Breaking my own rules
I usually have three or four arranging projects on the front burners and many more unfinished projects in my "work in progress" file.
Maybe because last summer I had just told the world that I don't like to transcribe orchestral music, thus suggesting to myself that I can't or shouldn't do that, I suddenly got interested in some big orchestral transcription projects. At the time, I was also preparing my friend Jeremy Van Hoy's fantastic Wagner Ring Cycle synopsis for publication, and accidentally fell in love all over again with the idea of the orchestral paraphrase.
I'll also admit that sometimes I hear or see projects done by other arrangers (not Jeremy's) and think "this was a great idea and kinda worked, but it just wasn't done well." Voila - the birth of a new arranging project.
I'm in the final stages of 're-imaginations' of Respighi's Pines of Rome, the first movement of Mahler's 5th Symphony, and a breezy paraphrase of Puccini's Tosca, all for a large brass ensemble (444.12 timp perc). This instrumentation is the same as I have used for my version of Pictures at an Exhibition and Ravel's Alborada del Gracioso (both NOT orchestral transcriptions but rather orchestrations of the composers' original piano scores). It's also the same instrumentation as Jeremy's Ring Cycle.
While all three projects are challenging in their own way, the Respighi is proving to be the most difficult, with each movement presenting unique obstacles. I got sucked in to the project by imagining the 2nd (Catacombs) and 4th (Appian Way) movements which were, relatively speaking, easy to score. Even the first half of the 3rd (Janiculum) movement was not too difficult. But once started, my OCD would not let me leave it alone. I'm just about there.
Respighi was a fantastic orchestrator and it's really tempting to emulate his orchestral colors. Frankly, this is what I dislike the most about most orchestral transcriptions. I love the sound of brass instruments playing well crafted music - I intensely dislike the sound of arrangements that try to make brass instruments sound like other instruments.
So, for me, the trick is to distill orchestral music down to the most important basic elements, and then recolor them; that is melody, counterpoint, harmony, texture, and balance. Respighi made his own two piano version of Pines and, while unsuccessful (IMO) as a piano transcription, it is useful in that it reveals the musical elements that Respighi himself valued as most important. As is usual for me, I refer to both versions (orchestral and piano) often to learn about the composer's own orchestration toolbox. There are tricks to be gleaned.
With orchestral transcriptions, balance and effect can be the trickiest to translate to brass.
Orchestra scores show the individual instrumental lines but fall short of describing the composer's thoughts about how the sound of a section of string instruments balances with a flute solo, for example. How many violinists did the composer hear in his head when he wrote the passage? Were they pros? Did they all play on well-crafted instruments? The twenty-four violinists of the Philadelphia Orchestra will not sound like the sixteen violinists of the University of Southern North Dakota at Hoople Symphony Orchestra (or USNDHSO for short - thanks to a friend for setting me straight on this), if I may drive my point home with a Schickele-ism.
You can chase yourself down the rabbit hole with the finer points of this consideration - the parable is you must consider these orchestral balances, at least generally, when coloring and balancing your arrangement.
Also keep in mind that tessitura plays a role in orchestral balance as it does in the final brass product. A single trumpet player on a high C can bury just about any ensemble. A section of violinists playing that same C or even an octave higher can penetrate, though not to the extent of the aforementioned trumpet player; OR it can have a fine gossamer sound. The gossamer is far more important than the actual tessitura of the note you're transcribing, so the idea is to create an effect, not faithfully register the pitch itself at register.
The same is true of florid passages of notes scored originally for woodwinds and celeste (I'm thinking of the 'wind in the trees effects' 2/3 of the way through the Janiculum movement of Pines). Brass simply isn't going to achieve this effect well, so if you try to score these notes faithfully for brass, it will just sound silly. Some brass nerds may ooh and aah, but a musical connoisseur will certainly be annoyed at the distraction. Another solution is in order. I'm using a combination of softly muted brass in medium high ranges, along with available keyboard percussion instruments to craft an effect - and it's hard damn work to sort it all out.
Epiphany - even just writing about these subjects and trying to crystallize more abstract thoughts into words has helped me, even as I type, to arrive at some solutions for these effects. Perhaps this is the point I'm trying to make - transcribing, orchestrating, and arranging isn't just about assigning notes. As I've mentioned before, getting the notes into the score is just the first baby step - the hard work comes from making thoughtful decisions based on desired outcomes.
Intent is everything.