I would appreciate a word of advice. I can find plenty of material about the techniques of orchestration, and I understand them well enough. What I can't seem to find is discussion as to how far an orchestrator is expected to go in adapting original material to its new orchestral medium.

For example, a piece might have a piano original, with chords that fade where an orchestra might, to advantage, swell. This, in turn, might suggest a succession of chords where the piano has just one, with further, consequential, changes down the line.

Obviously, an orchestrator can do what (s)he likes if the original work is out of copyright but I'd like to understand where the boundary between orchestration and adaptation is felt to lie. What's expected in the way of fidelity or freedom if someone is commissioned to orchestrate someone else's work?

There may not be formal protocols governing the orchestrator's craft but there may be norms, and I wonder what people think (or know) about the matter, or what I can read?

TDT

I think there are two takes on this:

orchestration - arranging a composition written for an ensmble (but more usually a solo instrument or in particell) for anything that can pass as an orchestra from the kind you would find in a provincial theatre to something huge such as Mahler would use.

arrangement - reworking a composition for solo/ensemble/orchestra to another.

The aim is to adapt the idiom of the original to the new instrument/ensemble.

The arranger/orchestrator goes as far as needed (eg to fulfil a commission) which doesn't mean the arrangement will be effective. I learn that someone has orchestrated Debussy's Preludes. I cannot believe they would be effective, just because the originals depend SO much on the colours and timbres of the piano. They could perhaps be regarded as different works but they wouldn't be Debussy's Preludes to me.

Orchestrating a piano piece can be tricky. Deep chords that sound fine on a piano would sound horribly thick literally transcribed for divisi cellos and basses. The sustaining pedal has to be taken in account in the arrangement/orchestration and, as you observe, the dynamics of a piano note/chord have to be taken in account.

A super piece of arranging I recently heard is Elgar Howarth's version of Mussorgsky's Pictures for a brass ensemble, played by the Wallace Collection. I believe it's available as an MP3 but the CD is out of print.

It works the other way round. Liszt arranged the Beethoven Symphonies and much Wagner for piano (the aim was to bring these works into musical households. Stravinsky wrote a version of the Sacre du Printemps for two pianos, for use in rehearsals.

So, really, it's what you dare. There's no hard and fast rule. Some arrangers are sensitive to that preservation of the idiom, others are not. Some specialise in different genres: jazz, classic, light music. Some have to get on with it, like it or not if it's a commission.

And the only way, really, to work out how it's done is to study an arrangement against an original. Unfortunately few orchestrations from piano scores are out there. Some, though. In the main, it's score study - that's a subject by itself!

But no boundaries or demarcations. There are some challenges that no one in their right mind would take: Beethoven's Moonlight; Ravel's Gaspard de la Nuit!

I did get to hear Yan Pascal Tortelier's orchestration of the Ravel Piano Trio, and it was very successful. The original piece seems to strain at the leash, and he avoided simply becoming indulgent with the orchestra as a result.

Historically, the terms and roles have changed, but I'd suggest the contemporary distinction is that 'orchestration' is taking a piece from a small score or piano sketch, and using the orchestra as an artist's pallette, from which one can obtain every colour.

'Arrangement' suggests either a reduction in the ensemble, inevitably finding compromises, or a change in structure.

There are quite a few orchestrations by Ravel of his own and Debussy's piano pieces. Some that spring to mind:

"Mother Goose"

There are two orchestral versions of this. The first is a straight transcription, keeping the movements in order; the second (the ballet) has a larger orchestra, the order of the movements is changed, and there is new material, forming bridges between them. The first version is available in Dover score and is probably the more useful for the immediate purpose.

"Tombeau de Couperin"

Four of the original six movements. The orchestral score is available as a download from IMSLP.

Debussy "Danse (Tarantelle Styrienne)"

See here for sources, including orchestration by Ravel, string quartet version by Steve Jones and expansion to piano, four hands, by Henry Woollett.

This advert lists some others.

Moonlight:

http://www.youtube.com/watch?v=SAIkeYnqu30

http://www.youtube.com/watch?v=uj_FTlF38Fs

http://www.youtube.com/watch?v=Qem_Wqx-s7s

http://www.youtube.com/watch?v=8YEqJ4Kmf-8

Gaspard de la Nuit:

http://www.youtube.com/watch?v=bMkhrfB7dfQ

http://www.youtube.com/watch?v=7BaOuXny66Q...;feature=relmfu

http://www.youtube.com/watch?v=UZ7qgB4yqZI...;feature=relmfu

Oh yes. There are plenty of musicians who are not in their right mind. What do you think? Are they wasting their time?

Do you truly think Ondine or Scarbo preserve Ravel's pianistic idiom in an orchestrated version? Ravel more than proved his ability with the orchestra. He would have written those works orchestrally if so inclined. Unfortunately they were written to test the mettle of pianists!



As I said in reply to the original post, there are no hard and fast rules. Hoffnung had someone rearrange Tchaikowsy's 1812 for a recorder quintet and cap-gun.

Many thanks for responses so far.

The general view seems to be that if someone is commissioned to orchestrate, say, a piano piece (s)he can freely change the harmonies, dynamics etc.

Obviously, the particular results would be open to critical appraisal but significant change would not, in itself, be a violation of what's implied in the term 'orchestrate.'

That's how I understand what's been said but please correct me if I'm wrong.

T

You've about got it right.

Until recently my job was composer/arranger. I was sometimes forced to commit acts of heresy for which you'd be burned at the double-bar-line in less enlightened times.

Adapting a few bars of a Mozart piano sonata, butting it on to Elgar's Nimrod variation 9 and things. Made me cringe at times. Wasn't always as bad, of course.

I would consider a change of harmony to go beyond the bounds of "orchestration". One might change dynamic markings in order to be sure that the new version preserved the intent of the original; one might add a tempo marking to show a widely agreed traditional interpretation;[1] one might elaborate an ornament where one did not trust the player's interpretation of it.[2]

[1] but remember Mahler's aphorism, "Tradition is laziness"; also listen to Kissin's version of the Chopin Bb minor scherzo, in strict tempo, instead of the enormous speed variations of all the other performers I have heard.

[2] E.g. the Liszt piano duet versions (c. 1860) of the Beethoven symphonies, in which he adds a "nachslag" (turned ending) to many of the trills, because he knew that Beethoven's musicians would have done that as a conventional practice, but that his customers, small-town and rural middle class musical amateurs, would not have known the convention.

I think Ondine is very well orchestrated. Ravel wrote Gaspard pretty much simultaneously with Daphnis et Chloe and the way in which the orchestrator has mimicked the style of Daphnis in Ondine seems apt to me.

I am not entirely convinced by Scarbo, to be honest. Scarbo is in a different league to the other movements and is in the same vein as Chopin/Liszt etudes and Islamey and other testing the mettle of pianist works, to borrow your term. (I have also heard an orchestration of Islamey and the Liszt B minor which are equally unconvincing).

My view is that as he wrote the two works at the same time, he put everything he wanted to do with the piano into Gaspard and everything he wanted to do with orchestra into Daphnis. Writing Daphnis in particular almost killed Ravel. I don't think it was a question of the musical ability to arrange each for the other, I think it was a question of how much Ravel could take as the mere mortal that he was.

Anyway...

All of my favourite orchestrations/piano transcriptions have added a taste of the arranger. The person who can literally take the notes and reposition them onto another instrument is different to the person who can recompose the whole piece (for example, Liszt-Saint Saens Danse Macabre and the Gryazanov (sp) arrangement of Daphnis et Chloe) to suit the new instruments.

Oooo I'd like to get my hands on that one...