of Penda in the recently re-edited Welsh compilation Englynion y Beddau (Stanzas of the Graves).2 Penda (d. 655) is the chronologically latest leader to be mentioned in this work, adding possibly to his significance as both pagan and Angle or, more generically, Saxon. The entry for Penda in this series of brief elegiac mentions of dead heroes, the manner of their death, and often their resting places is limited to two brief lines:
Bet Run mab Pyd in ergrid avon.
in oervel, ig gverid.
"The grave of Rhun ap Pyd (is) in the horror of a river,
in the cold, in the soil."3
Here the editor argues for Penda ap Pyd as the correct reading, Rhun being an erroneous substitution.4 In the subsequent discussion of Anglo-Saxon names that are found in what is conventionally known as Series III of the Englynion y Beddau the editor calls attention to "other seemingly vague references, like that to Penda's grave in 'the river's tumult,' which cannot have been composed near the river concerned, which was in England; such allusions must have ben intended to remind the audience of topographical traditions which they already knew" (281). Penda's mention in this series of englynion reads:
ed Panna vab Pyt
Yngorthir Aruon dan i oer ueryt
"The grave of Panna son of Pyd in the uplands of Arfon [recte, in a river's horror]
under its (or his) cold soil. (303-304)
As illustrated here, the challenging editorial work in Old Welsh proceeds one word, personal or place name at a time.
The present note will draw on Sims-Williams's edition and commentary in two areas: the problematics of the name Penda (seen just above in the variant Panna), and the sites proposed for Penda's fatal engagement, known in tradition as the battle at the river Winwæd, which in the above couplet has become a fluvial burial in Wales.5 Dynastic names such as Penda may reflect alliances through marriage of the royal house of Mercia with neighboring north British ruling families. This is better documented in the case of Penda's neighbor to the north, sometime ally, and ultimate opponent, Oswiu of Northumbria, whose wives included a Briton, Rhiainfellt from Rheged, a Gael, of the Cenél nÉogain dynasty of the Inishowen peninsula, and an Anglo-Saxon, Eanflæd, daughter of Edwin, raised in the Christian environment of Kent. One resolution, admittedly speculative, of the conflicting textual evidence, which includes genealogies, would be to make Pyd ('danger' in Welsh) the object of a generational remove and locate him as the maternal grandfather of Penda.6 One of his daughters, in this hypothesis, married the Anglo-Saxon Pybba.7 Another could have been married to Cadfan of Gwynedd. Cadfan's son Cadwallon and Penda would then have been cousins in a larger pattern of exogamy. A Welsh genealogy makes Pyd the maternal grandfather of Cadwalladr, the son of Cadwallon, which, modified to great-grandfather, supports the above hypothesis in part.8 Geoffrey of Monmouth states that Cadwallon married a sister of Penda, fully plausible in light of other such dynastic unions. Raised, we may imagine, by his British mother as a Christian, Penda's paganism may have been a court identity intended to maintain popular support among his largely Mercian and Anglian subjects. Speculation even extends to a functional bilingualism on Penda's part.9 Penda's court may have been less cosmopolitan than Oswiu's, with its learned Gaelic presence and the king's own Irish schooling and flattering epithet sapiens.10
"A dominant figure in English history between the early 630s and 655" (Sims-Williams, p. 155), Penda then had extensive political relations with both his British and Anglo-Saxon neighbors. These included his alliance with his cousin Cadwallon of Gwynedd against the Northumbrians in 633, which ended with Cadwallon's death in conflict with King Oswald of Mercia, and the legendary ascription to the son of the latter, Cadwalastr, of the removal of Penda's corpse to Wales for burial, whether in or by a river, as the Englynion y Beddau have it. Earlier, troops from Powys supported Penda at the battle of Maserfelth (641/2), where he slew the Mercian king Oswald, Oswiu's predecessor. British chieftains are also reported to have supported Penda at the battle of Winwæd, only one of whom escaped alive, Cadafael of Gwyneth, the likely successor to Cadwallon.11 In sum, the presence and treatment of Penda's name and patronymic in the Old Welsh poetry confirm that the nominally Anglo-Saxon ruler was known to Welsh tradition, may even have been considered part of it. Thus, both Penda's ethnicity and his paganism as recognized by current scholarship may have to be qualified.
To turn to more strictly onomastic matters, with the Welsh variant on Penda replaced by Rhun in some instances of the textual evidence, we do well to begin with Anglo-Saxon onomastics.12 The name Penda occurs in only one other context, a spiritual brotherhood mentioned in the Liber vitae Dunelmensis. Since no convincing Germanic analogues have been identified, a Brittonic origin for the name has found the strongest support.13 A British cognate of Welsh pen 'head, summit' naturally suggests itself but names with this root are generally compounds and it is the other element (the deuterotheme) that carries the greatest semantic weight as a distinguishing feature.14 Pen also meant 'chieftain, leader, lord' and a formulation *penn dda 'good ruler' (anglicized Penda) could be envisaged, although such general, non-specific designation might be expected for the founder of a dynasty.15 Hitherto unnoticed in this regard is the Gaulish root panto- 'suffering' (heroically borne or inflicted), which figures in personal names (Pata, Patto) and theonyms.16 If there were an original form Penna/Panna, a derivative Panda might have been judged more appealing to Germanic onomastic tastes. But this is doubtless carrying reconstructive speculation too far. Penda's father's name was Pybba (variants Pibba, Wibba, Wybba), replaced in Welsh lore by Pyd, as noted above. If the original form were Pybba, a cognate of Gaulish uepos 'voice, speech', especially forceful, as used in proper names may be considered, although Pybba is ostensibly an Anglo-Saxon.17 The alliteration in dynastic names, which extends to Penda's son Peada, along with a taste for geminate dental consonants plus -a (hypocoristic?), criteria which even Pyd satisfies, has counterparts elsewhere in western Europe. Given that none of the trio of names in P- has an assured Welsh or British counterpart, these may be hybrid Germanic-Celtic coinages, family-specific as it were.18 Such naming practices may even have had a propagandistic intent to assure broad support in a bi-ethnic society.
Penda of Mercia and Oswiu of the consolidated kingdom of Northumbria joined battle near a river called the Winwæd in November 655, after years of on-and-off conflict. According to Bede, with the death of Penda Anglo-Saxon paganism passed with what may have been largely its figurehead. Significant of the times and belying an idea of ethnic blocks, the belligerents were Northumbria and the British kingdom of Bernicia led by Oswiu on the one hand, and the kingdoms of Mercia and East Anglia, British Deira, and Welsh forces, all led by Penda on the other. Like the name Penda, Winwæd has been ascribed both Germanic and Brittonic etymologies. In the first instance the constituent Old English elements win 'strife, fight' and wæd 'shallow water' have been advanced. But was the river so named in memory of the the battle or does the name as understood represent its pre-eminent feature? The Brittonic candidate is a cognate of Welsh gwenwedd 'whiteness', more likely agitated 'white water' than some natural coloration.19 The Winwæd is generally thought to have been in the vicinity of Leeds, since Oswiu concluded his military campaign there. A plurality of scholars favors an identification of the Winwæd with the Trent. At this point, it will be pertinent to recall the characterization of the river (whether the site of death or of the ultimate burial of Penda) in the Englynion y Beddau: ergrid, variously rendered, literally and figuratively, 'tumult' and 'horror'. A fluvial configuration that accounts for much of this evidence is the necessarily agitated water at the confluence of the Trent and the Ouse, where they form the Humber estuary. Not only do we have this meeting of waters, both were tidal rivers, and, furthermore, the Trent was notorious for its flooding, to which phenomenon Bede attributes Penda's loss in the battle (Bede, III.24). Bede's fictions fed back into Welsh legendary history which was often ready to take them at face value. All this supports prior claims of an identification of the Battle of Winwæd as somewhere on the Trent, mostly like downstream, in the vicinity of the Humber to the east and Leeds to the west. Penda would have been fighting on his home territory, and Oswiu conceivably both the invader and aggressor. As for the name Winwæd, note the towns Winterton and Wintering to the east of the Trent and not far from the Humber.
The oldest of the Englynion y Beddau are judged by Sims-Williams to date to the eleventh century (48), although formal features that facilitate memorization suggest a centuries-deeper oral past. The legendary removal of Penda's body from the battle site, or a first burial, to north Wales by his ally Cadwaladr ap Cadwallon of Gwynedd as well as his very presence in Welsh texts has considerable symbolic significance. It complements Penda's hybrid name, his ties to, and possible maternal descent from, north British monarchs (leading to the substitution in Welsh sources of the Briton Pyd for the Anglo-Saxon Pybba as father), and his likely marriage alliance(s). As the last-named ruler in the Englynion y Beddau he seems, with his possible "public paganism," both to close off the pre-Christian period in northern Anglo-Saxon Britain and mark the terminus for legendary Welsh and British heroes in the Celtic literary tradition, at a time when Wales and Cornwall alone had the potential to survive as an independent matrices for Celtic language culture in Britain. Penda's presence in the Welsh work is now seen as less anomalous than previously, given his Welsh and Brittonic ties. The information on Penda in the Englynion y Beddau is scant in the extreme but does open a vista for further research on an England in the process of consolidation.
[1.] Bede, Ecclesiastical History of the English People, ed. and trans. Bertram Colgrave and R. A. B. Mynors (Oxford: Clarendon: 1992), II.20, III.7, III.16-18, III.21, III.24.
[2.] The Medieval Welsh Englynion y Beddau, ed. and trans. Patrick Sims-Williams, Studies in Celtic History XLVI. Woodbridge, UK: D. S. Brewer, 2023.
[3.] Englynion y Beddau, ed. Sims-Williams, Series I.10, p. 141.
[4.] See the fuller discussion in Englynion y Beddau under the heading "Run Mab Pyd," p. 154.
[5.] Relevant background study in Frank Stenton, Anglo-Saxon England (Oxford: Oxford University Press, 1943), pp. 45-47, 81-83, and Nicholas Higham, J. Ryan, and J. Martin, The Anglo-Saxon World (New Haven, CT: Yale University Press, 2013), p. 143.
[6.] Geriadur Prifysgol Cymru (Cardiff: University of Wales, 2024), s.v. pen; LINK.
[7.] On Pybba, see Higham et al., p. 143.
[8.] Englynion y Beddau, p. 44.
[9.] Speculative discussion on Oswiu's court and literary patronage planned in William Sayers, "Bede's Cædmon: A Bilingual Scop from the Cosmopolitan Court of Oswiu?" in progress.
[10.] Colin A. Ireland, The Gaelic Background of Old English Poetry Before Bede (Berlin: Walter de Gruyter, 2022), pp. 278-279 et passim.
[11.] Summarized in Englynion y Beddau, p. 156.
[12.] Concise review of earlier hypotheses as to the origin of the name Penda in "Penda of Mercia," LINK under "Etymology."
[13.] Summary of the debate in John T. Koch, Celtic Culture: A Historical Encyclopedia (ABC-CLIO: Santa Barbara, 2006), p. 60; and discussion in Englynion y Beddau, pp. 154-156.
[14.] Xavier Delamarre, Dictionnaire de la langue gauloise: Une approche linguistique du vieux celtique continental, 3rd ed. (Paris: Éditions Errance, 2003), p. 249, s.v. penno-.
[15.] I am grateful to an editorial reader of an early draft of this note for this suggestion.
[16.] Delamarre, p. 246, s.v. panto-.
[17.] Delamarre, p. 314 s.v., uepos.
[18.] For toponyms, see Graham Jones, "Penda's Footprint? Place-Names Containing Personal Names Associated with Those of Early Mercian Kings," Nomina 21 (1998): 29-62.
[19.] Andrew Breeze, "The Battle of the Uinued and the River Went, Yorkshire," Northern History 41 (2004): 377-383.