Notes to Poem VII

i. MS Text:  The text of this poem occupies the whole of p. 114 of the MS, and most of p. 115. The concluding letters of several lines on p. 115 are obscured by trimming and staining, but most of the poem is clearly legible. As well as scribal errors, the text contains emendations which suggest that more than one version was available to the scribe(s) (q. 9 cd). To the right of the ascription, “Ossianic” has been written by a later hand, possibly that of Dr Donald Smith (Black, Draft Catalogue).

ii. General background:  The poem tells how a fairy woman came to test the faithfulness of the wives of six of the Fian warriors. In a drunken state they had claimed that women as chaste as they were hard to find in the world (q. 4). No sooner was the boast made than the woman arrived, clad in a single-threaded mantle which could expose the misdemeanours of any wife who wore it (q. 9). The Fian women accept the challenge of wearing it, and they are suitably embarrassed, each in turn. Only Mac Reithe's wife escapes complete discomfiture, since the mantle covers all of her except for her little toe – and she later reveals that she has stolen a kiss from Diarmaid (qq. 18-19)! The woman then departs – with the parting shot that she herself has done nothing wrong but to sleep with Fionn (q. 20).

This ballad is distinctive within the surviving corpus of Fian verse because of its unusual plot, its strong element of mischief and its fluent use of dialogue. At the same time, it bears a striking resemblance to an Arthurian ballad, “The Boy and the Mantle”, in which the Arthurian heroines are subjected to a similar test (Child, Ballads, I, 257ff; Kinsey, Oxford Book of Ballads, 31-8). In the English poem the husbands are subjected to two cuckoldry tests; in addition to this, the owner of the mantle is a boy, only four Arthurian women are tested, and there is no mention of earlier boasting or drunkenness, although a feast is evidently in progress. Nevertheless, the women-testing in both poems is comparable, and the role of Craddock’s wife is directly parallel with that of Mac Reithe’s, even in the details of the toe and the kiss (Kinsey, 35):

            When shee had tane the mantle
                And cast itt her about,
            Vpp att her great toe
                Itt began to crinkle and crowt;
            Shee said, Bowe downe, Mantle,
                And Shame me not for nought;

            Once I did amisse,
                I tell you certainlye,
            When I kist Craddockes mouth
                Vnder a greene tree,
            When I kist Craddockes mouth
                Before he marryed mee.

In the English version, it is true that the great toe and not the little toe is mentioned, and that Craddock’s wife steals a kiss from her future husband; but the extent of the correspondence with the relevant section of the Gaelic poem is obvious. Occasionally, also, the two poems seem to correspond verbally; thus with “crinkle and crowt” we may compare the description of the mantle when worn by Maighinis, do chas is do chuar mar soin (BDL, q. 15 c).

The theme of the chastity test by means of an object which comes into contact with individuals is well known in Indo-European folklore, and this, together with its similarity to the English ballad, has given a special interest to the present poem. Particularly tantalising is the question of the origin of the Gaelic poem; is it an independent development of a traditional tale, or is it based on the English ballad, or do the English and Gaelic poems have a common ancestor? (See DF, III, for a useful summary of the debate; Bromwich, TYP, 512-4; Gillies 1981, 64-6.)

The possibility that the Gaelic poem derives from the English poem, or that both may come from an earlier Welsh source, has been suggested cautiously by Ludwig Stern, who draws particular attention to the correspondence of incident involving the wives of Mac Reithe and Craddock (1897, 294-326). T.P. Cross (1919, 649-58), on the other hand, has put the case for independent development, choosing to emphasise the superior articulation of the Gaelic poem, and the excellence of its characterisation, which is wholly consistent with Fian tradition. In Cross’s view, the motive for the stranger’s appearance (as a cast-off fairy mistress of Fionn’s) carries conviction in the Gaelic poem, whereas the Arthurian versions of the story offer less satisfying reasons.

Cross is right to draw attention to the literary sophistication and polish of the Gaelic poem. But it is just such distinction which sets it apart, not only from other non-Gaelic versions of the tale, but from the rest of the corpus of Fian verse. The prominence of the womenfolk of the Fian, the outstanding use of dialogue, the tight control of plot, and the psychological appeal to one’s sense of moral propriety – all are unusual in the Gaelic ballad tradition. It would seem wholly likely that the composer of the Gaelic ballad had a model from which to work, and there is no convincing reason to suppose that this model was not the English poem or an earlier common source. That the Gaelic poet could improve on his model is a comment not merely on his technical accomplishment, but on the temper of the Gaelic environment in which he operated, an environment which enjoyed “ribald guffaws at the expense of female virtue [and] deceived husbands” (Loomis, ALMA, 559). This, indeed, is the environment of courtly love poetry and satiric verse involving women – genres which are well attested in Irish sources and in BDL itself (Gillies, Courtly and Satiric Poems).

iii. Later versions:  This poem survives pre-eminently and popularly in Irish tradition only. In Scotland, one version is found in a MS of Irish provenance which was obtained by the collector, Peter Turner (Nat.Lib.Scot. MS 72.2.4 [Ed.]). Among the later versions two types of texts may be distinguished on the criterion of their length relative to the BDL version.

(1) Texts similar in length to that in BDL

This group is represented by DF LXV (19 qq.) and Ed. (21 qq.). Although these texts correspond fairly closely to BDL in the number of their quatrains, it is apparent that they do not share precisely the same quatrains with one another or with BDL, and their relationship to BDL may be demonstrated further on this principle.

(a) Quatrains peculiar to BDL: These consist of BDL qq. 3, giving the names of the individual wives; 7, telling how the stranger came into the presence of Fionn, and sat beside him; and 17, in which Mac Reithe’s wife, prior to her testing, affirms her faithfulness to her husband. All three quatrains make a contribution to the ballad narrative (q. 7 underlines the embarrassing boldness of the visitor, and q. 17 increases dramatic effect, while q. 3 gives a closer focus), but they can be omitted without damaging the essential structure of the poem. As these quatrains are in Deibhidhe, there is reason to believe that they were part of the original text.

(b) Quatrains in DF and Ed. but not in BDL: There are two such quatrains. The first of these is q. 13 of DF and Ed., which tells of the shaming of Oisean’s wife. As Oisean’s wife was clearly among the number, and is mentioned by name in BDL q. 3, the BDL text would appear to be defective in making no further reference to her. The evidence of the second group of texts might suggest that there was some sensitivity about portraying the wife of Oisean in a bad light, but it is difficult to know whether such an attitude or mere scribal carelessness is responsible for the loss of this quatrain in BDL. The second quatrain in this group is DF q. 18/ Ed. q. 20, in which the visitor states that she will now depart with a story to tell against the Fian women. The metre of this quatrain is a form of Rannaigheacht, and this might indicate that it is a later addition to the original text.

(c) Quatrains in BDL and Ed. but not in DF: Only one quatrain, BDL q. 20/Ed. q. 19, is in this category. In it the visitor identifies herself as the daughter of Dearg, and discloses her relationship with Fionn.

Ed. contains one quatrain (q. 19) not attested in BDL or DF, but it is clear from the foregoing that it is more closely related to DF than to BDL. This conclusion is borne out by the wording of individual quatrains also, although Ed. sometimes supports BDL against DF and vice versa.

(2) Texts considerably longer than that in BDL

Texts with a very much greater number of quatrains than BDL are fairly common in Irish tradition. These are represented by H and C, which occur in MSS from the first half of the nineteenth century. H has a total of 46 quatrains, and C has 44; both texts correspond closely in the general position of their quatrains. To obtain the higher number of verses, it is evident that the dialogue between the principal characters and their wives has been greatly expanded, and that more details have been provided of the testing of the women, sometimes leading to the duplication of incident. Thus Fionn, like Conan Maol, kills his wife when she has been shamed (H q. 38). Oisean’s wife, by contrast, is shown to be entirely virtuous. (See Robinson 1903, 145-52.)

The additional quatrains in H and C can be demonstrated to be accretions not only on stylistic grounds, but on metrical grounds also. The additional quatrains are generally in a poor form of Rannaigheacht Mhór, and some of the original Deibhidhe quatrains have been adjusted to conform to the newer, predominant metrical pattern.

An examination of the surviving original quatrains in H and C suggests that the texts have developed from a version of the type found in DF and Ed. Thus H and C omit the quatrains described in the preceding section as peculiar to BDL, and they include DF q. 18/Ed. q. 20.

iv. Metre:  Deibhidhe

v. Line annotation

Ascription:  The scribe has not inserted the name of the ughdar, but it is not immediately obvious why he has failed to do so. It is possible that he was uncertain as to his identity (note that Oisēan is referred to in the 3rd person (6), while the reference to mo bhean féin (10) does not clarify matters immediately), or he may have felt that the poem was worthy of being ascribed to a prominent non-legendary author, whose name was to be added at a later stage. At any rate, this ascription warns us that the names of certain authors in BDL may have been inserted after the poems were written in the MS.

2 nAlmhai[n]: The MS form nalwi shows no sign of an intended final -n.

’s: MS is requires to be regarded as ’s for line length.

6 Caoilt[e]: The text of the poem indicates (61-76) that the wife of Mac Reithe was the last to be tested; the wives of Conán,  Diarmaid, Osgar and Fionn (but not the wife of Oisean; see Section iii (1) (b) above) are previously tested. This would suggest, at first sight, that the BDL text is erroneous in referring to Caoilte at this point; both DF and Ed. read mac Reithe here, and HP thus emends the text. But the BDL reading is supported by other Irish versions: Cuilte Usgar 7 Diarmuid (H, and similarly C). We need not, in fact, conclude that the later references to Mac Reithe preclude the name Caoilte in this line. As Robinson pointed out (“Gaelic ‘Ballad of the Mantle’”, p. 7), a character Caoilte mac Reithe was known in later tradition; he occurs in Scottish versions of the Banner Quatrains, bearing the Dubh-Nimhe (see chart accompanying discussion of XVI).

7 <go moill[e]>: ġom mȧȧl MS. It is difficult to be certain precisely what the MS form, and particularly the second element, intends to convey. The presence of puncta over the initials of both words may show that lenition is intended; in the second word ȧȧ could be an attempt to represent the diphthongs /əi/ (as the restored form suggests) or /au/ (thus giving go mall, “slowly”, rather than go moill[e], “with slowness”). ȧȧ also represents /Eu/ in this text (57) as well as /a:/ (14), and cf. MS mc rȧȧ for Mac Reithe (61). Cf. too MS conane (mȧȧle) meil (45), where the scribe’s first attempt mȧȧle could represent either mall or an eccentric shot at conveying maol.

Professor Ó Maolálaigh, however, notes:

‘I think we can be reasonably certain that ġom ṁȧȧl is for go mall rather than go moill[e]. There are other examples (even though mawle, mawl and mall also occur):

‘Note also that aa occurs for feall, thall in BDL.

‘We might expect a short a-vowel for ṁȧȧl as the Irish versions seem to have a corresponding short a-vowel, e.g. Conán Maol nar lag ar muir (DF, ii, 330, 2c).’

11 áille: The scribe originally wrote ȝall (i.e. gheal), which was cancelled at some stage and aelli placed in superscript. The first attempt may be a scribal error, perhaps related to the occurrence of -gheal in the preceding line; its inclusion would lose a syllable in the line, although, if elision is allowed between áille and is, the line would be a syllable short in any case.

Dubh <Ros[g]>: dow ros MS. It seems most likely that the scribe has inadvertently omitted a final -k in the second word.

12 The line lacks a syllable as it stands, and HP reads agus for is (MS as).

14 The use of guth and rádh with the verb do-bheirim seems rather strange, although it is quite possible to understand the line as “they gave their voice and their speech [to express the view that...]”. Murphy (DF, III, p. 159) makes the suggestion, based on the BDL reading, that the line may once have had Tugsad cuir agus rátha, “They gave securities and guarantees” (cf. RIA Dict. s.v. I cor 13, I ráth (b)). do-chuadar dha iomrádh (DF); Do bhadair ag iomarbhaidh (Ed.).

15 <t[r]ic>: The MS form teg may be based on a miscomprehension of tic in the exemplar, with the BDL scribe failing to remember that a vowel in superscript indicated the presence of r, before or after that vowel, in the expanded form. tric (DF and Ed.).

16 an gcoimhionraic: in goyt Inrylk MS. Final -lk in the MS form Inrylk may simply be an attempt to represent palatalisation.

17 The line is hypermetric as it stands, and its restoration requires that Adubhairt be regarded as disyllabic; thus Dubhairt (HP), but the form may have been pronounced aduirt.

an innilt: Presumably this is the ainnear already referred to in 4, who now plays devil’s advocate.

18 cúileach cearnach: Cwlyt carnich MS. The line would mean: “The world is full of nooks and crannies”, i.e. “Don’t think that you have seen it all”. Cf. Sc. G. gach cùil is ceàrn. (I owe this restoration to Professor D.S. Thomson.) colach ceirdeach (DF); Corrach cerda (Ed.).

20 nach d’rinn feis: naċ drynn fes MS. The MS clearly shows the influence of vernacular Sc. G. in the form of the verb. nar fheis riam (DF); nach derna feis (Ed.). BDL appears to have a modified form of the construction found in Ed.

22 dar rochtain: dar rochtin MS. It is possible that final -r of MS dar appears in anticipation of the following initial, and that we should read da rochtain, “towards them”. Cf. da ffechain (DF, and similarly Ed.).

23 aonbhrat: The MS form Einwrata attests the growth of an unhistorical final syllable.  Cf. brata (61).

24 Although there is not complete certainty that MS E represents é and not í this line makes better sense if it relates directly to the brat rather than its wearer. The point would seem to be that the mantle is woven from a single thread; this would make it particularly special, and it is, after all, a magic garment. & í ’na háontshnáithe (DF, and similarly Ed.)

28 suidhis: After MS swis the scribe has written so<i>s (the third letter possibly having been a originally), which was later cancelled. It may be that so<i>s came about by dittography, or it could conceivably be suas.

32 The sense of this line would seem to be, “What has caused you to wear your single-thread[ed cloak]?” The peculiar DF reading gad bheir tú gan áontshnáite (transl. by Murphy as “what induces you to wear not a thread?”) may have come about by confusing an earlier con, “with”, with gan, “without”. Cread bheir tú ad taonshnáithe (Ed.).

34 This line is difficult to relate syntactically to the preceding line, and would appear to be a parenthesis; perhaps we should translate both as follows: “It is a magic property of my beautiful cloak [that when it accommodates] a woman within its single thread...” This involves regarding ach as virtually redundant in terms of sense, although required for metre, and relating aonshnáithe to the mantle rather than the woman; note that MS EInayġ need not necessarily imply a prefixed h in the restored form, any more than MS Iyn nayġ does in similar circumstances in 24; cf. MS In van for aoinbhean (22). DF seems confused: bean ann gan aontshnaithe (cf. 32 n); Aon bhen is í na haontsnaithe (Ed.).

35 <nochan fhaigh>: Noċt chay nayġ MS. As presently restored, the line lacks a syllable, and the problem seems to lie in this phrase. Given the MS form Noċt, it is not impossible that the scribe was thinking in terms of [a-]nocht chan fhaigh. However, HP restores to nocha nfhaghann, and it may be that the loss of the 3 sg. pres. indic. conjunct ending -ann in Sc. G. (IDPP, p. 132) has created the difficulty. nach bfaghann (DF); Nach faghadh (Ed.).

síoth dhen: This follows the original MS reading schee ȝin, which the scribe later cancelled, placing dein fame, i.e. díon fám in superscript. Transl.: “nobody gets peace from the mantle”; or, following the superscript, “nobody gets protection (i.e. complete cover) under my mantle”. a díol don (DF and Ed.).

bhrot: The scribe appears to have written wrat originally, and to have corrected to wrot for the sake of rhyme.

36 aonlocht: This follows the original MS reading aynlocht, which the scribe later cancelled, placing <râlocht>, i.e. rálocht (“excessive fault”), in superscript. rólocht (DF); áon lochd (Ed.).

38 <mear>: m<a>r MS. The second letter of the MS form would seem to be a rather than o. maol (DF and Ed.).

39 bhfeasmaois: westmist MS. Cf. vestmost (63). See II, 4 n.

40 a tug: This verb form is at variance with normal classical E. Mod. usage in two respects: the intrusion of the preverb do ( > a) as a rel. pron., and lack of concord with the subject, both points probably reflecting vernacular Sc. G. (which would have a thug). HP restores to thugsad. Do rádhsat (Ed.); glór na mban (DF).

42 racht: rachta MS. Cf. 23 n.

44 ruibh: ryf MS. The final letter of the MS form is clearly f. HP ris is probably based on a misreading.

45 Conán Maol: See 7 n.

48 an nighean: By an error of anticipation the scribe first wrote i’ neyn before the verb, and then cancelled it. A similar error occurs with am brat (50).

51 <a bha sí dhi>: a wassi ȝyi MS. The line would thus mean: “she for her part was no better”. However, Stern (“Die Gaelische Ballade”, p. 297) offers another possible restoration, a mheasadh dhi, which would give the line the meaning: “no better was it adjudged to her”.

61 The line is hypermetric as it stands, with no obvious solution. Murphy (DF, III, p. 160) suggests reading tug for tabhair, but this seems too drastic in view of the use of tabhair elsewhere (37).

64 dā h-iomlaoidibh: Problems have arisen in previous interpretations of this line (see HP n) because of splitting up MS hymlitdewe into two words, a reading encouraged by the slight gap between t and d. It is most satisfactory to regard the form as dat. pl.

65 Do-<bheirim>-se: Di wary’si MS. In the MS generally, and in this poem, it would seem that y’ is comparable with ym (cf. y’ brat (50, 52) and ym brat (60), ym brata (61)). It would therefore appear probable that 1 sg. pres. indic. is represented here, rather than 1 sg. condit. (cf. HP).

69 do <bhf>eis: di weis MS. The MS form weis seems to attest eclipsis, although this is not normal after the prep. do (“of”), which usually lenites. Thus do fheis (Stern, “Die Gaelische Ballade”, p. 298).

72 The line is hypermetric as it stands, and HP emends to go ladhair a lúdagáin, “as far as the fork of her little toe”. In the actual BDL text, the use of no go implies that the succeeding word is understood as a verb. This could well be a modification of the reading suggested in HP, resulting from the Sc. G. tendency to use ladhar for the hoof of an animal! dhi go ladhair a láodagán (DF); Di go lár a luidiocáin (Ed.).

lúdagnán: The MS form lwdignane may represent a dialectal variant of lúdagán.

73 d[o] fhuaras: The MS form doaris would appear to be responsible for the loss of a syllable in the line.

74 [o] Dhiar<maid>: Lenition of initial d is attested in the MS, and this suggests that the prep. o was understood to be present at this point in the line. do mac í Duibhne do Dhíarmaid (DF, and similarly Ed.).

75 do r<ei>seadh: Di reissi MS. The form of the verb would seem to be 3 sg. condit. of roichim (RIA Dict. s.v. reich, ro-saig). For the probability that MS -ei- in this instance represents -ei- in normal orthography, cf. di weis (67 n). do rachadh in brat dam go lár (DF).

81 ’s: MS is requires to be read as ’s for line length.