Notes to Poem VI

i.  MS Text:  The text occupies the whole of pp. 93-96 of the MS. Pp. 94-96 are clearly legible, but fading has seriously affected the outer margin of p. 93. Ultra-violet light is of limited assistance in retrieving the text on this page, and the edition consequently retains a number of gaps in qq. 6-10. Orthographic errors occur occasionally in the text. There is no evidence that the scribe had access to more than one version of the poem.

ii. General background: This unusual poem, which occurs only in BDL, is of particular interest since it consists mainly of an account of the Battle of Ventry, known in Irish and Scottish tradition as “Cath Fionntrágha”. The battle was the subject of a popular Early Modern prose tale (O’Rahilly; Meyer).

Before exploring the relationship between the BDL poem and the prose tale, we may look at the overall structure of the ballad. Its frame is that of an Oisean/Patrick dialogue. Patrick takes Oisean to his cell, and asks for an account of the person who caused greatest distress to the Fian (qq. 1-2). Oisean takes the opposite tack, and proceeds to emphasise the achievements of the Fian and the amount of territory which was under their sway (qq. 3-7). Patrick evidently tries to bring Oisean back to the original request (q. 8), but with apparently little success. He next asks the names of the best warriors in the Fian in the time of battle (q. 10). Oisean then provides a list of characters, who include the most prominent men of the Fian (qq. 11-16). Eventually (q. 17), Oisean asks Patrick to take his pen to write the names of the most accomplished characters who came to attack the Fian. The attack of Dáire Donn, localised at “tráigh Fionntrágha”, is narrated (qq. 18-38). The ballad concludes with quatrains in which Patrick asks Oisean to make his peace unto death. The impression given by the structure of the ballad is that of looseness, with little cohesion between its episodes.

(1) The relationship between the prose and verse accounts of Cath Fionntrágha (CF)

The prose version of CF is known in several different MSS, of which the earliest is Rawlinson B 847. Most of the MSS are of Irish provenance, but one was compiled in Scotland by the poet, Alexander MacDonald (Nat. Lib. Scot. MS 72.2.11). C. O’Rahilly ascribes the Rawlinson MS to the fifteenth century, and suggests a date 1460-1470 for the text of CF as it stands in that MS. The theme of a battle at Ventry, between Fionn and allmuraig, was however known as early as the twelfth century, since an account of it is given in Acallam na Senórach. This account gives little detail of the fight itself, concentrating instead on the death of Caol, a newly married warrior who, together with his wife, accompanied the Fian to take part in the battle. Caol was drowned on the last day of the fight, and his wife died of grief.

The Early Modern prose tale is on a much more ambitious scale than the Acallam description. It appears to draw on traditions relating to the Battle of Clontarf, the Second Battle of Mag Tuired, and the Ulster Cycle, to name only some of the sources utilised in its compilation. The tale narrates how Ireland is attacked by a vast host under the control of Dáire Donn, king of the world, who has come to avenge an insult given by Fionn to the king of France, whose wife and daughter had eloped with him. The host lands at Ventry (in Co. Kerry), and the Fian give battle. The tale consists largely of accounts of the individual encounters between warriors of the Fian and their opponents. The style is florid, with a general delight in exaggeration and tautology, to such an extent that the death of Fionn himself scarcely registers in the mind of the reader. The battle results in the virtual annihilation of both the Fian and the invaders.

Given the cataclysmic nature of the tale, it is scarcely surprising that CF should assume considerable importance as a grand statement of the heroic ideals of Fionn and the Fian. Such importance is indicated by the good number of surviving MSS, and by the frequency of oral versions from Connacht and Munster (Bruford, 253). The influence of the tale on the later ballad tradition is attested by the present poem, and apparently also by the poems on the Battle of Gabhair in BDL (see item XXII). We now turn to examine the similarities and differences between the BDL poem and the prose tale.

(a) Similarities: There is no doubt that the overall plot of the poem matches that of the prose tale. There are also several correspondences of detail. The invader has very numerous battalions (BDL qq. 21, 36; CFO ll. 852-3; CFM ll. 599-600); Conn Críthir is watching the harbour, and gives battle to the host (BDL q. 22; CFO ll. 122-272; CFM ll. 88-194); he also slays a king (ibid.); the flight of the king of France to Gleann Bolcáin is recorded (BDL qq. 28-9; CF) ll. 435-56; CFM ll. 306-20); a number of characters are shown in the same roles in both pieces, e.g. the Tuatha Dé, Cairbre Lifeachair, Goll mac ríogh Uladh (Conn, in the prose tale) (as allies of the Fian); Dolar, Orarmach (Ogarmach, in the prose tale), Foirne, the four sons of the king of Lochlann (all as enemies of the Fian).

(b) Differences: Some minor differences in the names of characters have been noted in the preceding section. The BDL poem also refers to characters who do not appear in the prose tale: the three sons of Dubhthach, Fear Teine, Ciorcall a’ Phluic (q. 24), the three Bailbh from Burren (q. 26). In the poem, Conn Críthir is said to have slain the king of India, and not the king of Spain, as in the prose tale. These differences are sufficient to indicate that the composer of the BDL poem was not drawing on a version of the tale identical with that in Rawlinson B 847. More substantial divergences may now be noted individually:

(i.) As one might expect, the poem is stylistically very different from the prose tale. It is bare and economical in its use of language, in keeping with the ballad tradition. More noticeable is the lack of reference to the many single combats which are a feature of the prose tale; only the encounter between Oscar and the king of France is mentioned in any detail (qq. 28-29). This is a reflection of the status of Oscar in CF. Otherwise, the poem gives only the names of some of those who have fallen in the battle.

(ii.) Dáire Donn is referred to in the first instance as rí Lochlann (q. 18), and then as rí an domhain (q. 27). In the prose tale, he is unequivocally the king of the world, and the king of Lochlann is included only as a member of his host, although the sons of the latter play an important part in the tale. The equation of Dáire Donn with rí Lochlann by the poet suggests that he tended to view the fight primarily as a Viking raid, although he was aware of the alternative designation of the king. Two oral prose versions also make Dáire king of Lochlann (Bruford, pp. 118-9).

(iii.) The poem makes no mention of the elopement of the wife and daughter of the king of France with Fionn, which the prose tale regards as the principal reason for the attack (CFO ll. 20-30; CFM ll. 17-24). The oral prose versions generally do not refer to this. In the poem, the explanation given is that Dáire Donn wanted to obtain the tribute of Ireland (q. 18).

(iv.) The outcome of the battle is portrayed very differently in the ballad. Although the composer emphasises the slaughter and the loss of five battalions of the Fian (qq. 31-35), he makes it clear that two battalions did survive in spite of being badly wounded (q. 33). Only the host of Dáire is wiped out (q. 36). Such a conclusion is in contrast to that of the prose tale, where the slaughter is total on both sides. Oisean is at pains to point out to Patrick that the Fian continued in existence until the Battle of Gabhair (q. 37). Clearly, therefore, the composer of the poem was conscious of the significance of Gabhair in the pseudo-historical scheme of the tradition, and he may well have adjusted the ending of his CF source to accord with this scheme.

(2) The relationship between the BDL poem and the ballad known as “The Lay of Airrghean the Great”, “Erragon” or “Teanntachd Mhór na Féinne”.

In his magisterial study of the treatment of the Viking theme in Gaelic heroic ballads, Professor Christiansen (pp. 97-131) pays considerable attention to “The Lay of Airrghean the Great”, the earliest text of which is DF LXVII. This ballad may be summarised as follows: Patrick comes to Oisean’s house, and he requests from Oisean an account of the greatest strait in which the Fian warriors found themselves. Oisean obliges, and proceeds to tell Patrick of how on one occasion Fionn forgot to invite certain of the Fian to a feast. Two of the forgotten Fian, mac Cronnchair and Aille, are insulted and take service with the king of Lochlann for a year. The wife of the king of Lochlann falls in love with Aille, commits adultery with him, and then escapes with him to Ireland. The enraged king, Airrghean mac Ancair, gathers a host and goes off in pursuit to Ireland. They encamp on the shore. Fionn offers a present to buy off the king, but when this is not accepted, battle is joined, and the Fian find themselves in difficulties. Goll mac Morna undertakes to meet Airrghean in single combat, and after a great struggle lasting eight days he manages to slay the king. The king’s host is destroyed, and the Fian themselves suffer the loss of more than half their number.

In the concluding chapter of his work, Professor Christiansen suggests that “The Lay of Airrghean” derived its theme from the prose version of CF. Christiansen was well aware of the tentative nature of his suggestion, and he goes on to explain (pp. 394-5): “The Erragon ballad could not be considered as a natural development of the Cath Finntragha, but the ballad-maker, a learned poet, evidently chose that single motif and made a poem on it … The plot in itself is sufficient to explain the whole story of the ballad, as very few particulars are given, and the sequel is only a development of a stereotype scheme for a Viking attack.” In the emergence of the ballad from the prose tale, Christiansen attached considerable significance to the BDL poem, since he regarded it as a “curious link in the development from the Cath Finntragha to the Teanntachd mór.”

There are, indeed, some broad similarities between the BDL poem and “The Lay of Airrghean”. Both are cast in the form of an Oisean/Patrick dialogue; in both, Patrick asks Oisean to tell him of the greatest strait ever known to the Fian; in both, a king of Lochlann attacks Ireland; and in both too, there seem to be correspondences of expression and detail. For example, the seven battalions who go to help Airrghean (DF LXVII, q. 21) are reminiscent of the seven battalions of the Fian who appear in the BDL poem (qq. 33, 35). “The Lay of Airrghean” also suggests that one of the only ways to escape alive was to fly through the sky (q. 22) – an idea which could echo the encounter between Oscar and the king of France in the BDL poem (qq. 28-29).

Yet, beyond the similarities in frame and occasionally in sentiment, there are substantial differences between the two poems. In the case of the battle itself, it is apparent that the BDL poem gives a vague and generalised account of the CF struggle. In the Airrghean ballad, the fight is particularised in the persons of the Norse king and Goll mac Morna. This presentation of the encounter with Airrghean links the poem with a small group of ballads which feature a struggle with an invader who causes great distress among the Fian until he is slain by Goll. In all the poems in this group, the outline of the fight is very much the same. It is suggested that these once formed part of a sub-cycle which celebrated the achievements of Goll himself.

The distinguishing feature of “The Lay of Airrghean” within the group centred on Goll is the elopement which caused the Norse king to take vengeance on the Fian. The BDL poem makes no mention of this matter, although the literary prose versions of CF refer to the elopement of the wife and daughter of the king of France with Fionn – an insult which brought about the attack by the king of the world. This could suggest a connection between CF and “The Lay of Airrghean” which by-passes the BDL poem. Yet, as Christiansen himself later pointed out, another ballad exists which combines the theme of elopement with the exploits of Goll. This is an item in the later Agallamh which begins, “A c[h]orr úd thall san léana”. Its contents are thus summarised by Murphy (DF III, pp. 163-4): “Fionn with fifteen of his men entered into the service of the King of Lochlainn. The King of Lochlainn's wife fell in love with Fionn. Fionn and his men were imprisoned. Goll came from Ireland to rescue them. Goll fought a hard fight with the King. Neither won a complete victory. In the night the King told his wife that only one weapon could kill him, his own sword. His wife stole the sword and gave it to Goll. Goll slew the King. Fionn and the fifteen men were released. Having plundered Lochlainn the Fian warriors return to Ireland, abandoning the faithless wife, who is drowned swimming after them.” The main difference between this poem and the others about Goll referred to above is the fact Goll is the invader, and not the king of Lochlann. Murphy considered that the language of the Agallamh poem was earlier than that of CF, and he concluded (ibid., p. 164): “Both the Lay of Airrgheann the Great and Cath Finntrága might with more probability be held to have been based on the Acallam poem, or on the stories which were its sources.” This would make the relationship between the BDL poem and the Airrghean ballad collateral rather than direct.

Murphy’s conclusion neatly illustrates the difficulty of pinning down the direction of borrowing, or the nature of correspondences between individual ballads or between such ballads and prose tales on similar themes. Unless there is a substantial body of detailed evidence, it is unwise to risk a firm conclusion. In the case of the BDL poem and the prose version of CF, such detailed evidence does exist; but in the case of the BDL poem and the Airrghean ballad, the evidence tends to be of a fairly allusive kind, and the problem is aggravated by the nature of the tradition, which produces poems that utilise a restricted range of plot types, as well as a pool of set phrases.

iii. Later versions: No later versions of this poem are known.

iv. Metre: With the exception of q. 15, the poem is in Rannaigheacht Mhór. Q. 15 is in Deibhidhe, and is likely to have strayed into the text from another source. It occurs as part of a catalogue of warriors, and the stereotyped nature of such catalogues could readily lead to some degree of confusion between quatrains from different poems, especially in later tradition.

v. Line annotation

Ascription: This has now faded to the point where little can be read with any certainty. The upper parts of surviving letters have also been obscured by binding. If the letters Auc have been read correctly, it seems likely that the ascription was of the type Auctor huius…, but it would be unwise to speculate beyond this.

4 The last few letters of this line cannot be retrieved even under ultraviolet light, and the proposed restoration ’na sgáth is therefore to some extent conjectural. Problems of legibility also affect the earlier part of the line; MS wes appears to have letters written in superscript directly above it which could be <.ct>, but it is hard to relate these to wes itself. While the other letters in the line are relatively clear, how they should be construed when restoring the line is not at all obvious, and the present restoration of the entire line is by no means definite. The rhyme mhuirn : cuid seems improbable, even by the relaxed metrical standards of such verse.

10 <réd>: Only the initial of the MS form is now clearly legible, although what appears to be the outline of -d can be distinguished. It seems probable that the MS originally had raid.

11 a <bhfoghar>: a woeyr MS.  HP’s restoration do bhréithir seems to be based on taking MS -o- in the second word as -r-. The present restoration suggests that the second element derives from foghar, “sound”, commonly referring to speech or natural sounds (RIA Dict. s.v. fogur); cuir a bhfoghar could mean “give expression to”.

19 ní h-<inneósad>: MS hy'nassit does not show the eó-fut. stem clearly, and -it for normalised -ad is unusual. Cf. hy'nosit (35) and 29 n.

20 a liuthad: a lwit MS. The line is hypermetric as it stands, and the main difficulty appears to lie in this phrase. MS lwit clearly attests the vernacular Sc. G. form of liú, the comp. of il, “many, numerous”, used substantivally (RIA Dict. s.v. il, liú). Line length may be restored by omitting a (? originally 3 sg. poss. pron., introduced by analogy with a leithid) and regarding liuthad as a monosyllable: HP emends to a líon éacht rinn Fiana Finn.

22 <féin>: f<enyt> MS. The MS form, if correctly determined, could represent féinnidh, “member of a fian, warrior, champion” (RIA Dict. s.v. féinnid), although this would evidently make the line hypermetric.

24 Fading and staining have removed the last three (?) syllables of this line, and the earlier extant portion cannot be restored with confidence; MS <teygwill> could conceivably stand for téagmhail as HP suggests, but it is hard to know how this should be accommodated within the line.

25 annsa[n]: a'syt si MS. The second element of the MS form may derive from dittography.

27 The line is hypermetric as it stands, with no obvious solution, unless one assumes that the prep. a was dropped.

29 Dā n-<inneósta>: Da ny' [no]s<y>t MS. The MS form poses several problems. First, there is the question of mood and tense. Strictly dán  was followed by the subj. in E. Mod. Ir. Here, however, the scribe’s final intended verb form was evidently the condit., signalled by -o- in the MS, representing normalised -eó-. In later E. Mod. Ir. the condit. did, in fact, tend to replace the subj. (cf. Des., pp. 260-261). But it should be noted that the letters -no- in the MS form occur above the line, and could be a later correction by the scribe. If we ignore the superscript letters, the MS form is ny's<y>t, which could reflect the influence of the mod. Sc. G. condit. form innseadh, with the scribe thinking initially in terms of nan innseadh, and then correcting to a more acceptable E. Mod. form. Or does an original subj. lurk in the background? Then, there is the difficulty of knowing which person of the verb the MS form intends to convey by -<y>t (y evidently being written over o, or possibly vice versa). Although the restored form assumes a regular 2 sg. condit. inflexion, the MS seems to suggest -ad or possibly -adh. If the latter is intended, it may be that the form is influenced by the Sc. G. condit. innseadh, which, although originally 3 sg., is used analytically with 2 sg., 2 pl. and 3 pl. pers. prons. It is therefore possible that the MS represents a hybrid E. Mod. Ir./Sc. G. verb form.

30 <tim>: The restored word is by no means certain, since MS <tem> might also be read as <senn>, possibly standing for sean.

32 Although individual letters in the earlier part of the line are legible, the later part is badly faded, and little restoration can be attempted. Similarly 38.

45 The line is hypermetric as it stands; correct length can be obtained only by regarding Feardhomhan as disyllabic, as the MS form Farȝone suggests.

49 <air bhleith mionn>: er wley mynni MS. This phrase seems likely to be a cheville; it may mean “after sharpening [his] weapons” (see RIA Dict. s.v. bleith I (b), 2 mind).

50 <gcionn>: genk MS. This word ought to provide aicill with the preceding line, and the present suggestion is made tentatively, with cionn being regarded as a variant of ceann, perhaps in the sense of “tip, point” (RIA Dict. s.v. cenn II.10) with reference to weapons. -nk of the MS form would indicate strong nasalisation.

53 mhac: wa<k>t MS. The third letter of the MS form appears to have been written over an original a, the scribe probably thinking initially of mhath, written elsewhere in the poem as waat (56).

54 The final part of this line seems to be i' deim <nawe> in the MS; the last word is unclear, partly because of cramming at the end of the line, but mainly because of scribal correction, perhaps of an original naave. It is by no means obvious how this is to be restored; an dtím námh, “in the time of enemy [attack]”, suggests itself, and this supplies satisfactory aicill and end rhyme; with the gen. pl. form námh (for historical námhad), cf. Psalm XXIII. v. 5 a: Dhomh dheasaich bord air beul mo nàmh.

57 On this quatrain, see Section iv above.

60 <theanntachd>: hentyt MS. The restoration can be made with reasonable confidence. HP restores as feadhma, but this would appear to be based on reading -nt- of the MS form as -m-.

61 m<é>: ma MS. For spelling, cf. III 8 n, 16 n.

63 cha: The MS represents three attempts at writing the negative particle. First cha was written and deleted; then not (perhaps to be read as nocha, with final -a from the original cha) was supplied in superscript; finally not was cancelled, and cha written to the left of the first attempt.

dhomhs[a]: The scribe first wrote ȝoos, cancelled it and wrote do<s>t, the third letter of which was evidently written over another.

64 The line is hypermetric, with no obvious solution.

65 MS is makes the line hypermetric, and requires to be read as ’s.

67 dha: The MS form ga may reflect a dialectal initial /g/, or it may assume /Ä/ without showing lenition.

74 Tuatha: t<wo>a MS. The scribe may originally have written toooa which he then altered by writing what appears to be w over the first two os.

75 Luachair: loeitchir MS. The MS form seems unequivocal, and is of some interest since the character is normally known as Cairbre Lifeachair. It is hard to know how the present epithet came to be applied to Cairbre; could it have been through confusion of his seat Teamhair (Tara) with Teamhair Luachra, a resort of the Fian evidently in the Luachair district of West Cork and Kerry (DF, III, p. 403, s.v. Lúachair; p. 407, s.v. Teamhair Lúachra)? Note that Luachair here is vouched for by line length, since Lifeachair would have made the line hypermetric.

79 Fading obscures the opening part of this line, and only Ne can be read with any confidence; one or two letters are suggested by the pen-marks thereafter, and then si or possibly fi follows. From this point, the line becomes legible, but other difficulties appear, principally in syllabification. Even allowing for only one illegible syllable, the line as it stands would have two additional syllables at least. Part of the problem may lie in the poss. pron. ar (MS [ir]), which seems redundant in view of dhinn[e] in the following line, and which may be intrusive (but how does one define its relationship to deach<ar>, if the final letters are indeed ar?). The removal of ar does not solve the difficulty, however. Was the line originally something like: Ní dheachaidh ceathramh ó thráigh?

80 <gháire Duinn>: ȝaryt dwn<n> MS. The MS form ȝaryt suggests that the scribe is thinking in terms of gáire, “laughter, a laugh”, rather than gáir, “shout, cry”. HP hesitantly restores to gháir an Duinn, which may well have been the original reading.

83 <trí catha> fichead: The scribe appears to have written first xxx feit, which would stand for tríocha fichead, “thirty score” (thus HP). However, an insertion mark follows MS xxx, and what appears to be ti caa is found in superscript. If ti is correctly understood, it could be an abbreviated form of trí, so that the superscript may well convey trí catha, “three battalions”. We are therefore left with the choice of reading either tríocha fichead or trí catha fichead, “twenty-three battalions”. As original xxx is not cancelled, we may suppose that the scribe was himself uncertain about the correct reading, although only trí catha fichead gives the proper line length. A similar problem is evident at 141, where the same number is required. The scribe first wrote xxx ca feiȝit, which represents tríocha cath fichead, “fifty battalions”. Then ti was written in superscript between xxx and ca, giving an ambiguous overall reading similar to that in 83. In both instances the trouble may have arisen because the scribe could not decide whether the reading in his exemplar was to be regarded as tríocha or trí catha; although it is possible too that the confusion may have stemmed from representing an original deich gcatha fichead (the number in the prose tale: CFO ll. 852-853; CFM ll. 599-600) by means of the Roman numeral xxx, and thereby losing certain syllables.

84 dan: The scribe first wrote dar which he later changed to dan.

87 The line appears to lack a syllable, and thus HP inserts the pron. sé after the verb. But if Innia is regarded as three syllables, and not as two, the line does not require correction.

95 Fear <Teine>: fer tenni MS. HP interprets the second element as Tuinne.

100 <sinn>: synn MS. HP reads sin, which is equally possible.

112 <teinn>: The initial of the original MS form tyn is intersected by an oblique stroke, the significance of which is uncertain.

115 <mar tha [a] theist>: mir ta hest MS. Although advanced somewhat tentatively, this restoration corresponds more closely to the MS than HP, mar do thosd. The phrase may be construed as a cheville: “as the account of him states” (see RIA Dict. s.v. teist II (c), (e)). If restored correctly, it could indicate that the composer of the ballad was consciously recasting an earlier version of the story.

116 <feis>: The MS form <fes> is faded, but seems reasonably probable. The word feis, “spending the night, sleeping, accommodation for the night” (RIA Dict. s.v.), would be appropriate here.

125 Fionntrágha: The -n of the MS form fintraichin anticipates the initial of the gen. pl. article.

128 <ar leinn>: er <lain> MS. HP cautiously restores this as ar léin, and translates “upon the field (?)”, evidently deriving léin from léana, “meadowland” (RIA Dict. s.v. lénae). The present suggestion would regard the phrase simply as a cheville meaning “we thought” (RIA Dict. s.v. dar 2).

131 eagair: The scribe first writes eggin (? perhaps thinking of éigin, “some, any, a certain”), and cancels the word.

134 buidhean: The second letter of the MS form b<o>ein seems to be written over an original e.

136 Chlannaibh: This form implies that the prep. de is assumed, although not attested in the line, which would be hypermetric if it were included. Cf. V, 3 n.

140 leagadh: It is difficult to know whether the second letter of the MS form l<e>gga is e or a; the one has apparently been written over the other.

141 <Trí cath[a]> fichead: See 83 n.

143 [a]: A form of the prep. de is required to restore line length. Cf. 136 n.

144 <d’ráinig>: The superscript r of the MS form, which is written directly above the a, suggests that it is a later addition, and that the scribe originally intended nach dtáinig, which would refer to the host coming back to Ireland; the implication of ráinig would be that they did not return to their homeland.

146 féin: pen̄ MS. Cf. II, 9 n.

sgéal: The second letter of the MS form looks more like a capital c than a k.

159 dhuinn: gin MS. Cf. 67 n.

160 The line lacks a syllable as it stands. HP restores line length by reading agus for is.

161-168 These lines may be a later addition to the text of the poem, but it is difficult to know when they may have been added; they may have been present in the exemplar used by the BDL scribe. That they are possibly a later addition is suggested by the final syllable of 160 which could once have formed the dúnadh of the poem, and which is repeated in the normal manner. It must be noted, however, that syllables corresponding to the openings of certain poems may be repeated in BDL texts at points where they cannot constitute a dúnadh: see XIII, 36, XXVII, 28, 60, 100. The quatrains also seem to show Oisean as having accepted the Christian faith. For a quatrain reminiscent of BDL q. 41, see Gillies, “Gaelic Poems of Duncan Campbell (I)”, p. 27, q. 7.

163 <Muire> is: A combination of fading and cramming have created some difficulty at this point in the MS. The final letters of MS mv<.yt> come right up against mcdey, suggesting that the scribe omitted a syllable which he later restored, but forgot is, which is required for sense.

164 <láimh>: The MS form laive is hardly to be restored to lá as in HP. Nevertheless, the present suggestion is made rather cautiously; transl. “by the hand”.