Chronological Synchronisation of the Irish Annals

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

21 March 2005

 

Edition History

Date

Description

1 Mar. ‘99

First Edition – the first edition of this collation and article accompanied the publication of Mc Carthy, ‘The Chronology of the Irish Annals’, PRIA 98C(1998)203–55, and collated essentially AT/CS/MB/AU over the years AD 307–722.

1 Sep. ‘99

Second Edition – this extended the collation backwards to cover the years AD 1–306, and it included where appropriate AR/AI/AB/CM. The tables were divided into c. 70 Kbyte units to facilitate Web downloading and on-going maintenance.

19 Sep. ‘00

Third Edition – this extended the collation of AT/CS/AU forward to cover the years AD 767–1178, thus fully covering the entire Christian era of the Clonmacnoise group of annals.

20 Mar. ‘05

Fourth Edition – this has extended the collation forward to cover the remaining years of the annals of the kalend tradition, viz. AD 1179–1590, by collating the Connacht group of texts LC/CT with AU. Additionally the following changes have been made to the earlier parts of the collation:

 

a) The misleading title ‘Annals of Clonmacnoise’ employed by Ware, O’Flaherty and Murphy for Conell Mageoghagan’s compilation of 1627 has been replaced by ‘Mageoghagan’s Book’=MB. While MB translates many Clonmacnoise entries it substitutes a comprehensive series of reigns of Irish kings for the kalends apparatus of the Clonmacnoise chronicle. The compilers of FM also employed this regnal canon, and a separate collation of this has been made available online, cf. d) below.

 

b) The annals of ff. 12–14 of TCD 1282, which had in 1858 been associated by J.H. Todd with AT, have been restored to AU to which they belong, cf. Mc Carthy ‘The original compilation of the Annals of Ulster’, pp. 77–84.

 

c) Small adjustments have been made to the locations of the six kalends restored over AD 612–64 in order to improve the relationship between the synchronised chronology and the independent chronology of known events over this interval, cf. the section ‘Restoration of the Chronological Integrity of AT/CS over AD 1–664’ below.

 

d) The regnal canon providing the primary chronological apparatus of both MB and FM has been collated for a wide range of chronicles including the Laud synchronisms, chronological poems, Lebor Gabála and Foras Feasa.

 

Sigla

The following have been used both in the collation and this article.

Siglum

Reference – see Bibliography for details

AB

Annals of Boyle, cf. Freeman Revue Celtique 41–4(1924–7).

AC

Annales Cambriae, cf. Morris Nennius (London 1980).

AD

Annus Domini, i.e. the years reckoned from the Incarnation.

AI

Annals of Inishfallen, cf. Mac Airt AI (Dublin 1951).

AR

Annals of Roscrea, cf. Gleeson & Mac Airt AR (1959).

AT

Annals of Tigernach, cf. Stokes AT (Felinfach 1993).

AU

Annals of Ulster, cf. Hennessy & Mac Carthy, and Mac Airt & Mac Niocaill AU.

B.B.

Brian Boru †1014, alias Brian mac Cenneitigh mac Lorcain.

BOT

Bedae Opera Temporibus cf. Jones (Cambridge, Mass. 1943).

CELT

Corpus of Electronic Texts, cf. the URL www.ucc.ie/celt/

CS

Chronicum Scottorum, cf. Hennessy CS (London 1866).

CM

Chronica Maiora by Bede, i.e. De temporum ratione cap. 66, cf. Jones (1977).

CT

Annals of Connacht, cf. Freeman CT (Dublin 1944).

EB

Encylopaedia Brittanica (1970).

FM

Annals of the Four Masters, cf. Donovan FM (Dublin, 1848–51).

H

Transcription by Ruaidhrí Ó Luinín, cf. Mac Airt & Mac Niocaill AU.

H1

Corrections/additions by Ruaidhrí Ó Luinín, cf. Mac Airt & Mac Niocaill AU.

H2

Interpolations by Ruaidhrí Ó Casaide, cf. Mac Airt & Mac Niocaill AU.

IC

Iona chronicle.

Kł/Kl/K

Kalends of January, i.e. 1 January.

LC

Annals of Loch Cé, cf. Hennessy LC (London 1871).

LG

Lebor Gabála, cf. Macalister (1938–56).

LP

Liber Pontificalis, cf. Mommsen (1898).

MB

Mageoghagan’s Book, cf. Murphy Annals of Clonmacnoise (Dublin 1896).

MC

Bartholomew Mac Carthy, editor of AU ii–iv.

MS/S

Manuscript(s).

MS H

MS TCD 1282 of AU.

MS R

MS Rawlinson B 489 of AU.

M.S.

Maolseachlinn mac Domnaill †1022.

ODCC

The Oxford Dictionary of the Christian Church, cf. Cross et al. (Oxford 1983).

URL

Universal Resource Locator

 

Introduction

The purpose of this article is to describe the principles used in the construction of the tables which synchronise the chronology of the Irish annals. Comprehensive examination of the chronological apparatus of Irish chronicles in general shows that two quite distinct traditions existed. One of these employed the kalend as its primary chronological element, writing it variously Kł/Kl/K and standing for Kalendae Ianuarii, i.e. 1 January. Thus I designate those annals that exhibit this feature as members of the kalend tradition. The second tradition employed an extended series of the regnal years of Irish kings, and hence I designate annals exhibiting this feature as members of the regnal canon tradition. Of these the kalend tradition is by far the oldest and since it dominated the compilation of Irish annals it is the most important for Irish historical chronography. Its origins can be traced to the third century Paschal tract of Anatolius, bishop of Laodicea †c.282 wherein it was combined with the feria or weekday of the kalends of January. The resulting chronological apparatus is thus closely linked to the Julian calendar and weekday cycle and may be succinctly described as a kalends+feria apparatus. The earliest chronicle employing this apparatus has been identified as an annalistic account of world-history compiled by Rufinus of Aquileia †410 in the final years of his life. Rufinus’ chronicle then formed the world-history basis for compilation of annals in the kalend tradition in Ireland, and continuations of it were made in Iona, Clonmacnoise, Armagh, Derry, and on the Connacht shores of the river Shannon. This Irish compilation had commenced by at least the beginning of the sixth century and it continued up until AD 1590, this being the last annalistic year of Brian Mac Diarmada’s compilation of the annals of Loch Cé. On the other hand the regnal canon tradition employed a canon of supposed kings of Ireland extending from Slainghe m. Deala, the legendary first king of the Fir Bolg, continued eventually to Ruaidhrí Ó Conchobhair †1198, regarded as the last high king of Ireland. The first version of the regnal canon was compiled as synchronisms by Irish scholars of the early eleventh century seeking to synchronise Irish history with Biblical and world history, and this was entitled the Psalter of Cashel. Later in the eleventh century its canonical status was established by the chronological poems of Flann Mainistreach and Gilla Cóemáin, and it furnished the chronological apparatus for Lebor Gabála in the twelfth century. In the fourteenth century it provided the chronological apparatus for the ‘old Irish book’ which Conell Mageoghagan translated into English in 1627, and shortly after that in c. 1636 Micheál Ó Cléirigh and his colleagues employed it in their compilation of FM. At about the same time Seathrún Céitinn also utilised it for his Foras Feasa ar Eirinn¸ and finally in 1685 Ruaidhrí Ó Flaithbheartaigh used it for his Ogygia. Thus the regnal canon tradition was substantially an Irish development of the second millennium, and although it came to dominate Irish historical literature of the second half of that millennium it is easy to show that it was profoundly inaccurate chronologically even for events well into the second millennium, cf. the section on FM below. Consequently the importance of the regnal canon tradition lies in its literary influence rather than its chronological, and it was the much earlier kalend tradition that, except for occasional disruptions, maintained an accurate historical chronology. The small number of chronological problems with the annals of the kalend tradition have largely arisen from the omission or interpolation of kalends in the transmission of the texts, together with the absence of an AD from most annals for the entire first millennium. The editors of the modern published editions of these annals have in most cases undertaken to supply a marginal AD for their readers, but in constructing these they have not included in their assessment a proper evaluation of the kalends or other chronological criteria, particularly ferial data. Also in many cases their judgments have been strongly influenced by the AD apparatus of AU, the only annals of the kalend tradition to systematically supply AD for the first millennium. The consequence of this has been widespread chronological conflict and confusion, with the various editions offering ranges of AD dates for the same event sometimes extending over a decade or more, cf. the discussion of Baethéne’s obit below. Nor have the problems been only confined to the first millennium for as we shall see AU’s AD chronology also exhibits serious problems for most of the thirteenth and fourteenth centuries.

In the evaluation of the chronology of the kalend tradition presented here the synchronization of these annals has been done by collating tokens of both their chronological entries, e.g. ‘Kl.ui.’, and their record entries, e.g. ‘Patricius pervenit ad Hiberniam …’, in parallel. Hence repeated references to ‘the collation’ will be found in this article. This process was first described in detail in pp. 215–21 of my paper ‘The Chronology of the Irish Annals’ published in the Proceedings of the Royal Irish Academy in 1998, henceforth cited as ‘Chronology’. This article serves both as an appendix and an update to that paper. The primary aim of this collation was to establish a single chronological structure, based on a critical examination of all the annalistic chronological and record criteria, into which all the annalistic records may be placed. This chronological structure is synchronized to the Annus Domini and feria of the Julian year and in this way a uniform, synchronized AD chronology has been established for all the Irish annals. A secondary aim of the collation was to provide a clear view of the relative content of the major annalistic collections, i.e. to identify as far as possible their common and their unique entries in order to gain insight into their common and individual text histories.

Because the character of the annals changes over time, for example in the early Christian period sustained ferial data is found in the Annals of Tigernach and the Chronicum Scotorum extending only up to AD 655, and because all the collections present lacunae, the basis for the synchronized chronology and the principles of collation must also change over time. I give below a summary of the basis and principles relevant to each important interval of collation:

AD_range

Basis for the synchronized chronology and principles of collation

1–664

The kalends and ferial data of AT/CS, together with a full collation of their entries with those of the other annals allows reconstruction of the chronology of their common source, the Iona Chronicle. However it emerges that thirteen kalends were lost from the IC which must be carefully restored and this process is described in the section ‘Restoration of the Chronological Integrity of AT/CS’ below.

665–766

The chronology is based upon the kalends and common entries of AT/CS/AU, wherever these are available. Because this interval covers the relocation of the IC to Ireland in c. AD 740 it is important to identify the common and unique content of each of the collections, so that nearly all of their entries have been collated.

767–803

Because of lacunae in AT/CS the chronology is based upon the kalends of AU and the entries that AU has in common with MB. It is confirmed by a number of events with independent chronology including eclipses at AD 773 and 788.

804–1013

The chronology is based upon the kalends and numerous common entries of AT/CS/AU, wherever these are available. Because AU’s content diverges from AT/CS over the tenth and eleventh centuries a substantial comparison of their entries has been recorded in the Remarks column.

1014–1192

The chronology is based upon the kalends, feria, epacts, AD and common entries of AT/CS/AU/LC, wherever these are available.

1193–1378

The chronology is based upon the kalends, feria and AD of LC/CT; AU, on the other hand, exhibits both the omission and interpolation of kalends which errors are reflected in its AD, with the result that its chronology is completely unreliable over this interval. These problems with AU’s chronology have been comprehensively documented

1379–1590

The chronology is based upon the kalends, feria and AD of LC/CT/AU wherever these are available.

 

Restoration of the Chronological Integrity of AT/CS over AD 1–664

The paper ‘Chronology’ established that the kalends and ferial data preserved in the Annals of Tigernach and the Chronicum Scottorum over AD 1–655 represent the oldest stratum of chronological apparatus in the Irish annals. These kalends and ferial data thereby offer us the best possible basis on which to recover the original chronology of the early medieval annals. As mentioned in the introduction the earliest known example of this kalends+feria apparatus occurs in the Paschal tract De ratione paschali of Anatolius of Laodicea, cf. Mc Carthy & Breen The ante-Nicene Christian Pasch: De ratione paschali The Paschal tract of Anatolius, bishop of Laodicea, pp. 51, 60–1, 68. This work was cited by Rufinus in the early fifth century in his edition of Eusebius’ Ecclesiastical History, and it appears that Rufinus subsequently employed it in the composition of a chronicle as is discussed in Mc Carthy ‘The Status of the pre-Patrician Irish Annals’, pp. 136–42, 148–9. The same kalends+feria apparatus was employed shortly afterwards by Sulpicius Severus in his composition of his Paschal table, the latercus, and then subsequently in the sixth century S. Columba extended Rufinus’ chronicle, and this extension was continued by his successor abbots at Iona until about the middle of the eighth century, cf. Mc Carthy ‘The origin of the latercus Paschal cycle’, p. 37–44. This compilation is known as the Iona chronicle and in c. 740 this chronicle was brought to Ireland where it became the source for the records of early medieval Ireland in all our surviving annals. The ferial data in AT/CS commence at the Incarnation and they continue with occasional errors in the placing of bissextile years up until AD 424 where a sudden disruption to the ferial sequence occurs, followed by another 208 years of cogent ferial data; this entire sequence is tabulated in ‘Chronology’, Appendix 1, and see pp. 221–3 for the relevant discussion. When the chronological accuracy of this kalends+feria structure was compared with that for those events with an independent AD chronology it was shown in pp. 223–9 that:

1.

Seven kalends are missing between AD 425–431.

2.

Five kalends are missing between AD 612–635.

3.

One kalend is missing between AD 635–664.

4.

The ferial data between AD 1–63 and AD 397–424 inclusive are properly synchronised with the Julian calendar.

5.

From AD 425 onward every fourth ferial datum had been decremented by one, excepting for the seventeen inclusive years from AD 524–540, which are correctly synchronised with the feria of the Julian calendar.

 

The purpose of this section is therefore to use this information to restore the chronological integrity of both AT and CS, and thus to provide a uniform chronological structure for the collation of all the other annals over these years. Up until AD 424 there is no difficulty about this and we simply use the Annus Kalendae defined by counting the series of 424 kalends tabulated in Appendix 1 from 1.1 to 16.4 inclusive, see ‘Chronology’ pp. 245–8. These data are mostly derived from AT, but from 12.28 to 13.22 they are taken jointly from AT and CS, and from 13.23 to 16.4 from CS alone. Immediately following AD 424 we must restore the first seven kalends for the years AD 425–431, and this is then to be followed by the series of kalends in Appendix 1 from 16.5 to 22.17. In this way all of the events from AD 432–612 are restored to their appropriate AD years, cf. ‘Chronology’ Table 5, p. 225.

Regarding the six remaining missing kalends these are clearly not consecutive and the best we may do in these circumstances is to distribute them so as to ensure that those events for which we have an independent AD are restored to that AD year. Between AD 612 and 664 we have ten such events and these are tabulated in Table 1 followed by their independent AD, their un-restored AD, and the difference between these.

 

Event

Independent

AD

Un-restored

AD

AD

difference

Solar eclipse 2 Aug. 612

612

612

0

Isidore’s Chronica Maiora

615

615

0

Volcanic eruption of AD 627

627

623

4

Baptism of Edwin

627

623

4

Lindisfarne founded

635

630

5

Heracleonas reigned

641

636

5

Death of Aidan

651

646

5

Death of Penda

655

650

5

Death of Cumine Fota

661

656

5

Solar eclipse 1 May 664

664

658

6

Table 1. Tabulation of those events in AT/CS between 612 and 664 for which an independent AD is available. Beside this is tabulated the AD of these events in the un-restored kalend chronology of AT/CS, followed by the difference between these.

 

As can be seen from the constant difference no kalends have been lost in the intervals 612–15 or 635–61, but four are missing in the twelve-year interval 615–27, one in the interval 627–35 and another in 661–4. Within these intervals the average error may be minimised by restoring the missing kalends at uniform periods. Distributed uniformly the first four kalends may be restored at three-year periods at AD 617, 620, 623 and 626, the next at 631 and the last at 662. To summarise concisely all thirteen restorations:

  1. Restored kalends nos. 1–7 placed at AD 425–431.
  2. Restored kalends nos. 8–12 placed at AD 617, 620, 623, 626, 631.
  3. Restored kalend no. 13 placed at AD 662.

With these thirteen restorations we find that all the events in the interval AD 612–664 for which we have an independent AD chronology are now restored correctly to these years as may be seen from Table 2:

 

Event

Synchronised

AD

Independent

AD

Independent

Authority

Solar eclipse

612

612

Mc Carthy & Breen

Isidore’s Chronica Maiora

615

615

ODCC

Volcano of AD 627

627

627

Mc Carthy & Breen

Baptism of Edwin

627

627

Bede HE v.24

Lindisfarne founded

635

635

ODCC

Heracleonas reigned

641

641

Enc. Brittannica

Death of Aidan

651

651

Bede HE v.24

Death of Penda

655

655

Bede HE v.24

Death of Cumine Fota

661

661

Walsh & Ó Cróinín

Eclipse of 1 May 664

664

664

Mc Carthy & Breen

Table 2. Tabulation of the synchronised AD against the independent AD following the restoration of six kalends at AD 617, 620, 623, 626, 631, 662. For each event the independent authority is indicated, cf. the bibliography for the details.

As may be seen, with these six kalend restorations all ten events for which we have an independent chronology are restored to their appropriate AD, and this arrangement represents an improvement upon that used in editions one to three of these tables. Whilst these restorations thus achieve precise synchronisation for those events with independently attested chronology, and they also maintain the annalistic sequence of all record entries, it must be acknowledged that there is significant uncertainty in the absolute chronology of the unattested events. Namely in the interval AD 616–26 there is a maximum uncertainty of up to four years in the restored AD for all unattested entries, and for AD 628–34 and 662–3 an uncertainty of one year. At the same time it should be emphasised that these kalend restorations do not perturb in any way the chronological sequence of the entries. I now discuss in turn the specific considerations arising in the collation of each of the annals; see also ‘Chronology’ pp. 229–39 for other relevant details.

 

AT/CS – Annals of Tigernach and Chronicum Scotorum

Since these are the primary witnesses to the chronology for AD 1–664 both their chronological and record tokens have been fully tabulated in the order in which they occur in their primary MS, and, as far as possible, these tokens have been aligned by placing tokens of corresponding events on the same line of the collation. Very occasionally, because one source has re-sequenced the event, they cannot be aligned and in these cases an opinion has been expressed in the ‘Remarks’ column, based on the order of the records in other sources and sometimes textual details, as to which source has re-sequenced the record. As will be seen the re-sequencing is most frequent in CS where it appears to be the result of subsequent restoration of records omitted from the initial compilation. Where it has been found that either of these sources has omitted or interpolated a kalend, then these have been emended in the conventional manner, i.e. either [Kl.] or (Kl.). Following AD 1024 both AT and CS additionally provide intermittent ferial, epactal, paschal cyclic, bissextile and AD data and of these the feria, epacts and AD have been tabulated wherever they occur. After AD 1071 these data are frequently presented with their numbers either partially or fully written in Irish words and in these cases for concision the numbers have been converted to Roman numerals so that they may be readily compared with other parallel series of these data. However this conversion has been indicated by writing it in italics, e.g. at AD 1074 is given ‘Kl.iiii.f.l.x.ix.’, representing AT’s ‘Kl. Enair for cétain .ix. dec fuirti’. This change to a literate rendering in Irish of the chronological apparatus was important, as it remains a recurrent feature of the annals thereafter. It is worth noting too that this change was synchronised to the start of the rule of Tigernach Ó Broin as comharb of Ciarán at Clonmacnoise.

Regarding the practical matter of locating the full-text entry associated with any record token cited from AT, in Stokes’ edition from AD 488 forward the parallel AU entry in the collations may be used as an index since immediately ahead of each kalend Stokes has listed cross references to Hennessy’s marginal AD for AU, see ‘Chronology’ p. 204 for an example. Before AD 358, i.e. for fragments one and two of AT, Stokes’ edition provides no such cross references, so the page numbers from the 1993 Felinfach edition of his work have been suffixed in italic to the token of the first kalend on that page, see for example AD 354. To similarly locate entries in Hennessy’s edition of CS the synchronised AD may be employed as an index using Hennessy’s marginal AD as an approximate guide. Finally, since a chronological feature of the primary MS for CS is Roderick O’Flaherty’s AD annotations, which Henessessy reproduced fully as footnotes to his edition, these have been recorded in the ‘Remarks’ column up until the CS lacuna beginning at AD 723. It will be seen from these that while generally they reconcile fairly well with the synchronised AD, on occasions they are quite erratic, giving insight thereby into O’Flaherty’s ability as a chronographer.

MB – Mageoghagan’s Book alias Ware/O’Flaherty/Murphy’s ‘Annals of Clonmacnoise’

This collection was first given the title ‘Annals of Clonmacnoise’ by James Ware in the mid-seventeenth century, which title was subsequently repeated by O’Flaherty. Then more crucially in the late nineteenth century Denis Murphy, with misgivings, employed it as the title of his edition. Ever since there have been repeated reservations expressed concerning this title which rests simply upon the number of its entries referring to the monastery of Clonmacnoise and its hinterland. But since this collection has completely lost the kalends apparatus which is so characteristic of the annals maintained in Clonmacnoise it is a very poor representative for those annals, and Ware’s title is consequently very misleading. Indeed as is discussed below the work is not in fact annalistic until the middle of the seventh century. In these circumstances I have resolved to abandon Ware’s title and since we know that the work was compiled by Conell Mageoghagan in 1627 I have entitled it Mageoghagan’s Book with the siglum MB.

As mentioned the primary chronological apparatus of this collection comprises a sustained regnal canon which extends from the first Fir Bolg king Slane to Rory O’Connor, and this has been separately collated against the principal chronicles that employ this canon, including the Laud synchronisms, the chronological poems of Flann Mainistreach and Gilla Cóemháin, Lebor Gabála, and FM. This collation is available at the URL:

https://www.cs.tcd.ie/Dan.McCarthy/chronology/synchronisms/Edition_4/RC_trad/RC_collation.htm

As well as this regnal canon MB also has an AD chronology which is inscribed in Arabic numerals in a separate column on the left-hand side of each page. This AD series commences at AD 425 and is intermittent as far 647 but thereafter the text becomes annalistic, excepting for lacunae, and the AD is normally recorded annually and continues to 1408. In his edition of MB Murphy took these AD from the MS TCD 673 and he placed them at the start of a line followed by a period plus hyphen, e.g. Murphy p. 113 ‘722.–’. Notwithstanding my statement to the contrary in previous editions of this article, having since had an opportunity to examine several MSS of MB I now consider that these marginal AD were indeed part of Mageoghagan’s original compilation of MB. Moreover I believe these AD to have been Mageoghagan’s own invention for they nearly always synchronise with the kalends of AT/CS and when MB has lacunae the AD often does not register it, cf. AD 725–7. Thus it appears that the ‘old Irish booke’ had still retained kalends which Mageoghagan counted as AD years and then used to locate his marginal AD. At other times Mageoghagan evidently missed kalends so that his marginal AD fell in arrears, cf. AD 762–69. As well as this his marginal AD is often conflict with AD data embedded in the entries, cf. Murphy p. 103–4 ‘642.– … The battle of Ossve … anno 625’, which entry is synchronised at AD 650. In the collation these AD have been cited from Murphy’s edition for the years AD 723–973, and it will be seen that they are practically all inaccurate, some seriously so as this example shows. However some of these errors may be restricted to MS TCD 673 or may have been introduced by Murphy; only an edition of the text which comprehensively collates all of the MSS of MB can resolve this question. While MB’s AD series is of no value whatsoever for chronological purposes, it does provide an efficient means of indexing the only published edition of MB from the mid-seventh century onwards.

Regarding the content of MB, over much of the first millennium its record entries are found to be virtually identical in semantic content and sequence to those of AT/CS. Consequently MB does provide insight into the original content and arrangement of the record entries of the Clonmacnoise chronicle. Another important consequence of the close correspondence of the sequence of MB’s records to those of AT/CS is that the synchronised AD may be used to provide an accurate AD chronology for it. To locate the full text of these tokens in Murphy’s edition of MB, the intermittent page references to records common to AT, CS and MB are given in italics following the ferial data up to AD 722. For AD 723–973 the marginal AD from Murphy’s edition have been entered in italic opposite the appropriate kalend. Collation of MB ends at present at AD 973 but since MB subsequently clearly preserves translations of many records found in AT/CS/AU/LC/CT it would be of great interest to extend this collation.

AU - Annals of Ulster

The pre-Palladian annals of AU were all written by Ruaidhrí Ó Luinín in ff. 12–14 of MS TCD 1282 where they now commence acephalously at AD 81 and are truncated at AD 387. Since the foliation of this MS was done by James Ware in the first half of the seventeenth century it is apparent that at that time eleven folios preceded f.12 and so it is virtually certain that originally AU commenced at or near Creation, cf. Mc Carthy ‘The original compilation of the Annals of Ulster’ p. 92. The surviving pre-Palladian annals were first published by Stokes who, following Todd, designated them the ‘Dublin fragment of Tigernach’, but subsequently Mac Airt realising that they were indeed part of AU included them in his edition and they were subsequently published by Mac Niocaill. Their inclusion was strongly attacked by at least one reviewer of the edition but a recent evaluation of the codicological evidence has emphatically confirmed Mac Airt’s judgement, cf. Mc Carthy, ‘The original compilation of the Annals of Ulster’ pp. 77–84, 87–93. In the edition of Mac Airt and Mac Niocaill they were given paragraph numbers §1–206 and in the collation these numbers have been used prefixed by the tag ‘U:’, e.g. U:1 is the obit of Iriel Glunmar at AD 81. It may be noted that many pre-Palladian entries from AU are closely matched by a corresponding entry in AB.

In the post-Palladian era we must clearly distinguish between the chronological apparatus written by Ruaidhrí Ó Luinín which is registered as H or H1 in the Mac Airt and Mac Niocaill edition, and the many subsequent chronological interpolations by Ruaidhrí Ó Casaide which are registered as H2. In his preface to this edition Mac Niocaill mistakenly identified H2 with the compiler, Cathal Mac Maghnusa, cf. AU p. ix and Mc Carthy ‘The original compilation of the Annals of Ulster’ pp. 73–7 for Ó Casaide. Since it is easy to show that many of Ó Casaide’s chronological interpolations in MS H are erroneous they have all been omitted from the collation. Subsequently Bartholomew Mac Carthy, mistakenly believing these interpolations to be the work of the compiler Cathal Óg Mac Maghnusa and trusting their accuracy offered his own correction to the chronology of AU over AD 431–1013 and made in turn his own interpolations to the chronology of AD 1169–1543. In particular Mac Carthy believed that a nineteen-year epactal series had formed the primary constituent of the chronological apparatus of AU and so he relentlessly interpolated synchronised epacts together with feria for AD 1169–1543, regardless of the MS evidence. These epactal interpolations are nearly all misguided and have therefore also been omitted from the collation. In fact a synchronised epactal series is found in MS H of AU only for AD 1014–1192, and indeed it is clear that epacts were recorded contemporaneously by annalists only for the two centuries following AD 1022.

When we try to collate AU’s records against those of AT/CS over the fifth and sixth centuries we repeatedly find that they occur either under different years or in a different sequence within a year. In these circumstances it is quite impossible over these centuries to collate AU’s record tokens in parallel with those of AT/CS. Fortunately Mac Airt and Mac Niocaill in their edition of AU have numbered the entries appearing within each year, and using these together with the MS AD we may uniquely identify each record. For example, AU 432.1 references the record ‘Patricius peruenit ad Hiberniam’. Note that if we are to locate the entry accurately in AU it is essential that the MS AD be used; the marginal AD notation supplied by Mac Niocaill does not run serially since it omits the number ‘488’. By using these MS AD in this way as numerical record tokens, we may simultaneously identify each record in AU and indicate precisely its location in AU’s chronological structure. Furthermore we may re-position these tokens so that they align with the corresponding token in AT/CS. This is effective because many of the records found in AU up to the tenth century are also found in AT/CS and in these circumstances there is no need to reproduce the textual version of the record token since it already appears under one of AT, CS or MB. Of course, in the event that the record is unique to AU we reproduce both the numeric and the textual record token. In this way we are able to show simultaneously, for all AU records, their chronological position in AU and their chronological relationship to that of AT/CS.

AU’s chronological apparatus as written by Ó Luinín in TCD MS 1282 typically consists of the words ‘Kł.Ian’ or ‘Kł.Ien’ followed by a blank space followed by the AD year in Roman numerals; for example, the record of ‘Patricius peruenit’ is preceded by the chronological apparatus:

Kl Ien. Anno Domini .ccccº.xxxº.iiº.

For most records over the years AD 431–56 AU’s AD chronology corresponds with the synchronised AD, but over the subsequent years AD 457–572 the relationship repeatedly changes and it is particularly perturbed over AD 493–536. In some cases displacements of up to eight years are found, cf. AU 526.1 at AD 518. At AD 573 AU restores the first of the six missing kalends and restores four more at AD 637, 642, 647 and 654 respectively. As a consequence not until the sixth kalend is restored to AT/CS at AD 662 does a stable chronological relationship between AU and the synchronised chronology emerge. However because AU restores only five of the six missing kalends its AD chronology remains in arrears by one year for AD 663 to 1013. This asynchronism is then rectified by AU’s omission of the year ‘1013’ from its AD series. It should be emphasised that the custom of incrementing AU’s MS AD by one year over this range to try to correct its chronological errors, which practice was introduced by James Ussher and endorsed by many subsequent authors including myself, is hazardous and should be strongly discouraged. In particular my rationalisation of this custom as an ‘Annunciation AD’ in ‘The chronological apparatus of the annals of Ulster AD 431–1131’ was mistaken and must be abandoned. The custom can exacerbate the error as for example in the interval AD 574–616 where AU’s AD is in fact one year in advance, and so the further increment then increases the error to two years, cf. the example of Baethéne in ‘Applications of the Collation’ below. The reality must be faced that in practical terms AU’s AD chronology is not reliable until AD 1014, and as we shall see further serious problems arise with it over AD 1192–1379.

If the order of the AU record tokens within each year is examined, starting from the middle of the sixth century, it will be seen that they regularly appear in counting sequence, viz. 1, 2, 3, …, showing that records that AU has in common with AT/CS/MB occur in identical order in all these collections. This correspondence is in fact first noticeable as soon as the disturbed chronology of AU over AD 493–536 ends. For example, at AD 569 AT/CS have four records in common with AU, viz 1) Iug. Fergus, 2) Aennu q., 3) Ite q., 4) Gillas [q.], which occur in this identical order in AU 569.1–3. It is evident that this correspondence in the ordering of their common records is the result of their having drawn from a common source, and the correspondence is strongly sustained until the early tenth century, though even after that a relationship is discernible up until AT ends. For example at AD 752 AT and AU have eleven records in common, all ordered identically, at AD 870 CS/MB and AU have five records in common, all ordered identically, and at AD 1173 AT and AU have three records in common, all ordered identically. In these circumstances the years at which this ordered relationship breaks down are of interest, and I here note two of these:

1. At AD 573 AU restores its first kalend, together with a unique record of the meeting at Droma Ceat, cf. ‘The Chronology’ p. 231, and its records for the subsequent year AD 574 are in complete disorder, viz. AU 575.4,2,1+3. It appears then that the interpolation of this kalend briefly disrupted the compiler’s systematic copying of entries in the order of their occurrence in his exemplar.

2. At AD 913 the four records in common with CS/MB appear in AU as AU 912.2,6,4,1, and following this year the textual and semantic relationship of entries in common between AT/CS/MB and AU become much more complex, viz. substantial differences, including contradictions, regularly occur. It is clear then that AD 913 marks an important horizon in the process of the compilation of AU.

In Mc Carthy ‘The chronological apparatus of the Annals of Ulster AD 82–1019’ pp. 278–81 I have argued that AU’s ‘Liber Cuanach’ attributions refer to a compilation completed in c. 1022 by Cuan Ó Lothcháin †1024 in which he introduced both systematic epacts of Dionysiac origin and also AD into annalistic chronology. Since this required the restoration of the kalends lost from the Iona chronicle I believe that Ó Lothcháin was responsible for the restoration of the five additional kalends found at AU 574, 634, 640, 647, and 654. However by omitting to restore the sixth kalend between AD 661–4 his AD fell one year in arrears until it was corrected at AD 1013 by a subsequent compiler. I consider it likely therefore that Ó Lothcháin was also responsible for the revision and reordering of the entries in AU for AD 574 and 913–1013.

A later phase of compilation is visible in AU’s apparatus over AD 1193–1378 where seven kalends were omitted at AD 1223, 1263, 1266, 1272, 1286, 1371 and 1373, and two kalends were interpolated at AD 1193 and 1314. These errors result in AU’s AD eventually falling five years in arrears which was then emended at AD 1379 by incrementing its AD by an additional five years. It should be noted that Ó Casaide repeatedly employed these erroneous AD when interpolating many entries over these years with the result that his interpolations are also displaced by up to five years. Mac Carthy introduced a subsidiary AD apparatus encased in parentheses which corrects some but by no means all of these displacements, e.g. the obit of Amhlaim Mag Uidhir in AD 1306, cf AU ii p. 418. Moreover examination of MS H shows that over AD 1343–78 the actual AD series written by Ó Luinín was very erratic and it has been substantially and silently emended by both Ó Casaide and Mac Carthy. In summary the AD chronology of AU is both complex and hazardous and it should not be used without reference to the chronology of AT/CS for AD 1–1013 and AT/CS/LC/CT for AD 1014–1543.

 

AI - Annals of Inishfallen

The considerations when collating the records of AI are similar to those of AU inasmuch as at least up to the seventh century it contains very few entries not already found in AT/CS. Thus numerical record tokens, analogous to those described for AU and derived from Mac Airt’s edition, have been used, however the following points should be noted. Up to and including the arrival of Patrick, Mac Airt used the serial numbers §1–391 to uniquely label all the pre-Patrician records, so for these records I have used these numbers prefixed by the tag ‘I§’ to identify AI’s records. In the case where there is more than one record at a serial number, then these are distinguished serially by the extensions ‘.1’, ‘.2’ etcetera; where there is only one record this extension is omitted for the sake of compactness. For example, Jerome’s death at AD 417 is identified simply as I§340, whereas the synchronism at AD 432 on the death of Connculaind is identified as I§389.5, being the fifth record at I§389. In the post-Patrician era Mac Airt supplied a marginal AD chronology based on that of AU, but he faced great difficulties due to the inherent chronological corruption in AI and its conflict with that in AU. However, as these AD are the only practical means of indexing the post-Patrician records in Mac Airt’s edition they have been used, analogously, with ‘I:’ prefixed. Thus at AD 456 the token I:453.1 refers to the imperial obit ‘Marciani mors qui regnauit annis .uii.’, which appears in Mac Airt’s edition as the first entry against his marginal AD of 453. It should be noted that in his endeavour to synchronise with AU's chronology Mac Airt followed the scheme published in 1939 by Paul Walsh, ‘The dating of the Annals of Inishfallen’, p. 677–82. In this Mac Airt freely both interpolated and omitted kalends, as Table 3 illustrates:

Mac Airt’s AD

Interpolated

Omitted

AI 473–480

6 kalends

 

AI 482–485

2 kalends

 

AI 489–491

1 kalend

 

AI 509–511

1 kalend

 

AI 519–521

1 kalend

 

AI 560A-B

 

2 kalends

AI 649–652

2 kalends

 

AI 667–669

1 kalend

 

AI 711–713

1 kalend

 

AI 744–747

2 kalends

 

AI 747a

 

1 kalend

Totals

17 kalends

3 kalends

Table 3. A tabulation of the kalends interpolated and omitted by Mac Airt over AD 477–747 in his edition of AI in his endeavour to align its chronology with that of AU, based upon Walsh (1939).

 

This table shows clearly that Mac Airt’s marginal AD cannot be regarded as an accurate count of the kalends in the MS, and given this degree of chronological distortion there appears to be little to be gained by tabulating the kalends of AI against those of AT/CS. Accordingly they have nearly all been omitted with exception of a few kalends between AD 433–454 which have some ferial and epactal data. These AI kalends have been aligned with the corresponding kalend from AT/CS and prefixed with their location in Mac Airt’s edition; for example, ‘I:440 K.ii.f.xii.l.’ is a kalend with feria .ii. and epact .xii. found at AI 440 and is synchronous with AT/CS at AD 440. The textual evidence suggests that these fifth century ferial and epactal data in AI date from the eleventh century. From the mid-eighth century onward AI increasingly registers the affairs of Munster and since these records generally find no counterpart in the other annals they have not been included in the collation.

AR/AB/FA – Annals of Roscrea and Boyle and the Fragmentary Annals

Collation of these collections has been done along the lines described above for AI. That is, the serial paragraph numbers provided in their published editions have been used with a single letter tag prefixed to the serial number to indicate the source from which it comes, cf. Table 6. Of these additional collections the only one to preserve anything of the chronological apparatus of the Iona chronicle is AR, which generally retains kalends, and between the years AD 445–9, 571–7, 583–8 and 596–601 also preserves ferial data. These are clearly cognate with the ferial data of AT/CS and in the collation these data have been included in a manner analogous to that of AI described above. e.g. at AD 445 ‘R:9 K.1.’. It is also noteworthy that notwithstanding its late date of 1641 AR preserves some remarkably early orthography, e.g. AR §76 ‘Quies Coluimbe Cille’ is the only annalistic obit of Columba to preserve both the Latin ‘b’ and final genitive ‘e’.

 

FM – Annals of the Four Masters

In the collation no attempt whatsoever has been made to include the entries of FM, and since FM is popularly regarded as epitomising the Irish annals it will be as well to explain the reason for this omission. As was mentioned in the introduction FM like MB belongs to the regnal canon tradition, but over the first millennium both have drawn most of their records from annals of the kalend tradition. When transcribing these the original compiler of the ‘old Irish booke’ of MB simply interpolated the regnal incipits of the regnal canon leaving the kalends and actual sequence of the records of his exemplar virtually undisturbed. Subsequently it was Mageoghagan who abandoned the kalends, replacing them instead with his marginal AD as discussed above. However the compilers of FM on the other hand repeatedly resequenced the records when transcribing them. As a result these transcribed records in FM are found to have random displacements right into the historical period. In order to assess the extent and amount of displacement I have made two separate surveys of FM. In the first of these I compared the chronology of a sample of records found in common in annals of the kalend tradition and in FM at about fifty-year intervals over AD 450–1000, and the results of this are given in Table 4. As can be seen the displacements range from FM being six years in advance to five years in arrears, and these displacements are both substantial and erratic at least into the tenth century, and they persist at least into the eleventh century.

AD

Common event

FM

D

453

Catrainodh Laegaire

453

0

502

B. Iubair

500

2

552

Feargna R. Ulad

551

1

600

Comgoll q.

600

0

644

Cuanu Cailchin m.

650

–6

702

Muiredhach do ég

700

2

755

Combustio Cluana M. Nois

750

5

805

Cuana ab Mainistrech

800

5

852

Lucht ocht fichit long

850

2

905

Mael Ciarain q.

900

5

952

Flann h. Becain m.

950

2

1001

Maolpoil q.

1000

1

Table 4. Tabulation of the synchronised AD of events found in common between annals of the kalend tradition and FM at about fifty year intervals between AD 453–1001. FM’s AD and its difference D=AD–FM are tabulated separately.

The second survey compared the AD chronology of every record found in common between annals of the kalend tradition and FM over the thirty-five years AD 562–596, and these are tabulated in Table 5.

AD

Common event

FM

D

 

AD

Common event

FM

D

562

Mo Laisse ob.

563

-1

 

581

M. Fearadhaigh

582

-1

563

Diarmait occ.

558

5

 

582

Q. Feargusa

583

-1

563

Colman R. Lagen

576

-13

 

582

Mo chaeme

584

-2

565

B. Domnaill

561

4

 

583

Q. Maic nisse

589

-6

567

B. Demain

565

2

 

584

Baedan m. Nindeadha occ

567

17

568

B. Ainmireach

566

2

 

585

C. Bhealaig

572

13

569

Iug. Fergus

568

1

 

585

Daigh ob.

586

-1

569

Aenu q.

569

0

 

586

Q. Cairillan

587

-1

569

Ite Cl. Credil

569

0

 

586

Q. Senaigh

587

-1

570

Maenu q

570

0

 

586

Guin Aedh Dubh

592

-6

571

Occ. Baetan & Eoch.

563

8

 

587

Q. Aedha m. Bric

588

-1

572

Brenaind q.

571

1

 

587

Aedha m. Brenuinn

585

2

572

C. Femin

571

1

 

588

Feidlimidh mor.

586

2

572

C. Tola

571

1

 

588

C. Muighe

586

2

573

B. Conaill

572

1

 

590

Ob. Lughdach

588

2

574

Brenaind m. Br. Ob

573

1

 

591

Gregorius s.

590

1

575

Aed m. Eachach

574

1

 

592

C. Eudhuind

590

2

575

Q. Brenaind Cl. Ferta

576

-1

 

592

I. Seanchain

590

2

576

Q. Etchen

577

-1

 

593

Aed Cerr q.

591

2

577

Cairech q

577

0

 

593

Q. Colum Cille

592

1

577

Feargus Scandal r.

580

-3

 

595

B. Cumascaigh

593

2

578

C. Droma

579

-1

 

595

C. Sliebe

593

2

579

Aed R. Ulaid

581

-2

 

595

B. Tibruide

593

2

579

Baedan R. ob.

585

-6

 

596

cec. Aedh m. Ainmireach

594

2

Table 5. A tabulation of the synchronised AD of every record found in common between the annals of the kalend tradition and FM over AD 562–596. FM’s AD is shown separately and its difference D=AD–FM. The obits of kings of the regnal canon are shown in bold.

 

A glance at the difference column D of this table shows the severity of the chronological conflict between FM and annals of the kalend tradition. Out of forty-eight records thirty-four are displaced by one or two years and ten more displaced by up to seventeen years in arrears and up to thirteen years in advance. Only four records are synchronised and in these circumstances FM’s AD chronology must be regarded as absolutely unreliable for historical chronology. In terms of relative chronology examination of the thirteen instances where the kalend tradition groups two or more records in a single year, FM agrees with only four of these, viz. AD 572, 588, 592, 595, always locating the group in arrears. Similarly at AD 583–5 where AT/CS/AR/AU/AB all agree in locating the obits of Mac Nisse and Baedan m. Nindeadha and Cath Bhealaig in three consecutive years, FM resequences all three and distributes them over twenty-three years, viz. FM 567–589. In particular FM has substantially displaced the obits of kings of the regnal canon, here highlighted in bold, cf. Diarmait by 5, Domnaill by 4, Baetan and Eochaidh by 8, and Baedan m. Nindeadha by 17 years. As a result Baedan m. Nindeadha’s reign is reduced from thirteen to one year, and Aedh m. Ainmireach’s reign extended from twelve to twenty-seven years. Furthermore the sequences of the obits of Ainmire m. Sedna and Baedan m. Nindeadha have been reversed. In a word FM’s chronology is in chaos.

Examination of the collation of the regnal canon from synchronisms, chronological poems, Lebor Gabála, MB and FM available at the URL:

https://www.cs.tcd.ie/Dan.McCarthy/chronology/synchronisms/Edition_4/RC_trad/RC_collation.htm

shows that these displacements of regnal obits in FM were in fact a consequence of substantial errors in the chronology of the regnal canon itself. Namely every source collated assigns Baedan m. Nindeadha a reign of one year, and Aedh m. Ainmireach a reign of between twenty-three and twenty-nine years. While these two instances represent the most substantial errors found, examination of the full regnal canon collation shows that discrepancies persist down to and including the reign of Brian m. Cinneitigh, i.e. Brian Ború †1014, cf. FM 1013.1. Thus we are obliged to conclude that the chronological structure of the regnal canon throughout the entire first millennium was absolutely incongruent to the chronology of the regnal obits in the annals of kalend tradition. It was as a result of imposing the chronology of the regnal canon upon the records taken from their kalend tradition sources that the compilers of FM randomly displaced most of these records. Hence examination of FM’s chronology gives no insight whatsoever into historical chronology and only shows us how profound was the chronological conflict between chronicles of the kalends and regnal canon traditions. It is of course straightforward to correct the chronology of all FM records found in common, but for the many unique records found only in FM we must assume that they too have been subject to the same randomising dislocations by their compilers. Consequently unless we have other independent chronological evidence we cannot correct the chronology of these unique records and must accept that the error margin on FM’s AD chronology is of the order of two decades. Moreover Table 5 shows that FM has repeatedly resequenced records with the result that one cannot use these records to safely argue from cause to effect. These then are some of the reasons why FM has not been included in the collation of synchronised annals.

CM – Bede’s Chronica Maiora

As discussed in detail in ‘The Status’ pp. 116–130 there are good reasons to believe that the pre-Patrician sections of AT and AI and Bede's Chronica maiora (CM) all derive from a common source, which I have identified as the chronicle of world history compiled by Rufinus in the early fifth century. The structure and content of this source is best preserved in AT and AI, and in order to show explicitly the relationship between AT, AI and CM, all the entries found in common have been collated using the serial paragraph numbers of Jones’ edition of CM, prefixed by the tag ‘C:’, cf. Table 6.

There is however a secondary stratum of this world history material found in the Clonmacnoise group which is best preserved in AT. If the Byzantine imperial regnal series commencing at Anastasius who reigned AD 491–518 is examined it will be found that AT incorporates a copy of CM’s imperial series extending to Leo whose rule commenced in AD 717, and in AT this secondary material has largely replaced the original imperial series of the Iona chronicle. However in this transcription the third ‘Constantine’ and his seventeen-year rule was omitted, i.e. Constans II Pogonatus who ruled AD 641–668, with the result that all subsequent transcribed regnal and other entries fell back by about seventeen years, cf. AT and AU at AD 673. This displaced material also appears regularly in MB but most of it has been removed from CS, however occasional survivals such as the reign of ‘Pilipus’ at AD 699 show that this stratum was in the source used by the compiler of CS. AU, on the other hand, preserves the original imperial series from the Iona chronicle and has none of this secondary stratum from CM.

AC – Annales Cambriae

This chronicle includes a considerable number of records of Ireland’s important saints such as Patrick, Brigit, Columba and Ciarán, as well as records of natural phenomena including eclipses, comets and plagues, all of which are found in the Irish annals. Thus it is clear that AC has taken these entries from an Irish source and hence all common records have been collated and their details suggest that this source derived from the tradition of Liber Cuanach. For example in AC Brigit’s natus is retarded by fifteen years to 454, cf. AU 456.1 and AI 455, and Patrick’s obit is retarded to 457, cf AU 461.2. As well as this in the earliest manuscript, Harleian MS 3859, exactly 533 years are tabulated which is just one complete Dionysiac cycle plus one year, so that AC like Liber Cuana also reflects a Dionysiac tradition, cf. Mc Carthy, ‘Annals of Ulster AD82–1019’, p. 264–9.

Reference and textual details

The complete list of tags and the edition with which the serial number is associated is shown in Table 6:

Tag

Source

Edition

Example

A

AC – Annales Cambriae

Morris

A:454 at AD 439

B

AB – Annals of Boyle

Freeman

B:146.1 at AD 440

C

CM – Bede DTR cap. 66

Jones

C:545 at AD 643

F

FA – Fragmentary Annals

Radner

F:22 at AD 663

I

AI – Annals of Inishfallen

Mac Airt

I:434.1 at AD 434

R

AR – Annals of Roscrea

Gleeson & Mac Airt

R:91 at AD 445

U

AU – Annals of Ulster

Mac Airt & Mac Niocaill

U:197 at AD 378

Table 6. Single letter tags used to identify various sources in the collation, cf. the bibliography for details of the editions.

 

Regarding the matter of recording a comparison of the semantic and their textual details of cognate records from two annals, because from the early tenth century onwards it is often found that annalistic records differ significantly from one to another, the following cryptic notation has been employed to register these differences in the Remarks column:

 

 

Semantic and textual relationships

Notation

Meaning

X=Y

Semantic content of X and Y is identical.

X» Y

Semantic content of X and Y is nearly identical.

X¹ Y

Semantic content of X and Y differs significantly.

X>Y

Semantic content of X is greater than that of Y.

X<Y

Semantic content of X is less than that of Y.

X »Y

Semantic content of X is much greater than that of Y.

X «Y

Semantic content of X is much less than that of Y.

Textual relationships, which refer to the word, not orthographic, level; note that if no textual relationship is expressed, X.tr.Y is assumed

Notation

Meaning

X.vb.Y

Text of X and Y are verbatim, and hence implies X=Y.

X.tr.Y

Text of X is related to that of Y.

X.tu.Y

Text of X is unrelated to that Y.

Examples:

 

X» Y, X.tu.Y

X and Y say nearly the same thing in different words.

X>Y

X says more than Y, and there is an implicit textual dependency.

X.vb.Y

X and Y are verbatim and hence X=Y.

X « Y

X equals a small part of Y and is implicitly textually related.

X=Y» Z

Z differs slightly from X and Y, which are close and are related.

X=Y>Z

Z has less than X or Y, and all are related

X=Y=Z

X, Y and Z all say the same thing and all are related

X>Y>Z

X says more than Y says more than Z, and all are related.

Presentation and Status of the Collation

The collation has been implemented as a series of tables in Word 6.0 which are designed to print on an A4 page in landscape orientation so as to provide the maximum width for parallel tabulation. Time is considered to advance down the page and all the tokens placed on one line are taken to refer to the same event. As no horizontal lines are printed it is recommended that a ruler be used to ensure correct alignment for tokens widely separated across the table. The arrangement of the columns is as follows: 1) Synchronised AD; 2) Feria of the kalends of January; 3) Two or more columns of synchronised tokens; 4) Remarks which register observations regarding the tokens, in particular the start of each solar cycle every twenty-eight years starting at AD 1, and the start of each lunar cycle every nineteen years starting at AD 1007. The annalistic version of the solar cycle comprises the following sequence of kalends and feria:

 

Years

1–4

5–8

9–12

13–16

17–20

21–24

25–28

Common

K.uii.

K.u.

K.iii.

K.i.

K.ui.

K.iiii.

K.ii.

Common

K.i.

K.ui.

K.iiii.

K.ii.

K.uii.

K.u.

K.iii.

Common

K.ii.

K.uii.

K.u.

K.iii.

K.i.

K.ui.

K.iiii.

Bissextile

K.iii.

K.i.

K.ui.

K.iiii.

K.ii.

K.uii.

K.u.

To compute the feria on the kalends of January for any Julian year first compute the remainder on division by twenty-eight and use it to index the above table. For example AD 664=28×23+20 so that the remainder is twenty and AD 664 is bissextile with K.ii. The epactal cycle specifies the age of the moon on the kalends of January according to the Dionysiac computation, and it comprises the following nineteen epacts indexed by the so-called Golden numbers (GN) 1–19:

GN

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

Epact

9

20

1

12

23

4

15

26

7

18

29

10

21

2

13

24

5

16

27

To compute the epact for any Julian year first increment the AD by one and then compute the remainder on division by nineteen and use it to index the above table. For example LC commences in AD 1014 with epact luna .xx.ui., and 1014+1=1015=53×19+8, so the remainder is eight and GN=8 and thus the epact is indeed 26. However it should be emphasised that notwithstanding the emphasis placed on epactal series by Bartholomew Mac Carthy in his 1893–1901 contributions to the edition of AU, in fact AU has correctly synchronised epacts only for AD 1014–1192, while in LC/CT they are found just for AD 1014–1234. The contribution therefore of epacts to Irish annalistic chronology as compared to that of feria is relatively brief and insignificant.

In order not to have large areas of blank space between the event tokens when there are lacunae in any of the sources AT, CS, MB, AU, LC and CT the collation has been split into a series of sub-tables based on these lacunae as shown in Table 7. The collations for these intervals have been further subdivided in order to obtain Word and HTML files averaging about 70K bytes in size, in order to facilitate their downloading across the Internet and their maintenance.

AD Interval

Collates ...

1–306

AT, AI, AB, AU, CM.

307–358

AT, CS, MB, AI, AB, AU, CM.

359–424

CS, MB, AI, AB, AU, CM.

425–487

CS, MB, AU, AI, AB, AR, CM, AC.

488–722

AT, CS, MB, AU, AI, AB, AR, FA, CM, AC.

723–766

AT, MB, AU, AI, AB, AR, AC.

767–803

AU, MB.

804–973

CS, AU, MB.

974–1013

AT, CS, AU.

1014–1150

AT, CS, AU, LC.

1151–1178

AT, AU, LC.

1179–1223

LC, AU.

1224–1316

LC, CT, AU.

1317–1412

CT, AU.

1413–1544

LC, CT, AU.

1545–1590

LC.

Table 7. Ranges of the collation of the different sources at March 2005.

 

The present collation therefore covers the entire Christian era up to the final year of the annals of the kalend tradition, viz. AD 1590. This collation is comprehensive up until AD 766, and substantial for AD 908–1178, and this coverage serves a number of purposes:

1.

It establishes a synchronized AD chronology for all these annals from the beginning of the Christian era up to AD 1590.

2.

It displays substantial detail regarding the structure and content of the Iona chronicle, which is considered to extend up to c. AD 740, and for the first 26 years after its removal to Ireland, allowing identification of its important features.

3.

It displays substantial details regarding the shared and contrasting features of AT/CS and AU over the interval of their divergence AD 913–1178, thereby displaying the significant features of the Clonmacnoise chronicle as preserved in AT/CS.

4.

It demonstrates that for AD 1014–1590 the kalends, feria and AD chronology of LC/CT are synchronized and reliable.

5.

It demonstrates that for AD 431–1013 and 1192–1378 the AD chronology of AU is unreliable.

Applications of the Collation

The principle application for which the collation has been designed is to provide a synchronised AD chronology for all the events recorded in the annals of the kalend tradition. For events occurring up to AD 766 this involves simply locating the record token of the event in the collation and reading off the AD from the left-hand column. After AD 766, for records which have been collated, this same procedure should of course be followed, however there is a substantial chance that the record of interest has not been collated and so no token for it appears in the collation. In this case the remaining records in the same year must be examined until one is found whose token does appear in the collation. For example the obit, ‘Illulbh, Rí Alban moritur’ appears in CS under Hennessy’s marginal AD of 960 but it does not appear in the collation. However a token of the immediately preceding record, viz. ‘Crech la Flaithbertach’ does appear in the collation at AD 962 and this therefore is the synchronised AD to be assigned to Illulbh’s obit.

An application complementary to synchronising their AD is to provide efficient identification of parallel records of events registered in the annals, and this may be done comprehensively up to AD 766. Thus someone interested in the death of Baethéne, Columba’s successor as abbot of Iona, may establish at a glance that the synchronised date is AD 596 and that parallel entries occur in AT, CS, MB, AU, AR, AI, AB, and the full text of all these entries may then be readily collated. It is to be hoped that this will facilitate the use of the annals for historical studies, which have hitherto been greatly impeded by the uncertainty attached to the marginal AD inserted in the various published editions. For example, between 1866 and 1983 the following were the editorial AD years assigned to Baethéne’s obit: Hennessy’s edition of CS placed it at 598, while his edition of AU placed it at 597 which Stoke’s edition of AT cited as ‘AU. 597’. The published editions of AB and AR cited AU in turn as ‘598’ and ‘AU 598’, which was Hennessy’s AU marginal AD incremented by one year. Mac Airt’s edition of AI placed it at 601, and finally Mac Airt and Mac Niocaill’s edition of AU placed it at 598. As can be seen these published editions provide three different years, viz. 597, 598, and 601, all of which differ from the synchronised year of AD 596. Moreover the pervasive influence of AU’s AD chronology together with the ill consequences of automatically incrementing its MS AD can be clearly seen in these editorial data.

A further application of this collation over the interval AD 1–766 is to use it to identify records of events unique to each source, since the token for such an event must stand on its own in a line. In this way it is straightforward, for example, to observe that AT presents a sequence of records which have been taken from Bede’s Chronica Maiora, at AD 492 (Trasamundus), 502 (Simacus P.), 526 (Iohannes) and 526 (Iustinus), to cite just a few of these. These records are important clues to the text history of the annal collection preserving them, cf. the discussion above regarding a secondary stratum from CM. Even after AD 766 where the record collation is partial, a good idea of the content relationship of the different annals may be obtained by examining the years for which a complete collation has been done, generally at decadal years; for example at AD 780 we see that MB has five entries in common with AU and they are in identical order, evidently reflecting their common source. Again, if AT/CS/AU are examined over the range AD 974–1023 it will be seen that they transmit numerous records in common, many of which are both semantically and textually closely related. Furthermore it will also be noted both that the first entry in AT/CS is normally also found in AU, and that sequences of mutually common records still are often presented in close or identical sequence in all three annals, see AD 1022 for example. Such widespread correspondences clearly point to a common source still underlying all three annals at this date. Hence the conclusion presented by Kathleen Hughes in Early Christian Ireland: Introduction to the Sources (London 1972) p. 107, viz., ‘To sum up, AU and Tig.. represent a Chronicle of Ireland which must have been drawn up some time before 913, for at this point the two families diverge. A copy was made then and was subsequently taken to Clonmacnoise, where additions were inserted’, is unsustainable. It emerges rather that in c. 1022 Cuan Ó Lothcháin based his compilation of Liber Cuanach upon the continuation of the Iona chronicle that had been maintained in Clonmacnoise from the mid-eighth century. This Clonmacnoise chronicle was continued there until nearly the end of the twelfth century.

A further application of the collation is to use it to observe closely what material is in common and what is omitted between cognate annals. For example at AD 588, relative to both AT and MB, CS has omitted the battles of ‘Leithrigh’ and ‘Muighe’ and the obit of Lughdach, so we may explore this to see whether a particular political outlook is implied by these omissions. Similarly at AD 622 CS omits the obit of Ailella and the battle of ‘Lindais’ from a long set of entries found otherwise fully in AT and AU. Historical evaluation of these omissions should sharpen considerably our knowledge of the editorial policy employed when CS was compiled. Similar methods may be applied to the smaller collections to identify what factors have influenced their compilation and hence to clarify their inter-relationships.

Work outstanding – the pre-Christian annals

It still remains to establish a corresponding synchronised chronology for the pre-Christian era for the annals of the kalend tradition. This will require the systematic collation of the kalends and patriarchal and dynastic reigns in AB, AI and AT, with the corresponding reigns in Bede’s De Temporibus and CM, together with the Annus Mundi apparatus of CM. This should yield further insight into the methods used by Rufinus in the construction of his chronicle, and also by Bede in the construction of his two chronicles.

Conclusion

This article has set forth the rationale and the methods used to construct the synchronised collation of the various annals over the years AD 1–1590. From this has been derived a synchronised AD chronology for all of the annals of the kalend tradition based upon the kalends+feria of AT/CS for AD 1–1013, except for their common lacuna of AD 767–803 where we rely on AU’s kalends. For AD 1014–1590 the synchronised chronology is based upon the kalends+feria+AD of AT/CS/LC/CT, wherever these are available. On the other hand the AD chronology of AU which in modern times has so frequently been cited for the chronology of Irish history has been shown to be reliable only for the years AD 1014–1192 and 1379–1543.

This collation has been done piecemeal by the author and while as much care as possible has been taken, given the sheer volume and complexity of the material it is virtually certain that it contains some omissions and mis-identifications. It will be greatly appreciated if these would be notified to him upon discovery in order that the accuracy of the collation may be improved. These may be emailed to him at mccarthy@cs.tcd.ie or posted to D.P. Mc Carthy, Department of Computer Science, Trinity College, Dublin 2, Ireland.

 

Acknowledgements

I wish to thank the following for their contributions to this work. Prof. John G. Byrne, of the Department of Computer Science at Trinity College Dublin, who provided both the multi-disciplinary ambience and the encouragement that enabled me to undertake this project. Prof. Dáibhí Ó Cróinín of NUI Galway who examined the first edition of this collation and offered constructive criticism and advice. The Librarians and support staff at the Armagh Public Library, Bodleian Library Oxford, British Library London, NUI Dublin Archives Department, DIAS Celtic Studies Library, and the Trinity College Libraries of Berkeley, Early Printed Books, and Manuscripts. In particular I wish to thank Dr. Bernard Meehan and Stuart Ó Seanóir of the Manuscripts Room for their assistance with my examinations of TCD 1282, i.e. MS H of AU. My colleagues at the Department of Computer Science who over the years helped solve a wide range of technical problems, and who by their interest and enthusiasm have sustained my commitment to the project. In particular I wish to thank Mike Brady, Alexis Donnelly, Tom Irwin, Tom Kearney, Stephen Kenny, John Keogh, Mary Murray, Willie O’Connor, Rosemarie Power, and Rosemary Welsh.

 

Bibliography

 

Encyclopaedia Brittannica 24 vol. (Chicago & London 1970).

Colgrave, B. & Mynors, R.A.B.

Bede’s Ecclesiastical History of the English people (London 1991).

Cross, F.L. & Livingstone, E.A.

The Oxford Dictionary of the Christian Church (Oxford 1983).

Freeman, A.M.

‘The annals in Cotton MS Titus A xxv’, Revue Celtique 41 (1924) 301–30; 42 (1925) 283–305; 43 (1926) 358–84; 44 (1927) 336–61. These are the Annals of Boyle.

Freeman, A.M.

‘The annals of Connacht’, Revue Celtique 50 (1933) 1–23, 117–42, 272–88,339–56; 51 (1934) 46111, 199–301.

Freeman, A.M.

Annála Connacht: The Annals of Connacht (A.D. 1224–1544) (Dublin 1944, repr. 1970).

Gleeson, D & Mac Airt, S.

‘The annals of Roscrea’, PRIA 59C (1959) 137–80.

Hely, J.

Ogygia, or a chronological account of Irish events (Dublin 1793). A translation of O’Flaherty Ogygia q.v.

Hennessy, W.M.

Chronicum Scotorum, a chronicle of Irish affairs from earliest times to A.D. 1135 (London 1866, repr. Wiesbaden 1964).

Hennessy, W.M.

Annals of Loch Cé: a chronicle of Irish affairs, 1014–1690 i–ii (1871 London, repr. Dublin 1939). This includes an edition of CT for AD 1316–1412.

Hennessy, W.M. & Mac Carthy, B.

Annála Uladh – The Annals of Ulster vol. i–iv (Dublin 1887, 1893, 1895, 1901), vol. i by Hennessy reproducing AD 431–1056, vol. ii–iii reproducing AD 1057–1588 with some lacunae, and vol. iv the introduction and index by Mac Carthy. Facsimile edition by Ó Muraíle, N. (Dublin 1998).

Hughes, K.

‘The Welsh Latin chronicles: Annales Cambriae and related texts’, in idem, Celtic Britain in the Early Middle Ages (Woodenbridge 1980) 67–100.

Jones, C.W.

Bedae opera de temporibus (Cambridge, Mass. 1943).

Jones, C.W.

‘Chronica maiora’ in Bedae venerabilis opera, CCSL cxxiii B (Turnhout 1977) 461–535.

Mac Airt, S.

The Annals of Inisfallen (MS Rawlinson B 503) (Dublin 1951).

Mac Airt, S. & Mac Niocaill, G.

The Annals of Ulster (to A.D. 1131) (Dublin 1983).

Macalister, R.A.S.

Lebor Gabála Érenn ITS vols. xxxiv, xxxv, xxxix, xli, xliv (Dublin 1938–56). Scowcroft’s 1987–8 articles q.v. are essential guides to this complex edition.

Mc Carthy, D.P. & Ó Cróinín, D.

‘The ‘lost’ Irish 84-year Easter table rediscovered’, Peritia 6–7 (1987–8) 227–42.

Mc Carthy, D.P.

‘The chronological apparatus of the annals of Ulster AD 431–1131’ Peritia 8 (1994) 46–79.

Mc Carthy, D.P.

‘Easter principles and a fifth-century lunar cycle used in the British Isles’, Journal for the History of Astronomy xxiv (1993) 204–24.

Mc Carthy, D.P.

‘The origin of the latercus Paschal cycle of the insular Celtic churches’, Cambrian Medieval Celtic Studies 28 (1994) 25–49.

Mc Carthy, D.P.

‘The Lunar and Paschal tables of De ratione paschali attributed to Anatolius of Laodicea’, Archive for History of Exact Sciences 49:4 (1996) 285–320.

Mc Carthy, D.P.

‘The Chronology of the Irish Annals’, PRIA 98C(1998)203–55. A pdf copy of this paper is available at:

www.ria.ie/publications/journals/ProcCI/1998/PC98/PC98.html

Mc Carthy, D.P.

‘The Status of the pre-Patrician Irish Annals’, Peritia 12 (1998)98–152.

Mc Carthy, D.P.

‘The Chronology of St. Brigit of Kildare’, Peritia 14 (2000) 255–81.

Mc Carthy, D.P.

‘The chronology and sources of the early Irish annals’, Early medieval Europe 10:3 (2001) 323–41.

Mc Carthy, D.P.

‘The chronology of St. Colm Cille’, An Tionól 24 Nov. 2001 DIAS www.celt.dias.ie/english/tionol/tionol01.html

Mc Carthy, D.P.

‘The Emergence of Anno Domini’ in Time and Eternity by G. Moreno-Riaño & G. Jaritz (edd.), (Turnout 2003) 31–53.

Mc Carthy, D.P.

‘The chronological apparatus of the Annals of Ulster AD 82–1019’, Peritia 16 (2003) 256–83.

Mc Carthy, D.P.

‘The original compilation of the Annals of Ulster’, Studia Celtica 38(2004)69–96.

Mc Carthy, D.P. & Breen, A.

‘An Evaluation of Astronomical Observations in the Irish Annals’, Vistas in Astronomy, vol 41, no. 1 (1997)117–138.

Mc Carthy, D.P. & Breen, A.

‘Astronomical observations in the Irish annals and their motivation’ Peritia 11 (1997) 1–43.

Mc Carthy, D.P. & Breen, A.

The ante-Nicene Christian Pasch: De ratione paschali The Paschal tract of Anatolius, bishop of Laodicea (Dublin 2003).

Mommsen, T.

Gestorum pontificum Romanorum i, in MGH (Berlin 1898).

Morris, J.

Nennius – British History and the Welsh Annals (London 1980).

Murphy, D.

The Annals of Clonmacnoise, being Annals of Ireland from the earliest period to A.D. 1408, translated into English A.D. 1627 by Conell Mageoghagan and now for the first time printed (Dublin 1896, repr. Llanerch, Wales 1993). This is MB.

Ó Cróinín, D.

Early Irish History and Chronology (Dublin 2003). Important collection of his articles published 1981–2003.

O’Donovan, J.

Annala Rioghachta Eireann – Annals of the Kingdom of Ireland vols. i–viii (Dublin, 1848–51; repr. N. York, 1966).

O’Flaherty, R.

Ogygia, seu, Rerum Hibernicarum Chronologia … (London 1685). See J. Hely Ogygia.

Ó Muraíle, N.

Annála Uladh – Annals of Ulster from the earliest times to the year 1541, vol i-iv (Dublin 1998). A facsimile reprint of Hennessy & Mac Carthy AU.

Ó Muraíle, N.

Irish leaders and learning through the ages – Paul Walsh essays collected, edited and introduced (Dublin 2003)

Radner, J.V.

Fragmentary Annals of Ireland (Dublin 1978).

Scowcroft, R.M.

Leabhar Gabhála Part I: The growth of the text’, Ériu xxxviii (1987) 79–140.

Scowcroft, R.M.

Leabhar Gabhála Part II: The growth of the tradition’, Ériu xxxix (1988) 1–66.

Stokes, W.

The Annals of Tigernach, first published in Revue Celtique 16 (1895) 374–419; 17 (1896) 6–33, 119–263, 337–420; 18 (1897) 9–59, 150–97, 267–303. Facsimile edition in two volumes by Llanerch (Felinfach 1993) whose pagination is used in this article and the collation.

Walsh, M. & Ó Cróinín, D.

Cummian's Letter De controversia paschali and the De ratione conputandi, Pontifical Institute of Medieval Studies (Toronto 1988).

Walsh, P.

The dating of the Annals of Inisfallen’, Catholic Bulletin 29 (1939) 677–82. Reprinted in Ó Muraíle, Irish Leaders, 477–83.