The collection of Irish Triads, which is here edited and translated for the first time, has come down to us in the following nine manuscripts, dating from the fourteenth to the nineteenth century:—
L, i.e. the Yellow Book of Lecan, a vellum of the end of the fourteenth century, pp. 414b—418a, a complete copy.
B, i.e. the Book of Ballymote, a vellum of the end of the fourteenth century, pp. 65b-66b (ends imperfectly).
M, i.e. the Book of Húi Maine, a vellum of the fourteenth century, fo. 190a[1]-fo. 191a[2]. A complete copy beginning: 'Ceand Erenn Ardmacha,' and ending: 'tri hurgairt bidh a caitheam díescaidheadh (sic) a chaitheam iarna coir a caitheam gan altughudh.' Then follow proverbial sayings from the 'colloquy of Cormac and Cairpre,' such as: 'Dedhe ara ndligh gach maith domelar ithe ⁊ altugud. Anas deach gacha fleidhe a cainaltughudh ⁊ a mochdingbail. Caidhe deach samtha. Ni hansa. Gal gan forran. Deasgaidh codulta frislige,' &c., ending: 'deasgaidh aineolais imreasain. Ni d'agallaim Cormaic ⁊ Cairpre coruici sin.'
Lec, i.e. the Book of Lecan, a vellum of the fifteenth century. The leaves on which the Triads are found are now bound up with the codex H. 2. 17 belonging to Trinity College. It is a complete copy beginning on p. 183b: 'Ceand erenn Ardmacha,' and ending on p. 184b: 'ceitheora aipgitri baisi baig connailbi gell imreasain.'[1] N, i.e. 23. N. 10, a paper ms. written in the year 1575,[2] pp. 98-101. A complete copy, the gap between pp. 100 and 106 being made up by pp. 7a-10b of the vellum portion of the manuscript.
[1] By an oversight I have referred to this ms. sometimes by Lec and sometimes by H. In some cases both Lec and H will be found quoted in the variants. The same ms. is always meant.
[2] As appears from the following colophon on p. 101: 'Oraoit uaim ar do lebor a hOedh in cédluan iar n-aurtach Johannes. Baile Tibhaird ar bla maige mo mendad scribne hi farrad Se(a)ain hi Maoilconari. Mese (Dubthach) do scrib in ball soin da derpiris ⁊ rlæ. Anno domini 1575. Guroiuh maith agat.
Hʹ, i.e. H. 1. 15, pp. 946-957. This is a paper manuscript written by Tadhg Tiorthach O Neachtain in 1745. It is a complete copy, with copious glosses in Modern Irish, the more important of which are printed below on pp. 36-43. At the end O Neachtain has added the following:—'Trí subhailce diadha: creidhemh, dothchus agus grádh. Trí a n-aon: athair, mac, spiorad naomh, da raibh gloir, mola[dh] ⁊ umhlacht tre bith sior tug ré don bhochtan bocht so. Aniu an 15 do bhealltuine 1745. Tadhg O Nechtuin mac Seain a n-aois ceithre bliadhna déag et trí fithchit roscriob na trithibh ṡuas.'
These manuscripts have, on the whole, an identical text, though they all occasionally omit a triad or two; and the order of the single triads varies in all of them. They have all been used in constructing a critical text, the most important variants being given in the foot-notes. The order followed is in the main that of the Yellow Book of Lecan.
There are at least three other manuscripts containing copies of the Triads. One of them I discovered in the Stowe collection after the text had been printed off. It is a paper quarto now marked 23. N. 27, containing on fo. 1a-7b a copy of the Triads, followed on fo. 7b-19a by a glossed copy of the Tecosca Cormaic. It was written in 1714 by Domnall (or Daniel) O Duind mac Eimuinn. Its readings agree closely with those of N. In § 237, it alone, of all manuscripts, gives an intelligible reading of a corrupt passage. For cia fochertar im-muir, cia berthair hi tech fo glass dodeime a tiprait oca mbí, it reads: cia focearta im-muir, cia beirthear hi tech fo glass no do theine, dogeibther occan tiprait, 'though it be thrown into the sea, though it be put into a house under lock, or into fire, it will be found at the well.' In § 121 for cerdai it reads cerd; in § 139 it has rotioc and rotocht; in § 143 for grúss its reading is grís; in §153 it has aibeuloit for eplet; in § 217 tar a n-éisi for dia n-éisi; in § 218 lomradh(twice) for lobra and indlighidh for i n-indligud; in § 219 it has the correct reading éiric, and for dithechte it reads ditheacht; in § 220 it reads fri aroile for fria céile; in § 223 after ile it adds imchiana; in § 224 it reads grís brond .i. galar; in § 229 for meraichne it has mearaigheacht; in § 235 it has mhamus for mám; in § 236 Maig Hi for Maig Lii; and for co ndeirgenai in dam de it reads co nderna in dam fria.
Another copy, written in 1836 by Peter O'Longan, formerly in the possession of the Earls of Crawford, now belongs to the Rylands Library, Manchester, where it was found by Professor Strachan, who kindly copied a page or two for me. It is evidently a very corrupt copy which I have not thought worth the trouble of collating.
Lastly, there is in the Advocates' Library a copy in a vellum manuscript marked Kilbride III. It begins on fo. 9b2 as follows:—'Treching breath annso. Ceann Eirind Ardmacha.' I hope to collate it before long, and give some account of it in the next number of this series.
In all these manuscripts the Triads either follow upon, or precede, or are incorporated in the collections of maxims and proverbial sayings known as Tecosca Cormaic, Auraicept Morainn, and Senbríathra Fíthil, the whole forming a body of early Irish gnomic literature which deserves editing in its entirety. It is clear, however, that the Triads do not originally belong to any of these texts. They had a separate origin, and form a collection by themselves. This is also shown by the fact that the Book of Leinster, the oldest manuscript containing the Tecosca Cormaic (pp. 343a-345b), the Senbríathra Fíthail (pp. 345b-346a), and the Bríathra Moraind (pp. 346a-b), does not include them.
It is but a small portion of the large number of triads scattered throughout early Irish literature that has been brought together in our collection under the title of Trecheng breth Féne, i.e., literally 'a triadic arrangement of the sayings of Irishmen.' I first drew attention to the existence of Irish triads in a note on Irish proverbs in my addition of the Battle of Ventry, p. 85, where a few will be found quoted. A complete collection of them would fill a small volume, especially if it were to include those still current among the people of Ireland, both among Gaelic and English speakers. I must content myself here with giving a few specimens taken at random from my own collections:—
Three kinds of martyrdom that are counted as a cross to man, i.e. white martyrdom, green martyrdom, and red martyrdom.—The Cambray Homily (Thesaurus Palæohibernicus, II., p. 246).
Three enemies of the soul: the world, the devil, and an impious teacher.—Colman maccu Beognae's Alphabet of Piety (Zeitschrift für celtische Philologie, III., p. 452).
Three things whereby the devil shows himself in man: by his face, by his gait, by his speech.—Ib., p. 453.
Three profitable labours in the day: praying, working, reading.—Regula Choluimb Cille (Zeitschr., III., p. 29).
Three laymen of Ireland who became monks: Beccan son of Cula, Mochu son of Lonan, and Enda of Arann.—Notes on the Félire of Oengus (Henry Bradshaw Society, vol. xxix., p. 112).
Three chief artisans of Ireland: Tassach with Patrick, Conlaed with Brigit, and Daig with Ciaran.—Ib., p. 186.
Three poets of the world: Homer of the Greeks, Vergil of the Latins, Ruman of the Gaels.—Book of Leinster, p. 354b.
The three worst counsels that have been acted on in Ireland through the advice of saints: the cutting short of Ciaran's life, the banishment of Colum Cille, the expulsion of Mochuta from Rathen.—Notes on the Félire of Oengus, p. 204, and Tripartite Life, p. 557.[3]
[3] Where for 'wrong stories' read 'wrong counsels' (sanasa sáeba). This triad is thus versified in the Brussels ms. 5100:—
Teora saoba sanasa Leithe Cuind roc[h]aras-[s]a:
Mochuda cona clamhra[i]d d'ionnarba a Rathain roghlain,
cur Coluim Cille tar sal, timdibhe saeghail Ciaráin.
Three things there are for which the Son of living God is not grateful: haughty piety, harsh reproof, reviling a person if it is not certain.[4]
[4] LB., p. 225 marg. inf., and Brussels ms. 5100, fo. 86a:
Fuil trí ní (a trí Br.) doná (danach Br.) buidech mac Dé bí:
crábud úallach, coisced (coiccsed Br.) serb, écnach duine mad inderb.
Three things there are for which the King of the sun is grateful: union of brethren, upright conversation, serving at the altar of God.[5]
[5] Edinburgh ms. xl, p. 28, and Brussels ms. 5100, fo. 86a:
Fuil tréide dianab buidech rí gréine:
óenta bráthar, comrád (fodail Ed.) cert, altóir Dé do thimthirecht.
Woe to the three folk in horrid hell of great blasts: folk who practise poetry, folk who violate their orders, mercenaries.[6]
[6] LB., p. 236, marg. inf.:
Mairg na trí lucht a n-iffirn úathmar anside:
óes dogní dán, óes choilles grád, óes amsaine.
Three things there are which do not behove the poor of living God: ingratitude for his life whatever it be, grumbling, and flattery.[7]
[7] LB., p. 238, marg. inf.:
Fuil trí ní ná dlegair do bocht Dé bí:
dimmda da bethaid cipé, cesacht ocus aibéle.
The following modern triads I owe to a communication from Dr. P.W. Joyce, who heard them in his youth among the people of Limerick:—
Three things to be distrusted: a cow's horn, a dog's tooth, and a horse's hoof.
Three disagreeable things at home: a scolding wife, a squalling child, and a smoky chimney.
The three finest sights in the world: a field of ripe wheat, a ship in full sail, and the wife of a Mac Donnell with child.[8]
[8] This triad comes from the Glynns of Antrim, the Mac Donnells' district.
In our collection an arrangement of the Triads in certain groups, according to their contents, is discernible. Thus, the first sixty-one—of which, however, the opening thirty-one are no Triads at all—are all topographical; and among the rest, those dealing with legal matters stand out clearly (§§ 149-172).
When the collection was made we have no means of ascertaining, except from internal evidence, such as the age of the language, and a few allusions to events, the date of which we can approximately fix.
The language of the Triads may be described as late Old-Irish. Their verbal system indeed is on the whole that of the Continental glosses,[9] and would forbid us to put them later than the year 900. On the other hand, the following peculiarities in declension, in which all the manuscripts agree, make it impossible for us to put them much earlier than the second half of the ninth century.
[9] I may mention particularly the relative forms téite 167, bíte 127, ata 75, 76, 224, &c., berta (O. Ir. berte) 109, 110, fíchte (145), coillte (166), téite (167), aragellat (sic leg. with N) 171; the deponent neimthigedar 116, &c.; ató, 'I am' (104), and the use of the perfective ad- in conaittig 77, 78.
The genitive singular of i- and u-stems no longer shows the ending -o, which has been replaced throughout by -a.[10] Now, in the Annals of Ulster, which are a sure guide in these matters and allow us to follow the development of the language from century to century, this genitive in -o is found for the last time in a.d. 816 (rátho, Ailello). Thence onward the ending -a is always found.
[10] rátha 56, foglada 92, flatha 151, 248, 253; dara 4, 34; Ela 31, 35, 44 (cf. Lainne Ela, AU. 816); átha 50, betha 82, 83, 249.
The place-name Lusca, 'Lusk,' is originally an n-stem making its genitive Luscan. This is the regular form in the Annals of Ulster till the year 880, from which date onward it is always Lusca (a.d. 916, 928, &c.). In our text (§ 46) all the manuscripts read Lusca.
In slender io-stems the dative singular in Old-Irish ends in -iu. I find this form in the Annals of Ulster for the last time in a.d. 816 (Gertidiu). Thence onward it is always -i, as in our text (hi Cúailgni 43, d'uisci 64).
The nasal stem léimm makes its nom. plur. léimmen in Old-Irish. In § 32 we find instead (tair-)leme. So also foimrimm makes its nom. plural foimrimme in § 163.
The word dorus is neuter in Old-Irish, making its nom. acc. plural either dorus or doirsea. In our text (§§ 173, 174) the word is masculine, and makes its nom. plural doruis.
Druimm is an i-stem in Old-Irish, but in the later language passes into an n-stem. In § 51 we find the nom. pl. drommanna.
The neuter grád in § 166 makes its nom. plur. grúda for O. Ir. grád.[11]