I have taken theory courses on harmonic analysis for a while, whereby we learned principles of analysing the harmonic structure of simple music pieces consisting of 4 voices only (mostly choral works): soprano, alto, tenor, and bass.
I am wondering how one would proceed into analyzing orchestral or opera scores, where you have 10 or 15 voices? Thanks for your comments.
One way is to reduce the score to fewer voices; three, four, or five. Things like octave doublings are collapsed to a single tone (like basses doubling the celli). In some modern orchestrations (like from Ravel) there may be doublings at the fifth or third; these are coloristic like tierces or quints on an organ. Then, an ordinary harmonic analysis is easier. A C-major chord with a low C in the bass is still a C chord even if there are lots of C's, E's, and G's played over it.
One example that may be helpful is to look at Liszt's piano arrangements of Beethoven's symphonies. Another is Ravel's piano version of his Bolero.
Even in (common practice) symphonic works, you seldom really have more than five independent voices. At every moment, each instrument will generally pick either one of 1-4 counterpointal voices (often a single such voice is doubled by many different, perhaps even most instruments!) or take part in some accompaniment ostinato like Alberti bass.
In that light, composing for orchestra isn't necessarily that different from composing for SATB, you just also get access to an enourmously rich palette of sound / dynamics options, given by which voice you map to which instruments at a given section. It's actually quite similar to an organ with lots of registers, except you also have the expressive/articulation freedom in all the individual voices like you'd have in e.g. a string quartet.
OTOH, there are some newer works that heavily rely on more or less complete independence of all single instruments. Ligeti was great at this. But this really isn't something that could quickly be explained here (and certainly not by me).
