I've absolutely had to impress the importance of dynamics on performers; in fact, I think I do that every time I coach one of my pieces. I encounter the "pitch first" mentality too, and I often have to describe a sort of shifting priority-scheme for the piece whereby some parameters have to be "righter" than others at certain times, and at certain other times, other parameters have to be "righter." Sometimes pitches are more important than dynamics for me, sometimes vice versa; sometimes the profile of the gesture is more important than either. I'm not sure that this approach isn't counterproductive, though.

Obviously, it would be optimal to work with performers who can give all the aspects of the piece the necessary attention, but that's been pretty rare in my career to date.

By the way, I know this isn't the best place to make such an announcement, but there's a very short piece of mine that's receiving its premiere in New York tonight. . .  If anybody on the BBC radio 3 message board happens to be in Brooklyn, I'd encourage you to check it out.

I think this is true even with my music, and I don't pay as much attention to dynamics as some others do. It can be difficult for the performer to tell which are the most important parameters at a given time - and sometimes I resort to telling them that the pitches are not as important as the speed of a phrase, or the dynamics, or the shape of the gesture, or whatever. At other times I have to insist on the pitches! But I've been fortunate enough to work with quite a number of musicians who seem to know intuitively (or perhaps rationally) what I'm getting at, and tailor their performance accordingly.

On the other hand, I doubt it if I've ever had a performance where every detail of pitch, rhythm and dynamics (as well as other parameters) has been realised exactly to my specifications, but deep down I'm a pragmatist and don't believe in being more fussy than is necessary to produce a musical performance (and I use that term deliberately, in the full knowledge that it has pejorative connotations for some...). I like the performer to take a personal approach to interpretation, and I've noticed that - at least with some performers and particularly with orchestras, where thinking as a group is the norm - the more detail there is in the score, the more detail gets left out, or smoothed out. And if different players in an ensemble happen to observe different details (which happens often due to insufficient rehearsal time) the performance can become grey, lacklustre and poorly integrated - and in the worst cases completely incoherent (who hasn't heard a few of those?).

Of course there are musicians who have the skill accurately to realise every notated parameter of the score, but I think those that can also bring personality and interpretative musicality (there's that word again!) are few and far between. And quite frankly I don't get asked to write for them, by and large! I'm much more likely to be writing for musicians who spend 90-99% of their time playing and learning older music, and who often say they don't have the time to learn really complex things, despite having an interest in performing new pieces. Of course, that's one of the reasons they decide to play my music - because it's not as complex as a lot of new music. (I'm not saying that's a virtue by the way, just the way it happens... )

And I frequently change dynamics in rehearsals if it occurs to me that something might work better another way...it's all about getting things the way you want them, isn't it?

I think (if I've understood you correctly) that I've been messing around with the same sort of thing lately.

Regarding everything to do with students and dynamics, I've noticed that in a lot of institutions there is a tendency to encourage student composers to treat workshops as performances (i.e. to submit complete works) rather than as opportunities to learn from performers.
I think it would also help if the students knew a little of what to expect before the session as well. I think that getting them to prepare passages in different ways is excellent and would really bring home the possibilities of the session.
I could go on about my workshop bugbears, but maybe not today!

One reason I would posit why students aren't so explicit when it comes to dynamics is that in Sibelius it's a bit of a pain   Sibelius puts in the notes for one just fine but as soon as one requires dynamical markings one finds unself rather alone in the world, dragging staves apart to make room for hairpins and never being sure if what you're writing is an indication of "expression" or "technique".  Or maybe it's just me.  I usually find myself scribbling most of my dynamic markings on the printed-out scores afterwards from the original manuscripts.

I see a hairpin intersect a p in the clarinet line, first bar. 

How do you work with quartertones?  (different people have very different philosophies when dealing with them, as they do with normal notes, I imagine).  I mean, do you treat them in terms of their sound, or do you have some methods for dealing with them?.

Or (possibly) more importantly so that the phrase doesn't cut across a change of clef.
The doubling indication is a variant on Richard's practice in Temptation (? I saw it in Toop's article on the Four Facets). Just using underlining (which Richard uses) didn't make it clear enough for my purposes so using the smaller font helped.

incre- thanks for spotting that problem in the clarinet part (and I'll need to fiddle a bit more with things as they are).
I'll come back to quartertones later. Got to go into town before the shops shut.

Just spotted this, hh - hope it went well!

Not sure it's my place, but I can say that it did indeed go rather well...congrats hh!!!