IMSLP Forums

In browsing through the site, I've noticed some errors in the instrumentation for some pieces. I'd be happy to make a contribution by correcting any I happen to notice, but I was wondering whether you have a standard format you want used?

For instance, the Schirmer website lists the orchestration for Tchaikovsky's 6th Symphony thus: 3222/4231/timp.perc/str
Kalmus shows it in this format: 3d1, 2, 2, 2 - 4, 2, 3, 1, timp, perc, str

IMSLP lists it as follows: piccolo, 3 flutes, 2 oboes, 2 clarinets (A), 2 bassoons + 4 horns (F), 2 trumpets (A, B♭), 3 trombones, tuba + timpani, cymbals, bass drum, tam-tam (ad lib.) + strings

The IMSLP format is more detailed and much longer, and as it happens this entry is not quite correct - the third flute doubles piccolo, so doesn't need an extra player.

Is there anything to be gained in terms of standardising the instrumentation format so that the instrumentation tag could be machine-parsed in future for a more detailed breakdown of pieces that have a specific instrumentation? As a member of an amateur orchestra, I know that being able to do a search for orchestral pieces that (for instance) have no more than 3 horns and no harp would be very useful when trying to build concert programmes that minimise the need for us to book extra players.

You are quite right, it was a mistake. I've corrected it now.

While it would be convenient in some ways to use just numbers, the percussion section can't be dealt with this way in any case. Apart from the fact that numbers won't tell you which percussion instruments are needed, percussion sections I've worked with usually decide on their own how many players they will use to cover the parts. I suppose in many cases, different orchestras will come up with the same number, but not always. I've never seen this section dealt with by numbers. The standard numbering system covers only woodwind and brass, and not the "extras" -- percussion, harp, keyboards, the odd mandolin part, etc.

Also, there are differing systems for dealing with doubling instruments, and even I (who regularly double within the clarinet family) sometimes can't remember how each system works, though I know the repertoire, so at least for my own section, I already know what's needed most of the time.

A numbering system has been discussed before, but never went anywhere, perhaps because it doesn't give a complete picture - if you have to look at a written out listing to get the number of percussion players and whether there is harp or keyboards, you might as well look at a written out list for woodwind/brass.

Another subsidiary issue where the written out list helps enormously is that for student (and some amateur) orchestras who not only need smaller orchestrations, but have horn players that don't read E flat horn, or clarinetists who don't have A clarinets, the written out list will tell them whether to skip a piece on the basis of availability of transposing instruments and/or transposition at sight.

I think it's a case of gaining some advantage, but at the loss of some other advantages.

By the way, mistakes will happen with any system, and it's not just our editors and librarians -- scores should give full instrumentation on the same page, and conductors/orchestra administrators, or anyone choosing repertoire shouldn't have to wade through to page 168 to find an instrument that wasn't listed on page 1, but it does happen. Anyone filling in instrumentation from a first page will most likely get a correct instrumentation, but not always . . .

Corrections to instrumentations are MOST welcome, and we will correct any mistakes people find.

Incidentally, I added a comment to the page about the common practice of using bass clarinet for 4 notes of the bassoon that diminuendos to pppppp. It is handed over from 1st clarinet anyway, so the timbres match better, and the dynamic is easier on bass clarinet.

I've no strong views about what format should be used. My main thought was that I didn't want to annoy people by making changes in a way that would have to be undone later because there was an agreed format that I didn't know about.

I see the change you've made to the Tchaikovsky. Now I see how you've notated the doubling, I can do the same with any changes I make.

Hi there. I have a habit of writing manual pages for this kind of standardization
This page http://imslp.org/wiki/IMSLP:Score_submi ... nformation should have what you're looking for. I have a template "I" — http://imslp.org/wiki/Template:I that automates this.

This guide is really good. I just want to question one thing: you deal with ad lib instruments in two different ways, which could be confusing:

a) Under chamber works, it says: "flute (+ piccolo), oboe" means that the piccolo part is optional."
b) Under orchestra, it says: "Optional instruments, which can be omitted, should be listed in parentheses. For example: "timpani (bass drum, cymbals) + strings" means that the bass drum and cymbal parts are optional." There is no "+" used here, except before "strings", which in any case might be confusing if in other contexts the "+" means ad lib. (and also conflicts with previous advice to use commas as separators).

I have never seen the "+" indicate "ad lib." To me, this designation would mean the part is mandatory. For instance, here is the relevant sentence from Luck's Music Library's orchestration guide: "a '+' if an extra player is required". So at least some systems don't use the "+" as an ad lib instrument. Apart from the fact that the second example is not consistent with the first, I also think it is not particularly clear, especially to an amateur.

I'm wondering if we shouldn't rather require the ad lib. designation to be written? Your examples would then be:

"flute, piccolo (ad lib.), oboe"
or
"timpani, bass drum (ad lib.), cymbals (ad lib.), strings

Slightly more work, but there is absolutely no ambiguity.

How about something like this:

"flute, (piccolo: ad lib.), oboe"
or
"timpani, (bass drum, cymbals: ad lib.) + strings"

... so that the ad lib. applies to every instrument inside the parentheses?