Where do accidentals go when voices overlap?

Question

When two different voices on the same staff have overlapping note ranges, they get shifted and written side-by-side. For example:

What happens if there is an accidental on a note of the voice that was shifted to the right? For example, if the E in the first illustration is flat, or if the right B in the second illustration is flat. Does the accidental go between the two stems or to the left? If the latter, is there a way to disambiguate to which note the accidental applies (as in the second illustration)?

Sibelius always puts the accidentals to the left but it looks a little odd and can be ambiguous. What's a good reference for this?

EDIT: To clarify, this is a totally invented example. I have not seen this case happening, but one could notate it using a notation program. The question is whether there is an standard way to notate this, and the answer could totally be that there isn't.

Answer 1

In the first example, there is no ambiguity and the accidentals should be entirely to the left. Although you're right that this might divorce them spatially from the notes they apply to, the line or space that the accidental is on makes it clear which note it applies to. They should not be put in between the C–G and the A–E, because that makes them no longer appear to be on the same beat.

The second example is more problematic, because there's a B in both the upstem and the downstem group. In this case, the standard solution is to still put the accidentals to the left, but you'll need to show a natural and a flat so it's clear that only one of the B's is being altered. So, in your specific question wherein only the downstem B has a flat, you would put both a natural and a flat sign (in that order) on the B line to the left of the whole thing. As several comments have pointed out, this is a highly unusual circumstance, but that's generally accepted convention anytime an augmented (or doubly, or triply augemented) unison is notated. You might look to some of the piano sonatas of Galina Ustvolskaya for some similar situations. Some of her solutions are unconventional, but are starting to be used more and more by other composers. They are almost always very clear.

If you want to really get into the nitty-gritty of notation like this, you should definitely get a copy of Gardner Read's Music Notation book, and Kurt Stone's Music Notation in the Twentieth Century.