DAVID LUMSDEN'S ACCOUNT OF THE lute sources in the 1950s provided prospective lute scholars not only with the first list of lute music concordances, but also dates and an approximate chronology for all the sources available at that time. With the progress that has relatively recently been made in paleographical technique coupled with more recent studies of specific manuscripts, it is evident that a large number of sources were wrongly dated, though not often by many years. However, as is discussed in Chapter 2, since the development and maturity of this repertory is compressed into such a relatively short time-span, a mis-dating of 15 years can be quite significant, particularly if the source is thereby inconsistent with its peers. In most cases, re-dating the sources with the new information to hand has not had unexpected repercussions, nor has the new evidence substantially contradicted existing research. On the other hand, some sources have raised issues, though not necessarily those of dating, that require detailed examination, particularly where established research appears to have explored all the necessary avenues of investigation fully. Manuscripts which have merited closer examination are discussed below.[1] Some of the scribal concordances are particularly controversial as they call into question many of the premises on which previous research has been based. It should be emphasized that the handwriting of the scribes is never the only evidence considered in any case.

It is frequently unnecessary to describe a hand in detail when a tabulation of its features is sufficient. In the case studies below, much descriptive detail has been omitted in favour of graphic demonstration, and only the points that are particularly salient are discussed, together with the resulting implications for those manuscripts.

There are numerous other examples of scribal concordances in currently unrelated sources that are not discussed here. The concordances do not, however, alter information about the sources themselves to any significant degree. These concordances are noted in table 19, p.ØØ. Samples of the work of each scribe are given in Appendix 1, where they may be compared.


LORD HERBERT OF CHERBURY'S LUTE BOOK: Cambridge, Fitzwilliam Museum Ms.Mus.689. Book written by two scribes with similar hands, and one further scribe. Currently dated 1624-40 as a result of some dated pieces of music composed by Herbert. Signed by Herbert on the front endpaper.

Three writers have examined this manuscript in some detail since Lumsden's brief and incomplete examination of the book in Sotheby's salerooms. As their work has been so central to the subsequent understanding (or misunderstanding) of its contents, these studies were reviewed in Craig 1991 when the manuscript was re-examined.[2] Thurston Dart[3] provided a comprehensive biography of the owner, and Curtis Price[4] made a study of the contents and discovered certain organizational features that accounted for some of the anomalies for which Dart was unable to find a satisfactory explanation. Both provided an index of the complete contents of the book[5] listing the original ascriptions, but neither attempted to list concordances either for those pieces Lumsden had examined, or for those that he omitted. Lumsden thought (falsely, as it later turned out) that the English music in the book was written in vieil ton but all the French music made use of transitional tunings, so he ignored a large portion of its contents as he believed they did not fit his vieil ton or date criteria. A further brief examination of the book was undertaken by Matthew Spring[6] attempting to fill in the gaps left by Lumsden, but in fact not doing so, as he also limited himself strictly to the music that fitted the scope of his thesis. He aimed to provide a study of the later English repertory that the book contained (i.e. that dating from after 1630), and the result was that he only examined the work of Edward Herbert himself and the otherwise unknown Cuthbert Hely. Price built on Dart's work, and Spring in his turn built on Dart, Price and Lumsden, so that together all four studies should constitute almost everything that can be ascertained about the compilation of the manuscript and its owner. Looking closely, though, it appears that all four writers overlooked or did not re-examine important elements in examining the manuscript.[7]

Before the study published in Craig 1991, the compilation of Edward Herbert's manuscript was generally accepted to span the years 1624 to 1640, and it was believed to have been written in three hands: Scribe A, Scribe B (the autograph hand of Herbert of Cherbury, 1583-1648) and Scribe C (responsible only for the music of Cuthbert Hely, and probably his hand).1624 was the year in which Herbert was exiled to Castle Island in Ireland--a location frequently cited by Scribe B, and possibly also referred to in the quotation on f.1r, from the second of Ovid's elegies, written from his exile near the Black Sea. This evidence, together with the probable time-filling purpose of the book, seemed good enough to Dart for the date of inception, but it seems unlikely that Herbert began the book before he made the inscription on the second flyleaf, naming himself as "Lord Herbert, of Cherbury and Castle Island". Until 1629, he could only claim the title of Herbert of Castle Island.

The terminal date relies on three factors. The date of Herbert's death, the dates provided in Herbert's autograph hand: 16[19],[8] 1626, 1627, 1628, 1639 and 1640;[9] and the assumption first made by Dart and maintained by Price that the two principal Scribes,A and B, were both Herbert of Cherbury, at different times in his life. There is no doubt that in several cases Herbert, whose autograph is only in the hand of Scribe B, added pieces to the collection after Scribe A had finished. The dates indicate the date of composition of their respective pieces, but do not necessarily also indicate the date on which they were copied. Indeed, the completed if inaccurate state of these copies, lacking any sign of ongoing compositional process, suggests that they were probably composed some time before they were added to the book. Further, the inscription on f.82r: 'Pavan of the Composition of mee Edward Lord Herbert 1627 ...' must have been written in after 1629, the date of his elevation to the English peerage.

It is also tempting to cast doubt on Dart's '1619' reading for one of Herbert's compositions, accepted but not verified by both Price and Spring. It has not been possible to clarify this, partly through a lack of ultraviolet equipment in the Fitzwilliam Museum, but mainly because the edge of the page is now missing. The doubt is crystallized when we note that the piece is essentially the same as that on f.82r (dated 1627), described by Dart as 'another birthday piece'. It is more likely that the damage to f.79r had obscured the date to too great an extent for Dart to make an accurate reading, even under ultraviolet light, and that in fact this piece is also dated 1627, leaving Herbert's earliest piece to date only from 1626.

Dart does not discuss the hands, simply assuming they both belonged to Herbert, despite the fact that the inscription on the front flyleaf and all Herbert's pieces were undoubtedly written by Scribe B, as is the first piece in the book. It was rare with any scribe from this period to leave his first leaf unused, since paper was by no means a cheap commodity, and when a scribe had a collection that was clearly going to require all the space in his book, he would not have left any leaf unused without very good reason.[10] Price's arguments for Scribe B having worked after Scribe A had finished are relatively unassailable, but still do not adequately explain how the first piece in the book came to be written by Scribe B. Perhaps Scribe B worked in the book both before and after Scribe A. On looking closely at both hands, several questions arose that seemed to call into question the rather seminal assumption that the whole book, with the exception of the music by Cuthbert Hely, was written by Edward Herbert.

ex. 47 Herbert f.46v, reduced to 82

Price describes the quite considerable variants between Scribes A and B and reasons that the differences were attributable to 'an unconscious change of writing style over the years', a reason that hardly seems sufficient for the quite radical habitual differences he describes. The one feature which was least likely to be an unconscious change, that is the alteration of the terminal word from fine (Scribe A)to finis (Scribe B), Price suggests may be deliberate, but does not offer a convincing reason for it. He discusses f.46v of the manuscript, shown here in example 47, with both the scribes at work on one page.

This also illustrates well the variation in slant between the two hands, as well as a number of other noteworthy elements not otherwise described by Price. Scribe A writes his letter 'b' with a single circular motion (much as a modern writer would shape the number 6), while Scribe B uses the down-up-down motion of a modern 'b'. Scribe A always writes to the end of the line, but never goes outside the ruled block, and thus does not always complete the line with a full bar. Scribe B always ends the line with a full bar, even when this means leaving a small space unused, or having to extend the pre-ruled lines. One final note on this page; although the two pieces were clearly copied by different hands, the titles were just as clearly written by the same hand. It could be argued that Scribe A was leaving a note for himself that this was the piece that he wanted copied next, but it may be taxing credibility to suggest that he did not fill in the piece for some 15 years. If the similarity of the hand in the titles is taken as an argument for the two scribes being the same, this does not explain why the tablature hand has changed so radically, but the title hand has not.

Not obvious from this page is the apparent inexperience of the copying of Scribe B (Herbert) when compared to that of Scribe A. Like most experienced copyists, Scribe A tailors his pieces to fit exactly the space for which they were intended, but Scribe B seems to lack this ability to judge, and ends up having to extend lines, add new staves, and sometimes meander onto unrelated folios to complete a piece. In many cases, this could be explained as the scribe 'squeezing' extra pieces into the book, but in as many cases that explanation is not sufficient. The lack of accuracy in Herbert's copying of his own pieces--all only in hand B--also confirms the probability that Herbert was not as good acopyist as the owner of hand A. If the hands both belonged to the same scribe, it is unlikely that his copying skills would have deteriorated over the years, rather the reverse. As a composer, Herbert was hardly to be considered competent.

By all accounts,[11] Herbert was somewhat arrogant and his pride in his own music was apparent, all signed in generous detail--but only by Scribe B. If Scribe A is Herbert copying 1624-8, why did he not include the Castle Island and earlierpieces at that time, and why did he not inscribe the flyleaf at that time? Perhaps he did. Dart was convinced that the two hands belonged to Herbert. Price introduced an element of doubt, but still concluded that the hands both belonged to him, and were entered into the book a number of years apart.In fact, this may only be the partial truth. It seems unlikely that the scribe who initiated the copying should begin anywhere but on the first folio of the book, and equally unlikely that Herbert should inscribe the flyleaf and put in the first piece in the manuscript some 20 years after the work had been started. What seems a more likely explanation, is that Herbert employed a copyist for much of the work, possibly an Italian as this is the only known source--English or Italian--that refers to a 'Sr Danielli Inglesi'.[12] He copied the first piece in the book, and then left the Scribe A to do the main work. When his copyist had finished, or during his work, Herbert then added pieces himself, most notably those of his own making. Herbert is known to have employed a secretary who would have used a counterfeit version of Herbert's hand as part of his job.[13] The most likely probability is that Scribe A was Herbert's secretary.

One further possibility exists: if it is accepted that Scribe A is not Herbert, then it is possible that Herbert could have obtained the partially complete book from a third party, and added his own pieces in accordance with the previous owner's key divisions. However, this still does not explain Herbert's activity on the first folio.


Dart, Price and Spring concur in the belief that the last 15 pieces in the book were copied later than the main part of it. Here, Herbert's autograph hand has clearly deteriorated from its earlier appearances, though its principal characteristics--the shape of the letters, the slant and the manner in which he covers the page--remain unchanged.

If Herbert inscribed the flyleaf when he started to compile the book, then the book cannot have been started before 1629, when he became Lord Herbert of Cherbury. On the other hand, we have no very good reasons for supposing that the book could not have been started until the first page was written. The only facts we can offer with certainty are that Scribe B, Herbert of Cherbury, used the book after Scribe A had finished working. We cannot tell how long a gap there was between the activity of the two scribes, if there was in fact a gap, and we have no good reasons for supposing that Scribe A was Herbert; quite the reverse in fact. It is almost certainly true that the last few pieces in the book are later than the main body, but again without knowing how late, or early, the main body is, we cannot say how much later--5 days or 15 years? If we follow Dart's reasoning regarding the Latin tags, then the book was probably started in 1624, but much of the work done by Scribe B must have been started after 1629. On the other hand, the Latin tags are written by Scribe B, which implies either that they date from after 1629, or that Herbert's hand did not change appreciably from 1624 to 1630. Perhaps the whole book dates from c1630, with the exception of the last 15 pieces,that date from 1640.

Dart gives very plausible reasons for the book having been started in 1624, but his similarly plausible account of its composition has been convincingly dismissed by Price, so his date of inception may not be as unassailable as if first appears. His main argument for the 1624-8 period revolves around the two Latin tags inscribed in the text, though Dart himself gives evidence that Herbert still felt the disgrace of his exile long after it was over, and thus it is possible that he may have written the tags later,particularly as the literary evidence points conclusively to Herbert's (i.e. Scribe B's) activity dating from after 1630.Assuming the tags are in Herbert's hand, which is likely, and we know that they date from the 1630s, then we now have no reason at all for assuming the book was copied before that date, as Scribe A gives no dateable elements. It seems highly unlikely that Scribe A is Herbert, as the scribe clearly has quite a different level of skill from Herbert and this copying lacks the evidence of his personality so abundant in the appearances of Scribe B. In this case we are not faced with the problem that concerned Price, that two hands belonging to the same scribe could be so different. The only question then, is to decide when Scribe A was writing. Like many manuscripts that are approximately dateable, this manuscript contains a mixture of retrospective and modern music in an old style of tuning, but one that had not entirely gone out of use. Dart may have been correct in suggesting that it was a copy of a loose-leaf collection, gathered over some years prior to copying.

It is rarely possible to reach a satisfactory conclusion in cases such as these. When faced with the evidence above, the obvious course is to conclude that Scribe A was probably copying just before Herbert started work, and the likelihood is strong that the gap between their respective periods of activity was quite small. Thus the major part of the manuscript probably dates from the early or mid 1630s, with the exception of the last 15 pieces, that date from 1640.

ex.47a: Edward, Lord Herbert of Cherbury

[1] I am particularly grateful to Dr Malcolm Parkes, Keble College, Oxford, who has given freely of his time to engage in detailed examination of these sources, and also to Detective Constables Hampshire and Short in the Oxford Constabulary Cheque Fraud Office for discussing modern techniques of examining hands to detect forgery.

[2] The following discussion is a summary of the findings discussed in the article: as only the conclusions are reproduced here, readers are referred to the original for detailed discussion.

[3] Dart 1957.

[4] Price 1969.

[5] Dart only in the reprint of his article, and with many inaccuracies in readings. Both Dart 1957and Price 1969 modernized or standardized the use of the letters 'u' and 'v', and also 'i' and 'j' which were interchangeable in the seventeenth century. Thus original readings of e.g. pauan became pavan. Scribal preference for either letter can, however, be significant, so original spellings are retained exactly in all transcriptions of primary source material.

[6] Spring 1987A.

[7] I am grateful to Dr Victor Coelho of the University of Calgary for his comments on the scribes in the book, which led me to re-examine their work in greater detail.

[8] Folio 79r. This is unreadable in the manuscript, but Dart in Dart 1957read the number '19' under ultraviolet light.

[9] On folios 13v, 82r, 78r, 90v and 90r respectively.

[10] The exception here is Sampson, in which the original scribe left not only the first two leaves unused, but also left unused pages later in the book.As the collection is comparatively short, it is not possible to tell whether these omissions relate to a copying order, or were more random in nature. This is particularly unusual for a pedagogical book and all other indications suggest this was its original purpose.

[11] Mostly written by Herbert himself, and reviewed in Dart 1957.

[12] Folio 3v. Other titles early in the book refer to English and foreign authors as 'Sr' rather than 'Mr', the form in which they appear later in the book in both hands A and B.

[13] See Chapter 4.

The Lute Page Julia Craig-McFeely's Thesis Index