Tuesday, October 06, 2009


Oh my god. Temperament regularly arouses heated passions and arguments among lutenists, especially on the lute mailing list. Intemperate is more like it. The arguments are futile because there is, mathematically, no right answer to the question of how different notes should be tuned relative to one another, so the different temperaments (tuning systems) are all flawed compromises. So it's largely a matter of which flawed compromise system you prefer, tinged with a little information about the systems used in the past.

Equal temperament (boo, hiss) is the modern standard. Mean temperament (1/5 comma, 1/6 comma, ...), just intonation, Pythagorean/ natural temperament, Valotti, Werckmeister, .... The debate rages. There are even i-phone applications for different historical tuning systems.

Refer elsewhere for more detailed explanations, for example Stuart Isacoff's book Temperament, aimed at non-specialist readers.

I'm not aware that the issue ever arises with modern guitarists. Firstly, fixed frets mean there's nothing to be done anyway. Secondly, guitarists seem to have accepted equal temperament long ago. Early nineteenth-century experiments with enharmonic guitars with crazy side-stepped fret dispositions gave impractical, not to mention unplayable, results.

On the lute, despite its movable frets, even the compromises have to be compromised. Each fret, as on the guitar, has to serve for five different notes. For example, the first fret on a G lute gives G#, C#, F#, A# and D#. Or, rather, ten different notes, since it also has to serve for Ab, Db, Gb, Bb and Eb. In systems other than equal temperament, these are not the same notes as the sharp equivalents. Which means, basically, that any tuning system other than equal temperament is shot to pieces.

A particular problem: Bb on the third course and F# on the fourth course require the first fret to be in different places. Both notes are frequently needed in the lute's favourite key of G minor. What to do? Some players have tried placing so-called tastini (little frets) just behind the real first fret, to allow both notes to be achieved. My own limited experience of tastini is that they are a serious impediment to accurate playing.

So, I'm with Vincentio Galilei (father of Galileo G) in advocating equal temperament. He put his theory into practice in his Libro d'intavolatura di liuto in 1584, with a mighty set of Passemezzos, Romanescas and Saltarellos, written in all the major and minor keys. This was a full century before the birth of J.S. Bach, he of the Well-Tempered Klavier. Various composers for baroque guitar (e.g. Bartolotti, Pellegrini) wrote similar cycles in the seventeenth century.

To be fair, I should allow the other side to put their case. Here are two well-written articles from Martin Shepherd and David van Ooijen. I'm not yet convinced, though...

Monday, October 05, 2009

Musica Ficta

Musica ficta is the practice in pre-1600 music of adding non-notated accidentals at cadences and elsewhere, usually to avoid ungracious harmonic and melodic intervals. Practical problem for performers, especially modern ones who have not been brought up with the medieval hexachord system: knowing where, how, and how much to do this.

Lute tablature comes galloping to the rescue! In tablature, pitches of notes are shown precisely. So a composer or arranger for the lute had to show his musica ficta decisions explicitly. This means that arrangements for lute of music for voices or other instruments can offer solutions, or at least insights, into musica ficta practice. There are hundreds of such arrangements, of both sacred and secular polyphonic music, throughout the sixteenth century.

Firstly: on this evidence, music ficta undoubtedly existed. And was heavily used.

But, secondly: there are no universal rules. Where different composers have written versions of the same piece, they often adopt different musica ficta solutions at the same points. For example, a cadence going from e-g-c to d-a-d might have E flat and C natural in the first chord, or E natural and C sharp.

Or even a mixture of both. Albert de Rippe's version of Janequin's chanson D'un seul soleil in this context consistently uses a chord of E flat - G - C sharp, with an extra A in the middle for good measure. An early example of what, a couple of centuries later, came to be known as a German sixth, sounding strikingly out of place in 1552 but undoubtedly authentic. And the first of his two fantaisies for guitar (also 1552: yes, the guitar has been around for a while) has some even more exuberant scrunches in it: accidental logic taken to extremes.

Sunday, October 04, 2009

Howard Mayer Brown

Instrumental Music Printed Before 1600: A Bibliography

An dry title for an astonishing book, first published in 1965 by the Harvard University Press. Howard Mayer Brown's aim was to catalogue and describe all of the instrumental music published before 1600. It has entries in chronological order for some 400 books, giving a brief description of each book and a full listing of the pieces of music in it. Equally valuable are a set of indexes (indices?) by library, type of notation, performing medium, composer, and titles of works, allowing easy access to the contents through different routes.

Now that these music books are widely available in facsimile, printed and online, Brown's book takes on a new life as a guide to these publications. The sparse examples he gives of the modern reprints available in 1965 show just how much things have moved on since then.

Index III is gratifying. It shows 206 volumes containing music for solo lute and only 76 for keyboard solo. Plus further volumes for similar instruments such as cittern, guitar, or vihuela. Music for other instruments is almost all generic ensemble music: there is very little that specifies named instruments, and very little solo music.

Brown's book is still available in a good-quality reprint from www.iUniverse.com

p.s. I've just discovered a review from 1966, when the book first came out, which starts: "This is a volume that I would not hesitate to call magnificent". Quite.


A nice name for a horrid phenomenon (pace Howard Mayer Brown). A concordance is another version of the same piece in a different source. A table of concordances is when someone goes through a book of music and lists the known alternative versions of every piece in it. A real enthusiast will highlight the differences between the different versions.

Tables of concordances feature heavily in modern publications of lute music, whether facsimiles or new editions. Of course they have a scholarly purpose. But they sit like a lead weight, dull, depressing, sapping energy, turning the publication into a cataloguing exercise rather than a musical one. And making the musical reader feel somehow inferior if he doesn't know his way round the arcane abbreviations for libraries, manuscripts, folios and sigla which are used to locate the pieces.

Worse, one forms the distinct impression that publications can be delayed, or simply fail ever to appear, because of the enormous time and labour required to assemble this catalogue of stuff that no-one wants to read. Not good.