Collation of the Irish regnal canon

D.P. Mc Carthy, Department of Computer Science, Trinity College Dublin 2.

Tuesday, 22 March 2005

Introduction

Comprehensive examination of Irish chronicles discloses a remarkably broad range of literary compilations such as regnal lists, synchronisms, chronological poems, prose histories and annals. The typical chronological apparatus or framework upon which all but the annals of the kalend tradition are constructed is a standardised list of kings and their reigns to which I shall refer as the regnal canon. This regnal canon was evidently not an Irish innovation since it is known from Greek chronography of at least the third century, and it provided the chronological apparatus for Eusebius’ chronicle of world history. Both Jerome’s translation of this in c. 381 and Rufinus’ subsequent adaption in the early fifth century maintained this regnal canon apparatus, and both of these chronicles were known in early Christian Ireland. While many of the Irish medieval regnal lists recite provincial kings, from about the tenth century on lists are found of the supposed ‘Rí Erenn’, i.e. king of Ireland, or ‘Rí Temrach’ used as a synonym, and it was these that formed the basis for the Irish regnal canon. In its mature version this canon commenced with Ceasair whose death was synchronised with the Flood, and extended to Ruaidhrí Ó Conchobhair †1198 who was considered the last king of Ireland, then followed by six kings ‘with opposition’.

The purpose of this collation is firstly to establish the salient features of this canon, viz. the names of the reigns, their order and regnal years, and secondly to identify the developments that took place in these features over time. To this end the regnal details of a number of the best-known sources have been collated together. These sources are arranged in what I consider to be the chronological order of their compilation as follows:

  1. While Bodleian MS Laud 610 is a fifteenth century MS, the synchronisms on ff. 112r–116v preserve what I consider to be an early eleventh century version of the regnal canon which recites reigns and regnal years from Slainghe m. Deala to Maolsechlainn m. Domhnaill †1022. The synchronisms present a number of very distinctive features as will be discussed below.
  2. The undated anonymous chronological poem Eremon is Eber ard recites reigns and regnal years from Ereamhón and Eber down to Eochaidh Feidhleach, while Flann Mainistreach’s two poems Rig Themra dia tesband tnú and Rig Themra toebaige iar tain, which recite reigns from Eochaidh Feidhleach down to Maolsechlainn m. Domhnaill are dated c. 1056. Thus these poems treat complementary periods and so I consider them to be approximately contemporaneous.
  3. Gilla Cóemáin’s two chronological poems Hériu ard inis na rrig and At-tá sund forba fessa recite reigns and regnal years from Ceasair to Maolsechlainn m. Domhnaill, while his Annállad anall uile recites synchronisms, and all are dated c. 1072.
  4. The Annals of Inishfallen include a canon of the kings of Christian Ireland and their regnal years which is located immediately before the incipit of Laoghaire m. Néill’s reign. This canon recites reigns from Laoghaire m. Néill to Maolsechlainn m. Domhnaill and this section of AI was written c. 1092.
  5. The various recensions of Lebor Gabála recite reigns and regnal years ranging in their maximum extent from Ceasair to Ruaidhrí Ó Conchobhair, and Mark Scowcroft dates the a and m recensions beween 1072 and 1166. The earliest MS of LG is that preserved in Lebor Laigin and is dated to the late twelfth century.
  6. Gilla mo Dubda Ó Casaide’s poem Ériu ógh inish na náemh recites the kings of Christian Ireland and their reigns, and Ó Casaide flourished 1143–8.
  7. Conell Mageoghagan’s source, the ‘old Irish booke’, recited reigns and regnal years from Ceasair to Ruaidhrí Ó Conchobhair, and the pre-Norman section of it was compiled in the early fourteenth century.
  8. Michéal Ó Cléirigh’s compilations Seanchas Riogh Ereann in 1630, Lebor Gabála in 1631 and Annala Rioghachta Eireann in 1632–6, i.e. FM, all recite reigns and regnal years from Ceasair to Ruaidhrí Ó Conchobhair.
  9. Seathrún Céitinn’s compilation of c. 1630 Foras Feasa ar Éirinn recites reigns and regnal years from Slainghe m. Deala to Ruaidhrí Ó Conchobhair. The evidence suggests that Céitinn’s work drew upon and hence postdated Ó Cléirigh’s Seanchas Riogh Ereann.
  10. Roderick O’Flaherty’s Ogygia of 1685 recites reigns and regnal years from Ceasair to Maolsechlainn m. Domhnaill.

Organisation of the collation

The reigns of the penultimate of these compilations, Ó Cléirigh’s FM, have been taken from the electronic edition at the invaluable CELT website www.ucc.ie/celt/ and enumerated as reigns 1–197 for reference purposes. The following orthographic modifications have been made to these names: a) All versions of ‘mac’ have been replaced by ‘m.’; b) All versions of ‘ua’ have been replaced by ‘Ó’; c) All instances of ‘cc’ have been replaced by ‘g’; d) The names have generally been shortened to uniqueness; e) Some names have been restored to their earlier readings, viz. Dathí m. Fiachrach rather than ‘Nath Í’, cf. reign 113. Since it transpires that FM includes all of the reigns of the earlier editions and also those of Ogygia it serves to provide a comprehensive list for the purposes of collation. Immediately following this list of reigns the regnal references have been collated and arranged left to right from the earlier to the more recent. In cases where sources reorder the sequence of reigns this has been emphasised by colouring the reference in red and placing a comment in the Remarks column. The section of regnal references is then followed by the regnal years as a second section, and it will be observed that some sources appear in only one of these sections. The reason for this is that for some sources it is relatively easy to locate the reign in question; for example in O’Flaherty’s poem ‘Carmen Chronographicum Ogygiae’ in Ogygia two lines are regularly assigned to each reign and so these are easy to locate and thus there is no need to identify the reference location. On the other hand some sources such as Flann Mainistreach do not systematically recite regnal years, or in other cases their regnal years closely relate to a series already given, e.g. the regnal years of Lebor Laigin correspond closely with those cited for Lebor Gabála. The regnal references are distinguished at the head of each page by their sigla shown in bold, whereas the regnal year sigla are shown in italic. These reigns and regnal references and years have been tabulated using the table function of Word and they have been arranged to print in landscape orientation in order to give the maximum page width for parallel collation. This collation has been divided into three sub-tables as follows:

Table 1. Kings of Ireland for the pre-Christian era – Ceasair to Eochaidh Feidhleach.

Table 2. Kings of Ireland for the pre-Patrician era – Eochaidh Feidhleach to Dathí m. Fiachrach.

Table 3. Kings of Christian Ireland for the pre-Norman era – Laoghaire m. Néill to Ruaidhrí Ó Conchobhair.

Location of the Incarnation in the regnal canon

Clearly the location of the Incarnation is critical for the chronological accuracy of all events which are dated from that epoch, and the reason for the division at Eochaidh Feidhleach between Tables 1 and 2 is that in the earlier versions of the regnal canon the Incarnation was synchronised with Eochaidh Feidhleach’s reign. Indeed the Laud synchronisms explicitly state that the Incarnation occurred in the third year of Eochaidh Feidhleach’s reign, and both Eremon is Eber ard and Flann Mainistreach’s poems observe this temporal boundary. Note that Eochaidh Feidhleach’s reign, i.e. reign 113, is repeated in both Table 1 and 2 because he is included at the end of Eremon is Eber ard and at the beginning of Flann’s compilation. However later versions of the regnal canon adjusted the location of the Incarnation. For example Gilla Cóemáin is silent in quatrain §118 of his poem Hériu ard inis na rrig wherein he describes Eochaidh Feidhleach’s reign. However in quatrain §25 of his synchronistic poem Annálad anall uile he located the Incarnation 79 years before the death of Conaire Mór, that is in the thirteenth year of Eochaid Airemh, reign 114, who was Eochaidh Feidhleach’s successor. The compilers of LG a and m located the Incarnation in the reign of Edescel m. Eoghain, reign 115, that is two reigns and at least twenty-four years after Eochaidh Feidhleach 3. The compiler of LG b, on the other hand, terminated his continuous regnal succession at Aenghus Tuirmeach, reign 101, that is twelve reigns and some hundreds of years before Eochaidh Feidleach. However in a series of synchronisms, LG §671 in Macalister’s edition, LG b synchronised Eochaidh Feidleach’s reign with that of Julius Caesar and the Incarnation with Octavian 41, from which it is clear that the compiler of LG b located the Incarnation significantly later than the compilers of LG a and m. Thus by the end of the twelfth century there prevailed considerable uncertainty amongst followers of the regnal canon as to the location of the Incarnation.

Later in 1627 Conell Mageoghagan acknowledged his uncertainty regarding the location of the Incarnation in the regnal canon writing in MB p. 47–8 that, ‘Some of our writers affirm that’ the Incarnation occurred in the sixteenth year of Fachtna Fathach, Eochaidh’s predecessor, and then that ‘some of the antiquarists affirm that’ the Incarnation occurred in the reign of Eochaidh Feidhleach. Then just three years later in his compilation Seanchus Riogh Ereann Michéal Ó Clérigh located the Incarnation precisely at the eighth year of Criomthann Nia Nair, reign 121. This location is eight reigns or about 120 years after that of Eochaidh Feidhleach, and accordingly represents a substantial divergence from all the above noted locations and the two alternatives cited by Mageoghagan. This is the more remarkable since in the preface Mageoghagan himself signed as a witness to the compilation. Also about this time Seathrún Céitinn also located the Incarnation in the reign Criomthann Nia Nair, but just four years later in year twelve of Criomthann’s reign. The close correspondence of Céitinn’s location together with other details suggest that he drew on Ó Cléirigh’s Seanchus Riogh Ereann. Finally about fifty years later O’Flaherty in his Ogygia located the Incarnation in the very first year of Conaire Mór, reign 117, that is three reigns and about eighteen years after Eochaidh Feidhleach 3. Thus it emerges that while in the early eleventh century the Incarnation was precisely located at Eochaidh Feidhleach 3, in the ensuing centuries various alternatives had been proposed with the result that by the seventeenth century there was no consensus whatsoever regarding its location.

The reason for the division in Tables 2–3 between Dathi m. Fiachrach and Laoghaire n. Néill is that many compilations observe this division between the kings of pagan and Christian Ireland, e.g. Ó Caiside, AI, Gilla Cóemáin, Lebor Gabála and Ogygia. Finally the regnal references in the annals of the kalend tradition are tabulated under the column headed ‘AD’ which cites their synchronised AD, while the regnal years implicit in their obits are tabulated under the column head ‘’. The annalistic regnal years generally assume that the reign ends at an obit but sometimes it is an abdication, e.g. Niall Frossach, reign 177.

The term used in Irish to identify this regnal canon from at least the late twelfth century was ‘réim rígraide’, see Lebor Laigin’s usage of it at line 3159, however it was already virtually implicit in Gilla Cóemáin’s §151, ‘ar n-árim ardrig Hérenn’; cf. also Michéal Ó Clérigh’s 1630 use of it in SR at p. 21, §59. While this is sometimes translated as the ‘Royal Succession’, I have entitled it the ‘regnal canon’ because as the collation shows most of the regnal names, their order and their regnal years were remarkably stable over nearly seven centuries, cf. Gilla Cóemáin versus O’Flaherty. Thus from the start of the second millennium for the poets and many chronographers of Ireland’s past this list served as a canon to provide a regnal framework for their compilations. However the relatively small number of observed changes in the regnal names, their order and regnal years are all of value to us in that they point to dependency relationships and developments between the different compilations.

 

Sigla, editions and references

The following are the sigla used for the different editions included in this collation, and also the methods used to reference them:

LS

Kuno Meyer, ‘The Laud synchronisms’, ZCP 9 (1913) 471–85, referenced by page and line number.

AI

S. Mac Airt, The Annals of Inisfallen (MS Rawlinson B 503) (Dublin 1951) pp. 42–4, referenced using Mac Airt’s paragraph numbers §345–86.

EE

The poem Eremon is Eber ard of which no published edition exists, so the facsimile in K. Mulchrone (ed) The Book of Lecan vol ii in Irish Manuscripts Comission Facsimiles in Collotype of Irish Manuscripts (Dublin 1937), fac. ff. 33r-v has been used. It has been referenced by its facsimile line number prefixed by a=33ra, b=33rb, c=33va, d=33vb. The earlier copy at ff. 14–15 is missing two verses.

FL

Flann Mainistreach – Best, R.I. & M.A. O’Brien The Book of Leinster vol. III, pp 504–15 for the poems Rig Themra dia tesband tnú §§1–35 and Rig Themra toebaige iar tain §§1–52, which has been referenced by the quatrain number.

GC

Gilla Cóemáin’s poems, Hériu ard inis na rrig §§1–151, and At-tá sund forba fessa §§1–37 which are numbered thus by Peter Smith in his monumental edition, Three historical poems ascribed to Gilla Cóemáin – A Critical edition of the work of an eleventh-century Irish scholar, which is in preparation for publication. See also the editions in Best & O’Brien The Book of Leinster III pp. 471–503.

OC

Ériu ógh inis na náemh in 83 quatrains by Gilla mo Dubda Ó Casaide fl. 1143–8, edited by Macalister in LG v, 540–65. Only his regnal years have been cited in Table 3.

LL

Book of Leinster vol. I by Best, R.I., Bergin, O. & M.A. O’Brien edition of Lebor Gabála whose reigns have been referenced by their line number. This is the earliest edition of LG a and its regnal years are virtually identical with those tabulated under LG.

LG

R.A.S. Macalister’s Lebor Gabála Érenn vols. i–v, (Dublin,1938, –39, –40, –41, –56) referenced by his paragraph numbers. Mark Scowcroft’s two papers Leabhar Gabhála Part I: The growth of the text’, Ériu xxxviii (1987) 79–140, and ‘Leabhar Gabhála Part II: The growth of the tradition’, Ériu xxxix (1988) 1–66 are essential guides for Macalister’s complex edition, and Scowcroft’s sigla LG a, b, c, m have been used rather than Macalister’s R1, R2, R3, Min. Both reign references and regnal year citations have been taken from LG a, but alternatives may be noted under Remarks.

MB

Mageoghagan’s Book, i.e. D. Murphy Annals of Clonmacnoise (Dublin 1896, repr. Llanerch 1993), referenced by page and line number.

SR

Michéal Ó Clérigh’s Seanchas Riogh Ereann of 1630 from P. Walsh Genealogiae Regum et Sanctorum Hiberniae (Dublin 1918) 11–35, referenced by paragraph number.

FF

Comyn, D. & Dinneen, P. (ed), Foras Feasa ar Éirinn le Seathrún Céitinn – The History of Ireland vol. I–iv, ITS 4, 8, 9, 15 (1902, 1908, 1908, 1915). His regnal years have been taken from the CELT edition, with any alternative years cited under Remarks.

FM

J. O’Donovan, Annala Rioghachta Eireann – Annals of the Kingdom of Ireland vols. i–viii (Dublin, 1848–51; repr. N. York, 1966). This has been referenced by the FM year and CELT entry number.

OG

Ogygia, seu, Rerum Hibernicarum Chronologia … (London 1685) by R. O’Flaherty; see also J. Hely Ogygia … (Dublin 1793). The regnal years have been cited from his marginal summary to ‘Carmen Chronographicum Ogygiae’ on pp. 443–60.

AT:

Regnal references and years found in these and other annals have been registered in the ‘Remarks’ column.

Observations and conclusions regarding the collated sources

The following represent my current conclusions (March 2005) regarding the distinctive features and inter-relationships between these collated sources.

LS – Laud 610 synchronisms

MS Laud 610 was written by two scribes, Seaan Ó Cléirigh and Gilla na Naem mac Aedagáin, and immediately after completing transcription of the synchronisms Gilla na Naem wrote, ‘We have in this book of the Ráith everything that we can find in the old book, the Psalter of Cashel, and whoever shall read and understand it, everything that is lacking at the end is supplied in the middle or at the beginning … And if it be not clearly understood, it is not the fault of the book but of the reader’, M. Dillon Celtica vi (1963) 153, cf. Meyer 485 n.1. Thus it is explicit that Gilla na Naem considered these synchronisms came from the Psalter of Cashel. In ‘The psalter of Cashel: a provisional list of contents’, Éigse 23 (1989) 107–30, Pádraig Ó Riain surveyed what is known concerning this work and pointed out that details of the genealogical material in Laud 610 showed evidence of a northern revision of an earlier southern compilation. Consequently he concluded that the exemplar for Laud 610 was a revised version of the Psalter, but he did not arrive at any definite conclusion regarding its original date of composition. Nonetheless, he did acknowledge that Conell Mageoghagan’s assertion in his preface to MB that it was compiled at the instigation of Brian m. Cinnéitigh alias Brian Ború was a possible, though untested, explanation of its origin. However examination of the details of the Laud synchronisms disclose both a Munster emphasis and a compilation date of c. 1014 which thus strongly support Mageoghagan’s assertion concerning the place and date of the Psalter’s compilation. On the other hand the details also show that the synchronisms were not in fact completed, so that Mageoghagan’s further claim that Brian had widely and authoritatively distributed multiple copies of the Psalter is not plausible. Thus it seems likely that the Psalter of Cashel was in the course of compilation in AD 1014 and that it was left incomplete by Brian’s sudden and unexpected death in that year. This corresponds with Bart Jaski's conclusion regarding the last stage of the compilation of genealogies of the Psalter, namely that those of the Ui Thairdelbaig were redacted 'not long after the battle of Clontarf', cf. Jaski 'The Genealogical Section of the Psalter of Cashel', Peritia (2003-4) 295-337:309. The synchronisms in Laud 610 represent therefore the oldest surviving version of this Psalter regnal canon known to me, and their Munster bias and incomplete state suggest that in contrast to the genealogical material they have not been revised. Since this Psalter canon appears to have provided the tradition from which sprang all subsequent compilations except for AI it is important to establish their salient chronological characteristics. An evaluation of the chronology of these Laud synchronisms discloses numerous problems and anachronisms, of which the following provides an indication by means of a summary analysis of the three intervals, Flood to Incarnation, Incarnation to Dathí m. Fiachrach’s obit, and Laoghaire m. Néill to Brian m. Cinnéitigh’s obit. These intervals correspond with the periods covered by Tables 1–3 respectively.

Flood to Incarnation. At LS 472.5 it is asserted that the Incarnation (gein Críst) was at AM 3952 which is indeed the Vulgate AM for this event, establishing thereby that the compilers were working within the Vulgate chronological tradition which placed the Flood at AM 1656. Thus this Vulgate tradition sets 3952–(1656–1)=2297 years reckoned inclusively from Flood to Incarnation. Now as mentioned at LS 472.10 the Incarnation is synchronised to Eochaidh Feidhleach 3, and at LS 471.25 it is stated that the Milesians invaded 1275 years after the Flood, and the summation of all LS regnal years from the first year of the first Milesian king Ereamhón 1 to Eochaidh Feidhleach 3 is 1309 years, yielding the total 1275+1309=2584 years for Flood to Incarnation. Now this figure is in excess of the Vulgate chronology by 2584–2297=287 years, and since in LS the regnal years have been omitted from six reigns the actual discrepancy must have originally exceeded 287 years. While some regnal years may have suffered scribal corruption these corruptions mostly affect minims ‘i’ and fives ‘u’, so that it is out of the question in my view to argue that an anachronism of this magnitude has come about by scribal corruption. I may add that if we ignore the Vulgate AM reference at LS 472.5 and assume that the compilers were working instead in the Septuagint tradition which set the Flood at AM 2242 and Incarnation at AM 5200, between which are 5200–(2242–1)=2959 years reckoned inclusively, then we find that there is a deficit of 2959–2584=375 years, an even greater anachronism. Thus the chronology of the Psalter over this interval was in severe conflict with both Biblical chronological traditions.

Incarnation to Dathí m. Fiachrach’s obit. As noted above the Incarnation is synchronised to Eochaidh Feidhleach 3 and at LS 478.24–9 Patrick’s uenit is identified with the advent of Christianity and synchronised with Laoghaire m. Néill 4. Since Patrick’s uenit is placed by annalistic chronology at AD 432 the obit of Laoghaire’s predecessor Dathí m. Fiachrach is implicitly located at AD 428. Summation of LS’s regnal years between Eochaidh Feidhleach 3 and Dathí’s death yields the total of 505 years, so that there is an excess of 505–428=77 years and again this is much too large to be plausibly explained by scribal corruption. The effect of this excess is to push Dathí’s obit into the early sixth century.

Laoghaire m. Néill to Brian m. Cinnéitigh’s obit. The summation of the LS regnal years from Laoghaire 1 to Brian’s obit yields a total of 568 years whereas the annalistic chronology places Briain’s death at 1014, making a total of 1014–(429–1)=586 inclusive years, so that there is an deficit of eighteen years exhibited by LS. While no regnal years are given for three reigns these otherwise only sum to eight years and are thus quite insufficient to correct this eighteen year discrepancy. Furthermore collation of the forty regnal years cited by LS against those derived from the obits recorded in the annals of the kalend traditions shows that while eighteen are identical, fifteen differ by one to two years, and the seven remaining reigns show differences of from three up to thirteen years. Moreover while the kalend tradition records the successive reigns of Ainmire m. Sedna followed Baodan and Eochaidh jointly, LS reverses this sequence, cf. reigns 155–6. Thus throughout the historical part of the first millennium the detailed chronology of LS, and hence that of the Psalter, which had evidently drawn upon the obits of the annals of the kalend tradition, repeatedly and profoundly conflicted with it.

In summary, over its entire range from the mythological to the historical LS repeatedly shows serious anachronisms and conflicts with both Biblical and historical chronology. The conclusion appears inevitable that the source of these canons, the Psalter of Cashel, was compiled by persons with an inadequate knowledge of and a naïve approach to chronology. This naiveté is well illustrated by the details of their synchronising equations for Christ’s life, for at LS 472.11 his Incarnation is synchronised with Eochaidh Feidhleach 3, and his Crucifixion with Conaire Mór 4, a total of (12–3=9) Eochaidh + 15 Eochaidh Airemh + 5 Eterscél + Nuadha Necht ½+ 4 Conaire Mór = 33½ years. This summation accounts for Nuadha Necht’s unique non-integral reign of one half-year, an idiosyncrasy maintained in every version of the canon collated here. But this incongruent reign then required that every AD interval reckoned from this Incarnation included Nuadha Nectht’s reign and should likewise register this half-year. But in practice of course this reign was simply treated either as zero or one and the half-year always omitted. Note however that in AT p. 39 s.a. AD 34 a lengthy interpolation in Irish in Rawlinson B 502 discusses the Paschal implications of an age of Christ of 33½ in detail. From this it is clear that by the late twelfth century this arbitrary Psalter chronology for Christ was intruding into the annals of the kalend tradition, cf. the section below 'Regnal canon intrusion into annals of the kalend tradition'. The appearance of this detailed treatment of the chronology of Christ’s life in LS together with its parallel synchronisation of the comharba of Patrick point to a substantial ecclesiastical contribution to the compilation of the Psalter. This too corresponds to Mageoghagan’s account of its compilation, cf. Murphy p. 7, ‘Bryan … assembled together all the nobility of the K.dome as well spirituall as temporall to Cashell in Mounster, & caused them to compose a booke … called by the name of the psalter of Cashell’. Indeed the very title ‘Psalter’ itself implies an ecclesiastical context for the work.

FL – Flann Mainistreach’s poems

Flann’s version of the canon commenced at Eochaidh Feidleach and omitted nearly every regnal year with just three exceptions, viz. reigns 141, 189–90, showing thereby that regnal years were recorded in his source. However he reproduced every reign 113–191 identically to that of LS and hence the Psalter, with only two exceptions, viz. at reign 159 he separated the joint reign of Aodh and Colman, and at reign 164 he reversed the names of Conall and Ceallach. Since Flann also compiled other poems reciting regnal series for provincial kings in which he systematically recited regnal years his substantial omission of this element when reciting the kings of Ireland suggests he harboured serious doubts regarding the validity of the Psalter data.

AI – Regnal canon in the Annals of Inishfallen

Since neither the sequence nor regnal years of AI’s canon reconcile with either LS or any of the other collated sources, cf. reigns 155–6, 160–72, I conclude that AI’s canon represents an alternative compilation independent of the Psalter of Cashel. It should be noted also that AI’s regnal canon does not reconcile either with its own annalistic obits for the kings of the reigns.

EE – Eremon is Eber ard

EE extends from a.11 to d.33 and it may be divided into a sequence of reigns a.11–d.7, followed by a recapitulation d.8–33 which recites the following genealogical sequences: Ir’s 22 descendants d.8–15, Eber’s 28 descendents d.16–23, Ereamhón’s 24 descendants d.24–31, and d.32–3 asserts a total of 74 kings of Ireland, ie. 74=22+28+24. This poem appears not to have been originally intended as a complete recitation of the regnal canon but rather a recitation only of the reigns of kings descended from Ereamhón and Eber, as indeed the first line suggests, and so it omitted those of Ir, i.e. kings of Ulster. But the Ulster kings were then interpolated subsequently at inappropriate locations with the result that EE’s sequence does not now correspond with the regnal canon. The recapitulation which now incongruently lists the descendant’s of Ir first, also appears to have been subsequently emended. If EE’s list of kings is broken into three equal sections then the sequence of the first and third sections agree with the regnal canon of LS, but the middle section is greatly disordered.

GC – Gilla Cóemáin’s poems

As mentioned in the introduction Gilla Cóemáin located the Incarnation at Eochaidh Airemh 13, and the summation of all of his regnal years from Ceasair to this year yields 2731 years. This is in excess of the Vulgate chronology for Flood to Incarnation, i.e. 3952–(1656–1)=2297, by 2731–2297=434 years, so that GC’s version of the pre-Christian canon conflicts even more profoundly with Vulgate chronology than does LS. In this interval while GC reproduced the same sequence of reigns as LS some of his regnal years differ significantly, cf. reigns 20, 31, 33, 96, 101, which exhibit differences of +5, +4, –9, +13, +20 years respectively. Since none of these differences are readily explicable as scribal errors they must be regarded as emendations in one source or the other, and a number of considerations lead me to conclude that Gilla Cóemáin was responsible for them.

In the interval from Eochaidh Airemh 13 to Dathí m. Fiachrach’s obit the summation of GC’s regnal years is 537 years, and it is therefore in excess of annalistic chronology by 537–428=109 years. Thus it likewise conflicted more seriously with annalistic chronology than did LS. In particular GC substantially emended four regnal years, viz. reigns 117, 131, 141, 146, exhibit differences of +56, –15, +10, –10 years.

Finally for the interval Laoghaire 1 to Brian’s obit the summation of GC regnal years is 618 which may be compared with the annalistic 1014–(429–1)=586 years, leaving an excess of 618–586=32 years. Thus in contrast to LS whose regnal years are insufficient to complete this interval GC’s regnal years are excessive. In this post-Patrician interval GC emended many regnal data, sometimes bringing them closer to the reigns of the kalend tradition but in some instances introducing substantial errors, viz. reigns 148, 163, 164, 179 exhibit differences of –4, +17, +12, +5 years with respect to annalistic chronology. In summary GC’s emendations to a relatively small number of regnal years generally exacerbated the chronological conflict between his version of the regnal canon and both Vulgate and annalistic chronology.

OC – Ó Casaide’s poem Ériu ógh inis na náemh

This poem recites the reigns and regnal years of the kings of Christian Ireland and so only appears in Table 3. While Ó Casaide has a considerable number of unique regnal years, cf. reigns 152, 158–9, 167–8, 170, 186–7 he also has retained a considerable number of LS regnal years that had been abandoned by GC, cf. reigns 154, 161, 162, 169, 171, 173, 174, 176, 178, 184, 188. It appears therefore that Ó Casaide’s principal source was the Psalter rather than GC. The summation of Ó Casaide’s regnal years from Laoghaire 1 to Brian’s obit is 560 years and so it is in deficit of the interval 1014–(429–1)=586 by 26 years, representing a further deterioration from that of LS.

LG – Lebor Gabála

Examination of LG’s regnal sequence and years shows that they follow GC very closely with the singular exception of reign 131, Conn Cedcathach, whose regnal years LG restored to the Psalter’s 35 from GC’s 20. The remaining substantive differences found in LG relative to GC’s canon were: a) To follow Flann’s three separate reigns, Domhnall and Feargus, Eochaidh and Baodan, and Ainmire m. Sedna, i.e. reigns 154–6; b) To restore Maoilseachlann after Brian, cf. reign 191; c) To extend the canon after Maoilseachlann’s death in AD 1022 with up to six ‘kings with opposition’, cf. reigns 192–7. As mentioned above LG a and m locate the Incarnation at Ederscel m. Eoghain, reign 115, but as I have not been able to resolve satisfactorily their periods for the arrivals of Partholon and Neimidh, cf. reigns 2–3, it is not possible to assess their chronology for the interval Ceasair to the Incarnation. However the summation of the LG a and m reigns from Ederscel 1 to Dathí’s obit is 553 years, and it is therefore in excess of annalistic chronology by 553–428=125 years, a substantial deterioration from GC’s figure of 109 years. However in the interval Laoghaire 1 to Brian’s obit LG a and m have a total of 589 years which corresponds closely with the annalistic 1014–(429–1)=586, so that it appears that an effort was made to bring these into overall coordination. Nevertheless over this interval LG shows similar major discrepancies to GC, cf. reigns 163–4, 179, and adds two of its own, viz. reigns 180 and 184 have discrepancies of –10 and +10 years respectively.

LL – Lebor Laigin

In a number of instances LL suffers from textual omission with the result that some of its reigns are either missing or incomplete, see reigns 32, 44, 95, 181. It appears then that LL represents a summary transcript of recension LG a, and this is also suggested by the remarkably cryptic version of LL’s entries for the post-Patrician period. Regarding its chronology LL exhibits similar characteristics to those discussed above for LG a and m.

MB – Mageoghagan’s Book

From passages in MB dealing with the pre-historic and mythic invasions of Ireland and the appearance of plains, rivers and lakes, it is apparent that textually MB has drawn upon LG. However Biblical regnal series derived from the Psalter synchronisms are also found, e.g. MB 14.1–13, so that it appears that its compiler has conflated material from the two separate sources. As discussed in the introduction in MB Mageoghagan expressed uncertainty as to whether the Incarnation fell in reign of Eochaidh Feidhleach, or in the last year of his predecessor, Fachtna Fathach 16, reign 112. Since the Psalter tradition placed the Incarnation in Eochaidh Feidleach’s reign it seems likely therefore that the ‘old Irish booke’ had placed the Incarnation at Fachtna Fathach 16 and Mageoghagan was remarking the conflict between the two traditions. The summation of the regnal years of MB from Ceasair to Fachtna Fathach 16 is 2264 years, but since the reigns 28 and 63, and also the regnal years of reign 62 are missing from Murphy’s edition, this figure must represent an underestimate. The regnal years recorded by other sources for these reigns are virtually unanimously 4, 16 and 12 years, suggesting that 4+16+12=32 years should be added to 2264, thus 2264+32=2296. Since this total corresponds remarkably closely with the Vulgate total of 3952–(1656–1)=2297, I conclude therefore that the frequent reduction of regnal years observed in MB was done deliberately with the purpose of obtaining congruence with Vulgate chronology. It is noteworthy for example that the work commences with a conspectus listing the Vulgate intervals from Creation to the time of Christ.

In the interval Fachtna Fathach 16 to Dathí m. Fiachrach’s obit MB’s regnal years sum to 468 and are thus in excess of the annalistic 428 by 40 years. For the interval Laoghaire 1 to Brian’s obit MB’s regnal years sum to 610 years and are thus in excess of the annalistic 1014–(429–1)=586 years by 24 years. Thus while neither of these intervals reconcile precisely given MB’s reconciliation for the pre-Christian period it seems more likely to me that these discrepancies may have arisen in either Mageoghagan’s translation or in Murphy’s edition. That said MB’s individual regnal years for the historical period correlate no better with annalistic chronology than do the earlier versions of the regnal canon, and in at least one instance, reign 170, it introduces its own error of +10 years which far exceeds the error of the earlier versions. In summary MB represents a successful fourteenth-century attempt to bring at least the pre-Christian section of the regnal canon into accordance with Vulgate chronology.

SR & FM – Micheál Ó Cléirigh’s compilations

Micheál Ó Cléirigh’s three compilations of the regnal canon in SR, his edition of LG (not collated here) and FM are all identical in structure, so I consider it certain that in 1632–6 Ó Clérigh based FM’s chronological apparatus upon his SR which he had completed in November 1630. SR is preserved in the UCD MS Killiney A16 ff. viii–lii which was written by Micheál Ó Cléirigh with subsequent systematic AM and AD additions by Conaire Ó Cléirigh. It is worth repeating that Conell Mageoghagan was a witness and signatory to the completion of this compilation. It emerges also that Ó Clérigh compiled SR using the MS Br2 of GC, i.e. Brussels Bibl. Royale 4640 (2569–72) ff. 149–59, which appears to be a descendent of MS G = NLI G2, saec. 14–15 c., of which Carney wrote, ‘this MS is the earliest surviving compilation of traditional material after the Book of Leinster, and antedates considerably the Books of Lecan, Ballymote, Uí Maine etc’, J. Carney Ériu 21 (1969) 122–47:123. MS NLI G2 has the additional two quatrains, 80A and 92A, cf. reigns 68 and 87, and also a number of its regnal names and year readings correspond with those used by Ó Clérigh, cf. Adamair, Ionnat Mar and Dati. These and other critical details appear in Micheál Ó Cléirigh’s own transcription of GC’s poem Hériu ard inis na rrig in this same collection, i.e. Brussels 4640 ff. 10–15. In this transcription Ó Cléirigh made a number of deliberate modifications to the regnal years, some of which he repeated in his conspectus on f. 15v immediately following his transcription. The first of these was to reduce GC’s interval of 311 years from Ceasair to Partholon to 278 years, cf. reign 2. In this way he arranged that the Milesians arrived precisely at AM 3500, which synchronism was written out in full by Ó Cléirigh on f. xxiir. Thereafter by arbitrarily increasing a relatively small number of reigns, notably reign 47 from 21 to an unprecedented 150 years, reign 99 from 11 to 17 years, by introducing reign 68 with its 10 years and an inter-regnum reign 27 of 7 years, and by retarding the Incarnation to Criomhthann 8, i.e. reign 121, Ó Cléirigh arranged that the interval Ceasair to Incarnation was 2959 years. This value corresponded precisely with the Septuagint chronology of Flood at AM 2242 and Incarnation at AM 5200, an inclusive interval of 5200–(2242–1)=2959 years. In the preface to his edition of LG Ó Cléirigh had simply stated that ‘We give the computation of the Septuagint for the first four ages of the world …’, cf. O’Curry Manuscript Materials p. 172. Ó Cléirigh justified this with a formidable list of authorities, not one of whom were Irish since the evidence of the Irish chronicles themselves is that Irish chronographers had followed the Vulgate chronological tradition. What Ó Cléirigh had done in fact was to extend the already inflated chronology of GC and to retard the Incarnation until they corresponded with the Septuagint tradition. This introduction of Septuagint chronology was Ó Cléirigh’s own innovation.

In the interval Criomhthann 8 to Dathí m. Fiachrach’s obit Ó Cléirigh continued his emendations of GC, principally by reducing reign 131 from 35 to 20 years and then reducing reign 138 from 27 to 17 years. Chiefly by means of these emendations to regnal years and by his relocation of the Incarnation to Criomhthann 8, Ó Cléirigh located Dathí m. Fiachrach’s obit exactly 428 years after his Incarnation. Finally between Laoghaire m. Néill 1 and Brian’s obit Ó Cléirigh made numerous emendations to GC’s regnal years, notably reign 163 from 30 to 16 years, and reign 164 from 22 to 17 years, with the result that Brian’s obit fell 585 years after that of Dathí and hence at 428+585=AD 1013. This was one year in arrears of the annalistic year of AD 1014. Taken as a whole Ó Cléirigh’s post Incarnation chronology was a considerable improvement over all earlier versions of the regnal canon, but that said it must be emphasised that Ó Cléirigh made no attempt whatsoever to bring the individual regnal years into correspondence with their annalistic chronology, several copies of which he used in his compilation. Indeed several of his emendations exacerbated the chronology of individual reigns, for example his version of reigns 159, 162, 165, 171, 173, 183 and 184 in each case increased the anachronism in their regnal years. Over the historical period AD 428–1014 Ó Cléirigh’s regnal chronology is generally in advance of annalistic chronology by up to 17 years, cf. reign 157, but it is in arrears at reigns 164 and 190, i.e. Brian m. Cennéitigh.

FF – Céitinn’s Foras Feasa ar Eireann

FF has some unique regnal years e.g. reigns 71, 125, 140, 142, and it sometimes substantially reduces long reigns or offers alternatives, e.g. the following cite EE’s regnal years first and then GC/LG regnal years – reign 15: 70 vs. 80 years, reign 26: 50 vs. 77 years; reign 52: 20 or 60 vs. 40 years; reign 101: 30 or 60 vs. 60 years. Judging by these profoundly contrasting alternative regnal years Céitinn himself does not appear to have done any systematic analysis of the chronology of his compilation. This strongly suggests then that his location of the Incarnation in the reign of Criomhthann Nia Nair is likely to have been a consequence of his having had access to Ó Cléirigh’s compilation of SR, completed around the same time.

OG – O’Flaherty’s Ogygia

O’Flaherty’s version of the canon in OG shows many instances of regnal years found elsewhere only in MB which accords with the fact that O’Flaherty annotated the Armagh MS of MB very intensively with many chronological marginalia. Moreover like the compiler of Mageoghagan’s ‘old Irish booke’ O’Flaherty also made a concerted effort to impose Vulgate chronology on his pre-Christian era. This can be clearly seen in the marginal AM and AD chronology that O’Flaherty added to his ‘Carmen Chronographicum Ogygiae’ in which he set AM 1656 beside the Flood and AM 3950 as ‘primus aerae Christian’, cf. p. 443. Following this while there are some small errors in his summation of the regnal years his marginal AM continues to AM 3949 at the reign of Nuadha Necht, reign 116. Then follows the reign of Conaire Mór of which O’Flhaerty wrote, ‘Conarius, quocum coeperat aera Dei’, and accordingly his marginal chronology changes to AD, implying thereby the equation AD 1 = AM 3950 exactly as he had announced at the beginning. This equation is thus just two AM years less than the equation found in the annals and Bede that sets AD 1 = AM 3952, but O’Flaherty offered no explanation for this discrepancy. Nevertheless his marginal AD chronology clearly sets the Incarnation at Conaire Mór 1 and thereafter this continues with some minor summation errors until the reign of ‘Dathyas’, i.e. Dathí m. Fiachrach concludes at AD 428. Thereafter it carries on and sets Brian’s obit at AD 1014 followed by that of Maolsechlainn m. Domhnaill at AD 1022. Thus O’Flaherty’s marginal AM and AD show that he made an independent attempt to construct a version of the regnal canon that closely corresponded overall with Vulgate and annalistic chronology. However like all those who preceded him he made no attempt whatsoever to restore the individual regnal years of the historical period to correspondence with their annalistic chronology in the kalend tradition.

Summary and conclusions

In conclusion, this review of the development of the regnal canon shows that from its very inception as the Psalter of Cashel it was deeply flawed and in profound chronological conflict with both Biblical and annalistic and hence historical chronology. Gilla Cóemáin’s subsequent poetic version substantially exacerbated these chronological conflicts, but nevertheless his poems served as the authority for subsequent compilations, most importantly that of Lebor Gabála. In consequence all of these compilations present deeply flawed Biblical and historical chronology. There were however at least two attempts made to reconcile the canon with Vulgate chronology, the first in the early fourteenth century by the compiler of Mageoghagan’s ‘old Irish booke’, and the second by O’Flaherty in Ogygia in 1685. Micheál Ó Cléirigh on the other hand undertook to reconcile the canon with Septuagint chronology by further expanding the pre-Christian regnal years and significantly retarding the Incarnation. At the same time Ó Cléirigh attempted to make the overall summation of the regnal years over the historical period conform to the annalistic chronology of the kalend tradition, but he did not succeed.

It is truly remarkable that neither Ó Cléirigh nor any other revisor of the regnal canon made any serious attempt to reconcile the regnal years of the individual reigns over the historical period with their annalistic chronology in the kalend tradition. Thus any chronological analysis based upon the regnal canon over the historical period conflicted seriously with the annalistic chronology. Also remarkable is the extraordinary burst of compilation activity in the decade 1627–36 when between them Mageoghagan, Ó Cléirigh and Céitinn prepared a total of five new versions of the regnal canon. It was these late versions, particularly those of Foras Feasa and FM, that have been the most influential in Irish literary and historical studies over the last four centuries. However none of these studies have acknowledged how deeply flawed was the chronology of these sources. Finally it is worth remarking that from the twelfth to the sixteenth century while the regnal canon tradition appears to have dominated much of Irish historiography, in Connacht particularly but also west Munster the kalend tradition was maintained. That is in Connacht various members of the Uí Maoilchonaire, Uí Duibhgeannáin and Mac Fhirbhisigh families either extended annals of the kalend tradition, e.g. the annals of Loch Cé and Connacht, or copied and used earlier texts, e.g. Chronicum Scottorum and the annals of Tigernach. At the same time the annals of Inishfallen were continued in a number of monasteries in west Munster up until the early fourteenth century. Then at the end of the fifteenth century a resurgence of interest in south Ulster inspired Cathal Óg Mac Maghnusa to gather the important annals of the kalend tradition and compile them into what we now call the annals of Ulster.

The Collation

This is tabulated in the three following tables and it must be emphasised that this collation represents work in progress and so it almost certainly contains some errors.

Table 1. Kings of Ireland for the pre-Christian era – Ceasair to Eochaidh Feidhleach.

Table 2. Kings of Ireland for the pre-Patrician era – Eochaidh Feidhleach to Dathí m. Fiachrach.

Table 3. Kings of Christian Ireland for the pre-Norman era – Laoghaire m. Néill to Ruaidhrí Ó Conchobhair.

 

Regnal canon intrusion into annals of the kalend tradition

Examination of all the annals of the kalend tradition that cover the first millennium leads to the conclusion that following compilation of the regnal canon in c. 1014 they rapidly assimilated some of its features. In the pre-Patrician period AT/CS/AU/AI/AB all reproduce to some extent regnal incipits with virtually the same regnal names, sequence and comparable regnal years, cf. the Remarks column for reigns 117–147. Of them all AT preserves the most substantial series of these entries and in places it is possible to see that AT reproduces details of LS and hence Psalter chronology. For example AT sets the obit of Conaire Mór, reign 117, at AD 44 and since it also sets the Incarnation at AD 34 these two reconcile precisely with LS which assigns Conaire Mór a reign of fourteen years and LS 473.4 asserts that the Crucifixion occurred in his fourth year. That is Conaire Mór died ten years after the Crucifixion. In most of these annalistic pre-Patrician regnal incipits the reigns are designated kings of Tara, but in some cases Emain is given, cf. reigns 124, 126, 135.

In the post-Patrician period the intrusion consists principally of the interpolation of the regnal title ‘rí Temrach’ or ‘rí Erenn’, or Latin equivalents, into the obits of most individuals listed in the regnal canon. However no systematic effort was made to introduce the regnal years or regnal incipits and in this way the profound chronological conflict between the two traditions was passed over. In one instance the interpolator of the regnal canon series remarked the then current uncertainty concerning the succession, viz. at AD 644 following the death of ‘Domnaill m. Aed, king of Ireland’, we find in both AT and AU the interjection: ‘Here it is uncertain who reigned after Domnall. Some historiographers state that four kings, i.e. Cellach, Conall Cael and two sons of Aed Sláine, i.e. Diarmait and Blathmac, reigned in mingled rule’. This interjection represents a conflation of the reigns 164–5, and since it appears in both the Clonmacnoise and Cuana annals it must have been interpolated before Cuan Ó Lothcháin’s compilation of c. 1022, and hence within a decade of the compilation of the Psalter of Cashel. Thus I conclude that AT and AU both preserve essentially an eleventh century interpolation of the regnal canon. AI, on the other hand, has abandoned the regnal titles and indeed some of the obits and has instead interpolated an alternative version of the post-Patrician canon immdiately before the incipit of Laoghaire m. Néill's reign. This incongruent and anachronistic interpolation whose reigns neither agree with the obits preserved in AI itself nor with the mainstream regnal canon appears to be the result of the compiler of AI struggling with the two irreconcilable traditions.

Later interpolations of the canon were much more disruptive, as for example the early thirteenth century compiler of Mageoghagan’s ‘old Irish booke’ who systematically intruded a regnal incipit with regnal years for virtually every reign from Cesair to Brian m. Cinnéitigh. In doing this he made no attempt to make the regnal years and kalends reconcile so that in some cases the reigns and obits are impossibly separated. For example, Baedan m. Nindeadha’s one-year reign by the regnal canon appears at MB 88.27, whereas his obit is located many years later at MB 90.6. Since this compiler retained both kalends and obits his compilation presented a confused conflation of the two chronological traditions, which confusion was concealed to some extent by Mageoghagan’s subsequent deletion of the kalends. Finally the compilers of FM abandoned the kalends obit chronology altogether and imposed Micheál Ó Cléirigh’s version of the regnal canon as discussed above. Thus from the eleventh to the seventeenth century a progressively more disruptive intrusion can be seen of the regnal canon tradition into the annals of the kalend tradition.

Acknowledgements

I wish to acknowledge with gratitude the following contributions to this work. Trinity College Dublin for awarding me a Berkeley Fellowship for 2002–3, and the School of Celtic Studies, DIAS, for admitting me as a visiting scholar for 2003–4, during which period most of this work was undertaken. Peter Smith of Magee College, Derry for introducing me to the work of Gilla Cóemáin, for patiently answering many questions, and generously presenting me with a draft copy of his important edition of the chronological poems of Gilla Cóemáin. Bart Jaski of Utrecht for emphasising to me the importance of the Laud synchronisms, for providing me with his annotated copy of Meyer’s edition, and sharing with me his important work on the evolution of the Irish origin myth. Mark Scowcroft of Washington for his wonderful guide to Macalister’s complex edition of LG, and for sharing with me in conversation his deep knowledge and feeling for the text and tradition of LG. Bernadette Cunningham of NUI Dublin for her assistance with identifying the hand of the marginal AM and AD in UCD MS Killiney A 16. Donnchadh Ó Corráin and his team of scholars in UCC who maintain the invaluable CELT website which provided a great deal of the working material for my study of FM and FF. The Librarians at the School of Celtic Studies, DIAS, for their assistance with access to essential printed sources, and to Pádraig de Brún’s splendid collection of microfilms of Irish MSS.