Orchestration?

[c#minor]

Where can you figure out orchestration techniques? Are there any books that help with this, "Orchestration for Dummies "? I know if i attended a music school i could take classes but i would like to at least have a grip on it and written some orchestral music before i apply anywhere.
Thanks,
c#minor

[johnQpublic]

Study scores, not books.

[lukeottevanger]

Was going to say the same thing. The more the better.

[c#minor]

Do you recommend any scores that are fairly easy to analyze, to get a grasp of it? I really have never analyzed sheet music outside of piano so an orchestra is quite daunting.

[lukeottevanger]

Well, you could take the plunge wherever takes your fancy - as I said, it's important to just immerse yourself in all sorts of stuff. Having said that, it's not a bad plan to start on something colourful, well-orchestrated, containing unusual points of detail and written for a standard Romantic orchestra, so as to give yourself a good base to work from. Something like Scheherezade is ideal - it is at the same time lucid and fascinating, with all sorts of interesting points to note, from the orchestral layout of the opening chords onwards.

[greg]

you can start with the easier stuff, like a Baroque or Classical score, which is pretty easy to read and follow along. Vivaldi, Mozart, Beethoven, when it comes to reading scores that's what you start with. After that, you can move on to Brahms, then maybe Scheherazade, impressionist scores, and lastly, the really complex ones (Strauss, Mahler, 2nd Viennese, modernist scores, etc.)

[lukeottevanger]

Personally, I don't think this graded approach is worth it. It's certainly not how I did it (and my orchestration is the one thing about my composing skills which I know is good!). I just started by surrounding myself with as many scores as my little teenage hands could grab hold of, and also, for a short while, reading some orchestration textbooks just to help me know what sort of thing to look for. I could go into detail about why I think that the graded approach is not necessary, but I haven't got time right now. 

[Montpellier]

How you study the scores is important.  It's rarely satisfactory to just go through the score while listening because too much is happening.  The music usually sounds simpler than it is.  Better to choose a brief passage, listen to it extensively then study the passage in the score with the performance fresh in your mind.   Then perhaps work with book and music together.   
Why not start with something straightforward then move onto more complex composition later.  I think I started on Eine Kleine Nachtmusik and moved on to Elgar's Enigma Variations.   

However, if you want to study a selection of examples against a commentary and like Rimsky Korsakov, try the online book:

http://www.northernsounds.com/forum/forumdisplay.php?f=77

It's an expansion of Rimsky's own book so you'd have to turn to other examples/scores as quickly as possible.   

[lukeottevanger]

Why not start with something straightforward then move onto more complex composition later.  I think I started on Eine Kleine Nachtmusik and moved on to Elgar's Enigma Variations.   

Why not? Well, IMO, doing something like this gives a rather misleading image of what modern orchestration is - at least, a rather staid one. Unlike harmony, which is best studied in a quasi-chronological way, because in that case this also = both natural harmonic 'evolution' and greater harmonic complexity, orchestration developed to an extent at the whim of instrumental technology, and so a chronological study (which is essentially the same thing as starting with small, simple orchestration and moving on to more sophisticated ones) gives the misleading impression that more recent instruments and effects are whimsical add-ons. Starting with Bach, Haydn, Mozart and Beethoven one absorbs the idea that double basses are made to generally double cellos, that trombones and triangle are for special effects, that trumpets are best reserved for triadic figures and so on and on. Of course you can shed this later on, but these features are best viewed as part of the historical context from the off. Whereas if you start with something a little more colour-oriented - like my previous R-K example - from the very off you are aware of the orchestra as a great resource to be plundered. FWIW, my first score was Brahms PC1, followed by Bruch VC1, Beethoven PC5, Vaughan Williams symphony 2, Scheherezade, The Planets... My score library now is hundreds or perhaps thousands of books (see the 'mystery scores' thread!), but I can still remember the thrill of these first volumes - and particularly, the thrill of my first 'colouristic' scores, which were what really got me going on orchestration. That's why I'm making this fairly minor point on this thread.

[longears]

Are there any books that help with this

See Walter Piston, Orchestration

[greg]

How you study the scores is important.  It's rarely satisfactory to just go through the score while listening because too much is happening.  The music usually sounds simpler than it is.  Better to choose a brief passage, listen to it extensively then study the passage in the score with the performance fresh in your mind.   Then perhaps work with book and music together.   
Why not start with something straightforward then move onto more complex composition later.  I think I started on Eine Kleine Nachtmusik and moved on to Elgar's Enigma Variations.   

However, if you want to study a selection of examples against a commentary and like Rimsky Korsakov, try the online book:

http://www.northernsounds.com/forum/forumdisplay.php?f=77

It's an expansion of Rimsky's own book so you'd have to turn to other examples/scores as quickly as possible.   

what a course!
thanks for the link, i'll have some fun reading through this one....


well, whatever works for C#m..... whether it be starting with any score or easy scores. I know I didn't learn guitar by starting with Mary Had a Little Lamb...... either way, more important than which score you start with is to have tons of scores, because if something doesn't make sense to you in one part of the score you may eventually come to understand it through studying other scores. I don't know if you already have a ton of orchestral scores or not, since IMSLP is down....

[greg]

See Walter Piston, Orchestration

that's a good one, too, first one i read

[(poco) Sforzando]

Personally, I don't think this graded approach is worth it. It's certainly not how I did it (and my orchestration is the one thing about my composing skills which I know is good!). I just started by surrounding myself with as many scores as my little teenage hands could grab hold of, and also, for a short while, reading some orchestration textbooks just to help me know what sort of thing to look for. I could go into detail about why I think that the graded approach is not necessary, but I haven't got time right now. 

So study scores and books - of which there are a number of standard texts, ranging from the classic Berlioz/Strauss and Rimsky, to the slightly musty Victorian treatises of Forsyth and Prout, to more modern books by Piston, Kennan, Blatter, Adler, Rauscher, and doubtless more. I have most of these, and have learned from them; perhaps the greatest benefit are things like fingering charts for the woodwinds, slide positions (for the trombone), and pedal diagrams for the harp.

But there are two things missing in all these texts that I think could be of great benefit to the student orchestrator. One is that I've never seen an orchestration book that has sections written by experienced instrumentalists themselves, to tell the student what can best be managed on each instrument and where the pitfalls lie. I would love to read an essay on the violin by a renowned concertmaster, where he indicates which orchestral passages are notoriously tough, or which need touching up because of composer error. (There are always sections on the various instruments, but they are always by the single author, who may or may not play most of the instruments himself.)

The second deficiency is that I've never seen a book that shows examples of problematic orchestration and how to correct it. Toscanini premiered a piece by Abram Chasin once, and during rehearsal Chasin was chagrined to find one section sounded just awful, not at all what he had expected. Afterwards Toscanini told the composer: "Chasin! that middle section was not what you wanted, was it?" And that night Toscanini went home and rescored the entire passage. Lessons like that, or the kinds of lessons where Rimsky set Stravinsky to orchestrating passages of Rimsky's own unpublished works, and then the master showed the pupil what he did differently, could be worth their weight in gold.

[greg]

The second deficiency is that I've never seen a book that shows examples of problematic orchestration and how to correct it. Toscanini premiered a piece by Abram Chasin once, and during rehearsal Chasin was chagrined to find one section sounded just awful, not at all what he had expected. Afterwards Toscanini told the composer: "Chasin! that middle section was not what you wanted, was it?" And that night Toscanini went home and rescored the entire passage. Lessons like that, or the kinds of lessons where Rimsky set Stravinsky to orchestrating passages of Rimsky's own unpublished works, and then the master showed the pupil what he did differently, could be worth their weight in gold.

good thing we have notation programs and MIDI nowadays!
(i'd end up making my whole scores sound like something that i didn't want  )

[Mark G. Simon]

good thing we have notation programs and MIDI nowadays!
(i'd end up making my whole scores sound like something that i didn't want  )

The MIDI playback is useful up to a point. But the balance is artificial. You don't hear the various timbres and lines blend together the way they would in a real orchestra. For that, there's no substitute for a real orchestra.

I have found that the most useful experience for me in learning to orchestrate, has been simply to play in an orchestra, sit through the rehearsals, and listen to what kinds of passages require constant rehearsing, what kind of writing tends to create train-wrecks, which pieces fly without effort. You do that for a number of years and you can accurately judge how your orchestration is going to sound; you can actually hear the orchestra in your head.

[Montpellier]

Why not? Well, IMO, doing something like this gives a rather misleading image of what modern orchestration is - at least, a rather staid one. Unlike harmony, which is best studied in a quasi-chronological way, because in that case this also = both natural harmonic 'evolution' and greater harmonic complexity, orchestration developed to an extent at the whim of instrumental technology, and so a chronological study (which is essentially the same thing as starting with small, simple orchestration and moving on to more sophisticated ones) gives the misleading impression that more recent instruments and effects are whimsical add-ons.

Certainly won't disagree, these things are opinions, and it does partially rest on what the person wants.  An interest in (for example) identifying instruments and doublings while listening would probably be different from a composer wanting to learn to write orchestrally.   
The arranger is another case, attempting to transfer music from one instrument/combination for another - it may be their own music.  The tactics are different.
                 
The ultimate aim of being able to construct the sound of a score in one's mind is, to me, what studying scores is about.  But if an enquirer is totally immersed in something like The Rite, then enthusiasm alone might make that the best bet.  It's a question of balance in a way - how much a newcomer can handle without getting frustrated.  If a composer, what sort of orchestral resources do they have at their disposal?   There's no sense composing something on Mahlerian scale if the available orchestra is something like 2221 2220 a couple of timps, cymbal and strings of maybe 86441. 

I also agree about harmony and there'll always be a debate about how much theory one ought to have for composing.  My thoughts are divided.  Having enough to be able to handle one's material is about right but that's 'how long is a piece of string'. 

[johnQpublic]

or the kinds of lessons where Rimsky set Stravinsky to orchestrating passages of Rimsky's own unpublished works, and then the master showed the pupil what he did differently, could be worth their weight in gold.

I did something similar to that in my orchestration class. The professor would give us a piano reduction of an interior portion of an unidentified orchestral work and have us score it. After turning them in, we would then see the original orchestral score and make notes as to what we did well and poorly.

Learning to play (amateurishly of course) many instruments like Hindemith is a very good way to understand the ease and difficulties.

[(poco) Sforzando]

I did something similar to that in my orchestration class. The professor would give us a piano reduction of an interior portion of an unidentified orchestral work and have us score it. After turning them in, we would then see the original orchestral score and make notes as to what we did well and poorly.

Are there any patterns that emerged to suggest what the student orchestrators did well or not as a rule?

[johnQpublic]

Patterns, no, I don't recall so (I had that class in 1974, so my memory might be a little fuzzy.....little?....did I actually was the word "little"?... )

But what was hardest to figure out were things like how composers would create harmonic filler and percussion parts.

[Montpellier]

Balance is a problem for beginning students.   Likewise laying out the parts to get the desired effect and textures.  A solution at pianissimo will be different from fortissimo, so will the dovetailing woowinds.  Anything above mf in a full orchestral tutti and every important note in the harmony must be represented in the brass - they swamp everything else so the other instruments just double as best they can in their most effective register.