(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "An Elizabethan virginal book; being a critical essay on the contents of a manuscript in the Fitzwilliam museum at Cambridge"

UNIVERSITY OF TORONTO 




Presented to the 
EDWARD JOHNSON Music LIBRARY 

by 
Mrs. Schaeffer 

in memory of 
Dr. Myron Schaeffer 



AN 
ELIZABETHAN VIRGINAL BOOK 



All Rights Reserved 



AN 

ELIZABETHAN 
VIRGINAL BOOK 

BEING A CRITICAL ESSAY ON THE CONTENTS 

OF A MANUSCRIPT IN THE FITZWILLIAM 

MUSEUM AT CAMBRIDGE 



BY 



E. W. NAYLOR, Mus. D. 

Author of "Shakespeare and Music" 
Organist and Lecturer 

of 
Emmanuel College, Cambridge 




WITH 
ILLUSTRATIONS 



LONDON 

J. M. DENT 6f CO. 

NEW YORK : E. P. BUTTON & COMPANY 

1905 



Go tbc 

MASTER AND FELLOWS 
OF EMMANUEL COLLEGE, CAMBRIDGE 

WHO HAVE MAINTAINED 

A COLLEGE LECTURESHIP IN MUSIC 

DURING THE PAST FOUR YEARS 



ML 



v 



\D 



' 



OCT 22 



'S/TY OF 



101G354 



PREFATORY NOTICE 

IT is hoped that this work may be of real use to two distinct 
classes of readers 

(a) Students of the history of music who have access to 
the published Fitzwilliam Virginal Book, but have been 
prevented from giving it the attention it deserves, by reason 
of its great size and various contents. 

A collection so large must necessarily include a consider- 
able proportion of uninteresting work. This book is offered 
as a guide to what is most worthy of notice. 

() Students of the Elizabethan and Jacobean Drama, who 
daily increase in numbers, and are often at a loss for musical 
illustrations such as are necessary for the representation of 
these works, even on a humble scale. 

This book contains a complete set of instrumental pieces, 
strictly contemporary with Shakespearian times, which are 
sufficient for almost any occasion of the sort, and may be 
written out for strings or otherwise, according to convenience. 

The description of the dancing steps will be found useful 
for the same purpose. 

As for the general object of the book, the author trusts 
that no reader will mistake him for a mere antiquarian, 
praising days long since outworn. 

The Elizabethan time has passed, but its spirit is not dead ; 
and these chapters are meant to do further battle with the 



viii PREFATORY NOTICE 

ignorance which still attributes the invention of the main 
features of Modern Music to the eighteenth century. 

The method of the book is simple, consisting mainly in 
a careful study of nearly Three Hundred pieces of the Tudor 
period, which are contained in the Fitzwilliam Virginal 
Book, and which, although published some years back, are 
still practically unknown, and almost altogether neglected. 



I have great pleasure in thanking Mr C. F. Bell of Oxford 
for kindly placing at my disposal the photographic plates 
from which the printed edition of the Fitzwilliam Virginal 
Book was prepared. 

Messrs Adam & Charles Black were good enough to 
allow me to use the colour -print of Queen Elizabeth's 
Virginal, from Mr Hipkins' well-known book, in the pro- 
duction of my frontispiece. 

The author of the Preface to the Fitzwilliam Virginal 
Book, Mr Barclay Squire, will, I hope, allow me to thank 
him for the valuable information which forms part of 
Chapter II. 

Mr H. Ellis Wooldridge, who was, until lately, Slade 
Professor at Oxford, did me the great favour to read 
through my work in manuscript. He, of course, is not to 
be held responsible for any mistakes that may be found in 
the book. 

E. W. N. 

CAMBRIDGE, April 5, 1905. 



CONTENTS 

CHAP. PAGE 

~~ I. VIRGINAL BOOKS CONTENTS OF FITZWILLIAM VIRGINAL BOOK I 

II. FRANCIS TREGIAN DANCES PAVAN GALLIARD DOWLAND . IO 

III. DOWLAND, TABOUROT, PASSAMEZZO, COURANTE, ETC. . 28 

IV. ALMAN, BRAWL, GIGGE, ETC. . . . . 38 

V. TOYE, CORANTO, LAVOLTA, GALLIARD, ROUND, SPAGNIOLETTA, 

MORISCO, ETC. . . . . 47 

VI. FUGAL FANTASIAS (ONE SUBJECT) . . 57 
VII. FUGAL FANTASIAS (SEVERAL SUBJECTS) . . 65 

VIII. SONGS OF THE SIXTEENTH CENTURY . . 79 

ix. "FANCY" PIECES . . . . 94 

X. HEXACHORDS . . , . . . 1OO 

" XI. FITZWILLIAM VIRGINAL BOOK AS A HISTORICAL DOCUMENT . 112 

XII. EARLY WESTERN HARMONY . . . .14! 

XIII. THE ELIZABETHAN VIRGINALS . . . .149 

XIV. ORGAN PIECES, PRELUDES, MADRIGALS (TRANSCRIPTIONS) . 169 
XV. THE COMPOSERS IN THE FITZWILLIAM VIRGINAL BOOK . 198 



LIST OF ILLUSTRATIONS 

QUEEN ELIZABETH'S VIRGINAL (in colour) . Frontispiece 

" THE IRISH HO-HOANE " 

Facsimile of a page in the MS., being piece No. xxvi. 
in the Fitzwilliam Virginal Book (Anon.}. The upper 
part of the plate shows a portion of an anonymous 
Prelude, piece No. xxv. On the harmony of the final 
bar, see p. 68 . . . . facing page 8 

FRONTISPIECE OF " PARTHENIA " (1611) 

Showing a lady playing Virginals. Observe the 
position of the hands, and compare pp. 157-168, about 
fingering ..... facing page 149 



LIST OF MUSICAL ILLUSTRATIONS 

PAGE 

Pavana. Thomas Tomkins . . . . .16 

Galliard. Oystermayre, J. . . . . .22 

Alman. Hooper . . . . 31 

Coranto. Hooper . . . . . '35 

Reduction of Allemande. Bach, J. S. . . 39 

Coranto. Bach, J. S. . . .41 

Gigg. Byrd, W. . . . . .43 

Specimens of rhythm in other Giggs. Bull, Byrd, Farnaby, G., 

Farnaby, R, . . . . 44~4^ 

La Volta. Morley, Thomas . . . .48 

Tordion (Galliard or Cinquepace). Arbeau . .52 

La Volta. Arbeau . . . . 53 

The Haye. Arbeau . . . . -54 

Specimen of subject and answer. Sweelinck . . .61 

Specimen of " Handelian " material. Philips, Peter . . 62 

Fantasia (extracts). Strogers, N. . . . 63-64 

Remarkable Pedal. Bull . . . . -65 

Dux and Comes, example of . . .66 

Subject (Mixolydian). Farnaby, G. . .68 

False Relation. Tallis, Farnaby, Pur cell . . .68 

Fantasia, extract, false answer. Farnaby, G. . .69 

second subject. Farnaby, G. . . .70 

Mixolydian specimen. Byrd, W. . . 72 

The same altered into key of G . . -73 

Tonal Fugue, specimens. Byrd and Bull . . 74'77 

Variations on songs ft Daphne." Farna&y . . 79-8 1 

Walsingham." Byrd . 81-83 

" Barafostus' Dreame." Byrd 83-84 

xiii 



xiv LIST OF MUSICAL ILLUSTRATIONS 



" Jhon come kisse me now," harmonised 

"Rowland." Byrd 

Giles Farnaby's Dreame 

Hexachords, example 

Just intonation, specimen of possible chords . 

Enharmonic modulations. Bull 

Hexachord, fantasia on. Sweelinck . 
other extracts. Stveehnck 

Variation on " Woods so Wild." Byrd 

Pavana and Galliard en suite. Johnson, E. . 

(t Mai Sims," song in Pavan form . . . 

"Walsingham," tune and words 

Development by sequence (melody). Anon. 

(harmony.) Oystermayre . 

Alman, showing key of 4 sharps and " mode." Anon. 

" Pawles Wharfe," showing key of D major. Farnaby, G. 

" Quodling's Delight," showing " mode." Farnaby, G. 

Alman, showing tonic and dominant in key of D. 
Johnson, R. 

Alman, showing Modal Harmony (extract). Anon. 

The same altered, showing key of G. . . .139 

Early Western Harmony, examples of, I3th century 

onwards, Burdens, etc. . . . . 141-147 

Short Octaves, use of. Philips, P. . . . . 153 

Virginals, compass of. Bull, Tomkins, Philips . 154-156 

Compass of clavier (1636) ..... 157 

Fingering of Emmanuel Bach . . . .159 

Couperin and J. S. Bach . . .160 

John Munday .... 162-163 

J. Bull . . . 163, 165-166 

,, uncertain author . . . .164 

Left hand thumb, marked 5, etc. . . . 166-167 

Slide, with fingering. Farnaby, Giles . . .167 

Fingering, 4 crosses 5. Byrd . . . .167 

Organ music, specimen. Anon. . . . .170 



PAGE 

90-91 
92-93 
94-95 

. 100 

(note) 102 
. 106 
. 107 
108-109 

"3 
114-115 
117-118 
119-120 

. 121 
. 122 
I3I-I32 
134-135 

'35-I36 

137 
. 138 



LIST OF MUSICAL ILLUSTRATIONS xv 

PAGE 

Harmonies of "Veni" reduced from organ music . 171-172 

Specimens of " Felix Namque." Tallis . . I73-J75 

" Christe Redemptor," as given by Bull . . .176 

Specimen of Bull's setting of it . . . 176-177 

" Gloria Tibi Trinitas," first notes of . . .178 

Bull's setting of the same (^ time) . . 178-180 

Sesquialtera, proportion of. Bull . . . .180 

" Salvator Mundi " or " Veni, Creator," Bull's setting 181-184 

Bull's arpeggios, etc. . .181 

,, harmonies . . .181 

,, mordents . . .182 

,, and broken octaves . . .183 

altered plainsong . . .184 

tritone not avoided . ' . .184 

Blitheman's harmonies to " Gloria Tibi Trinitas " . .185 

Bull's . .185 

" Ground," with variations. Farnaby . . .187 

"Up Tails All," used as "ground." Tomkins . 188-189 

Technical passages, i6th century, fiu/1 and Siveelinck . 190-192 

Madrigals, Caccini, Philips' arrangement . . 193-194 

Marenzio, . . 195-196 

Counter subject formed from subject. Siveelinck . . 204 



AN 
ELIZABETHAN VIRGINAL BOOK 

CHAPTER I 

VIRGINAL BOOKS CONTENTS OF FITZWILLIAM VIRGINAL 
BOOK 

THE subject of the following chapters is the important 
collection of clavier music preserved at the Fitzwilliam 
Museum in Cambridge, and, known as the FitzwilJiam 
Virginal Book. Until a few years ago, before its origin 
and history had been studied critically, it bore the name or 
"Queen Elizabeth's" Virginal Book, an erroneous notion 
having gained credit that it had been in the possession of 
the Virgin Queen, who was without doubt an amateur 
player on the virginals. This name must, however, be 
entirely put away from our minds, as the evidence which is 
now at hand shows quite clearly that the book can never 
have belonged to Queen Elizabeth. 

The Fitzwilliam Book is one of several well-known MS. 
collections of clavier music and arrangements for the clavier, 
which were brought together in Elizabethan times. The 
others deserve mention here : they are 

I. Benjamin Cosyn's Virginal Book, which is preserved 
at Buckingham Palace, and has a date somewhere about 
the year 1600. It contains as many as 98 pieces for the 
popular clavier of the time, i.e. the virginals, by such 



2 AN ELIZABETHAN VIRGINAL BOOK 

composers as Orlando Gibbons, Dr Bull, W. Byrd, T. 
Tallis (some of which are also found in the book which 
is the subject of this volume), and a number of arrange- 
ments of morning and evening " services " by Strogers, 
Weelkes, Bevan, etc., which are well known to students of 
old Cathedral music. 

2. Will Foster's Virginal Book, also at Buckingham 
Palace, and containing about 70 pieces, by similar composers, 
including Morley and John Ward. Also arrangements of 
sundry anthems. 

Several of the pieces in this book, again, are found in the 
Fitzwilliam Book. 

The writer himself gives us a date in his own hand, viz., 
1624. 

It is convenient, though not accurate, to speak of all 
these collections of music as belonging to Elizabethan times. 
Although in part gathered together in the earlier years of 
James I., they belong in essence to the previous reign, and 
it would be misrepresentation to allow them to be closely 
associated with the Stuart period. 

3. My Lady Nevell's Book, containing nearly 40 pieces, 
all by one composer, viz., William Byrd, who was one of 
the very greatest musicians of his time. This book is the 
property of the Marquis of Abergavenny, and is preserved 
at Eridge Castle. It was written out by John Baldwin of 
Windsor, who lived in Elizabeth's reign. Several of the 
pieces in this book also are in the Fitzwilliam Book. 
There are some features of interest in the titles, which 
appear in more complete form in this book ; e.g., the piece 
called " Walsingham " in the Fitzwilliam Book, is in the 
Lady Nevell's Book called in full, "Have with you to 
Walsingham" (which, by the way, is different from the 
first line of the ballad as it appears in W. Foster's Virginal 
Book or Percy's Reliques, "As I went to Waisingham ") ; 



AN ELIZABETHAN VIRGINAL BOOK 3 

and "The woods so wild" appears as "Will you walk the 
woods so wild ? " 

The fourth piece in the Lady Nevell's Book should be 
amusing, as it contains a suite of movements dealing with 
Army Life, by Byrd : 

(a) The Soldier's Summons. 

(b) The March of Horsemen, 
(r) The Trumpets. 

(d) The Irish March. 

(e) The Bagpipe, 
(/) And the Drone. 

(g) The Flute and the Drum (spelt Droome). 

(See the "Oxford History of Music," vol. iii. p. 95, 
which prints a few bars.) 

(j>) The March to the Fight. 

(i) The Retreat. Now followeth a Galliard for the 
Victorie. 

Yet another collection may be briefly referred to, Par- 
thenia, the first music ever printed for the virginals (date 
1611), containing 21 pieces by Bull, Byrd, and Gibbons. 

But all these are quite eclipsed by the Cambridge Book, 
which contains nearly 300 pieces by over thirty different 
composers, not reckoning the misty authors who appear 
as Anon., and who are responsible for more than 40 
pieces in this collection. Nearly all these composers are 
Englishmen. 

There is no need to labour the point that a collection of 
such size, representing the work of so large a number of 
writers, is of the greatest value historically, and I suppose 
there will be no difficulty in persuading the reader of this 
fact. In order to drive it home, I recommend him to try and 
bring to mind the names of between thirty and forty 
reputable composers of modern music, say from the year 
1800 to 1900, and then, if he succeeds, think whether his 



4 AN ELIZABETHAN VIRGINAL BOOK 

memory would serve him to mention even the names of goo 
separate compositions by them ! 

In other words, the Fitzwilliam Book can tell us more 
about the state of music in Elizabeth's days than many of us 
have ever known about our own times. Some figures will 
convince the student of this. 

The Fitzwilliam Book contains over 130 Dances ; 17 Organ 
Pieces; 46 arrangements of 40 different Songs; 9 arrange- 
ments of Madrigals; 22 Fantasias or Ricercare, by 9 com- 
posers ; 7 Fancy Pieces, descriptive or otherwise, by 4 
composers ; 19 Preludes by 6 writers, including 6 by Anon.; 
and 6 exercises on the Hexachord, by 4 different authors. 

It is not going too far to say that if all other remains of 
the period were destroyed, it would be possible to rewrite 
the History of Music from 1550 to 1620 on the material 
which we have in the Fitzwilliam Book alone. 

\Ve may look at the contents of the book in yet another 
light. Thus, the 130 Dances and the 46 arrangements of 
Songs are mostly used as an excuse for numberless variations. 
Here, therefore, we have an early exposition of a musical 
form that was highly thought of, and brought to its climax of 
possibility, three centuries later, by such a man as Brahms. 
In the Fitzwilliam Book we have the advantage of seeing this 
form in its infancy. 

Again, the association of certain Dances, particularly the 
Pavan followed by the Galliard and preceded by a Prelude, 
the whole forming a series of movements with a certain 
connection, which is observed in Parthenia (1611) and the 
Fitzwilliam Book, presents us with a most interesting pheno- 
menon, viz., the origin of the suite, with its series of dance- 
named movements, all in one key, and subsequently of the 
sonata of Beethoven, Schumann, Brahms, Strauss. Here, in 
this Elizabethan clavier music, we see the thing at its very 
beginning, and we realise perhaps for the first time, that 



AN ELIZABETHAN VIRGINAL BOOK 5 

the vile howlings and drum-thumpings of a Central African 
dance of savages are in a tolerably close connection with 
the refined inspirations of such poetical natures as are re- 
presented by the names which I have just referred to. It is 
instructive, sometimes, to be reminded, in the midst of our 
spun-sugar civilisation, of the pit from which we have been 
digged. (Any one who wishes to study this subject may 
read "Primitive Music," R. Wallaschek, pub. Longmans, 
1893; Camb. Univ. Lib., M.H. 22, 33.) 

Once more, the 17 Organ Pieces in the book, which are 
counterpoints on ecclesiastical melodies, are of interest not 
solely as examples of the organ music of the 1 6th century, 
but as examples of the stage which had been reached in such 
matters as modal tonality, and other marks of transition from 
the mediaeval system. 

The 22 Fantasias are examples of a fairly early, but 
already well-developed form of a kind of composition which 
we have since known as fugue. Here, again, we see in the 
specimens provided in the Fitzwilliam Book, the great form 
of Bach, in a comparatively infantile state. 

Further, the appearance of such compositions as I have 
referred to under the name of " Fancy Pieces," is in itself 
remarkable, showing that the spirit of the Romantic School 
was early present in the history of modern music, although 
much hampered by the Temporal Power, and therefore 
somewhat disabled. Still it was there ; and we can contem- 
plate some quite sufficient instances of work of this sort in 
the pages of the Fitzwilliam Virginal Book. Giles Farnaby's 
Dream, His Humour, His Rest, His Conceit, etc., are 
examples, and Byrd's " The Bells." 

But we have not exhausted the general views that may 
be taken of this collection. The " arrangements " (as we 
call them now the Elizabethan title, in Italian, was Inta- 
volatura) of Madrigals, in conjunction with the similar reduc- 



6 AN ELIZABETHAN VIRGINAL BOOK 

tions of ecclesiastical music found in other Virginal Books, 
are striking evidence that the serious amateur of music in the 
1 6th century desired to have the best works of his time 
put into such a shape as could be made useful in his 
home. 

So also, these arrangements of Madrigals in the Fitzwilliam 
Book show that the influence of Italy was being felt in England 
in Elizabeth's reign. The names of the original composers of 
the 9 Madrigals (or parts of Madrigals), of which arrange- 
ments appear by Peter Philips, are the well-known ones, 
Orlando Lasso, Striggio, Marenzio, Caccini. By the way, 
here we have dates given which are worth notice in connec- 
tion with the history of the MS. itself. Peter Philips' 
arrangements are carefully dated, e.g., 1602, 1603, 1605. 
This at least means that the pieces dated were not written 
in the book before those particular years.* Thus the MS. 
is plainly of post-Elizabethan origin ; but it must not be 
forgotten that its general contents are Elizabethan, or earlier 
still. 

Other countries also appear to have had a connection 
with musical England about 1600. One of the most pleasant 
pieces in the book is a Galliard by Jehan Oystermayre, 
ii. 405, ) (see below) ; and the 4 pieces by J. P. Sweelinck 
of Amsterdam are excellent examples of a writer whose 
reputation was equalled by the value of his work. Bull, 
40 pieces by whom are in the Fitzwilliam Book, was also 
associated with the Netherlands (he was organist of Notre 
Dame, Antwerp, in 1617); likewise Peter Philips, who was 
an English Catholic priest, and lived in the Netherlands. 

Again, we can practically study the theatrical instrumental 
music of the time from this collection. For instance, the 

* On this point see below. 

f The references are to the edition of Messrs Fuller Maitland and Barclay 
Squire, published by Breitkopf and Hartel (1894-99). 



AN ELIZABETHAN VIRGINAL BOOK 7 

Dances by Giles Farnaby, ii. 264, 265, 273,* are all called 
" Maske," f and most probably were part of the music of a 
" Masque." Other pieces which might well have completed 
the work are close to these, viz., seven anonymous short 
dances, Almains and Corantos, ii. 266, 267, 268. See also 
the series of seven excellent short dances by Hooper and 
Anon., ii. 309-13. 

The connection of England with Ireland is marked by the 
titles of three pieces in the Fitzwilliam Book : 

(a) The Irish Dumpe, ii. 236. Anon. 

(b) The Irish Ho-Hoane (Ochone), i. 87. Anon. 

(c) Callino Casturame, ii. 1 86. Arranged by Byrd. 

(Corruption of " Colleen oge asthore.") 

The last of these is referred to in Shakespeare, " Henry 
V." iv. 4, 1. 4, where it is spelt " Callino, Castore me ! " 

The scene is a Field of Battle alarums, excursions. Enter 
French soldier, Pistol, and Boy. 

Put. Yield, cur ! 

Fr. Sold. Je pense, que vous estes le gentilhomme de bonne 
qualite. 

Pist. [misunderstanding] Quality ? 



Cal - Ii - no, Cas - to - re me ! 

Art thou a gentleman f 

The second of these Irish tunes, the Ho-Hoane, is a very 
first-class example of real pathos, and should be heard. It 

* The Maske, ii. 273, ascribed to G. Farnaby in the Fitzwilliam Book, is 
labelled Cupararee = by Coperario, i.e. John Cooper, in MSS. Brit. Mus. 
10,444 Plut. It seems to have been one of the tunes in the "Maske of 
Flowers," 1613, and in the above MS. it is also called Graysin, i.e. Gray's Inn. 

f The accustomed order in a Maske was three stately dances, e.g., as 
described above, followed by " character " dances, such as of Satyrs, 
Baboons, or other mountebank displays. 



8 AN ELIZABETHAN VIRGINAL BOOK 

will be found a true expression of tender sorrow. To me 
the piece is a wonder, for I believe it to be about 400 years old. 

The first, the Irish Dumpe, ii. 236, is a cheerful tune, 
by no means like the English Dump, which (according to 
Shakespeare himself in Lucrece, line 1127) was a doleful 
ditty, hence the " doleful dumps," which we still acknow- 
ledge in 1904, although they were well-known under the 
same title by Richard Edwards, who was born in 1523, nearly 
four centuries back. 

Once more, works by at least ten different composers may 
be found in this book, which contain evidence of a growing 
sense of what we call Key, as distinguished from modal 
transition or chromatic alteration. 

And again, works are presented to us here which can only 
indicate that the questions associated with the idea of tem- 
perament were not unknown difficulties in the Elizabethan 
mind. The remarkable work of Bull on the Hexachord 
(i. 183) frankly recognises the necessity of enharmonic 
modulation through all the tonalities, more than a century 
before the appearance of the "Well-tempered Clavier "in 
1722. 

Other general features (e.g., notation) might be mentioned, 
but as it is nearly impossible for any one to find this of use 
without a considerable knowledge of the contents of the 
book, I refrain from any further introduction, and proceed to 
describe the book itself. 

It is a small folio, about 13 inches high and 9 inches 
broad, and contains 220 leaves, with music (closely written) 
filling up 209 pages. The writing is on six-line staves, as 
usual at the beginning of the 1 7th century. 

The binding is English, of the 1 7th century, crimson 
morocco, gold tooled, with fleur-de-lis ornaments. The 
paper probably came from Basel, as the water-mark, a 
crozier-case, appears in the arms of that town. 



$W?pR 



m 

* [ $ * ***'*! ' 



i* 




SLIGHTLY REDUCED FACSIMILE OF "THE IRISH HO-HOANE " 
IN THE FITZWILLIAM MS. 



AN ELIZABETHAN VIRGINAL BOOK 9 

The history of the book is very meagre indeed. It was 
written at some time between 1600 and 1620, but the earliest 
mention of it is 1740, in a book by Ward, "Lives of the 
Gresham Professors," of whom Bull was one (in 1596). 
Ward gives a list of the compositions by Bull in the Fitz- 
william Book, which, at this time, belonged to Dr Pepusch. 
In 1762 Robert Bremner bought it at the sale of Pepusch's 
collection (Pepusch died 1752), and he gave it to Lord Fitz- 
william, who had it in 1783. Sir John Hawkins (" History," 
1776) seems to be responsible for the connection of the book 
with Queen Elizabeth. Dr Burney (' History," also 1776) re- 
marked that " if Her Majesty was ever able to execute any of 
the pieces that are preserved in a MS. which goes under the 
name of Queen Elizabeth's Virginal Book (i.e. the Fitzwilliam 
Book), she must have been a very great player, as some of the 
pieces are so difficult that it would be hardly possible to find a 
master in Europe who would undertake to play one of them at 
the end of a month's practice." Dr Burney was exaggerating, 
probably having only glanced at the book, but his remark is 
nearly correct. Some of the pieces, in particular certain by 
Bull, are very nearly impracticable. 



CHAPTER II 

FRANCIS TREGIAN DANCES PAVAN GALLIARD DOWLAND 

CHAPPELL (" Pop. Mus. of Olden Time," 1859, 1st ed.) 
suggested that the book was made in the Netherlands by or 
for an Englishman, and that Dr Pepusch got it there. His 
guess was founded on the appearance of Mr Tregian's name 
in the MS., often abbreviated (as if the writer knew it too 
well). Moreover, a book published at Antwerp in 1605, viz., 
Richard Verstegan's " Restitution of Decayed Intelligence," 
contains a sonnet by " Fr. Tregian." The name appears 
variously in the MS. as Treg., Ph. Tr., F. Tr. ; and there is 
a piece by Tisdall called "Mrs Katherin Tregian's Pavan"; 
also S. T. is mentioned, who may have been Sybil Tregian, 
a member of the same well-known Roman Catholic family, 
who belonged to Volveden (or Golden) in the parish of 
Probus in Cornwall. 

" Towards the close of the i6th century the head of the 
family was named Francis Tregian ; he was the son of 
Thomas Tregian and Catherine, daughter of Sir John 
Arundell of Lanherne, and his wife was Mary, daughter of 
Charles, Lord Stourton. In the year 1577 the members of 
the Tregian family seem to have become suspected, probably 
as much on account of their wealth as of their religion, and 
(according to one account) a conspiracy was planned for their 
ruin. On June 8 the house at Golden was searched, and a 
young priest of Douay, Cuthbert Mayne, who acted as 
steward to Francis Tregian, was arrested and imprisoned, 
together with several of the household servants. At the 



AN ELIZABETHAN VIRGINAL BOOK n 

following assizes, Mayne was convicted of high treason, and 
on November 29 of the same year he was executed with 
hideous barbarity at Launceston. Mayne was the first priest 
to suffer under the long persecution which the English 
Catholics endured during the reigns of Elizabeth and James 
I., and his name was included in the list of martyrs beatified 
by Leo XIII. in 1 886. Tregian himself, who had been 
bound over to appear at the assizes, was committed a close 
prisoner to the Marshalsea, where he remained for ten 
months. He was then suddenly arraigned before the King's 
Bench and sent into Cornwall to be tried. For some time 
the jury would deliver no verdict, but after having been 
repeatedly threatened by the judges, a conviction was 
obtained, and Tregian was sentenced to suffer the penalty of 
pramunire and perpetual banishment. On hearing his sen- 
tence, he exclaimed : ' Pereant bona, quae si non periissent, 
fortassis dominum suum perdidissent ! ' Immediately judgment 
was given, he was laden with irons and thrown into the common 
county gaol ; his goods were seized, his wife and children 
were expelled from their home, and his mother was deprived 
of her jointure. After being moved from prison to prison 
and suffering indignities without number, Tregian was finally 
confined in the Fleet, where his wife joined him. He 
remained in prison for twenty-four years, during which time 
he suffered much from illness, occupying himself by writing 
poetry. In 1 60 1 he petitioned from the Fleet that for his 
health and upon good security being given he might * have 
the benefit of the open air about London (not exceeding five 
miles' circuit), yielding his body every night to the Fleet,' 
and also for leave on certain conditions to visit Buxton or 
Bath, having of late been * grievously punished with 
Sciatica.' His petition seems to have been granted, for on 
25th July 1602, he wrote from Chelsea to Sir Robert Cecil 
to the effect that the day on which, through the Queen's 



12 AN ELIZABETHAN VIRGINAL BOOK 

clemency, he came from the Fleet to Chelsea, he was 
enriched with a litter of greyhound whelps,' a brace of 
which he designed for Cecil, they being now just a year old. 
In 1606 he left England and went to Madrid, visiting (July 
1606) Douay on his way. In Spain he was kindly received by 
Philip III., who granted him a pension. He retired to Lisbon, 
where he died September 25, 1 608, aged 60. He was buried 
under the left pulpit in the church of St Roque, where a 
long inscription to his memory is still to be seen. At Lisbon 
he soon came to be regarded as a saint ; his body was 
said to have been found uncorrupted twenty years after his 
death, and it was alleged that miracles had been worked at 
his grave. Francis Tregian had no fewer than eighteen 
children, eleven of whom were born while he was in prison. 
The eldest son, who bore his father's name of Francis, was 
educated first at Eu and entered the college of Douay, 2pth 
September 1586. On the occasion of a visit of the Bishop of 
Piacenza (l4th August 1591) he was chosen to deliver a Latin 
address of welcome. He left Douay on nth July 1592, and 
was afterwards for two years chamberlain to Cardinal Allen, 
upon whose death in 1594 ^ e delivered a funeral oration in 
the church of the English College at Rome. This was 
probably the 'Planctus de Morte Cardinalis Alani' which, 
according to some accounts, was written by Charles Tregian, 
another son of the elder Francis Tregian. In a list of the 
Cardinal's household, drawn up after his death and now 
preserved in the archives of Simancas, Francis Tregian the 
younger is described as ' molto nobile, di 20 anni, secolare, 
di ingenio felicissimo, dotto in filosofia, in musica, et nella 
lingua latina.' In a draft petition of the year 1614, pre- 
served in the House of Lords, it is stated that he had borne 
arms against the friends of Queen Elizabeth, but eventually 
he returned to England, where he bought back some of his 
father's lands. The details of the transaction are somewhat 



AN ELIZABETHAN VIRGINAL BOOK 13 

obscure, but it seems to have led to his being convicted in 
1608-9 of recusancy, and to his imprisonment in the Fleet, 
where he remained until his death, about 1619. From a 
statement drawn up by the Warden of the Fleet Prison 
(apparently about 1622), it seems that at his death he owed 
over 200 for meat, drink, and lodging, though in his rooms 
there were many hundreds of books, the ownership of which 
formed a matter of dispute between his sisters and the 
Warden. It may be conjectured with much plausibility that 
the present collection of music was written by the younger 
Tregian to wile away his time in prison. The latest dated 
composition it contains is the * Ut, re, mi, fa, sol, la ' by the 
Amsterdam organist Sweelinck, which bears the date 1612, 
while the series of dated pieces by Peter Philips, who was an 
English Catholic ecclesiastic settled in the Netherlands, the 
note to Byrd's Pavan, before referred to, and the heading of 
Bull's Jig,* all point to the conclusion that the collection 
was formed by some one who was intimate with the Catholic 
refugees of the period. In this respect the evidence of 
Philips's pieces is especially important, as MSS. by him are 
hardly ever found in contemporary collections formed in 
England. The handwriting also bears out the theory that 
the MS. was written in the manner suggested; though 
obviously proceeding throughout from the same hand, the 
characters gradually become larger as the work goes on. 
In the absence of any undoubted specimen of the younger 
Tregian's writing, the point must remain for the present 
unsettled." 

Enough has now been said about the origin of the MS. and 
its history. All that is important to remember is, that the MS. 
was made by one writer, copying various MSS. of various 
dates (for some of the pieces may well be a century before 
his time), during several years. If this writer was Francis 
* Bull fled to the Netherlands in 1613. 



14 AN ELIZABETHAN VIRGINAL BOOK 

Tregian the younger, it is extremely likely that the MS. 
was written between the years 1608 and 1619, during which 
time F. Tregian was in prison at the Fleet, for his religious 
opinions. 

We now proceed to learn something about the contents of 
the MS., remembering always, that although it was written 
after 1600, the pieces in it represent, generally speaking, 
the music of the Tudors, and by no means that of the 
1 7th century. 

I have already named some of the classes of pieces, f.g., 
Dances, Madrigals (arranged), Songs, Fantasias (or Fancies), 
Preludes, Organ Pieces on plain songs, and so on. 

The Dances outnumber all the rest, so we will begin 
with them. 

The Fitzwilliam Virginal Book contains about 130 dif- 
ferent dances, and these are of a round dozen different 
sorts. Most of them are the subject of innumerable 
variations, and thus take up a good deal of room in the 
book. 

It will be useful to give here a rough idea of the different 
classes of Dances and their numbers. 

The largest class is naturally 

No. of Exx. 

. ~ . . IPav. 26 

1. Pavan and Galliard en suite} ^ .. , 

} Gall. 26 

There are about 20 more Pavans 
and Galliards in the book, but 
independent, not en suite. 

2. Alman ( Allemande, " German" 

dance) . . . -24 

3. Coranto (Courante) . . 14 

4. Gigge (Gigue) . . .5 

5. Maske (meaning Dance for a 

Masque, or a song-dance) . 4 



AN ELIZABETHAN VIRGINAL BOOK 15 

No. of Exx. 

6. Toye (of various character) . 6 

7. Lavolta (the Galliard with 

"caper") .... 2 

8. Round (Dance) ... 2 

9. Spagnioletta (various) . . 2 

10. Braul (Bransle) ... I 

11. Morisco .... I 

12. Muscadin I 



114 + about 20 other 
- PavansandGalliards. 
Rough total, 134 different Dances 
in Fitzwilliam Virginal Book. 

There are, besides, others called merely " Daunce," which 
are to be remembered. 

Some of the Dances are also songs, e.g., the Muscadin 
mentioned above (i. 74, and ii. 481, which is the same) is 
known as " The Chirping of the Lark," see Chappell 
(Wooldridge ed.), vol. i. 177. 

Again, some of the Dances were sung by the dancers 
under certain circumstances, and sometimes did duty under 
different names, e.g., the " Coranto," ii. 305, is really the 
Pavan (altered to time) which Arbeau (1588) gives to 
be sung in four vocal parts by the dancers, to the words 



qu tens ma vie 
Captiv-? dans tes yeulx," etc. 

(See Grove, vol. ii. 676.) 

The Pavan was the most important slow Dance in the 
1 6th and 1 7th centuries. It was written in semibreves and 
minims as a rule, and was far more like a hymn tune than 
anything else, to our flighty modern minds. 



1 6 AN ELIZABETHAN VIRGINAL BOOK 



PAVANA (FITZ.WILLIAM VIRGINAL BOOK, ii. 51). 
^Without the 'variations on each strain.] 

THOMAS TOMKINS (Elizabethan). 



Slow 



xv -^^ w w 






w 



pe 

\v l^ 



J *l - w ^- ^ - <L ^- 

^^y=g=E-fbg:^^?=^ 





1 M . .IC^> 



AN ELIZABETHAN VIRGINAL BOOK 17 
2. 











-- >^t 4 jt -^-, 






r 






te- 



= .r j J-" 







tzst 









1 8 AN ELIZABETHAN VIRGINAL BOOK 
. 3. 






m 

& 



Q?E 







$F 



_, mm- 

^TT 



*?=*? 

^ _Q _, 

*^f"=f^ 




^ 



^ 



ap 




^^ 



^nr? 



rl 



-Jt^ U^ *^' url 

T-^T~*T~ -up 



r 



AN ELIZABETHAN VIRGINAL BOOK 19 




N.B.(a) The ornaments, mostly mordents and slides, are 
given in the first strain of the above Pavan. In the second 
and third strain they are omitted, being of no real musical 
interest, and generally superfluous when played on the 
modern pianoforte. 

(b) The florid passages at the end of each strain should 
be played with delicacy, so as not to draw attention from 
the plain chords of which they are decorations. 

(c) The contrapuntal nature of each strain should be 
noticed. There is one motive running through each, in 
imitation. In (2) the alternation of the motive between 
treble and tenor is remarkable, e.g., bars 3, 4, 5. 

(d) The chromatic passages in (3), beginning at bar 4, 
are worthy of attention. However, this is not a solitary 
specimen of the kind, but will be sufficient to astonish 
students who may have been brought up to believe that 
such progressions were entirely the invention of Spohr, 
who lived two and a half centuries later than Tomkins. 

(e) Notice the irregular use of the bars, e.g., the two final 
bars in each strain contain 'only one semibreve instead of two. 

(/) Notice the irregularity in the lengths of the three 
strains, viz., 1 6 semibreves, 20 semibreves, 26 semibreves. 



20 AN ELIZABETHAN VIRGINAL BOOK 

Of the Pavan (commonly danced before the Basse-dance), 
Arbeau says it is very easy, consisting only of " two simples 
and a double " advancing, and again " two simples and a 
double " retiring. It is (as we already know) in Binary 
measure, and the careful Capriol (his pupil) here joins in 
with a calculation of the steps, saying that he makes the 
Pavan 8 measures (semibreves) " en marchant," and 8 measures 
" des-marchant." 

The master also gives particular instructions about the 
form and manner of dancing the Pavan. Noblemen dance 
these Pavans and Basse-dances " belles et graves," with cap 
and sword ; others in long robes, " marchants honnestement, 
avec une gravite posee." And the damoiselles with an 
humble countenance, " les yeulx baissez, regardans quelque- 
fois les assistans avec une pudeur virginale." Kings, princes, 
and " Seigneurs graves," in dancing the Pavan on great occa- 
sions, wear their " grands manteaux, et robes de parade." 
Also, queens, princesses, and ladies accompanying them, 
have their robes " abaissees et trainans," " quelquefois portees 
par demoiselles." The Pavan on these occasions is called 
LE GRAND BAL, and the music is provided, not by simple 
flute and drum, but by " haulbois et saquebouttes," and they 
continue the tune until the dancers have made the circuit of 
the " salle" twice or thrice. 

Besides this state dancing of Pavans, this Dance was used 
in Mascarade, when triumphal chariots of gods and goddesses 
enter, or of emperors and kings " plains de maieste." 

On p. 29 ff., Arbeau gives the vocal Pavan for four voices, 
" Belle qui tiens ma vie," which is quoted in Grove. The 
proper drum accompaniment, continued throughout the 
32 bars (|) is f> ff^ff^ff etc - He also g ives 
seven more verses of words to it, and says if you do not 
wish to dance, you can play or sing it. Moreover, he adds 
that the drum is not a necessity, but is good to keep the time 



AN ELIZABETHAN VIRGINAL BOOK 21 

equal ; and that for dancing you may use violins, spinets, 
flutes, both traverse and " a neuf trous " (nine-holed flute 
i.e. a recorder), hautboys, and, in fact, "all sorts of 
instrument? " ; or you may sing instead. 

Arbeau's account of the Passemeze, or Passy-measures- 
pavin of Shakespeare, is very simple. He says that the 
instrumentalists increase the speed of the Pavan every time 
they play it through, and by the time it has reached the 
moderate speed of a Basse-dance, it is no longer called Pavan, 
but Passemeze. 

And now, to explain what is meant by Pavan and Galliard 
en suite (see a specimen below, 114), Morley, 1597 (pp. 206, 
207, of the reprint of 1771), says, " After every pavan we 
set a galliard (that is a kind of music made out of the other], caus- 
ing it to go by a measure, which the learned call trochaicam 
rationem, consisting of a long and short stroke successively 

This is a lighter and more stirring kind of 

dancing than the pavan, consisting of the same number of 
strains ; AND LOOKE, how many foures of semibriefs you put in 
the strain of your pavan, so many times six minims must you 
put in the strain of your galliard." 

The meaning of this in modern words is simply that the 
most correct Elizabethan Galliard was made of the same tune 
and harmony as its own Pavan, but with the time changed 
from Quadruple to Triple. 

A good example is in "Parthenia," viz., Bull's Pavan and 
Galliard, St Thomas Wake (the Galliard is in the Fitzwilliam 
Book also), pp. 16, 17, etc., Mus. Antiq. Soc. edition. 

Bull has here carried this principle of unity backwards 
into the Prelude, and thus presents us with a real suite of 
three pieces, which have a vital connection one with the 
other. It was published in England (1611) before Bull fled 
to the Netherlands (1613). 

The 4< steps " of a Galliard are six in number ; see " Shake- 



22 AN ELIZABETHAN VIRGINAL BOOK 

speare and Music," pp. 142, 143, where Jehan Tabourot's 
account of it is quoted. 



GALLIARD (FITZWILLIAM VIRGINAL BOOK, ii. 405). 

OYSTERMAYRE (i6th Century) 




it tzzStzzzsziz: 

jl=-J -f-fb 



1 



^~ 




* Nothing seems really known of this composer. I give, for what it is 
worth, the following statement about a musician of the same name, from the 
Allgemeine Deutsche Biographic, vol. xxiv. (1887), Hieronymus O. Oster- 
mayer, described as "ein Siebenbiirgisch-Sachsicher Chronist des 16 Jahr- 
hunderts," i.e. "a Transylvanian historian of the sixteenth century," 
Siebenbiirgen being the German name for Transylvania. This Ostermayer 
was organist at Kronstadt in Transylvania (Cronstadt in Siebenbiirgen), 
where he died in 1561. He seems to have been a good fellow, and a 
teetotaler. Hieronymus = Jerome ; Jerome begins with a "J"; so does 
Jehan. It is possible the two names may represent one man. 



AN ELIZABETHAN VIRGINAL BOOK 23 







j. 






1 



-tJL 

^?%>- 



** 



J | 

a! 










^E 



~ f3 f3 m (~3 






' 

Rhythm changes to 



\ 1 



i r rr 

^^^^ 



24 AN ELIZABETHAN VIRGINAL BOOK 



/TN 







* - f^ I_ c^ ^ 

pEEE:=iES5E5Gfcg=i 

~T~ ' i 

?. The first and second strains contain eleven dotted 
semibreves. The third strain has four, plus eight dotted 
minims. 

Many of these Pavans and Galliards are curiously named, 
sometimes after persons, to whom perhaps they were 
" dedicated," to use our modern phrase. 

Amongst many instances in the Fitzwilliam Book are the 
following : 

Bull. Lord Lumley's Paven, i. 149. 
The Quadran Paven, i. 99. 
The Spanish Paven, ii. 131 (said to be more elaborate, 

and called the Grand Dance). 
Byrd. Pavana Bray, i. 361. 

Pavana Fantasia, ii. 398, apparently so called because full 

of imitations, like a " Fancy." 
Pavana Lachrymae * (Dowland's tune, see Grove, ii. 677), 

ii. 42. 

Lady Montegle's Paven, ii. 483, 
Quadran Paven, ii. 103. 
Others are : 
Giles Farnaby. Farmer's Paven, ii. 465. 

The Flatt Paven, ii. 453 (probably because in G minor, 

as ive say). 

Walter Erie's Paven, ii. 336. 

* J. Dowland's " Lachrymx, or seven teares, figured in seven passionate 
Pavans with divers other pavans, galliards, and almands," 1605. 

Settings of this tune are also found by G. Farnaby and Thos. Morley. 



AN ELIZABETHAN VIRGINAL BOOK 25 

William Tisdall. Pavana Chromatica, ii. 278, being " Mrs 
Katherine Tregians Paven," so called because using a large 
number of accidental sharps, the average " key " being 
E major, as ive say. 

Peter Philips. Pavana Dolorosa, i. 321. 
Pavana Paggett, i. 291. 

(Philips' " 1st pavan," dated 1580, see i. 345-) 

Similarly, various Galliards, en suite or not, are given 
distinguishing names, as 
Anon. Nowel's Galliard, ii. 369. 
Bull. Piper's Galliard, ii. 242, immediately preceded by 

" Piper's Paven," by Peerson, ii. 238. 
Byrd. Sir John Graye's Galliard, ii. 258. 
G. Farnaby. Rosseter's Galliard, " set by " Giles Farnaby, 

ii. 450. 

Thos. Tomkins. Hunting Galliard, ii. loo. 
W. Tisdall (see above), Pavana Clement Cotton, ii. 306. 

Here then are over twenty examples of this custom of 
naming Dances. 

Bull's "Piper's Galliard," just mentioned, is another ex- 
ample of how the composers of this time would "set" a 
melody, i.e. would " translate " it, and make it do duty in 
various shapes. I have already mentioned a 1 6th century Pavan, 
that of Tab6urot's book, which appears as a Coranto in the 
Fitzwilliam Book, with its rhythm changed from f to f-. Here 
we have Bull making his variations on "Piper's Galliard" 
out of the music of a Madrigal which some of us know 
familiarly, "If my complaint could passions move," by 
Dowland. It is " Captain Piper's Galliard," and appears in 
the original issue of " Lachrymae" (1605) under that name. 

I say that Bull presents us with a Madrigal in the guise of 
a Galliard. But it may very well be otherwise. I have no com- 
plete evidence. Certainly the name of Bull in the Fitzwilliam 



26 AN ELIZABETHAN VIRGINAL BOOK 

Book is not conclusive, for there are demonstrable mistakes 
of this sort ; and even the quaint case may be found of a 
piece being copied twice and ascribed to different composers, 
e.g., " Go from my window," i. 153 and i. 42 (Munday, and 
Morley). The Madrigal is printed as No. 4 of Dowland's 
"First Set of Songs," 1597. (See the edition of the Mus. 
Antiq. Soc., p. 13, or the Choir and Musical Record, Oct. 17, 
1863, Metzler). 

The chances are that the original is Dowland's and that 
Bull, who was the virtuoso of his time, took it to exercise his 
skill upon. We have modern, and much more disgraceful 
instances, e.g., where a Schumann song appears on a pro- 
gramme as by " Liszt " (the "Bull" of the Ipth century china 
shop), or where a movement of Bach thinly varnished over 
with an implicit (or explicit) melody is labelled " by 
Gounod." 

This being so, we add one piece to the small list of 
Dowland's compositions in the Fitzwilliam Book. (These 
are merely three " settings," by Byrd, Farnaby, and Morley, 
of Dowland's "Lachrymae," already alluded to.) 

I shall make no apology for saying a few words here about 
this remarkable man, who left his mark, not only on his time, 
but upon the great men of his time, which is far better. Men 
like Middleton, Jonson, Fletcher, Massinger, Shakespeare, 
mention John Dowland with familiar affection or admiration ; 
and before he had reached his 47th year he had travelled 
into every corner of Europe, had met everybody and seen 
everything, just as did the Abt Vogler in the l8th, and the 
Abbe Liszt in the Ipth century.* 

* The address " To the Courteous Reader," given below, which is prefixed 
to Dowland's "First Booke of Songes" (1597), shows that at the age of 39 
he had already visited France, Germany, and Italy, spending " some months " 
under the patronage of the Duke of Brunswick and the Landgrave of Hesse, 
and receiving " favour and estimation " in such important places as Genoa, 
Ferrara, Padua, Florence, and Venice. A similar preface to Dowland's 



AN ELIZABETHAN VIRGINAL BOOK 27 

References to Dowland are frequent in the poetry of his 
time. One of the most familiar is in the eighth verse (1. 103) 
of the "Passionate Pilgrim" which is printed in our copies of 
Shakespeare. As is well known, this particular sonnet is by 
Richard Barnfield, not by Shakespeare, and was published in 

i 59 8. 

His Pa van " Lachrymae " was clearly very popular. As I 
have said, the Fitzwilliam Book contains three different 
arrangements of it ; and the Mus. Antiq. preface to Dowland's 
First Booke of Songes (pp. 2, 3) gives several lines from con- 
temporary Plays, in which " Lachrymae" is referred to by the 
characters. William Byrd's setting of " Lachrymae" may be 
found in the Fitzwilliam Book, ii. 42. 

English edition of the " Micrologus" of Andreas Ornithoparcus (first 
original edition, 1517), which was printed in London, 1609, informs the 
reader that the author, at the age of 47, was tired of roving, or, to use 
his own words, was "now returned home to remaine," and to live a quiet 
life at "my house in Fetter-lane." 



CHAPTER III 

DOWLAND, TABOUROT, PASSAMEZZO, COURANTE, ETC. 

IN the last chapter our heroes were John Dowland, composer 
of celebrated Pavanas, and Jehan Tabourot, writer of instruc- 
tions for dancing to the same. Before we go on to other 
matters, it will be interesting to refer to the fact that their 
names were the subject of anagram, but of a different type in 
each case. Jehan Tabourot, being a Churchman, canon of 
Rheims indeed, naturally (I suppose) thought it somewhat 
decent not to print his own name without disguise on the 
title-page of his " Orchesographie " (1588), and so with 
simple cunning, transposed the letters thereof to 

THOINOT ARBEAU 

(the "i" in Thoinot represents the " J" in Jehan), 

and under the name of " Arbeau " he is best known. 

The anagram on Dowland's name, however, is of a far 
superior brand. 

Fuller says that John Dowland was " a cheerful person, 
passing his days in lawful merriment, truly answering the 
anagram made of him 

JOHANNES DOULANDUS 
ANNOS LUDENDO HAUSI" 

" Jack Dowland wasted all his years in playing " 

(referring to his lute-playing). 

Fuller also tells us that this anagram was made by Ralph 
Sadler, Esq. of Standon in Hertfordshire, and that he was 

with Dowland at Copenhagen. 

28 



AN ELIZABETHAN VIRGINAL BOOK 29 

Dowland, like other musicians of the late Tudor times, 
was not only a musician, but a scholar, for in 1609 he 
published a translation of the " Micrologus" of Andreas 
Ornithoparcus (Vogelgesang) [orig. pub. at Leipzig, 1517]- 
A proof of his excellent general education is found in the 
strength and simplicity of his English prose, as it may be 
studied in the dedication and preface to " The First booke of 
Songes of foure parts" (1597): 

THE First Booke of Songes or Ayres of foure parts 
with Tableture for the Lute. 

So made that all the parts together, or either of 
them severally may be song to the Lute, Orpherian 
or Viol de gambo. Composed by John Dowland, 
Lutenist and Batcheler of Musicke in both the 
Universities. 

Printed by Peter Short the afTigne of Th. Morley, 
& are to be sold at the Signe of the Starre on Bred- 
street Hill. 1597. 



TO THE COURTEOUS READER, 

How hard an enterprise it is in this skilfull and curious age to 
commit our private labours to the publike view, mine owne disability 
and others hard successe do too well assure me : and were it not for 
that love I beare to the true lovers of Musicke I had conceald these 
my first fruits, which how they wil thrive with your taste I know 
not, howsoever the greater part of them might have been ripe enough 
by their age. The Courtly Judgement I hope will not be severe 
against them, being it selfe a party, and those sweet springs of 
humanity (I meane our two famous Universities) wil entertain them 
for his sake [D. himself apparently], whome they have already grac't, 
and as it were enfranchisd in the ingenuous profession of Musicke, 
which from my childhood I have ever aymed at, sundry times leaving 
my native Country, the better to attain so excellent a science. 

About sixteene yeres past, I travelled the chiefest parts of France, 



30 AN ELIZABETHAN VIRGINAL BOOK 

a nation furnisht with great variety of Musicke. But lately, being of 
a more confirmed judgement, I bent my course toward the famous 
provinces of Germany, where I found both excellent masters, and 
most honorable Patrons of Musicke : Namely, those two miracles of 
this age for vertue & magnificence, Henry Julio, Duke of Brunswick, 
and learned Maritius Lantzgrave of Hessen, of whose princely vertues 
and favors towards me I can never speake sufficiently. . . . Thus 
having spent some moneths in Germany, to my great admiration of 
that worthy country, I past over the Alpes into Italy, where I found 
the Citties furnisht with all good Artes, but especially musicke. 
What favour and estimation I had in Venice, Padua, Genoa, Ferrara, 
Florence and divers other places I willingly suppresse, least I should 
any way seem partiall in mine owne indevours. Yet can I not dis- 
semble the great content I found in the proferd amity of the most 
famous Luca Marenzio, whose sundry letters I received from Rome, 
and one of them, because it is but short I have thought good to set 
downe, not thinking it any disgrace to be proud of the judgement of 
so excellent a man. 

(Here follows a complimentary epistle from Luca Marenzio, 
dated Rome, 1595.) 

Not to stand too long upon my travels, I will only name that 
worthy master Giovanni Crochio Vicemaster of the chappel of S. 
Marks in Venice, with whome I had familiar conference. And 
thus what experience I could gather abroad I am now readie to 
practise at home, if I may but find encouragement in my first 
assaies, etc. 

Really, a man of such varied accomplishment and excellent 
ability deserves to have anagrams made on his name. 

Excellent examples of Dowland are his Galliard for two 
players on one lute, "My Lord Chamberlain his Galliard," 
which is printed at the end of the Mus. Antiq. Soc. reprint 
of the First Book of Songs ; also the *' Lachrymae" of which 
three different settings appear in the Fitzwilliam Virginal 
Book, ii. 42 CByrd), ii. 472 (Farnaby), and ii. 173 (Morley). 



AN ELIZABETHAN VIRGINAL BOOK 31 



A delightful, and I daresay unexpected, contrast to these 
slow mournful Pavans, is furnished by the Meridian Alman, 
ii. 477, "set" by Giles Farnaby. 

This brings us nearer to the second most numerous class 
of Dances, the Almans, twenty-four of which may be seen 
in the Fitzwilliam Book. Here is one of the best in the book. 

ALMAN (FiTZ. BOOK, ii. 309). 

Brightly HOOPER (i6th Century). 

NjfcJ k v|^ * 



Q J- 



JL. 




32 AN ELIZABETHAN VIRGINAL BOOK 

ST\ 




(a) The ornaments are given in the above piece for the 
sake of completeness. In playing these old works on the 
pianoforte most of the ornaments may be omitted with a 
good result. In the Fitzwilliam Book the following sign is 



used for the upper or lower " mordent " S meaning 




any of these -ypj-p^- J J J- 1 and this <d for 

B 5 

a " slide," meaning 



F= 



(b) It will be noticed that the first " strain " contains nine 
semibreves, instead of the orthodox eight. 

(c) For the final "pause" chord, see below, pp. 95, 133^. 

The Alman seems to have had a fairly close relationship with 
the Pavan and also with the Brawl (Bransle) as described in 
Arbeau's " Orchesographie." The steps appear very similar, 
being arrangements of " simples" and " doubles " (see below, 
p. 50), and the tunes themselves (as Morley says) are similar 
in construction, i.e. consisting of strains of eight semibreves 
(see below, pp. 131, 137, for two examples). Here it is in- 
teresting to recall a passage in Shakespeare's "Twelfth 



AN ELIZABETHAN VIRGINAL BOOK 33 

Night" v. I, 1. 197, where the Clown and Sir Toby bandy 
words about the Doctor, who was drunk some time 
before. 

Sir To, (Drunk). Sot, didst see Dick surgeon, sot ? 

C!o. O ! he's drunk, Sir Toby, an hour agone ; his eyes were 

set at eight i' the morning. 
Sir To. Then he's a rogue, and a passy-measure s pavin. 

Here Sir Toby, being drunk, turns the Clown's expres- 
sion about the Surgeon's eyes being "set" so early as eight 
o'clock in the morning, into a reference to the method of 
composing a Pavan, which is set in phrases or strains of eight 
semibreves, A passy-measures pavin (Passamezzo^ was a Pavan 
played so quickly that the steps had to be cut short, thus Sir 
Toby ingeniously twists this musical allusion into his re- 
flections on the speedy intoxication of the Doctor. 

The Passamezzo Pavan occurs under this name in the 
Fitzwilliam Book, e.g., Byrd, i. 203; Peter Philips, i. 299; 
in each case with a Galliard en suite, which is similarly called 
Passamezzo Galliard. 

These two examples let us a little further into the secret 
of the Passamezzo ; for in both Byrd and Philips the Passa- 
mezzo Pavan is very similar in every way to the Pavana 
without any qualifying term, until near the end, where 
the " square" time in semibreves and minims breaks into a 
triple measure, ^ or f. The Passamezzo Galliard likewise 
shows this tendency to increased speed and liveliness, for in 
the example by Byrd, the sixth section contains passages in 
a triple crotchet rhythm (marked ^); and the Philips 
Passamezzo Galliard frankly ends with a " Saltarella " 
(i. 310), the style of which points in precisely the same 
direction. 

The example of the Passamezzo Pavan and Galliard by 
Philips is well worth playing (i. 299). 



34 AN ELIZABETHAN VIRGINAL BOOK 

While we are still in sight of the Pavan and Galliard, 
I take the opportunity of pointing out that the examples in 
the Fitzwilliam Book seem to show that only in a compara- 
tively few cases was the Galliard en suite made of the same 
melody as its Pavan, with merely the alteration of the rhythm 
from 4 to 3 time. An example by Bull, " St Thomas 
Wake " (Parthenia has it complete ; the Galliard only is in the 
Fitzwilliam Book) is a specially plain instance. Two others 
must be mentioned here, viz., the Pavan and Galliard by 
Edward Johnson, called "Delight" (ii. 436, 440), and by 
Ferdinando Richardson (i. 27, 32). 

The former (Johnson, " set " by Byrd) is a first-class 
instance, as each of the three sections of the Pavana is almost 
literally reproduced in the Galliard ; the latter (Richardson, 
whose real name was Sir Ferdinand Heyborne) at least 
makes a good attempt to develop the one music out of the 
other. 

See below, p. 114, where a strain of Johnson's Pavan and 
Galliard is quoted. 

We now go on with the numerous Almans. 

It is worth noting, as a confirmation of my remark above, 
concerning the close relation between the Pavan and the 
Alman, that Byrd's "Monsieur's Alman," i. 234, is remark- 
ably like the Pavan, " St Thomas Wake," by Bull, already 
referred to. 

Thus we already have cases showing that the purely 
musical difference between Pavan, Alman, Coranto, was not 
very marked, although the " style " of the dancing might be 
and was various. Even the Pavan might (as a Passamezzo) 
introduce triple rhythm ; just as we have seen a Pavan trans- 
formed into a Coranto on its way from Arbeau (Tabourot, 
1588) to the pages of the Fitzwilliam Book; and here again 
we find what is, in every essential, a Pavan, appearing under 
the name of Alman. 



AN ELIZABETHAN VIRGINAL BOOK 35 



CORANTO (FITZWILLIAM BOOK, ii. 312). 
\A mistake in the Index ascribes this to Giles Farnaby.] 

HOOFER (i6th Century). 

Rather quick I 

4 I J J^- 






JT 



is 



i 






J. 



iJ^^Mw 



- 



-\ & 



m 



g r 



F- 



^^^ 

-H _ ^ ^1 



i r -p 






r r ^ r i i F r F ^ 



36 AN ELIZABETHAN VIRGINAL BOOK 







r 






i=i gj: 



4 



Hooraa 



i i 



(a) This is about the best of the thirteen or fourteen 
Courantes in the book. It is curious that most of them are 
not such good music as the rest of the Dances. 

() The ornaments are omitted, except the " slide " in bar 
g of the second strain. 

(c) The number of " bars " in each strain is orthodox, i.e. 
8 or 16. 

(d) The final pause chord is merely decorative. See 
below, pp. 95 and 132. 

The " steps" of the Courante as given by Arbeau are the 
same as those of the Pavan, i.e. left foot out, right foot up ; 
right foot out, left foot up ; and then left foot out and right 
foot up twice over, the total result being to place the dancer 
further along on the floor. This was known as " two simples 
and a double," and was written " s,s,d." The difference was, 
that the dance was much faster. A Shakespeare quotation 
is evidence here " Henry V." iii. 5, 1. 32 : 

Bourbon. They bid us to the English dancing-schools, 
And teach lavoltas high and swift corantos. 



AN ELIZABETHAN VIRGINAL BOOK 37 

Sir Toby in " Twelfth Night " seems to regard the Coranto 
as even gayer than a Galliard, which again was far more 
lively than the Pavan. He says (" Twelfth Night," i. 3, 1. 
123): " Why dost thou not go to church in a galliard, and 
come home in a coranto ? " Obviously the journey home (to 
dinner) would be the faster of the two. 

A capital quotation from Selden is worth printing here, 
which gives a general idea of the way in which the Eliza- 
bethans used these dances. 

" The court of England is much alter'd. At a solemn 
dancing, first you had the grave measures, then the 
Corantoes and the Galliards, and this kept up with 
ceremony v and at length to Trenchmore, and the Cushion 
dance : Then all the company dances, lord and groom, lady 
and kitchen-maid, no distinction. So in our court in queen 
Elizabeth's time, gravity and state were kept up. In King 
James's time things were pretty well. But in king Charles's 
time, there has been nothing but Trenchmore and the 
Cushion-dance, omnium gatherum, tolly polly, hoite cum 
toite." 

Finally, Morley (1597) describes the French Bransle and 
the Alman as being equally in the " time of eight, and most 
commonly in short notes " ; and goes on to say that " voltes 
and courantes" also are "like unto this," but are " danced 
after sundrie fashions " " the volte " (La Volta, a Galliard 
danced so fast that only two steps could be squeezed in, in 
each bar, and with a high leap every other bar. Also the 
dancer had to whirl round, waltz fashion. See " Shakespeare 
and Music," p. 148) " the volte rising and leaping, the courant 
travising and running." 



CHAPTER IV 

ALMAN, BRAWL, GIGGE, ETC. 

Wsfhave now had sufficient description of the Alman and 
Coranto. The latter dance is well represented in the Fitz- 
william Book by fourteen examples. A whole "Suite "of 
them may be played from the second volume of the book 
(ii. 266), where an anonymous series of two Almans and three 
Corantos is found in one opening of the pages. 

Other excellent Almans are, ii. 171 (Morley), i. 65 (Anon.), 
and i. 75 (Anon.). It will be observed that the general char- 
acter of them all is lightness, as compared with the Pavan or 
even the Galliard of state ; and that the Alman has a " square " 
rhythm, while the "Coranto" has a "trochaic" measure, 
although the general lines of the time are still " set at eight." 

A very quaint Alman is the "Duke of Brunswick's 
Alman" by Bull (ii. 146), which may be taken as an ex- 
ample of the sort of variations which the Elizabethan com- 
posers were in the habit of writing on the various sections 
or strains of dances. This of Bull's has two " strains," with 
a variation on each, and concludes with an extra " Rep.," as 
it was called, showing Dr Bull's nimble fingers working at 
one of his great tricks, viz., quick repetition on one note. 

Byrd's "The Queenes Alman," ii. 217, is also worth 
looking at. 

It will be of interest here to point out that the unlikeness 

between these Almans of the Tudor times and those of J. S. 

Bach's Suites and Partitas, more than a hundred years later, 

is only a superficial one. The fact is merely that Bach 

38 



AN ELIZABETHAN VIRGINAL BOOK 39 

wrote out the "variation" on his work as the original 
Alman. Thus 



f 



etc. 



-I 1 



might well have been a 16th-century Alman, if one could 
forget the positive key effect of the scale of Bb major (utterly 
unknown to the Elizabethans). And a variation might be 
superimposed of the guileless sort which is frequent in the 
Fitzwilliam Book, e.g. 



Agrtmcns or " Rep" 




I will here give what Bach really wrote in the first four 
bars of his Allemande in Partita I. 



ALLEMANDE FROM BACH'S PARTITA I. 



40 AN ELIZABETHAN VIRGINAL BOOK 




A similar experiment may be made with the Courante of 
the same work. 



fcq 



3fc 



^-kj-r jp^rr-fri^r f 

*** I L^.J Lgj 1 F? |T ii 




This is of course nothing more than an ornament on 



AN ELIZABETHAN VIRGINAL BOOK 41 




and so we see the essential features are still there. 

"We have traced a close connection or relation between the 
Pavan and the Alman, and Morley (1597) has pointed out to 
us that the Alman is related to the FRENCH BRAWL. We 
can compare an Alman by Thos. Morley himself (a capital 
tune, ii. 1 71) with the solitary example of the Brawl (so named) 
which occurs in the Fitzwilliam Book (ii. 269), by Thos. 
Tomkins, " Worster Braules." I dare not, in the face of the 
immense variety and uncertain nomenclature of the Eliza- 
bethan Dances, lay down any sort of rule, however modified, 
as to a difference in the method of dancing to these two 
tunes. But there is an apparently "characteristic" differ- 
ence, viz., that in Morley's Alman the " strain " has the 
quick notes first, and ends with a long sound ; whereas 
Tomkins's Brawl has the heavy accented part of the tune 
first, and then the shorter and livelier notes. 

The subject would have to be treated much more 
thoroughly than is possible in these pages, to arrive at a 
conclusion. For Arbeau (1588) gives no less than seventeen 
different kinds of Branle before he arrives at one which was 
called the BRANLE DES SABOTS (p. 88 of Arbeau), the tune 
of which is well known to most students in one form or 
another. It is printed in "Shakespeare and Music" (Dent), 
where the steps also are given (" Shakespeare and Music," 
p. 147), which, as far as they go, could be danced quite well 
to the tune " Worster Braules " in the Fitzwilliam Book. 

There are some examples in the Fitzwilliam Book of Almans 



42 AN ELIZABETHAN VIRGINAL BOOK 

which do not altogether follow Morley's rule of " containing 
the time of eight, and most commonly in short notes." The 
particulars are here given, for the benefit of students : 

i. 75. Anon., containing two " twelves," instead of 
" eights," but in " short notes," as Morley says. 

i. 65. Anon., containing two "eights" and one "ten," 
but still in " short notes." This is curious, as appar- 
ently the longer third strain is the result of following 
out a " sequence," which again leads to the weighting 
of the Form at its close. 

i. 234. "Monsieur's Alman," by Byrd, which we have 
already looked at, and suspected of a close relation- 
ship with Bull's " St Thomas Wake " Pavan in Par- 
thenia (1611). This piece is built in two." sixteens," 
in " short notes " mainly. 

ii. 1 60. Robert Johnson, set by Giles Farnaby, which con- 
sists of a "ten" and an "eight," and (curiously) 
begins with an "outside" note, just as Bach's Alle- 
mandes and Corantos in the Suites do. 

Other examples of the " outside" crotchet are, " Nobody's 
Gigge," ii. 162, and the Coranto, ii. 268. 

A good Alman is the one by Marchant (ii. 253). 

Nothing seems known about this composer, except that he 
was in the service of Lady Arabella Stuart, a cousin of King 
James the First. 

And now we naturally proceed to the Gigge. 

There is plenty of confusion for the student to face in the 
Gigges of the older days. This dance, as one knows it in 
Corelli, Handel, Bach, Scarlatti, etc., is generally in some 
sort of triplet time, ^ 2 , f , or even T \. Queer exceptions are 
found, e.g., Bach's Partita VI. (in E minor) ends with a Gique 
(sic), in a square time of 4 minims in the bar, marked CC. 
But even in this out-of-the-way case, the rhythm sticks to the 



AN ELIZABETHAN VIRGINAL BOOK 43 

use of "pointed notes " (Hawkins, writing in 1776, speaks 
of the pointed note, meaning " dotted," and says there was no 
authority for a Jigg having this as a general feature). 

All I know about the history of the Jig is in my book, 
"Shakespeare and Music," pp. 124, 125, and 117, with the 
two examples on p. 205. 

A GIGG (FITZWILLIAM BOOK, ii. 237). 

WILLIAM BYJLD, 1538-16x3. 















44 AN ELIZABETHAN VIRGINAL BOOK 

(for repeat), 







WILLIAM BYRD. 

(a) The solitary ornament given in the MS. of this piece is 
printed above, in bar I. 

(ft) The two strains are plainly meant to " repeat." See 
the final crotchet preceding the first double bar, and the two 
quavers in the bar preceding the " pause chord " at the end, 
which clearly lead back to the repetition of strain 2. 

(c) Something seems wanting in bars 6 and 10. 

(d) The inquiring student will rejoice to find several con- 
secutive fifths in three consecutive bars. 

But the very various rhythms and times of the Gigges in 
the Fitzwilliam Book are worth looking at. All the five 
instances are in the second volume, and I will here give 
examples of their rhythm formulas. 

Thus, whereas the "Cobbler's Jig" (1622), and Bull's 
"King's Hunting Jig" (1580 or later) are in 4 time, and 
have no suspicion of triplets in them, we find that in the 
l8th century (Bach, etc.), the Gigue seems to have taken the 
triplet rhythm as its principal characteristic. 

The examples in the Fitzwilliam Book are as follows : 

ii. 257 "DOCTOR BULL'S MYSELFE* -GiGGE. 

Time Rhythmical Scheme. 

3 r r rr rr- r r r rr rr- r etc - 

A I I 



AN ELIZABETHAN VIRGINAL BOOK 45 

ii. 258 (Also by BULL). 



Time 

3 



r r- pr r- 



Rhythmical Scheme. 



r r r r r 



etc. 



etc. 



This latter is a curious experiment in mixed rhythm, quite 
in Bull's characteristic manner, and is worth playing. 

ii. 237 WILLIAM BYRD (with F. Tr. in margin). 

Time Rhythmical Scheme. 

> J JJ - JVJ lrrrr r 



etc. 



j=;t J J EE T I * 



letc. 



ii. 416 GILES FARNABY A very curious casein two ways : 
(a) It is in 4 time, and 
(If) The strains are built in Nines, not " eights." 



Time 



Rhythmical Scheme. 



etc. 



r r r r r r r r 






:fe=5=:r 



There are four sections of this Gigge, and all of the last 
three are variations on No. I. 



46 AN ELIZABETHAN VIRGINAL BOOK 

ii. 162. RICHARD FARNABY. " Nobody's Gigge." A 
good tune. 



n 




Rhythmical Scheme. 


a 


r 


r 


r r 


cr 


r 


r 


r 


r 


etc. 


Y^ , ^Jfc _). 


fN 


H- 


-=s 








.*_ 


F= : 


^F^F 


**_ 
~" U" 


p -- etc. 






. ^ 


=**-( 


i 


- 


L 


1 


P 




JCE 


F 1 - 



in "eights," with an "outside" crotchet (see above). 

Richard Farnaby was the son of Giles. 

The Farnabys came from Cornwall, and therefore Mr 
Tregian may have known them personally. At any rate 
there are 50 pieces by the father, and 4 by the son, in 
the Fitzwilliam Book. 

Giles Farnaby took his Mus. B. degree at Oxford, from 
Christ Church, in 1592, and in his "supplicat" stated that 
he had studied music for twelve years. 

Twenty-nine years later (1621), he was one of the con- 
tributors to the harmonised tunes in Ravenscroft's Psalter. 

I have more to say about the Farnabys, which must be put 
off till a later occasion. 

Both Giles (the father) and Richard (the son) were 
persons of great gifts, and are certainly amongst the most 
important of the authors represented in the Fitzwilliam 
Book. 



CHAPTER V 

TOYE, CORANTO, LAVOLTA, GALLIARD, ROUND, SPAGNIOLETTA, 
MORISCO, ETC. 

I HAVE mentioned the small class of dance pieces called 
"Toye." There are six pieces given with this name; one 
has been already spoken of, as it appears in two places (ii. 
260, "Toye," and ii. 267, "Coranto"), there being slight 
differences between the copies, which seem to show that Mr 
Tregian did not recognise the piece, as he reproduced it 
almost immediately under the name of Coranto. It is pretty 
certain that he must have had it in tiuo MSS., from different 
sources. 

But other examples are to hand, ii. 412 and 413, which 
make it clear that a " Toye " need not be a Coranto, nor a 
Coranto a " Toye." Both these pieces are called " Toye," 
but the one (Duchesse of Brunswick's Toye, by Bull) is in 
-J time, the other (Anon.) in quadruple time. 

Both are worth playing. The harmony of the second strain 
of the Anon. "Toye," ii. 413, is remarkable : ii. 421 contains 
a " Toye " by Giles Farnaby, the first strain of which is 
practically the same as Bull's "Duchesse of Brunswick," and 
on the whole is a better piece than Bull's. 

Here we may consider a particularly charming and beauti- 
ful Galliard by Giles Farnaby (ii. 419), which happens to be 
copied into the Fitzwilliam Book alongside of these " Toyes." 
It consists of the usual three strains, beginning with a sweet 
melody in A minor, which is not at all unlike what Rameau 
(1683-1764) would write, more than a century later. The 



48 AN ELIZABETHAN VIRGINAL BOOK 



third strain (p. 420) is very remarkable for the beauty of the 
antiphonal use of the two hands. 

We now proceed with miscellaneous Dances, such as the 
Lavolta, Round, Spagnioletta, Morisco, and Muscadin. 

LA VOLTA (Lavolte), FITZWILLIAM BOOK, ii. 188. 

\T.he quid-time 



THOMAS MORLEY (1563-1604). 
Set by W. BYRD. 







J. 



J=J= 



2-d A J dg'ip^-^g * i G* JQ 

3^ IB ^_ C2 - t0 | x*y i^ 

^=E=iE3?=g=z= Yl I p-^H^-p-gzg- 

^^ I^ZtlJg-^ M ' 1 1 I i i ^P. 




The variations on the two strains are omitted. 

In connection with the two Lavoltas (both by Byrd) found 
in the Fitzwilliam Book, I will say a few words about the 
class of triple-time Dances, of which the Lavolta is one. 

The father and mother of all these trochaic Dances is the 



AN ELIZABETHAN VIRGINAL BOOK 49 

BASSE Dance, of the I5th century. (" Basse" means par has, 
i.e. sliding along the ground.) 

I here quote from " Shakespeare and Music " Arbeau's 
account of this Dance, as it is given in " Orchesographie " 

(1588):- 

II On p. 25, Capriol (the Pupil) asks his Master (Arbeau) 
to describe the steps of the 'basse' dance. This was the 
' danse par bas, ou sans sauter,' which was of the 1 5th 
century, was in triple time, and contained three parts, A, 
basse dance ; B, Retour de la basse dance ; C, Tordion. 
This third part, or tordion, n'est aultre chose qu'une 
gaillarde par terre ' ; i.e. the Tordion of a Basse dance was 
simply a Galliard par terre, without the leaping or ' Sault 
majeur.' 

" Before Arbeau answers his pupil, he gives him some pre- 
liminary instruction as to the etiquette of the ball-room. He 
says * In the first place . . . you should choose some virtu- 
ous damsel whose appearance pleases you (telle que bon vous 
semblera), take off your hat or cap in your left hand, and 
tender her your right hand to lead her out to dance. She, 
being modest and well brought up, will give you her left 
hand, and rise to follow you. Then conduct her to the end 
of the room, face each the other, and tell the band to play a 
basse dance. For if you do not, they may inadvertently play 
some other kind of dance. And when they begin to play, 
you begin to dance.' 

" Capriol. If the lady should refuse, I should feel dreadfully 
ashamed. 

" Arbeau. A properly educated young lady NEVER refuses one 
who does her the honour to lead her out to the dance. If 
she does, she is accounted foolish (sotte), for if she doesn't 
want to dance, what is she sitting there for amongst the 
rest ? 

" The Master then gives his pupil an account of the basse 

D 



50 AN ELIZABETHAN VIRGINAL BOOK 

dance, the 1st and 2nd parts of which are composed of 
various arrangements of the following movements 

1. La reverence, marked with a big R. 

2. Le branle (not the dance of that name), marked with b. 

3. Deux simples, marked ss. 

4. Le double, marked d. 

5. La reprise, marked with a little r. 

"The 'chanson* 'ue. the dance tune, was played on the 
flute, and accompanied by the ' tabourin ' or drum, which 
beats all the time. Every ' bar ' of the music is called 
either a ' battement ' of the drum, or a ' mesure ' of the 
chanson. 

" Now Arbeau explains the steps and time of each of the 
above five movements. 

" I. R. This takes four bars. Begin with left foot forward, 
and in doing the reverence, half turn your body and face 
towards the Damoiselle, and cast on her ' un gracieux 
regard.' 

" 2. b.* Also takes four bars. Keep the feet joined together, 
then for the first bar, swing the body gently to the left side ; 
second bar, swing to the right, while gazing modestly upon 
' les assistants ' ; third bar, swing again to the left ; and 
for the fourth bar, swing to the right side, looking on the 
Damoiselle with an ' oeillade defrobee, doulcement et 
discretement.' 

"3. ss. 1st bar, left foot forward; 2nd bar, bring right 
foot up to the said left foot; 3rd bar, advance the right 
foot ; 4th bar, join the left foot to the said right foot ; et 
ainsi sera parfaict le mouvement des deux simples. 

" N.B. Always suit the length of your steps to the size of 

*The branle (not the dance, but as used here) is called Congedium by 
Anthoine Arena. Arbeau thinks because the dancer appears about to take 
leare of his partner i.e. prendre conge. See Hen. VIII., iv. 2. 1. 82, stage 
direction, " congee." 



AN ELIZABETHAN VIRGINAL BOOK 51 

the room, and the convenience of the Damoiselle, who cannot 
with modesty take such big steps as you can. 

" 4. d. 1st bar, advance left foot ; 2nd, advance right 
foot ; grd, advance left foot ; 4th, join right to left. For 
t*wo doubles (dd) do it over again, but contrariwise, beginning 
with the Right foot. For three doubles (ddd), the form of 
the third will be, 1st bar, advance left foot; 2nd, advance 
right foot ; ^rd, advance left foot ; 4th, ' puis tumbera 
pieds joincts comme a este faict au premier double.' And 
thus (he carefully adds) the three doubles are achieved in 
1 2 * battements et mesures du tabourin.' 

" 5. The Reprise (r) is commonly found before the branle 
(b), and sometimes before the double (d) [see the Memoires]. 
In it you have to cultivate a certain movement of the knees, 
or feet, or ' les artoils seullement,' as if your feet were 
shaking under you. 1st bar, ' les artoils' of the right 
foot ; 2nd bar, do. ; 3rd bar, of the left foot j 4th, of the 
right foot again. 

" The Memoir e of the movements of the basse dance i.e. 
its first Part, is 

R b ss d r | d r b ss ddd rdrb|ssdrbC. 
The C means the * conge,' or * leave ' which you must take 
of the Damoiselle ; salute her, and keep hold of her hand, 
and lead her back to where you began, in order to dance the 
Second Part namely, the Retour de la basse dance, the 
Memoire for which is 

b | d r b ss ddd r d r b | C. 

[The nine movements enclosed between the upright lines, are 
the same in both parts.] 

" Capriol now remarks that he has been counting up, and 
finds that the music of the basse dance proper (part l) has 20 
' fours ' (vingt quaternions), and the retour (part 2) has 
1 2 ' fours.' 

" Arbeau then describes the Tordion, which is Part 3 of the 



52 AN ELIZABETHAN VIRGINAL BOOK 

basse dance. He says it is still in triple time, but plus 
legiere et concitee,' and does not consist of * simples, 
doubles, reprises,' etc., like the first and second parts, but is 
danced almost exactly as a Galliard, except that it is par terre 
I.e. without any capers, and low on the ground, with a 
quick and light step ; whereas the Galliard is danced high, 
with a slower and weightier * mesure.' 

" He gives the following tune, which will fit to any of the 
innumerable diversities of Galliard. If played fast, it is a 
Tordion, if slower, a Galliard. [There are, of course, no 
bars in the original.] 

TORDION OR GALLIARD (CINQUEPACE). 



123 456 









" Here are the Steps of the Galliard, consisting of five move- 
ments of the feet, and the caper, or * sault majeur.' The 
five steps give the Galliard the name of Cinque pas. 

1. Greve gaulche. ['Greve' is explained as a 'coup 

de pied.'] 

2. Greve droicte. 

3. gaulche. 

4. ,, droicte. 

5. Sault majeur. 

6. Posture gaulche. 

"I, 2, 3, 4, 6, are the ' Cinq ' pas, and 5 is the character- 
istic leap or caper. 



AN ELIZABETHAN VIRGINAL BOOK 53 

" The next six minims are danced to the Revers, which is 
just the same, except that the words ' right ' and ' left ' (droicte 
and gaulche) change places all the way down. Then repeat till 
the tune is finished. 



AIR D'UNE VOLTE. [LA VOLTA.] 




^1" 



9 CM-C* & 

1. Petit pas, en saultant sur le gaulche, pour faire pied 

en 1'air droit. 

2. Plus grand pas du droit. 

3. Sault majeur. 

4. Posture en pieds joincts ; 

etc., all over again every two bars. 

" The sault majeur of the high lavolt' comes at the semi- 
breves in this tune." 

Thus, the Tordion is Part III. of the BASSE Dance, which 
was par terre. The Galliard is the Tordion with the sault 
majeur or " leap " or " caper." The Volta (Lavolt) is the 
Galliard cut short as to its steps, and with most of the 
importance laid on the " caper," which was " high " (see 
" Twelfth Night"), and adorned with a whirling movement 
in addition. (See ii. 1 80, 1 88, for Lavoltas by William Byrd. 
The tune of the latter is by Thomas Morley.) 

The Round, i. 248, Byrd's " Sellinger's Round," also called 
" The Beginning of the World." Chappell, vol. i. 256. 
[Sellinger = St Leger (not far from Cornwall again.)] 

And ii. 292, Byrd's " Gipsy's Round." 
Both are in f , and with f * T f as a characteristic. 



54 AN ELIZABETHAN VIRGINAL BOOK 

Both are worth playing. The former has a curious like- 
ness to a well-known hymn tune, " St Theodulph." (See the 
S. John Passion, No. 28. Date about 1725.) 

All that seems known about the Dancing of the Round is, 
that it has some relation to the Hay, or Raye, which was 
popular both in England and France in the 1 6th century. 

Haie means Hedge in French, and " faire la haie" seems to 
be explainable as follows : 

Arbeau says first the dancers dance alone, each separately , 
then together so as to interlace, " et font la haye les uns parmy 
les aultres." That is, during each batch of four steps, the 
dancers change places one with another, so that if there are 
three dancers, A, B, C, in the first four steps, B and A 
change places, and make B, A, C ; in the next four steps, C 
and A change places, and make B, C, A, etc. 

Here is the tune and the formula of steps 

THE HAYE 







Beginning at the first complete bar, and reckoning one 
step to each semibreve I. Deux simples (ss). 2. Double 
(d). 3. ss. 4. d. 5. ss. 6. d. 7. ss. 8. d. 



AN ELIZABETHAN VIRGINAL BOOK 55 

The formula of steps ss d ss d (repeat) would at all events 
carry the dancers "round" the room. 

The Spagnioletta, of which there are two examples, i. 199, 
by Giles Farnaby, and ii. 471 (the Old Spagnoletta) by 
the same. The first is in 4 time, but the second in triple, 
besides showing a very strong family likeness to the " Sel- 
linger's Round." 

We now come to the 

Mcrisco, ii. 373, a solitary but excellent specimen in the 
Fitzwilliam Book. Anon., and called " The King's Morisco." 
It has six short strains, which admirably illustrate the im- 
mense difference between the key effect of our modern 
Scales and that of the old Modes ; some of them ending on 
a chord of C major, others (of course) on a chord of A major. 
The square time continues until the sixth section, when 
time begins. 

Other Moriscos with which these may be compared are in 
"Shakespeare and Music," p. 151, Arbeau's tune (the steps 
given seem to bring the Morisco very near to the modern 
clog dances of the music-hall), and the Moresca of Monte- 
verde (Orfeo, 1 608), the latter being all through in triple 
time (" Shakespeare and Music," p. 206). 

This last is a very ingenious piece, as it consists simply of 
8 bars (or 4, according to the barring), which are repeated 
in 4 different keys, so as to produce 32 bars of music out of 
the same phrase ; and it manages to finish the fourth repeti- 
tion on the same chord on which the first begins : so the 
dance can be repeated ad placitum. 

The Muscadin, \. 74 and ii. 481 (which gives variations on 
the same tune by Giles Farnaby), is not unlike the accepted 
Hornpipe tunes. This particular example (a solitary one) is 
found in Chappell's "Popular Music of the Olden Time" 
under the name of "The Chirping of the Lark" (Woold- 
ridge's ed., vol. i. p. 277), and thus furnishes us with one 



56 AN ELIZABETHAN VIRGINAL BOOK 

of the many cases where a dance tune is used as a song, or 
vice versa. 

I will close my account of the numerous Dances of the 
Fitzwiiliam Book by referring to some of the best examples 
which are contained in this work. 

Pavana by W. Byrd, ii. 384. 

Galliard by J. Oystermayre, ii. 405 (see above, p. 22). 

Pavana by Thomas Tomkins, ii. 51 (best played without 
agremens, see above, p. 1 6). 

Galliard by Thomas Morley (ii. 177). 

Two Almans by Robert Johnson, ii. 158, 159 (see below, 

P- 137)- 
Coranto, Anon., ii. 310. 



CHAPTER VI 

FUGAL FANTASIAS (ONE SUBJECT) 

UP to this point we have tried (and, of course, failed) to 
deal completely with the Dances of the Fitzwilliam Virginal 
Book. We have seen something at least of about a dozen 
sorts of dance tunes which were current in Elizabethan 
times ; and made some inroads amongst the 130 to 140 
different examples of these which are preserved in this one 
book alone. 

We still have before us the consideration of 46 Songs, 17 
Organ Pieces, 9 arrangements of Madrigals, 22 Fantasias, 7 
" Romantic" Pieces, 19 Preludes, and the important class of 
6 works on the Hexachord, besides the general questions 
concerning Tonality, Acoustical Temperament, Musical 
Form, and other historical matters, which can all receive 
light from a careful study of the nearly 300 pieces in this 
volume. Furthermore, at least 28 different composers 
whose names and works appear in the Fitzwilliam Book 
must have some notice given to them of a directly personal 
kind, and some attempt must be made to appreciate and com- 
pare the various features of their compositions. 

I have already pointed out that in the Fitzwilliam Book 
various "forms" of our modern music are to be seen in 
their infancy, and that this circumstance alone would make 
the book invaluable to the student, e.g., in Chapter I., where 
the "variation form" of the l6th century was brought 
before the reader. Numberless " variations" on Elizabethan 
songs and dances occupy a large proportion of the pages 

57 



58 AN ELIZABETHAN VIRGINAL BOOK 

of the Fitzwilliam Book. Specimens are given below, pp. 
70-84. We will now proceed to the examples of another 
form of modern music which can be studied in an early 
stage in the Fitzwilliam Book, viz., that of Fugue. 

The thing called Fugue existed in the 1 6th century under 
the name Fantasia, or Ricercare. The name "Fancy" was 
common in England in the 1 7th century, i.e. "Fantasia" 
anglicised. The Ricercare was simply a more elaborate ex- 
ample of fugal writing, and its name (" sought out," 
recherche, exquisite) merely indicates that every possible 
contrapuntal ingenuity would be exhibited. But the names 
were used with no particular care, and in any case the great 
point to remember is, that both practically mean "Fugue"; 
but "Fugue" as it was practised a hundred years before 
J. S. Bach was even born. The word " Fugue" in the 1 6th 
century did not indicate what is now understood by the 
term. When Morley (1597) gives us an example of what 
he calls a Fugue, it turns out to be what we describe as a 
Canon, or Imitation. 

So, some time between 1600 and 1700, the name of these 
contrapuntal works was changed from Fantasia or Ricercar 
to the name of Fugue, which was certainly used in Germany 
during the latter years of the 1 7th century (Buxtehude), and 
in England earlier still, viz., in 1665, when John Milton used 
the term in the Eleventh Book of "Paradise Lost." 

" the sound 

Of instruments that made melodious chime 
Was heard, of harp and organ j and who moved 
Their stops and chords was seen ; his volant touch, 
Instinct through all proportions low and high, 
Fled and pursued transverse the resonant fugue." 

Almost all the best music of the i6th and 1 7th centuries, 



AN ELIZABETHAN VIRGINAL BOOK 59 

except the single-voice songs, or duet dialogues, or dances, 
was fugal. Whether vocal or instrumental, the "form" of 
part music was practically only one, that of the fugal Fantasia 
or Ricercar. The church music for voices, or the chamber 
music for viols, were not only singularly alike in general 
tone, but were most usually constructed in the same way, 
i.e. the fugal way. There might be one subject, or several, 
or a large number, introduced in turn, either singly or in 
combination with others ; but the general plan was that 
known as Fantasia ; and that, again, was in its principal 
features almost the same as what has been known, now for 
two hundred years, to musicians as FUGUE. 

We shall find it interesting, both musically and historically, 
to examine in some detail a few of these older examples of 
Fugue taken from the Fitzwilliam Virginal Book, which 
contains twenty-two of them, by eight different composers, 
who were all living at the same time as William Shakespeare. 
It is worth remembering here, that, although Shakespeare 
uses musical terms with such familiarity, he never mentions 
the word Fugue ; but he does refer to the musical form we 
are considering, under the term we should expect, viz., 
Fancy in the second part of " Henry IV." iii. 2, 1. 320, 
where Falstaff meditates on Mr Shallow's lies about his 
wildness in early youth. 

Fal. He [Shallow] came ever in the rearward of the 
fashion, and sung those tunes . . . that he heard the carmen 
whistle, and sware they were his fancies, or his good nights 
. . . etc." 

The same kind of music is referred to in " Cymbeline," 
ii. 3, 1. II, where Cloten provides instrumental and vocal 
music as a serenade to Imogen, and asks for two pieces : 

"First, a very excellent good-conceited thing: after, a 
wonderful sweet air, with admirable rich words to it." 

This "good-conceited thing " is nothing else than a "Fancy " 



60 AN ELIZABETHAN VIRGINAL BOOK 

or " Ricercar " for viols, and is aptly described by Cloten, 
being made always as full of "good conceits " (/.*. contra- 
puntal tricks, as of diminution, augmentation, imitation, 
answer per Arsin et Thesin, or Recte et Retro, etc.) as it would 
hold, or as the ability of the composer would permit. 

The names of the composers of such pieces as this, 
which are found in the Fitzwilliam Book, are : Bull, 
Byrd, Giles Farnaby, Morley, John Munday, Peter 
Philips, Nic. Strogers, and J. P. Sweelinck, the Dutchman. 
(The index of the printed Fitzwilliam Book credits W. 
Tisdall with a Fantasia. This is merely a mistake for 
Tisdall's Galliard, ii. 486.) 

All are English except Sweelinck of Amsterdam, who is 
connected with two of the others, Bull and Philips, in a 
peculiar way, for Bull at one period of his life lived in 
Antwerp, being organist of Notre Dame there in 1617; 
and Philips, an English Catholic priest, also lived in the 
Netherlands. 

We will make a good beginning of this subject by studying 
Sweelinck's Fantasia in ii. 297 of the Fitzwilliam Virginal 
Book, and a Fantasia by Philips, i. 335* both of which 
exhibit some of the most characteristic features of this 
" form " as it was practised in England, Italy, and the 
Netherlands (not to mention Germany), in the latter part of 
the 1 6th century. 

Both these works are lengthy, and might, with some 
justice, be called "dull"; but there is no doubt of their 
historical value, and I imagine it is very largely the change 
in the mental attitude of modern audiences that denies such 
music its artistic success. However that may be, let us look 
at them with reverent interest as examples of the practice of 
the older method of contrapuntal writing which made J. S. 
Bach a possibility, and (finally) an actual master in this kind. 

These two Fantasias (Dutch and English) have more than 



AN ELIZABETHAN VIRGINAL BOOK 61 

one thing in common besides their great length. Both have 
one mam subject, and they stick to that one subject in an admir- 
able way. 

Both works are written for a keyed instrument, e.g., the 
harpsichord or organ, whereas many works of the sort 
were scored for strings or wind. 

Both of them are full of the quaint conceits alluded to 
above, presenting the subject in Imitation by direct motion, 
by contrary motion, by augmentation, by diminution, by 
double diminution (i.e. the subject introduced in notes of 
a quarter the original length), in stretto, and in combination 
with subsidiary motives. 

Both of them are true Fantasias, as they are built on 
original subjects, not on ecclesiastical plain-songs. 

Here is the "subject" of Sweelinck's Fantasia, together 
with its " answer" in contrary motion. 

J. P. SWEILINCK. 

\_Ansiver in Contrary Motlon.~\ 

* ^ ' 

J J '* ^ J J J 

*\ <P ~ m 



This subject is a long one, and the very first thing Sweelinck 
does is to give the whole of it in " answer " by contrary motion, 
at the distance of a single bar (ii. 297, line l). 

On line 4 an episodic passage begins, and the Fantasia is 
fairly under way. 

Some of the " scientific " features are : Page 298, bottom 
line, subject by augmentation (treble) ; 299, line 4, answer 
of the same by contrary motion (bass) ; 3* line 4 (end), 



62 AN ELIZABETHAN VIRGINAL BOOK 



augmentation (tenor); 301, bottom line, an irregular sort of 
augmentation, combined with close answers in crotchets (bass); 
302, line 5, etc., close imitation on second part of the original 
subject ; 303, line 4, eight bar " Pedal" in treble part. 

This Fantasia is in the Dorian (transposed) mode, with one 
flat, showing the changed pitch. It is important to remember 
that the notion of " key " (in the modern sense) was not 
invented in the Elizabethan times. It is merely an accident 
that this piece of Sweelinck's happens to begin with a phrase 
founded on a chord of D minor, and that the signature seems 
to correspond. All the flat means here is, that the Dorian 
mode is moved up a fourth. 

Philips' Fantasia (i. 335) is almost equally ingenious in its 
construction. [Byrd's Fantasia, ii. 406, is on the same subject.] 

An interesting feature is (p. 335, 1. 3, etc.) the use in 
episode, of a passage familiar enough one and a half centuries 
after in Handel's "Amen" Chorus, and in many other iyth 
and 1 8th century contrapuntal works. 

SPECIMEN OF "HANDELIAN" MATERIAL, FROM THE 
l6TH CENTURY. 

Taken from Fantasia (Fitzivilltam Book, i. 335)- 

PETER PHILIPS. 







etc. 



Another curious thing is the fact that the one subject is 
brought in thirty-nine times (each time carefully numbered). 
This, there is little doubt, Philips would regard as " an excel- 
lent good conceit," referring, in the crooked mystic manner 



AN ELIZABETHAN VIRGINAL BOOK 63 



of the time, to the words of St Paul in 2 Cor. xi. 24, " Of 
the Jews five times received I forty stripes save one." 
References to contrapuntal subtilties in this piece are : 

Page 339, 1. 3, subject by Augmentation. 
338, 1. 4, Diminution. 
342, 1. 4, Double Diminution. 
341, 1. 4, Close Stretto. 



EXTRACT FROM FANTASIA (FITZWILLIAM BOOK, i. 357) BY 
NIC. STROGERS (showing system of numbering the 
successive appearances of the subject). 

N. STROGKRS. 

1. 3. 




l f 






64 AN ELIZABETHAN VIRGINAL BOOK 

The last bars of the same, as beautiful as the former are 
ugly. 








J_!UL_ i l fc L_ 




NICHOLAS STROGERS. 

Other instructive features of this example of Strogers are 
(a) The uncompromising nature of the " false relations," 
e.g., in bars 4 and 8 ; (b) the varying lengths of the " bars," 
some containing six minims, others four or two, showing that 
the "signature" refers to the relative lengths of semibreves 
and minims, not to the size of bars. 



CHAPTER VII 

FUGAL FANTASIAS (SEVERAL SUBJECTS) 

WE now will consider two other Fantasias, in which several 
subjects are used, instead of one only, by Bull (i. 423) and 
Byrd (i. 1 88). 

In the former (which is in two parts only, except 
quite at the end), no less than four different subjects 
appear on p. 423, and several more occur in the following 
pages. This work of Bull's is, however, very remarkable, 
on account of the unusual nature of p. 425, where a pedal 
is introduced, consisting of continual repetition of the fol- 
lowing passage in the bass 






This motive is repeated more than twelve times, and must 
have been a wonderful, if not alarming, exhibition of con- 
tinuous purpose to those who first heard it. 

Byrd's piece (i. 1 88) also has several subjects I., p. 188, 
11. I, 2, 3 ; II., end of 1. 3 to p. 189, 1. 3; III., p. 189, 1. 3, 
and onward. 

An interesting and very unusual thing is presented at the 
top of p. 189, viz., a case of literal repetition of a phrase 
which has already been given at the bottom of the previous 
page, last bar, etc. 

IRREGULAR ANSWERS 

We must now pay some attention to the system of Subject 
and Answer, or Dux and Comes, as it was often called. As 

E 65 



66 AN ELIZABETHAN VIRGINAL BOOK 

is well known, the method of bringing in the various vocal 
or instrumental " parts" in a Fugue of the 1 8th century was 
cut and dried. One feature, as students are aware, was, that 
the voice or part which came second in order of entrance had 
to imitate the melody enunciated by the first voice or part, 
but at a different pitch. To put this commonplace piece of 
knowledge in a concrete form, supposing the Fugue to be in 
two " parts," X and Y : X begins, e.g., with a tune made of 
the first four notes of the major scale of C, viz., c d e f e dd c. 

That is the Subject or Dux. 

Y then joins in and imitates these notes, but instead of 
beginning on the note C, she begins on the note G, 
and produces this g a b c b a a g. This is the Answer 
or Comes. In the meantime X proceeds, accompanying the 
answer with a "Counterpoint" (i.e. a melody which fits the 
answer, but runs " counter " to it, if possible furnishing a 
contrast of style), for instance : 

Subject (Dux) 



=: 



m 



r 

Answer (Gomes') 




Here the lower "part" gives out the subject, beginning in 
.the key of C, on the note C. The upper part "imitates" 
this, beginning its " answer " on the note G, and eventually 
turns the " answer" into the key of G, as we call it nowadays. 

This method was, as I said, cut and dried in the i8th 
century, and no one, as a rule, would have dared to 
vary it. 



AN ELIZABETHAN VIRGINAL BOOK 67 

But this plan was not always so firmly established ; and in 
this connection we will study a few examples, illustrating 
what may be called the History of Irregular Answers, taking 
instances which occur in the Fitzwilliam Virginal Book, in 
Fantasias by Bull and Byrd. 

In considering these irregularities (by which word I merely 
mean cases which do not follow the rule of the 1 8th century 
with unanimity), it is well to recollect a trick of G. F. Handel, 
examples of which may be found in a good many of his fugal 
oratorio choruses, viz., the method of a " false answer " to 
the fugue subject. A familiar case is in the "Messiah" (1741), 
" Blessing and honour " (Worthy is the Lamb). Here the 
basses (and tenors) give out the subject, "Blessing and 
honour," in the key of D. It is answered " falsely " by the 
sopranos, who give out the same notes, still in the key of 
D. Other cases are easily found in Handel's works, e.g., in 
" Samson," the final chorus, " Ever to sound His praise," the 
fugue subject of which has a " false answer " of the same sort. 
These things are cases of the survival of antique methods in 
Handel's more modern works. 

This irregular method may be seen in the Fantasias of the 
1 6th century, preserved for us in the Fitzwilliam Book. Good 
specimens are the Fantasias of Bull, i. 423 ; and of Giles 
Farnaby, ii. 343. 

In the example of Bull the "answer" (p. 423, 1. 2) is 
merely the same notes as the " subject," repeated in a 
lower octave. This Fantasia has several subjects, e.g., a 
second and third melody make their appearance on 1. 3 ; 
both answered regularly. These, however, are succeeded 
by a fourth subject (end of 1. 5), which has a "false" 
answer, reckoning from the last two crotchets of 1. 5. 
The same thing is observed on p. 424, where the regular 
"answer" seems almost forgotten. 

Giles Farnaby (ii. 343) supplies a like case in his first line. 



68 AN ELIZABETHAN VIRGINAL BOOK 

This is an interesting piece. The subject has been well worn 
since the l6th century. 



Amongst the many who have used it, we remember J. S. 
Bach, who composes one of the best-known of the forty-eight 
Preludes and Fugues upon these very notes. Let us not forget 
that the English Giles Farnaby knew the phrase more than a 
century before J. S. Bach was heard of. Another point is 
noticeable at the bottom line of p. 343, bar 2, where a harmony 
occurs 

GILES FARNABY (100 years before Purcell). 



f 55 11 ^3 


:=j_ g^ 


^ T 


.* * *J 

j2J * fl^ 

-i IT i 

rj ' ' o ' 

tL> 



which our simple historian gossips have identified with the 
name of Purcell for so long that it will perhaps never be 
possible to undo the false impression they have produced, 
and establish it as dating more than a century earlier. Of 
course, all who know Tallis's five-part Litany (it is almost 
universally neglected, by the way) know that this harmony 
occurs there also, in the setting of " We beseech Thee to 
hear us, good Lord." This takes the passage back still 
further in point of time, as Tallis was living in the early 
half of the 1 6th century. 

The first three pages of this Fantasia of Giles Farnaby's 
are worth playing over. A second subject is introduced on 

P- 344> 1- 3- 



AN ELIZABETHAN VIRGINAL BOOK 69 



EXTRACT FROM GILES FARNABY'S FANTASIA (FITZWILLIAM 
BOOK, ii. 343) showing " false answer " of the second 
voice. Mixolydian Mode. 

G. FARNABT (i6th Century). 
Very slow 



















rrr 



etc. 



After seventeen more bars like this, a second subject 
is brought in. A good sample of its working is the 
following : 



7 o AN ELIZABETHAN VIRGINAL BOOK 

G. FAKNABY, 




EEg=J^=J^^=F=^E 




Byrd's Fantasia (i. 1 88) in the ^Eolian Mode is also worth 
playing through. The fact that it has several subjects has 
already been mentioned, and that an unusual case of literal 
repetition takes place on p. 189, 1. I. But it is interesting 
mainly in that the " answers " are regular. For instance, 
p. 1 88, top line : four different voices come in, one after the 
other, in the order Tenor, Bass, Alto, Treble. The first 
notes of each, in order, are, E, A, E, A. This, of course, 
is quite " regular," except that the last comer begins at short 
notice, only allowing the alto to give out three notes of the 
" subject," before chiming in with the " answer." The next 
line again (line 2, bar i) is nearly regular; the order, Bass, 
Tenor, Treble, (the tenor introduces an imitation of the 
subject, beginning on the final crotchet of line 2, continuing 
on line 3), Alto (line 3, bar i) ; first notes (reckoning from 
line 2, bar i, bass), E, A, E (treble last bar, line 2), A (alto, 
line 3, bar i). 

A second subject in crotchets begins at the end of line 3 
in the alto (the three crotchet /s), answered by the tenor, 



AN ELIZABETHAN VIRGINAL BOOK 71 

and introducing a " tonal " alteration of the "subject," i.e. 
e e e f is " answered "by a a a c. This may be only 
an accident, but it will serve to direct the student's attention 
to a new department of the History of Fugues, viz., the 
origin of Tonal Fugue as opposed to Real Fugue. (The 
latter simply means that the " answer" is a "real" literal 
imitation of the " subject," merely altered as to the part of 
the scale in which it is written; whereas "Tonal" Fugue 
means that the "answer" is altered, more or less, in its 
intervals, on account of the thing called " Key," or " Tone.") 

Now the fact really is, that the Fantasias of the i6th 
century were very generally what we call " Real " Fugues. 
Those in the Fitzwilliam Book are so, almost without ex- 
ception. That is to say, the " answers" are literal transcrip- 
tions, without alteration (except of pitch) of the " subjects." 
This was the case throughout the early history of Fugue. 
Now, we must remember always that there were no such 
things as " keys " in those days, at least in our modern 
sense. The fact that a piece began on the note G, and 
ended on a chord of G, did not mean, as it must mean now- 
adays, that it was in the key of G. The fact that a piece 
had a flat in its signature did not mean that it was in the key 
of D minor or F major. Such ideas were in their early 
infancy in Elizabethan times. 

And apparently, it was not until the older methods 
associated with " Mode " had been overlaid by the notion of 
"Key," that the difficulty of "Tonal" answer in Fugue 
made itself felt. 

An example will make the meaning of this clear to any 
musical person. I take Byrd's Fantasia (i. 37) in the Fitz- 
william Virginal Book. The subject consists of a portion of 
the scale of G, including G a b c d. We note, by the 
way, that there is no sharp in the signature, although the 
piece strikes our modern eye and ear as being in the key 



72 AN ELIZABETHAN VIRGINAL BOOK 

of G major ; also it ends on a chord of G ; nevertheless, 
Byrd thought of it as in the Mixolydian Mode, the " final " 
of which is the note G ; and indeed even a beginner will 
easily notice that the note F is as frequently natural as sharp, 
and as a fact that the use of the F| is much circumscribed 
throughout. 

Now, Byrd makes this a " Real " Fugue, i.e. he " answers " 
the " subject " literally, viz., G a b c d is " answered " 
by D e f$ g a, twice in the first two lines. Nevertheless, 
Byrd felt a difficulty at bar 2, for he wished to bring in the 
bass at bar 3, and apparently felt that to answer d b G by 
a fj D, would somehow be out of place, and accordingly he 
alters the answer at bar 2. However, he does not do it 
again in line 2, so we will let that pass. What it is necessary 
for us to grasp now is, that the practice of the 1 8th century 
(Handel, Bach, etc., etc., etc.) would have been different. 

Here is Byrd's practice, and (below) that of the men 150 
years later. 

FANTASIA IN MIXOLYDIAN MODE. 

W. BYRD (i6th Century). 




TENOR. 

BASS. 
TREBLE. 



JR-r^Tj 


D 


^-V? 

f^* "* 


^^ 

etc. 


L e <s> 


L ^ 


=i =* 





* " Accidental alteration " by Byrd. Compare with the following. 



AN ELIZABETHAN VIRGINAL BOOK 73 



Fugue on Byrd's subject, as it might have been done by 
men of the l8th century, who thought in the " key" of G 

major. 

E. w. N. 




Byrd's Fugue is Real, the alternative version is " Tonal," 
i.e. the " answer" is altered from the " subject " in order to 
fit with the more modern conception of the "key" of G 
major, which was not in the mind of William Byrd. In this 
example of the 1 8th century method I have taken care not to 
alter the " style," e.g., the dulness of the alto part, continually 
returning to the note D, is quite of the 1 6th century, not of 
the 1 8th, which, whatever faults it developed, on the whole 
avoided dulness in Fugue. 

An example of a different kind, which seems to confirm 
this view, is found in Wasielewski (" Hist. Instr. Mus. XVI. 
Cent."), who gives an instrumental Fugue (dated 1532) in four 

* Compare these with the corresponding notes in the Jirst bar. The 
crotchet D is altered, the minim F| is not. F# is the major third of D, and 
is the proper tonal answer to the B in bar i, which is the major third of G. 

f These two are the Tonic and Dominant of the scale of G, and are the 
* regular" answers to the Dominant and Tonic which are observed in bar i. 



74 AN ELIZABETHAN VIRGINAL BOOK 

parts: where the treble begins the "subject" on C, the alto 
"answers" it, beginning on F; but the tenor and bass, 
instead of doing exactly the same over again, and thus pre- 
serving the " key effect," go on from the alto with further 
intervals of a fifth, and begin their repetition of " subject " 
and " answer " on Bb and Eb respectively. 

This, of course, has the natural result of lowering the 
general pitch of the piece by a whole tone at the second 
appearance of the " answer." The composer seems, however, 
not to feel the smallest embarrassment in this predicament. 
Nor is his procedure in any way remarkable, except in that it 
very naturally fails to exemplify the practice of a coming 
century, which did not draw its life blood either from Modes 
or Hexachords, but from major and minor scales. 

Some few instances from the Fitzwilliam Book will be use- 
ful as illustrations of this matter, which is still dealt with in 
University examinations in Counterpoint. 

I will give Jive examples of what appear to be hints of the 
practice of "Tonal" Fugue in the 1 6th century. I lay no 
stress on four out of the five, but give them as possible illus- 
trations, no more. 

Byrd. Miserere, ii. 232, where, in the first line, A Bb is 
answered in the bass by D F, not D E. 

MISERERE. 

W. BTUD (Fitzwilliam Book, ii. 131) 



The Alto is the 
" Canto Fermo." 




AN ELIZABETHAN VIRGINAL BOOK 75 

On p. 233, 1. 3, bar I, there is an obvious mistake in the 
copy, which causes an apparent case of altered "answer"; 
the alto part should read (in crotchets), A B C D, not 
A C D, and should begin at the fourth beat of the bar, not 
the fifth. 

There is, however, one more genuine alteration of the 
kind, p. 233, 1. 4, last bar and following, where the bass 
G B C D, etc., is answered by the treble A B C D, etc. 

But there is reason for not attaching much importance to 
either of these examples. 

Byrd, Fantasia, ii. 408, top line, where C G, in the alto, 
is answered by G C, in the treble. 



etc. 




ffrtT 



But this is apparently set at nought by the " Real " 
answers which bristle elsewhere in the next two lines, e.g., 
the D G, etc., in the lowest voice of the final bar of the 
extract given. 

Bull, Fantasia on Canto Fermo, i. 138, where in bars i 
and 2, C B A is answered by F E C|, instead of FED. 
Perhaps some little reliance can be placed on this, as the Canto 
Fermo does not make it impossible for the answer to be 
" Real " ; in fact it would go better and more pleasantly 
thus : 

j~~ E. w. N. ~'\ 

i 



C.F. 



: r 



76 AN ELIZABETHAN VIRGINAL BOOK 



J_ 



jt: ^ * * J J~ 


h^ g 


-^ 


^_ ft 


rd 
JZL 


>-^ p_p_M 


1 


MTT 



etc. 



Compare bars 2 and 3 (treble) of this alteration with the 
original. 

FANTASIA ON CANTO FERMO 

JOHN BOLL (Fitzwilliam Book, i. 138). 



C.F. 



<sU 



->- 



etc. 



It is quite clear that the alteration, with " Real " answer, 
makes the passage more easy and natural. Therefore it 
seems possible that here Bull really had some slight intention 
of giving the subject a " Tonal " answer, even at the cost of 
graceful melody or pleasing harmony. 

Mor/ey, ii. 57, gives what is apparently a " Tonal " reply on 
line I, A to E being answered by E to A. This is a trivial 
instance. 



AN ELIZABETHAN VIRGINAL BOOK 77 



The only case, however, which appears really certain, and to 
be mentioned with confidence, is 

JByrd, Fantasia, i. 406, where the (comparatively) unim- 
portant scale of eight notes from C is answered by a 
corresponding scale of eight notes from G ; but in the 
subsequent crotchets which complete the subject, the 
notes C A B G, etc., are answered by G D E C, etc., 
thus bringing the music back into the right key. 

FANTASIA. 

W. BYRD (Fitzwilliam Book, i. 406). 



Brightly. 




t&rr-rT-^T 



etc. 



Also see the last three bars, on p. 410, where the treble 
C E F G is answered in the alto by G A B C. 

WILLIAM BYRD. 
>_J N/JJ 



End of 

the same 

piece. 



J- 



a 



J- 



=t=te= 



g 



II 



An apparently plain case of a plagal answer to a Fugue, 
is that of Thomas Tallis, " Felix Namque" (No. 2), on page 
I of volume ii., and dated 1564. 

The first four and a half lines of this work are a fugal 
exposition of a subject, three semibreves in length, where 
the main notes, A D, are answered by D A, with almost 



78 AN ELIZABETHAN VIRGINAL BOOK 

perfect consistency. The one irregularity is in the treble, 
I. 2, bar I, where the A is reached from E instead of 
D. This is a good example, as the subject or the answer 
appears no less than seven times in the four and a half 
lines. 

A difficulty in reconciling the practice of Tallis in this piece 
with the Modal Theory seems to be, that as the work is in 
the -/Eolian Mode, the plagal answer to the note A would 
be E, not D. It is sufficient to say here that the answer is 
unusual, and doubtless to be regarded as an effort towards 
the creation of a new " phase " of the mode. 



CHAPTER VIII 



SONGS OF THE l6TH CENTURY 

OF popular songs, Shakespearian and otherwise, a large 
number are given in the Fitzwilliam Book. As with the 
dance tunes, so with the songs j they are used for the pro- 
duction of variations, mostly of the kind already familiar to 
those who have studied the dances of the Fitzwilliam Book, 
i.e. the variations are merely ornamental, consisting of running 
passages and florid semi-contrapuntal overlayings of the tune. 
This as a general rule; there are a few cases where the 
possibility of other methods seems to have been felt, e.g., 
Giles Farnaby's "Daphne" (ii. 12): 

FIRST PHRASE OF THE FIRST STRAIN OF "DAPHNE," SET 
BY GILES FARNABY (FITZWILLIAM BOOK, ii. 12). For 
the words, see Chappell, i. 150 (Wooldridge ed.). 




The " Rep." or Variation on the same. 




8o AN ELIZABETHAN VIRGINAL BOOK 




Last phrase of the third strain of " Daphne." 




Varied thus on p. 14, with the melody in the left hand 
part : 




AN ELIZABETHAN VIRGINAL BOOK 81 







r 

where on p. 14, line 5, the melody is in the tenor, and the 
right hand has a florid passage, indicating the harmony very 
clearly, and going in a contrary direction to the melody. 
Again, "William Byrd's " Walsingham," i. 267 (a much more 
artistic work than Bull's on the same tune, i. i), where 
variations 2, 3, 4, treat the melody by imitation, instead of 
mere agr emeus or ornamentation ; also variation 5 (pp. 267-68), 
where the melody is in the bass, and the other voices have 
imitations founded on the melody itself. 

VARIATION 5 ON "WALSINGHAM" (with the tune in the 
Bass, and accompanied by imitations). See below, p. 1 19, 
for the plain melody. 

BYRD (1538-1623). 
Andante tranqtiillo. 

-&L *- , 1_ 




rrftt 



82 AN ELIZABETHAN VIRGINAL BOOK 



r 



=g 






> ' ' 

(a) In bar 2, the low D is omitted by Byrd, as it is out of 
reach. 

(b) In bar 4, the bass is ingeniously made to answer an 
imitation leading to the coming phrase of the melody at 
bar 5. 

(c) The melody relapses into the conventional bass of a 
cadence at the third crotchet of bar 7. 

(d) Observe the time signature, which neither prescribes 
nor disallows triple rhythm. 

This work of Byrd's is of great interest and value, and shows 
Byrd at his best, which means a good deal. Other interest- 
ing passages are, variations 13 (tune nearly always inside the 
chords), 14 (curiously like a well-known passage in Wagner's 
" Siegfried Idyll," or would be, if the tune was played on an 
oboe, and the left-hand part legato on strings), 

ANTICIPATION OF WAGNER'S "SIEGFRIED IDYLL" IN BYRD'S 
" WALSINGHAM " VARIATIONS (FITZWILLIAM VIRGINAL 
BOOK, i. 270). 

BYKD (1538-1623). 
Variation 14. 

How should I know your true 



7iki 


-p ? J & 


==3E 


/" 


\j m 

Andante. 
ft m 


9 9 


ff&i 


-7~f Z3-I* hS * m + 


-tr^ 


ys|E 


-E^-^dr rfi= 




AN ELIZABETHAN VIRGINAL BOOK 83 



love From anoth 



er one 







16 (not unmusical in its use of the rhythm of 3 against 2, 
whereas Bull's exploits in this line are often merely clever 
and ugly), and 22 (where at page 273, line 3, bar 2, the 
melody is missing in the copy, bed, being two crotchets 
and a minim). Still again, the anonymous setting of "Bara- 
fostus' Dreame," i. 72 (? whether Dreame = Dreme, i.e. 
" Song," for the only words that seem to be known to it 
are those in Chappell's " Old English Popular Music," vol. 
i. 148, called "The Shepherd's Joy"). In this work the 
third variation (last line of p. 72) hides the tune in the tenor, 
and introduces above it a sort of canon in quavers between 
treble and alto. 



FIRST LINE OF " BARAFOSTUS' DREAME " (set by an anony- 
mous composer of the l6th century, who may be 
William Byrd, judging by the style). 




84 AN ELIZABETHAN VIRGINAL BOOK 
1 







THE gRD VARIATION ON THE ABOVE (showing the tune in 
the tenor, accompanied by a well-marked "figure" in 
imitation). 




(dr) In line i, bar 3 (containing three dotted minims, in 
contrast to the bars preceding and following), two crotchet 
rests appear wanting in the R. H. 

(b) In line 2, the treble resumes the melody in bar 3 at 
the ffcj- 

(c) There is probably some correction to be made in the 
last bar given above, marked (?). Most likely the first 
quaver R. H. ought to be " f," not " g." 



AN ELIZABETHAN VIRGINAL BOOK 85 

(d) These extracts contain two instances of a sign or 
obliteration of a sign which has puzzled students of the 

MS., viz., 1^1 (applied to the crotchet " g " in the R. H. 

of bar 3 of the former line, and gg (occurs with quaver 
"b flat" in the last bar R. H. of the latter specimen). 
Nothing seems known of them, if they really are ornaments. 
Probably they are merely obliterations of the single and 
double line, explained above, p. 32. 

However, on the whole, the variations on these songs are 
not of any greater interest than those we have already con- 
sidered in connection with Pavans and Galliards, etc. 

The real interest of these songs is that they must have 
been popular, since so many variations were written on them. 
Besides this testimony, the allusions to many of them in con- 
temporary plays are numerous, and we remember that a 
song would have to be a household word with the public 
before it rose to be mentioned familiarly on the stage. 

Some of them are actually songs used in plays of Shake- 
speare, e.g., " O mistress mine " ; more commonly, however, 
they are alluded to, or quoted, e.g., Callino Casturame, 
Walsingham, Fortune, Robin. 

The total number of songs in the Fitzwilliam Book is about 
fojty^two. The number is slightly doubtful, because one or 
two are not certainly songs, for no words seem to be known, 
nor any connection with words; e.g., "Heaven and Earth" 
(i. 415), by F. Tregian, which is almost surely an ecclesiastical 
piece, and not a song. 

Six of the forty-two songs have two arrangements, this, I 
suppose, tending to show that they were even more popular 
than the others. Such are, " Go from my window " (Morley, 
i. 42, and Munday, i. 153 ; but these are nearly the same, and 
probably the double ascription is a blunder), " The woods 



86 AN ELIZABETHAN VIRGINAL BOOK 

so wilde " (Byrd, i. 263, and Gibbons, i. 144), " Wai sing- 
ham" (Bull, i. i, and Byrd, i. 267), "Why aske you" 
(Anon., ii. 192, and G. Farnaby, ii. 462), " Barafostus' 
Dream" (Anon., i. 72, and T. Tomkins, ii. 94), and 
" Bonny Sweet Robin" (G. Farnaby, ii. 77, and J. Munday, 
i. 66). 

We will make a beginning with five songs from the Fitz- 
william Book which are used or alluded to in Shakespeare. 
(I omit " Hanskin," ii. 494, which may belong to " Jog on," 
Autolycus' song in " A Winter's Tale.") 

The very first piece in the Fitzwilliam Book is the raria- 
tions by Bull on " Walsingham." "Walsingham" is an 
ancient tune, probably belonging to the quite early 1 6th 
century,* and is associated with various sets of words 
dealing with the visits of pilgrims to the shrine of Our 
Lady of Walsingham in Norfolk. This shrine was cele- 
brated in the time of Henry III. (who was there in 1241), 
of Edward I. (who visited it in 1280 and 1296), and of 
Edward II. (in 1315). Henry VII. spent Christmas of 
1486-87 at Norwich, and made a pilgrimage from there 
to Walsingham ; and next summer, after the battle of 
Stoke, he sent his banner to be offered before Our Lady 
of Walsingham. 

It was said that the Milky Way pointed directly to the 
home of the Virgin, to guide pilgrims on their road ; 
hence it was called the "Walsingham Way." Also crosses 
at every town pointed the way to Walsingham along the 
earthly counterpart of this heavenly path. 

Catherine of Arragon, when dying, recommended her 
soul to Our Lady at Walsingham ; and so did her royal 
husband. Thus the reputation of the place was very 
great at the beginning of the 1 6th century, and had been 
so for 300 years. But this came to an end in 1538, when 

* For the plain melody of " Walsingham," see below, p. 119. 



AN ELIZABETHAN VIRGINAL BOOK 87 

the priory was dissolved, the image of Our Lady being 
burned at Chelsea by the commissioners ; and the gains 
of the Norfolk people were stopped at their source. 

The Percy Folio Manuscript (ed. Hales and Furnivall, 
Triibner, 1868), vol. iii. p. 471, gives the words of an 
ancient ballad which goes to the tune as it is in the Fitz- 
william Book, and contains a line or two of Ophelia's 
scattered scraps of songs in " Hamlet," e.g., " How shold I 
know your true loue | that haue met many a one," etc. 

Dent's " Shakespeare's Songs " gives the three verses (p. 
77) sung by Ophelia in " Hamlet," iv. 5> which are part of an 
old ballad on similar lines to that in the Percy Folio. 

Caulfield's "Collection of Shakespeare Vocal Music" (1864) 
gives the tunes and portions of tunes to which actresses 
traditionally sang Ophelia's quotations. These are printed 
in "Shakespeare and Music" (Dent, 1896), p. 196, where 
the traditional air for " How should I your true love know ? " 
etc., is clearly a corrupt form of the tune " Walsingham " in 
the Fitzwilliam Book. 

Percy's Reliques (Bohn, vol. i. p. 312) gives a single verse, 
quoted from the Pepysian collection, which also goes to this 
tune. 

The old ballad from the Percy Folio is a dialogue between 
a lover who has lost his love, and a pilgrim whom he meets 
returning from Walsingham, which was known as the English 
" Holy Land." 

I venture to suggest that Byrd had this dialogue form in 
his mind, when he set his first harmony of "Walsing- 
ham," as it may be seen in the Fitzwilliam Book (i. 267), 
obviously portraying two persons singing alternate lines 
of the poem. 

Callino Casturame (corruption of Irish, " Colleen oge 
Asthore"). It has already been mentioned that Shake- 
speare's Pistol refers to this song in " Henry V. " iv. 4, 



88 AN ELIZABETHAN VIRGINAL BOOK 

1. 4. It is a charming tune, and the words may be seen 
in Chappell, i. 84. The words of the title are used as 
a refrain, so the tune has to be sung nine times to get 
through the verses. 

The words as given there were printed in 1584. 
" Fortune my Foe" or " Fortune." This tune is set by 
Byrd (i. 254). One verse is given in Chappell (i. 76), but only 
for want of anything older, for it is probably not the original 
set of words. Shakespeare mentions the tune in " Merry 
Wives," iii. 3, 1. 62 : 

Fahtaff\\.Q Mrs Ford]. I see what thou wert, if Fortune 
thy Foe were not, Nature thy friend. 

A number of songs were sung to this tune, and in the 
1 7th century it was used largely for doleful and lugubrious 
verses. Thus in 1634 ' ll was s P oken of as " the hanging 
tune," and in 1641, " that preaching tune," in connection 
with songs concerning the execution of criminals. 

A much earlier instance of this is dated 1588-89, "The 
judgment of God shewed upon Dr John Faustus : tune, 
Fortune my Foe." This has woodcuts of Dr Faustus signing 
the contract with the devil, etc. 

Moreover, the old ballad of " Titus Andronicus," on which 
the so-called Shakespeare play was founded, was also sung to 
" Fortune," and with propriety, judging by this rule, for 
nothing could be more gloomy than the story (Percy Reliques, 
Bohn, vol. i. p. 162), or more dreary, for there are thirty 
verses of it. 

Robin, or " Sonny Sweet Robin" set by Giles Farnaby (ii. 77) ; 
and John Munday (i. 66). Farnaby's the longest and best. 
The tune is in "Shakespeare and Music," p. 198. 

Scarcely any words remain, but the one line sung by 
Ophelia in Hamlet, " For bonny sweet Robin is all my joy," 
and a line which may or may not have some connection with 
it, " My Robin is to the greenwood gone." 



AN ELIZABETHAN VIRGINAL BOOK 89 

For the sake of having some words, I have strung together 
two verses, including these two lines, which can be used for 
this tune: 

" My Robin is to the greenwood gone, 
My Robin has left me quite alone. 

Sad are the days, alas, 

Slowly the hours do pass : 
Bonny sweet Robin is all my moan. 

My heart is sore with all annoy, 
My thoughts are set in one employ. 

Weeping I grieve for him, 

All would I leave for him : 
Bonny sweet Robin is all my joy." 

" O Mistress mine " This, of course, is a complete Shake- 
speare song, " Twelfth Night," ii. 3. The clown sings it 
to Sir Andrew and Sir Toby. 

The tune is set by Byrd (i. 258), and there are six varia- 
tions, well worth hearing. It was also set by Morley, but 
this version is not in the Fitzwilliam Book. 

Of those songs which are not Shakespearian, one of the best 
is " Barafostus* Dream" already mentioned. It is set by Tomkins 
(ii. 94), whose work is difficult to play; also by an anonymous 
writer (i. 72), whose composition is more valuable artistically, 
though much shorter. I suspect it may be Byrd's, on account 
of the general style, which is more concerned with the display 
of the possibilities in the given melody, than with the display 
of the virtuosity of the performer ; also on account of vars. 3 
and 4, which tend to "disguise" the tune rather than to 
over-ornament it. (See above, pp. 83, 84.) 

Chappell (i. 148) gives a set of words called " The Shep- 
herd's Joy." 

Another beautiful tune is Daphne (see above, p. 79), set by 



90 AN ELIZABETHAN VIRGINAL BOOK 

Giles Farnaby (ii. 12). The melody has three strains, and a 
variation on each strain is introduced before the next is 
played. Interesting passages are: p. 14, 1. 5, where the 
tune is in the left hand ; and p. 15, 1. 5, where the rhythm of 
three against two occurs in the bass staff, accompanied by a 
passage in broken thirds, which is really an ornament of the 
plain harmony set to the last two bars of sect, i or sect. 3. 

Chappell (i. 150) gives two pleasant verses. 

A good example of the humorous song of the l6th 
century is Martin said to his Man. 

This is set (ii. 275) by Anon, with one variation only, and 
the words are supplied by Chappell (i. 140), who says they 
were printed in 1588. 

Under this heading may also be included Go from my 
window (i. 42 or i. 153, either by Morley or Munday), 
some words of which are in Chappell (i. 146, 147); and John, 
come kiss me noiv, set by Byrd (i. 47)* one verse being in 
Chappell (i. 269). Various other verses are preserved in MS. 
in the library of Trinity College, Dublin. 

These were printed in " Longman's Magazine," vol. xxvii., 
November 1895, an d are here given, with the tune, from the 
Fit z william Virginal Book. 

"JHON COME KISSE ME NOW." 

Harmony imitated from WILLIAM 
BYRD, in Fitzwilliam Book, i. 47, 

by E. W. N. 
\Wife\ v. i. Jon come kisse me now, now, Jon come kisse me now. 




AN ELIZABETHAN VIRGINAL BOOK 91 



Jon come kisse me by - and - by, and mak no mor a - dow. 




r^-r 



This is followed by His answer to yt sam toone. 

[Husband] 

v. 2. Peace Pm angrie now, now, peace I'm angrie now, 

Peace I'm angrie at the hert, and knowe not qt to dow. 

Here is an example of the way this tune may be fitted to 
the next eleven verses. 

Husband] 
v. 3. Wyfs can faine and wyfs can flatter : heav I not hit them now : When 





















v. 4. Wyfs ar good and wyfs are bad . . . heav I not hit them now . . . 
once they beginn they still doe chatter, and so does my wyf too. 

Wyfs . . . can make ther hus - bands mad : and so does my wyf too. 

The rest of the song (verses 5-14) is as follows : 

5. Wyfs can sport and wyfs can play : heav I not, etc. 
And wt little work passe ower the day, and so does, etc. 

6. And wyfs hes many fair words and looks : heav I not, etc. 
And draws sillie men on folies hooks : and so did my 

wyf too. 

7. Wyfs will not ther meetting misse : heav I not, etc. 
A cowp of seek they can well kisse : and so can, etc, 



92 AN ELIZABETHAN VIRGINAL BOOK 

8. Wyfs can dance and wyfs can lowp : heav I not, etc. 
Wyfs can toome the full wyne stowp : and so can, etc. 

9. Wyfs can ban and wyfs can curse : heav I not, etc. 
Wyfs can toome ther husband's purse ; and so can, etc. 

10. Wyfs can flyte and wyfs can scold : heav I not, etc. 

Wyfs of ther toungs they heav no hold : and nane hes 
my wyf too. 

11. Wyfs they'r good then at no tym, neither is my wyf now, 
Except it be in drinking wyn, and so is my wyf too. 

12. Some they be right needful evills, so is my wyf now, 
Wyfs ar nothing else but divles, and so is my wyf too. 

1^. Now of my song I make ane end, lo heir I quyt the now, 
All evill wyfs to the divle I send, amongst them my 
wyf too. 

14. Peace Pm angrie now, now, peace Fm angrie now, 

Peace I'm angrie at the hert and cannot tell qt to dowe. 



One more good tune is Rowland, set by Byrd (ii. 190) : 

"ROWLAND," OMITTING -REP." 

(FITZWILLIAM BOOK, ii. 190). 

W. BYRD (1358-16x3). 
Not too fast. tf | 



1 



AN ELIZABETHAN VIRGINAL BOOK 93 

IV I ^ 




Chappell gives two verses of the song " Lord Willoughby's 
March," as belonging to this tune, but they fit badly enough. 

" The woods so ivi/de," or " As I walked the wood so wild," 
is set by Byrd(i. 263) and Gibbons (i. 144). A short extract 
is given below, see p. 145. 

The words are not extant, but the poetry of the subject 
was plainly realised in the setting of Byrd, who shows un- 
mistakable signs of picturing the rustle of the great forest, 
and the woodland atmosphere generally. Orlando Gibbons' 
setting is more wooden ; still there is something poetical 
about it, as one would expect from him, and as is not to 
be expected from Bull. 



CHAPTER IX 

"FANCY" PIECES 

A SMALL but remarkable class of works in the Fitzwilliam 
Virginal Book is that of " Fancy Pieces." 

Some of these are, Byrd's "The Bells" (i. 274), John 
Munday's Fantasia on the Weather (i. 23), and Giles 
Farnaby's three pieces (ii. 260-62), called 

(1) " Giles Farnaby's Dream" (a Pavana). 

(2) " His Rest " (a Galliard). 

(3) "His Humour." 

GILES FARNABY'S DREAME (FITZWILLIAM 
BOOK, ii. 260). 

G. FARNABY (i6th Century). 
Slow. 



i'r-^f^-^^^- 

, , 


e_ ^-J--^^ 

1 | i 


T 4 t~d i i 1 rh 


r rr r 

: fl 


^-^'^f^f- f 



r 1 r 



AN ELIZABETHAN VIRGINAL BOOK 95 



ogx 




The mordents marked in the MS. (only two) are given. 
For other examples of the final repetition of the chord of the 
key, see above, p. 32, and below, p. 132. 

There is also a piece by Munday (ii. 449) called 4< Monday's 
Joy," but it is of no special interest except in the form of 
the title, and may be classed with (ii. 257) " Doctor Bull's 
my selfe," a Gigge, which consists of a couple of ordinary 



96 AN ELIZABETHAN VIRGINAL BOOK 

variations on a rather comical J tune, otherwise not re- 
markable. 

Another little piece is ** Farnabye's Conceit " (ii. 424), not 
very remarkable. 

Monday's " Fantasia on the Weather," though interesting, 
is mere experiment when compared with Byrd's " Bells." 

The various meteorological events, Faire Wether, Light- 
ning, Thunder, Calme Wether, Lightning, Thunder, Faire 
Wether, Lightning, Thunder, Faire Wether, Lightning, 
Thunder, and finally " A CLEARE DAY," are successively 
imitated by Mr J. Munday ; and the piece is about as truly 
affecting as the simple list of phenomena named above. (The 
very first note of this piece probably ought to be A, not G. 
Also the note CJ in line 3, queried by the editors, may well 
be right.) 

I class it with the "Battle of Prague," a very popular 
after-dinner piece about 1850, containing indications (in print) 
of the passages which were supposed to represent the advance 
of cavalry, the rattle of musketry fire, the groans of the 
wounded, and the other noises of war ; the whole con- 
cluding with "Malbrouk" and " God save the King," as in 
duty bound. 

Before leaving this, we may remind ourselves that Munday's 
attempt is a very early specimen of " programme music," and 
thus is worth looking at, in spite of its comparatively small 
musical value, as representing the l6th century. Later 
composers whose names are associated with the growth of 
programme music, are Kuhnau (lived late in the lyth century), 
whose "Biblical Sonatas" (pub. 1 700), aiming at the re- 
presentation of well-known scriptural stories, are interesting 
both as music and as exemplifying a stage in the history of 
descriptive music (a modern reprint is to be had), and J. S. 
Bach, who in the early part of the 1 8th century wrote his 
" Capriccio on the Departure of a Friend," representing the 



AN ELIZABETHAN VIRGINAL BOOK 97 

farewell speeches of acquaintances, the laments and tears of 
friends, their final leave-taking, the arrival of the coach, and 
the journey, enlivened by an active posthorn. 

Later still, we have Beethoven's Sonata in Eb, represent- 
ing not only the departure, but the absence and return of 
a friend. 

Thus John Munday's l6th century piece deserves more 
than passing mention, although there were others of about 
his time, e.g., Frohberger (d. 1667), who tried their hands 
at programme music with such success as was possible. 

But Byrd's " Bells " (i. 274) is in every way more valuable. 
This piece convinces me that Byrd was more of a musician 
than Bull. It is a genuine attempt to embody the romance 
of church bells, in terms of the art of the l6th century, and 
without any of the vanity of technique which stares one in 
the face in almost all Bull's compositions. Points of interest 
in this piece are : 

(1) P. 274, line I, the bells "ring up," beginning with 
the " tenor " (or largest bell). This is still the practice of 
change ringers. 

(2) Throughout the whole piece Byrd never fails to re- 
member that the lowest bells, especially the " tenor," are 
the most impressive. This is common knowledge to persons 
who have often listened to " changes," especially on large 
peals of ten or twelve bells. It is even the case that the 
three or four lowest bells in such a peal are the only ones to 
pierce through the general jangle, and preserve their separate 
existence. This is very apparent at a considerable distance. 
Byrd attempts, with some success, to realise this general 
effect. 

(3) An example of the effect of wind or atmospheric 
change, such as is commonly noticed. It is indicated, not 
by / and p, for the virginals could not achieve this, but by a 
difference in the octave. See 11. 3 to 6. 

G 



98 AN ELIZABETHAN VIRGINAL BOOK 

(4) P. 279, var. 9, where the gradual weakening of the 
"peal," produced by "ringing down," is depicted. This is 
especially clear in the last line but one, where the little in- 
complete descending scales, tumbling over one another, are 
quite characteristic of the effect referred to. 

"The Bells" may well be associated with "The woods so 
wilde" (i. 263) of the same composer (dated 1590), or 
" Walsingham " (i. 267) in this respect. All three pieces 
stamp Byrd as a "romantic." For a contrast with Bull, the 
settings of "Walsingham" may be directly compared, as 
both writers have dealt with this melody (Bull's, i. i). 

All three of these compositions of Byrd are worth hearing, 
and deserve any pains to bring out the expressive power that 
they certainly possess. 

The " Romantic " quality of Byrd is well shown in his 
" Cantiones Sacrae " (some of which are still sung in Cathedral 
services as anthems) better shown in these than in the 
virginal music, inasmuch as the voice was a perfect instru- 
ment even in the 1 6th century, whereas the virginals were 
far below the pianoforte, even below a bad pianoforte. One 
must remember (it is easy to forget) that the virginals (also 
the similar ordinary harpsichord of a later time) had no ex- 
pressive power whatever, no possibility of soft and loud, or 
mechanism which would allow of different vocal "parts" 
being made prominent or the reverse. Consequently the 
virginal music does not do the same justice to its composer 
as the vocal compositions do. 

Giles Farnaby's three little " fancy " pieces are all very 
short, but undoubtedly are what is now called " characteristic." 

(i) Giles Farnaby's " Dreame " (ii. 260) is clearly a poetical 
effort, and contains three sections, all of a placid character. 
" Formally," i.e. in respect of its time and number of strains, 
it is a Pavan ; but its " character " is certainly not that of 
the usual square cut dance. 



AN ELIZABETHAN VIRGINAL BOOK 99 

(2) ' His Rest." This, in form, is a Galliard, i.e. it has 
three strains and is in triple time, but the title is easily 
justified by sympathetic playing. 

(3) " His Humour," on the next page, is only longer than 
the two former pieces, because variations (called " Rep.") are 
added to some of the phrases. 

The word "humour" in Shakespeare (which should give 
us Giles Farnaby's general understanding of the word) in- 
dicates a "wayward" fancy. 

Hence this piece is named. It is not Pavan, Galliard, 
Fantasia (or Fugue), Alman, Gigge, Air, or what not ; it is 
" His Humour," or wayward fancy. The " Humour " is 
plain, and is seen in the chromaticism of line 2, and the 
sudden slow pace of the minims, after the crotchets and 
quavers of line I, and in the mild joke of setting the 
Hexachord in section 4. 



CHAPTER X 

HEXACHORDS 

THIS brings us to a very remarkable class of six works on 
the Hexachord,* or 

Ut, Re, Mi, Fa, Sol, La. 

One is by Sweelinck (ii. 26), two by Byrd (i. 395; i. 401, 
Ut, mi, re, fa, etc.); two by Bull (i. 183; ii. 281), and 
the passage of Farnaby already named (ii. 262). 

* The word Hexachord means "six notes," and is the ancient (nth 
century) name for the three scales which the mediaeval writers referred to 
in their "Gamut." It is easily understood by an illustration 

i. n. IL -*"- 



ut re mi fa sol la | ut re mi fa sol la | ut re mi fa sol la 

I. is the Hexachoraum Durum. 

II. is Hexachordum Naturale. 

III. is Hexachordum Mollt, or " soft" Hexachord, the word "soft" referring 
to the Bemol, or Bfat. 

It will be observed that these scales of six notes are all exactly similar 
in respect of their tones and semitones, viz., they are the same as the first 
six notes of our modern Major Scale. 

The first G, a later addition (for obviously the original scale must have 
been named from A), was known as Gam-Ut, being marked by the Greek 
letter gamma, whereas the other two Ut's were C fa ut, and F fa ut, the 
double "solfa" names having the effect of deciding the actual octave in 
which the note referred to was to be found. "Tenor" C, for instance, is 
Fa in. Hexachord I, and Ut in Hexachord II, thus it was known as C fa ut. 
A more complicated case would be the C above, viz., " middle" C, which 
would be called C sol fa ut, a compound of the names of C in all three 
Hexachords. 



AN ELIZABETHAN VIRGINAL BOOK 101 

These pieces introduce us to a subject which is as difficult 
as it is interesting to the student of musical history, namely, 
the origin of " Key " in the modern sense. In close 
attendance on this, there is also presented the question of 
the History of Temperament (i.e. the question of how 
musical instruments should be "tuned"), which has always 
been a trouble to the practical musician, and probably always 
will be, in spite of the labours of such enthusiasts as Perronet- 
Thompson and other acousticians of the ipth century. 

In the first place, these pieces are a living demonstration 
of the fact that the Hexachord was the only notion of a 
scale, in our sense, that the classic i6th century possessed. 

This is important, as it shows us one reason why the 
notion of " Key " was so slow in coming to the front. 
What I mean is, that the Hexachord could not "define" a 
" Key " so completely as our more recent scales do, inasmuch 
as it was without a " Leading" Note, i.e. the top note but 
one, which we regard as a strong characteristic of any key, 
major or minor. 

Second.ly, Bull's work on this subject (i. 183) introduces 
in a remarkably bold manner the question of Temperament,* 

* A most valuable experiment may be made by any student who wishes 
to have a clear notion of "Temperament," and the physical limits of 
Harmony. All that is necessary can be done on a small harmonium with 
one set of reeds, or on one set of pipes in a church organ. The method of 
tuning the small number of twelve semitones in the octave, which such an 
instrument presents, so as to give "just" intonation, as far as those twelve 
sounds can do it, is described in the following diagram given to me by 
Mr G. T. Bennett of Emmanuel College, who has himself tuned, or mis- 
tuned, a harmonium in the Cavendish Laboratory in this very way: 

A E B Fjf 
F C G D 
D|, Aj, El, B|, 

This scheme, read hornnntally, gives a series of perfect 5ths which are to be 
tuned exactly "just," i.e. as a violinist tunes his 5ths, in perfect " intona- 
tion," so that there are no beats, and the vibration numbers of the two notes 



102 AN ELIZABETHAN VIRGINAL BOOK 

and shows plainly that the difficulties of "Enharmonic" 
change of key had presented themselves very completely 
to the musicians of the Tudor period, long before John 
Jenkins wrote his Fancy for three viols (1667 (see the MS. 
at York) which modulates through flat keys from F to 
(nearly) D flat ; and longer still before J. S. Bach wrote the 48 
Preludes and Fugues in every key for which a name existed. 

bear the precise relation of |. e.g., in the top line, A E, E B, B FJJJ, 
where the second note of each pair is the higher in pitch. 

Read vertically, the scheme shows "just" major 3rds, which are to be 
tuned similarly until no beats are perceived, exactly in the proportion 
JjL e.g ., in the fourth column, Ftt D, D B(j, where the second note of each 
pair is the lower in pitch. 

Proceed in this way until the " octave " is " set," and the medium portion 
of the keyboard is tuned "just"; then try simple chords of three notes 
in close harmony. 

It is no exaggeration to say that the unaccustomed observer will find 
himself provided with a new sensation in every combination which he cares 
to test. The perfect musical beauty of a few (a very few), and the dreadful 
cacophony of most harmonies which can be produced in this way, is quite 
beyond description. Here is Nature's final word to the artist in sounds, 
an " everlasting No." In practical music, perfect intonation cannot be, except 
occasionally, and more or less by accident, in the case of voices and certain 
orchestral instruments. In the case of pianoforte and organ, from 72 to 
100 separate sounds in each octave would be required to ensure the result. 
This clearly cannot be thought of except as a scientific toy. Compromise 
is the only resource; either "mean-tone" temperament (where certain 
chords are better than others, and some are impossible) or " equal" tempera- 
ment (where all chords are possible, and none are perfect). 

The following particulars as to the major and minor common chords on 
the harmonium tuned as described above, will be of interest : (a) C G F 
have both major and minor triads good; (/>) D F B[> have both major and 
minor triads bad; (t) Djj E|> A|> have major good, minor bad; and (</) E A B 
have major bad, minor good. Thus iz out of 24 common chords are unusable ; 
and chords of greater complexity than these are naturally worse still, as a 
rule. This series of chords sounds beautifully smooth, 

4- 




but if transposed into A major it is unbearable. 



AN ELIZABETHAN VIRGINAL BOOK 103 

In a sense, Bull's piece recognises the question more com- 
pletely than Bach's work, inasmuch as he tries to use the 
idea of " enharmonics " to produce ''modulation" (i.e. a 
change of "key"), which was only in a limited sense the 
object of Bach. 

Bull's work is not alone here, for Byrd's composition (i. 
395) bears evident marks * of a leaning towards a practical 
adoption of some scheme of " temperament " which should 
make a free " modulation " possible such a modulation as 
he, above all others of his time, was capable of imagining, if 
not of completely realising. 

Byrd's "Ut, Mi, Re" (i. 401) also has hints of a "key" 
of D major. 

William Tisdall's "Pavana Chromatica" (ii. 278) (Mrs 
Katherin Tregian's Paven) is to be considered another piece 
of evidence, so much of it being in a "key" requiring five 
sharps. This is not a mere literal " transposing " (in our 
modern sense) of a " Mode," for the 2nd strain (p. 279) is 
frankly Mixolydian, and only has one accidental sharp, viz., 
F$, as is so common in that Mode. 

Bull's experiment (i. 183) follows the plan of giving the 
Hexachord (harmonised), beginning first on G (the ancient 
Hexackordum Durum of the nth century), and then on 
A, B, etc., a tone higher each time, regardless of con- 
sequences. Almost at once he is confronted by heavy 
troubles. G A B C D E is all very well ; A B Cf D E F# is 

* E.g., 5. 397, 1. 4, where the chord of A flat major occurs. This was 
unbearably out of tune on the old "unequal" (mean-tone) temperament, 
which was still in use in English church organs last century, even in the 
memory of many persons now living. Also, p. 398, last line, where the 
chord of E major is used. This also was painfully bad on the "unequal" 
temperament. Therefore it seems more than likely that both Bull and Byrd 
used a modified form of tuning, closely resembling the " equal " temperament 
now universally adopted for all our keyed instruments, e.g., pianofortes and 
harmoniums. 



io 4 AN ELIZABETHAN VIRGINAL BOOK 

not very unfamiliar to the Elizabethan mind, but B Cjf Dj E 
FJ G$ is decidedly eccentric ; while the next repetition (No. 
4), beginning on Cjf, puzzles the composer with its necessary 
sharp to every one of the six notes. Wherefore he tries 
changing the names of the notes (in practical music that is 
really all that is meant by "enharmonic" change), and calls 
the 4th repetition Db Eb F Gb Ab Bb. Here, one would say, 
Bull was little better off, for the number of flats is all but as 
great as the number of sharps would have been. But, owing 
to the nature of the mediaeval system of music, flats were 
more familiar to composers, e.g., the system of Modal Trans- 
position even resulted in a signature of as many as two flats, 
apart from " accidentals." 

But at this point Bull finds another difficulty ; on the very 
chord marked by the 4 indicating the beginning of the 4th 
Hexachord, he is brought up with a round turn by the 
necessary names of the notes in the " common" chord, which, 
with the Db above mentioned, completes the harmony of the 
moment. These should have been, according to our more 
modern practice, bass and tenor (same note), B double flat, 
treble F flat. One can imagine Bull thinking over this little 
problem in notation. "Who" (he might say), " who ever 
heard of such a thing as F flat or B double flat ? They 
will say John Bull is mad. Call them E and A respectively." 

Having repeated the Hexachord six times, rising a whole 
tone each time, Bull finds his rope has no loose end, but leads 
back to his starting-point, so he makes a dive from F to 
Ab (in the bass this time, p. 184, line 5), whence he starts 
a new series of repetitions from Ab, Bb, C, D, E. Having 
got to E, he is involved in a maze of sharps, as formerly of 
flats, and cuts the knot, after putting in Fj Gj Ajf B CJ D#, 
by slipping from the Fjf to Gt|, and so preparing the way for 
home at var. 14, which, with 15, 1 6, and 17, are all har- 
monisations of the Hexachord on G, and finish the work. 



AN ELIZABETHAN VIRGINAL BOOK 105 

The example by Byrd, however (i. 395), is at once less 
complete and more musical than the experiment of Bull. In 
fact, Byrd's setting of the Hexachord is much more truly 
described as a "Fantasia" (Fugue) on the "subject" 
G A B C D E. This is well enough shown on the very first 
page, both in the "answers" of various "parts" and in the 
use of the " subject " in different forms, while the notes of 
the hexachord are preserved, but not necessarily in the same 
lengths or the same rhythmical formula. Another feature is 
(e.g., p. 397, lines I, 2, 3) where the hexachord is 
accompanied by " imitations," in the other " parts," of a 
secondary " subject," in crotchets, dotted crotchets, and 
quavers. 

Similar remarks are to be made on Byrd's other 
composition on this subject (p. 401), where the order 
of the notes in the hexachord is changed, i.e. instead 
of G A B C D E, the order is GBACBDCE, thus 
introducing a new source of variety in the setting. Here 
again we see the essential difference between Byrd and 
Bull. 

Bull's mind was logical and complete in its own circle, not 
liable to be turned aside by any considerations of traditional 
beauty. Indeed, in the piece considered above, he shows 
himself a ruthless artist, and for that reason I can hardly take 
the result too seriously as a contribution to the New Music 
that was surely passing its obscure infancy in the England of 
Elizabeth, James, and Charles. 

Still, we must take what is offered, and make what can be 
made of it, without going too far in an over-estimate of its 
value as evidence. Page 183, line 5, contains a convincing 
transition from what we must call the " Key of E major" to 
that of D flat. 

This is startling, but the effect is more than fairly smooth 
and reasonable. 



io6 AN ELIZABETHAN VIRGINAL BOOK 

JOHN BULL (1563-1628). 

4. 






The figure 4 marks the beginning of a Hexachord (the 
fourth repetition) on Db in the alto (here altered for clearness 
of reading to CJf tied to Db). 

The chords are more easily appreciated as a modulation in 
this simplified form : 

The same, ii 




Page 184, 1. 5, presents us with a sudden, but by no 
means harsh modulation, from the harmony of "F major" 
to "D flat," which is perfectly satisfactory to the ear. 

JOHN BULL. 




If these instances seem trivial to any student, he may take 
it as a proof that he is sufficiently corrupted by the music of 
the Ipth century, and that he should by all means take a 
course of Italian 1 6th century madrigals or masses, to con- 



AN ELIZABETHAN VIRGINAL BOOK 107 

vince him of his highly coloured misunderstanding. In his 
present state he is incapable of dealing with the subject, 
whether he wishes or no. 

But this matter of "modulation" (in our modern sense) 
must be left for discussion elsewhere. 

The work of J. P. Sweelinck on " Ut, Re, Mi, Fa, Sol, 
La" (ii. 26) is a very good piece, and worth playing. It is 
not only an exercise on the Hexachord, but (as with Byrd's, 
see above) is a regular fugal " Fantasia" or " Fancy," with 
independent subjects which maintain their existence, along 
with and in conjunction with the Hexachord. The first two 
lines are enough to show this clearly 

" Ut, Re, Mi, Fa, Sol, La," a 4 voci, 

Dated l6lX. 

J. P. SWEL1NCK. 

Slow. 



J/_k-yj, t j. ~\~~r3 ^ & 


T& ~~ ^ n : 


I J.JU.JJM ^^ 


1 1 




^^^^^^^^* 

^^EE^E=p===^EEEEEEE^ 



G A BtT D 

where the tenor and treble start a " subject " in a fugal 
manner, and the bass joins in at bar 3 with the Hexachord 
F G A Bb C D,* proceeding then in the next bar to " answer " 
the subject already given out twice. 

* F G A B[> C D was the "Hexachordum Molle " of the nth century, so 
called because of the B[>, or Bemol, " softened B." 



io8 AN ELIZABETHAN VIRGINAL BOOK 

An excellent specimen of the cleverness of contrapuntal 
device, combined with good musical effect, which was 
attained by the best l6th century writers, is the passage 
on p. 27, line 5, where the alto, tenor, and bass have 
the "subject" given above, in very close "stretto" (i.e. 
" answered " as close together as possible), and at the same 
time in " contrary motion," accompanied simultaneously by 
the Hexachord (descending) in the treble. 

J. P. SWKELINCK (from the same work). 
D C Bfr 




I I 

; Subject " or " Dux " in contrary motion (see p. 107). 

G F 




Once again it is found interesting and curious to notice how 
very old is the "material" of G. F. Handel (150 years later 
than Sweelinck), and to compare Handel's " He led them," 
" as through a wilderness " (Israel in Egypt, No. 13), with 
the Fitzwilliam Virginal Book, ii. 31, line 2, and following 
lines. 

The "Coda" on p. 33 is worth attention, presenting the 
Hexachord in close imitation, in quavers and crotchets, 
instead of semibreves, and accompanied by a bass of 
immense vigour and strength, which is repeated twice 
in F, and twice in Bb, during lines 3 and 4. 



AN ELIZABETHAN VIRGINAL BOOK 109 

J. P. SWEJLINCK (same work). 




The whole passage is repeated literally in B flat (a fourth 
higher), thus preparing the way for a "Piagal" Cadence, 
which brings the piece to an end. 

Students of composition and of the history of " school " 
counterpoint will be delighted with the second bar of 

P- 3" 

J. P. SWEELI.NCK (same work). 






12 

4 



12 

4 



which shows with beautiful clearness that the great Nether- 
lander of the 1 6th century cared no more than J. S. Bach in 
the l8th century for consecutive octaves and fifths, if they 
interfered with his plans. Here are six of each sort, one for 
each accent of the bar. 

All that is necessary has now been said about the " Hexa- 
chord" pieces in the Fitzwilliam Book. They show very 
clearly the essential difference in the notion of a " Scale " 
which the classical 1 6th century held, as compared with 
that which obtained only a few years later, viz., that their 
scale was Ut, Re, Mi, Fa, Sol, La, and did not essentially 
include the remaining two notes. 



no AN ELIZABETHAN VIRGINAL BOOK 

They also help to show that certain musicians of the 1 6th 
and early iyth centuries had a system of tuning not far 
different from the " equal " temperament which we now use 
for all keyed instruments, viz., neither " just" nor " mean- 
tone" tuning, but a compromise, in which all the twelve 
semitones in the octave are "equal" in magnitude. 

But the most important thing is the witness borne in them 
to the new growth of harmonic modulation, or, in other words, 
the growth of the motion of "Key," which eventually destroyed 
the Modes of the Middle Ages, and the appearance of which 
marks the beginning of modern music. 

In this connection, the following list of references in the 
Fitzwilliam Virginal Book, which seem to supply evidence on 
the question, may be useful to the student. 

LIST OF REFERENCES in the Fitzwilliam Virginal Book, of 
possible evidence as to a growing sense of " Key," 
as distinguished from " Modal " transition, or chromatic 
alteration, in the music of the l6th century. 

Bull, i. 183. F to Db> p. 184, 1. 5. 

E major to Db> p. 183, 1. 5. (Bull's methods ruthless.) 
Byrd, i. 395. p. 397, 11. 3, 4, 5> G to G minor, C minor, 
C major. 

i. 401. ? Key of D major, p. 401, 1. 4. (Byrd musical.) 
Tisdall, ii. 278. ? Key of B major or thereabouts, p. 280, 
1. 2, etc. good : 

ii. 307. G minor, p. 308, a great struggle for some- 
thing, difficult to describe. 
Hooper, ii. 309. A major, most of first half, and all second 

half. (See above, p. 31.) 
Snueelinck, ii. 31. ? Eb. Not convincing, probably merely 

the result of the Hexachordal repetition. 
Warrock, i. 389. ? Minor " scale " (chords of). 

i. 384. ? Bb major. 



AN ELIZABETHAN VIRGINAL BOOK in 

Anon,, ii. 312. E major, with Tonic, Dominant, Subdominant, 
quite plain in first section only. (See below, p. 1 31*) 

Oystermayre, ii. 405. G major, real " modulations " in line 3. 
One of the most likely cases. Section 2 goes through 
D major, A major, E major, B minor, FJ minor, etc. 
(See above, p. 22.) 

Giles Farnaby, ii. 270, bottom line to top of p. 271. G. D, A, 
E, B. and back to C. But this seems to be a result of 
taking Ut, Re, Mi, Fa, four times up. (See the bass 
of the passage, and tenor of the first two quavers on 
p. 271.) 

Anon., ii. 375. N.B. The Alman on ii. 375 is a useful 
example of what distinctly Modal harmony is, and the 
hopeless difference between that and our modern sequence 
of chords is felt easily. Even playing over the tune 
alone, simple as it is, shows the distinction. See 
below, p. 138. 

Giles Farnaby, ii. 17. " Pawles Wharfe." This tune is 
certainly in D major. Tonic, Subdominant, Dominant. 

Robert Johnson, ii. 159. Alman, in D major. Tonic and 
Dominant, quite strong. An excellent case. A beauti- 
ful and graceful tune. (See below, p. 137.) The 
copy is incorrect at 1. 2, bar 2, where the tenor should 
be e d e (minim); also in 1. 4, bar 2, where an alto part 
is badly missing, viz., from the third crotchet, g fj e 
(minim), in tenths with the low tenor. 

Richard Farnaby, ii. 162, 163. "Nobody's Gigge," in C. 
A quite useful instance. Tonic and Dominant. 



CHAPTER XI 

FITZWILLIAM VIRGINAL BOOK AS A HISTORICAL DOCUMENT 

IT is now necessary to put together the materials available 
in the Fitzwilliam Virginal Book for a chapter on the state 
of English music in the latter half of the i6th century. 

The collection is most interesting to the student of musical 
history, containing, as it does, so many examples of every 
kind of music that was current during the Tudor Period in 
England. 

We have here nearly 300 pieces of music by over 30 
composers of that period. Thus the collection is more 
correctly regarded as a library than as a mere book, for it 
contains more direct evidence of the musical practice of the 
Tudor times than most of us have of the music of our own 
century, as any reader may prove to himself by honestly 
enumerating the composers of the Ipth century whose names 
he knows, and such of their works as he can call to mind. 

A thorough knowledge of the contents of this " book," 
especially when accompanied by contemporary illustrations 
in other forms (e.g., the original Madrigals, of which 
" arrangements" are found in the Fitzwilliam Book), would 
entitle a musician of a critical turn to an opinion as to the 
state of composition in Elizabeth's time, the main conclusions 
whereof would be second in value to none that we have been 
able to secure for the more recent centuries where so much 
larger quantities of material are at hand. 

The first thing that looms large on the student of the 
Fitzwilliam Book is the " dead end " at which the " classical " 
train had brought up about the year 1600. 



AN ELIZABETHAN VIRGINAL BOOK 113 

We consider the Dances, especially the Pavans and 
Galliards, and the numerous Songs, which have received 
the adornment of variations for the delight of the virginal 
player of the Elizabethan day, and what do we find ? 

At once we must honestly say, that on a broad inspection 
of the very large number of specimens of this sort of thing, 
it is only clear that none of these excellent composers could 
think what to do next in the direction of Variation of Melody. 
It is all agremensy and everybody does it alike. A few 
examples suffice for the whole to be fairly appraised.* 

* One example, of a very limited number, where a " variation " is achieved 
without simple ornamentation of the melody, or contrapuntal imitation of 
the melody, is the fourth variation on "The Woods so Wild," by William 
Byrd. The plain tune is to be found below, p. 145, and the student will 
observe that the fourth variation scarcely alludes to the tune, but builds an 
imitative passage taken from it, on to the same harmonies which are set to the 
original melody. 

FOURTH VARIATION ON " THE WOODS so WILD," by W. Byrd (i. 163). 
See below, p. 145. 




I 

Here we see that although the rhythmical figure is got from bar ^ of the 
original tune, the tune itself is disregarded. The extract is also useful as 
showing the time signature 3, meaning here or l^; the irregular use of 
bar lines ; and the absence of the flat in the signature, where the modern ear 

H 



ii 4 AN ELIZABETHAN VIRGINAL BOOK 

There are only a few hints in the book that even the best 
of the Tudor composers had any idea of "Development" in 
the sense that has been the common ground of the best 
writing since their day ; but those few hints are very valuable 
to the student ; slight as they are, they contain the germ of 

new life. 

FORM (External). 

The most important of these may be directly named ; it 
is, that the theory and the practice of the time agreed in 
supposing that the Galliard en suite might be made from the 
same melody and harmony as its proper Pavan, with the 
necessary alterations induced by the change to triple rhythm 
from quadruple. This is, of course, a perfect, though 
simple, case of pure modern " Development." 

And equally, of course, any fool can " play upon this pipe," 
when he has once been shown how ; so let us not forget to 
thank our Tudor forebears for a method that has helped to 
make Brahms a possibility, three centuries after their time. 

Some examples are as follows : 

Edward Johnson, ii. 436 and 440. A striking case, where 
all the three sections of the Pavan are paralleled in the 
Galliard following it. 

FIRST "STRAIN" OF PAVANA (DELIGHT) 
(FITZWILLIAM BOOK, ii. 436). 

EUWARD JOHN-SON (Elizabethan). 
Quadruple Time. 

\. W 1 1 1 r 1 1 tr-H 1 



'W 



d 










\ 



would (at least in bar i) diagnose the key of F. 
in a " key " at all ; see elsewhere.) 



(Of course the piece is not 



AN ELIZABETHAN VIRGINAL BOOK 115 










le 'f g 

J -4-Y^ 

=p=3& 



I 'ketc. 

A J. j P J. 1 
IS 



FIRST " STRAIN " OF GALIARDA (belonging to, and made 
from the Pavana), FITZWILLIAM BOOK, ii. 440. 

Triple Time. 




In these examples the corresponding chords in the Galliard 
and Pavana are marked with the same letters of the alphabet. 

Ferdinando Richardson, i. 27 and 32. 

Dr Bull, i. 131, and the Pavan belonging to it, which is in 

Parthenia, published l6ll (see i. 234, which is practically 

the same). 
Peter Philips, i. 299 and 306, where several sections in 

Pavana and Galliard are clearly parallel. 
Also Byrd's Passamezzo Pavana and Galiarda, i. 203, 209 ; 

and Philips Pavana Paggett and Galiard, i. 291, 296. 



n6 AN ELIZABETHAN VIRGINAL BOOK 

The practice was, however, by no means universal. Still, 
it was not uncommon. 

The arrangement of Pavan and Galliard with a Preludinm 
preceding them, which seems to have been done " on purpose " 
by the compiler of Parthe'nia, "the first virginal music ever 
printed" (l6ll), appears to indicate that the notion of a 
"Suite" extended even further than the mere relation of 
Pavan and Galliard. This again is an undoubted hint of the 
future possibility of the Dance-named Suite of the early 
1 8th century, and thus of the Sonata and Symphony of 
several movements, which is still the noblest achievement of 
pure music in the 2Oth century, 

FORM (Internal). 

Another sign of undeveloped life, a few instances of which 
are to be found in these pieces, is the attempt at Internal 
Unity in certain of the Dance Tunes, by Repetition of some 
kind, apart from " variation." 

As we know already, the interesting beginnings of certain 
well-established features of modern " Form" are plainly seen 
in some quite ancient melodies, *.., church plain songs, which 
now and then show a tendency to " repeat" a portion of the 
tune at the end which has already appeared nearer the be- 
ginning. An instance is " Ein* Feste Burg," the last line of 
which is a literal repetition of the second line. Another 
example is " Christe Redemptor," quoted further on, p. 176. 
Also, for a secular case, compare the Fitzwilliam Book, ii. 
162 and 163 ("Nobody's Gigge," by R. Farnaby). 

This is very elementary in appearance, but it is doubtless 
the germ of the great principle which underlies at once the 
most cut-and-dried formula of a second-rate 1 8th century 
Sonata and the apparent rhapsody of the 2Oth century Ballade 
or Tone-picture. None of them has yet succeeded without 
this principle of Repetition in some shape. 



AN ELIZABETHAN VIRGINAL BOOK 117 

Accordingly, so far back as the 1 6th century, we find the 
principle already (if not recognised in theory, still) adopted 
in the practice of composers. 

We may give as an example " Mai Sims," a Pavan, i. 68 or 
ii. 448 (where the same tune is varied by Giles Farnaby). 
In this Pavan, which is Binary in form, like most Sonata 
movements up to our own time, we find the main features of 
the first half reproduced or alluded to in the second half, 
and it is hardly too much to say that this i6th century melody 
is a more creditable specimen of coherent form than many a 
Sonata movement of two centuries later. It even corresponds 
to the most modern form, in the introduction of something 
like a " free" section (p. 69, last two bars of the top line and 
first two bars of line 2), which makes the second half longer 
than the first, thus weighting it suitably and convincing the 
hearer of a conclusion. 

MAL SIMS (FITZWILLIAM VIRGINAL BOOK, i. 68 AND 
ii. 447). 

\_A song, in Pavan form."] 
Brightly, but not too fait. 







f 4 -^ j j j^Aflj. i tJ 






r --jf-r- rg-*i 



n8 AN ELIZABETHAN VIRGINAL BOOK 



_a_ ^Aa.J_u_ 4 _ r L' 

-g- -j- _l_J_*LJ_J_ J^-l_ J_J. 



^TrTrTTf- 

^-^ ^- a -4-J-|L-J- r J-^-J^.l_J_J_ 

'v? S ^ Oi S . 







I/ I 



" 




The final two bars are in the most intimate connection with 
the first and last bars of the first strain. The one bar tends 
towards the "Dominant," the other towards the "Tonic," 
so that, in a limited sense, the first and second " subjects " 
of modern works are exemplified on a small scale. The 
rhythmical character also of bars I and 2 of the second strain 
is precisely that of the corresponding bars of the first strain. 
The low pitch of bar 2, in the second strain, is an echo to 
the similar passage in line I, bar 3. The two bars marked 
by round brackets are what I refer to as the " free" portion. 

Other instances are: ii. 188, La Volta (i.e. the "high 
lavolt," or quick Galliard, in which the "caper" was the 



AN ELIZABETHAN VIRGINAL BOOK 119 

main feature), a tune by Mor/ey, and arranged by Byrd. 
This short piece is frankly in G major. The first half begins 
on the Tonic and ends on the Dominant ; the second half 
favours the Dominant, and ends on the Tonic. The main 
lines of the " Old Binary " form had not gone beyond this 
l6th century example under Bach in the 1 8th century. 
Butts " Gigge" (ii. 258) has much the same features, and 
another good case is "Pawles Wharfe" (ii. 17), set by Giles 
Farnaby, already mentioned as plainly showing the Tonic, 
Subdominant, and Dominant chords of the "Key" of D 
major. The final two bars of the second half are the same 
as the corresponding part of the first half, but the sequence 
of keys does not include a modulation to the Dominant at the 
end of the first half (see below, p. 1 34). 

Another, and in some respects even a better, instance is 
the Alman (ii. 375), which is strictly Binary in form, and has 
the latter half of the first section reproduced literally (except 
for an ornament of semiquavers in one bar) in the latter half 
of the second section. (The second section also begins on a 
chord of the " dominant" key, if one may use the term here ; 
but I do not lay any stress on this feature.) This work, 
however, as has already been mentioned, is altogether 
" Modal " in its harmonies, and is to be thought of, not as 
in a " key " of G, but as in the Mixolydian Mode. 

Similar coherence, in a probably older tune, may be observed 
in " Walsingham" (i. I, or i. 267). Short as it is, the 
second four bars depend wholly on the first four, as is easily 
seen. 



" WALSINGHAM " (early i6th Century, at latest). 

i. As ye came from the Ho - - - ly Land 

?====== 






a. How should I know your true love, 



120 AN ELIZABETHAN VIRGINAL BOOK 



Wai - - sing - - - ham, 



Met you not 



with 




Ho - 



Land, that have come, that have gone? 



(Words: Percy Folio, vol. iii. 471.) 

To speak in our modern way, this " Walsingham" has two 
bars in A minor, and then two bars in C major (the relative 
major of A minor). Here ends the first section. This 
sequence of keys was not improved upon in Beethoven three 
centuries later. The second section, as may be seen by mere 
inspection, is formed entirely on the model furnished by the 
first section, with the two interesting alterations (l) the 
line begins on B instead of C, thus indicating a harmony of 
E major (what we now call the "dominant" of A minor); 
and (2) the last two bars are altered so as to finish, not in C 
major, as the first line did, but in A major itself. Thus the 
key sequence, and the interdependence of the different 
portions of the tune, almost perfectly correspond with the 
practice of the ipth century. 

The "Duke of Brunswick's Alman" (ii. 146) should be 
compared with " Walsingham," partly because the curious 
change (curious to our ears) from minor to major, in the last 
phrase, occurs in both. This is not to be considered as a 
peculiarity of Bull, simply because both pieces have his name. 
Byrd has the same thing in his setting of " Walsingham," i. 
267 ; also G. Farnaby, ii. 19, line 3, last bar. But the " Duke 
of Brunswick's Alman" is more valuable, as another good 
example of interdependence between the sections of a Binary 
movement. So complete is it in this case, that the grd and 
4th bars of Section I. can almost be played together with the 



AN ELIZABETHAN VIRGINAL BOOK 121 

corresponding bars of Section II., in spite of the tune and bass 
differing considerably. 

The same thing can nearly be said of the 1st and 2nd bars 
in each section. At least, they are connected most intimately 
in their harmonic tendencies. 

These specimens are sufficient in number, and, I think, quite 
convincing. 

DEVELOPMENT. 

The next point is, that " symptoms" of the simpler notions 
of " Development," as used in modern writings, are to be seen 
in another anonymous Alman (i. 65), containing three sections 
(3 sections is the most common arrangement in these old 
dances), the grd of which introduces two phrases, the first of 
two " bars," which is used again in " sequence," a tone higher ; 
the second being a shorter phrase of one " bar," which is 
doubled on itself by ornamental repetition, and then repeated 
in sequence like the first, thus producing a section of ten 
"bars," a mighty contrast with the former two divisions of 
the whole movement, which have only eight " bars" (the most 
usual number). 

MELODY OF THE 3RD "STRAIN" OF AN ALMAN 
(FITZWILLIAM BOOK, i. 65). 

ANON. (i6th Century). 







122 AN ELIZABETHAN VIRGINAL BOOK 



Here are lo " bars" instead of the regular 8. 

Bars I and 2, founded on a chord of F major, are used in 
sequence to produce bars 3 and 4, founded on a chord of G 
minor. 

Bars 5 and 6 (harmony of C major) are similarly made to 
give forth bars 7 and 8 (harmony of D major) ; and thus the 
piece presents a somewhat advanced case of development of 
phrases, resulting in the lengthening of the whole division. 

The student will carefully note that the signature of one 
flat merely indicates that the Dorian Mode has suffered " trans- 
position," i.e. that the notes of that mode have been raised in 
pitch from D up to G, a perfect fourth, according to the practice 
of the time. The piece is not in what we call G minor, though 
we probably should consider its general character to be fairly 
indicated by that name. 

A nice instance of a similar "development" is in ii. 405, 
section 2, of Oystermayre's Galiard. 

SECOND "STRAIN" OF JEHAN OYSTERMAYRE'S GALIARDA 
(FITZWILLIAM BOOK, ii. 405). 

(i6th Century.) 
Slow. 

6 
2 







etc. 




AN ELIZABETHAN VIRGINAL BOOK 123 

The characteristic motive of this example, taken from the 
second strain, appears in bar 3 of the// strain ; moreover, 
another sequence is found in the third strain, the melody of 
which seems to be a sort of " development " of the melody in 
the first bars of the first strain. Thus the solitary specimen 
of Oystermayre's work which is given in the Fitzwilliam 
Virginal Book, may be regarded as a particularly remarkable 
example of modern tendencies (also see below, p. 139, and 
above, p. 22, where the whole piece is given). 

A final instance may be taken of William Tisdall's " Pavana 
Chromatica," where (ii. 280, lines 2 and 3) a phrase is re- 
peated in ascending sequence in three successive bars, and 
even a fourth bar is made by the same trick, in a less com- 
plete form. 

Thus we are able to point out slight but quite unmistakable 
evidences that English (and other, e.g., German) composers of 
the l6th century were striving, darkly but surely, after three 
main features of modern music : 

(1) Coherence, by means of "formal" repetition. 

(2) Development of melody by agremens, by altered rhythm, 

or by sequence. 

(3) The larger Coherence of a series of movements en suite, 

from which we derive Symphony, Sonata, Quartet, etc. 

THE NEW MATERIALS. 

It is plain that a considerable growth in the method of 
Instrumental Music might have taken place on the general 
lines which are indicated in these three main orders, quite 
apart from any startling change in the simple materials of 
music, the Melody, the Harmonies, or the scales to which 
these last are referred. For this reason I have noticed 
first the three features named above. 

The modes clearly did not prevent the practice of tune, 
chord, or design, in music. 



i2 4 AN ELIZABETHAN VIRGINAL BOOK 

But, with the close of the Tudor period, we notice that 
great changes are at hand in other matters besides those 
of melody, harmony, or form. 

The " ecclesiastical " influence was dying down, and the 
" secular " characteristics of the music of the people were on 
the eve of asserting themselves, in a more positive manner 
than had hitherto been possible to ages which reckoned the 
" secular" musician no better than a jack-pudding or common 
street tumbler, who varied his witticisms or acrobatisms with 
fiddling or piping of " Dances " and " Gigges." 

It was beginning to appear to composers in general that 
contrapuntal exercises were not inexhaustible fountains of 
inspiration. Thus they tried "variations" on well-known 
melodies. Later on they found those melodies to contain 
within themselves the possibilities of spontaneous growth, 
the most complete exemplification of which we now possess, 
three hundred years later, in (say) Brahms's Violin Sonatas. 

Now, what ivere the " secular characteristics of the music 
of the people," mentioned above ? 

We have already seen some of them in the simple cheerful- 
ness and artistic coherence of the anonymous Dance tunes 
which are to be found in the Fitzwilliam Virginal Book. 
They depend on rhythm and natural tunefulness, not upon 
harmony, whether attained " accidentally " by contrapuntal 
setting, or by design of bass and super-imposed chord. 
They are constructed on a plan that knows more about 
the Tabor - pipe than of the Schoolmen's Modes ; more 
about the modern Scale of eight notes than of the 
Hexachordum Durum; more of natural Accent than of 
the "Proportions" enumerated by GafFurius (1496). (The 
hopeless character of some of these "Proportions" is 
excellently demonstrated by Dr John Bull's experiments, 
e.g., i. 186, 1875 66 ; " 39> l ine 2 > bar 2, etc.) 

It is easy to show how little in common these " popular" 



AN ELIZABETHAN VIRGINAL BOOK 125 

melodies have with those of the Church. The following 
may be confidently referred to as specimens of no uncommon 
order, which will convince any one that it was not the nature 
of the words only which made the difference between sacred 
and secular music in the l6th century. 



LIST OF EXAMPLES of secular tunes in the Fitzwilliam Vir- 
ginal Book, which show modern tendencies of Scale or 
Key. 

1. " Jhon come kisse me now," set by Byrd, i. 47. 

2. " Nancie," set by Morley, i. 57 ; affected by Mode, 

however. 

3. Pavana, by Bull, i. 62 ; except the chord of F in bar 2. 

4. Muscadin, Anon. 1 . 

5- Alman, Anon. \ '' ?4' 75- Parts of these. 

6. Galliard, St Thomas Wake," Bull, i. 131. 

7. " The King's Hunt," set by Giles Farnaby, i. 196. 

8. Spagnioletta, set by Giles Farnaby, i. 199. 

9. " The Carman's Whistle," set by Byrd, i. 214. 

lo " Now God be with old Simeon," introduced by Byrd 
in his " Hunt's up," i. 223. 

11. " O Mistris Myne," set by Byrd, i. 258. 

(Compare the tune of " Woods so wilde," p. 263, and 
see how great the difference.) 

12. " Pawles Wharf," set by G. Farnaby, ii. 17 ; positively 

in the Key of D major. A good piece. 
(Compare " Pawles Wharf" with " Quodling's 
Delight," p. 19, and observe the contrast.) 

13. Alman, by Robert Johnson, ii. 159; in D major. 

14. " Nobody's Gigge," Richard Farnaby, ii. 162, 163 ; in 

C major. 

15. Alman, Thomas Morley, ii. 171 ; in key of C. Tonic, 

subdominant, and dominant quite clear. 



126 AN ELIZABETHAN VIRGINAL BOOK 

16. La Volta, Byrd, ii. 180, in G. (Compare with Alman, 

p. 182.) 

17. La Volta, Morley, arr. by Byrd, ii. 1 88. In G, with 

"modulation" to D. The "Form" very nearly 
Modern Binary. 

18. Lady Zouches Maske, G. Farnaby, ii. 350. 

19. " Tower Hill," G. Farnaby, ii. 371. 

20. The Earle of Oxford's Marche, Byrd, ii. 402. In G, 

D, and G. 

21. The Old Spagnoletta, G. Farnaby, ii. 471; in "G 

minor," with "relative" major, "dominant" 

major, and return to " G minor." 

N.B. With the exception of these twenty or so pieces, 
all of which are "popular" in character, the Fitzwilliam 
Virginal Book simply shows that the Modes reigned very 
completely in the minds of musicians as late at least as the 
year 1600. 

We can now add something to our former proposal, and 
may, with some confidence, hold an opinion that the " Music 
of the People," in the l6th century, as shown in the Fitz- 
william Book, was in favour of 

(1) The adoption of the Major and Minor SCALES, instead 

of the Hexachords. 

(2) The adoption of KEYS, instead of the Modes. 

(3) The use of the characteristic CHORDS of the Key, as we 

now understand them, namely the TONIC, SUBDOMI- 
NANT, and DOMINANT. 

(4) The composition of melodies in BINARY FORM, with 

internal unity of the keys and of the constituent 
sentences, corresponding very nearly with the general 
features of a Beethoven Sonata movement two cen- 
turies later. 

But, as with other great changes, the transition to modern 
ways was effected gently, and without any alarming explosion 



AN ELIZABETHAN VIRGINAL BOOK 127 

on the part of the talkers. This kingdom also came " not 
with observation." 

Even now, in spite of all that has been written since the 
time of Henry VIII., the Modes are still with us. Every 
time Tallis's " Festal Responses " are sung in church, we 
hear how impossible it is to put entirely away from us this 
ancient practice of harmony. 

In our own time alone, leaving out all mention of the cen- 
turies between, Dvorak and Verdi, Wagner and others, 
cannot, under certain circumstances, refrain from using 
chords in series, which owe their grandeur even more to the 
continuous tradition of the familiar things of the i5th and 
l6th centuries, than to their intrinsic worth, high though that 
may be. 

" Parsifal " alone is enough to prove this point, to any 
student who knows that work as it should be known. 

" Pieces Justificative*" 

It will not be improper here to consider, with some care, 
certain examples in the Fitzwilliam Book which may be 
regarded as the "pieces justificatives" of my case. 

Bull's very remarkable experiment on the Hexachord (i. 
183) supplies us with some examples of " modulation " which 
are quite outside the regular practice of his time, and which 
(as we have seen) can be regarded as successful in a some- 
what modified degree. Nevertheless, as already has been 
pointed out, Bull's general characteristics make it right to 
suspect (not his bond fides, but) a certain measure of insincerity 
in the production of this work. If Bull's work stood alone, 
it could hardly be accepted as a proof that "enharmonic 
modulation " and " equal temperament " formed part of the 
direct heritage of the iyth century. But it does not stand 
alone. 

William Byrd's exercise on the same theme (Hexachord) 



128 AN ELIZABETHAN VIRGINAL BOOK 

is thoroughly characteristic of the composer, and furnishes a 
much-needed foil to the work of Bull. Byrd is always a 
musician first, and a technician afterwards if he feels inclined. 

The signs of his utter unwillingness to be ugly because his 
first plan leads logically to ugliness, are plain all through his 
works, whether in the Fitzwilliam Book or elsewhere : i. 397, 
line 3 onwards, shows us the way from G major to G 
minor and C minor, but in the spirit of the true composer, 
not of the teacher of Harmony. 

In spite of the innocent ease of these " transitions " of 
Byrd's, they are, to my mind, more really to be regarded as 
evidence of Byrd's growing feeling for " Key" as opposed to 
" Mode," than Bull's more complete and startling efforts. 

Bull, in fact, behaves like his namesake in the china shop, 
and sends everything flying that he can get his horns any- 
where near. As I have said before, he is a "ruthless" 
artist. 

Byrd, on the other hand, has the gentleness which makes 
for beauty ; and he cares for beauty equally with progress, 
though he was by no means a Church and State Tory, but, on 
the contrary, a Roman Catholic Nonconformist who gave 
some trouble to the authorities in his time. 

Clearly, the evidence of the growth of the new methods is 
greatly strengthened when two men of such widely con- 
trasted tendencies show themselves at work in the same 
direction. 

Nor are these two giant explorers alone in their newly 
discovered country. 

William Tisdall may perhaps have doubt cast upon his use 
as a witness to the facts. The Pavana Chromatica, ii. 278, 
is very full of sharps and of semitonal progressions (line 2), 
but, the student might object, it is Modal after all in char- 
acter. So it is ; and what else could be expected ! Take 
the reverse case. Set a man, born in the Ipth century, to 



AN ELIZABETHAN VIRGINAL BOOK 129 

imitate the Modal progressions of the 1 6th century. In 
spite of an immense knowledge of all the particulars of the 
difference between the one style and the other, every line 
of his production betrays him. Every line speaks loud of 
the later period, and the work does not even deceive the 
expert of to-day, still less could it have deceived those of 
the earlier time, in whose atmosphere it vainly seeks to 
breathe again. 

Well may we find that Tisdall and his contemporaries 
of the Elizabethan time, with no complete prophetic know- 
ledge of what was to come, indeed with only a dim shadowy 
notion that anything new could come, do as a rule fail in 
their efforts to leave behind them the old practices which 
they seem to have felt were becoming insufficient for their 
needs. 

But TisdalFs Chromatic Pavan is of great value for our 
purpose, if we care to remember that parts of it (e.g., section 
l) require as many as four sharps (i.e. those of the key of 
E major) ; that it begins and ends on a chord of B major, 
and constantly uses that chord, so unusual to the age in 
which this piece was written j and that on the other hand, 
the second section is entirely in the Mixolydian mode, re- 
quiring only one sharp note. We cannot suppose that Tisdall 
was otherwise than gratified at his work, or that he was 
less than pleased to find that the harmony of B major was 
so nearly related to that of G. 

The same piece, p. 280, lines 2 and 3, affords a good 
"sequence" of four bars, in which the harmonies are suc- 
cessively G major, E minor, D major, B major. 

In a Pavana, ii. 307, Tisdall appears at first to be aiming 
at the scale of G minor ; while on p. 308 there is manifestly 
a great struggle towards something which is scarcely well 
enough defined to receive the name of " Modulation." 

The passage from Sweelinck's " Ut Re Mi Fa Sol La " 

I 



1 30 AN ELIZABETHAN VIRGINAL BOOK 



(the date, 1612,* may be the year when the piece was copied 
by the writer of the Fitzwilliam Virginal Book), vol. ii. p. 3 1, 
where there is apparently a " modulation " from the region 
of F major to A minor and to E flat, is, I fear, merely a 
formal consequence of the subject on which the piece is 
written. 

A similar remark applies to Giles Farnaby's apparent pro- 
gress through the keys of G D A E B, and back to C, 
in ii. 270, bottom line to top of next page. This is only 
the result of writing Ut, Re, Mi, Fa four times over in scale 
form, and then adding upper parts to it (p. 270, last bar, 

* It would appear, however, that the dates given to certain of the works 
in the MS. are the dates of their composition. 

The following short list of such dates seems to make this plain : 

Date 
given. 

1590 (the first 
date in the book) 
1591 
1602 



Reference in the printed 
copy of the MS. 

i. 266 



Name of Composer 
(or arranger). 

Byrd 



35 
. 320 
.3*6 

33 1 

334 
345 

35 
.356 

436 

ii. ii 
ii. 33 



1603 
1605 
1580 

15^5 

1582 

1561 (the earliest 

date in the book) 

1564 

1612 



Philips 

Philips 

Philips 

Philips 

Philips 

"Philips' first" 

Philips 

Philips 

Tallis 

Tallis 
Sweelinck 



Other more or less "datable" pieces are 

ii. 273, a " Maske " ascribed to G. Farnaby, which appears to have been 
a dance in a Gray's Inn Masque in 1613 (viz., The Maske of Flower) 
bearing the name of Coperario. See above, p. 7. 

ii. 330, G. Farnaby's arrangement of his own "Canzonet," "Ay me poor 
heart," date 1598. 

i. 131, Bull's Galliard "at Thomas Wake," which was published (with it 
Pavan) in 161 1 (Parthenia), and 

ii. 479, Gibbons' Pavana (jEolian), also in Parthenia, 1611. 



AN ELIZABETHAN VIRGINAL BOOK 131 

tenor, affords another instance of the uncertain notation of 
such passages : E|> for D$). 

On the other hand, the anonymous Alman, ii. 312, is 
written (at least in the first " strain") definitely in the " Key" 
of E major. (The author, of course, knew of no such term, 
but he clearly knew the thing signified by the term.) It 
is useful to contrast the 2nd and grd strains with the 1st 
in this case. The little piece is so interesting that I give 
it complete. 

ALMAN (FITZWILLIAM VIRGINAL BOOK, ii. 312). 

[Piece No. 22y.J 




* The alto part is slightly corrected here, and the uniformity of the synco- 
pated imitation preserved. 



1 32 AN ELIZABETHAN VIRGINAL BOOK 









g 



r 



l*=|^33E5EEf!* 



^g? ft*-f 



i y^F j s^ p f-i 







I ) 












1st 



r 



In this one short piece the student may see illustrated 
several of the characteristic features of the clavier music of 
the 1 6th century, e.g. 

(a) The notation. Although the four sharps of the " key " 



AN ELIZABETHAN VIRGINAL BOOK 133 

of E are wanted everywhere, they are written throughout 
as " accidentals." 

(b) The straightforward construction of the work, in 
three strains, each consisting of eight semibreves. 

(c) The irregular use of bars, some containing 4 crotchets, 
others 8 crotchets. 

(d) The fashion of playing the chord of the "key" at 
the end of the piece, reminding one of the Wiltshire (and 
other) countrymen who invariably announce the title of the 
song they have just sung, before sitting down. 

(e) Most important of all, the mixture of "modal" har- 
mony and " modal " tune with " scale " harmony and " scale " 
tune, e.g., the first strain, entirely in the modern " key " of 
E major; the second strain altogether "modal," whether 
the chords or the melody itself are considered ; the third 
strain, where (in the last line) the immense difficulty of 
persuading the music to return to the "key" of E major 
is very strangely shown. 

(/) The extreme compass of the Elizabethan virginals (as 
given in the Fitzwilliam Book) is found in this piece, viz., 
4 clear octaves, from A under the bass staff, to a over 
the treble staff. 

After contrasting the 2nd and 3rd strains of the above 
Alman with the 1st strain, and noticing the essential differ- 
ences, the reader should play through the Alman (anonymous), 
ii. 375 (see below, p. 138), which is "modal" in tune and 
harmony alike, and contrast it also with " Pawles Wharfe," 
ii. 17, by Giles Farnaby. The latter is as certainly in the 
"key" of D major as can possibly be: the Tonic, Sub- 
dominant, and Dominant characteristics are plainly present, 
and are used in order, just as we have been accustomed 
to them now for many years back. 

Equally certain is it, that " Quodling's Delight" (G. 
Farnaby again), ii. 19, is not in the modern "Key" of A 



134 AN ELIZABETHAN VIRGINAL BOOK 

minor. As will be perceived, it is founded on the notes 
la, sol, la, mi, which are characteristic of many early " modal" 
tunes. 

It is of the greatest consequence to the student that he 
should have a clear idea of the distance between these two 
methods of harmony ; therefore I here give these two tunes, 
shorn of repetitions. 

"PAWLES WHARFE" (FITZWILLIAM BOOK, ii. 17). 

[_\6th century example of the modern "key" of D major, with itj 
essential harmonies of Tonic, Dominant, Sub- dominant.^ 



GILES FARNABY. 



Not too quick. 

-%=&=: 




c 



-I g- 



?^ : 



r 







AN ELIZABETHAN VIRGINAL BOOK 135 







r 

Here in section I we have a tune made of the plain " scale " 
of D major, accompanied by " tonic" and " dominant " chords 
in such order as modern ears have come to expect. In 
section 2, bar I, we find the " subdominant " chord (of G) 
introduced in such a manner as to emphasise it as a 
characteristic of the " key." 

On the other hand, " Quodling's Delight," set by the 
same composer, is definitely wanting in certain features of 
our modern " minor scale," and is definitely not wanting in 
the characteristics of "modal" harmony, e.g., the basis La, 
Sol, La, Mi, which constantly appears in ^Eolian melodies. 

" QUODLING'S DELIGHT" (FITZWILLIAM BOOK, ii. 19). 

[Showing " Modal' 1 characteristics.'] 

GILES FARNABY. 

izd 




136 AN ELIZABETHAN VIRGINAL BOOK 




r 

\ 

The "modal" basis is well seen in section I, marked by 
capitals under the music. Had this piece been as truly in 
the modern "minor scale" of A, as the former ("Pawles 
Wharfe") is truly in the modern "major scale" of D, we 
should have perhaps found A E F E in place of A G A E. 
The general resemblance of the ^Eolian mode to the modern 
minor scale is no doubt striking ; how otherwise ! for in 
it we have the lineal progenitor of all modern minor scales ; 
but in particulars the difference is very plain. The 
" essential " nature of the G natural in bar I should be 
compared with the "accidental" nature of the G sharp in 
bar 2. The one is of " Mode," the other of " Harmony." 
But in the "key" of A minor, both would be alike. Thus 
the distinction is clear, even to modern ears. 

This piece is almost the same as " Fayne would I wedd," 
ascribed to Richard Farnaby, Giles's son (ii. 263). 

Robert Johnson's Alman, ii. 159, a beautiful and 
graceful dance, furnishes an excellent instance of the strength 
of Tonic and Dominant in the "key" of D major; and 
Richard Farnaby's "Nobody's Gigge," ii. 162, 163, is quite 
as clearly in the key of C, with the Tonic and Dominant 
harmony plainly recognised. 

Here is Johnson's piece : 



AN ELIZABETHAN VIRGINAL BOOK 137 

ALMAN (FITZWILLIAM BOOK, ii. 159). 



Not too fast 



ROBERT JOHNSON. 




^^pLj^ \ 






r r 

j L 






r 




138 AN ELIZABETHAN VIRGINAL BOOK 

The final chord is a formality, as has already been 
observed. The probable alto part between * and * has 
been restored, and an obvious mistake in the tenor part 
of the final bar of the first strain corrected. 

We remark in this Alman, (a) the "key" of D major, 
with scarcely a hint of the smallest kind that Robert Johnson 
lived amongst the Modes, (b) The chords of the Tonic, 
Subdominant, and Dominant throughout, used as they were 
used a century later, e.g., bar I, where all three occur in 
an order very familiar to modern hearers, (c) The modern 
touch of " Form," at the beginning of the second " strain," 
where a " free" section is actually modelled on the opening 
bars of the first "strain," just as it would be in a ipth 
century sonata movement. 

The student may now turn back to the anonymous Alman 
(above, p. 131), and compare the 2nd and 3rd divisions of it 
with this of Johnson. A good experimental way of realising 
the abyss which lies between " Mode " and " Key " would be, 
to play over the first strain of Johnson's Alman, and then, 
while its general character is fresh in the memory, play the third 
strain of the anonymous Alman (above, p. 132), which begins 
on a chord (A major) closely related with the key of John- 
son's piece, but leads to a very different sort of conclusion. 

If this is not sufficient, the reader may play again the first 
strain of Johnson, and contrast with it the following two bars 
of another Alman (ii. 375), which has already been named more 
than once as having the Modal character strongly marked, 
ALMAN (SHOWING "MODAL" HARMONY). 

ANON. fi6th century). 
# 




AN ELIZABETHAN VIRGINAL BOOK 139 

The difference between this " Modal " harmonisation and 
the harmony of modern " key " is seen as follows : 

THE SAME MELODY (SHOWING " KEY" HARMONY 
OF G MAJOR). 
[With Signature.'} 
Ni * 41 






*ZI 



^M^L~^^= J j J 



etc. 



One chord only (marked *) is altered, but that is more 
than enough. 

One of the most convincing specimens, however, is the 
Galiarda of Jehan Oystermayre, ii. 405, in which the Tonic, 
Subdominant, and Dominant are well marked in the key 
of G major; and (in the second "strain"), a sequential 
passage (quoted above, p. 122) leads the music naturally 
and pleasantly through D major, A major, B minor, D major, 
E, and back to D. Moreover, the general key effect in 
the three sections is, as in a modern work it would, or might 
easily, be, viz., G j D ; G . (See p. 22, where the 
whole piece is given.) 

It is a pity more of this composer's works did not reach 
Mr Tregian. 

A final example is that of Thomas Warrock (organist of 
Hereford, 1586), whose Pavana, i. 384, and Galiarda, i. 388, 
besides showing smoothness and grace beyond most of the 
music of the time, gives us some hope that the composer 
believed in the scale and key of B flat, and (in the Galiarda) 
even in the possibility of certain Minor Chords in the key 
of C major (p. 389, top line). 



1 40 AN ELIZABETHAN VIRGINAL BOOK 

It is indeed possible to say that Thomas Warrock's 
signature of two flats is merely an indication of " Double 
Transposition " ; but to call a piece Ionian does not do away 
with its general characteristics, and these seem to be decidedly 
separate from Mode, and equally akin to the idea of Key. 

We have now seen practically all that can be found in this 
large collection of nearly 300 pieces, in the way of evidence 
for the early growth of modern methods of composition in 
the later 1 6th century. 

It must not be pressed too far, but I venture to think that 
it will not be possible to reject most of it. 

When all deductions are made, it seems clear to me that 
the idea of major and minor scale, of key, and of the charac- 
teristic chords of a key, ivas gaining recognition in England 
and the Netherlands about the time of Elizabeth, and that the 
germs of modern composition, in all essentials, were to be 
seen developing in the practice of that time. 



CHAPTER XII 

EARLY WESTERN HARMONY 

IT will now be interesting to look backwards from the Fitz- 
william Virginal Book, and to study some pieces in it which 
seem to have a relationship with the earliest Western Har- 
mony, i.e. which appear to contain relics of the Origins of 
Harmony as it was already practised in the 1 2th century in 
England. 

Some references to passages of this nature are : i. 54 5 
i. 263 ; i. 42 ; i. 47 ; ii. 128 ; ii. 387 ; ii. 422. 

Before looking at these the reader's attention must be 
directed to a peculiarity of many of the most ancient popular 
English melodies (which are often either Dorian or ^olian), 
namely, that they are founded on a simple " burden " 
(bourdon, drone," buzzing bass"), consisting of two alternating 
notes, adjacent to each other. 

Sometimes these are what we should call the first and 
second notes of the scale, as in the case of " Sumer is icumen 
in" (l3th century), which has this characteristic, and any 
person may join in it by merely singing F G, F G, F G, etc., 
throughout as a sort of Ground Bass. 

CANON ON " BURDEN" OF I3TH CENTURY. 

Sum - - - er is i - - cum - - <n 

' J I J 

TUNE. I/, t\ Qiri: 



BURDEN. ::z^: r? 



a 



1 42 AN ELIZABETHAN VIRGINAL BOOK 

in Lhu . . . de sing cue - - cu. 

* 



In the actual work itself, which requires 6 voices (4 to the 
tune in Canon, and 2 to the " Pes " or "Pedal part"), the 
" burden " is given in a slightly more elaborate shape, the 
F G, F G, appearing first in one of the two " Pedal " voices 
and then in the other. But the gist of the thing is accurately 
represented in the above example, which shows that the 
writer of the 1 3th century had two chords in his mind. Here 
it is by no means trivial or without occasion to remind the 
reader that the common street accordion preserves this 
ancient feature, and that the players of this barbarous instru- 
ment do still try to fit the tunes they know on to the two 
chords which are at their disposal, according as the bellows 
are blown inwards or outwards. 

The same remarks apply to a very old catch, " Heigh ho, 
nobody at home," where it is only necessary to alternate 
A G, A G (supposing it to be ^Eolian), in order to " take 
part " effectually. 

CATCH ON " BURDEN," OF EARLY DATE. 

(Date probably early i6th century.) 



no - - body at home! Meat nor drink nor 




mon-ey have I none: Yet will I be mer - - ry! 



Here are 5 bars, and 5 voices may take part in the Canon 
of this catch, each beginning a bar after the other. The 
result, in harmony, is simply the alternation of two chords 



AN ELIZABETHAN VIRGINAL BOOK 143 

IgJZ^ZCg^ ^f3l 




and, as will be seen, the two notes, A G, A G, can be 
continually alternated throughout. 

Another case of a Northern tune, which has this charac- 
teristic, may not be so well known, though many country 
people still sing it in Yorkshire and Cumberland to a set of 
humorous words, beginning " When a liv'd at yam wi' ma 
fadher an' mudher," etc. The tune is ^olian in its tonality. 
A sufficient specimen is the chorus (transposed to a lower 
pitch for convenience, so that the burden A G A becomes 
D CD). 

ANCIENT NORTH ENGLISH SONG ON " BURDEN" 
(Cumberland and elsewhere). 






Singing Reyt fol-lood - le lay. Reyt fol-lood - le lad-di-ty Aw 1 

Burden D C 




Reyt fol-lood - le lay, Singing Whack follood - le lad-di - ty Aw ! 

D D 

Those who are not used to dialect may be glad to know 
that " Reyt" is the correct Northern for " Right," and that 
" Aw" means " Oh." 

The burden D C D, etc., is marked in capitals under the 
music, and it is extremely probable, if this melody ever was 
sung with an instrumental or vocal accompaniment, that this 
" burden " would have been regarded as the correct thing. 

It is well to notice that alternative drones of this nature 



144 AN ELIZABETHAN VIRGINAL BOOK 

were found in the Northumbrian bagpipe, which possessed 
an arrangement for changing the note of the drone or drones j 
and that something of the same sort is still heard in the 
Italian bagpipe performance with two players, one of whom 
plays the tune on a " chanter," or rough kind of oboe, the 
other accompanying him on a larger instrument which 
supplies a limited pedal bass. 

A few remarks on the possible connection of the ancient 
melody and harmony of the bagpipe, with the passages in 
the Fitzwilliam Virginal Book referred to above, will not be 
altogether out of place. 

My suggestion is, that these ancient " burdens " of two 
alternating notes lie at the very root of the mediaeval notion 
of Harmony (apart from the harmonies produced by Counter- 
point, or the combining of melodies), and therefore that the 
student must regard any possible evidence as to their develop- 
ment with careful interest. 

The first example is by Bull, the Galliard to my Lord 
Lumley's Pavan, i. 54. The tune of the first strain is 
remarkably like a real bagpipe tune in every 'way,* and a 
double drone-bass of the kind already described is indicated 
in the harmony given by Bull. 

BULL (1563-1628.) 




* A curious comparison may be drawn between this tune and the second 
section of the castanet song in Bizet's "Carmen," which is strangely like 
Bull's melody, and stands on an alternate drone bass of F G F G. See 
'* Carmen," No. 17, the eight bars preceding the entry of the trumpets. 



AN ELIZABETHAN VIRGINAL BOOK 145 




A G 

The most likely "burden" seems to be G and A, as 
marked. Another case is " Goe from my window " (i. 42) by 
Morley (or Munday), where the harmony is alternately based 
on G and A throughout. 

Byrd gives us an instance in his " John come kiss me now" 
(i. 47), where the burden or "bussing bass" of G F G D, 
would be quite in character. Compare the burden of La, 
Sol, La, Mi (A G A E), which underlies the catch " Heigh 
ho, nobody at home," quoted above (p. 142). 

The same composer harmonises " The Woods so Wild" 
(words unknown, see Chappell, i. 119) on practically two 
chords, those of F and G, thus 

"THE WOODS so WILD" (FITZWILLIAM BOOK, i. 263). 

[Showing drone bass or burden.~^ 
Slo-w WILLIAM BYRD (1538-1623). 




146 AN ELIZABETHAN VIRGINAL BOOK 




Here we have F and G in alternate bars. But the student 
will carefully note that it is not the same case as that of 
"Sumer is icumen in" (p. 141 above), which is written in 
the key of the "softened B," or bemol, alluded to by 
Giraldus Cambrensis as being characteristic of the part sing- 
ing in Northern England in his time (end of the 1 2th 
century). 

Dr John Bull's " Juell" (ii. 128) is founded on the same 
bass as " Heigh ho," etc., only beginning on C, and using 
another order, viz., instead of C, B flat, C, G, it has C, G, 
B flat, C, etc., according to the demands of the tune. The 
second strain is even more convincing than the first. 

DR BULL'S " JUELL " (SHOWING " BURDEN ") 

(FlTZWILLIAM BOOK, ii. 128). 
Fit*t strain JOHN BULL (1563-1618). 




AN ELIZABETHAN VIRGINAL BOOK 147 



Second strain 




^|3 I J-*= 
f ' *,? 



.^. The minim note "g" in the left hand part of bar 4 
in the second strain is probably an error for " b." Compare 
the variation which follows on p. 129, where the correspond- 
ing bar is founded entirely on B[>. 

Byrd's Galiarda (ii. 387) has the word "Drone" as it 
were written on it in large letters. The first strain has, 
in common with other pieces already mentioned, the " burden" 
La, Sol, La, Mi (A G A E), as its harmonic basis. 

Martin Peerson's "The Primerose " (ii. 422) is worth 
mentioning on account of the alterations of the chord of C 
with that of Bb, and back again, which may be seen in line 4 
and onward. 

A final instance is found in Peter Philips, the English 
Roman Catholic priest, who supplies us with a Passamezzo 
Galliard, en suite with the Passamezzo Pavan preceding it 
(dated 1592). The reference to the Galliard is i. 306, and 
the student will observe that the basis of the first four bars 
is G, F, G, D, i.e. once again, La, Sol, La, Mi. 



1 48 AN ELIZABETHAN VIRGINAL BOOK 

These ten examples of " popular " harmony seem to point 
in one direction, and to indicate the source of chord music, 
by which I mean harmony which was not the result of 
"contrapuntal" combination. 

In all likelihood they fairly represent the four centuries 
from John to James I. in this respect, and they certainly en- 
courage the student in a belief that the modern notion of a 
"Bass," implying harmonies and implicit melody, is directly 
traceable to the drones of the bagpipe family. 



PARTHENIA 



or 



THE MAYDENHE^JD 

of the first musicke tnat 




^ 

fLLjr v c. 



- ^ 



CHAPTER XIII 

THE ELIZABETHAN VIRGINALS 

THE clavier used by the Composers of the pieces in the 
Fitzwilliam Book was known as " the virginals." For the 
plural form of the word one may compare such cases as 
"scissors," or "bands" (the ecclesiastical neck- wear). It 
was common also to speak of " a pair of virginals," for 
which, again, we compare "a pair of scissors," "a pair of 
bands." The name itself seems to point to the domestic use 
of the instrument. It was in fact the ladies' clavier, and so 
got its graceful title. Shakespeare's "Winter's Tale" (i. 2, 
1. 125) gives us an instance of the name used as a verb, 
where Leontes the King of Sicilia speaks of Hermione and 
Polixenes 

" still virginalling 
upon his palm ? " 

Thus, in Shakespeare's mind, the word " virginals " clearly 
pictured the action of musical fingers on a keyed instrument. 

The " virginals " of the latter half of the l6th century 
had much in common with the harpsichord of later times. 
That is to say, it had a keyboard similar to that of our 
modern pianoforte, organ, or harmonium, but of short com- 
pass, viz., about four octaves at the most : it had wire 
strings, similar to those of the modern pianoforte, but very 
much thinner, shorter, and slacker. 

The harpsichord, on the other hand, was made (as its 
name implies) in the shape of a harp (i.e. its ground plan 



150 AN ELIZABETHAN VIRGINAL BOOK 

was much the same as that of the modern grand pianoforte) ; 
whereas the " virginals " had a flat rectangular case like 
that of the old-fashioned " square " piano. 1 

The grand difference between harpsichord and pianoforte 
was, that whereas the action of the latter consists in the 
striking of the strings with a hammer, the action of the harpsi- 
chord consisted in plucking the string with a plectrum (or 
" spike " of quill, metal, or other suitable material). 

The action of the l6th century virginals was essentially 
the same as that of the 1 8th century harpsichord, that is, 
each string was "plucked" by a horizontal plectrum carried 
on a "jack" (a vertical slip of wood, moved upwards by the 
far end of the key lever, and called "jack " because it spent 
its life in "jigging" up and down in its groove). Thus the 
quality of tone was not unlike that produced by the plectrum 
of the mandoline on a single string. Also (more important 
still), the virginals could make no difference whatever 
between loud and soft, for the action could do no less and 
would do no more than simply " pluck " the strings ; thus, 
however hard the blow administered by the player, no 
" dynamic " expression whatever could be got out of the 
instrument, which merely continued to produce an average 
sort of mezzo-forte tone, louder or softer according to its 
individual nature, but never altering. 

We now proceed to gather together the information 
respecting the " virginals " which can be found in the 
Fitzwilliam Book, by a careful examination of the pieces 
given in that MS. 

In the first place, there was a variety of compass in these 
old claviers of the Tudor times. (This was also the case 
with the older pianofortes, e.g., a little square piano by Zumpe 
of London, dated 1766, has five octaves, less by one note, 

1 The virginals was described first by Virdung (Basel, 1511), who gives a 
picture showing the rectangular form of the instrument. 



AN ELIZABETHAN VIRGINAL BOOK 151 



beginning at G below the bass staff, whereas another old 
square by Wilson of Scarborough, dated 1791, has the full 
five octaves from FFF to f 3 . Even in our own time exten- 
sions of compass have been introduced.) 

Peter Philips appears to have had an instrument with a 





somewhat meagre-looking keyboard, viz., having 

A 



as its lowest key, and 



f r i* s topmost note. 



But Philips had other resources which do not meet the eye 
at the first inspection. His instrument was provided with 
the curious arrangement known as the " Short Octave," by 
which, although the keyboard came to an apparent end at 

fn\ Tj 

low E jSj . ..i =|, the complete diatonic scale of C was 

available in the lowest octave. The idea was, that low 
F| and G| were seldom found, and therefore that their 
mechanism might as well be used for other notes. 
The arrangement is best shown as follows : 



" SHORT OCTAVE." 



Notes actually 
played. 






Notes actually 
heard. 



RESULT IN PLAYING. 






152 AN ELIZABETHAN VIRGINAL BOOK 

Thus we find that Peter Philips's "virginals" had a 
compass extending over three octaves and a sixth, i.e. from 
C under the bass staff to A above the treble staff, but alto- 
gether omitting the " black notes " Cf Df F| and Gf in the 
lowest octave. Hence the term " short" octave. 



SHORT OCTAVES. 

Such an arrangement of the notes of the lowest octave 
was/ quite common in church organs from 1500 to 1800 or 
later. There are even persons still living who remember the 
" short octave " as an actual fact of ecclesiastical musicianship. 
It was managed somewhat differently in more extended key- 
boards, e.g., if the keyboard really had the visible semitones 



down to C JS? "7=j|, the organist was likely to find that 




the apparent C sharp would produce 

much more useful note in the days of unequal temperament, 
and so on. 

Peter Philips's pieces, Pavana Doloroso and Galiarda 
Doloroso (en suite), dated 1593, te ^ us Vei 7 Painty of tne 
short octave on his clavier. He ingeniously makes use of the 
peculiarity (which was quite common) by bringing in " long " 
chords, such as cannot be played by one hand nowadays, but 
which are perfectly easy under the " short-octave" system. 

Instances are, i. 325, top line, 326, line 4, etc., 328, 329, 
all in the left-hand part, where Philips uses a whole series of 
chords including tenths, some of which at least are impossible 
now, unless played arpeggiando. 

An example from the end of the Galiarda Doloroso 
(p. 329) will make the method quite clear. 



AN ELIZABETHAN VIRGINAL BOOK 153 
TENTHS MADE EASY BY " SHORT OCTAVE." 

PETER PHILIPS (dated 1593). 



?fi ^ 


1- 


g! 


M 


! 


^ 




-*= 


g 


>-v i 

* 


= *1 


F f r 

{ 1 


r r 

IL 1 J Jfr 


~r 

i j 


r-r 

i ! 


r 


r 

1 


1 


r 


i 
I 


@:_d 




p "=fr 


*. 


J 

- 







-J- 


j 


,^J 


^ 



f 



r f 



Fr 

| 


r-r 


r 

j j i 


-t 


=*= 


P 

1 1 


S 


^__ 




g g 1 



r r f 

Here the left-hand part is rendered possible for the 
smallest hand by the tuning described above, for all the 
tenths simply become ordinary octave stretches, the player 
merely following his thumb with his little ringer. (The 
tenth marked * would be less still, viz., Fj F, a stretch of 
a major seventh.) 

Philips uses the "short octave" elsewhere (i. 287, an 
arrangement of a madrigal) to make the left hand play a big 



chord of 4 notes, G 



which of course is 



quite feasible, for the low GJf key (see the explanation above, 
p. 151) has the low E on it, so all the player has to do is to 

.(wr yo 
strike ^^ *gz=~. 

IT 

It is obvious that Philips was much struck with the value 
of this common peculiarity of the virginals of his time, indeed 



154 AN ELIZABETHAN VIRGINAL BOOK 

he uses it so frequently in the Pavan and Galliard above named 
as to lead one to believe that he wrote these two movements 
with the special purpose of showing how these large chords 
or intervals could be made available. 

On the other hand, Bull and Tomkins appear to have pos- 



sessed virginals of four clear octaves from 



to 



i ^ with or without a short octave at the low 



end. I cannot clearly decide on the last point. The highest 
note in the whole book (unless I have missed some very rare 
case) is the treble a named above. 

References to the use of the A beneath the bass staff 
are: Bull, i. 177, 179; ii. 22. Tomkins, ii. 56, 96, 102. 



Bull, in ii. 22, has this chord -- gdzzzi, which 



(with the other passages mentioned) makes it likely that he 
had all the semitones down to low A, on their proper keys. 

ii. 56 shows on lines 2 and 6 that Tomkins had at least 
these notes in the lowest octave of his keyboard : 

LOWEST OCTAVE. 

TOMKINS (Elizabethan). 






-^^ -+ ^ 

An anonymous writer (ii. 312), in a short Alman (already 
quoted, p. 131), uses the following: 

ANONYMOUS " SHORT OCTAVE." 

(Elizabethan or earlier.) 



AN ELIZABETHAN VIRGINAL BOOK 155 

which would be possible on a keyboard with a short octave 
ending with the C below the bass staff, having the low A 
tuned on to the CJf key, a common arrangement on the organs 
of the 1 8th century and earlier. 

These last two writers may therefore have used instru- 
ments like " Queen Elizabeth's Virginal" (South Kensington 
Museum),* the keyboard of which possesses 5 keys below the 
E 



and would most likely be tuned thus : 
NAMES OF THE KEYS. 






NOTES PRODUCED IN PLAYING 




The top note of this clavier is high C, 



Here must be mentioned a curious difficulty which arises in 
considering Philips's pieces in connection with the " short 
octave," which his instrument certainly had. Plainly, it is 
this, that whereas the arrangement described on p. 151 pro- 
vides only C D E F G A as the lowest notes which could 
really be sounded by the player, nevertheless Philips uses the 



low Gtt 4>! ^~ JL more than once in the Pavana Doloroso, 
viz., i. 322 and 323, bottom lines. The two cases are 

* See Frontispiece. 



156 AN ELIZABETHAN VIRGINAL BOOK 

properly only one, as the second is merely a repetition or 
variation of the first. Still it is a difficulty, and requires 
explanation. 

The passage is as follows, in the simpler form of i. 322 : 

PETER PHILIPS (1593). 



m 



3 



^rrr^rr 



r 



-- 




J- 



etc. 



The GJf in the bass, as before explained, had no place in 
the " short octave " scheme, upon which the rest of this piece 
is ordered. The first suggestion would naturally be that the 
sharp is simply a mistake. This is not impossible, for the 
passage is quite smooth and right without the sharp, as may 
be seen best by playing it over so, as far as the asterisk. 

But this explanation is untenable (a) because the G| is 
well defined two lines above, in a similar passage, and () 
because the MS. itself is perfectly clear about the sharp. 

Another explanation might be found in the rather remark- 
able fact that there are 19 pieces by Philips in the Fitzwilliam 

Virginal Book, and that the note jSj _ z=i is found 

\ & 

nowhere in the whole 19, except in this one place. May we 
suppose, then, that Philips wrote the passage, including this 
apparently impossible note, knowing quite well that it was 
impossible, just as Beethoven or Schubert did in the case of 
their contrabasso parts ? 

But a simpler explanation is still at our hand, which may 
solve the difficulty. It is quite likely that Philips's instru- 
ment had a " split " key on Gf , that is, a key made in halves, 



AN ELIZABETHAN VIRGINAL BOOK 157 

one part of which would play the low E (as already ex- 
plained), and the other G*. Double keys of this kind were 
fairly common in the old " short octave " claviers. Mersenne 
(1636) apparently speaks of a divided key below the C under 
the bass staff, the two halves of which sounded GG and 
FFF, the AA and BB being on CC# and DD|, so that a 
complete scale of "white" notes was available down to 
FFF 



i 
*5& 

Additional evidence as to the practice of the middle 
iyth century is furnished by Mr A. J. Hipkins (Grove, IV. 
305^, note), who speaks of split keys in short octaves, with 
date 1664. 

Thus it is very probable that Peter Philips wrote his 
Pavana Doloroso (1593) ^ or an instrument which had its low 
G| key " split," so as to provide for the note G$ as well as 
the note E. 

Finally, although the G| may have been a novelty (indeed, 
Italian instruments seem never to have provided for it when 
the lowest key was E), the short octave itself was none, for 
examples are to be had of this arrangement of the bass notes 
before 1500. 

FINGERING IN THE SIXTEENTH CENTURY. 

We have not yet reached the height of von Billow's ideal, 
namely, that the correct fingering of a pianoforte piece is that 
which makes it possible to play the piece in any key without 
further cogitation on the part of the performer. On the con- 
trary, there are still with us excellent persons who would 
hesitate to use the thumb on a black key, if there was any 
alternative. 

This being so in the 2Oth century, let us be careful in 



158 AN ELIZABETHAN VIRGINAL BOOK 

our judgment of a much earlier time, with its even more 
elementary notions, and a far less satisfactory instrument 
on which to exercise those notions. 

A book was published at Leipzig in 1571, by a German 
called Ammerbach, entitled " Orgel oder Instrument Tab- 
^ latur," which tells how the scales were fingered in his day. 
Mr W. H. Cummings (" Proceedings of the Musical Associa- 
tion," London, 1893, page II) gives the important details of 
this scale fingering as follows : 

1. The right hand thumb was not used at all. 

2. The little fingers very rarely employed. 

3. When the thumb of the left hand was occasionally used, 
it was marked with a o. 

4. The index finger was marked I, and the others 2, 3, 4. 

Thus we see that what is now called " English " fingering 
was "German" fingering in 1571. And this was still the 
case in 1697, when Daniel Speer published a book called 
" Das Musicalische Kleeblatt," which gives the same signs, 
o, I, 2, 3, 4, and shows that the right thumb was still 
entirely neglected, although the left-hand thumb was more 
frequently used. 

Various other books, including Mattheson's " Kleine 
Generalbassschule " (1735) and Maies's " Musiksaal " (1741), 
show that the o, I, 2, 3, 4, was still used in Germany, and 
that the thumbs were generally unused, especially the thumb 
of the right hand. 

An Italian book, "La prima albori musicali, etc.," by 

Father Lorenzo Penna, printed five times between 1656 and 

1696, gives fingering allowing of the use of the three middle 

} fingers only, and ignoring both the thumbs and both the 

little fingers. 

While this sort of thing was going on on the Continent, 
we find that England always used the thumbs and little 



AN ELIZABETHAN VIRGINAL BOOK 159 

fingers quite freely, and marked the fingering with I, 2, 3, 
4, 5. This was so in 1599, the date of a MS. book of 
lessons for the virginals, fully fingered, and it was still so in 
1757, the date of the last edition of " The Muses' Delight." 

The system of naming the fingers -J-> I, 2, 3, 4 seems to 
have been first brought to England by a German, Rudolph 
Falkener, who published " Instructions for playing the 
Harpsichord" in 1762 and 1774. Another German called 
Heck tried to counteract these evil ways, and to re-introduce 
the old English fingering, I, 2, 3, 4, 5, speaking on his 
title-page of " such as have accustomed themselves to a 
wrong way of fingering." Heck did not succeed in his 
effort, and dementi (who came to London in I777)> together 
with Dussek (who arrived in 1788), completed the fell work 
by setting up as music publishers and marking their fingering 
+ ,I, 2, 3, 4. Thus the original German fingering came to 
England, and has since been wrongly known as " English." 

In the meantime the Germans had very sensibly adopted 
the English method, for Emmanuel Bach's " Versuche," the 
letterpress of which is dated 1759, uses x 2 3 4 5> anc * 
fingers the scale of C in many various ways, including 




And thus the " English " fingering has, ever since, been 
known as " German." For the sake of completeness it may 
be stated here that Couperin's "L'Art de toucher le 



160 AN ELIZABETHAN VIRGINAL BOOK 

Clavecin " (Paris, 1717) marks the fingering in the English 
way, viz., I, 2, 3, 4, 5. 

A slight notion of the principal difference between modern 
pianoforte fingering and that of such men as Couperin and 
J. S. Bach may be got from these two passages, fingered by 
Couperin : 
RIGHT HAND. 4345432 3 (1717) 



and 




234345 

In the first example, 3 is passed over 2. Emmanuel Bach, 
forty years later, disallowed this, but his father John 
Sebastian still practised it. Emmanuel also said that only the 
thumb might pass under. His father, however, allows even 
the little finger to pass under the third (i.e. 5 under 4). 

John Sebastian Bach told his son Emmanuel that when he 
was young, the thumb was only used for great stretches. 
In this connection, it may be worth while pointing out that 
in the Fitzwilliam Virginal Book it is the rule for the big 
chords to be in the left-hand part ; and this may have led to 
the habit of using the left-hand thumb more freely than that 
of the right hand, which seems to have been a long estab- 
lished custom in Germany, judging from the account of 
Ammerbach's fingering (1571), given above (p. 158). 

The student cannot be too careful in keeping clearly before 
him that the essential difference between the modern piano- 
forte and the old " plectrum" claviers renders it misleading 
to judge of the ancient method of fingering from his know- 
ledge of what is suitable on the keyed instruments of our 
own day. 



AN ELIZABETHAN VIRGINAL BOOK 161 

In the first place, " touch," in our sense, could not exist 
on the virginals, spinet, or harpsichord (all of these were 
"plectrum" claviers), except with a limited meaning. A 
" smart " downward pressure, and an equally " smart " 
return of the finger in an upward direction was necessary, on 
account of the mechanism itself. What we call " evenness " 
of touch mattered not at all, again, on account of the 
" plectral " action. " Repetition " of one note, however rapid, 
was perhaps better done with one finger than with several.* 
Again we see the essential difficulty of judging the 1 6th or 
1 7th century fingering, when the judge knows only the 
pianoforte, and that in its perfected form. 

Again, a method, whether of " touch" or of mere finger- 
ing, suitable to the harpsichord class of keyed instrument, 
might well be unsuited to the clavichord, which had a 
" tangent" action and not a " plectral," and which, therefore, 
required both evenness and gentleness in the playing. Once 
more we perceive how misleading it may be to judge from 
the pianoforte to the older claviers. In this case (the German 
" tangent" clavier) we find an instrument possessing at any- 
rate two features which are not present either in the piano- 
forte or the harpsichord, viz., the possibility of inducing slight 
temporary alterations in the pitch of a note, by pressing it 
more or less strongly ; and the " Bebung," or vibrato, which 
could be got on a single note by rocking the finger on the 
key. 

These facts must be remembered in studying the specimens 
of 1 6th century fingering which have come down to us 
in the Fitzwilliam Virginal Book. 

It is fortunate that examples of Bjull's fingering occur in 
this collection, for he probably represents the high-water 

* This is the experience of modern players on the virginals. The Eliza- 
bethans themselves, however, seem to have thought differently ; see p. 164 
below. 

L 



1 62 AN ELIZABETHAN VIRGINAL BOOK 



mark of 1 6th or early iyth century agility. But in addition to 
Bull, instances are found in pieces by Munday, Byrd, and 
others. 

John Munday's " Fantasia" in the ^Eolian Mode (i. 19) is 
not very interesting musically, but will serve to introduce us 
to the " Fingering of the Sixteenth Century," as marked in 
the Fitzwilliam Book. 

Munday's piece, above named, has this passage for the 
right hand on p. 19. 




with fingering as marked. 

I suppose that the complete fingering may have been 

_ J -^~"f"* g i^ f i* f -rf3 

i 345 345 345 4 3 

A similar passage in the next line for the left hand is 
marked with a 3 on the corresponding note, and thus was 
most likely fingered 32 I, 32 I, 3 2 I, just as we should 
do it now, with the thumb on the first quaver of each beat. 

The same composer's " Fantasia on the Weather," i. 23, 
has a suggestion that Munday used the five-finger exercise as 
we do : 



32 i 



1234 5 



12 34 



LEFT 
HAND. 




AN ELIZABETHAN VIRGINAL BOOK 163 

The fingering given in the MS. is merely the 3 in bar 2. 
It seems clear that the fingering printed above the notes is 
implied, therefore John Munday probably used I, 2, 3, 4, 5 in 
order over five adjacent notes, and understood the " con- 
traction " of the hand implied in the change from 5 to 3 on 
the same note, and accepted the most modern dictum that 
similar passages should have similar fingering if possible. In 
the same piece (p. 26) we find the following for right hand, 




The 5, 4, 3 underneath is given in the MS., and I presume it 
to imply what is printed above the notes. 

Bull has an interesting line (i. 70) in a Galiarda, which 
seems to show that he too thought it well that passages of a 
similar outline should have the same fingering; and, at the 
same time, tends to prove that he avoided the thumb except 
for chord playing. 

As in MS 

BULL'S FINGERING. 

JOHN BUIL (1563-1628.) 




S 2 

-[-m *. -^~m ^-m-^^-m -f 

:etc. 







52 52 



The continual recurrence of 5, 2 is instructive. The note 
marked with the asterisk is acfifriculty. If correct, it must 
mean that the note before, e, was intended to be played by 2, 
and not by I, which one would certainly have expected. It 



1 64 AN ELIZABETHAN VIRGINAL BOOK 



seems likely to be a mistake, for the result of having 3 on 
F| prevents 5 being on the following B, and breaks the clear 
rule of the whole passage. 

(N.S. The treble part of line 4, bar I, on p. 71, is plainly 
incorrect, and should read 



j 



et,:. 



as is seen by comparing line I, bar 3, of p. 71.) 

An unknown author provides another puzzle, reflecting 
the want of certainty in the use of the thumb, in a Praeludium, 
i. 82, lines 2 and 3, where the change of Angers for repeated 
notes is marked with a thoroughness which is quite of our 
own time, but the thumb apparently avoided, as a rule, in 
spite of its obvious value. 

FINGERING (UNCERTAIN AUTHOR), FITZWILLIAM BOOK, i. 82. 






3242322525421424 52422522 




5 3 4 



2_2 5 3 2 4 2525 



Some of this is quite in the modern style. But it is hard 
to see why the thumb should have been disregarded in the 
three cases underlined, where 2, 2 is used, although every 
consideration would have indicated 2, I as the right fingering. 

Bull seems to have believed entirely in changing fingers 
for rapid repetition of the same note, e.g. 9 in his Pavana, 
i. 125. 



AN ELIZABETHAN VIRGINAL BOOK 165 
BULL'S FINGERING. 



Right Hand. 






5 2 4 1 



524 



3 * S 5 




A passage on the previous page (i. 124) illustrates other 
probable features of his system : 

Right Hand. 

2 4 




2 3 






Bull's own fingering is under the notes : my guess at the rest 
of it is given above. 

On the next page of the same piece (i. 126) is a passage 
showing Bull apparently avoiding the right-hand thumb, but 
using that of the left hand freely. 

VARIATION ON STRAIN 2 OF PAVANA, WITH FINGERING AS 
IN MS. 

JOHN BULL. 




2 3 



1 66 AN ELIZABETHAN VIRGINAL BOOK 




The use of the left-hand thumb in the short ascending 
passages in bar I is quite plain, as is the avoidance of the 
right-hand thumb in bars 3 and 4. 

On p. 128 there is evidence that Bull was uncertain in his 
use of the figures for the left hand, and that he, with others 
(such as Byrd and G. Farnaby), sometimes used I for the 
little finger, and 5 for the thumb of the left hand, t.g. 9 



J 



and 



3 5 






Bull's 



which can only mean what I have put over the notes, 
fingering (as usual) I have put underneath. 

Other singular instances of the fingering of this piece must 
be passed over. 

Byrd, in a Pradudium (i. 83, line I, and i. 84, line 4), shows 
that 5 meant " thumb " in the left hand ; and similarly Giles 
Farnaby, in the second line of " The King's Hunt," i. 196. 
An interesting feature of the latter case is the ornament in- 
terpreted by Mr Dannreuther as a " slide." The fingering 



AN ELIZABETHAN VIRGINAL BOOK 167 



curiously confirms the interpretation, while the interpretation 
naturally explains the fingering, which would otherwise have 
been inexplicable. The bar runs thus : 
FINGERING AS IN MS. 

GILES FARNABY. 

Left hand 
5 




A similar " slide " appears in the right-hand part of the 
same bar, fingered in a more comprehensible manner, i.e. 

123 

^ * : 

:, meaning 



Byrd seems to have the crossing of 4 over 5> ' 4545 
ascending. 

FITZWILLIAM BOOK, i. 84. 

BYRD (1538-1613 
4 




R. H. rttZZKZ5t3C=lZ=t3LI 




which I take to mean something like the following 
R.H. 







1 68 AN ELIZABETHAN VIRGINAL BOOK 

This is from a Praeludium (printed also in "Parthenia" as 
Prelude to the "Carman's Whistle"), the reference in the 
Fitzwilliam Book being i. 84. 

The reader now has before him most of the typical speci- 
mens of 1 6th century fingering contained in the Fitzwilliam 
Virginal Book. 

We gather from them : (i) That the English writers of 
the Elizabethan time used i, 2, 3, 4, 5 to denote the fingers 
of each hand, but that there were various uncertainties, e.g., 
some composers used I for the little finger of the left hand, 
and 5 for the thumb. (This was also the arrangement 
contained in a "Collection of Ayres," published in London 
in 1700.) 

(2) That the use of the right-hand thumb was avoided, 
but by no means forbidden. (This apparently was still the 
case under Mattheson in 1735 and Maies in 1741.) 

(3) That the fingering of scale passages was uncertain, and 
generally based on the crossing of fingers, rather than on the 
" passing " of the thumb. 

(This also appears to have gone on until the time of 
Emmanuel Bach, i.e. the middle i8th century.) 

(4) Rapid repetition was practised with change of fingers. 
With regard to this last, the student must be warned again 

that repetition on a harpsichord is not in the least like repeti- 
tion on a pianoforte. 



CHAPTER XIV 

ORGAN PIECES, PRELUDES, MADRIGALS (TRANSCRIPTIONS) 

THERE still remain three classes of pieces in the Fitzwilliam 
Book which have not been reviewed in these chapters, viz., the 
17 " Organ" pieces (settings of ecclesiastical melodies, etc.) ; 
the 19 Preludes (13 by six writers, and 6 more anonymous); 
and Philips's arrangements of 9 Italian Madrigals or parts of 
Madrigals, with which may be included Giles Farnaby's own 
arrangement of his Canzonet, " Ay me, poor heart " (ii. 330, 
followed by another movement, ii. 333), and Bull's variations 
on Dowland's Madrigal, " If my complaint " (ii. 242 and 244), 
which was published in 1597 as one of Dowland's First Set 
of Songs, and as " Captain Piper's Galliard" in his " Lach- 
rymae," 1605. 

ORGAN PIECES. 

The class first named, which seems on the whole to be 
referable to the organ, consists mainly of 13 pieces founded 
on Church Melodies, and comprises examples by Tallis, Bull, 
Parsons, Byrd, Blitheman, and an anonymous writer. 

These pieces are amongst the earliest examples of the 
method of embroidering an Ecclesiastical hymn tune with a 
contrapuntal accompaniment, which was practised in the l6th 
century by Netherlanders (Sweelinck), Italians (the two 
Gabriellis, of Venice), and by various Englishmen. Relics 
of the practice are to be found in J. S. Bach's organ works 
more than a century later. 

They are naturally founded on the general method of the 

169 



1 70 AN ELIZABETHAN VIRGINAL BOOK 



mediaeval vocal part music, where a Church Melody was taken 
as a basis (known as Canto Fermo) for an extended move- 
ment or movements, e.g. y the Mass. 

Naturally, also, they are altogether " Modal " in character, 
and are further from modernity than any of the dances or 
other pieces which are in this collection. 

As a rule, these " organ " pieces will be found without 
interest to the modern mind ; but much may be gathered by 
a careful study of their main features. Anon. (i. 421) sup- 
plies us with an example of the simplest way of doing these 
things, namely, the plain song in the right hand, and a con- 
tinuous " run " in the left hand, thus : 

" VENI." 

ANON (date ? 1550). 




Here the running counterpoint defines the implied harmony 
conveniently for us moderns, who have generally lost the 
tradition, e.g. 

THE SAME, PLAIN. 




rcto 



j.. A. A 



t \JJ * i 

^^ \J 



AN ELIZABETHAN VIRGINAL BOOK 171 

This, of course, is got by taking the principal notes of the 
" run " as the bass, and occasionally a hint from the same 
source concerning the precise form of the chord indicated by 
the bass. 

The progression of the two first chords given above would 
probably be fatal in a modern " examination " ; however, it 
was not avoided in the l6th century, in spite of the " tritone" 
contained therein (i.e. the augmented 4th, with the B in the 
one chord, the F in the other). It may be seen again, 
quite clearly implied, on p. 422, bars 4 and 5, where the 
tritone occurs in its most plain-spoken shape, in the extreme 
parts, viz., the treble and bass. I now give the whole of 
this Plain Song, with the harmonies implied by this anony- 
mous 1 6th century composer. 



"VENl" (FlTZWILLIAM BOOK, i. 42 1). 



Harmonies taken from the counter- 
point of ANON. 1550? 



3=^?S=^33ES5=g33 

S5=eEzESESE=Bz==t: 




1 72 AN ELIZABETHAN VIRGINAL BOOK 



^. A 



rp-r i 

J_J_J: 



^ 



F 



A- 



r 



' jjzzJz^ 
fr"73 FQ j=n 

11 ' r r^ 



p*- 



* 



Q t. LZ 



. A 



j 



r 
^= 



E.W.N. 



AN ELIZABETHAN VIRGINAL BOOK 173 

The final pause chord is given as in the original. The 
rest of the "reduction" probably does not much misre- 
present the implied harmonies of the Fitzwilliam author. 
The " flat" seventh note of the .^Eolian mode is well shown 
in the grd bar from the end, where the harmony indicated 
by the counterpoint of the original piece absolutely prevents 
the G from being anything but " natural." Also the tritone, 
as already stated, is to be seen more than once, undisguised 
and apparently without blame. The same Canto Fermo is 
set by Bull (i. 138); a short extract is given above, p. 76. 

Two works of this kind, by Thomas Tallis, both entitled 
" Felix namque," are given. The first (i. 427) is the earliest 
dated piece in the book, 1562; the second (ii. i) is dated 
1564. A specimen of the former will be sufficient to give 
an idea of both. I select passages where the plainsong 
happens to repeat itself more or less, so that the composer 
may be seen treating the same notes in two ways, viz. 



PART OF "FELIX NAMQUE" 
(FITZWILLIAM BOOK, i. 429, 1. 5). 

THOMAS TALLIS (dated 1562). 




174 AN ELIZABETHAN VIRGINAL BOOK 







The chromatic harmony (bars 5 and 6) is worth noticing. 
The chord marked * really has the previous minim "g" 
sustained over it, and should be compared with the asterisked 
chord in the next extract. 

The same notes as set on p. 431, 1. I. 



THOMAS TALLIS (dated 156*). 



T^-- 



AN ELIZABETHAN VIRGINAL BOOK 175 






I r 



r 



m 



"CTT ' ~~|" i" 'T r Y i f 

jj/y 



:^r:-^-* 




-r^-r^^rfVfj 



rt 



M 



i^dEfe^E 



t 



176 AN ELIZABETHAN VIRGINAL BOOK 



The chord marked * is very strange. Also I suspect the 
final crotchet (marked f) in the left hand of the bar before 
should be D F, instead of E G. 

Several examples are by Bull. " Christe Redemptor" 
(ii. 64) has the Canto Fermo in the tenor, with treble and 
bass counterpoint added, in continual imitation. It is dry, 
which is a pity, for the plain song is a fine one, and I give 
it here as Bull has it, omitting the repetition of notes less 
than breves. 

"CHRISTE REDEMPTOR" AS GIVEN BY BULL 

(FlTZWILLIAM BOOK, ii. 64). 



-*** 



101 



A very short quotation will be sufficient to show the 
aridity of Bull's treatment. 

" CHRISTE REDEMPTOR." 

JOHN BULL. 




C. F. in TENOR, one note in each bar. 



SES33Ei 



AN ELIZABETHAN VIRGINAL BOOK 177 







This plainsong is known elsewhere as " Jesu dulcis 
memoria," and may be seen in " Hymns Ancient and Modern " 
(No. 177), fitted with harmonies which in many instances 
can only be regarded as out of place. 

The plain song " Gloria Tibi Trinitas" is set no less than 
five times in this series of pieces, three settings being by 
Bull, the others by Parsons (Robert the father, who died 
1570, or John his son) and William Blitheman, who is said 
to have been Bull's master. In four out of the five cases 
the piece is called 

"In nomine" 

which may perhaps be explained by the dialect of North- 
East Yorkshire, where, till recent times, it was not un- 
common to speak of any set of verses as a " nomminy," with 
a reference to words such as had to be committed to 
memory. This title, " In nomine," may therefore merely 
mean " Hymn." 

Blitheman's effort (mostly in 3 parts) is dry (i. 181). 
Some patience is required to find the notes of the plainsong 
in Parsons' setting (ii. 135), which is somewhat confused, 
and mostly in four parts. 

Two of John Bull's settings (i. 135 and i. 160) of this 
hymn are not very exhilarating, but the remaining one 
(ii. 34) requires some notice, partly because of its greater 
musical value, partly because it is a characteristic example 

M 



178 AN ELIZABETHAN VIRGINAL BOOK 

of Bull's fondness for experiment. In this case the whole 
piece (six heavy pages) is in V time. The final page is 
further complicated by the dotting of the notes, and the 
last bar but one on p. 39 (being part only of one of the V 
bars, printed so for convenience of reading) contains the 
puzzling combination of 9 quavers (R. H.), a minim and 
two crotchets (tenor), and a dotted minim and dotted crotchet 
(bass). 

The rhythmical experiment, however, is not without suc- 
cess, and a specimen of the first few bars is now given. The 
Canto Fermo is in the bass. 

FIRST NOTES OF * GLORIA TIBI TRINITAS." 



FIRST BARS OF BULL'S SETTING (FITZWILLIAM BOOK, ii. 34). 

Canto Fermo in the Bass (each note 1 1 beats). 



11 

4 
11 




q^. J^-_^ 


t= ' *3-d ife 




r * * 4 

f r-f=-ft 


E=P=EE3E 
J. * N I J J 


/j 5? 


^ m 


:-'.. ^ J -\ 


-^5i= ij/ ^- 


f- 


, 


Plain Song in 

-1 


BASS, one note in each bar. 

=--' J^Mrt 


r r 


|/T\ I 

r 


* f SEE 

r I 

- ^^-- 


* r r~ 

^ N J j j- 
P & \ 



AN ELIZABETHAN VIRGINAL BOOK 179 

* J ) K i_ J^T} J^rJ^ 



J 






f ' 










r r 



r- 



i8o AN ELIZABETHAN VIRGINAL BOOK 

I. ^J J_:J^^J_J_4 



i 




i=j J 



i 






/U 






etc. 
j*. 



Bull's liking for complications may be illustrated from the 
" Gloria Tibi Trinitas," found in i. 160, where on p. 162, 
lines 2 and 3, the bass and alto exhibit the "Proportion" 
known as " Sesquialtera," i.e. 2 equal notes against 3 of the 
same. 

EXTRACT FROM ANOTHER SETTING OF " GLORIA TIBI 
TRINITAS," SHOWING " PROPORTION " OF " SESQUIAL- 
TERA " (3 to 2). 

(Canto Fermo in the treble. One note to each bar.) 

Dr BULL. 




-f-rf rfT r ~i 



AN ELIZABETHAN VIRGINAL BOOK 18 



Another well-worn plainsong is set by Bull under the 
name of " Salvator Mundi" (i. 163). 



!zetc. 



This is very generally set to " Come, Holy Ghost " (Veni, 
Creator) in English hymn-books. 

Bull produces three variations, with the Canto Fermo in 
the treble. The first variation is in two parts, the left-hand 
part consisting mainly of runs. An interesting extract, how- 
ever, is the following : 

EXAMPLE OF ARPEGGIOS AND BROKEN CHORDS 
(FITZWILLIAM BOOK, i. 163). 

(From near the end of line z of plainsong "Salvator Mundi," or "Come, 
Holy Ghost.") BULL, 1563-16x8. 




1 82 AN ELIZABETHAN VIRGINAL BOOK 



The sturdy harmonies of the following (i. 166) portion of 
the 2nd variation are worth considering. (I begin the 
extract where the plainsong joins on to the corresponding E 
which ends the above quotation from variation I.) 

FROM THE SAME. VARIATION 2. (FITZWILLIAM BOOK, i. 166.) 

BULL. 




' V 



*mp 



E i r"? r i i =i~ r r= 



-? 



r 



^TrrTr 



=& 



etc. 



I have purposely simplified the bass here, in order that the 
reader's attention may be fixed on the musical interest of the 
passage. As a matter of technical interest, it may now be 
mentioned that the bass is in the form of broken octaves, 
ornamented by mordents thus 



etc. 



After ten bars like this, the form of the bass becomes even 
more complicated. 



AN ELIZABETHAN VIRGINAL BOOK 183 
THE SAME, i. 167. 

G] C 



BOLL. 




Here, in the first bar, the four crotchets in the right-hand 
part denote that Dr Bull suddenly forgot how slow the 
Canto Fermo was, and managed to have four notes of it in 
one bar. Consequently he had to pull up quickly, and do 
the three notes C A G over again at the proper pace, in the 
next three bars. 

It may be of value to point out, that although Bull har- 
monises this plainsong sometimes with Fj|, sometimes with 
F$, e.g., the opening notes 



184 AN ELIZABETHAN VIRGINAL BOOK 
Harmony Implied, i. 163, 1. I. 



or 



From i. 167, 1. $. 



he shows no sign of altering a somewhat critical Btj near the 
end of the plainsong, hampered as it is by an F, which to 
more modern ears would present a difficulty in harmonisation. 



THE SAME, i. 164. 

(Implied Harmonies, end of var. i.) 



BULL. 






f=rf 



_C2- 



M=l 



g 



AN ELIZABETHAN VIRGINAL BOOK 185 



The harmonies given by Bull in the second and third 
variations at this point, make B[? altogether impossible. 

On the other hand, Bull's setting of " Gloria Tibi Trinitas " 
(i. 137, 1. 4) introduces an altered note in the plainsong 
which is not found in the other four settings of that melody 
mentioned above, two of which are by Bull himself, and the 
others by Blitheman and Parsons. The phrase (^Eolian, 
" transposed," i.e. with one flat in the signature, the pitch 
being lowered a perfect fifth) is as follows : 

LAST PHRASE OF " GLORIA TIBI TRINITAS " (TRANSPOSED). 
Harmonies implied by Blitheman, i. 182, 1. 4. 

-&- 




i 

Bull's harmonies are not widely different from this, but he 
alters the Eft to Eb, thus 

LAST PHRASE OF " GLORIA TIBI TRINITAS," WITH BULL'S 
CHORDS AND ALTERED E (FITZWILLIAM BOOK, i. 137). 







etc. 



1 86 AN ELIZABETHAN VIRGINAL BOOK 

These harmonies are merely simplified from the original 
florid passages (i. 137), and are well worth study, as is the 
whole treatment of the last six bars of this piece. 

Lastly, in Bull's setting, i. 1 60, quoted from above (p. 180), 
another alteration is allowed, viz., the last note but one (e.g., 
in the last extract the note C) is sharpened (i. 162). 

Alterations of this kind are always to be found in music of 
the period, and the practice seems to depend entirely on 
convenience. The same remark applies to similar cases in J. 
S. Bach's Choral Preludes on well-known hymns, and to his 
use of the Canto Fermo in his Church Cantatas, where the 
actual tones and semitones of the plainsong are not allowed 
to interfere with Bach's added vocal and instrumental parts. 

Three pieces of this class, called " Miserere," are included 
in the MS. The one by Bull, ii. 442, with three variations 
on the plainsong, in 3 parts, is very dry indeed. 

The other two are by Byrd, and are worked on a different 
melody from that given by Bull. The setting with the 
melody in the alto (ii. 232) is in four parts, and has already 
been referred to (p. 74 and following) and quoted from as a 
possible example of " tonal " answer in fugue. A rather 
important correction of ii. 233, 1. 3, in this piece has also been 
given in the same place. The arrangement in three parts by 
Byrd is in ii. 230, but is not interesting enough to deserve an 
extract. 

"Heaven and Earth" (i. 415) probably belongs to this 
group of ecclesiastical pieces, as it seems to be written round 
some plainsong or another. The author may be Mr Tregian, 
but this is doubtful. Musically, the piece is without value. 
The words may have something to do with the " Sanctus " 
or " Te Deum." 

Four pieces called " Grounde " may be considered here. 
They are, of course, variations on " ground bass," and there- 
fore somewhat related to the exercises on Canto Fermo 



AN ELIZABETHAN VIRGINAL BOOK 187 

which have just been dealt with. Giles Farnaby's " Groude" 
(ii. 353) consists of 14 variations, some of them very 
difficult, on the following " grounde." 

FITZWILLIAM BOOK, ii. 353. 

GILES FARNABT (i6th century). 



<" 

I give this mainly as a good example of the absolute want 
of "key" feeling in the "grounde" itself, which may be 
useful to the student of these matters (see above, p. 138 ff.). 
But it will also serve as an instance of the use of " bars " in 
these early days, e.g., the " grounde " is apparently in triple 
time, whereas the piece immediately proceeds (at the entrance 
of the tenor part) in duple or quadruple time. Again, the 
student will observe that the sign ( does not refer to the 
probable contents of any bar, but simply and solely to the 
relative values of notes occurring in any bar, e.g., that a semi- 
breve contains two minims, and not three.* 

*The following cases, gathered from the Fitzwilliam Book, will be of 
interest to students : 

Sign used in MS. Modern Meaning. Reference. 

c f i. 267 

$ i H. 453 

ii. 489 
iL 5 i 
ii. 450 

G f or f or f i. *i8, i. 54, 1. 4*1 

3 f ii. 486 

3 ii. 471 

and the one just given, where (p is used indifferently for either 3 or 2 semi- 
breves in the " bar." 

Thus, to take one example, the sign ( may mean either ^, ^,4, ; 
or, | may be expressed by 3,( , (J), or C. 



f 

I 



188 AN ELIZABETHAN VIRGINAL BOOK 



Byrd's variations on " Tregian's Ground" (probably not 
Tregian's but Hugh Ashton's, see the Nevell book) are 
good music (i. 226), but are better classed with the varia- 
tions on Dances, as the ground is practically a Galliard. 
William Inglott's "Galliard Ground" (ii. 375) is dry. 
Much the best of these pieces is Thomas Tomkins's (ii. 87) 
" A Grounde." The variations are interesting, and are 
founded on a very short but quaint subject which seems 
to be taken from the popular tune called " Up Tails All," 
set by Giles Farnaby (ii. 360). 







Here are short examples of the many ways in which 
Tomkins uses this little subject : 



THOMAS TOMKINS ( 1 6th century). 



i j i j , J i rjv^L 



In the bass. 




AN ELIZABETHAN VIRGINAL BOOK 189 

In the treble. 




PRELUDES. 

There are nineteen " Preludes " in the Fitzwilliam Book, 
six by anonymous writers, seven by Dr Bull, two by Byrd, 
and the rest by Galeazzo (an unknown person), Sweelinck, 
Giles Farnaby, and Thomas Oldfield (another unknown). 
As a class, these Preludes are not interesting, as they 
generally consist of chords ornamented by florid passages, 
runs, shakes, and arpeggios, which were merely meant to 



1 90 AN ELIZABETHAN VIRGINAL BOOK 

warm the player's fingers before embarking on a more serious 
performance. 

Few of them can be regarded as part of a "suite" of 
pieces. Amongst those which are so, however, we find the 
two by Byrd ; i. 83, which is known (from " Parthenia") to 
be the Prelude to " The Carman's Whistle," and i. 394, 
which is en suite with the Fantasia in i. 1 88. 

Bull's Prelude in the Dorian mode (i. 158) is said (by 
Ward) to be the "Prelude to Gloria Tibi Trinitas." If this 
is the case, the particular setting of that plainsong which 
it belongs to may be the one in i. 135, and not the one 
which immediately follows it in the volume (i. 1 60). There 
is, however, nothing in common between the melody of 
"Gloria Tibi Trinitas" and the music of this particular 
Prelude. 

Bull's Preludes are mainly valuable as supplying specimens 
of the high technical standard which he must have attained. 
Among such are the following : 



TECHNICAL PASSAGES FROM BULL'S PRELUDES 
I. (FITZWILLIAM BOOK, i. 419 Mixolydian.) 



Ritfit hand 




etc. 



AN ELIZABETHAN VIRGINAL BOOK 191 

2. (FITZWILLIAM BOOK, ii. 260 Mixolydian.) 



BULL. 



- * 




3. (FITZWILLIAM BOOK, ii. 274 ^Eolian.) 



BULL. 




--*--*-_}-" 



From these three sample passages we have (a) rapid 
repetition in both hands, (b) chord passages for right hand, 
(c) broken octaves and other broken intervals, (d) extended 



1 92 AN ELIZABETHAN VIRGINAL BOOK 

scales, in both hands, and in contrary motion, and in similar 
motion at the interval of a sixth. 

An example of arpeggio has already been quoted. See 
above, p. 1 8 1. 

Giles Farnaby's Prelude in the Ionian Mode (ii. 372) is 
worth naming with Bull's efforts, for it is really difficult, and 
only a thoroughly neat player could do it well. 

The example by Sweelinck (i. 378) is hardly in the same 
class with the others, for it is called " Preludium Toccata," 
and is far more a " Fantasia " than a Prelude, being fugal in 
character and six pages in length. Amongst other difficulties, 
it contains the following for left hand : 



FITZWILLIAM BOOK, i. 383. 



Left hand 



J. P. SwEEMNCK. 







The Anonymous Preludes (six in number) are more 
sensible than any of the others, especially the Ionian one at 
i. 85, and the ^Eolian at ii. 169, which latter piece has some 
real musical interest over and above the scales and runs in 
which it abounds (e.g., one passage in semiquavers, for the 
left hand, goes on for fourteen bars). An extract has already 
been given from the Prelude at i. 8 1, showing some curious 
fingering of a difficult passage (see above, p. 164). 



AN ELIZABETHAN VIRGINAL BOOK 193 

MADRIGALS. 

It is not necessary to say much about the transcriptions 
of Madrigals which are given in the Fitzwilliam Book. They 
are not really of any great importance. The method of 
"arrangement" is what might be expected, viz., the plain 
harmonies of the original vocal parts are used by the tran- 
scriber as a skeleton on which to hang elaborate passages, 
shakes, runs, and other ornaments. A few bars illustrating 
this method are given in the " Oxford History of Music," vol. 
iii. p. 94, where an extract from Philips's arrangement of 
Cost moriro (by Luca Marenzio) is printed, with the corre- 
sponding bars of the original. 

The same work (vol. iii. p. 42) quotes a passage from 
Caccini's Madrigal (so named) for solo voice and accompani- 
ment, " Amarilli, mia bella," an arrangement of which 
appears in the Fitzwilliam Virginal Book, by Peter Philips, 
dated 1 603. Here is Caccini's original, with Philips's tran- 
scription printed immediately below it. 

" AMARILLI, MIA BELLA " for solo voice and accompaniment, 
by Caccini (Julio Romano), 1558-1640. 



ma - ril - li, mia bel - la, non ere - di del mio 

i . J. 




i 9 4 AN ELIZABETHAN VIRGINAL BOOK 

The same, transcribed for the Virginals by Peter Philips, 
dated 1603. 

FITZWILLIAM BOOK, i. 329. 

znJ-jiLz^zq^ 







The lengthening of the original Caccini by Philips may 
point to the method of singing the madrigal, e.g., the slight 
pause after " bella," which he fills up with a run ; and the 
extension of the second syllable in " desiro," which may well 
represent some singer's trick lingering in Philips's memory 
from an actual performance of the piece. 



AN ELIZABETHAN VIRGINAL BOOK 195 



The following short extract is not without interest : 
TIRSI " (the " first part " of a madrigal in three movements, 

"FRENO" and "Cosi MORIRO " being the other two. 

From " Musica Transaipina," published in London, 

1588). 

LUCA MARENZIO (1550-1599). 



SOPRANO. 



ALTO. 



TENOR. 
QUINT us. 
BASS. 



Thir 



to 




Thir 



jp=|E|=p \ ^ ^ ^ 


dye de - sir - ed. Thir - 




sis. etc. 


m i 1 1 \ 






^/ J " fy ' rzn 
dye de- sir - ed * " ( ' 


IC-pJ! 
- SIS 

, r-ftj-^ 


(Wj i 






Peter Philips transcribes the passage thus : 
FITZWILLIAM BOOK, i. 280. 

The same arranged by PHILIPS. 




1 96 AN ELIZABETHAN VIRGINAL BOOK 







r*=jz2=*: 






m 





Philips's fifth bar is a rather clever attempt to render the 
difficulty presented in the Soprano and Alto at the words 
" dye desired," where the two voices cross and the melody 
consequently is hidden. The largeness of effect produced 
by the full vocal harmony at the repetition of the word 
" Thirsis " is indicated by the florid passage in Philips's 
7th bar. 

Besides the 9 arrangements of madrigal movements by 
Peter Philips, there may be included in this class of pieces 



AN ELIZABETHAN VIRGINAL BOOK 197 

Giles Farnaby's own transcription of his Canzonet " Ay me, 
poore heart"* (ii. 330, 333), and Bull's Galliard called 
(Captain) Piper's Galliard (ii. 242), which is a well-known 
madrigal by Dowland, and is found in Dowland's " First Set 
of Songs "(1597), set to the words " If my complaint could 
passion move." 

* Date 1598. 



CHAPTER XV 

THE COMPOSERS IN THE FITZWILLIAM VIRGINAL BOOK 

THE following paragraphs, in the shape of notes on about 
30 of the composers represented in the Fitzwilliam Book, 
are intended as a practical guide to the student who wishes 
to examine the printed edition for himself. Out of such a 
mass of music (nearly looo folio pages) it is difficult to 
extract that which is most valuable, unless much time is 
available for the purpose. 

It may be really useful, then, to indicate shortly the pieces 
which appear to deserve attention in the first place. 

I. Bull (said to be a pupil of Blitheman, see above, p. 185) 
stands almost in a class by himself, as the Liszt of the 
1 6th century. He was a real virtuoso. There is scarcely 
one, of over 40 pieces by him in this book, that does not 
show Bull in this light. He is not a man of sentiment, and 
thus stands in contrast with Byrd, Farnaby (Giles), and others 
of the Anonymous composers. 

1. 70 is a good Galliard (see above, p. 164, for correction). 
" Dr Bull's Juell," ii. 128, is good. The " Spanish Pavan," 
ii. 131, is a capital tune and somewhat unusual. " The King's 
Hunt," ii. 1 1 6, is worth attention, especially the French Horn 
imitations. Another interesting piece is " The Duke of 
Brunswick's Alman," ii. 146. 

2. Byrd. There are more than 70 pieces by Byrd, and 
most of them show signs of the Romantic spirit which is 
mainly absent from Bull's compositions. Although Byrd's 

198 



AN ELIZABETHAN VIRGINAL BOOK 199 

pieces are sometimes difficult, they are never so difficult as 
those of Bull, and apparently are never made difficult on 
purpose. 

The Pavana, ii. 384, is beautiful. " The Mayden's Song," 
ii. 67, is quaint, and the first page or so is worth playing.* 
" Walsingham," i. 267, and "The woods so wilde," i. 263, 
are excellent. " John, come kisse me now," i. 47, is capital, 
also " Callino Casturame" (Colleen Oge Asthore), ii. 1 86. 
A pleasant " suite" may be made of "The Lavolta," ii. 180, 
"The Alman," ii. 182, and " Wolsey's Wilde," ii. 184. 
Finally, see i. 223, where, in his " Hunt's Up," Byrd has 
used the old catch " Now God be with old Simeon," at sec. 9, 
in wrong rhythm. (The same tune may be seen in ii. 435, 
in Byrd's "Pescodd Time," which is apparently another 
version of the " Hunt's Up.") 

3. Giles Farnaby is as interesting as any composer in the 
whole collection. More than 50 pieces by him are included, 
and from them it may be gathered that he was a more clever 
player than Byrd, though nowhere near Bull in this respect. 
In sentiment and musical feeling Giles Farnaby's music is 
comparable with Byrd's. 

The Galiarda, ii. 419, is of the very best, especially the 
antiphonal passages for the two hands on p. 420. 

" Rosasolis," ii. 148, " Up Tails All," ii. 360, " Tower 
Hill," ii. 371 (short), and " Pawles Wharfe," ii. 17 (see above, 
p. 134), are worth playing. 

"Put up thy dagger, Jemy," ii. 72, is a very good tune, 
and variations I, 2, 3, 5 are of some interest. 

The Duet for Two Virginals, i. 202, should not be over- 
looked, as a curiosity. 

"Woody Cock," ii. 138, has a characteristic melody; and 

* Lilly's " Ancient Ballads and Broadsides" (1867) gives a 16th-century 
Ballad, "The Marchant's daughter of Bristow," "To the tune of The 
Mayden's Joy." Perhaps this "Mayden's Song" is the same, 



200 AN ELIZABETHAN VIRGINAL BOOK 

pp. 143, 144 show a rather unusual "augmentation" of the 
second half of the tune. 

" Quodling's Delight," ii. 19, is also very good, and may 
be usefully compared with a well-known and beautiful piece 
by Rameau (1683-1764), viz., the Gavotte and 6 variations 
in A minor. Quoted above, p. 135. 

Also compare " Fain would I wed," ii. 263, by Richard 
Farnaby (the son). 

4. Anonymous composers supply over 40 pieces, some of 
which are among the very best in the book. 

Examples are : " Barafostus' Dream," i. 72 (see above, 
p. 83), the " Irishe Hohoane" (Ochone), i. 87, see plate 
facing p. 8 ; the Alman and Muscadin, i. 75 and 74 ; the 
Alman, i. 65 ; the Galliard, i. 77 ; and the King's Morisco, 
ii. 373. 

5. Thomas Morley. The author of one of the most sensible 
works ever written on Counterpoint and the rest of the 
subjects usually called "theoretical" (lucus a non lucendo, 
probably, such things being entirely " practical "). Morley's 
"Plain and Easy Introduction, etc." (1597) would entitle 
him to our respect, even if his compositions did not compel 
our admiration. But as a composer he is more than re- 
spectable. Some of his Madrigals are well known, and are 
among the best of their kind. His 9 pieces in the Fitzwilliam 
Book, though not showing him at his best, are good work, 
especially the Dances here named. 

We may suppose a " Suite" in the three pieces (a) 
Alman, ii. 171 ; (b) Pavana, ii. 173 ; (c) Galiarda, ii. 177. 

(a) is in the Ionian Mode, (b) is ^Eolian, and is Dowland's 
celebrated " Lachrymae." Also " set " by Byrd, ii. 42, and 
G. Farnaby, ii. 472. (c) The Galliard partly fulfils the old 
condition of using the melody of the preceding Pavana. (The 
same holds good of the Pavan and Galliard by Morley at 
ii. 209 and 213.) 



AN ELIZABETHAN VIRGINAL BOOK 201 

ii. 177, line 3, bar 2, may be compared with Bach's 
Corantos, in respect of the uncertain rhythm (? f or ) in the 
final bars of sections (Suites Anglaises, and some Partitas). 

6. Orlando Gibbons. A great name. It is strange, the book 
contains only 2 pieces by Gibbons. Both, however, are 
more than worth hearing : the Pavana, ii. 479 (also found 
in Parthenia, 1611), and the setting of "The woods so 
wilde," i. 144, of which the most attractive portions are 
those numbered I, 4, 6, 7, 8. These are specimens of the 
best quality of the Elizabethan school. 

7. Thomas Tomkins. By whom are 5 pieces ; one, a setting 
of "Barafostus* Dreame," ii. 94, which is creditable, 'but 
not Romantic, as is the arrangement of Anon, at i. J2. The 
" Hunting Galliard," ii. loo, has a reminder of the melody 
of "Barafostus' Dreame" which immediately precedes it. 
It is worth playing up to line 2 of p. lol. Tomkins's 
" Grounde," ii. 87, is of some interest. It has already been 
described above, p. 188. Generally, the parts without 
"runs" are worth playing. 

The Pavana, ii. 51, is very good indeed, and the portions 
without agremens on pp. 51, 52, 54, are well worth playing 
for their harmonies. See above, p. 1 6, where it is quoted 
in its plainest form. 

" Worster Braules," ii. 269, has a capital tune and good 
variations. 

8. John Munday provides 5 pieces. " Munday's Joy," 
ii. 449, short and quaint;* "Bonny sweet Robin," i. 66 
(Hamlet ; and see above, p. 88), set also by G. Farnaby, ii. 
77, whose work is altogether better stuff, see ii. 77, ?8,f 
80 (line 2). The first page of Munday's may be played. 

* In bar 2 the minim in R. H. should be B G|, and the quaver F preceding 
it in the alto will require a sharp. 

f I suggest that on p. 78, 1. l, bar 3, the quaver rest in L. H. should be 
omitted as a mistake. The tenor E D may then be quavers, as in the MS. 



202 AN ELIZABETHAN VIRGINAL BOOK 

" Goe from my window," i. 153, is uncertain, as it is 
almost the same as i. 42, where Thomas Morley is said to be 
the author. The Fantasia on the Weather, i. 23, is rubbish.* 
But see above, pp. 162, 163, concerning " fingering." 

9. Martin Peerson (or Pierson). By whom are 4 pieces, 
not the worst in the book by any means : i. 359, " Alman," 
an excellent tune, and sensible variation; ii. 238, "Piper's 
Pavan" (Captain Piper). See Bull, ii. 242, for "Piper's 
Galliard? which, however, has nothing in common, and is a 
transcription of Dowland's familiar madrigal, " If my com- 
plaint could passion move." 

"The Fall of the Leafe," ii. 423, and "The Primerose," 
ii. 422, have pretty tunes, and are worth hearing. These 
two little pieces of Peerson are in excellent company, as 
some particularly good short dances, etc., are to be found 
from ii. 412 to 421, as follows : 

(a) The Duchesse of Brunswick's Toye (Bull), ii. 412. 

(b) A Toye (Anon.), ii. 413. 

(c) Three Corantos (Anon.), ii. 414, 415. 

(d) A first-rate and very difficult "Gigge" (G. Farnaby), 
ii. 416 (see above, p. 45). 

(e) A Toye (Anon.), ii. 418. 

(/) Galiarda; a beautiful tune (G. Farnaby), ii. 419. 
(g) A Toye; see likeness to Bull's p. 412 (G. Farnaby), 
ii. 421. 

10. Robert Johnson. Four pieces (2 arranged by Farnaby). 
Two Almans, one in the Dorian Mode, ii. 158; the other, 
ii. 159, certainly not so. For the latter, see above, pp. 137, 
138. A beautiful Pavan (set, however, by G. Farnaby) is 
that at i. I41.f 

* The first note should be A, not G. 

f In R. Johnson's Pavan, i. 141, 1. 4, bar i, the final crotchet G in L. H. 
should be omitted ; and in the next bar the hypothetical crotchet rest should 
be replaced by the crotchet C, which follows it in the tenor part. 



AN ELIZABETHAN VIRGINAL BOOK 203 

11. William Inglott. Born 1554. Organist of Hereford and 
Norwich. Died 1621, and is buried in Norwich Cathedral: 
ii- 375> a l n g " Ground " in Galliard time ; ii. 381, variations 
on " The Leaves be greene ; " tune mostly in bass or tenor ; 
see var. II, p. 383, 1. 2, bar 2, where a semibreve, A, is 
missing in the Alto. Good work ; better than some of the 
more pretentiously difficult pieces. 

12. Peter Philips. Besides his 9 Madrigal movements, 8 
of which are arranged from the work of four Italian com- 
posers, there are in the Fitzwilliam Virginal Book lo original 
pieces by Philips; two Fantasias (Mixolydian, i. 335;* 
Ionian, i. 352); two Galliards (Mixolydian, i. 351; Dorian, 
doubly transposed, i. 296). The Mixolydian one is 
remarkable in its rhythm. The Pavana and Galiarda 
"Dolorosa" (? Tregian), dated 1593, i. 321 and 327, 
are particularly good, played " plain." The Galiarda 
is truly one with the Pavana. With the exception of 
section i, the "Pavana Pagget (? Paget)," i. 291, is 
dull. 

The " Passamezzo " Pavan and Galliard, i. 299 (date 1592) 
and i. 306 (with Saltarello at end of Galliard), are good 
and worth playing. 

The Mixolydian Pavana, i. 343, is dated 1580, and is 
noted as " The first one Philips made." 

The interesting connection of the Pavana and Galiarda 
"Dolorosa" with the mysteries of "short octaves," has 
already been fully dealt with above, see p. 153, etc. 

13. /. P. Siueelinck (of Amsterdam). 4 pieces. 

(a) Pradudium Toccata, i. 378. A good example and 
worth playing through. More a Ricercar in places than a 
Toccata. See lines I and 2, contrary motion of answers in 
imitation. Vigorous passages for L. H, p. 382 and 

* See ii. 406, Byrd's Fantasia, on the same subject. 



204 AN ELIZABETHAN VIRGINAL BOOK 

(specially) 383 (see above, p. 192). See also L. H. passages, 
ii. 153, lines I, 2, 3. 

(^) Fantasia, ii. 297 (Dorian transposed). Contrary motion 
of answer again, in all four parts, during the whole piece. 
The "augmented" subject will be found pp. 298, 299 
(contrary motion in Bass, line 4), and in " Tenor," p. 300, 
line 3, last semibreve. Once more in original shape on 
p. 303, line 4, tenor, on a Pedal in the Treble part. A 
good piece, worth playing. 

(c) Psalme [140], ii. 151. Five variations on the Plain 
Song, in 2, 3, and (at end) 4 parts. Rather dry, but not so 
dry as Talk's and Bull in this kind. Line I, etc., notice J. S. 
Bach's trick, of making the Counter Subject out of the Canto 
Fermo, already in full work, e.g. : 

J. P. SWEELINCK. 






Counter Subject, 



f r r r T~f~r-r~ rrr-f 





i 


-^ 




.^ 






ff ft- 


t 


^T 


i Q 


m 

F 
i 


[ [ 







etc - 



Also, the 4th variation contains the subject in diminution 
(treble), imitated by the tenor in augmentation. 

(d) Ut, re, mi, fa, sol, la; ii. 26, in 4 parts, dated 1612. 
Fully described elsewhere, see p. 107, etc. 

The various spellings of Sweelinck's name are somewhat 
comical. In (a) he is described as " Jehan Pieterson 
Swellinck"; in () the form changes to "Jhon Pieterson 
Sweeling," "Organista a Amstelreda " ; in (c) we have " Jehan 
Pieterso Swelling"; and in (d) the "Pieterson" is spelt 
"Peterson." 



AN ELIZABETHAN VIRGINAL BOOK 205 

Thus the accepted spelling "Sweelinck" does not appear 
in the Fitzwilliam Book. 

14. Thomas Warrock (organist of Hereford, 1586). A Pavan 
and Galliard, i. 384, 388, is all that we have by this writer, 
but although small in quantity, it has considerable interest 
with regard to the growth of the notion of " Key," as 
opposed to the idea of " Mode." See above on this question, 
pp. 119, etc., 125, etc., 127-140. 

Thomas Warrock was a remarkably original writer. The 
two pieces mentioned deserve careful attention. He shows 
modern tendencies of key and of harmony, besides a general 
smoothness quite foreign to England, the Netherlands, or 
even Italy. 

15. Ferdinando Richardson (Sir Ferdinand Heyborne). 8 
pieces, all Pavans and Galliards, with " variations" on them. 

None of these are worth playing, in comparison with the 
other things in the book. 

i. 27 and i. 32 are Pavan and Galliard en suite, and the 
Galliard really is the same tune as the Pavan, according to 
the recognised rule. See above, p. 114, or compare Bull's 
Pavan and Galliard, " St Thomas Wake," in Parthenia 
(quoted in " Shakespeare and Music," 201, 202). 

1 6. Richard Farnaby (son of Giles Farnaby). 4 pieces, 
ii. 374, a " Duo." Dry and short, ii. 263, " Fayn would 
I wedd," a good example of simple musicianly variations on 
a typical melody 400 years old. (Rather like his father's 
"Quodling's Delight," ii. 19.) 

This one page (ii. 263) is valuable to a student. In it are 
seen (a) the ancient Dorian or ^Eolian burden or bagpipe 
bass of la sol la mi, which can be sung with the first half of 
each section, (b) The ordinary Dorian harmonies, with the 
accustomed "relief" in what we call "the relative major." 
(c) The obvious way of the invention of " Canons," by 
accident. See, for instance, the melody of section 2, where 



206 AN ELIZABETHAN VIRGINAL BOOK 

each of the first three pairs of bars harmonises with the 
other ; or, the first four bars of all three sections, which 
will fit together well. The latter example shows clearly 
how the discovery of Canon might be made. 

ii. 162, "Nobody's Gigge." An excellent piece. See 
above, p. 46. 

ii. 494, Variations on " Hanskin." The tune is associated 
with Shakespeare's song, " Jog on," in the Winter's Tale. 
See "Shakespeare and Music," 192, for a different version of 
the first half, and indeed of the whole tune. "Hanskin" 
is the last piece in the Fitzwilliam Virginal Book, and is 
numbered CCXCVII. These four pieces show Richard 
Farnaby to have been a man of great gifts, and a worthy 
son of his father. 

17. Edward Johnson. 3 pieces. ii. 366, " Jhonson's 
Medley." This is really a Pavan and Galliard in one piece, 
with the usual variations. It has, however, other curious 
features, e.g., the approximation to the homophonic madrigal, 
at the top of p. 367 (including the last chord of the previous 
page), and in the passage labelled 3, beginning on the fourth 
line of that page. ii. 436, Pavana " Delight," with Galiarda 
ii. 440 following, en suite, " set by Byrd." The Galiarda 
(see above, p. 115) is properly allusive of the Pavana in all 
three sections. On p. 438 is again seen the homophony already 
noticed in the " Medley." 

1 8. Francis Tregian (who may have written out this collec- 
tion in the Fleet Prison, between 1609 anc ^ 1619). * 4 J 5 
" Heaven and Earth," may be by Tregian. It is an orna- 
mented version of some ecclesiastical melody (?). i. 226, 
This is probably not by Tregian, in spite of its title " Treg 
Ground," as in the Nevell Book it is called " Hugh Ashton's 
Grounde." Also, it is " set " by Byrd, so, after all, very little 
can be by Tregian himself. 

For i. 321 and i. 327, see under Peter Philips. 



AN ELIZABETHAN VIRGINAL BOOK 207 

Thus Tregian's name is of no importance, except as that 
of the possible compiler and writer of the Fitzwilliam Book. 

Other names of Composers in the book are Tisdall, Oyster- 
mayre, Oldfield, Pichi, Galeazzo (of these nothing seems 
known). Again, Strogers, Blitheman, Rosseter, Parsons, 
Marchant, Hooper, Harding, are little more than mere names 
to us. The Italians, Caccini, Lasso, Striggio, Marenzio, 
whose Madrigals appear in Philips's arrangements, can hardly 
be considered here. 

19. John Dowland is almost the only other well-known 
musician we have not yet noticed. There is, however, very 
little of his work in the book. One piece is the famous 
" Lachrymae Pa van," which was first published in 1600, and 
is "set" in the Fitzwilliam book by Byrd, Farnaby, and 
Morley, ii. 42, 472, 173. Also see the note on "Lach- 
rymae," in the printed Fitzwilliam Virginal Book, ii. p. vi., 
and the Musical Antiquarian Society's edition of Dowland's 
" First Set," etc. (1597), p. 2, where the dramatic quotations 
alluding to it are given. It is worth playing "plain," as 
evidently its popularity must have been considerable. 

ToDowland's name we must also add "Piper's Galliard," 
ii. 242 (Bull), which is simply a version of Dowland's mad- 
rigal (publ. 1597), " If my complaint " (" First Set of Songs," 
No. 4). Further remarks will be found above, pp. 27-30. 

20. Thomas Talks y the best-known name of the time, at least 
to modern ears. Only 2 rather dry ecclesiastical pieces, 
i. 427 and ii. I. The latter is dated 1564, and is terribly 
long (about eleven minutes). It is almost worth playing to 
remind one of the unwisdom of judging a man by one speci- 
men of his worse work (compare, perhaps, Browning's 
" Sibrandus Schafnaburgensis "). 

2 1 . William Tisdall. By whom 5 pieces (a mistake in the 
Index makes a Fantasia of the Galiarda) : ii. 276, a good 
Alman, worth playing "plain"; ii. 486, Galiarda, quite 



2o8 AN ELIZABETHAN VIRGINAL BOOK 

good : ii. 278, Pavana Chromatica, called " Mrs Katherin 
Tregian's Pavan." Seems to recognise a " Key" of E major 
and minor (see above, pp. 128-9); u - 3^ Pavana, 
" Clement Cotton," quite good ; ii. 307, Pavana, Dorian 
transposed or (?) G minor. 

22. Edmund Hooper. Two short dances (Index gives only 
one)i ii. 309, Alman, very excellent; ii. 312, Corranto, for 
which see above, pp. 31 and 35. 

23. Marchant. One piece, ii. 253, a good Allemand. He 
is said to have been a musician in the service of Lady Arabella 
Stuart, King James I.'s cousin, who died in 1615. 

24. Nicholas Strogers. One piece, Fantasia, i. 357. 

Full of queer "false relations"; but a capital piece of its 
kind, by a good man. See above, p. 63. 

An evening service in D minor, by Strogers, was rescued 
by my father, John Naylor, the late organist of York Minster, 
from the MSS. of the Cathedral Library, and formed part of 
the repertoire. 

25. Jehan Oystermayre. Galliard, ii. 405. A charming 
piece. (See above, pp. 22 and 122). I rely on this as an 
example of modern key effect. See sec. 2. 

26. Philip Rosseter. ii. 450, A good Galliard, " set " by 
G. Farnaby. Worth playing throughout (something is want- 
ing in bar 2).* 

Rosseter published "Ayres" in 1601, and "Consort 
Lessons" in 1609. 

27. Parsons (? Robert or John), ii. 135, "In Nomine." 
Not very interesting. 

28. James Harding, ii. 47, Galiarda "sett foorth by" 
Byrd. Very good. 

* In the 3rd section of this Galliard, ii. 452, top line, last bar, the 2nd, 3rd, 
and 4th crotchets in the left hand ought to be C D E F, as may readily be 
seen by comparing the four previous bars. 



ADDENDA 

Nicholas Strogers, see above, p. 208. The "Gostling" MS. 
Choir Books at York contain, besides the " service " named above, 
an anthem by Strogers, " O God be merciful." 

Edmund Hooper, see above, p. 208. The same MSS. at York 
contain three anthems by Edmund Hooper (1553-1621), viz., "O 
Thou God Almighty," Behold, it is Christ," and " Teach me, O 
Lord." 

John Doivland, see above, pp. 26-30. Mr Barclay Squire's 
interesting article in the "Musical Times" (vol. 37, Dec. 1st, 
1896) gives a curious letter from Dowland to Sir Robert Cecil, 
dated Niirnberg, Nov. 10, 1595? concerning the '* villainy of 
these most wicked priests and Jesuits," and warning Queen Elizabeth 
to beware of them. 

Mordents, and the sign for them, used in the Fitzwilliam Virginal 
Book, see above, pp. 32, 85, 182-3. 

I find that Dr James Nares (1715-83) used the same sign for the 
mordent, viz. a double stroke, more or less sloping, but with this 
difference, that it was written over the note, not across the tail. The 
sign occurs frequently in the MSS. in the Fitzwilliam Museum at 
Cambridge, of organ and harpsichord music by Nares, and in the 
engraved copy of his " Lessons " for Harpsichord. The dates of 
these are marked in the handwriting of Lord Fitzwilliam, 1767 
and 1768. 



209 



INDEX 



./EOLIAN, see Mode. 

Allgemeine Deutsche Biographic, 22 

note. 

Almain, see Alman. 
Alman, a dance, 7; 31 (Hooper's); 

related to Pavan and Brawl, 32, 34 ; 

Elizabethan compared with Bach, 

38 ff. ; related to French Branle, 41 ; 

form of, 119, 1 20 ; " key" in, 125 ; 

other examples, Anon. 131, R. 

Johnson 137. 
"Amarilli, mia bella," Caccini's 

madrigal, arranged by Philips, 

193-4- 

Ammerbach (1571), on fingering, 158. 

Anagrams, on Tabourot and Dow- 
land, 28. 

Anonymous composers, 200, best 
works named. 

Answer (fugal), contrary motion, by 
Sweelinck, 61 ; dux and comes, 65 ; 
false answer, 67, 70 ff.; real and 
tonal. 72, 73 ff. 

Arbeau, 15, 20-22, 25, 28 ; galliard, 
etc., 49; also see " Orcheso- 
graphie." 

Army Life, piece by Byrd, 3. 

Arpeggios, in Bull's " Salvator 
Mundi," 181. 

Arrangements for virginals, of ' ser- 
vices ' and anthems, 2 ; of madri- 
gals, 5, 6, 130 note, 169 ; examples, 
193-6. 

Ashton, Hugh, "ground" varied by 
Byrd, 1 88. 

" Ay me poor heart," arrangement of 
G. Farnaby's canzonet, 130 note, 
169, 197. 

Ayres, Dowland's first book of, 29. 

310 



B 



BACH, Emmanuel, fingering in " Ver- 
suche," 159. 

Bach, J.S. , Capriccio on the Departure 
of a Friend, 96 ; B. and " Tempera- 
ment," 102-3 ; " key," 102 ; on 
fingering, thumb, etc., 160 ; pre- 
decessors (organ), 169 ; alteration 
of plain song, 186; uncertain 
rhythm in corantos, 201. 

Bagpipe, see Bass ; also 144, Bull's 
Galliard. 

Baldwin, John, 2, copies " Lady 
Nevell's Book." 

Barafostus' Dream," 83, 84, 86, 89, 

200. 

Barnfield, Richard, sonnet on Dow- 
land, 27. 

Bars, irregular use of, 63, 840, 113 
note, 133^, 187. 

Bass, Drone, 141-148, 205 (16). 

Basse-dance, 20, 21, 49 ; steps of, 50. 

" Bebung," on clavichord, 161. 

Beethoven, 4 ; programme sonata, 

97- 
" Bells, The," romantic piece by 

Byrd, 5; described, 97-8. 
Bevan, 2, " services " arranged for 

virginals. 

Binary Form, 119, 126. 
Bizet, see Carmen. 
Blitheman. organ piece, 169. 177; 

harmonies implied in his "Gloria 

tibi Trinitas," 185 : 207. 
" Bonny sweet Robin," see Robin. 
Bourdon, Burden, see Bass. 
Brahms, 4; development of themes, 

124. 
Brawl (Bransle), related to Alman 

and Pavan, 32, 41 ; Tomkins' 



INDEX 



211 



Brawl, 41, 20 1 ; steps of, 41 ; 
peculiar forms of, 42. 

Bremner, R., and the Fitzwilliam 
Book, 9. 

Broken octaves, sixths, and thirds, 
examples by Bull, 191. 

Bull, Dr John, 2, 3 (Parthenia), 6 
(Antwerp), 8 (modulation), 9 
(GreshamProfepsor), 13,21 (suite), 
24, 25 and 197 (arranges madrigal) ; 
remarkable pedal, 64 ; subject and 
answer, 67 ; fantasia, 75, 76 ; 
"Walsingham," 86; Hexachord, 
101-6; Bull's characteristics, 105, 
1 27-8 ; St Thomas Wake, 115, 1 30 ; 
music of his "Juell," 146; his 
fingering, 163, 165-6; his organ 
pieces, 169; specimen of his 
"Christe Redemptor," 176-7; of 
his" Gloria tibi Trinitas" 178-180 ; 
his fondness for curious "propor- 
tions," 124, 178; " Sesquialtera," 
example, 180 ; " Salvator Mundi," 
extracts from, 181-3 5 his Preludes, 
189-90; examples of difficult pas- 
sages, 190-1 ; best works reviewed, 
198. 

Biilow, von, on fingering, 157. 

Burden, see Bass. 

Burney, Dr, and the Fitzwilliam 
Book, 9. 

Byrd, William, 2 (Lady Nevell), 3 
(Parthenia), 13, 24, 25, 26 (Lach- 
rymas), 30, 33 (-^ time), 43 (Gigg), 
48 (Lavolta); iantasias, 65; fan- 
tasias, 70 ff. ; Miserere, 74 ; fan- 
tasia, tonal answers, 75, 77 ; 
variations on "Walsingham," 81- 
83; 85 ("The woods so wild"); 
"O mistress mine," 89; "Jhon 
come kisse me now," 90-92 (music 
and words), 125 (modern tendency); 
93 (see Gibbons); "The Bells," 
97; Byrd as Romanticist, 93, 98; 
Hexachord, 105 ; uses old song in 
"Hunt's up," 125; his charac- 
teristics, 93, 98, 105, 128 (contrasted 
with Bull); his fingering, 167; his 
organ pieces, 169, also see his 
" Miserere," 74 ; variations on 
"Tregian's Ground," 188 ; his Pre- 



ludes, 189-90 ; best works reviewed, 
198-9. 



CACCINI, 6 ; his Amarilli, mia bella," 
quoted, with Philips' transcription, 

J 93-4" 

"Callino Casturame," 7, 85, 87-8. 
Canon, in variation by Tomkins, 189 

invention of canon, 205-6. 
Canto Fermo, pieces on, see Organ 

Music. 

" Carman's Whistle, The," 125. 
" Carmen," song from, with drone 

bass of ancient type, 144 note. 
Chappell, on Fitzwilliam Book, 10 ; 

words of songs, 15, 79, 83, 88, 89, 

9> 93- 

" Chirping of the Lark, The," 15. 

Chord passages, Bull, 165, 190. 

Chords, source of, as harmony, 
141-8. 

"Christe Redemptor," 116; quoted, 
176. 

Chromaticism, i6th century; Tom- 
kins, 18, igd; Tisdall, 128; G. 
Farnaby, 99 ; Tallis, 174. 

Cinquepace, dance ; steps of, 52. 

Clavichord, action and touch, 161. 

" Come, Holy Ghost," see " Salvator 
Mundi." 

Compass of virginals, I33/., 150-2, 

1 S4-S> 57- 
" Conceits " in Fantasias, 59-63 ; 

mystical conceit by Philips, 62-3. 
Congee, dancing term, 50 note. 
Coperario, 7 note, 1 30 note. 
Coranto (courante), 7, 15 ; Hooper's. 

35 ; steps, 36 ; Shakespeare, 36, 

37 ; Selden, 37 ; uncertain rhythm, 

201. 
Corrections in text of Fitzwilliam 

Virginal Book, 75, 147, 191, 201 

notes, 202 notes, 203 (n). 
" Cosi moriro," by Luca Marenzio, 

arranged by Philips, 193 ; 195. 
Cosyn, Benjamin, his Virginal Book, 

I, 2. 
Couperin (1717), on fingering, 159- 

60. 



212 



INDEX 



Croce, Giovanni, 30 ; friendship with 
Dowland. 



DANCES, suite of, 4; in Masque, 7, 
and notes; in Fitzwilliam Book, 
4; classes of, 14-15; instruments 
used in, 20-21 ; dances sung, 15, 
20, 55 ; dances named after persons, 
etc., 24, 25 ; list of useful short 
dances, 202. 
Dannreuther's explanation of " slide " 

confirmed, 166-7. 
Daphne," song, 79, etc. (variations); 

89-90. 

Dates given, of pieces in Fitzwilliam 
MS., 6, 13, 25 ; list of such dates, 
130 note; extracts from earliest 
dated piece,Tallis' " Felix namque," 
173-6. 
Development, in form, 114 n. ; 

sequence, 121 ; 124. 
Dominant harmony, see Tonic; also 

117, 119, 120, 125, 126, 138. 
Dorian, see Mode. 

Dowland, J., 24-6 (Lachrymas), 28, 
29 ; his life and wanderings, 26-30 
madrigal arranged, 169, 197, 200 
work renewed, 207 ; 209. 
Drum, in i6th century dance, 20. 
Duet for two virginals, 199. 
Duke of Brunswick's Alman, form of 

120. 

Dump, an English dance, 8. 
Dumpe, The Irish, 7, 8. 
" Durchfiihrung," or free section in 
binary form, 117,118, 1 6th century 
example. 
Dvorak, modal relics, 127. 



EDWARDS, Richard, 8. 
"Ein'Feste Burg," 116. 
Elizabeth, Queen, a player on th 

virginals, i, 9. 
"English" fingering (i599-i?57 

158-60. 
Enharmonics, see Notation, Key, als 

1 06, 127 (Bull). 



ALKENER, Rudolph (1762), and 
" English" fingering, 159. 
Fall of the Leafe, The," set by 
Peerson, 202. 

alse relations; Strogers, 63, 64; 
Farnaby, Tallis, 68. 
'ancy pieces, i6th century, 5 ; music 

of Farnaby's Dreame, 94; 94-99. 
'antasias (Fancies), fugal, 5, 57-78, 
102; in Shakespeare, 59; writers 
of, 60; Munday's Fantasia on the 
weather, 96, 202 ; correction in 
" The Fantasia on the Weather," 
202 note. 

Farnaby, Giles ; romantic pieces, 5 ; 
Maske, 7 and notes; Pavans by, 
24 ; Galliard by, 47 ; setting of 
< LachrymaG,' 24 note ; 25, 26, 30, 
31 ; curious Gigue, 45 ; G. Far- 
naby's Fantasia, 67-70 (music); 
"Daphne," 89, 90; variations on 
"Daphne," 79-81 ; his "Dreame," 
94, 98; his "Conceit," 96; his 
"Robin," 201; King's Hunt and 
Spagnioletta, etc., 125; Canzonet 
arranged, 130 note and 197; his 
variations on a ground bass, 
187; prelude, 189, 192; belt 
works reviewed, 199-200. 
Farnaby, Giles and Richard, father 

and son, 46. 

Farnaby , Richard, " Nobody's Gigge," 

46, 1 1 6, 136; "Fayne would I 

wedd," 136, 200; work reviewed, 

205-6. 

"Fayne would I wedd," 136, 200, 

205. 
" Felix namque," plain song ; set by 

Tallis, 173-5. 
Fifths, consecutive, Byrd, 440, 

Sweelinck, 109. 

Fingering (1571, I599> l6 5*> 1697 
,735, 1741), .58 ff., "English" 
and " German," 158-60; Falkener, 
Heck, Clementi, Dussek, 159; E. 
Bach, scales, 159; Coupenn, 
examples, 159-60; E. Bach on 
passing of fingers and of thumb, 
160; J. S. Bach on the same, 160; 



INDEX 



213 



examples, Munday, 162-3 5 Bull, 
163, 165-6; Byrd, 166-167; Anon., 
164; von Biilow, 157 ; Repetition, 
161 ; examples of, 164, 165; other 
examples, 162-8, 189, 190; curious 
use of the figures, 166-7. 

Fitzwilliam, Lord, and the Fitz- 
william Book, 9. 

Fitzwilliam Museum, Cambridge, i. 

Fitzwilliam Virginal Book ; con- 
tents, 3 ff. , 57; classes of pieces, 
4, 14, 15 ; date of MS., i, 6 (dates 
of Philips' pieces) ; list of dates 
in MS., 130 note; description of 
MS., 8; history of MS., 9 ff. 

Flat ([^), use of, in Modal Transposi- 
tion, 122, 185, also see Signature. 

Form; external, 114-16; internal, 
116-123; Binary, 117, 137; free 
section, 118; sequence, 121 ; 
general tendencies in i6th century, 
123, 126, 138. 

" Fortune," song, 85 ; account of, 
88. 

Foster, Will, his Virginal Book, 2. 

France, music in, i6th century, 29, 

3- 

" Free " section in binary form, 117, 
118, 138. 

" Freno," part of a madrigal of 
Marenzio, arranged by Philips, 
195. 

Frohberger, programme music, 97. 

Fugue, 5 ; 57-78 ; examples by 
Sweelinck, Philips, Strogers, Bull, 
Farnaby, Byrd, 61-78 ; history of 
the name, 58 ; Real, and Tonal, 
71-78; on Hexachord, Sweelinck, 
107-109. 

Fuller, on Dowland, 28. 



GABRIELI, organ music, 169. 

Gaffurius, on "proportions," 124. 

Galeazzo, composer of a Prelude, 189, 
207. 

Galliard ; with Pavan, 4; made out 
of Pavan, 21, 34, 114 ff. , 200 (5^, 
205 (15); steps of, 21 f. ; music of 
Oystermayre's, 22-24 ? " Captain 



Piper's," 25 f. ; Galliard en suite, 
E. Johnson, 115 (music) ; Richard- 
son, Bull, Philips, Byrd, 115; 
Morley, 200 ; Galliard en suite with 
Prelude and Pavan, n6(Parthenia); 
Bull's, with bagpipe harmony, 144; 
Farnaby's, 202. 

Gam-ut, 100 note. 

Germany, music in, i6th century, 

3- 
"German" fingering (1571-1774), 

158-160, etc. 

Gibbons, Orlando, 2, 3, 86 and 93 
(Woods so Wilde), 130 note ; work 
reviewed, 201. 

Gigue ; usual form of, 42 ; music of 
Byrd's, 43-4; various forms of, by 
Bull, Byrd, G. and R. Farnaby, 
44-46 ; Bull's " myselfe," 44, 95 ; 
Richard Farnaby's "Nobody's 
Gigge," 46, in, 1 16- 

Giraldus Cambrensis (end I2th cen- 
tury) on bemol, 146. 

Gloria tibi Trinitas," set by Bull, 
Parsons, Blitheman, 177; speci- 
men of Bull's setting in ^ time, 
178-180; harmonies of Blitheman 
and Bull, 185 ; prelude to, 190. 

" Go from my window," song, arr. 
Morley and Munday (?), 85, 90, 
202 ; on "burden," 145. 

Gray's Inn, Masque, 7, 130 note. 

Graye's Galliard, Sir John, 25. 

" Grounde."pieces on, 186-88; extract 
from Giles Farnaby, 187; Byrd's, 
on Tregian or Hugh Ashton, 188; 
Inglott's, 188, 203 ; Tomkins', 188, 
201. 

H 

" HAMLET," see Ophelia. 

Handel, material borrowed from 
previous centuries, 62 ; false 
answers, 67 ; Sweelinck, 108. 

" Hanskin," song, 86; R. Farnaby's 
setting, 206. 

Harding, James, 207 ; work re- 
viewed, 208. 

Harmonic basis; .flSolian, 135, 145, 
147 ; Mixolydian, 144. 



214 



INDEX 



Harmony, early Western, 141 ff. ; 

of 1 3th century, 141 f. ; connection 

with Drone bass, 144-148. 
Harmony, growth of modern, i lo-iu, 

118-120; Tisdall, 129; Farnaby, 

133-5; Oystermayre, 139; causes 

alteration of plainsong, 183-4, 185, 

186. 
Harmony, Modal, 131-4 (music), 

135-6 (music), 113 note (music); 

contrasted with key harmony, 

138-9 (music). 
Haye, dance ; tune and steps, 54 (for 

explanation of steps, see p. 50). 
Hawkins, Sir J., and the Fitzwilliam 

Book, 9, 43. 
" Heaven and Earth," Tregian's (?), 

85, [? Te Deum or Sanctus], 186. 
"Heigh ho, nobody at home," old 

catch, on burden, 142. 
Hexachord, pieces on the, 4, 8, 

100-110; Bull's, as evidence of 

modern modulation, 127; Byrd's, 

127-8; Sweelinck's, combined with 

Fugue, 107-9. 
Heyborne, Sir F., Pavan and Galliard 

en suite, 34; work reviewed, 205. 
History of Music, The Oxford, 3, 193. 
; Ho-hoane, The Irish," see " Ochone." 
Hooper, 7; Alman, 31 ; Coranto, 35; 

key with 3 sharps, no ; 207 ; work 

reviewed, 208 ; anthems, 209. 
Humorous Songs, i6th century, 90-92. 
"Hunt's up," Byrd's, 125 (No. 10), 

199. 

I 

" IF my complaint," madrigal ar- 
ranged by Bull for virginals, 25-6, 
169. 

Inglott, William, his "Galliard 
Ground," 188; work reviewed, 
203; correction of text, 203 (n). 

"In Nomine," see "Gloria tibi 
Trinitas." 

Intonation, just, etc., 101 note, 103 
and note ; also see Temperament. 

Irish music in Fitzwilliam Book, 7, 
and see "Ochone." 

" Irishe Ho-hoane, The," see Ochone. 

Italian fingering, i7th century, 158. 



Italy, music in, i6th century, 30. 
Ionian, see Mode. 



JKNKINS, John, Fancy for viols, 102 

(modulation). 
" Jesu dulcis memoria," see " Christe 

Redemptor." 
" Jhon come kiss me now," humorous 

song, tune and words, 90-92 ; set by 

Byrd, 125 (No. i). 
Jig, see Gigue. 

" Jog on," Shakespeare's, see Hanskin. 
Johnson, Edward, Pavan and Galliard 

en suite, 34; music of, 114-115 ; 

work reviewed, 206. 
Johnson, Robert, Alman, 1 1 1 (example 

of modern key), 202; music of the 

same, 137; work reviewed, 202; 

correction in text, 202 note. 
" Juell, Dr Bull's" (Jewel), 164, 198. 

K 

KEY, signs of, in i6th century, 8, 25 
(four sharps, by Tisdall), 71 (fugal 
answers), 71 (early signatures); 
list of passages, no-ni. 

Key, enharmonic change of, 102-3; 
origins of, 101, no f., 125. 

Key, C major, in, 125; G major, 
119, 126; G minor, 126, 129; 
D major, 119, 125, 126, 138; 
A major, 31, no; E major, 131; 
E or B major, 103 ; B^ major, no; 
E and B major, Tisdall, in relation 
to Mixolydian mode, 129. 

Key, List of pieces shewing "key," 
125-6; characteristics of "key" 
contrasted with those of "mode," 
I 35"6> 138; example of absence 
of "key" effect, 187; also see 
Modulation. 

"King's Hunt, The," G. Farnaby, 
125 ; fingering, 167. 

Kuhnau, Biblical sonatas, 96. 



LA, Sol, La, Mi, harmonic basis, 135. 
142. 



INDEX 



215 



"Lachrymae," by Dowland, 24 and 

note, 25, 26, 169, 200. 
Lasso, Orlando, 6, madrigals arranged. 
Lavolta, method of dancing, 37 ; 

Morley's, 48; steps of, 53; old 

binary form, 118-119; B yd, 126; 

Morley, Modern Binary, 126 (No 

I 7> 
"Leaves be Greene, The," set by 

Inglott, 203, tune in bass or tenor. 
Lumley's Paven, Lord, 24. 
Lute, as song accompaniment, 29. 



M 

MADRIGALS, arranged for virginals, 4; 
Peter Philips' transcriptions of 
Lasso, Striggio, Marenzio, Caccini, 
6; Bull's of Dowland, 25-6; 112; 
Giles Farnaby's of his own, 169; 
examples from Philips' arrange- 
ments, 193-6. 

Maies (1741), on fingering, 158. 

" Mai Sims," form of, 117 (music). 

March, The Earl of Oxford's, by 
Byrd, 126 (No. 20). 

Marchant, composer, 42, 207 ; ac- 
count of, 208. 

Marenzio, 6 (madrigal arranged) ; 
30 (corresponds with Dowland); 
Philips' transcription of " Cosi 
moriro," 193, of "Tirsi," 195-6. 

"Martin said to his man," humorous 
song, 90. 

" Maske of Flowers," 7 note, 130 
note. 

Masque, dances for, 7 ; order of, 7 
note 2 ; Pavan in Masque, 20 ; 
Lady Zouche's, 126. 

Mattheson (1735), on fingering, 158. 

" Mayden's Song, The," 199 and 
note. 

" Micrologus " of Ornithoparcus, 27 
note, 29. 

" Miserere," plain song, set by Byrd, 
74, 186 ; by Bull, 186. 

Mixolydian, see Mode. 

Mode; Dorian, 61 (example), 62, 
122, 202 ; Dorian bass and har- 
monies, 205 (16); Mixolydian, 69 
(example), 72 (example), 103 ; 



JEolian, 70 (Byrd, fugal answers), 
78 ('1 allis, fugal answers), 130 
note, 142 and 143 (examples), also 
200, 205 (16); Ionian, 140, 200; 
characteristic harmonies of Mode, 
in, 119, 122, 125 (Nancie), 138 
(example); effect of Mode on the 
development of modern music, 1 23 ; 
"Mode" mixed with "Scale," 
131-2 (example), 133*,- "Mode" 
contrasted with "Key," 129, 138-9; 
-/Eolian harmony, 135, 136, 142-3, 
185; Eolian 7 th, "flat," 173; 
Relics of Modes surviving, 127; 
also see Transposition. 
Modulation, enharmonic, 8, 106 



Modulation of Key, early history of; 
Jenkins, 102; Bull, 105-6; Byrd, 
no; Oystermayre, in; "Wal- 
singham," 120; Byrd contrasted 
with Bull, 127-8; Tisdall, 129; 
Oystermayre, 139. 

Montegle's Pavan, Lady, 24. 

Mordents in Fitzwilliam Book, 32, 
85 ; examples from Bull's "Salvator 
Mundi," 182, 183; in Nares, 209. 

Morisco, dance, 55 ; the King's 200. 

Morley, T., 2; on galliard, 21; his 
setting of "Lachryma*," 24 note, 
26, 200; publisher, 29; 30; on 
eight-bar strains, 32 ; on the Volte, 
Courante, etc., 37; Tonal "answer" 
in his Fantasia, 76 ; "Go from my 
window," 85 ; modern tendencies 
of key in "Nancie" and Alman, 
125 ; modern tendencies of Modu- 
lation and Form, in Lavolta, 126; 
best works reviewed, 200-1. 

Munday, John; "Go from my win- 
dow," 26, 85; "Munday's Joy," 
95 ; his Fantasia on the Weather, 
96; fingering, 162-3; corrections 
in text, 201, notes; work reviewed, 

2OI-2. 

Muscadin, dance, 15 ; also as ong, 
15; variations on, 55; modern 
tendencies of key, 125 (No. 4); 200. 

Music on the Continent, i6th cen- 
tury, 6, 29, 30. 
Musica Transalpina" (1588), 195. 



2l6 



INDEX 



Musical Antiquarian Society's publi- 
cations, 21, 26, 30. 

N 

NAMES of persons, attached to Dances, 

24-5; also see Ashton. 
" Nancie," set by Morley, 125. 
Netherlands, The, associated with the 

Fitzwilliam Virginal Book, 6, 10, 

I 3- 
NevelFs, My Lady, virginal book, 2, 

188. 
" Nobody's Gigge," tune quoted, 46 ; 

shewing key of C, etc., in; 

shewing formal repetition, 116. 
Notation of time signatures, C = |, 

81, 82, 119, 144; G = |, 170; 

<JJ = J^L, 178 ; list of instances, 187. 
Notation of enharmonics, 104, 106, 

131. 

Notation of " key," 31, 131-3. 
"Now God be with old Simeon," 

introduced in Byrd's " Hunt's Up," 

125 (No. 10), 199. 



" O MISTRESS Mine," 89, 125 (Key of, 
contrasted with Mode in "Woods 
so wild "). 

" Ochone, The Irish," 7, and photo 
of MS. facing p. 8. 

Oldfield, Thomas, composer of a pre- 
lude, 189, 207. 

Ophelia's songs, remains of in Fitz- 
william Book, 87, 88, 119. 

" Orchesographie," Arbeau's, on the 
Pavan, 20 ; on Passamezzo, 21 ; 
anagram on name Tabourot, 28 ; 
Alman and Bransle, 32 ; Pavan 
turns into Coranto, 34; steps of 
Courante, 36; "simples" and 
"doubles," 36; Branle des Sabots, 
41 ; Bas^e dance, Tordion, Galliard, 
Lavolta, 49-53, steps and tunes ; 
Round, 53 and 54 ; the Haye, steps 
and tune, 54. 

Organ music, i6th century, 5, 169 ff. ; 
founded on Plain Songs, 169 ; 
names of composers, 169; examples. 
Anon, on " Veni," 170; harmonies 



of the same, 171-2; extract from 
Bull's Fantasia on the same, 76 ; 
Tallis, "Felix namque," 173-6; 
Bull, "Christe Redemptor," 176-7 ; 
Bull, on " Gloria tibi Trinitas," in 
JJ^ time; 178-180; another setting 
of the same, with " Sesquialtera," 
1 80; specimens from Bull's " Sal- 
vatorMundi," 181-3 ; implied har- 
monies, Bull and Blitheman, 184-5. 

Ornaments, mordents, slides, 190, 32, 
166-7, JtBz-S* 20 9- 

Ornithoparcus, Andreas, 27 note, 29. 

Oystermayre, Jehan, 6 ; his Galliard, 
22; 22 note; modulation, in; 
"development," 122-3; modern 
"key" effect, 139; 207, 208. 

Oxford History of Music, 3, 193. 



"PARSIFAL," tradition of modal har- 
monies, 127. 

Parsons, organ piece, 169, 177, 185 ; 
207. 

Parthenia, 3 ; Bull's St Thomas 
Wake, 21 ; 115 ; 'suites 'of 3 move- 
ments, 116; Pavana, Gibbons, 130 
note; prelude to "Carman's 
Whistle," 168, 190 ; 201. 

Part-songs, i6th century, 29. 

Passamezzo (passymeasures pavan), 
21 ; and galliard, 33, 147. 

Patrons of music, 30 ; also see Names. 

'Pause' chord, final, 32, 36, 95, 

*3*~3> *37- 

Pavan, associated with Galliard, etc., 
4, 14; Morley, 21 ; Edward John- 
son's, en suite, 114-115; Tl6; 200 
(5 c.); related to Brawl and Alman, 
32 ; altered to Coranto, 15 ; Pavan 
by Tomkins, 16-19, 2O1 ? by Bull, 
modern key effect, 125 ; by Bull, 
bagpipe tune in galliard, 144; 
Pavana Chromatica, by Tisdall, 
103, 128-9; steps of, and manner 
of dancing, 20 ; instrumental music 
for, 21 ; " Lachrymas " Pavan, 24, 
and note; Robert Johnson's, 202. 

" Pawles Wharfe," in D major, etc., 
in, 119, 125 (No. 12); contrasted 



INDEX 



217 



with modal pieces, 133 ; music of, 

'34 5I99- 

Pedal bass, see Bass. 

Peerson, Martin, "The Primerose," 
147, 202 ; work reviewed, 202. 

Penna, Father L. (1696^), on finger- 
ing, 158. 

Pepusch, and the Fitzwilliam Book, 
9, 10. 

Percy's Reliques, 2, 87, 88. 

" Pescodd Time," see " Hunt's Up." 

Philips, Peter, arrangements of mad- 
rigals, 6, 169; specimens of, 
193-6; Philips and "short 
octaves," 153; and date of Fitz- 
william Book, 13 ; dates of his own 
pieces, 130. 203; his Pavans, 25, 
33 ; his Fantasia, 62 ; his Pavana 
Dolorosa, 203 ; his Passamezzo 
Pavan and Galliard, with Saltarello. 
203 ; list of his original pieces, 203 ; 
best work reviewed, 203. 

Pichi, an unknown composer, 207. 

Piper's Galliard, Captain; arranged 
by Bull from Dowland's madrigal, 
25 f.; 169. 

Piper's Pavan, arranged by Peerson, 

202. 

Plain Song, harmonies set, by i6th 
century writers: Anon., 171-2; 
Tallis, 173-5: Bull, 176-7, 178-80, 
180, 181, 182, 183, 184, 185; 
Blitheman, 185. 

Popular music, its influence on the 
classical school, 124; list of ex- 
amples, 125. 

Preludes, en suite, 4, 116; in Fitz- 
william Book, 169; account of 
them, 189-192; examples of 
technical passages; Bull, 190, 191 ; 
Sweelinck, iyi;Anon., 164. 

" Primerose, The," set by Peerson, 

202. 

Primitive Music," Wallaschek, 5. 

Programme music, i6th century, 96; 
I7th century, 96,97; i8th century, 
96-7 ; I9th century, 97. 

Proportion (time) ; Bull and Gaffu- 
rius, 124; Bull, strange combina- 
tion, 178; example of " Sesqui- 
altera," 180 ; relation to signatures, 
187. 



Put up thy dagger, Jemy," 199. 



" QUADRAN " Pavan, 24. 

Queen Elizabeth, i, 9. 

" Quodling's Delight," modal tune 
and harmony, contrasted with 
"Pawle's Wharfe," 125 (No. 12); 
not in "key" of A minor, 133-4; 
music of, 135-6; shows charac- 
teristics of ./Eolian mode, 136 ; com- 
pared with Rameau, 200. 



RAMKAU (1683-1764), 47, 200. 

Repetition in fingering; Anon., 164; 
Bull, 165-6; 168 (No. 4); ex- 
amples by Bull, 190 f. 

Rhythmical balance in Form; "Mai 
Sims," 118 ; " Walsingham," 119. 

Ricercare, 4, 58 etc., see Fantasia. 

Richardson, Ferdinando, see Hey- 
borne. 

" Robin," song ; in Shakespeare, 85 : 
set hy Farnaby and Munday, 86 ; 
words practically lost, 88-89; cor- 
rection in Munday's, 201 note. 

Romanticism, i6th century, 5 ; Byrd 
and Gibbons, " Woods so wilde," 
93; G. Farnaby's "Dreame," etc., 
94-5 ; Byrd's " Bells," 97-98 : 
Byrd's "Walsingham," etc., 98; 
Farnaby's " Humour," 99 ; " Bara- 
fostus," Anon.,-Lo\. 

" Rosasolis," Farnaby's, 199. 

Rosseter's Galliard, 25, 'set' by 
Farnaby ; 208 ; correction in text 
of Galliard, 208 (26) and note. 

Round, dance; Byrd's Sellinger's,' 
53-4 ; relation to Haye or Raye, 54. 

"Rowland," song; tune, 92. 



SALTARELLA, ends Philips' Passamezzo 

Galliard, 33. 
"Salvator Mundi," setting by Bull, 

181-183. 

Scale, technical passages, 191. 
Scales, origin of; hexachords, 100- 



2l8 



INDEX 



101, 109; eight-note scales, 124; 
list of examples tending to employ 
modern scales, 125-6; also Byrd's 
Fantasia, 77; " Pawles Wharfe," 
134; R Johnson s Alman, 137. 

Scales mixed with Mode, Alman by 
Anon., 131-2 and, 133*; example 
of " keyless " ground bass, 187. 

Schumann, descent from i6th cen- 
tury, 4. 

Selden, on court dances, James I. and 
Charles I., 37. 

Bellinger's Round, 53-54, by Byrd. 

Sequence, repetition in ; Anon , 121 ; 
Oystermayre, 122; Tisdall, 129. 

Shakespeare ; " Callino castore me," 
7 and 87-8 ; Dumps, in Lucrece, 
8 ; Barnfield and Dowland, 27 ; 
Sir Toby on eight-bar strains in 
pazzamezzo, 33 ; Shakespeare songs 
in Fitzwilliam Book, 85 ; Ophelia's 
songs, " Walsingham," 86, and 
tune, 119; " Fortune," " Robin," 
" O mistress mine," 88-9. 

Signature, modal; sign of "trans- 
position," 62, not of "key," 71; 
signature of t-wo flats for " double 
transposition," 104; example of 
" transposed " Dorian mode, with 
signature of one flat, 121-122. 

Signature,time; C = -f,8i,82, 119, 144; 
3 = f or -\ 2 -, 113; "e=|, 170, 171; 
( = 3J-, 178 ; list of instances, 
187. 

Sinkapace, or Galliard, see Cinque- 
pace. 

Six-line staves for virginal music, 8, 
and photo, of MS. facing p. 8. 

Slide, ornament, in Fitzwilliam Book, 
32, 166-7. 

Sonata, origin of, 4 ; form, 117, 119, 

I2 3- 

Sol Fa, 100, 135, etc. 

Songs, arranged for virginals, 4 ; 
book of songs by Dowland, 29 ; 
treatment in Fitzwilliam Book, 
examples, 79-93; Shakespearian, 
85-89; humorous, 90 ; words and 
music of " Jhon, come kisse me 
now," 90-92. 

Spagnioletta, dance ; Farnaby's, 55, 



125; Farnaby's 'Old Spagno- 
letta,' with modern "minor key" 
features, 126. 

Speer, Daniel (1697), on fingering, 
158. 

Spohr, his chromaticisms anticipated 
by Tomkins and others, iyd; see 
music, p. 18. 

"St Thomas Wake," Pavan and 
Galliard, with Prelude, en suite. 21 ; 
modern "key," 125 (No. 6); date, 
130 note ; 205 (15). 

Steps of dances ; Pavan and Cou- 
rante, 36 ; Basse dance, 50 ; Gal- 
liard or Cinquepace, 52 ; Lavolta, 
53; Haye, 54; Morisco, 55. 

Strains; in pavan, irregular number 
of ' bars,' 19 ; ditto in galliard, 24 ; 
ditto in alman, 32^, 42 (list); 
regular strains, 32-33, 131-2, 
133^. 

Strauss, R. , 4. 

StriggiO) madrigal arranged, 6. 

Strogers, N., " services " arranged, 2 ; 
Fantasia, 63, shewing numbering 
of appearances of subject, and 
" false relations ; " final bars of 
same, 64, shewing irregular use of 
bars, church music by Strogers, 
208, 209. 

Sub-dominant harmony, in, 119, 
125, 126, 134, 138. 

Subject and answer; in Fantasia, 
appearances numbered, 62, 63 ; 
irregular, or false, answers, 65-69, 
in Bull, Byrd, Farnaby, and Han- 
del, also see 187; "tonal" answers 
result of "key," 71-73; "real" 
fugue in Fitzwilliam Book, 71-73 ; 
"real" fugue (1532), 73-74; 
"tonal" fugue, 74-78, Byrd, Bull, 
Morley, Tallis. 

Suite, origin of, 4, 21, 1 14-116 ; pos- 
sible " suite" by Morley, 200. 

" Sumer is icumen in," early Western 
harmony, 141-2. 

Sweelinck, 6 ; date of his " Ut, re, 
mi, fa, sol, la," 13; Fantasia, 
answer in contrary motion, 61 ; 
list of scholastic conceits in the 
same, 61-2 ; his piece on the 



INDEX 



219 



Hexachord, 100 ; extracts from the I 
same, 107, 8, 9; 'modulation' 
doubtful, 129-30; date, as throw- 
ing light on date of Fitzwilliam 
Book, 130 note ; organ pieces, 169 ; 
prelude by, 189; passage from, 
192; work reviewed, 203-5; example 
of counter-subject derived from sub- 
ject, 204^; various forms of Swee- 
linck's name, 204-5. 



TABOR-PIPE, and popular music, 124. 
Tabourot, tee Arbeau. 
Tallis, Thomas, pieces in Cosyn's 
Virginal Book, 2; his Litany, 68; 
plagal answer, 77 : dates of pieces, 
130 note; organ pieces, 169; 
specimens of his setting of " Felix 
namque," 173-6 ; work reviewed, 
207. 

Technique, i6th century, see Arpeg- 
gio, Broken octaves, Chord pas- 
sages, Fingering, Mordent, Scale ; 
examples, 190-192, 162-167. 
Temperament, acoustical, i6th cen- 
tury ; enharmonic modulation, 
Bull, 8 ; Perronet-Thompson, 101 ; 
practical directions for study of 
"just" tuning, 101 and 102 note; 
"mean-tone" tuning, 103 note; 
" equal " temperament, 102 note, 
and no; Bull's Hexachord as 
evidence, 127. 

Text of Fitzwilliam Virginal Book, 
corrections in, 75,111,138, 147,164, 
191, 201 notes, 202 notes, 203 (n). 
Thumb, in fingering, unused in Ger- 
many and Italy, 158 ; used in Eng- 
land, 158-9, but R.H. avoided, 164; 
L.H. not avoided, 165. 
Time, see Notation, Signature, and 

Proportion. 
" Tirsi," Marenzio's Madrigal, ar- 
ranged by Philips, 195-6. 
Tisdall, W., " Pavana Chromatica" 
and " Pavana Clement Cotton, 5 ' 
25; "key" with 5 sharps, 103: 
sequence, 123; Mixolydian mode 
and" key "with 4 or 5 sharps 
128-9; work reviewed, 207-8. 



" Titus Andronicus," ballad, see 

Fortune. 

Tomkins, Thomas ; his Pavan, 16-19 ; 
notes on the same, 19 ; his Hunting 
Galliard, 25, 201 ; his arrangement 
of " Barafostus' Dream," 86 (see 83), 
201 ; his variations on "Up Tails 
All," 188; work reviewed, 201. 
Tonal fugue, examples, 74-78. 
Tonic, dominant, etc., harmony, in, 

119, 120, 125, 126, 138. 
Tordion, dance, 49 ; galliard far tirre. 
Touch, on virginals, etc., 161. 
Tower Hill," Farnaby's, 126 (No. 
19), 199. 
Toye, dance, 47 ; four examples 

named, 202. 
Transcriptions for virginals, see Ar- 
rangements. 
Transposition, of modes, 62, 103 ; 
" double" transposition, 104; 122; 
Thomas Warrock's signature con- 
sidered, 140 ; jEolian transposed, 
185. 

Tregian, F., family history of, 10-13 ; 
possible connection with Fitz- 
william Book, 13-14; Mrs K. 
Tregian, 25 ; Sybil Tregian (?), 
10 ; Mr Tregian as copyist, 47 ; as 
composer, 85, 186, 206-7. 
Tritone (augmented 4th) in i6th 

century, 171, 184. 
Tuning, see Temperament. 

U 

"Up TAILS ALL," set by Farnaby, 188; 
variations on, by Tomkins, 188-9 ; 
Farnaby's, 199. 

Ut, re, mi, fa, sol, la: date of 
Sweelinck's, 13 ; settings by Swee- 
linck, Byrd, Bull, Farnaby, 100- 
no; 'modulation' a 'result' of 
this subject, in Sweelinck and 
Farnaby, 129-30. 



VARIATION form in i6th century, 4; 
variations on songs and dances, 57, 
79 ; specimens, variations on 
" Daphne," by G. Farnaby, 79-81 ; 



22O 



INDEX 



on " Walsingham," by Byrd, 81 
83; on " Barafostus' Dream," 
Anon., 83-84; on "Woods so 
wild," by Byrd, 113, ste 145-6 for 
the tune ; melody in tenor, 80, 84; 
melody in bass, 81, 188, 203 
(Inglott) ; melody within the 
chords, 82; method by agremens, 
113; unusual method, Byrd, 113 
note; "variation" leads to true 
"development," 124; variations 
on a " Grounde," by G. Farnaby, 
W. Byrd, W. Inglott, T. Tomkins, 
187-9; specimens of Tomkins' 
variations on a "ground," 188-9. 

" Veni," plain song; set by Anon., 
170; reduction of the harmony, 
170-2 ; the same set by Bull, 76. 

Verdi, and tradition, 127. 

Verstegan, Richard, and Tregian, 
10. 

Viol da gamba, as accompaniment for 
song, 29. 

Virdung (1511), on virginals, 150 
note. 

Virginal Books, 1-3. 

Virginals; account of the instru- 
ment, 149-157; case of, 150, and 
frontispiece; compass of, 132, 133 f. ; 
variety of compass, 150-1, 155, 
157; jacks, and action, 150; 
methods of fingering, 157-168 ; 
"short octaves," 151-155; split 
keys, 156-7 ; " touch," 161 ; also see 
plate facing p. 149. 

W 

WAGNI:R, anticipation of, 82 ; tradi- 
tion in, 127. 

Wallaschek, "Primitive Music," 5. 

"Walsingham"; in Lady Nevell's 
Book, 2 ; Byrd's variations on it, 
81, 82; in Shakespeare, 85, 86, 
87; the Shrine, 86; words of the 
songs, 86, 87 ; plain tune and 



words, 119-120; set by Bull and 
Byrd, 120. 

Ward, John, Elizabethan composer, 
2. 

Ward, 1740, "Lives of the Gresham 
Professors," 9, 190. 

Warrock, Thomas (1586); ? B|j 
major, no; ? minor chords in 
major key, 139-40; his two flats 
not a sign of "double transposi- 
tion," 140; work reviewed, 205. 

Wasielewski, on i6th century instru- 
mental music ; fugal answers 

OSS 2 )* 73> 74- 

Weelkes, "services" arranged for 
virginals, 2. 

"Well tempered Clavier," anti- 
cipated, 8, IO2. 

" When a liv'd at yam," Northern 
Song on Burden, 143. 

"Why aske you," song, set by 
Anon, and G. Farnaby, 86. 

Wilson, of Scarborough (1791), 
pianoforte by, 151. 

"Woods so wild, The"; in Lady 
Nevell's Book, 3 ; set by Byrd and 
Gibbons, 85-6 ; modal harmony, 
125 (No. n); variation on, 113 
note ; music of, 145-6 ; 201. 

"Woody Cock," Farnaby's, 199-200. 

Wooldridge, H. Ellis, editor of 
Chappell, 15, 79. 



YORK Minster Library, MSS.; of 
John Jenkins' Fantasias for strings, 
102; of service in "D minor" by 
Nicholas Strogers, 208 ; of anthem 
by Strogers, 209 ; of three anthems 
by Edmund Hooper, 208, 209. 



ZoucHE'sMaske, Lady, 126 (No. 18). 
Zumpe (1766), pianoforte by, 150. 






TURNBULL AND SPEARS, PRINTERS. EDINBURGH. 



BOOKS ON MUSIC AND MUSICIANS 
By the Author of"An Elizabethan Virginal Book " 

Shakespeare and Music 

BY 

E. W. NAYLOR 

With Photogravure Frontispiece, $s. net. 

"A little manual which all readers of Shakespeare will heartily 
appreciate. . . . Printed in such a style as to worthily accompany 
the * Temple Shakespeare.' " The Academy. 

"An excellent little work." Musical Opinion. 

" The scope of the work is not only perfectly original in design 
and thoroughly well carried out, but it is written throughout in a clear, 
concise, and most interesting manner. It will be found of invaluable 
use to the Shakespearean scholar, as the subject on which it treats is 
one upon which some of our greatest commentators have come 
sorely to grief." St Paul's. 



Richard Wagner 

BY 

HOUSTON STEWART CHAMBERLAIN 

TRANSLATED FROM THE GERMAN BY 

G. AINSLIE RIGHT 

AND REVISED BY THE AUTHOR 

With Numerous Illustrations, Medium 4/0, 25^. net 

The Times says : " The four parts of the book contain elaborate 
and exhaustive essays on Wagner's life, writings, and teachings, 
his art-works and Bayreuth. The illustrations are beautiful as 
well as plentiful, and the book is one of deep interest and great 
weight." 

J. M. DENT & CO., 29 AND 30 BEDFORD ST., W.C. 



BOOKS ON MUSIC AND MUSICIANS 



A History of the Pianoforte 
and Pianoforte Player 

TRANSLATED FROM THE GERMAN BY 

DR OSCAR BIE, E. E. KELLETT, M.A. 

AND 

EDWARD W. NAYLOR, M.A., Mus.D. 

Cloth extra, Royal 8?>o, I2s. 6d. net. Illustrated 

" Not merely a handsome volume, with a multitude of interesting 
and effective illustrations, but a most valuable compendium of piano- 
forte literature/' Times. 

" A valuable and interesting book of which the illustrations alone 
constitute a real treasure." Birmingham Daily Gazette. 

" Very comprehensive in scheme, and intelligent in treatment." 
Globe. 

" A distinct gain to musical literature." New Age. 



God and Music 

BY 

JOHN HARRINGTON EDWARDS 

Cloth, Crown 8u0, 4-r. net 

<f He has much to say that is interesting and convincing in proof 
* that there is a Supreme Being making for harmony, happiness, and 
spiritual perfecting.* " Western Daily Press. 

" It should be read with interest by readers anxious to encourage 
an appreciation of music by reasoning out a cause for its charm. "- 
Scotsman. 

" The argument, in a series of carefully written chapters, is 
developed with a wealth of quotation and a felicity of language 
which makes most agreeable reading. It is refreshing to have so 
clear and convincing a statement which makes for faith from a side 
which has been so little heeded." 

J. M. DENT & CO., 29 AND 30 BEDFORD ST., W.C. 




ML 

703 

N3 

Maaic 



Naylor, Edward Woodall 

An Elizabethan virginal 
book 



PLEASE DO NOT REMOVE 
CARDS OR SLIPS FROM THIS POCKET 

UNIVERSITY OF TORONTO LIBRARY