Dowland: Farwell (fantasie) (P 3)
|The last two lines of Matthew Holmes’ neatly written manuscript of John Dowland’s piece for lute Farwell.
At the bottom is a rare example of Dowland’s signature, presumably indicating his approval.
They must have been wonderful players to give a performance from such a compressed and enigmatic tablature.
John Dowland wrote two extraordinary pieces that are unusual in being based on fragments of the chromatic scale. In Farwell (his spelling), set for the lute in G, he chose the 6 notes ascending from A to D. In this ukulele version I have maintained his fingering on the top 4 strings as closely as possible, so the scale goes from B to E on the 1st string. This fragment is used throughout the piece: Diana Poulton (in John Dowland, 1982, Faber & Faber, p 115) counted 14 instances.
Structure: One can regard this piece as being written in segments of about 3 (sometimes 6) bars, each expressing a particular idea or development of it.
Bars 1 – 3: the chromatic scale segment that “seeds” the piece.
Bars 15 – 17: what could be regarded as the statement of the “theme tune”, in fugue format.
Bars 23 – 24, 27 – 28: variations on the above.
Bars 35 (2nd half), 39 (1st half): the “Solus cum sola” motif (see relevant post).
Transcription for the ukulele presented more than the usual problems in reducing from 8 courses on the lute (unusually, JD specified fingerings on the 7th and 8th courses) to 4 strings. Fortunately, Sarge Gerbode has published a fair copy online, but there were so many unexpected notes and harmonies that I also referred to facsimilies of the the MS in the Matthew Holmes Lute Book in Cambridge, and there the notes really were. (The oddest note is the D at the end of bar 30, but the task of harmonizing the chromatic figures leads to interesting voicings throughout.)
I transferred all the notes I could into TablEdit, moving the lower notes an octave higher where I could fit them in. Once they were in the right position on the “timeline”, I then had to decide on lengths. (I have mentioned elsewhere that lute tabs at this time showed where to put your fingers, and when to pluck, but not when to take your fingers off.) The next decision was to sort out the 4 voices and reduce them in number to something fingerable on the ukulele. Fortunately, I then found a transcription made by the English composer Peter Warlock in the last century, and this helped a lot.
The principal changes are:
Bar 9: scale fragment on string 2 is in bass voice in original.
Bar 12, 2nd half: upper voice raised an octave.
Bars 35 – 37: upper voice raised an octave.
Finally, I listened to Nigel North’s performance repeatedly, and then removed as many less important notes as I could, to help distinguish the voices with the most melodic lines that carry the piece. Thus, what we now have is basically a two-voice arrangement, with bass notes added to emphasise the rhythm and enrich the harmonies. It just needed minor adjustments to make the score reasonably playable, but I still find it very challenging. Poulton does say that the lutenist requires “superb technique” to play this piece, so I wonder if I will ever be able to do it justice.
By the way, you may think that a tempo of 36 bpm is very slow at the beginning of the piece, but just wait until you get to the meaty bits!
You can find the piece in the following formats:
pdf (quick preview.)
pdf (tabs and notation; 9 pp.)
MIDI (the intentional discords sound very harsh, but don’t let this spoil your appreciation of the writing.)
Note added 11 Sept 2018: I came back to this piece after a year, and found 2 misprints and a few infelicities. These I have corrected, so the links above refer to the new version. I have also removed the fingerings, which were presumptuous: also, I kept changing my mind.