Thank you again for all your ideas. I realize now that my original problem was not well-formulated: it is not so much polyphony that is a problem , as there are work-arounds that can make this possible, if not very conveniently. Rather, it is that NMC thinks in terms of staves. This is fine for instrumental and vocal ensembles, but is not appropriate for keyboard, where a melodic phrase may need to wander back and forth between two staves: a couple of well-known Bach examples are attached, which I simply do not think NWC can reproduce. Even Finale seems to suffer from the same problem. At the cost of making the calculation of the temporal durations quite demanding (but clearly doable, or we would not be able to calculate it ourselves!) I feel that were one to be starting all over again it would be good if the user could designate a pair of staves as 'keyboard', and be free to place notes one either of them without redundant rests or unnecessary - and sometimes conflicting - changes of clef.
Thank you to all of you for your very prompt and thoughtful responses. I agree that layering forces one to think genuinely polyphonically, but if one takes the example of a prelude from the 48 for instance, the number of parts will be varying, and often need to switch back and forth between treble and bass clef. When the layers are collapsed, the redundant rests and superimposed clefs are simply unacceptable. I will have a go with the specialist tools mentioned ...
There are many aspects of NWC I love, but trying to enter more than one part to a stave is not one of them. I realize this CAN be done - clumsily - with layering, or even more clumsily by pretending it's a lot of chords. Is there really a fundamental technical reason why NWC could not cope with entering multiple lines on a single stave? Could a note not be tagged as belonging to one part rather than another? E.g. by temporary colour-coding?
