j.p.

More knowledge is nearly always helpful but that might not be the problem here. Many of the best pieces of music don't actually have the best musical ideas contained within them but they work so well because they have a strong sense of structure and dramatic flow. After all, music is a temporal art. The listener wants to be taken on a journey of some sort. When I'm stuck, I often find it helpful to listen back to what I've written so far and ask myself a few questions from a listener's perspective: Where, as a listener, would I like this piece of music to go next? Should it lead to a climax then fade out? Would the piece benefit from a calmer section? Should the piece explore contrasting material or expand on current material? Should the new material be reached suddenly or do I need a transition? Is there a sense of growth/progress to the piece so far? etc.

For a masterclass on how to continually introduce new material without rambling, I'd recommend listening to some Stravinsky: especially Petrushka, Rite of Spring and Pulcinella. All three are great examples of how to effectively move seamlessly between seemingly unrelated ideas. When you look closely at the scores, you realise that the material you just heard that sounds as though it came out of nowhere has actually been carefully building momentum in the background for some time. Incidentally, all three of these pieces are ballets, meaning that the structure is dictated by the drama to a certain extent. Maybe it would be useful to look at the structure of some of your favourite movies, plays, musicals, operas etc. and see if there are any structural aspects that would translate well to your music?

Maybe you could try putting a fairly easy piece at the start of your recital to give you a chance to ease into it. Maybe something around grade 5/6 that you feel extremely comfortable with. I find that my nerves settle if I get through the first piece without too many problems. Also, you have to remember that the guitar is incredibly hard to play perfectly. I don't think I've ever seen a guitar recital without a mistake in it somewhere. The audience cares about mistakes much less than you do. Unless your audience is mostly classical guitarists, they are going to be impressed by even the simplest classical guitar piece. Most people have picked up a guitar and may even play it to a decent standard, yet classical guitar seems impossibly difficult to them. You may know otherwise but put that to the back of your mind and embrace the fact that the audience are going to be impressed because you are playing classical music on a guitar. Not many people can do that, believe it or not.

I think it's amazing how the accepted canon has taken over to the point that anything outside of it is virtually unknown. I wasn't really aware of Cherubini or his influence myself but I'll be sure to have a listen now. I think even good grad programs can get these things wrong though. I'm a grad student myself. It seems to be taken as a given sometimes in musicology seminars that program music didn't exist before the 19th century when this just isn't true. I don't know if you've ever heard of Heinrich Biber? He was a 17th century composer who was way ahead of his time. In this piece: http://www.youtube.com/watch?v=Z9DJpaxT7wg, he uses Ives-esque polytonality to mimic the sound of a soldiers' camp, with lots of different folk melodies in different keys being played against each other out of sync. He also attempts to represent arrows being shot in the music. Quite amazing considering the date it was written. As with Cherubini, it just goes to show that you shouldn't take the standard history of music as gospel.