An extensive corpus of genealogies maintained in writing for a thousand years from the seventh to seventeenth centuries has been transmitted to us in the extant Irish manuscript codices, a body of connective material that allows the history of early medieval Ireland to be elucidated with an uncommon clarity and depth. These codices functioned as miniature bound libraries containing collections of laws, genealogies, wisdom texts and narrative tales alongside religious, classical and annalistic texts. The legal, genealogical, wisdom and narrative collections are the product of native secular druidic/poetic schools whose orally transmitted lore was committed to writing in the seventh century, though the making of lists is known to be one of the earliest uses of literacy and so the genealogies may have been recorded in writing from a much earlier date. The importance of the genealogical collections in a society in which political power, legal liability and property rights attached to the extended kin group is obvious as is the existence of a conflicting imperative that they be made to reflect aspiration rather than reality. The extensive crossover of secular and ecclesiastical learning that occurred prior to the thirteenth century repudiation of native secular learning by the Church as it came under greater foreign influence has to some extent mitigated the loss of the earlier writings of the secular schools so that the corpus is transmitted in the first instance in the twelfth century codices of the ecclesiastical schools but thereafter in the codices of the secular schools.
Agnatic lineages are expressed in narrative form, the two main formulae being the genelach ('genealogy') ascending vertically in the form A m B m C m D ('A son of B son of C son of D') and the cráebhcoibneasa ('nearest branching') descending horizontally in the form m A X Y Z ('sons of A: X, Y and Z') where m is the appropriate inflection of the noun mac ('son') for number and case and the personal names of fathers are given in genitive form.
D | C | B | A A_______ X Y Z Ascent Descent
Given in isolation these describe a sparse, one-dimensional genealogical space but when grouped in interconnected series they effectively describe the complex, two-dimensional, triangular space inhabited by an extended family. Brief descriptors of an individual (eg: epithet, office held) are often given but dates are almost always absent as are the names of mothers, wives, sisters and daughters.
The compiler would base his work upon one or more pre-existing genealogical collection. An earlier edition of a genealogy would be located, edited and merged with contemporary oral genealogies captured by the compiler and presented as part of a new genealogical collection updating the pre-existing genealogical scheme. Over time successive collections tended to constrain their size through the extensive abridgement of extraneous portions of the scheme so that the genealogy of earlier groups often survives to us as only a single genelach. Nonetheless, by their copying of earlier materials successive compilers have tended to ensure the survival of the contents of collections now lost and by their geographically asymmetric development of these materials they have tended to ensure an extensive coverage of families of both local and national importance.
Lineages descending from or ascending to a nodal ancestor are gathered as blocks of text under a header to form the genealogy of a named family group. Group names within headers take the form of the dynastic collectives Uí, Cenél, Clann, Síl or surname prefixes Ua, Mac applied to a personal name in genitive case (or in the earliest remnants of the corpus the tribal collectives Dál, Corcu and suffixes -rige, -ne, -acht applied to a person, animal, divinity or occupation) and indicate group descent from an eponym. Note that the eponym is not always set as the nodal ancestor and that there is no neccessary inconsistency in this as the eponym may validly be an ascendant or descendant of the nodal ancestor. The major dynamic to be accommodated within the structure is the segmentation from an older group Y of a newer group Z descending from a more recent nodal ancestor. This is achieved by gathering the segmenting lineages to form group Z and adding a descriptor such as otait Z ('from whom exists Z') or sunn condregait Z ('here joins Z') to the appropriate individual in group Y (often as part of a cráebhcoibneasa) though the intervening lineage is not always made explicit. A compiler would often be unable to capture the genealogy of an extinguished or obscured line leaving a stratum of prematurely terminating lines marking the extent of the earlier edition within his own compilation. The corpus can be seen to narrow at several points as large numbers of these previously traced lines fall into oblivion.
/\ / \ / \ / \ / Y \Segmentation / /\ /_________/ \ Stratum / \ / \ / Z \ / \ /____________\
A genealogy is easiest understood when recast as a chart with names converted from genitive to nominative form and with lines of ascent converging wherever a shared name sequence or a shared earliest name indicate a common ancestry. The lexicon of the Irish genealogist is relatively small and so presents few problems to the determined modern reader.
The material relating to the fifth century onwards is basically sound but is not free from inconsistency and so a genealogy should be controlled against other historical records whenever possible. Some of the inconsistency is the result of scribal error, being the natural degradation found within any corpus transmitted across several centuries by many different hands, but some is arguably the result of the deliberately introduced error of fabrication. While we should never fall into the trap of imposing our own preconceptions upon a historical text, casually assuming that we possess a genius superior to its historical witness, there are in some cases structural clues indicating that a family (with what we must imagine to have been 'unsuitable' ancestry) has been converged vertically into a formerly prominent lineage now obscure/extinct by a fictional continuity or horizontally into a prominent lineage still extant by a fictional brotherhood inserted into the past. The fabrication of such spurious convergence is typically manifest at a point or range in the genealogy rather than generally along its length so we should be careful to recognise that even a fabricated genealogy is capable of authentication within the limits of the group spuriously linked, otherwise the very purpose of fabrication is denied. In examining the structure of a genealogy both deliberate fabrication and scribal error will tend not to have propagated uniformly across the genealogical corpus and even less into other classes of historical record and so a genealogy can be authenticated by comparing it in earlier and later editions, by dating the individuals named within it by cross reference to their appearance in other historical records and by dead reckoning from these dated points using average generation lengths. This approach highlights structural weaknesses in the genealogy caused by:
A better reading may then be conservatively restored. In the genealogical record, then as now, the written account is secondary being driven by a primary oral account and so the quality of the oral account captured and embedded within the written account also requires assessment. Anthropological studies have shown that the near-present portion of an oral genealogy is likely to be historically accurate but that beyond lies an area of ambiguity prone to foreshortening, adjustment and resolution into a single line of fathers and sons. If the interval between successive strata in the written account falls within the historical time of the oral account captured at each stratum then the written account can be said to be the interlinked product of historically accurate oral accounts and so be historically accurate throughout its full length. Given that texts on Irish customary law indicate the operation of kingroups spanning six generations (the indfhine with ingen ar méraib) then at least six generations would appear to constitute historical time for this purpose.
Due to the heterogeneous nature of the corpus the historical horizon can be determined only at the level of the individual genealogy. On this basis the historicity of the corpus is found to diminish steadily beyond the twelfth century upon discovery of much spurious convergence so that just a handful of major lineages remain capable of authentication back into the fifth century. At this point an absolute historical horizon is met as the historical sources fail and we encounter instead the work of seventh century ecclesiastical schools who fabricated a pseudohistory projecting into the distant past known as Lebor Gabála Érenn ('Book of Invasions of Ireland') by which the Irish nation were inserted into the Christian scheme of universal history set out in the Etymologiae of Isidore of Seville within which the peoples of Europe were made to descend through Japhet from Noah and Adam. The genealogical component of this pseudohistory became attached to the corpus at an early date in a scheme that can be seen to proceed from the remote biblical lineages of the Etymologiae through various learned fictions, euhemerised divinities and legendary figures to the handful of prominent individuals known to have existed in the fifth and sixth centuries. The creative process becomes apparent when comparing earlier and later editions of the genealogies to find the earlier reaches adjusted and readjusted to fit the latest version of the pseudohistory. Unfortunately this pseudohistorical scheme is so early and so thorough that only tantalising fragments of a superseded pre-fifth century genealogical scheme can be discerned from careful analysis of tribal and dynastic origin legends embedded within the early narrative literature.
The pre-Norman Irish genealogies, J.V. Kelleher, Irish Historical Studies 16.
The Irish genealogies: their value and defects, K.W. Nicholls, The Irish Genealogist 5 (2).
Kingship, genealogies, and regnal lists, D.N. Dumville, Early Medieval Kingship, P.H. Sawyer and I.N. Wood.
Irish origin legends and genealogy: recurrent aetiologies, D. Ó Corráin, History and Heroic Tale: A Symposium, T. Nyberg et al.
The Irish genealogies as an onomastic source, N. Ó Muraíle, Nomina 16.
Kings, chronicles and genealogies: reconstructing Celtic mediaeval dynasties, D.E. Thornton, Family Trees and the Roots of Politics, K.S.B. Keats-Rohan.
Creating the past: the early Irish genealogical tradition, D. Ó Corráin, Peritia 12.
Orality, literacy and genealogy in early medieval Ireland and Wales, D.E. Thornton, Literacy in Medieval Celtic Societies, H. Pryce.