7c

SSBOARD AND HIRSCH

THE MARGARET BOARD LUTE BOOK: Private library of Robert Spencer, Woodford Green, Essex. Probably dating from 1620-35. Book written mainly in the hand of Margaret Board [Boord] (see end-papers, initials on cover) baptized Lindfield, Sussex, 19 Nov 1600, m Henry Borne between 1623 and 1631 (see f.32v),[1] with a section of later additions c1630-35 towards the end of the book, possibly in a French hand.

THE HIRSCH LUTE BOOK: London, British Library Ms Hirsch.M.1353. Undated, but currently believed to originate c1595. Named after Paul Hirsch, twentieth-century owner of book. Written by at least five anonymous scribes. Initials on cover H O. No other information about original owner.

Board and Hirsch are two sources of mainstream English repertory, dated 1620-30 and c1595 respectively, since the publication of facsimiles of each manuscript.[2] According to the research published in the facsimile introductions, the manuscripts appear to have been written about 30 years apart. However, closer examination of the details given in this research reveals a number of anomalies that bring the dating of Hirsch into question. The situation is further complicated by the apparent presence of a scribal concordance between the manuscripts. This is neither unusual or problematic, particularly as the identity of scribes is often not known. However, in this case the scribe in question is Margaret Board, born in 1600,and therefore unlikely to have been one of the scribes in Hirsch, if it does indeed date from c1595.

Margaret Board began copying her lute book around 1620. Her initials are stamped on the cover, and her name appears frequently on the end papers of the book together with names of family servants, and her married name: Margaret Borne. Margaret married some time between 1623 and 1631,[3] so the book was clearly begun before 1623, but the repertory indicates that it cannot have been much earlier. The dating is well-justified in Spencer 1976, where he also traces the year of Margaret's birth: 1600.

Dowland himself copied two pieces among those of Margaret, and added hold signs to some of her work, indicating that he was probably teaching her at some point in the book's history, though his activity in the body of the book is only evident between folios 10v, where she adds his 'Lachrimae' to her collection, and 16v, where Dowland adds hold signs to her copying. This part of the book at least must pre-date his death in 1626. From the consistency of Margaret's hand, it seems that her copying occupied a fairly short period of time, and it is likely that the fascicle in her hand dates from about 1620-25.

Board was almost certainly originally sold as a bound and ruled lute book. Margaret copied from the first folios of the book up to 30v, and from here to 45v at least three further scribes are at work. There appears to be no link between Margaret's work and these later copyists, who probably date from after 1630.

Hirsch seems to date from the last decade of the sixteenth century. The initials stamped on the cover are H. O., but there is no internal evidence of its original ownership and even the music is for the most part untitled and unascribed.This manuscript was originally dated 1597 by David Lumsden among others, though evidence for such a specific date is lacking. Spencer suggests that it may relate to song versions of two pieces, first published in 1597,[4] but widens his field of evidence in reaching his own conclusions about dating:

All English manuscripts before about 1590-95 are quarto in form,but Hirsch is folio. The style of binding is very similar to that of the Welde lute book, which has been dated c1600. The paper could have been made between about 1577 and 1596.

The majority of the known composers represented date from the 16th century ... and only Dowland and Byrd were still alive in 1613. If the MS had been written out after 1611 (see piece 44) [apparently a copy of a 1611 version of a Byrd fantasia] one would expect to find more French music and pieces by Robert Johnson, Sturt and Rosseter. Pieces 19 and 20 were not copied from Barley 1596.

I suggest the MS was begun c1595.[5]

Since the Byrd fantasia (No.44) raises serious questions about the early dating of the manuscript, Spencer takes care to justify his excluding its publication date from his evidence, continuing:

In The Consort and Keyboard Music of William Byrd (Faber, 1978) Oliver Neighbour writes about this Fantasia: "Though published in 1611, Fantasia 4/g can be dated with some certainty about 1590 or a little earlier" (p.92). Having seen a transcription of the Hirsch version (made by Peter Trent, who first identified this text), Mr Neighbour informs me that it is the 1611 text with some errors. This suggests but does not prove, that it was written in the Hirsch manuscript after 1611. Byrd could have made the revision some years before the 1611 publication; likewise the copyist could have added this piece some years after beginning his manuscript collection.

This discussion is plausible, though it seems that the argument may have been tailored to fit a preconceived notion of the date of the book. The exclusion of French music could be attributed to a conservative or specifically English taste, but the real influx of French music in these sources did not begin until well after the publication of Robert Dowland's Varietie of Lute Lessons in 1610, and was only usual from 1620 with scribes who had some personal connection with the continent, such as Edward Herbert. The French additions to Board and Pickeringe date from as late as 1630 and 40 and, in the case of Pickeringe, up to 1650. In fact, there are three token 'French Corantos' on f.10v of Hirsch that are found in other English sources, particularly for keyboard. The first was based on the pavan Belle qui tiens ma vie, that was included in Arbeau 1588, and is concordant with f.4 of Montbuysson (1611).

Spencer lists John Sturt among the composers one might expect to find in a manuscript of later date. This is dubious, as Sturt's music only survives in a very small proportion of the manuscripts contemporary with him, and it appears that the circulation of his work was rather limited. The Holmes books, compiled by a professional singer and probably also city Wait in Oxford and London, which otherwise contain virtually the gamut of contemporary lute music, only contain three isolated pieces by Sturt, in Nn.6.36 (c1610-15). Spencer also mentions Robert Johnson and Philip Rosseter, who were more popular than Sturt,although Rosseter's surviving repertory is substantially smaller than Sturt's, and is found in only a few of the many sources dating from the first quarter of the seventeenth century.[6] Rosseter only has two pieces in only one of the Holmes books, and he is not found in Marsh, Folger, Wickhambrook, Trumbull, 31392, Mynshall, Brogyntyn, Welde, Euing, Sampson, Cosens or ML, the other major sources from this period. Robert Johnson is also poorly represented in the Holmes books.7 Comparison of the contents of Hirsch with Pickeringe, dated 1616, shows a similar range of concordant sources. Pickeringe, though it dates from the second decade of the seventeenth century, contains 11 pieces by John Johnson (d 1594), but only one piece by Robert Johnson, so it is clearly not usual to find music by Robert Johnson in sources from the 1620s. The exclusion of Sturt, Robert Johnson and Rosseter from Hirsch therefore has considerably less significance than Spencer's comments would suggest. Their absence from the manuscript certainly does not prove that Hirsch pre-dates their working lives.

All these factors could argue against an earlier provenance for the book, but Spencer's dating is not unreasonable. Therepertory as a whole certainly seems to date from the late sixteenth century, as the number of concordances with Dd.2.11 (shown in table 31) demonstrates. However, it is not the date of the repertory that is in question but its date of copying, since the probable purpose of the book makes an anachronistic repertory virtually certain. Spencer points out that the format makes an even earlier provenance unlikely, though even for an upright folio format book this one is very large, and closer in size to those dating from the 1620s such as ML rather than earlier sources like the Holmes books. Spencer also advances the theory that the book may have had pieces added to it for many years after its inception. However, the arrangement of the scribes suggests that they were copying concurrently, and the later inclusion of the Byrd fantasia as he proposes would have involved the scribe leaving the book andreturning to it at least 15 years later without any alteration in copying style, and without noticeable intervening interference or addition from any of the other scribes involved in the compilation. The currency and obvious interrelationship of this group of scribes indicates conclusively that in fact the book was compiled over a relatively short time-span, and not the 15 to 20 years which Spencer concedes must be possible in order to fit the Byrd anomaly into his overall scheme.

The Byrd fantasia is not the only piece to appear in printed sources, and other concordances might also shed further light on the average date range of sources containing this repertory.

Table 31 lists the other lute sources with which Hirsch has concordances (in date order as far as this is possible),and the quantity of those concordances. Cognates with consort, keyboard or versions for other instruments are not shown.

TABLE 31

LUTE SOURCES CONCORDANT WITH HIRSCH

Source
date
No. of cons.
Folio numbers in order of concordances in Hirsch
2764(2)
c1585-90
1
[5]v-[6]
Dallis
c1583-5
1
93-94
Adriansen 1584
1584
1
5
Dd.2.11
c1585-95
21
101v, 48, 71, 41, 44, 60v, 72v, 57v-58, 32, 21v, 57-56v, 75v-77, 37v, 40, 50v-51, 22, 65, 54v, 28, 18, 16
Trumbull
c1595
4
16v-17, 6, 4, 1
Marsh
c1595
1
247-6
Wickhambrook
c1595
1
11
Dd.5.78.3
1595-1600
7
33, 29v-30, 15v, 10v, 20, 14v-15, 69
Barley 1596
1596
2
3-4v, 1v-3
Welde
c1600
1
8
Dd.9.33
c1600-1605
5
59v-60, 60v, 17, 86v-87, 84v-85
Herhold
1602
1
13v-14
Besard 1603
1603
1
32
31392
c1605
3
30v-31, 32v-33, 29v-30
Euing
c1610
2
39, 32
Dowland 1610B
1610
1
1-1v, Sig.G
Nn.6.36
c1610-15
1
14v
Montbuysson
1611
1
4
Schele
1613-19
1
p.52-5
Mertel 1615
1615
4(5)
p.146-7, 223-4, 191, 222-3,

(cf: 148-9)

Fuhrmann 1615
1615
1
p.59
Pickeringe
1616
1
17
ML
c1620
1
13v-14
Herbert
c1630
1
44v
The largest single group of theseconcordances is with Dd.2.11 which dates from the period around 1595. The other concordances are spread fairly evenly over sources dating from the period 1570-1630, with the exception of the other Holmes books, Dd.5.78.3 and Dd.9.33, and the music found in Mertel 1615. The Holmes collection is undoubtedly the most comprehensive from this period, and so a large number of concordances is to be expected. Mertel's fantasias Nos.17, 56, 79 and 80 are concordant with Hirsch folios 13r (Scribe A), 65r (Scribe B), 65v (Scribe B) and 64r (Scribe A) respectively. The fantasia on 65r is a tone lower than Mertel's version, but otherwise is precisely concordant. The two fantasias copied by Scribe B are almost exactly concordant with those in Mertel 1615, while the one on f.13r shows quite substantial re-writing of the harmony. The last piece noted is concordant with one of the Hirsch fantasias, and may not therefore be related.

Mertel 1615 probably had a fairly limited circulation as very few copies survive when compared with other publications. We cannot say for certain that it did not reach this country in some form, particularly as its contents indicate that the compiler was familiar with English music, though his sources are not evident. It is far more likely that music in a manuscript source should have been copied from a printed source, than that the printed source was copied from a manuscript as diverse as this one. These four pieces were not copied in by the same scribe, and the third also has concordances with Dd.2.11 and Herhold and is parodied in one of the fantasias in Hove 1601.8 Unfortunately, it is not possible to demonstrate conclusively that the Hirsch versions of the fantasias were copied from Mertel 1615. All that this demonstrates is that these fantasias were clearly still in circulation and popular enough in 1615for Mertel to be familiar with them, and to print them in his Magnum Opus.

The paper of Hirsch, though not precisely identifiable, may date from the period 1590 to 1610. Restoration work carried out by the British Museum conservation department between 1958 and 1960 included re-stitching and re-backing, when the positions of the end-paper stubs was altered, and water stains were removed from a number of leaves. Since then there has been quite substantial offsetting of the ink onto facing pages of the treated folios[9] affecting legibility quite badly. A microfilm made before the conservation work was done has survived, and it is from this that facsimiles of the damaged folios have been made.

It is difficult to tell now whether Hirsch was bound before writing, but there is no loss of material in the gutters or from cropping, even where the writing extends right out to the edges of the paper. The likelihood is that, like Board, it was sold ruled and bound. In 1947, before its acquisition by the British Museum, Daphne Bird, in a note now bound as folio ii, dated the manuscript 1597, and this is the date that David Lumsden repeated, neither presenting supporting evidence. It is possible that during the restoration work the evidence for this date was also removed or trimmed away, but it is more likely to relate to the song versions of two pieces.

Table 32 is an inventory of the contents of the Hirsch manuscript. In the second column, the letters A to E indicate the work of its five scribes. A large gap of 42 ruled but otherwise unused folios between 21v and 63v is indicated with a double horizontal line. The single horizontal line indicates the point where dance music gives way to fantasias.

TABLE 32

CONTENTS OFHIRSCH

Folio Scribe Title Composer

[1v-]2/1 A Pavan [last five bars only] John? Johnson

2/2 Galliard William Byrd arr.

2v B In Nomine Pavan, duet part Nicholas Strogers

3/1 In Nomine Galliard, duet part Nicholas Strogers

3/2 A Lady Laiton's Pavan/Dream ?John Dowland

3v-4/1 B Pavan Richard Allison

4/2 C Galliard

4v-5/1 Sharp Pavan Richard Allison

5/2 A Hasselwood's Galliard Anthony Holborne

5v-6/1 Pavan Ambrose [?Lupo]

6/2 Galliard

6v/1 Fantasia, duet part Marchant

6v/2-7/1 Galliard Francis Cutting

7/2 Galliard John Dowland

7v/1 Clark's Galliard/Quadran Galliard/Jest Anthony Holborne

7v/2-8 Last Will and Testament Pavan Anthony Holborne

8v-9/1 Fantasia Alfonso Ferrabosco

9/2 D Galliard Richard Allison

9v (A)B Pavan Francis Cutting

10 (A)B Pavan Francis Cutting

10v/1 E Ground

10v/2 Three French Courants

11/1 Galliard Edward Pierce

11/2 Galliard

11/3 Captain Digorie Piper's Galliard John Dowland

11v/1 Earl of Essex's Galliard/Can She Excuse John Dowland

11v/2 Captain Candish's Galliard John Dowland

11v/3 Lachrimae Pavan John Dowland

12 Groninge's Pavan Francis Cutting

12v-13/1 B Pavan Francis Cutting

13/2 A Fantasia

13v-14/1 Fantasia Alfonso Ferrabosco I

14/2 Fantasia Alfonso Ferrabosco I

14v-15/1 Fantasia

15/2 Prelude

15v Fantasia

16 Fantasia

16v-17 Fantasia

17v Fantasia Renaldo Paradiso

18 Fantasia

18v-19 Intabulation of Verdelot's Ultimi Miei Sospiri Alfonso Ferrabosco I

19v-20/1 Fantasia

20/2 Fantasia 5 Emmanuel Adrianssen

20v-21

21v Fantasia William Byrd arr. ?

63v C Sharp Pavan Richard Allison

64 Fantasia Anthony Holborne

64v/1 Fantasia Ut Re Mi Fa Sol La Alfonso Ferrabosco I

64v/2 Fantasia Francesco da Milano

65 B Fantasia Anthony Holborne

65v/1 Fantasia Francesco da Milano

65v/2-66 Fantasia

66v-67 Fantasia

67v/1 Fantasia

67v/2-68 Fantasia

68v/1 A Fantasia

68v/2-69 Fantasia

Scribes D and E are comparatively anomalous: D copies only one piece in what appears to be a blank space left at the end of music copied by Scribe A, while E occupies two full openings, and inserts a series of very short pieces without divisions, apparently unrelated in style to the music or copying of any of the other scribes. It appears very much that D and E could have been scribes who used gaps that had been left in the compilation organised by the first three scribes. Scribe B is the putative Board scribe, and it is clear from this table that B and A had some connection, confirmed by their joint activity on folios 9v and 10. The activity of Scribe C is more difficult to pin down, but becomes clearer if we examine the compilation of the book.

The copying began on at least one lost folio, indicated by the presence of the last few bars of a pavan on the folio now marked as folio 2. Since this pavan, in the form that it takes elsewhere, would not have occupied a full page itself,[10] there must have been further missing pieces, and the collation of the book, in 11 regular gatherings of eight folios, suggests that there was probably at one time a further gathering at the front of the book which is now missing, thus making the complete number of gatherings 12 rather than 11. Since the manuscript binding has been altered, it is not known whether there was originally any looseness at the front, or a series of stubs that might have indicated the removal of some sheets. The foliation in the upper right hand corner of the recto side of each folio was added in pencil by Daphne Bird, and begins with the number two, presumably because she could see that there was a page missing.

Examination of the layout reveals that the book appears to have been used at both ends at once. Scribe A copied two dances at the front of the book, and two Fantasias onto the final opening. He then handed over to Scribe B, who copied two pieces at the front of the book, counted back four folios from the end and then filled in a group of six fantasias copying towards the back. Scribe B even appears to have been using the same pen as Scribe A. After two more entries by A and B at the front, Scribe C added two dances and a pavan, and three fantasias at the back, before A once again interrupted with two large groups of pieces. At this point, his largest group of fantasias was added, but this group does not appear to fit into the 'front and then back' sequence, as they were copied into the front section of the book.

There are several possible reasons for this interruption to the scheme previously followed by all the scribes. It could simply be a mistake, or Scribe A may have decided that he did not intend to add any more dances, and the remaining music was to be only fantasias. This last proposition seems the most plausible, since on folio 8v, Scribe A had already changed to fantasias, only to be interrupted by B and E.

Scribes A, B and C seem to have copied continuously from each other. The probability that at least A and B were working together is confirmed by the first lines of the pieces on folios 9v and 10, shown in example 53. Here, Scribe A has written the first few notes of two pavans at the top of each page, and Scribe B has filled in the remainder, suggesting that although there were many copyists, there was one principal scribe overseeing the contents of the book, and its compilation was far from random. The point where the scribes change is indicated with a double vertical line.

ex.53 first three bars of the pieces on folios 9v and 10

It appears that Scribe A was the principal scribe in this collection, responsible for the choice of its contents, and their layout. Scribe B seems to have been influenced by him, as he (or she if it is Margaret Board) imitates the terminal flourish employed by A, illustrated in example 54.

ex.54 terminal flourishes of Hirsch Scribes A and B

Scribes D and E do not fit so neatly into this organization.Scribe D, particularly, seems to have been filling in a gap left at the end of a piece copied by A and, unlike A, B and E, he appends both title and composer to the music. The work of E occupies two complete openings with short versions of dances. Unlike the other dances in the book, these lack divisions.

>From the title index in table 32 it looks as if by the end of his first significant group of copying from folios 5 to 9, Scribe A had decided to conclude the group of dances, and continue with fantasias. Scribes B and E nevertheless added some further dance music, before A finally took over and added his largest group of fantasias. The blank opening between ff.20r and 21v may have been a device to indicate that one or two other pieces in the sequence were intended to be filled-in later. There seems to be no obvious reason for it though, and the handwriting of Scribe A is unchanged.

The most significant organizational feature of this manuscript is not, however, the activity of the scribes. The presence of a number of copyists is fairly commonplace in manuscripts of this period though rarely with such clear links as in this case. What is surprising is the organization of the genres into two definite sections. The order envisaged and fairly consistently maintained by Scribe A was known and followed by Scribes B and C, and possibly also D and E. This type of preconceived layout is very unusual in English lute sources .

The contents of the book and the arrangement of the scribes working in it indicate that it was almost certainly copied in a fairly short space of time, though Scribes D and possibly also E may have been working later. Though some organization is present in varying degrees in late sources such as Herbert, it is unknown in English manuscript sources before 1620 or printed English sources before 1610, in spite of foreign precedents. Robert Dowland adopted the layout common on the continent for the Varietie of Lute Lessons, where organization by genre in printed collections had been established for some 50 years by 1600. The Varietie of Lute Lessons is fairly conservative in content though, despite the inclusion of a section of French dances, and it is likely that if the foreign sources influenced its organization, then they also influenced its repertory. In addition, foreign collections always place fantasias at the head of any collection, not at the end as in Hirsch.

The arrangement of Hirsch indicates particularly that it was almost certainly not a pupil's book copied for the purposes of learning the lute. In a teaching book, we might expect to find some duet music,and a more balanced mix of genres of approximately graded difficulty with considerably fewer fantasias, if any.

The immediate conclusion that arises from the activities of Scribes A and B in Hirsch, is that A may have been the teacher of B. In Hirsch, one scribe (A) is clearly influencing the order of the compilation. Despite his uniform and compact copying style,he still ends up having to attach an extra piece of paper to f.19r in order to accommodate the length of the intabulation of Verdelot's Ultimi Miei Sospiri. This is also evidence that he intended to play the music, since if he had not intended to play from the copy, Scribe A would not have been concerned about avoiding a page-turn in this rather clumsy manner. In addition, the edges of the pages containing music are noticeably grubbier than those that remained blank. Scribe A wrote for a six-course lute, and used x and # graces, but no fingering; Scribe B also wrote for six-course lute, used + and . graces, and . and .. for fingering; Scribe C wrote for six and seven-course lute,and gave titles in an apparently Italian style in the fantasia section. He used +, # and .graces and . and .. fingering; Scribe D also wrote for six-course lute, and used .and .. for fingering; Scribe E, again writing for six-course lute,used no fingering or grace signs.

In the section of fantasias at the end of the book, three of the pieces copied by Scribe A have time signatures: two of c-slash, and one of 3.[11] Time signatures are not found in English lute music before 1620 unless it was copied (or published) by John Dowland, and are usually associated with the later French repertory in which they were employed as a matter of course. The English scribes up to the mid-1620s, and sometimes later, relied on the placing of the bar-lines and the grouping of the flags to convey the metre of a piece.

There are clearly some aspects of the compilation of the Hirsch manuscript that make its present dating suspicious: the organization by genre, the presence of one piece apparently copied from a 1611 print, the use of time signatures and the extremely up-to-date repertory (if it was copied in 1595) since every source, no matter what its purpose, has a significant proportion of earlier music. However, the dating of the greater part of this repertory to c1595 is not unreasonable.

Table 33 is an inventory of Margaret Board's copying in Board, laid out in the same way as that for the Hirsch lute book. The section added by later scribes and some blank folios are omitted here and the omission is indicated with a double line near the bottom of the second page of the table. In this manuscript, Scribe A has been identified by Spencer as Margaret Board, and Scribe B as John Dowland. There is no overlap between the Hirsch and Board repertories.

TABLE 33

CONTENTS OF BOARD

Folio Title Composer

1/1 A Duet treble

1/2 Duet ground

1/3 Orlando Sleepeth/Orlando Furioso John Dowland

1v/1 Delacourt Pavan

1v/2 Galliard Marc Antoine

2/1 Ruggiero

2/2 Pavan

2v/1 Flat Pavan John Johnson

2v/2-3/1 The New Hunt's Up, duet treble John Johnson

3/2 The Hunt's Up, duet ground John Johnson

3v/1 Pavan Ambrose

3v/2 Study

4/1 Galliard Ambrose

4/2 Maske

4v-5/1 Spanish Measure, duet treble Richard Allison

5/2 Spanish Measure, duet ground Richard Allison

5/3 Light of Love/The Countess of Ormond's Galliard

5/4 The Scolding Woman

5/5 Prelude

5v-6/1 Pavan Peter Phillips

6/2 La Rossignol Almain, first part of duet

6/3 Toy/Ballad Tune

6v-7/1 Delight Pavan John Johnson

7/2 Courant

7v/1 Delight Galliard John Johnson

7v/2 Loth To Depart Galliard John Dowland

8/1 Lord Hay's Courant

8/2 Earl of Derby's Courant/French King's Maske

8v-9/1 Passamezzo Pavan Richard Allison

9/2 Singer's Jig Jo Singer

9v-10/1 Passamezzo Galliard Richard Allison

10/2 Lady Banning's Almain John Sturt

10/3 Go From My Window Richard Allison

10v-11/1 Solus cum Sola John Dowland

11/2 Home Again, Market is Done, Ballad tune

11/3 I Cannot Keep My Wife at Home

11v-12/1 Lachrimae Pavan John Dowland

12/2 Sellenger's Round/Est-ce Mars/The French Tune

12v/1 B Almain Robert Dowland

12v/2 A Robin is to the Greenwood Gone/

Bonny Sweet Robin/Robin Hood John Dowland

13/1 Almain John Dowland

13/2 La Volta Courant William Byrd arr.

13v-14/1 Primero Richard Allison

14/2 Flow Forth Abundant Tears

14v-15 Delight Pavan, duet treble John Johnson

15v/1 Study

15v/2 Courant

16/1 To Plead My Faith Galliard Daniel Bacheler

16/2 The Prince's Almain Robert Johnson

16v-17/1 Galliard on a Galliard of Daniel Bachleler John Dowland

17/2 My Mistress Farewell

17/3 The Lady Phyllis's Maske

17v-18 Battle Galliard/King of Denmark's Galliard/

Mr Mildmay's Galliard John Dowland

18v/1 Almain ?Anthony Holborne

18v/2 Mrs Lettice Rich's Courant Julien Perrichon

18v/3 Mrs Lettice Rich's Lavolta ?Julien Perrichon

18v/4 Courant

19/1 Galliard Robert Johnson

19/2 Courant

19v-20 Quadran Pavan Richard Allison

20v/1 Galliard Heigh Ho Holiday Anthony Holborne

20v/2-21/1 Dream

21/2 Lord Burgh's Galliard/Johnson's Jewel Galliard Johnson

21/3 Almain Lusher

21v/1 Captain Digorie Piper's Galliard John Dowland

21v/2-22 The Gordian Knot Anthony Holborne

22v/1 Dowland's First Galliard John Dowland

22v/2 Courant

22v/3 Courant/Branle

23/1 The Prince of Portugal's Galliard/

La Bergera Galliard Philip Rosseter

23/2 Poor Tom of Bedlam

23/3 Branle de la Torche

23v/1 Sir Walter Raleigh's Galliard Francis Cutting/William

Bradbury/Richard Allison

23v/2 Fair Ministers Disdain Me Not

24/1 Courant

24/2 Queen's Galliard John Dowland

24v Maske Tune?

25/1 Sellenger's Round/Est-ce Mars/The French Tune Pierre Guedron

25/2 Courant

25/3 The Eglantine Branch

25v/1 The Wood Bind

25v/2 The Gillyflower

25v/3 More Palatino/En Me Revenant Daniel Bacheler

26/1 The Witches Dance from the Maske of Queens

26/2 Gathering Peascods John Johnson

26/3 Almain?

26/4 Ballad tune/Toy

26v/1 Marigold Galliard

26v/2 Midnight John Dowland

27/1 The Prince's Courant Robert Johnson

27/2 The May Pole/The King's Morisck

27v/1 Hunter's Carrier

27v/2 Ballad Tune

27v/3 First Tune of the Lord's Maske Robert Taylor

28/1 The Prince's Maske ?Robert Johnson

28/2 Almain Robert Johnson

28v/1 Almain Philip Rosseter

28v/2 Almain

29/1 Volt/Courant Julien Perrichon

(/Mathias Mason)

29/2 Prelude John Dowland

29v Fantasia Alfonso Ferrabosco/

Richard Allison

30/1 Courant John Dowland

30/2 Almain Johnson

30/3 Ballet/Almain Robert Johnson

30v/1 Almain Jenning

30v/2 The Prince's Maske/Lady Elizabeth's Maske ?Robert Johnson

30v/3 Mall Peatly

83v/1 B Prelude

83v/2 A Dulciana/Dance

Not apparent from the two tables is how little information is given by the Hirsch scribes in comparison with Board. Only three of a possible 56 titles and ascriptions in the Hirsch index were provided by the scribes, whereas most of the information in the Board list was supplied by Margaret--genres and composers.

A cursory glance through the titles of the Board inventory reveals a large number of popular songs and ballad tunes, and French genres that only became popular in England in the early 1620s. On the other hand, the Hirsch book consists almost exclusively of pavans, galliards and fantasias; all relatively complex compositions demanding quite a high level of technical skill from the player.

The dimensions of the two books are shown in Table 34.

TABLE 34

DIMENSIONS OF BOARD AND HIRSCH


Board
Hirsch
Binding
353 x 217 mm
343 x 218 mm
Page
339 x 205 mm
332 x 204 mm
Width of stave
12.3 mm
17 mm
Distance between staves
13 mm
15 mm
All surviving English lute books from before about 1590 are quarto in format, but Hirsch is folio, and very close in size to Board. The Board book is ruled with 12 six-line staves to a page, but Hirsch was ruled with a wider rastrum, allowing only 10 six-line staves. The rastra used for ruling the Board and Hirsch papers are quite different in size, and the spacing between them also differs. Clearly a different instrument was used to rule the two papers, and the possibility of a single stationer being responsible for the two books is cast into doubt by this disparity. Unfortunately, we do not know enough about ruling practices to say whether one stationer always used one rastrum. The quality of metals and the nature of manufacture of precision instruments at that time would suggest that a stationer would have kept a number of rastra in use at one time, and the possibility of their having the same dimensions would be extremely small.

The Hirsch music was intended for six- and seven-course lutes, and Margaret Board's repertory suggests its later provenance by being written for a lute with nine courses. Hirsch appears to have been copied quite fast, and apparently not subsequently corrected by players, who may have admitted the errors rather than clutter the text with messy corrections. The written pages of the manuscript are significantly grubbier than the unused folios, indicating that there was a fair amount of handling,so the music was not simply deposited here, but played from, probably quite extensively. There is no sign of the idiosyncrasy in playing style that can be seen in Board. In Board, Margaret's copious relishes reflect not only her florid musical taste, but also demonstrate a considerable dexterity, while Dowland's irritably repeated hold signs suggest a lack of understanding of the importance of holding down some notes. Stretches caused by holding down bass notes were unlikely to have been a problem, as some of the chord shapes and cadential configurations imply a reasonable hand span--or at the very least, a small lute. Margaret's book shows every evidence of slow and careful copying, with decorative final bar-lines and carefully placed graces. Together with the progressive difficulty of the music, this indicates that Board was a collection compiled at the instigation of a teacher, reflecting the influence of the teacher as well as the tastes of the scribe, and quite extensively used to play from.

There is no evidence of this sort of use in Hirsch, copied fast and with a scheme in mind that would have been entirely inappropriate for a teaching collection. What we are able to surmise from household accounts, letters and diaries about the manner of teaching the lute[12] suggests that a lute teacher visited his pupil every day, and probably supplied him with one piece at a time on a loose leaf, to be copied and learned before the teacher's next visit, when it would be exchanged for the next in the progression. This practice may lead to an explanation for the compilation of the Hirsch lute book.

It is possible that Hirsch was compiled under the direction of a teacher, in order to preserve his collection of loose-leaf music that was beginning to suffer from its itinerant lifestyle. The different scribes may have been pupils whom he considered could copy accurately and neatly enough to perform a copying task that he would have found tedious and needlessly time-consuming. This explanation would account for the inclusion of the Byrd fantasia in an otherwise early repertory. That fantasia may have been a later addition to the master's teaching repertory that had probably remained largely unchanged for many years.

The same explanation would also account for the organisation of the layout which one would not expect to find in a late sixteenth-century source, nor for that matter in any source compiled as a personal or teaching collection. Most collections appear to have grown more or less spontaneously through a copying period. The sort of organization one might expect to find in a printed source, where the contents are determined before the book is produced, is not generally applicable to the more personal collections that these manuscripts represent. It could perhaps be that Hirsch was intended to end up as a printed lute book, but there are a number of factors that make this unlikely. Firstly, printer's fair copies were usually, for practical reasons, written out by one scribe, most often the compiler. Secondly, the only commercially produced book of solo lute music to be published in this country before 1650[13] was Dowland 1610B which suggests that the market was very poor for this sort of publication, as opposed to the immensely popular lute song.

Also accountable is the influence exerted by A over B and the speed of copying. The earlier type of six- or seven-course lute intended for Hirsch may be explained as the lute for which the exemplar was prepared, and from which the music was copied. This may also be the case with the limited use of fingerings and graces, but can also be explained if the book was an 'archive' rather than a performing or teaching collection, and thus never accrued those signs that would have been particular to the owner's or scribe's style of playing.

So far, there is nothing in either the putative dating or their respective compilations to link the Hirsch and Board manuscripts except, possibly, Margaret Board herself.

The Board lute book is one of a relatively small number of manuscripts for which we are able to put a name to the principal scribe. We have reasonable grounds to suppose that Margaret Board was born in 1600, and began copying her lute book in her early 20s. For at least a short period she was taught by John Dowland, probably at her family's London residence, as Dowland was living in Fetter Lane at the time. Her hand is extremely distinctive among those in the lute repertory for its size, weight, regularity and legibility, and this is apparent in example 55, which illustrates as much of a page of her copying as will fit onto an A4 page at actual size.

In titles, Margaret Board uses an italic script, commonly used by women in the early seventeenth century. Professional handwriting analysts emphasise the importance of habit over all variations in detail when examining any sample of handwriting. Margaret Board's tablature hand is flamboyant and fairly heavy, and some letters betray secretary rather than italic forbears. The sharp differentiation in shading suggests a wide and very soft quill. The size of the letters does not seem to take the width of the ruled lines into account, and in many ways they look too large for the stave, as the minim stroke entirely fills the space between the lines.

The letters are generally written vertically upright, above the lines of the system rather than on top of them, and the script covers the page very heavily both vertically and horizontally. When dotting flags, Margaret tucks the dot under the lowest curve of the flag. Noticeable is the recurved shape of the top of single flags as opposed to the `grid' pattern of flag groups of the same value. The manner in which the flags fill the space between the systems entirely,frequently overlapping the stave above, is particularly noteworthy, since it is particularly distinctive. This overlapping is unknown in the copying of other lute scribes--not only those active in England. Margaret uses a very large vocabulary of grace signs, many also used, and possibly added, by her teacher, John Dowland.

Two more detailed characteristics may be noted at this point: the vertical angle of the short stroke of the 'e', probably a hangover from a pure secretary form, and the manner of writing the letter 'a', with the straight back of the letter higher than the rounded lobe.

One highly individual peculiarity of Margaret Board's hand is the manner in which she appears to roll the pen slightly when writing, so that the wide part of the stroke is not consistently at one angle. This is particularly noticeable in the letter 'a' where the angle of the wide stroke of the nib used for the lobe of the letter is different from the angle for the backstroke.

Altogether, it is a distinctively modern and extrovert hand for its time, and one of the largest of any lute scribe surviving from this period. There are several other scribes active in this lute book, all added at a later date with the exception of John Dowland, who copied during Margaret's active period and corrected her work by adding hold signs and some graces (Margaret herself uses no hold signs at all). There seems no reason to question the resulting supposition that Dowland taught Margaret Board at some time during her learning period.

Example 56 shows part of a page of the Hirsch lute book at actual size, copied by Scribe B. Although the pages are similar in dimension, the staves are ruled further apart and the rastrum has wider-spaced tines than in Board.

ex. 55 Hand of Margaret Board

ex. 56 Hirsch Scribe B

The hand is again large and the covering of the page is very heavy. The flags again intrude on the stave above, though the slightly wider space between these staves is more accommodating to the size of the script. All the English lute scribes are extremely careful to ensure that their flags do not intrude on the tablature line above the one to which they apply, with the notable exception of these two hands. Here, as in Board, the flags are very large and overlap the stave above, but the letters, although still very large, are more suited to the spacing of the individual lines.

Most of the characteristics of Margaret Board's hand are evident here: the habit of size and layout, the similar type of quill, the angle of multiple beams, the shape of the tops of flags and positioning of dots, and the more specific details in the letters a and e.

Most lute scribes attempt to end each line with a full bar, but neither of the scribes here makes this effort. It is simple to isolate various letters which are very similar, or identical, in these hands. In table 35 though, whole chords that appear in both sources have been reproduced, as each scribe will deal with the problems of letter combinations in a different and idiosyncratic way.

TABLE 35

COMPARISON OF FIGURES FROM BOARD AND HIRSCH

There are strings of higher letters (those above e) in a set of divisions together with both forms of letter 'd'and fermatas.

Margaret Board
Hirsch B


The roll of the pen is evident, and the angle of the ascender of the 'd' in the second chord (line two) shows a similar stroke, though the wider spacing of the Hirsch lines has meant that the stroke has not become flattened to quite the same extent. Both the flat and upright forms of the letter 'd' appear in each manuscript, and the heavy and slightly bowed shape of the added backstroke is repeated in each instance. In examining any renaissance hand, the higher letters above f tend to be more idiosyncratic than the lower letters, since these letters have descenders and ascenders which allow a certain amount of flourish that is not possible in letters like a, c and e. The angle of the curve of the descender, and the amount the tail returns to the starting point varies widely among scribes, and is often the surest way of identifying a hand. The ascenders and descenders offer more scope for originality and embellishment, particularly in the secretary forms that can be seen in the progression of letters 'e-g-h'.

Finally in this table there are fermatas employed by both scribes. This circular shape is again unusual, and the central placing of the dot is again limited to the use of the two scribes in Hirsch and Board.

The flags are as individual as the letters, and the recurved shape of the top as well as the return at the bottom of the beam towards the stem, with any dot tucked under the final tail, is noticeable in both hands; features unusual enough in lute usage to be noteworthy here. The angles of the multiple beams in the two samples are the same, and closer examination shows that this is partly due to the scribe aiming for the top of the last downstroke with the first beam of each group, but not drawing the two without a pen-lift, as some scribes do. The first beam and the last stem are not in these cases drawn in the same motion. Placing single elements such as the dots under a magnifying glass--literally this time--shows the pen with the same touch and angle to the paper.

One of the features often viewed as a 'signature' when comparing hands is a scribe's use of the final bar and also the fermata or final note and flag. Some scribes always use the same terminal bars, some use the same termination for one period of their copying, and then change to another, while others are inconsistent. The terminal bars in these two sources do not match: that of Margaret Board is more meticulously executed,with decorative features, whereas that of the Hirsch scribe is faster and more spontaneous, and appears to have been influenced by the usage of the principal scribe in that book. It is not unusual for a scribe to use several forms of terminal barline, although most do practise just one form. Changes usually occur when a scribe is copying at different periods,[14] or from different exemplars.

More significant in this case is the use of a fermata on the last note, which is an integral part of the duct of the hand, and offers more scope for idiosyncrasy than letters or flags. Margaret Board uses a fermata at the end of virtually every piece, but the Hirsch scribe uses one only twice: these are isolated at the end of table 35. Among the other scribes in this repertory, several[15] draw fermatas in this manner. Several omit a dot altogether, but those who use one place it less uniformly centrally, and none of them makes the outer bracket almost completely circular. Although the shapes are not uniform, they have similarities, and the pattern of shading is unique.

The hand of each scribe develops quite noticeably through each manuscript, although the change in handwriting is more evident in the Hirsch book, where it becomes larger and inclines slightly to the right towards the end of the book so that the scribe appears to be writing faster.

Returning to examples 55 and 56, the samples of the hands of Margaret Board and Hirsch B: assuming they were written by the same scribe, they appear to have come from two different periods of copying. However, there are essential differences in the speed of writing and the intended purpose of the two sources which may explain the differences in fluency. In Board, the copying is clearly intended for the use of the scribe;she has added graces and other signs which give a personal flavour to the collection, and John Dowland has obviously taught from the book,copying one of his son's pieces in himself, and adding hold signs to Board's work. In Hirsch the copying appears to be much faster, probably simply reproducing the exemplar without spending time learning or practising each piece.

The letters are written closer to each other in Hirsch, but the wider spacing of the ruling means that the hand is better accommodated. All these samples exhibit strong similarities in habit, and many of the minor alterations seem to be due to the difference in the spacing of the tablature lines, and the different speed at which the two samples were written. The shapes of the letters and their distinctive shading remain unaltered, as does the very particular placing of the 'e' on the line: just scooping through it. The scribe also uses connecting dashes in both samples, though many other scribes also do so.

The string of higher letters--e, g and h--is so similar that they might have come from the same page, and are the most eloquent indication that these two hands were written by the same scribe. The Board lute book is a slowly and carefully written collection that clearly shows the personality of the owner. In Hirsch, the scribe was working in somebody else's book, and this seemed to influence her copying style. The book as a whole lacks the personality evident in many books of this period, and was probably never played by Margaret or Scribe A. Since Dowland only seems to have influenced her for a short time during the compilation of the Board book, it is possible that Hirsch Scribe A could also have been one of Margaret's teachers, and was using her as one of his trusted copyists. In a repertory that boasts some 200 scribes, none shows the heavy flamboyance of these two scripts, and none is even remotely similar in any respect.

Table 36 shows the types of corrections made to their copying used in the two books by the scribes in question. Again, this is revealing, as habit plays an indispensable role in emendation. A scribe rarely thinks very hard about the manner in which he emends, so that the way he does it becomes habitual. A quick glance down the table shows three types in common, the most significant being that on folio 1 of Board: other scribes in this situation, where too many beams have been drawn, scribble out the bottom one, as in the example at the end of the Hirsch corrections, rather than inking in the gap between the bottom two beams. In short, all the paleographical evidence is in favour of the conclusion that the hand of Margaret Board does appear in the Hirsch lute book.

One further possibility could be explored. Was it possible for two scribes to develop these same characteristics of duct independently, or could Hirsch B have been an older relation of Margaret? The conclusion that has to be reached is that this was not a viable possibility, since the handwriting tutors seem to be unquivocal about the indispensability of originality in the development of a good hand. Since the hand should be a reflection of the writer's personality, imitation of another's hand would not have been complimentary to the writer of the model, and would also reflect badly on the copier.

TABLE 36

EMENDATIONS MADE TO COPYING IN BOARD AND HIRSCH

Board

Hirsch
4v, 10v,

11, 13v

2v,

66v

1, 2v, 7,

11v, 13v,

15, 16v,

20v

10,

12v

83v
65v, 67v,

68

1, 1v,

3, 5v,

7, 14v

9v
1v
9v,

10

2v, 3v
65v, 67v,

68

3v, 5v,

6

66v
25,

26v

2v
Robert Thompson[16] asserts convincingly that papers with precisely the same mark were manufactured in only small quantities, since the amount of paper that could be made from a single mould was extremely limited. The Hirsch book contains only three watermarks, all from the mill of Nicolas Lebé of Troyes (1550-1605). The Board lute book contains four papers, one which may be Burgundian or English, and three further, all from Troyes, made by Edmond Denise, Jean Nivelle and Nicolas Lebé. The Lebé mark is the same type, and very similar to that in the Hirsch book, but is clearly not from the same mould. Their arrangement suggests that both books were bought bound.

Practices surrounding the sale of music paper in the early seventeenth century suggest that the paper of the Board and Hirsch lute books probably came from the same supplier, imported from France or the Low Countries. What is interesting is that virtually all the imported papers in Board show marks that are either similar or identical to marks dating from the 1580-1605 period, not the second decade of the seventeenth century.[17] Clearly an old batch of paper was used for Board, and a similar watermark turns up in Hirsch, for which the date of writing is in question. It is reasonable to assume that the paper was bought from the same seller, though not necessarily at the same time. From the ruling and watermark evidence, we can say with some certainty that they did not come from the same batch. This does not rule out the possibility that they were bought at the same time, but it makes it considerably more unlikely. The ruling also places these books in different batches as does the tooling on both covers, which also makes their origins with the same binder unlikely. This is not surprising, since the books were clearly not bought or owned by the same person.

So far, the purpose and date of the Hirsch lute book have not been considered other than superficially. None of the scribes apart from John Dowland have been traced in different manuscripts up to this time. There were a number of anomalies which did not seem to require immediate answers: the very large upright format, the organization of the genres which one would not expect to find until after 1620, the presence of an 'overseer', the use of time signatures, the modern appearance of the hands of the scribes--most notably Scribe B--and the anomalous Byrd piece. The sum of these points is sufficient to raise serious questions about the current dating of Hirsch. We know that it is highly unlikely, if not impossible, for two independent hands with the same unusual and distinctive nuances to have developed at any time, and particularly not 30 years apart, so if we add the probability that one of the scribes cannot have been copying earlier than about 1620, then not only is its earlier date incompatible, but there is new evidence to suggest a 'correct' date.

It is possible that the two books were copied during the same period, especially if Hirsch A were Margaret's original teacher, but the appropriateness of the Hirsch version of her hand to the size of the ruling, suggests that her tablature hand may have evolved in this book first.

There is no reason to suppose that the majority of the music in the Hirsch book was originally compiled later than 1595. On the other hand, the evidence points towards the fact that at least one of the scribes is from a later period than that suggested by the book's contents, so there must be another explanation. The most plausible explanation, and one which is supported to a great extent by the internal evidence, is that the book may have been a copy, or partial copy, made in the early seventeenth century of an exemplar dating from 1595, with some later pieces included.

One of the pieces of evidence to support this hypothesis is the apparent presence of the Boardscribe. The possibility that the Board scribe may have added those pieces at a later date than the rest of the book is unlikely, since some of the pieces copied by the Board scribe were clearly started by Hirsch Scribe A, and apparently using the same pen.

It is fairly likely that Margaret Board was born in the year she was baptized (1600), and the Hirsch book is unlikely to pre-date the inception of the Board book by many years without suggesting an unusual degree of precocity. As the Hirsch hand is more fluent, fast and consistent than Board, it is reasonable to suppose that it is more mature, and therefore dates from the same period or a later one than the copying of Board. The fact that the Board book is more carefully prepared and has a more personal connotation for the scribe than the work in Hirsch, may make the likelihood of their being contemporary greater, particularly if Hirsch Scribe A was Margaret's first teacher, and guided the first lessons that Margaret copied into Board. The absence of titles and graces from Hirsch, which proliferate in Board, is probably due to their absence from the examplar, rather than being a personal preference on the part of the scribe. The Board book was her own, but clearly Margaret Board was copying for someone else in Hirsch: specifically, Scribe A (anonymous) who wrote a more conservative secretary hand.

Conclusion

It is unlikely that the copyist of Board was not Margaret Board, born at Lindfield in Sussex in 1600. The date of Board was written on the verso of the first flyleaf, but it has been scratched out, and is now unreadable. She may have begun to copy as early as 1615, but in the absence of firm evidence, a fairly accurate approximation would be c1620. It is difficult to doubt that the hand of Margaret Board appears in the Hirsch lute book, currently dated c1595, particularly in view of the uniqueness of her hand when compared with all other surviving tablature hands, which makes the present dating unlikely. The argument concerning the dating of Hirsch does hinge to a great extent on this probability though, and such a controversial element qualifies any judgement, no matter how seductive the arguments, since it is impossible to prove. Taking into account all the internal 'literary' and paleographical evidence, it would seem likely that Hirsch was copied in the early seventeenth century, possibly as a way of preserving a master's exemplar, or simply as an organised copy of parts of one or more earlier sources, thus accounting for the early nature of the repertory it contains. Thus, the conclusion is that the Hirsch lute book is a copy of an exemplar or exemplars dating from the late sixteenth century, and [1] Chichester, West Sussex Record Office, Par. 416/1/1/1 (Lindfield parish register), f.8v.

[2] Board: Spencer 1976C; Hirsch: Spencer 1982.

[3] See Spencer 1976C.

[4] 'Captain Digorie Piper's Galliard', and 'Can She Excuse' (ff.11-11v).

[5] Spencer 1982, xiii.

[6] Sources of music by Philip Rosseter: Barley 1596 (Orph.) 55-56, 57-58, 59; Board 23/1, 28v/1; Herbert 10/2, 21-21v/1; Dd.3.18 41/2; Dd.9.33 46/2-46v, 47v-48; Montbuysson 99/2;Pickering 25v-26/2; Vilnius 8/2.

7 See table 21.

8 Hove 1601 sig.A f.3v.

[9] ff.2v-5 and 65 almost illegible, others affected: ff.5v-6, 12, 19-19v and 68v.

[10] I am grateful to Dr Richard Rastall for bringing this to my attention.

[11] ff.68v-69r.

[12] Many dating from later in the seventeenth century, and these including the letters and accounts of Lord Danby and his guardian during their stay in Utrecht, 1706-10. See Tim Crawford: `Lord Danby, Lutenist of "Quality"', LSJ xxv/2 (1985), 53-68. See also Chapter 3.

[13] Maynard 1611 was privately published by the author.

[14] e.g. the Holmes books (Dd.2.11, Dd.3.18, Dd.4.23, Dd.5.20, Dd.5.21, Dd.5.78.3, Dd.9.33, Dd.14.24, Nn.6.36), Sampson Scribe B.

[15] Principal scribe of ML, Scribe C in Dd.9.33, Scribe D in Folger.

[16] Thompson 1988A, and Thompson 1988B. See Chapter 5.

[17] See Spencer 1976C, Introduction: The Watermarks. its date of copying is more likely to be c1620 than c1595.
The Lute Page Julia Craig-McFeely's Thesis Index