Throughout my music education I was taught to furnish my compositions with enough information to ensure that any performer would be able to reproduce my musical intentions. So I suppose it goes a bit against the grain to write a 40-minute Requiem for choir, organ and solo instrument where the solo part isn’t even written down. Not a single note. But is this just taking the easy way out?

Improvisation in classical music isn’t new. Solo concertos of the Classical and Romantic era were often vehicles for performer-composers to show off their formidable talent. The jewel in the crown of a performance would be the cadenza in which the performer’s technical brilliance had free rein. Improvisation has always played an important role in so-called 'classical' music.

Improvisation gives a performance an element of spontaneity, freshness, surprise. The improvising performer is more often than not probably the creator of the work, in a sense leaving the final touch of the composition process until the concert itself.

But what if a composer deliberately relinquishes compositional control of part of the work and allows somebody else to contribute to the creative process?

When I composed my Requiem, that was the basic concept from the very start. I wrote down the choir and organ part, envisaging the sound of jazz trumpeter Arve Henriksen’s improvisations moving in and out of the notated score. I could have maintained full control of the process and notated the solo part in every possible nuance I could imagine with my inner ear. Well, I had to swallow my pride and admit that Arve Henriksen (or, as in the case of the recording, Trygve Seim on saxophones) would be able to improvise a solo part far better than I could notate one. It’s not that it isn’t possible to capture the many different inflections of tone, pitch and rhythm, it’s just that complicated notation would be counter-productive to the free spirit of the solo part.

That’s not to say that the solo part hasn’t been planned; we went through the work together and agreed on where the solo instrument would play (or, more specifically, where it wouldn’t – such as during shorter sections where the choir sings alone). The Requiem is pretty much tonal/modal throughout, and its melodic material lends itself to the kind of improvising I had in mind. For the recording we were fortunate to have with us Ståle Storløkken, an organist/keyboard player best known for his jazz and free improvisations. In addition to adding extra colour to the organ part here and there, he and Trygve Seim improvised transitions between some of the movements. Giving them this amount of freedom meant that they were able to invest considerable emotional energy in the music.

It’s a piece with a strong emotional core, dedicated to young victims of conflict and in particular to those who lost their lives at Utøya, Norway, in July 2011, and the freedom of an improvised solo part brings the audience much closer to the emotion. You remove the barrier of notation, and there’s an immediacy and freshness to the music.

In the end it still feels like 'my' work because it’s how I imagined it, more or less, and I know that the final result is greater than the sum of its parts, or at least better than anything I could have put down on paper.

Andrew Smith’s Requiem, with the Nidaros Cathedral Girls’ Choir, Ståle Storløkken (organ), Trygve Seim (saxophones), conducted by Anita Brevik, will be released on the 2L label [2L-150] in mid December; the score is published by Norsk Musikforlag, available direct from www.musikkforlagene.no