Medieval Welsh Genealogy: An Introduction and Textual Study

This substantial book does two jobs. It undertakes the first full textual study of Welsh genealogical literature in the Middle Ages, and it provides a new critical edition of the most important texts. In the second of these roles it replaces Peter Bartrum’s Early Welsh Genealogical Tracts (1966), the workhorse on which everyone relied till now. In the first role, however, it has no predecessor. Bartrum offered only a modest commentary and apparatus. That cannot be said of Ben Guy’s book. The task of reviewing this imposing volume calls to mind a certain early Welsh poem in which an inferior warrior takes on the hero, like ‘a shrew that scrabbled against a cliffside.’

Readers may want to know what is so important about genealogy, and also why such basic source criticism is still needed in 2021. Genealogical thinking pervaded medieval Welsh views of the past and there was a dedicated literature of genealogy from a quite early date. There is plenty of material, therefore, and the general shortage of historical sources from early medieval Wales means that genealogies play an outsize role in the reconstruction of the country’s political history. As to why the texts were still in such a deplorable state of confusion, that is a consequence of an abundance of material combined with a shortage of investigators. Few historians have the time or inclination for this work. The astonishingly productive Peter Bartrum was an amateur scholar who worked in his professional life for the Meteorological Office. He has had few emulators, though the name of David Thornton deserves honourable mention, and other historians have dealt with individual problems. As a corpus, however, the earliest Welsh genealogical literature has never been reduced to textual order – until now.

Medieval Welsh Genealogy has been shaped by the decision to take the genealogical collection as the ‘basic unit of analysis’ rather than the individual genealogy (p. 49). Most medieval Welsh genealogy occurs in specifically genealogical texts, or at least in isolatable genealogical sections of longer texts, rather than scattered about. A proper analysis of these texts, their manuscripts, their transmission, and their origins, is a sine qua non for progress, without which historians have struggled to know what to make of genealogies and have used them in an ad hoc manner. This is the situation that Guy set out to address.

The introduction (Chapter 1) is a wide-ranging discussion of genealogy in medieval Wales. Its bears the imprint of two outstanding commentators on Insular genealogy, David Dumville and the late Donnchadh Ó Corráin, in its insistence on the textuality of genealogies and their close links to study of the bible. Right at the outset Guy distinguishes between the uses of genealogy within medieval Welsh society and the much narrower set of reasons for which genealogies were made into written texts—literary genealogy. Only biblical characters, kings and would-be kings, saints and nations as a whole were the proper subjects of literary genealogy. Non-royal freemen are absent from the Welsh collections before the 13th century. The purposes too were narrower: (i) to assert the royal status of individuals or lineages, or the close royal connections of saints; and (ii) to embody nationhood by treating lineages together, and especially by tracing them to a common ancestor. Literary genealogies were rarely if ever recorded during legal cases or inheritance disputes: such day-to-day genealogy remained oral. On the other hand, literary genealogy typically involved collecting and studying many more pedigrees than the average medieval Welsh person would have needed to know, and they could be much longer, ultimately winding their way back to the bible. Literary genealogy, being literary, took specific textual forms: descending and ascending genealogies, and segmentary subtypes of each. All of these, it is now generally agreed, had their inspiration in biblical models. Guy is agnostic about the form that oral genealogy took, which is sensible since we have no access to it.

Chapter 2 deals with the collection known as the Harleian Genealogies, from British Library Harley Manuscript 3859, written around 1100. By Welsh standards this is a heavily researched text and both Guy’s discussion and his footnotes are densely filled with references to a century of scholarship. Notwithstanding this, the treatment is clear and compelling, combining established arguments with progress. The manuscript itself was written by an English or Anglo-Norman scribe at Canterbury; its exemplar was of Welsh origin, probably from St Davids and written in or very shortly after 954—the ‘St Davids Recension’. Some of the genealogies are brought down to that period but most terminate earlier and reflect a longer period of accumulation. The oldest discernible stage of compilation is a collection made in north Wales about a century earlier, which Guy calls the ‘Gwynedd Collection of Genealogies’. A solid case is made for associating it with the reign of Rhodri Mawr (d. 878), whose territorial sway it seems to have documented. The Harleian Genealogies contain other entries besides those taken from the Gwynedd Collection, though it is not possible to be certain of the exact boundaries of all the constituents.

In the second half of Chapter 2 Guy treats various offshoots of the St Davids Recension. All are less familiar and less well understood than the Harleian Genealogies. Much the most interesting are the genealogies attached to the Vita S. Cadoci, a Life of St Cadog of Llancarfan in south Wales, dated roughly 1100 and attributed to Lifris. Guy provides an masterly analysis of these. The immediate ancestry of the saint, from the royal lines of Glywysing and Brycheiniog, appears to have been established doctrine, but by using the St Davids Recension the genealogist was able to extend the saint’s various lineages far back in time. Importantly, Guy demonstrates that the extant St Cadog Genealogies can hardly be the work of Lifris, since there is a discrepancy between the Life and the genealogies: the attachment of Cadog’s mother to the line of Meurig ab Enynni pushes Meurig eight generations earlier than Cadog, whereas in the Vita S. Cadoci (§25) he actually meets the saint. Whoever attached Cadog’s mother to Meurig’s line in this clumsy fashion ignored or missed Meurig’s appearance in the Life.

It is worth noting that, without this join, the genealogies still make Cadog and Meurig descend from a common ancestor, Anna, and if we count down from her, we find that Meurig is one generation older than Cadog, which fits extraordinarily well with the account in the Life, where Meurig marries Cadog’s aunt. Some at least of the genealogical structure may therefore be as old as Lifris’s time if not older. The meeting of St Cadog and Meurig ab Enynni was an incident of importance for Llancarfan since, as Guy notes, Meurig became king at the word of the saint, and that advertised the dependency of Meurig’s descendents, the rulers of south-east Wales, on the authority of the saint’s church. It is often the case in Wales that the earliest documents emerge in the decades around 1100 out a vacuum of earlier evidence, and yet on close inspection they show signs of a complex gestation.

Other 12th-century texts that drew on Llancarfan’s genealogical expertise include Geoffrey of Monmouth, the Book of Llandaf, the Cornish Vita S. Petroci and the Breton Vita S. Gurthierni, all of which are discussed in some depth. A web of circumstantial evidence links the genealogical activity at Llancarfan to the hagiographer Caradog of Llancarfan. Cumulatively Guy’s case that Caradog was involved is convincing, but it must be admitted that none of it is quite proof. Indeed, proof has always been elusive for scholars trying to trace the activities of Caradog beyond the two saints’ Lives that he wrote and Geoffrey of Monmouth’s tantalizing mention of him. What is not lacking, however, is suggestions for things that Caradog might have done, such as authoring part or most of the Book of Llandaf or several of the saints’ Lives of south-east Wales. Guy had now added yet another plausible hypothesis to this list. Caradog was not, however, the only scholar at Llancarfan. Leaving aside Lifris, Guy has convincingly proposed another: Iuthael ab Aeddan, a Welshman who provided genealogical information to the author of the Breton Vita S. Gurthierni in the 1120s or 1130s. The identification of Iuthael with a son of Aidan presbiter Cadoci, who twice witnessed charters alongside Lifris of Llancarfan (p. 90), helps to solidify the line of transmission between Llancarfan and the abbey of Quimperlé where the Life originated. I would add that Iuthael was not only responsible for the genealogy of St Gurthiern but also for supplying details of the saint’s life, for that is how I would understand the phrase Incipit vero conversatio sancti Gurthierni secundum traditionem ejusdem; ejusdem can hardly refer to Gurthiern, who did not tell his own story, nor should it mean ‘about the same person’, i.e. Gurthiern himself, for that would be tautological.

Few aspects of medieval Welsh legend are more tantalizing than the tangle surrounding the Roman emperors Constantine (d. 337) and Magnus Maximus (d. 388), Eudaf Hen, Cynan and the founding of Brittany. Pieces of the puzzle are scattered not just through the Harleian and St Cadog Genealogies but also in the tracts that are the subject of Chapters 3 and 4. The problem looks as though it should be resolvable but the missing links in the chain of fabrication are just a little bit too extensive. Guy has attacked this from various angles in other places and the bibliography promises yet another forthcoming essay on the problem, to which we look forward. The very important move by which Maximianus (who derives from the historical Magnus Maximus) was made the son of Constantine surely arose from reading Historia Brittonum where the two occur next to one another in the list of emperors who ruled in Britain (ignoring confused doublets). The Llancarfan genealogist had already absorbed the idea that the Roman emperors formed a patriline from the St Davids Recension, so nothing could be simpler than to assume that Maximianus was Constantine’s son. We know that the genealogist had Historia Brittonum in front of him because he appropriated its account of Maximianus’ founding of Brittany. Guy is probably right, therefore, in thinking that the joining of these two unrelated emperors occurred at Llancarfan in the 12th century, but it remains slightly troubling that a more distant link between Maxen Wledig and Constantine is already present in Harleian Genealogies, §2, and that the Llywelyn ab Iorwerth Genealogies (see below) give Maxen a son called Custennin.

Chapter 3 moves much further into the unknown. The genealogical collection in Jesus College 20, copied c. 1400, has never been properly analysed. All that was clear till now was that it is obviously complex and composite, is related to other Welsh genealogies but also contains unique material, is rather corrupt, and is clearly older than its sole manuscript. In a clear and convincing analysis Guy shows that Jesus 20 drew on two older collections, one a version of the Gwynedd Collection of Genealogies revised in the 12th century and again in south Wales in the thirteenth, and the other closely related to the St Cadog Genealogies from Llancarfan, behind which can be discerned multiple other sources, the most interesting of them being an important 10th-century act of genealogizing from Glamorgan. Readers will definitely need the diagram on p. 157 which sets all this out clearly.

Jesus 20, though blessed with a tangled prehistory, at least has only one copy. In contrast, the Llywelyn ab Iorwerth Genealogies (Chapter 4) have more than ninety. Most are in early modern genealogical compilations where the text has been broken up and scattered amidst other material. The scale of the textual work that went into this chapter is remarkable. It could not have been done without Peter Bartrum’s pioneering efforts, which Guy acknowledges, but he has turned up a wealth of new copies not seen by Bartrum. Moreover, this is the text least satisfactorily presented by Bartrum in 1966. He considered it to be a series of smaller tractates and so broke it up, giving each a separate title. It is all but impossible to work out the presentation of the genealogies in the manuscripts from Bartrum’s edition. Only now does the shape of this compilation become clear: it is a tract made in Gwynedd in the reign of Llywelyn ab Iorwerth (d. 1240) and very probably by his court poet and minister Einion ap Gwalchmai, whose family comes in for detailed treatment. The lower reaches of these genealogies are very important for understanding the politics of native Wales in the twelfth and thirteenth centuries, especially such matters as marriage ties between the royal dynasties and the fate of cadet branches, and for the first time they begin to document families beneath the level of royalty. Moreover, being contemporary and closely dated, they are likely to be trustworthy for this late period, in contrast to much of the earlier material, which is legendary (though still fascinating for mentalities). As a text, the genealogies present Llywelyn as the head of native Wales, sitting atop a web of kinship that binds all the other royal lines to him, ruling over the landowning families of north Wales, and personifying the Welsh nationhood implied by the legendary sections on saints and heroes. The Llywelyn ab Iorwerth Genealogies thus emerge as a major source for the political and landowning structure of Gwynedd and the parts of Wales under its influence in the first half of the 13th century, and in their political thought they compare with other literary products associated with Llywelyn’s court such as the Dream of Maxen.

Along the way, a path is cut through the wilderness of the early modern genealogical manuscripts. Bartrum had already made a good start, fully acknowledged by Guy, but it is now much clearer who was copying from where. The abbeys of the north-east, especially Valle Crucis, can be seen to be the fount of a great deal of the genealogical sources that passed into hands of the 16th century writers, while the poet, copyist, genealogist and historian Gutun Owain, already a major figure in late-medieval Welsh culture, becomes ever more prominent. Many of Gutun’s manuscripts survive, so it is frustrating that some of the particular ones that were most crucial in this case do not, but their contents can in large measure be established from the later tradition.

In the last chapter, Guy moves from individual collections to tracing the fate of a single genealogy, that of the rulers of Gwynedd, through its development. It is not merely the most intensively worked genealogy in the Welsh tradition but one that broke through elsewhere because of the importance of the dynasty. Guy explores the political agendas that lay behind various attempts to extend the pedigree backwards in time, as well as its influence on English royal pedigrees and those of the Mortimer family. The chapter is a forerunner of work promised for a second, more discursive volume.

After the analytical chapters there comes a ream of ‘supporting material’, largely tables comparing names and spellings across different genealogies, supporting the arguments put forward in the analysis. The decision to put these here was a good one as they are less disruptive to the flow of the arguments here, though readers will have to flip back and forth to find the evidence on which Guy bases his arguments. Finally come the editions, filling more than a hundred pages. There are no translations, an understandable omission since most of the contents are personal names, but readers will need some knowledge of Latin and Welsh to deal with the textual frame and the narrative remarks which some of the texts provide.

Boydell have done a remarkable job in condensing such a huge work into a single volume. Doing so has meant not only allowing the book to be much longer than is usual for the Studies in Celtic History series, but also lengthening and widening the pages, and setting the editions in a smaller font than the main text. Certainly it is a great convenience to have everything between two covers, but the result is a very dense book, and I am not wholly confident that the binding will survive heavy use. Medieval Welsh Genealogy is a case where an ebook may be more convenient than the printed version. The same applies even more strongly to the problem of reference. Indexing these texts full of proper names, variously spelled, poses insuperable problems. Guy has chosen the only realistic option, which is to index names in modernized Welsh orthography, except for those whose modern form is difficult to determine; these are left in the original spelling, but italicized. This decision allows people who know something about Welsh history and literature to find characters in whom they are interested, but it opens a gulf between the index and the edited texts which non-experts may struggle to traverse. There is no help for it, as an index which incorporated all the variants would be impossibly long.

In his introduction Guy cautions against the view that genealogy was omnipresent in medieval Welsh learning, and in particular against the belief that the professional poets carried reams of genealogical lore in their heads, or in such books as they possessed. Appeals to the vague term ‘Welsh tradition’ should be rejected in favour of close study of the extant texts. These reveal a much more modest enterprise: individual acts of genealogizing, scattered in time and space and drawing heavily on earlier work, the result being a fairly tenuous thread of mostly interrelated texts visible from the mid-9th century onwards, and never broadening into an uncontrollable mass of material. The warning is well-aimed. Nevertheless, the many loose ends (which Guy scrupulously notes, by the way) tell their own story. As is the case with other kinds of medieval Welsh writing, the origins of genealogical literature are lost. The Gwynedd Collection that is discernible in the 9th century was already quite a complex text. Much doctrine was already established by then, notably the importance of characters such as Beli Mawr or Coel Hen. The narrative world that must have sustained these genealogies—for they make no sense unless these apical figures had their own stories—is lost or recoverable only in fragments. Moreover, material continues to enter the record in subsequent centuries that looks just as complex, or more so. The Brychan texts, which exist in independent versions as well as forming part of the Jesus 20 and Llywelyn ab Iorwerth collections, exemplify this, for by the time they appear to us the kingdom of Brycheiniog, whose origins they trace, was already defunct. Guy is correct to try to cut nebulous ‘Welsh tradition’ down to size and define it through close study of the extant texts, but there was formerly much more material out there than we have today. The message of our texts is not that genealogical writing was restricted to what we have, but that it usually disappeared. Certain texts that passed out of the hands of their immediate creators, such as the St Davids Recension, thus seem to play a domineering role in the extant tradition, but that is because their sources and congeners did not make it even as far as 12th-century Llancarfan, let alone down to Gutun Owain or ourselves.

Medieval Welsh Genealogy is a major contribution to a vital but neglected field. It bears few of the traces of a doctoral thesis turned rapidly into a book. Elegantly and clearly written, in spite of its almost impossibly complex material it manages to set out a clear and convincing story of the development of this particular way of fashioning the past in medieval Wales, as well as providing solidly grounded editions of the relevant texts.