...is a brilliantly ambiguous dynamic marking. It can mean that something should sound louder than the precedent, as well as softer – depending naturally on the precedent dynamic; p in the former, f in the latter. It cleverly provides a more street-credible (or noble, as it were) version of the somewhat uninteresting mf. It can also help to override the danger of the music becoming dynamically too polarised or megalomaniac in nature, although it’s use in full orchestral tutti (as seen in some of the later works by Sibelius, namely 5th and 7th symphonies) communicates a feeling of grandeur that doesn’t even sound withheld or reserved, to name but one contradicting example.
The obvious danger in using this marking is that the performer is left 'a son propre chef’ [rt. from his own initiative] to decide in what mood and volume to play – but this is the case (but not too strictly) anyway, at the end of the day.
Or at least, more often than composers would like to admit. But then I wouldn't know, I'm on the hoodblinked side.
