These days there seems to be a positive Urtext mania among guitarists. I am as fanatic as anybody about respecting the composer’s ideas, but I am referring to something else. Sorry if I somehow contributed to this trend (although I was not the only one) through insisting (20 years ago) that the 1929 version of the Villa-Lobos Études was worth considering (the new edition by Frédéric Zigante in Max Eschig has solved practically all the textual problems and the typos of the 1954 edition, so if you want to know how this particular thriller ended, go look).
Every musician knows there are invaluable clues to be found in the original manuscripts. But I am not referring to the police-work part of it, just looking at the notes to see if changes have been made. Manuscripts preserve the creative élan, even if the composer was very disciplined in his calligraphy. And even more if the composer wasn’t: anybody who has seen the “Appassionata” manuscript will be permanently affected. In the guitar literature, if we look at the Giuliani manuscripts, the elegance and lightness of the calligraphy is already saying something about how the music should be played. Sor wrote a very good hand too. I would love to see some manuscript of the Legnani Caprices, also. So, is it a good idea to publish manuscripts? Yes, absolutely. Is it a good idea to play only what is there, when the publication of the work has a complicated history? Not necessarily. Urtext is not always the solution. Life is a bit more complicated than that.
In the literature for guitar, we have special problems. This is probably because the guitar had such a lowly status that nobody bothered to write about it in the treaties of instrumentation, and when they did, it was –like Berlioz did – to say that you need to play the instrument to write well for it. Which is certainly not true, but who was going to argue. So composers were not informed about the guitar, and in many cases still aren’t. There are indeed composers who understand inmediately the nature of the guitar and what would work well on it – Britten is of course the example, but he was not the only one. When Tosar wrote a piece for me, there was nothing to change, and I am sure that if Ligeti ever had written for guitar, the same would apply, not to speak of Lachenmann’s “Salut für Caudwell” and the wonderful guitar part in his opera. Ginastera actually bought a guitar to try the positions when he was writing the Sonata. But this is not always the case. So, when the composer was not a guitarist, it often happened that the work had to be “revised” by a guitarist to make it playable, or even just to make it better suited for the instrument. This is the case with practically all the repertoire written for Segovia; the works, as published, are “virtual transcriptions” of an ideal “original” which usually was completely unidiomatic. Just check the differences between what was and was not revised by Segovia in the works of Ponce or Castelnuovo-Tedesco, both fine composers but completely obtuse as far as guitar idioms was concerned. They both trusted Segovia to find the way to express their ideas more or less well translated; specially Ponce, who came to form a kind of symbiosis with Segovia when writing for the guitar.What we have is a result that they (C-T and Ponce) approved.
So, in many cases, if you ignore the Segovia revisions, at least of composers which were close to him, you are playing a first draft instead of the finished work. And it would be nuts to deny that Segovia’s revisions improved the result. (If you doubt this, the manuscript of the “Concierto del Sur” is circulating in Internet. Go have a look. Segovia should have been given a medal.) You say, his decisions were radical and sometimes debatable? When he was in good synthony with the composer, he made the works what they should have been in the first place.
Of course you can go overboard in editing, and you can also do the opposite mistake. I have seen the manuscript of the “Sonata” by Antonio José. In the title page it says “Regino, haz lo que quieras con esto” (Regino [Sainz de la Maza] do what you want with this). I have to mention that Moreno-Torroba once gave me about two kilos of guitar works with the same comment, and I am sure I was not the only one. Well, Regino didn’t. So we are stuck with a very unidiomatic major work, which has caused more tendinitis than a million video games.
One mistake which appears often in composers who do not know the guitar is as follows: there is chord A and chord B. Both are playable. But you just cannot go quickly or comfortably from A to B. The composer asks a guitarist friend about playability, and the g.f. says, yes, of course both chords can be played. But the important question was never asked. Another usual mistake is to trust that it is enough to write “ffff” to ensure that it will sound loud. Some things will just not sound loud, no matter how many “f”s you write next to them.
Another example: Piazzolla’s “Five Pieces”. I am not acquainted with the details of the creation of the work, but I do know he had the possibility to consult good guitarists. And of course Bèrben published what Piazzolla gave them. Nevertheless, the final result just needs some kind of revision to work, at least, I feel it does. It is one of Piazzolla’s finest works, and it is a pity that it is so uncomfortable to play. Sometimes the writing is unnecessarily timid, as if he feared that continuing an ostinato and doing something else at the same time would make it unplayable; sometimes it asks for impossibilites. Another point in this connection, and I am going wildly out of the subject, but after all this is my blog, not yours: if you don’t know Piazzolla’s tango work, or more in general, the style of tango playing, and you play this work “as written” (without modifying durations, articulations, tempi), you are missing the very essence of the music. Context is invisible text, and of course not only in Piazzolla’s music.
Tansman’s guitar works (the ones not edited by Segovia) are another case in point. The “Passacaglia” is a wonderful piece of music, but it is just not playable “as written”, at least, not in a way that can be musical, it is just too uncomfortable. The same thing happens with Rodrigo’s “Elogio de la guitarra”, maybe potentially his best solo guitar work – but it needs either to be transposed, or to use some kind of unorthodox tuning. Who can think of a pedal of fast repeated notes in B (2nd fret of the 5th string) and complicated chords going on above it? This is physically impossible. No need to mention Aranjuez, where everybody changes something here and there, without anybody noticing (I do it too). Question: why didn’t anybody tell these composers that changes were needed? Awe? Timidity? Lack of imagination? Or maybe they did speak up and were ignored?
The problem is compounded to the nth when we consider that now, many of the manuscripts are available in Internet. So, second-guessing Segovia has become almost a parlor game, by people who are (in 99.9 % of the cases) not qualified to untie Segovia’s shoes.
So, what does it mean to be “true” to the composer’s idea? It is a very complex problem and the answer might be different for every case. I am sure I will return to this, and it is not only a case of which notes do you play. But to think that just playing what is written in the manuscript is the answer (specially when the manuscript was revised with the consent of the composer) would be, at least, extremely naïve, and potentially very wrong. Playing a first draft instead of the finished work does not mean you are being true to the composer’s “original” conception. It is just a big mistake.