The available music by French composers, such as Le Roy, is written in tablature, and is clearly printed. In French tablature, also used in Britain,
a indicates the nut (fret 0),
b is fret 1,
c is fret 2, and so on. The letter
j was not used, so
k = 9,
l = 10 and
m = 11. There is a full list in the lute section
below.
The first string is represented by the space above the top line of the stave, down to the 4th string at the bottom. The facsimile above shows one and a half pieces in Adrian Le Roy's
Quart livre de tablature de guiterre, pub 1553. The letter forms are squashed but clear. The
c has a flat top to distinguish it from the
e. (When we get to manuscripts later, we will see that letter forms are not always so clear.)
The "flags" above the letters indicate how long a note sounds until the next note is plucked, but not necessarily how long the note should be held. See the table below.
 |
Concordance of the various symbols used in Renaissance lute and guitar music in print and MSS (French format). One may need to halve the lengths of the notes in the originals, as Poulton and Lam did in most of their transcriptions. The first finger dot indicates un-accented notes, assuming that it is weaker than the thumb.
|
In some setting styles, every note will be given a length flag, as in modern music. In others, as in the Le Roy example above, if there is a series of notes having the same lengths (or, strictly, are plucked at the same intervals), only the first note is flagged. This leads to a less cluttered appearance, but makes it harder to appreciate complex rhythms at a glance.
Having got the notes down in the right place on paper or in your favourite music setting package, you then has to decide how long a note should be held.
To set the music, I have been using the slightly idiosyncratic TablEdit app, which seems perfectly designed for setting this early music, because note fingerings are in effect entered into a tablature grid, with a regular time axis along the bottom. Once the notes are in the right place, their lengths can be easily adjusted.
Looking at bars 1 and 5 in line 1 in the facsimile above, it is pretty obvious that the notes on strings 2, 3 and 4 should be held through the bar whilst the melody is played on the top string. In line 4 bar 5, the melody is held on the top string throughout the bar.
When it comes, however, to the bass line (on the 4th string) in this bar we have a problem. In this notation system, rests are implied by a gap, just as an extended note is. So:
(a) is there a minim (half note) rest in the first half of the bar, before we play two crotchets of c a, or
(b) do we extend the e from the previous bar?
The obvious thing to do is to try it out on your ukulele and see what seems best. Or, you can find a recording online by an expert, and see what he or she does. Valéry Sauvage (a k a "Ukeval") is a really excellent and productive exponent of this music, and it is worth searching his YouTube channel.
There are occasions where a note followed by a space does not imply that it should be extended. In line 3 bar 5, for example, the a on the first beat on the first string is obviously part of a scale that continues down the second string (d b a) so the a will last just one beat. The notes on the third and fourth strings will, however, be held through the bar.
Clef: it is traditional and convenient in music for guitar and banjo to write the music on the treble clef, but an octave higher than it actually sounds. When I started transcribing this is what I did, as you will see from the first posts on the blog. There were, however, just too many ledger lines.
Later transcriptions I have posted are true pitch, with middle C (the one on the first ledger line below the stave) = C4 = string 3. This means the low G string for which I write only needs 2 ledger lines. I am grateful to Bill1 on the Ukulele Underground forum for pointing out the inconsistency, and the value of true pitch notation.
(For the lute transcriptions by Poulton and Lam, shown below, the great stave is used, which is OK for understanding a composition, but impossible as a performing copy for a lutenist.)
Here is a worked example of how I adapted a piece by Le Roy to be played on a ukulele with a low 4th string. I've used this one as it is one of the first that I worked on. You will see how compact the original format is. I read somewhere that in this period books were so expensive that they cost as much as a lute, so conciseness was obviously an imperative.
My version below should be self-explanatory, if not perfectly set. Note the attempt in places to discern three voices: melody on top, bass on the bottom, and filled out with harmonies in between. Understanding these is a great help when interpreting some complex lute pieces.
Baroque guitar
Arranging baroque guitar music for ukulele is a minefield. Different composers used different, often unspecified, tunings. The instrument has five courses, of paired strings, except sometimes for the first. The courses were often in octaves, or the lower strings could be re-entrant, like the modern ukulele. This can make the music seem rather jumpy in playing a particular voice. If you ignore the octaves, the tuning was like the first five strings of the "normal" guitar.
I have made a few arrangements and published them on this blog.
For more information I refer you to A guide to playing the Baroque guitar by the late James Tyler (Indiana University Press, 2011), which I found invaluable.
There is quite a body of lute music available: as facsimiles of MSS (more on reading them later), as transcriptions (tablature and notation), and as arrangements for guitar (some with the third string tuned a semitone flat to give the same intervals as a lute). I am assuming here that the French form of tablatures described above is used.
Much of this account also applies to music written for the vihuela, a Spanish instrument tuned like a lute, but with a body more like that of a guitar. The tablature was written in numbers rather than letters, and the string order was reversed, with the treble at the bottom. This post is long enough already, so I will just deal here with lute music in French tablature.
If you are lucky enough to have a Renaissance guitar (I recently treated myself to one) most of the transcriptions will fit perfectly. The only problem is that on a tenor uke you can stretch to an extra fret, so in a few places my arrangements feel a bit uncomfortable on the guitar.
The first step is to transfer the fret/string positions in the lute as directly as possible to the ukulele. The challenges include:
(a) The 3rd string on the lute is 1 semitone flat in comparison with the 3rd string on the ukulele or guitar. This is a pain because if there is an open 3rd string in a lute piece, this has to put on string 4 fret 4 on the uke, which can mess up the bass line.
(b) The uke obviously has no 5th and 6th strings, so where possible we have to move notes on them up an octave and enter on strings 3 and 4, which might already be occupied.
The following equations (sorry if you hate algebra) summarise what I do in my head when transcribing. If U1...U4 = fret positions on ukulele strings 1...4, and L1...L6 = fret positions on lute courses 1...6, then:
U1 = L1
U2 = L2
U3 = L3 – 1
U4 = L4
U3 = L5 + 2 [8va]
U4 = L6 + 2 [8va]
You will see that the 3rd and 4th strings could get a little crowded.
(Thanks to Mauro Padellini for pointing out errors in the formulae.)
The second step is to identify the voices. If you have a piano transcription in parallel with the lute tabs (as in Poulton and Lam, see below) this is quite easy. In general there will be a "melody" line on top, a bass line (guess where), and one or two middle lines or chordal harmonies, all rather like the Le Roy pieces mentioned above, but more complicated. It's probably best to metaphorically ink in the top voice first, then the bottom voice, and finally see what one can do to fit in the middle parts. It will never be as full a version as the original, but it does help in our understanding of the music. By studying the tabs of composers such as Le Roy and Morlaye for the Renaissance guitar you can see how the masters managed the compromises.
The third step is to refine the arrangement by
(a) Playing it through, on the ukulele, adjusting the fingering and string on which a note is played to make it as easy and efficient as possible to perform. At least we can stretch to a few more frets than the classical guitarist!
(b) Listening to the piece played by a skilled performer on the lute, to get the feeling of the music, and to understand the structure and emphasis.
It may make the piece easier to play, or more complete, if you transpose the piece to a higher key, to make more use of the higher frets on the ukulele or to fit some lower notes in.
My observation of lute pieces is that the chords (in our modern way of looking at music) are generally rooted in the bass. Often in one's uke arrangement one wants a bass note as part of the rhythm (the
boom in
boom ching), and the ukulele doesn't have it. I then use another note from the chord on the bass (4th) string. So, in the chord of D or Dm, I will use an A rather than the absent D – not perfect, but better than nothing. (This is also what composers for the Renaissance guitar did.)
Addendum added Nov 2023I have recently been trying out a modification of the above workflow:
1. Make a full transcription for lute – surprisingly easy in TablEdit. This means I always have a full copy to work with.
2. Duplicate the lute version, and copy notes on the 4th to 7th strings an octave higher. Play back each bar in MIDI and listen for and correct any discords.
3. Eliminate the bottom strings, leaving just the top four strings. This gives a version playable on a ukulele with the 3rd string tuned down a semitone.
4. Make a copy of the above, adjust the tuning to that of a conventional low-G ukulele. All the notes on the third string will have to be dropped by one fret. Annoyingly, this means representing an open 3rd lute string at the 4th fret on the 4th string.
5. Now comes the refining stage: correcting the note lengths, and further modifying the fretting to fit more easily under the fingers. The upper voice will always be maintained, but the skill is in making the lower voice(s) comprehensible.
6. I have now started making two versions of the arrangements:
(1) a fully-voiced version, to aid understanding of the musical lines, and
(2) an unvoiced version with all the stems pointing up, and every note on the same stem of the same length. This sounds horrible in MIDI playback but it does help getting the timing right in more complex passages. And, of course, this is much as the Renaissance lute player would have dealt with - see the illustrations in
THIS POST.
Symbols and handwriting in MSS
The table above shows a concordance of fret numbers, French tab letters, and examples from facsimilies of MSS around 1590. And here is the secretary hand from the fount "Secretary hand ancient":
The forms are pretty close to those used in fingering MSS scores. It is confusing at first that the c-symbol resembles r, the f can look like ß, the i like y, and the k like a mistake in writing an l. Even now, in my head, I say "r = 2" rather than "c = 2". Why they weren't written as clearly as those in Le Roy's books from the 1550s, I don't know – tradition, I suppose.
By the 16th century the elegant Italian chancery hand (later to become "Italics" in print) was influencing handwriting, and the text in lute MSS was written in the more legible form, or a hybrid between the two.
An example: John Dowland's Lachrimae Pavan
I have chosen this piece as it was enormously popular in his time and was his signature composition. Here we see the first four bars.
 |
One of many MS versions of Lachrimae Pavan. Note that the second flag looks a bit like a ß with a long stem: this is a shorthand version of the crotchet (quarter-note) flag, which should look rather like an "F". The MS versions of even shorter notes look like a stem with a wiggly line, and are not always easy to interpret. In flagging adjacent notes of the same length, the tails are joined as in modern notation, to give the "gate" or "grid-iron" format.
This facsimile can be seen in full here: https://cudl.lib.cam.ac.uk/view/MS-DD-00005-00078-00003/18
 |
The first 4 bars of Dowland's Lachrimae as transcribed and interpreted by Poulton and Lam. I think this must be a consensus version from a number of sources. A modified version of the Secretary hand is used for the tabs. The # symbol indicates an ornament of some kind (it's up to you), and the slanting lines indicate how long to hold the bass.
|
|
You will see that in the lute version there are as many as four voices (musical lines), as in bar 3. In the uke version below I have simplified this to two or three voices: the upper stems-up, the others stem-down. When the stems-down notes are of the same duration it is neater to combine them, but where durations vary they may be displaced to left or right, whichever is the less cluttered.
The tablature setting I have used is as complete as possible within the constraints of the format. This means that you can play directly off the tab stave, rather than use the tabs for fingering and the notation for duration. The notation does give a much better visual impression of what the music sounds like.
 |
The same 4 bars in my arrangement for ukulele, with a low 4th string.
|
This example shows amongst other features:
(b) Bar 2, beat 1: The ligatured B♭5 on lute string 1 is implied, not written in the tab.
(c) Throughout, there are inevitably fewer notes in the harmonies in the uke version, and the lower voices are sometimes combined.
Concluding remarks
Despite the length of this post, the process is really not that difficult. Why not have a go?
Arranging can be just as much fun as playing, and it does help you to understand a piece.
References
