Scribal features of BDL

Classical approaches to compilation

In Ireland the written ballad tradition was maintained until the nineteenth century by scribes who compiled small poem-books in beautiful Gaelic script with hardly a blot.  DF is securely within this scribal convention, and appears to consist of several smaller poem-books, or gatherings of poems, collected into the one manuscript.  When a second, significantly different version of a poem became available to a scribe operating within this tradition, he would usually write out the second version in its entirety, rather than emend the first draft in the light of the second.  This approach is illustrated by DF, where two versions of the same poem are preserved (DF II, LXVI).  In different MSS compiled by the Ó Longáin family, one often finds variant texts of the same ballad, but there is no evidence to indicate that the scribes attempted to collate their versions at any time.

Scottish approaches to compilation

On the Scottish side, principles similar to those in Ireland are evidently followed in those surviving MSS which are compiled according to classical conventions.  In ballad collections made in Gaelic Scotland in the eighteenth century, however, one can find variant readings inserted between the lines or in the margins of certain texts.  The text of 'The Death of Fraoch' preserved by the Rev. James McLagan in MS 245, for example, clearly indicates that the compiler worked over his first draft in the light of a version (or versions) received at a later stage.  This is apparent in the cancellation of words and phrases in the first draft, and the substitution of preferred or alternative readings in the superscript or in the margins.

Distinctive features of BDL

As has already been indicated, BDL is distinctive as a MS within the classical tradition.  Its script and orthography set it well apart from the main stream, but so also does the general presentation of the material.  Unlike the clean pages of DF, BDL has no less than eight poems which contain untidy 'corrections' and 'emendations'.  These are Poems V, VII, IX, XX, XXIII, XXVI, XXVII and apparently XXII.  In poems V and XX, further quatrains are added to the original MS drafts, with editorial marks in BDL XX to show the required position of these quatrains, but neither text attests the substitution of alternative words and phrases within the original draft. With the other emended texts, however, such a process of substitution is very obvious; individual words, phrases, lines, and even entire quatrains, can be cancelled with preferred readings placed in superscript.  The extent of emendation varies, from the comparatively few alterations attested in BDL VII, to the remarkable flurry of secondary scribal activity found in BDL IX, in which substitutions abound.

Types of emendation found in BDL

The emendations attested in BDL fall into two broad groups.  The first group comprises alternative readings which are found as part of the normal texts of certain ballad versions later than BDL itself.  Emendations of this kind are evident in Poems VII (l. 36), XXIII (ll. 2, 8, 30), XXVI (ll. 2, 3), and possibly XXVII (ll. 11-12).  The fact that these alterations are attested in later versions indicates clearly that the scribes were operating within an evolving tradition, in which there was no concept of a fixed text, and that they were accessing particular poems at different stages in their creative development.  As the MS was compiled over a period of at least thirty years, and as the scribal catchment area was extremely large, and embraced both Ireland and Scotland, it is entirely likely that the scribes would have encountered variant versions of their poems.

Within this first group we can also place the additional quatrains which are found in the MS texts of Poems V and XX.  These quatrains are all found in the later versions of the poems concerned.  Poem V, however, attests the addition of an adventitious quatrain which may originally have belonged to another poem.

The second group of alternative readings consists of changes which cannot now be paralleled in surviving versions.  Poem XXIII, for example, shows scribal instructions for the transposition of qq. 3 and 4 of the first draft, but no version with the quatrains in the alternative order is known to survive (see also 6.6).  Such emendations may appear idiosyncratic in the light of modern versions, but they may once have been well attested in the textual family being tapped by BDL.

Poem IX, with its conspicuous amount of emendation, furnishes further significant insights into the material available to the BDL scribes.  In comparison with the text preserved in the late Acallam, which is likely to be earlier, the first BDL draft has certain distinctive structural features (e.g. the attestation of qq. 60-66).  The version(s) supplying the emendations evidently shared these characteristics, since alterations are made to the distinctive sections of the first draft.  This would suggest that, in this instance, the BDL scribes were working within a well defined textual family. Where the scribes made alterations to the original BDL draft, in the majority of cases it can be shown that the readings of the original draft are closer than the superscript readings to those attested in the earlier Acallam version.  Only in a minority of cases are the superscript readings closer to the Acallam.  The effect of emendation is therefore to take the BDL text away from the stratum shared with the Acallam version.

BDL emendations and textual development

There is some evidence to suggest (at least in the case of Poem IX) that the development of variants within ballad texts may be governed by regional factors.  Some alterations appear to reflect dialectal differences.  There may also be a link between the popularity of certain ballads and their tendency to produce variants.  The Frequency List of BDL ballads indicates the popularity of certain poems in later tradition, but the evidence which it furnishes may shed some light on the emendations in BDL.  The list shows that five of the eight BDL ballads with emended texts (Poems VII, XX, XXII, XXIII and XXVII) were very popular in Ireland or Scotland, or in both countries, after 1600.  The emendation of the texts in BDL could thus indicate that they were already very popular when BDL was compiled.  Their popularity may have contributed to the proliferation of versions.

On the other hand, two of the ballads emended in BDL, poems V and IX, appear only occasionally in later tradition.  Indeed, poem IX disappears from written record after DF (1627), but it has the most extensively emended text of all the BDL ballads.  Such apparently contradictory evidence may demonstrate that a change in taste came about at the end of the classical period; it is by no means improbable that both poems, and the genres that they represented, were very popular when BDL was compiled, but that they subsequently lost their appeal.  The developing tradition tended to favour narrative ballads with strong storylines, and was less inclined to support vague, generalised plots or esoteric catalogues of place-names.

If there is a link between the popularity of certain texts and the development of variant versions, this is not exemplified by the scribal presentation of all the BDL texts which later became very popular.  Poem XIII, on the death of Diarmaid, which may well have been composed in Scotland, was a firm favourite in later Scottish tradition, but the BDL text shows no sign of emendation, despite the clear evidence of variation within the later versions. Similarly, the 'banner quatrains' in poem XVI are not emended in the MS, despite their inherent complexity and their popularity (and substantial refashioning) in later Scottish tradition. We may suppose either that the Diarmaid ballad and the 'banner quatrains' were composed so close to the period of compilation of BDL that they had no time to develop significant variants, or that the BDL scribes were simply at the mercy of their suppliers.  If no variant version of a particular text came to hand, no variants were recorded in the MS, and arguments about the relative age of texts are irrelevant.  There is little internal evidence to indicate the relative ages of any of the BDL texts; none of the texts, as presented in BDL, need be any earlier than 1400, and it is hard to determine any chronological point at which it would be expected that such texts should develop variants.

BDL emendations as evidence for the emergence of Scottish versions

The emendations made to certain BDL texts are potentially useful in demonstrating the growth of regional variants of ballad texts.  It would be difficult, if not impossible, to identify all the regional features of the texts, but in three texts there is sufficient evidence to suggest that, in altering their texts, the BDL scribes were reflecting distinctively Scottish versions of their poems.

BDL XXIII, on the death of Conlaoch, was evidently composed in Scotland, though versions of the poem were also known in Ireland.  After the insertion of the first draft of the poem in the MS, the text was emended in the light of a version (or versions) which bore a close relationship to those circulating in Scottish tradition, although the scribal instructions for the reversing of the order of qq. 3 and 4 appear to lack the support of a comparable Scottish version.

The emendations made to two other poems suggest the emergence of distinctively Scottish text-types.  BDL V, describing the chase of Slievenamon, contains 17 quatrains which are very close to the version preserved by DF LVIII and in later Irish MSS.  Indeed, there are points of detail in the BDL text (e.g. the use of the ampersand, see l. 38 n.) which suggest that it goes back to an exemplar of the type represented by RIA MS 23 A 47.  However, the appearance of an adventitious quatrain (q. 18 of the edition), evidently appended to the text after the first draft was written, adds another dimension to the BDL version. Only the later Scottish versions contain the equivalent of the additional BDL quatrain, although we cannot exclude the possibility that Irish versions containing this quatrain may once have existed.

A similar pattern seems to be attested by the text of BDL XX, telling of the Fian's adventure at Eas Ruaidh.  The original BDL draft of this poem lacked several quatrains which are regularly found in later Irish and Scottish versions.  These were subsequently supplied by a BDL scribe who had access to a longer text, and they appear as qq. 32, 36, 38 and 39 of the present edition.  The original BDL draft (which lacks these quatrains) is probably closer to the archetype of the ballad, since the additional quatrains provide details which are not essential to the plot of the poem.  This expansion of the text was absorbed by the tradition in both Ireland and Scotland, and there is very little evidence to indicate whether the BDL scribes derived their additional quatrains from Irish or Scottish tradition.  However, there is another additional quatrain in the text, represented by q. 26 of the present edition, which is found only in surviving Scottish versions.  This could indicate that all, or merely one, of the additional quatrains in the BDL text were derived from a Scottish source or source(s).  The slender evidence provided by q. 26 does seem to support the possibility that a distinctively Scottish version of the poem was in existence by the early sixteenth century.

Very cautiously, therefore, we may conclude that the alterations made to the texts of BDL V and XX do offer some grounds for believing that, in the case of certain ballads common to Scotland and Ireland, distinctively Scottish versions were emerging by the time BDL was compiled.  The fact that the distinguishing features of these texts are additions to the original BDL drafts would suggest that differentiation had come about only comparatively recently, perhaps within the period during which BDL itself was compiled.  It is, however, important to note that these apparent Scottish characteristics are relatively minor, and that the BDL texts are structurally and often verbally very close to surviving Irish versions.

Scribal significance of BDL emendations

As has been noted in Distinctive features, the scribal emendations attested in BDL are without parallel within the classical tradition of MS compilation, and are not matched on the Scottish side until the eighteenth century.  Why should the BDL scribes have chosen to insert emendations of this kind, when it was not normal practice within Gaelic scribal tradition, and what may have been their overall purpose in altering their texts?

It does seem evident that the BDL scribes were particularly aware of, and responsive to, textual variation, and that this is to be ascribed to a general Scottish perspective whose origins cannot be pinned down with any precision.  It is at least possible that their approach reflects the influence of Renaissance humanism, emanating from the eastern side of Scotland, and mediated through centres of humanist scholarship such as King's College, Old Aberdeen.  Basic to humanism was a critical approach to texts, and an interest in Latin and Greek classical literature.  The achievement of Erasmus in producing a critical edition of the New Testament based on Greek manuscripts symbolises the aspirations and methodology of continental humanism, and these were reflected in a Scottish context.  While it is obvious that the BDL scribes cannot be compared with Erasmus, it would seem that they shared a similar consciousness of textual variants within the genres of literature which were their particular concern.  Overall, BDL does seem to represent an attempt to gather some of the best products of medieval bardic talent in Scotland and Ireland.

It does not seem at all likely, however, that the BDL scribes regarded themselves as textual editors in the more modern sense or that they were aiming to produce a variorum edition of their texts.  The impression given by their general practice of deleting original readings is that they were operating in terms of texts 'most recently to hand', and attempting to fuse their material so as to remove inconsistencies and produce a single version which was both up-to-date and fairly representative of the textual family within which they were working.  If BDL as it now stands had been recopied, it is highly likely that the scribe(s) would have followed the 'top line' only, and that the cancellations of earlier readings would have been obliterated.