TWO

The Manuscripts and
the Songs

In quality, extent, and range, Agnes Lyle’s repertoire constitutes a major piece of data for students of balladry, oral tradition, and folklore. While William Motherwell probably took down only a portion of the songs this singer knew, this surviving repertoire is still remarkably extensive for the period. Motherwell was principally interested in ballads, and he found in Agnes Lyle a singer of diversified ballad-singing technique. Her standard quatrain ballads reveal the full control of compositional elements usually associated with formulaic oral singing. Her songs in other meters reveal adaptations, adjustments, and even failures of oral technique. And a few items bear all the earmarks of aurally memorized pieces. But Motherwell’s one extant notebook and his ballad collection between them authenticate each individual item, and guarantee that these items, taken together, constitute an oral repertoire, accurately recorded. The reader, then, once he understands the nature and significance of these two important documents, can have assurance that a text has been recorded accurately and attributed properly.

Motherwell housed his formal folksong collection in a single document usually called the Motherwell Manuscript. The early pages of the Manuscript are heterogeneous in character, but the last six hundred and more pages he devoted almost entirely to ballad material, each text written out with great care. In the Motherwell Notebook, on the other hand, the collector created a much more informal document. Through its pages he scattered ballad texts, lists of singers and their repertoires, statements of cash outlays and loans, outlines and notes for the “Introduction” to the Minstrelsy, and antiquarian matters. On page 176, to take an extreme example, he entered brief journal entries for a week in August of 1825 and for a weekend in July of 1826, followed by the detached and unexplained name “D Boone.” Obviously Motherwell, like many people, used notebooks in a rather casual and unorganized way, writing notes relative to different matters in different sections of the same book.

Thanks to the efforts of Francis James Child, each of these documents survives in two versions, the autograph original and the Child-commissioned copy. The autograph Manuscript was in private hands in Child’s day (M. C. Thompson, Esq., of Glasgow) but is now part of the Glasgow University collection. The Notebook may have belonged at one time to Motherwell’s printer, John Wylie. In 1874, when it was copied for Child, the copyist noted that the original belonged to “Mr. J. Wylie Guild [sic.].” It has disappeared from circulation more than once but at last report was in the library of Pollock House, Glasgow (Montgomerie 1966, 5; Lyle 1975, xii). The Child copies of both documents are in the Houghton collection, Harvard.

The Manuscript is a hefty folio, its pages numbered slightly irregularly through 697. The Notebook is a small octavo of 178 numbered pages. For the Harvard copy of the Manuscript the folio has been copied onto quarto sheets and bound into two volumes. At first Child told the copyist, J.H.M. Gibbs, to make excerpts from the Manuscript. But when Gibbs had copied about fifty pages of excerpts, Child changed his mind and ordered a complete copy. Gibbs went back and added every item he had skipped, completing the copying on January 7, 1874. Child had Gibbs’s work collated by M. H. Dalziel, who rearranged the pages so that the items would come in the order they fall in the folio, not in the order in which Gibbs had copied them. Interestingly enough, Gibbs received £4.1.9 for copying, but Dalziel received £.5 for collating. In 1886 James Barclay Murdock collated the copy and original a second time, inking red numbers in the margins to indicate the original pagination. Murdock was the same meticulous worker who had produced in 1875 the Harvard copy of the Notebook, an exact facsimile in size, pagination, and even the use of pencil or ink, of the autograph. For most purposes the Harvard copies of the Manuscript and Notebook are still used, as they were in Child’s day, interchangeably with the originals.¹

The Notebook offers occasional vivid glimpses of Motherwell the field collector in action, but the Manuscript offers a gauge of his development as collector over the entire period in which he was most involved with balladry. For purposes of analysis, the Manuscript divides into three segments, pp. 1-60, pp. 61-251, and pp. 252-end. The first sixty pages of the Manuscript contain songs of every description: true ballads from old women, bawdy songs for “The Paisley Garland,” children’s songs such as “The Frog and the Mouse,” sentimental songs such as “Somebody,” and historical songs such as “Captain Kidd.” Though Motherwell collected many of these first songs in the field, he found others in manuscripts or printed collections. It is impossible to know precisely when Motherwell began filling in these early pages of his Manuscript. The first text, “Lady Maisry” (Child 65E), was collected from Mrs. Thomson of Kilbarchan, 25 February 1825, but the earliest dated texts are “Child Maurice” (Child 83B) and two other ballads collected from Widow M’Cormick of Paisley, 19 January 1825, entered on page 255 ff. By the end of 1824 Motherwell was involved with printer John Wylie and other friends in producing a ballad book, Minstrelsy Ancient and Modern. At the same time, as he confided in letters to Robert Smith and Charles Kirkpatrick Sharpe, he had in mind a pamphlet or “garland” of less edifying pieces for a more select audience. In a letter of 2 January to Smith (Montgomerie 1958, 153), he calls it “The West Countrie Garland,” but in the Manuscript he changes the name to “The Paisley Garland.” The change from the January title for the garland, the February text on page 1, and the heterogeneous nature of the early contents all indicate that Motherwell began the Manuscript, sometime around March 1, as a convenient repository for pieces which interested him, especially if they might be used in either the Minstrelsy, of which he was probably not yet the editor, or the proposed garland.

After about 60 pages, however, the Manuscript shows a change in Motherwell’s conception of his collection. He includes no more “Paisley Garland” items (indeed “The Paisley Garland” never saw print, so far as is known). He now focuses exclusively on ballads, though a few of these ballads, such as Mrs. Storie’s variant of “Paper of Pins” and Mrs. Birnie’s variant of “Pretty Peggy O,” are pieces that Child did not elect to canonize. Motherwell now seems dedicated to collection for its own sake and seeks out multiple variants of the ballads, even though he obviously can not secure publication of every single variant. Most important, he confines himself to what he considers oral texts. Although many of these are supplied by his friend Peter Buchan, more than two hundred come from his own field collecting. He gives the names of some twenty-five informants, many of whom he visited repeatedly. Thus he transforms himself from a culler of old volumes to a cultivator of old singers, and some not so old.

In the first two segments of the Manuscript, Motherwell concentrates on text rather than singer. Seldom does he give more than one song at a time from a singer and never more than two songs in succession. Instead, he attempts to organize the material by variants. Pages 196-218, for instance, contain two versions of “Clerk Saunders” (Child 69) followed by three versions of “Johnie Scot” (Child 99). Similarly, Agnes Lyle’s version of “The Wee Wee Man” (Child 38) follows two versions of “Kempy Kay” (Child 33), another ballad about a creature of strange physiognomy. In these pages Motherwell is not always careful to record sources. Seventeen ballad texts in the second segment as well as a number of the songs in the first segment have no ascription of any kind.

In the third segment of the Manuscript, as in the first two segments, Motherwell is careful to preserve texts just as he heard them. But he now pays much more attention to details of singer, time, and place. In more than four hundred pages he errs occasionally, but omits ascription only twice. In each case the unascribed text follows a text sent to Motherwell by Peter Buchan, and in each case the unascribed text is probably to be understood as coming from the same source.² This care to identify an authentic text with a particular singer and an exact place and time is what sets Motherwell apart from the collectors of his day. And one ballad made the difference.

In the early months of 1825 Motherwell was almost fanatic on the subject of “Child Maurice” (Child 83). He saw in the ballad the basis for Home’s popular tragedy The Douglas, widely regarded at the time as one of Scotland’s greatest claims to literary distinction. Percy’s version of “Child Maurice” was still a popular stall ballad of the day, but Motherwell in his youth had heard an older, more “traditionary” version from a singer now dead. He kept finding other singers who remembered having heard “Child Maurice” the old way, but none could sing it that way themselves. What a coup it would be to print in the Minstrelsy “the only pure traditional version” (Letter to Scott, quoted in McCarthy 1987, 301) of the ballad which had inspired The Douglas. Letters to friends uncovered nothing. But then, on January 19, the very day on which he had written Charles Kirkpatrick Sharpe for help, he found an old lady, Widow M’Cormick, who sang him the song. For three months Motherwell sat on the song, hoping to find other versions or extra stanzas. Finally he wrote to the great Scott himself:

My reason for troubling you in this little matter is that if perchance any such copy may be now in your possession it would in all likelihood serve to correct some evident errors which occur in this one, and which though trifling I would rather wish to have it in my power to amend by the assistance of other recited copies than trust to my own judgment in doing so. Other copies too may possibly supply preferable readings in many places, and contain additions of material moment. . . . I mean to get this ballad inserted in a small 4to collection of Ballads now in the course of publication by John Wylie & Co. of Glasgow. Before doing so however I am anxious to learn whether you were ever acquainted with the ballad or had any copy of it which could rectify the text of the version now sent. (McCarthy 1987, 302)

Apparently Motherwell had already had one result of his collecting “inserted” into the Minstrelsy. The second fascicle opens (1827, 35) with a composite version of “Hind Horn” (Child 17) based on the version in Cromeck’s Select Scottish Songs, with additions from other versions, “which, joined to the stanzas preserved by Mr. Cromeck, have enabled us to present it to the public in its present complete state.” It seems safe to assume that Motherwell was responsible for the composite, since the versions from which the additions come are contained in his Manuscript. The headnote, too, has his tone and style, and promises the “Introduction” to the Minstrelsy, which he wrote, and the tune for the musical appendix, which he provided. This head-note betrays Motherwell’s first conception of ballad editing. There is, he implies, only one correct text of a ballad, and that text is often corrupted in printed sources. Fieldwork may turn up something of value to correct a text, but the great contribution an editor can make is to take the versions with significant variants and from them print a better composite, a more “correct” version than any previous text. He sees “Hind Horn” as a keystone in the fabric connecting balladry (“Minstrelsy Modern”) to romance (“Minstrelsy Ancient”), and he is obviously proud to have produced a more “correct” version or “set” for the Minstrelsy.

“Child Maurice” seemed such another important item because of the connection with The Douglas. The letter to Scott suggests that Motherwell wished to give it the “Hind Horn” treatment, but he found it more difficult to construct a composite text. Widow M’Cormick’s version had many characteristics, notably a more unadorned style, which made Motherwell hesitate to piece it out with stanzas from the Percy text.

Motherwell’s editorial philosophy at this point was the standard editorial philosophy of the day, confirmed by the tremendous critical and popular success of Scott’s Minstrelsy, first published in 1802-1803. But Scott had had twenty years to meditate upon that philosophy, twenty years of second thoughts. His reply to Motherwell’s letter must have come as quite a shock to the young collector. Though the great man had nothing of substance to say about “Child Maurice,” he had a good deal to say about editing:

Yet there are so many fine old verses in the common set [i.e., the Percy version] that I cannot agree to have them mixed up even with your set, though more ancient, but would like to see them kept quite separate, like different sets of the same melody. In fact, I think I did wrong myself in endeavoring to make the best possible set of an ancient ballad out of several copies obtained from different quarters, and that, in many respects, if I improved the poetry, I spoiled the simplicity of the old song. There is no wonder this should be the case when one considers that the singers or reciters by whom these ballads were preserved and handed down, must, in general, have had a facility, from memory at least, if not from genius (which they might often possess), of filling up verses which they had forgotten, or altering such as they might think they could improve. Passing through this process in different parts of the country, the ballads, admitting that they had one common poetical original (which is not to be inferred merely from the similitude of the story), became, in progress of time, totally different productions, so far as the tone and spirit of each is concerned. In such cases, perhaps, it is as well to keep them separate, as giving in their original state a more accurate idea of our ancient poetry, which is the point most important in such collections. . . . This reasoning certainly does not apply to mere brief alterations and corruptions, which do not, as it were, change the tone and form of the original. (M’Conechy 1865, xxxii-xxxiv)

Hustvedt, in Ballad Books and Ballad Men, cautions against giving undue credit to this letter (1930, 76-77). But the difference between the second segment of the Manuscript and the third segment indicates that the letter did have a strong impact. The occasional dates provided in the early pages of the Manuscript suggest that when Motherwell received Scott’s letter, sometime in May of 1825, he had filled in most of the first 251 pages, leaving only a few pages blank in hopes of finding further variants of particular ballads. But Scott’s letter and Motherwell’s own meditations seem to have shown the collector that what he had done so far was inadequate, if not misguided. He had overlooked the fact that a ballad is the product of a certain singer at a certain time and place. Henceforth his collection must include such data. The subsequent difference in the manuscript is modest but notable. The head-note on page 252 is typical: Mrs. Crum, Dumbarton, 4/7/25. In the remaining four hundred pages all items, with but two exceptions, are assigned to identified singers, with identified residence, and most items are dated as well.

This attention to details of ascription brought about a further shift of emphasis in the Manuscript. Gradually Motherwell became more interested in performers than in variants. Pages 255-264 contain three texts from Widow M’Cormick, with two more on pages 290-296. Pages 271-279 contain three texts from Marjery Johnston. Page 433 begins a long section of ballads from James Nicol. Page 550 begins yet another section of texts which seem to represent a single singer.

Motherwell’s handling of the ballads of Agnes Lyle, his most prolific informant, shows how gradual was this shift from a variant to a performer orientation. Presumably he entered the Widow M’Cormick texts as a unit shortly after receiving Scott’s letter. But two months later, when he was collecting from Agnes Lyle, he still felt the attraction of organization by variants, hence the placing of Lyle’s “Wee Wee Man” (Child 38E, Ms. p. 195) following “Kempy Kay,” and her “Babylon” (Child 14D, Ms. p. 174) following John Goldie’s text of the same song. And yet at the same time, impressed by the sheer number of songs that Agnes Lyle knew and the uniqueness of so many of those songs, he had already begun to enter the rest of her repertoire as a unit. This unit begins on page 331 and runs, with but one interruption, to page 397.

Motherwell was not entirely unique among Scottish ballad collectors in his interest in repertoires. Jamieson before him had set about collecting the ballads of Mrs. Brown of Falkland. Peter Buchan showed a similar interest and awareness with regard to the ballads of James Nicol. And Andrew Crawfurd, Motherwell’s protege, returned again and again to Mary Macqueen (Mrs. Storie) to collect as much as possible from her. Motherwell, however, collected the repertoires of a significant number of singers, noting, as Dave Harker points out “matters of performance style . . . and the importance of particular texts to particular singers” (Harker 1985, 65). He likewise understood the significance of differences in a given ballad as sung by different singers. And he recognized, influenced perhaps by Scott, that the record of a singer’s full repertoire serves to authenticate individual items in that repertoire: when the repertoire as a whole shows that a singer is representative of a singing tradition, each item in that repertoire demands respect and consideration.

Motherwell began as a dilettante antiquarian especially devoted to song. In time he came to recognize the intrinsic interest and importance of certain ballads that had first attracted him because of supposed literary connections, simultaneously becoming aware of highly traditional versions of these ballads being sung everywhere around him. Scott’s letter opened his eyes to the necessity of authenticating such versions by noting singer, time, and place and of publishing them in unaltered form. Finally, his experience with the singers of these versions led him to focus in his collecting not on the diachronic issue of how variants reflect the history of a ballad, but on the synchronic issues of what songs compose a singer’s repertoire and how those songs differ from other contemporaneous versions. He became interested, in other words, in the individual singers and the distinctive characteristics of their repertoires. The Motherwell Manuscript, in its three segments, documents this transformation of William Motherwell from haphazard collector of popular song to serious student of ballad repertoires.

The Notebook, containing as it does explicit lists of informants and of songs in the repertoires of informants, pays tribute even more than does the Manuscript to Motherwell’s awareness of the importance of individual repertoires. One section of the Notebook, pages 17-51, is especially rich in data about repertoires. This section contains Motherwell’s field notes for nine days in August of 1825. It opens with Mrs. Rule’s text of “Geordie” (Child 109G), collected on 16 August, and closes with Agnes Lyle’s text of “Turkish Galley” (“The Sweet Trinity,” Child 286Cf), collected on 24 August. Among other items included in this section are the rest of the songs collected from Agnes Lyle on 24 August, and lists of songs in the repertoires of Agnes Lyle and of another Kilbarchan singer, Agnes Laird. The Harvard copy of the Notebook bears the date “About 1826-1827” on the title page. Bibliographers usually repeat this date,³ which was probably assigned to the volume by an early owner who had not read it carefully enough. The use of the Notebook as a field book in August of 1825, however, demonstrates that the dates on the title page are not accurate.

The texts in the Manuscript are fair copies written out neatly in that penmanship that was Motherwell’s only fortune when he set out in the world to make a living. But the Notebook texts are working copies, with misunderstandings corrected, omitted words inserted, and easily understood words and stanzas abbreviated; they present, in fact, just the characteristics one would expect of texts written rapidly from oral sources. As such, they are closer to the performance, in fact, a part of the whole performance context. In transcribing these texts from Notebook to Manuscript, Motherwell, like anyone, made mistakes. In questions of disagreement, therefore, whether about a textual reading or, more importantly, about the ascription of a ballad to a particular singer, the Notebook texts take precedence over the neater Manuscript texts.

Despite Motherwell’s heeding of Scott’s words, there are some problems in identifying the ballads of Agnes Lyle. Child, on the basis of the Manuscript evidence, accurately identified 15. Kenneth Thigpen, the only scholar to publish a figure based on Child, finally settled on a count of 17 (1972). The actual figure seems to be 21 complete ballads, 1 nearly complete ballad, 7 fragments, and 3 titles.⁴ The 22 long items are all entered in the Manuscript and are all ascribed, although not unambiguously.

The 15 ballads which Child identified as from “Agnes Lyle” are the following:

1. “The Twa Sisters” (Child 10F)
2. “Babylon” (Child 14D)
3. “Sheath and Knife” (Child 15B-16F)
4. “Hind Horn” (Child 17C)
5. “Cruel Mother” (Child 20E)
6. “Fair Janet” (Child 64B)
7. “Earl Richard” (“Young Hunting,” Child 68D; 22 stanzas but incomplete, “the catastrophe wanting”)
8. “Little Musgrave and Lady Barnard” (Child 81J)
9. “Johnie Scot” (Child 99G)
10. “The Baffled Knight” (Child 112E)
11. “The Gypsy Laddie” (Child 200C)
12. “Geordie” (Child 209F, and fragment 209K)
13. “Bonnie Baby Livingston” (Child 222C)
14. “Lord William” (Child 254A)
15. “The Sweet Trinity” (Child 286Cf)

Five of these ballads, “Hind Horn,” “The Cruel Mother,” “Johnie Scot,” “The Baffled Knight,” and “The Sweet Trinity” appear in both the Notebook and the Manuscript. The rest appear only in the Manuscript.

Child identified two more ballads, following the Manuscript, as from “Agnes Lile, Kilbarchan.” The difference in spelling however proves to be immaterial: Motherwell was simply inconsistent in writing the name of his best informant. On page 331 of the Manuscript, for instance, he spells the singer’s name Lile, but spells her father’s name Lyle. Conversely, on page 367 he spells the singer’s name Lyle, but on page 370 spells both her name and her father’s name Lile. Lyle seems to be the more appropriate spelling. It is the spelling preferred for this family name in the Kilbarchan parish register, and it is the spelling Motherwell uses in the list of Kilbarchan singers on page 52 of the Notebook. The point would not need mentioning were it not for the fact that Child, by repeating Motherwell’s inconsistencies, creates the impression that there might be two singers with nearly identical names. The two ballads Child attributes to the singer under the name “Agnes Lile” are

16. “Lord Derwentwater” (Child 208A)
17. “The Braes o Yarrow” (Child 214C)

For one text the Notebook and Manuscript provide conflicting ascriptions:

18. “The Wee Wee Man” (Child 38E)

This was one of six songs which Agnes Lyle sang for Motherwell on 24 August 1825. Although Motherwell correctly attributed the text to Agnes Lyle on page 40 of the Notebook, when he copied it into his Manuscript he made the understandable slip of attributing it to Agnes Laird. The confusion is not a case like the preceding case of two names for one person. Agnes Laird is not simply Agnes Lyle under a married name. She is identified as a separate singer in the list of singers on Notebook page 52, and the Notebook also includes consecutive repertoire lists for the two women. Moreover, two of Agnes Laird’s six ballads are distinctive versions of ballads that Agnes Lyle also sang.⁵ Motherwell made another slip of the pen, the converse of this one, in attributing Agnes Laird’s “The Gay Goshawk” (Child 96D) to Agnes Lyle when he copied it into the Manuscript. Again, the Notebook evidence is clear, and the Manuscript attribution is in error.⁶

Child overlooked the fact that four other ballads which he printed from the Motherwell Manuscript also originated with Agnes Lyle. On page 331 of the Manuscript Motherwell had written:

The six following ballads I took this day from the recitation of Agnes Lile Kilbarchan a woman verging on 50 daughter of ———— Lyle a customary weaver of Locherlip who died 14 years ago aged 80. She learned these ballads from her father.

Child correctly attributed the first two ballads, “Lord Derwentwater” and “The Braes o Yarrow,” already mentioned. He published the remaining four without attribution:

19. “Sir Patrick Spens” (Child 58E)
20. “The Eastmure King” (“Fause Foodrage,” Child 89B)
21. “Mary Hamilton” (Child 173B)
22. “Jamie Douglas” (Child 204G)

Of these 22 items, 6 had appeared in print before The English and Scottish Popular Ballads: “The Wee Wee Man,” “Gypsy Laddie,” “Lord Derwentwater,” “Braes o Yarrow,” “Bonnie Baby Livingston,” and “Lord William.” Motherwell had included them in his Minstrelsy, making minor editorial changes for the reading public in accord with the license given him by Scott’s letter to correct brief “alterations and corruptions” so long as he preserved the “tone and form of the original.” These corrections, of course, make the Minstrelsy texts less authoritative than the Child texts derived from the Manuscript and Notebook.

The 22 ballads and the “Geordie” fragment do not constitute the whole known repertoire of Agnes Lyle. The Notebook adds the first stanzas of six songs and the titles of three more, making a total of thirty-two items. This material is contained in a memorandum similar to the Agnes Laird repertoire memorandum already mentioned. This second memorandum follows immediately upon the first, and may well have been drawn up on the same day, 18 August 1825. The full text follows:

Memorandum to take from Agnes Lyle Kilbarchan the ballad she has beginning

This memorandum and the pages which bracket it provide a vivid glimpse of the ballad collector at work in the field. Motherwell had already made several visits to Kilbarchan, notably on 18, 19, and 27 July, when he collected a total of fifteen ballads from Agnes Lyle and one from Janet Holmes, who may have been the sister mentioned in one of the Notebook memoranda. In Kilbarchan again on 18 August, he dropped in on Agnes Laird and Agnes Lyle to find out what further songs these two women knew. In something of a hurry, apparently, by the time he got to the home of the second Agnes, he entered only the titles of familiar songs, but wrote out first stanzas of six less familiar broadside ballads. On 24 August he returned to Kilbarchan to collect full texts. The first item on his list, “There was a Lady Lived in Luke” (i.e., “The Cruel Mother”), was the first song he collected from Agnes Lyle that day. He entered it beginning on page 33 of the Notebook, immediately following the memorandum. The second song he collected, “Johnie Scot,” was the third song on the list. After that, singer and collector abandoned the list altogether for “Hind Horn,” “The Wee Wee Man,” and “The Baffled Knight.” After singing five stanzas of “The Baffled Knight,” Agnes Lyle stopped. In all probability she was unable to finish the song at that time, but promised Motherwell she would work it up for a later performance. Motherwell left three pages blank to allow room to add the missing stanzas and returned to his list. The next song on the list, “Turkish Galley” (“Sweet Trinity”), was the last song collected that day. As Agnes Lyle completed this song, Motherwell discovered he had no more room for songs in that section of the Notebook; beginning on page 52 was an earlier series of notes on singers and repertoires, including the list referred to earlier of “old singing women” of Kilbarchan and other towns. Motherwell’s list or memorandum would seem to indicate that he collected no more songs from Agnes Lyle that day. “Turkish Galley” is the last song marked “D°” (that is, ditto: got). According to the Manuscript he went on to collect at least three ballads from Agnes Laird that day. He was able to squeeze in “The Gay Goshawk,” already referred to, between the Laird and the Lyle memoranda. For the others he changed notebooks.

Field notes from that Agnes Laird session, and indeed from the rest of Motherwell’s collecting career, have disappeared. Yet these few pages provide a fascinating glimpse of the great collector. Like a good fieldworker he prepared the ground carefully and made preliminary notes. In the midst of the collecting session he could abandon those notes to let the singer sing the songs as inspiration suggested them. When inspiration failed, as it finally did, he had the list to fall back on. Furthermore, he paid Agnes Lyle a modest compensation for her help on this and other days—a total of 8/6 according to the expense account on page 156 of the Notebook. Thus the Notebook reveals Motherwell to be, in field technique as in critical understanding, a folklorist ahead of his time.

Expert though he became in field technique, Motherwell left no systematic description of his collecting. In the margin opposite stanza 10 of “Sheath and Knife,” in a copy sent to Charles Kirkpatrick Sharpe, he did provide one other vignette of his experience with Agnes Lyle: “The poor old woman while singing this verse absolutely wept and the ‘Adam Scrivener’ who wrote and listened did all but evince similar emotion.” A rather rueful footnote in the Minstrelsy may refer to Agnes Lyle too. Commenting on how ballad singers tend to regard their songs as true accounts, identifying for the collector the very castle, river, or oak where ballad events transpired, he says:

I have, unfortunately for myself, once or twice notably affronted certain aged virgins by impertinent dubitations touching the veracity of their songs, an offence which bitter experience will teach me to avoid repeating, as it has long ere this, made me rue the day of its commission, (xxvii)

These affronted virgins taught Motherwell a hard lesson, but he learned it well. The Minstrelsy makes no further mention of fieldwork, but its editing reflects the understanding of balladry which the collector reached through his field experience.

Motherwell returned to Agnes Lyle on 25 September and secured a full text of “The Baffled Knight.” The fact that he never collected—or at least never entered into the Manuscript—the broadside pieces indicates how clearly he now distinguished between older ballads and broadside texts such as these. Nevertheless, the titles and fragments in Motherwell’s memorandum, all that remain of nine songs out of thirty-two in Agnes Lyle’s known repertoire, can not be ignored in any analysis of that repertoire.

“Johnie Armstrang” is doubtless “Johnie Armstrong” (Child 169). The spelling is Ramsey’s. Motherwell repeats the title and spelling when he cites the Ramsey text (Child 169C) in the “Introduction” to the Minstrelsy (lxii). The ballad tells a tale of teachery in which the king invites Johnie to court, only to have him killed. The three Child versions use the King’s Letter “theme” which Agnes Lyle uses in “Sir Patrick Spens,” “Johnie Scot,” and “Lord Derwentwater.”

The song “Slippings o Yarn” has so far eluded identification. Clearly it is a weaving song, probably not the only weaving song in Agnes Lyle’s repertoire. As a weaving song it provides a direct link between the repertoire and the Lyle family profession, and this is its principal significance. Fortunately the song traditions to which the other bits and pieces belong can be identified at least tentatively. Consequently, plot summaries can help supply the lack of full texts, while clues in the surviving stanzas can hint at particularities in this singer’s treatment of these songs.

Motherwell himself identified “Fair Margaret of Craignargat,” the first title on the list, as “a common stall ballad sixty years ago.” The notation “got” is puzzling, for there is no text in the Manuscript. There is, however, a full text printed in Charles Kirkpatrick’s Sharpe’s A Ballad Book (1823; 1976). When Motherwell noted that he had “got” this piece, he may have meant only that he had found the Sharpe text. This text tells in typical broadside language the tale of a mother who dreams that a raven carries off her baby. A wise woman tells her that the dream is a prophecy of unhappy love for the child. When the child reaches marriageable age she passes over many suitable suitors to choose a young thief. She leaves with him, bearing her father’s curse. The ship carrying the pair sinks, and the girl dies convinced that her father’s curse has been carried out. The ballad is unusual among ballads of family opposition in that the opposition is justified. The catastrophe, too, is unusual, relying as it does on coincidence rather than the evil machinations of the unworthy bridegroom.

It is possible that the first fragment (“But sixteen years of age she was . . .) represents Agnes Lyle’s version of “Fair Margaret of Craignargat.” The phraseology, especially in the last line of each stanza, is reminiscent of that of the Kirkpatrick Sharpe text, especially stanzas 1, 2, 20, and 25. If so, her version is quite different from the stall copies and the memorized text which Sharpe printed, despite the phraseological echoes. The long prologue is missing, and the young man takes a more active role. It seems more likely, therefore, that the fragment represents some one of the many songs about a girl who loves indiscreetly and dies at the hands of her lover. There are several good candidates from this category, and the verbal parallels are even more striking than those with “Fair Margaret.” The first two lines of the fragment, for instance, recall lines 7-8 of an American text of “The Perjured Maid,” also known as “The Gentleman of Exeter”:

The same sort of diction occurs in “The Ballad of the Ladies Fall”:

Perhaps the closest fit is “The Oxford Tragedy.” The first five lines of the fragment recall lines 13-16 of the contemporaneous Peter Buchan text:

The plot of the fragment too, so far as it can be guessed at, parallels that of “The Oxford Tragedy,” but the parallel is not perfect. The Oxfordshire rogue must resort to threats of suicide to gain his wicked way and achieves the deed in kind while the erstwhile maid is in a swoon. The youth in Agnes Lyle’s version needs nothing more than blandishments to achieve his ends, for the maid is only too ready to be deceived. But whatever ballad it represents, it clearly comes from a print tradition: the elaborate syntax—especially the use of the word but three times in seven lines, with two different meanings—and the sententious moralizing betray the pretensions of a broadside hack.

The second fragment, unlike the first, is easy to identify. “The Week Before Easter,” or “The Forlorn Lover” as it is also known, has been popular throughout the documented history of English folk song. The Euing collection contains a broadside version from the late sixteenth or early seventeenth century (1971, no. 112). But even this early text bears evidence of previous circulation. Stanza 2 has lost the phrase “with oats,” giving that stanza a rhyme scheme of x a a b instead of the regular scheme of a a a b, and a similar loss has occurred in stanza 5. Moreover, the final lines of many successive stanzas show a vestigial secondary rhyme scheme. The final lines of stanzas 1 and 2 rhyme posies . . . roses; stanzas 9, 10, and 11 rhyme favour . . . ever . . . ever; and stanzas 14, 15, and 16 rhyme accuser . . . misuse her . . . abuser. This rhyme scheme seems too consistent to be accidental, but too haphazard to be complete. Apparently the Euing text is an abbreviated and otherwise modified reprinting of an earlier and lengthier version in which stanzas were linked together by a pattern of rhymes in the final line of each stanza. It seems safe, therefore, to place the origins of the song somewhere in the sixteenth century.

Post-Euing printings show further effects of circulation. An eighteenth century broadside reduces the text to ten stanzas and eliminates the interlocking rhymes completely (Roxburgh 1966, 6: 233-235). Baring-Gould in the nineteenth century prints only six stanzas (Baring-Gould and Sheppard 1895, no. 97). Cecil Sharp, collecting in the early twentieth century, found three-and four-stanza versions in which the song, always more lyrical than narrative in tone, finally became almost pure lyric (1974, no. 59). Despite its continuing popularity in Great Britain the song has apparently not been collected in North America west of Newfoundland and has no Laws number (cf. Karpeles 1970, no. 31; Peacock 1965, 2: 441-442).

The situation in the song is clear enough. A young man has given his love to a girl who proves faithless. In a fit of masochism he forces himself to attend the wedding and watch all the details of the celebration. The usual last stanza, in which he prays to die and be buried, introduces the association of weddings with funerals which will come up over and over in the analysis of Agnes Lyle’s ballads.

After these tales of faithless lovers it is a relief to turn to the third fragment (“A fair maid walking in a garden”). It is the first stanza of a classic tale of heroic fidelity. The ballad, variously known as “Pretty Fair Maid,” “The Single Sailor,” and “The Broken Token” (Laws N42), is widespread in both North America and the British Isles. The Cecil Sharp collection, for example, includes nine British versions (1974, no. 144). No early broadsides are known, however, and this fragment from Agnes Lyle seems to be the earliest record of the ballad. In the ballad a young man, usually a sailor or soldier, approaches a girl in a garden and asks her to marry him. She replies that she is already pledged to another, a man who has been gone for seven years now. In some versions, usually British, the man promises the girl riches if she will give up her old love, but she reasserts her fidelity. In other versions, usually North American, he suggests that her lover may be dead or even married to someone else, but she replies that if he is dead she will be true to his memory, and if he is married she wishes him happiness. In either case her intention is to marry no one else. Moved by her fidelity, the young man, in both British and North American versions, takes a love token—a ring or divided ring—from his pocket and reveals himself as the long-lost lover. In an ironic way this ballad too unites the leitmotifs of marriage and death spoken of before. But in its emphasis on fidelity and its provision of a happy ending it is unusual though not unique in the repertoire.

The fourth fragment (“There were two sailors were lonely walking”) seems to belong to a comic ballad, “The Basket of Eggs.” A Cecil Sharp text is remarkably similar:

The differences in numbers are immaterial, since one sailor and one girl are all that are involved in the subsequent plot. Agnes Lyle’s phrase “cauler air” may even echo some vague aural memory of the word “gold” in the second line. In the ballad one of the sailors offers to carry a basket of eggs for one of the girls. She asks him to leave it at the inn if he walks ahead too fast for her to keep up. Arrived at the inn, the sailor thinks to treat himself to fried eggs but, uncovering the basket, discovers a baby. Of course it turns out to be his baby, and he is happily reunited with the girl he has seduced and abandoned but not forgotten. On the face of it, this would seem to be a cheery piece. But an Irish treatment of the motif is more cynical: The sailor, completely innocent, is left holding the basket (which he thinks contains a bottle of rye), and the girl disappears. Agnes Lyle, as will appear, is quite capable of darkening the usual tone of a ballad. “The Basket of Eggs,” as the Irish treatment of the motif shows, offers possibilities for such an alteration.

The fifth fragment (“It was in the middle of fair July”) illustrates the process of association so important in the transmission of folksongs. The stanza, though probably a ballad fragment, derives most of its diction from popular and broadside song. The first two lines are usually associated with “Two Rigs of Rye” (Laws 0-11), about a man who, taking his constitutional on a July morning, overhears a pair of lovers. The boy, to test his sweetheart, declares he can not marry her because of parental opposition or some such obstacle. The girl expresses such genuine distress at the announcement that the boy, satisfied, promises to marry her. Four of the five versions of the song examined for this analysis begin with the same words as the Agnes Lyle fragment.⁷ The fifth, from Gardner’s Michigan collection, preserves only three lines of the first stanza, but one of these lines corresponds to the second line of all other versions and of the fragment. Despite these correspondences of diction, however, the Lyle fragment does not seem to fit with the plot of “Two Rigs of Rye,” nor does the phrase “rigs of rye,” or an equivalent, found in the first stanza of all five variants, appear in the fragment.

The third line of the fragment, with its “glint and a glancing eye,” certainly has a formulaic ring, but it has resisted all efforts to match it to a parallel in any traceable folk song. Parallels to the fourth line, on the other hand, “Abroad as I was walking,” are ubiquitous in British folksong and yet do not seem to occur in any songs which have the initial situation of this fragment. The phrase may have come to this song from “The Chain of Gold” (Sharp 1974, no. 80), which opens with two lines parallel in content to the first two lines of “Two Rigs of Rye” and which does include the phrase “Abroad as I was walking.” The parallel with “Two Rigs of Rye” even extends to mention of parental hostility, but in “The Chain of Gold” the hostility is real, and the ballad ends tragically. Nevertheless it is possible that some singer familiar with both songs could have substituted the opening line of “The Chain of Gold” for the tag line of “Two Rigs of Rye.”

If this is not a fragment of “Two Rigs of Rye,” and certainly not of “The Chain of Gold,” what song is it? A good candidate is “The Beggar-Laddie” (Child 280). The fragment preserves, in addition to the opening lines, the meter and rhyme scheme of “Two Rigs of Rye,” suggesting that it is a song associated with that tune. Keith, in his discussion of “The Beggar Laddie,” mentions “ ‘The Rigs o’ Rye,’ to the tune of which this ballad was sung” (1925, 228). Bronson confirms that the melodic tradition of “The Beggar-Laddie” is “perfectly distinct and unusually stable, and that this distinct and stable tradition is identical with the more popular of the tunes to which “The Two Rigs of Rye” is sung (Bronson 1959-1972, 4:250). The Child D text of “Beggar-Laddie” begins:

The parallels with the fragment do not seem forced; the walk on a summer morning, the mention of the sun, and the glimpse of someone attractive. The fragment, then, may tentatively be identified as the first stanza of “Beggar-Laddie,” sung as is traditional to the tune of “Two Rigs of Rye.” This ballad tells the story of a girl who elopes with a beggar, learns to regret the hardship, but finally has cause to congratulate herself when her beggar reveals himself to be a great lord in disguise.

Agnes Lyle’s fragment combines an introductory situation widespread in Anglo-Scottish balladry and song, a tune and two lines from “Two Rigs of Rye,” and two additional formulaic lines, one of which may have been suggested by its occurrence in a somewhat similar song. The thread of associations which holds this stanza together provides an outstanding example of one of the basic compositional techniques of traditional ballad making.

The final item (“As I Went Out on a May Morning”) seems oddly out of place—a Buckinghamshire local hunting song—and yet is in some ways the most exciting discovery among the fragments. It is a traditional reworking of a Stuart broadside text, “The Fox-chace or the Huntsman’s Harmony by the Noble Duke of Buckingham’s Hounds, etc.” (Roxburghe 1966, 2: 360). Baring-Gould places the hunt in the reign of James I (1895, xxxvii), but Chappell (Roxburghe 1966, 1: 359, note) and Maude Karpeles (Sharp 1974, no. 267) each place the events in the reign of Charles II. The first record of the song, after the Stuart broadside, is this fragment collected from Agnes Lyle. Next, late in the nineteenth century, Baring-Gould collected it in Cornwall, and not long thereafter Sharp collected it right in Buckingham. This record is skimpy, but the wide geographical distribution and the persistence through nearly three hundred years would seem to indicate that the song has been more popular than the record reveals. Roger deV. Renwich, in English Folk Poetry (1980), mentions songs of this type and reasons why collectors have overlooked them:

The eminent Yorkshire collector, Frank Kidson, summed up the feelings of many people—including folksong collectors—when confronted with a local song pertinent only to a certain district. In his Traditional Tunes [1938, Oxford: Charles Taphouse], he printed but two stanzas of a North Riding hunting song, allowing a comment to fill in for the rest of the text: “I have no wish to inflict upon the readers more than two verses of this effusion. Like all songs of its class, it runs to about twenty verses, and the prowess of every fox-hunting squire and yeoman of the district is chronicled; highly interesting to those who know the descendants of the persons mentioned, but rather monotonous to the general reader.” Monotonous to the general reader, perhaps, but evidently not to the folk of the relevant district itself. (113-114)

The present song, however, has circulated far beyond its “relevant district.” Furthermore, Agnes Lyle’s repertoire as a whole is not particularly friendly to the squirearchy.⁸ But Yorkshire, like Renfrewshire, was known for its handweaving. Renwick associates songs like this one with the Yorkshire weavers’ penchant for poaching. Perhaps this item is all that remains of a similar lively hunting-song tradition in Renfrew. Both the Baring-Gould and the Sharp versions rework the catalog of dogs, which the Stuart broadside confines to one stanza, into a rollicking chorus:

This jolly, convivial chorus surely deserves some of the credit for the broad enduring popularity of the song.

Like the preceding fragment this stanza from Agnes Lyle’s version of “The Fox Chase” is remarkable as a demonstration of the recreative handling of broadside material in traditional fashion. The text derives ultimately from stanzas 1 and 2 of the broadside. It begins with a formulaic “As I walked out. . . .” It compresses the content of twelve lines into four. Finally, it shifts the meter from a rather elaborate six-line stanza to a more conventional four-line stanza with internal rhyme in the third line.

The six fragments and the one identifiable title all belong to broadside material. Even the text of “Beggar Laddie,” if that is what the song is, has been shaped by a broadside hack at some point. Probably all the songs have more or less fixed texts, unlike the orally recreated pieces in the repertoire. And yet all show that the singer was no slave to any printed text. Taken line by line the fragments demonstrate varying degrees of reworking by a traditional singer or line of singers.

Twenty-two ballads, seven fragments (including the extra stanzas of “Geordie”), and three titles remain of the textual element of Agnes Lyle’s repertoire. The musical element has not fared as well. In fact, a casual glance through Child might lead one to think that Agnes Lyle was not a singer at all. In the headnote of “Twa Sisters,” the first Agnes Lyle ballad in The English and Scottish Popular Ballads, Child says, “From the recitation of Agnes Lyle, Kilbarchan.” In the headnotes of “Babylon” and “Sheath and Knife” he says the same thing. The word recitation comes from the Manuscript, but by that word Motherwell seems to mean oral performance. Thirty years ago William Montgomerie pointed out that a letter from Motherwell to his friend Robert A. Smith clarifies this usage:

I have been long searching for some person who can sing the ballad of Jamie Douglas. A copy of this ballad you will find in Finlay’s collection or as you may not get Finlay’s book so readily you will find it in Gilchrist’s ballads. Perhaps about Edinburg you may light on some one who recites it; hitherto in this quarter I have been unsuccessful. The reason I am so anxious to recover a recited copy is that I believe and have frequently been told that the song of “Waly Waly up yon bank” is part and portion of the same ballad but how incorporated some of my informers were uncertain others told me it was the concluding part of it. This is not improbable as you will find they assimilate in subject well, and their tune is the same. In one of my rambles lately . . . [I met a woman who] concurred in assuring me that she always sang and always had heard the ballad sung with Waly at the end. (1958, 154; italics added)

In the above letter Motherwell opposes a recited copy, obtained from oral performance or recitation, to a printed copy, obtained from a book such as Finlay or Gilchrist, and he implies that the recitation will have a tune, that is, it will be sung. Indeed he cites identity of tune between the ballad in question and the related song as a partial explanation of the peculiarity he would look for in a recited copy of the ballad.

In the case of Agnes Lyle there is really no doubt. In fact, Motherwell explicitly mentions her singing in the letter to Sharpe quoted above. At least two tunes have survived in transcriptions from her singing, and some other tunes can be identified as approximations of what she would have sung.

The two definite Agnes Lyle tunes are included among the 33 that Motherwell had fellow Paisley enthusiast Andrew Blaikie transcribe for the Minstrelsy. Because tunes had to be printed from engraved plates rather than from set type, Motherwell gathered these tunes into an appendix rather than affixing them to the ballads to which they fit. In each case, however, he provided a cross-reference, if one was appropriate, as well as printing the exact stanza to which the tune was sung. In emphasizing tunes as well as texts, in transcribing from some stanza other than the first, and in printing the exact stanza to which the tune fit, the Motherwell-Blaikie team was ahead of its time. The tunes from the appendix that can be assigned with confidence to Agnes Lyle texts are Tune IV to “Lord Derwentwater” and Tune XIX to “Lord William.” Emily Lyle (1972) has succeeded in identifying many of the other Blaikie tunes with appropriate texts in the Motherwell and Crawfurd manuscripts. Of the eight orphan tunes in the Minstrelsy, Tune XXVI might be linked tentatively with Agnes Lyle’s “Babylon,” and Tune XIII even more tentatively with “Hind Horn.” In addition, Motherwell left in his notes and Manuscript a few comments about musical aspects of her songs.⁹ So few and so faint, these echoes can tell us little about Agnes Lyle the musician. But from time to time they can amplify the text and must be attended to.

Music is closely associated with metrics. In the case of Agnes Lyle’s ballads, meter may be the single most important aesthetic element. Metrically, the ballads fall into four groups:

1. Ballads composed in quatrains of alternating four- and three-beat lines, with masculine rhyme a b c b, so-called ballad quatrains.
2. Ballads composed in couplets of four-beat lines with some sort of intertwining refrain.
3. Ballads composed in quatrains of alternating four- and three-beat lines, with feminine rhyme a b c b.
4. Ballads composed in quatrains of four-beat lines, with rhyme a b c b.¹⁰

Preliminary prospecting among these groups revealed a remarkable coincidence. The oral patterns discussed in the “Introduction,” that is, the binary, trinary, and annular patterns of organization, and the formulaic and commonplace patterns of diction and narration, appeared in abundance in the ballads composed in ballad stanzas. But they appeared less consistently in the ballads composed in couplets or feminine-rhymed stanzas. They appeared not at all in the two ballads composed in four-beat quatrains. This remarkable pattern has suggested the basic hypothesis of the present study: The consistent way in which Agnes Lyle uses oral techniques in the standard quatrain ballads, that is, her consistent oral style, indicates that she is indeed an orally recreative, or oral-formulaic ballad singer, one whose technique depends heavily on the standard quatrain. When she sings as an orally recreative ballad singer in other meters, the strain of working in a less congenial medium shows in inconsistent technique. Finally, in singing certain songs she does not use orally recreative technique at all. She sings them instead from memory.

Determining the orality of a repertoire is really a two-step process—and of course even then the determination can not be definitive. One must first determine that in style and technique the repertoire exhibits internal consistency—in the present case, consistency in its inconsistency. Then one must demonstrate that the repertoire is distinctive of and appropriate to a single singer in a single time and place. Part Two of this study deals with the question of stylistic consistency. Stylistic success is always the result of a precarious balance of forces in the oral architectonic of the ballad. Chapter 3 will look at the standard quatrain ballads, in which that balance is constant, in order to highlight the basic forces at work in Agnes Lyle’s style. Chapter 4 will look at ballads in which at least one factor is always upsetting the balance. In couplet ballads the factor is uncongenial meter, and the singer has to compensate, with varying degrees of success. In feminine-rhyme quatrain ballads, not only does the singer have to deal with an uncongenial meter, but in each case there is a further complicating factor for her to handle. Chapter 5 will look at ballads in which, despite complicating factors, she has achieved notable aesthetic success, suggesting why this should be so. Part Three will deal with the question of whether these assembled versions exhibit characteristics of a single controlling personality. Chapters 6 and 7 will show how Agnes Lyle’s distinctive treatment of two ballad leitmotifs, love and perfidy, is characteristic of this daughter of a weaver, singing in southwestern Scotland in the mid-1820s.