Monday, February 12, 2007

practical considerations for transcribing piano literature for guitar

Boy! Doesn't this sound exciting. Well, maybe it will interest a handful of people and if not I don't expect my blog to get a lot of traffic.

The most basic questions of what kind of piano literature to transcribe for solo guitar are as follows: in no particular order, and all of these are basic.

1. What is the compass of the writing?

Let's be reasonable. If the range of the piano writing goes beyond five octaves odds are pretty good that unless you're Katsuhiro Yamashita you better not transcribe this stuff. If the range if five octaves and you don't mind simplifying the harmonic language to its most essential elements you can get away with transcribing something like the first movement of Hindemith's first piano sonata for solo guitar. Not that I'm suggesting you should actually try that, exactly.

2. What is the contrapuntal density of the writing?

A certain composer I won't name said that the human brain can't meaningfully perceive more than three completely independent melodic lines and that rule holds mostly true for solo piano literature. In short fugues are out with the exception of fugues written for the violin, which has strings that are arrayed in an even less inviting pattern for contrapuntal music than the guitar. So Bartok or Bach fugues for violin can be played on the guitar. For that matter basically anything written for the viola can PROBABLY be played on the guitar just be playing everything at pitch. You'll never transcribe Hindemith's Op. 22 viola sonata, movement 4, of course, unless you choose to transpose down to E.

Even two voices with a high separation between parts or a lot of activity are not practical. No one transcribes a solo guitar version of the two voice fugue in E minor from the Well-Tempered Klavier and for good reason!

3. How extensive is accompanimental figuration used to support the melodic line?

A tight, close harmonization and simple arpeggiations or block chords are relatively easy to adapt. For instance, if you take something like, oh, Hindemith's Sonata II for piano and just play what's written in the treble staves as guitar music you can actually get through the whole first theme just by playing it on solo guitar. This is because the separation between melody and accompaniment isn't that severe, there aren't any abrupt changes of register, and the writing is in a simple key. When you CAN make a transcription that leaves the original key in tact it's preferable to do so even if it costs you a certain amount of effort.

That's my opinion anyway. Even if you have a piece in B flat major and it's tough to play you should still consider keeping it in the key if at a contrapuntal and textural level you can preserve the essense of the music which gets to the next question, jumping past piano music to other types of music.

4. How broad is the timbrel pallete of the music you plan to transcribe?

Transcribing piano music and string music is a relatively simple affair for solo guitar because of the homogenous tone colors. Transcribing trio music is trickier because while the three strings may have a consistent sound you'll make sacrifices in the contrapuntal and harmonic range of the piece you're transcribing. If you have an unusual instrumentation for a piece of chamber music (say, violin, clarinet, and cello) you can still transribe away despite the drastic differences in timbres among the instruments if the contrapuntal density and harmonic pallete are simple enough to let you distill its musical essence on to solo guitar. What you will lack in the timbrel variety of the instrumentation above you can make up for in using guitaristic effects like harmonics or pizzicato, ponticello, and the like.

But what this by and large does NOT mean is that you have any business transcribing symphonic works unless the musical form and structure is plain enough to survive the transition. One of the most overblown and meaningless declarations in guitar circles is that the guitar is like a miniature orchestra. Toru Takemitsu could say something like that and get away with it because he had some clue how the guitar was played and anything he didn't grok he could run by his good pals Leo Brouwer and Julian Bream. When you're friends with THOSE kinds of guitarists you can choose to not sweat the hard stuff. It doesn't hurt if you're a musical genius, either. Takemitsu had a keener ear for the tonal colors of the guitar than most guitarists so he could write stuff with unusual and complex timbrel shifts and have them work.

Anyway, end of that particular rant.

To throw a bone to orchestral transcriptions for solo guitar to the extent that a score is more pianistic it could work. Pictures at an Exhibition was originally a piano score and that gets back to my earlier observations about homogenous timbre and the compass of the writing. Once you eliminate the artificial extension of range created by orchestration you may find that music that spans a bunch of octaves can be reduced to maybe three or four, which puts it in a reasonable range for transription for the guitar. An example of this might be a work by Haydn. You could get at least a decent approximation of Haydn's Op 76 string quartet, movement 1 by keeping the points of melodic interest and simplifying the accompaniment patterns. Oh, wait, that's not an orchestral score. Oh well, I think the point probably speaks for itself. If Yamashita can put The Firebird on solo guitar an orchestral score can be transcribed for guitar.

But something tends to be sacrified. If you've heard even the fastest guitarst blazing through Esturias or an Albeniz piano piece transcribed for guitar it won't prepare you for someone playing the piano scores as they are originally set up. Every guitarist owes it to their art and to an understanding of Albeniz' piano music to get familiar with performances of the original piano versions. It might give you some incentive to look for more original works for guitar. :)

Transcriptions aren't bad or anything but I feel like when I attend guitar recitals I so often hear the same old pieces. Not saying I'm even remotely a better guitarist than any of these people, I'm just saying that as a paying audience member I can't help but agree with Matanya Ophee. I could hear even a great guitarist play a Bach partita or I could hear Hilary Hahn play the same thing. I'm a loyal guitarist and a composer who loves the guitar but when push comes to shove I'm hard pressed to find a guitarist whose version of a Bach partita for violin beats Zoltan Szekely's older recordings.

Something I don't see happening much in transcriptions for solo guitar is choral literature. This is a little odd because in a way choral literature is governed by restrictions at least as harsh as those the guitar presents to a composer. I suppose there are Bach chorales but I notice not much is done with more recent choral stuff. Is there anything wrong with transcribing a motet by Poulenc? Even I would concede that a solo guitar transcription of O Sacrum Convivium would need to transpose away from F sharp major ... but even allowing what for me is an unusually liberal concession to the reality that some things are easier to play in keys adjusted for the guitar it seems like there's some choral repertoire ripe for transcribing if we're going to keep playing transcribed music along with stuff origianlly written for the guitar.

Another little tangent. Things aren't that much different for two or more guitars. I can't shake the convention that the more guitars you multiply for a performance the lamer the music sounds. A guitar quartet has no sustain except through tremelo and tremelo doesn't sound that hot compared to the sustain of bowed string instruments and woodwinds or ... really any other instrument in the orchestra. Maybe it's more honest to say the guitarist is not a miniature orchestra but a really great harpsichord you can carry around with you that has some special effects the harpsichord doesn't have. This seems more honest than the dubious claim that the guitar is a miniature orchestra. The guitar and harpischord even have somewhat similar ranges overall ... which gets me back to why Bach gets transcribed so much. Ah, well, Badh is as fun as he is unavoidable if you don't overdo him.

1 comment:

James "Shoes" Walker said...

WTH,

Thanks for the insights, like your stuff.

James "Shoes" Walker
Seattle