I know that when you have one stave with S/A or T/B combined you have the convention that the sopranos (or tenors) always have tails up, and altos (or bases) have tails down.

However, when it comes to 8-part harmony compressed into 4 staves, does that convention still apply?

I mean, would Sop 1 be tails up, and Sop 2 be tails down (and the same for Alto 1 / Alto 2, Tenor 1 / Tenor 2 etc), or is the split usually done purely on note pitch (the "ones" always get the higher pitched note) regardless of the tail direction?


enter image description here

First note: is the Bb or the Eb part 1?

Also, for bar 3, is it safe to assume that the pitch defines the part, since there are no other clues to go on?

1 Answer 1


The first way you describe it is standard: S1, A1, T1 and B1 will always have upstems, and the rest have downstems. This will be true even if, for example, Soprano 1 temporarily goes lower than Soprano 2—in fact, that's the point. Otherwise, a person singing soprano 1 wouldn't know which line they were supposed to sing, or at least wouldn't be able to tell with ease. Of course, if Soprano 1 is almost always below Soprano 2, you should switch the parts. The "ones" should be higher than the "twos" more often than not.


In the first beat, the Bb is part 1 and the higher Eb is part 2.

  • @PatMuchmore "you shouldn't connect the notes to the same stem ever" - Does that "never" only apply to vocal scores? It is very common in instrumental scores to have two instruments (e.g. orchestral woodwinds) on one staff written as "chords", to save vertical space. But of course vocal parts may be divided in which case both the 1's and/or the 2's are written as "chords," and that doesn't apply to monophonic instruments with only one player per musical line, as in a conventional orchestra.
    – user19146
    Oct 20, 2017 at 20:19
  • @alephzero, that's right, I meant in vocal scores. Your point about Soprano 1 potentially splitting hadn't even occurred to me, that's one good reason. But the larger reason is simply that a vocal score, unlike an orchestral score, is very often a performance document. Individual soprano 1s and soprano 2s will most likely be reading from the score, and thus need to be able to easily tell which part is theirs. Trumpet 1 and trumpet 2 will almost certainly be performing from separated parts and won't see the chords that are shown in the score. Oct 20, 2017 at 20:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.