The Llyfr Aneirin (henceforth LA), now housed in the National Library of Wales (Cardiff MS 2.81), is one of the most celebrated but controversial collections of early Welsh poetry. LA contains the famous elegies collectively referred to as Y Gododdin, four interpolations and four gorchanau. Y Gododdin is the standard modern title derived from a rubric preceding the text that translates: ‘This is The Gododdin. Aneirin sang it’. These poems are attributed to the sixth-century poet Aneirin, and according to conventional understanding, they are laments commemorating the warriors of the Guotodin people and their allies who fell in the battle of Catraeth in the late sixth or early seventh century. The manuscript dates to c. 1250-1265. It was compiled in North Wales, and though the foundation that produced the work is unknown, it might have been the Cistercian house of Aberconwy. The manuscript Peniarth 17 has recently been attributed to one of LA’s scribes, and because of the Gwynedd interest in Peniarth 17, it has been suggested that this scribe belonged to Aberconwy (Mittendorf, 129; Huws, 75).
LA was written in two hands now known as A and B. The hands are contemporary with one another, and exhibit thirteenth-century orthography. Scribe A wrote the majority of the manuscript. Scribe A’s orthography is Middle Welsh. Scribe B’s text, however, preserves elements of archaic orthography and language. Based on linguistic analysis of the B text, two distinct sources can be traced in B (Isaac 1993: 83-9; Isaac 1999), and this has led some editors (e.g. Koch 1997) to categorize the B text as B¹ and B², B² being earlier. There is little doubt that Scribe B was copying from an earlier source (or possibly two), but the dating of B’s lost exemplar(s) is problematic. Many of the poems in the A text are also repeated, with variations, in the B text, which implies that these poems were derived from different exemplars with a common archetype. To summarise the arguments for the dating of the texts, careful analysis of the language and corresponding stanzas between A and B indicates that the exemplar behind B’s text cannot be dated any earlier than the ninth or tenth century, the latter being most likely (Charles-Edwards 1978: 50-1; Simon Evans, 81-7; Padel, 132).
The LA, particularly Y Gododdin, is controversial because most of the elegies appear to be set in the sixth- to seventh-century Old North. However, the LA — the only manuscript to preserve the corpus — dates to the thirteenth century. This is a large chronological gap, and opinions on the antiquity of Y Gododdin are divided.
For decades the assumed date of composition was c. AD 600 (Williams 1938: xl-xlii; Jackson 1978: 56-67). Some have recently argued that the majority of the poems were composed in the Brittonic North from c.540-c.570, and only later transferred to Wales (Dumville, 3-4; Koch 1997: III, IX and xxx, lxxx). Others have argued that the entire corpus is a later Welsh literary creation, reflecting on events of the Brittonic Old North (Padel, 117, 132).
Most of the place-names in LA have been subject to orthographic modernisation, and few are identifiable. Regardless of the difficulty in interpreting the onomastic material, even a cursory reading of Y Gododdin demonstrates that place-names were an important, even integral feature of this style of poetry. Of the 134 stanzas in LA (including the gorchanau and englynion) there are only 29 which do not record place-names or folk-names. This is less than a fourth of the entire corpus. The place-names associated with the individual(s) commemorated were, therefore, an important feature of this poetic tradition.
The following section will consider the majority of the place-names and folk-names that occur in both the A and B texts in Y Gododdin, then those that occur only in B because it is earlier, followed by those in A. This article will conclude with a brief discussion of the place- names and the text itself.
The Guotodin people are recorded 20 times in the A text and 11 times in the B text. The Guotodin are identified with the Votadini, an Iron Age tribe first recorded by Ptolemy, who locates them in Northumberland, Berwickshire, and Lothian to the Firth of Forth. The linguistic transition from Iron Age Votadini > ninth century Guotodin > MidW Gododin > ModW Gododdin is unobjectionable (Jackson 1953: 653; Jackson 1978: 69).
The English are the main antagonists, and the Guotodin are never specifically depicted in conflict with other Britons (contra Koch 1997: xxxii-xxxiv). A variety of terms are used for the English, including Lloegr, Lloegrwys, Saesson and Eingyl. A distinction is made between the Bernicians (Brennych) and the Deirans (Deivyr), which may be significant as the Anglo-Saxon kingdoms of Bernicia and Deira did not amalgamate to form Northumbria until AD 654.
The most frequently mentioned place-name is Catraeth. It is recorded 23 times in the A text and five in B. It is the name of a battle and also a place-name. The name occurs in the formulaic first line gwyr a aeth gatraeth ‘warriors went to Catraeth’ nine times in the A text; in all other instances it is also associated with a battle. The identification of Catraeth has been problematic, but is now accepted to be Catterick on the River Swale in North Yorkshire (Dunshea, 84). Jackson (1978: 83) notes: the ‘identification of Catraeth with Catterick is unobjectionable philologically’. Catterick was strategically important as it guarded the major routes to the North: it was the location of a Roman fort, and is a military base even today.
Eidyn, recorded five times in B and seven in A, is identified with Castle Rock in Edinburgh (Alcock, 165-6). The stanzas recording Eidyn are independent of one another, namely, there is not an equivalent stanza in A or B recording Eidyn. In Y Gododdin, Eidyn is depicted as a political centre in Guotodin territory and under Guotodin control. Some stanzas refer to a battle around Eidyn itself. LA 36 (B) records ‘at the resounding of his shield there was flight, | innumerable men fleeing before the hill (vre) of Eidyn’. This implies a conflict at Eidyn, and the context implies that the Guotodin were the victors.
Nouant and Aeron
Koch (1997: xxx) suggests that Nouant is to be identified with the Novantae, an Iron Age tribe located in south-western Scotland. The name Aeron, which also occurs in each stanza recording Nouant (with the exception of Gorchan Kynfelin), has been used to support this theory. Aeron has traditionally been equated with the River Ayr or Earn Water. However, these are not philological matches for Aeron, though either identification is possible if the Welsh scribe confused this name with the River Aeron in Ceredigion (James, s.v. ajr).
Gwanannon is unidentified. It has an –on suffix, which suggests it was possibly a river-name, and in LA 19 (A) it is described as ‘the borderland of the Guotodin’. Gwanannon is not identified with any known river, nor do we know the extent of Guotodin territory. This name also occurs in LA 17.11 (A), but this particular stanza is unusual in structure compared to the other Gododdin poems, and is likely a later interpolation (Koch 1997: 224).
Gwynedd and Place-Names in Northern Wales
The early medieval kingdom of Gwynedd is recorded on five occasions in the A text and once in B. Two almost identical passages in A and B also record the cantref name Rhufoniog (Rywynyauc) and the River Aled (Alet), which flows through Rhufoniog; the person commemorated is known for his participation ‘in the second battle seen around the Alet’. This is a specific reference to a battle that is not associated with Catraeth. These stanzas do not record any place-names outside of North Wales, and they might be interpolations.
Lleuvre, Lleu-tut and Din Dywyt
These names occur in three stanzas (one in A and two in B), and are derived from a single original that has diverged during transmission (Isaac 1993: 83-9). Lleu-tut is from *Lugu–totā ‘Lleu’s tribe’. The second element, –tut, has the same meaning and interpretation as the Irish cognate túath ‘people, country, territory’. Lleuvre is from *Lugu-*briga meaning the ‘hill of Lleu’, and from context was in the ystre ‘frontier’ of Guotodin. The first element of these names occurs in Lleudonia and Lleudinyawn, the name applied to the region of Lothian in Welsh sources. Lleu-tut and Lleuvre are likely located in Lothian.
Din Dywyt is unidentified. Grugyn, the commemoratee, is said to have come from Din Dywyt over the merin ‘firth’, perhaps the Firth of Forth. The geographical focus of this poem is unquestionably Northern (Lothian and perhaps Pictland), and from the context it appears to have nothing to do with Catraeth (Isaac 1993: 83).
Place-Names in the B Text
Bannauc (‘the peaked hill, or range of summits’) are the Touch and Gargunnock Hills near Stirling. The name is still preserved in modern Bannockburn (Watson, 195-6).
Merin Iodeo is the Firth of Forth. This form also appears in urbs Iudeu ‘city of Iudeu’ in the Historia Britonum, and as Giudi urbs in Bede, which is now generally agreed to be Stirling (Fraser 2008). The spelling Iodeo is also Pritenic, or Pictish, in form (Koch 1997: 136; James s.v. jṻd). Note also that the metre of this stanza is unusual for the LA corpus, and Koch (1997: 135-6) suggests it might be an interpolation.
Place-Names in the A Text
Koch (1980) has convincingly identified Gwynngwn as a later, significantly altered form of Uenicones, a tribe first attested by Ptolemy, located between the Forth and the Dee and particularly around Tayside. Maen Gwynngwn means ‘stone of the Uenicones’. Llif ap Cian, the individual praised in the poem, was from Maen Gwynngwn and probably a Pict. He is perhaps the same ‘son of Cian’ from ‘beyond Bannauc’ (see above).
Gweiryd is the River Forth. Guerit is also the spelling of the River Forth in the thirteenth- century Black Book of Chirk. This identification is further supported by the appearance of Eidyn gaer ‘the fort of Eidyn’ in the preceding line.
The first element, bann-, is ‘horn, peak, summit’, and the second element carw means ‘deer, stag’. Banncarw was likely in Wales. A place called Bancarw is recorded in an early thirteenth- century charter of Llywelyn ab Iorwerth confirming the lands of Aberconwy Abbey; it is associated with the area of Nantgwynant (Gresham 1939 and 1983). The Bancarw of this charter is likely identified with the ridge now called Cerrig Cochion near Snowdon (Gresham 1939: 156). Is it possible that the Banncarw of LA and the Bancarw of the Aberconwy Charter are one and the same place? This is perhaps a poem recording a place in Aberconwy’s holdings that was added to the corpus; as noted above, LA was possibly compiled at Aberconwy Abbey. Koch (1997: 216-17) notes that the dating of this piece is complicated, and that Scribe A does not follow his ‘normal practice’ in this stanza. There is no association with Catraeth, and this poem is perhaps an interpolation.
This unidentified name means ‘ford at the head of the Clwyd’. Clwyt in Penclwyt cannot be identified with the River Clyde philologically. Clwyd is a common Welsh river-name, the most notable of which is the Clwyd in Denbighshire. There are no other place-names in this stanza that associate Ryt Benclwyt with the events around Eidyn or Catraeth, and it may refer to a ford on the River Clwyd in North Wales.
Though Manawyt is also a personal name, the preceding term breithel ‘land, region’, indicates that it was a geographic area. Manawyt may be a mistake for Manau, the region around the head of the Firth of Forth, which was possibly a sub- division of Guotodin territory (Watson, 103; Charles-Edwards 1974: 41). This name is still preserved in Slamannan and Clackmannan.
Jarman (1988: 80, n.45) observed that the battle alluded to in this stanza has nothing to do with Catraeth, and the line in which this name is found suggests that the enemy, who is not named, fled before the Guotodin army.
Despite the complicated nature of this material, examining the contents of Y Gododdin alongside its place-names can shed new light on the history of the poetry and LA. Conventionally, Y Gododdin has been viewed as a coherent poem commemorating warriors who fell in the battle of Catraeth. Many of these poems focus on Catraeth, and it was probably a significant battle, but some place-names and their contexts indicate that it is not the only conflict recorded.
A few stanzas refer to a battle around Eidyn. The stanzas that record Lleuvre, Lleu-tut and Din Dywyt are not associated with Catraeth. Likewise, the stanzas recording place-names in Wales are not associated with Catraeth, and one refers to a battle around the River Aled. The conclusion we should draw is that at least three conflicts are recorded in the Gododdin poems: Catraeth, one around Eidyn, and at least one in North Wales. These conflicts were probably independent, and this implies that the poems in Y Gododdin likely reflect more than one time period.
Conventionally, Y Gododdin is attributed to a single poet, Aneirin. But this is questionable, given that so many interpolations have made their way into the text. The Reciter’s Prologue, Pais Dinogad, Gorchanau and Strathcarron Stanza are known interpolations, but it has also been suggested that stanzas recording the names Rywynyauc and Alet, Merin Iodeo, Banncarw, and one recording Gwanannon are also interpolations. The place-names suggest that we have a body of poems concerned with three areas: Catterick, the region around Edinburgh, and a third group focusing on Gwynedd in North Wales. This is a very broad geography.
What are the consequences of this analysis? Conventionally, Y Gododdin is viewed as early or late, but this all-or-nothing viewpoint prompts the question: why does the entire collection have to be either early or late? Why does it have to be the work of a single poet? The place-name evidence suggests that what we have in LA is an anthology, made of stanzas that probably derive from quite early Northern material, and stanzas that are later compositions which have become attached to ‘Y Gododdin’ because of the shared theme of lament. Lewis (1986: 7-8) was the first to suggest that Y Gododdin as we have it today is a cumulative work based around a common theme, and this theory deserves more attention than it has generally received. The stanzas focusing on the region around Edinburgh and Lothian are probably derived from early written, Northern, material.
It is especially difficult, however, to decide which poems are early and which are late (though much effort has been expended over the past few decades to organise these based on language and orthography). The orthography of many place-names (especially those in the A text) was updated, but the names themselves often preserve elements of their earlier forms, e.g. Maen Gwynngwn.
If we abandon the assumption that the elegies of LA are either all early or all late, then we are freed from many interpretative constraints, but this also has significant consequences for our understanding of early Scottish history, Y Gododdin and LA. Whether or not we view the contents of LA as early, late, or a combination of both, the place-names themselves will likely form a key part of this ongoing debate.
Bibliography and Abbreviations
LA Llyfr Aneirin, ed. D. Huws, Llyfr Aneirin: A Facsimile, reproduced from the facsimile; ed. J. Gwenogvryn Evans 1904-22 (Aberystwyth, 1989).
Charles-Edwards, T. M. (1978), ‘The Authenticity of the Gododdin: An Historian’s View’, in R. Bromwich & R. Brinley Jones (eds.), Astudiaethau Ar Yr Hengerdd (Cardiff), 44-71.
— (1974), ‘Native British Political Organization in Roman Britain and the Origin of MW brenhin,’ in M. Mayrhofer (ed.), Antiquitates Indogermanicae (Innsbruck), 35-45.
Dumville, D. (1988), ‘Early Welsh Poetry: Problems of Historicity’, in B. F. Roberts (ed.), Early Welsh Poetry: Studies in the Book of Aneirin (Aberystwyth), 1-16.
Dunshea, P. M., ‘The Meaning of Catraeth: a revised early context for Y Gododdin’, in A. Woolf (ed.), Beyond the Gododdin: Dark Age Scotland in Medieval Wales (St Andrews), 81-114.
Fraser, J. E. (2008), ‘Bede, the Firth of Forth, and the Location of Urbs Iudeu,’ The Scottish Historical Review, 87, no. 223 (April), 1-25.
Gresham, C. (1939), ‘The Aberconwy Charter,’ Archaeologia Cambrensis, vol. 94 pt. 2, 123-162.
Gresham, C. A. (1983), ‘The Aberconwy Charter; Further Consideration,’ Bulletin of the Board of Celtic Studies, vol. 30, 311-347.
Huws, D. (2000), Medieval Welsh Manuscripts (Cardiff).
Isaac, G. R. (1999), ‘Readings in the History and Transmission of the Gododdin,’ Cambrian Medieval Celtic Studies, vol. 37 (Summer), 55-78.
– (1993), ‘Canu Aneirin Awdl LI,’ Journal of Celtic Linguistics, vol. 2, 83-89.
Jackson, K. H. (1978), The Gododdin: The Oldest Scottish Poem (Edinburgh).
– (1953), Language and History in Early Britain (Edinburgh).
James, Alan (2014), Brittonic Language in the Old North. Available online at: https://spns.org.uk/resources/bliton
Jarman, A. O. H. (1988), Aneirin: Y Gododdin (Llandysul).
Koch, J. T. (1997), The Gododdin of Aneirin: Text and Context from Dark-Age North Britain (Cardiff).
– (1980), ‘The Stone of the Weni-kones,’ Bulletin of the Board of Celtic Studies, vol. 39 pt. 1, 87-9.
Lewis, S. (1986), Braslun o Hanes Llenyddiaeth Gymraeg (Cardiff).
Mittendorf, I. (1999), ‘Sprachliche und orthographische Besonderheiten eines mittelkymrischen Textes aus dem 13. Jahrhundert (Gwyrthyeu e Wynvydedic Veir),’ in S. Zimmer, R. Ködderitzsch & A. Wigger (eds.), Akten des Zweiten Deutschen Keltologen-Symposiums (Tübingen), 127-48.
Padel, O. J. (2013), ‘Aneirin and Taliesin: Sceptical Speculations,’ in A. Woolf (ed.), Beyond the Gododdin: Dark Age Scotland in Medieval Wales (St Andrews), 115-52.
Simon Evans, D. ‘Iath y Gododdin’, in R. Bromwich & R. Brinley Jones (eds.), Astudiaethau Ar Yr Hengerdd (Cardiff), 72-88.
Watson, W. J. (1926), The Celtic Place-Names of Scotland (Edinburgh; repr. 2005).
Williams, I. (1938), Canu Aneirin (Cardiff; repr. 1978).
Kelly A. Kilpatrick
NOTE: This article is a brief summary of a work- in-progress, as outlined at the May 2018 conference. To cite this article, or to discuss the fuller work, please contact the author at: