As Bruce Ellis Benson explains in chapter two of his book, The Improvisation of Musical Dialogue, we tend to think that a musical composition is finished when the piece in its “final” version is written down. However, there are a number of assumptions that we should question in connection with such a conclusion. First, why assume that a process of revisions always leads to a better version, much less to the “perfect” version? Beethoven, for example, was known for ceaselessly revising and offering a number of variants for musical passages and even entire sections of his symphonies. Even if we grant that his revisions generally improved his work, why should we necessarily conclude that they always did? Second, is it not the case that pressing deadlines, familial responsibilities, or creative inertia also factor into to a piece coming to completion. That is, the artist may not in fact be satisfied with his or her final version, and yet the work must be brought to a close. If this is the case, then we might even say that the composer is aware of the imperfections in his or her work-the places that at some later time, he or she if given the time, would want to change or develop the work. Third (and closely related to the second point), is there a sense in which a composition becomes “fixed” and definite when written down, or is it the case that even for the composer there is a certain “indefiniteness” and indeterminacy involved in his or her work even when the composition is “finished”? Arguing for the latter, Benson states that although composers have “reasonably” definite intentions, “it would be impossible for their intentions to encompass all of the details of any given piece.” In other words, often or perhaps even most of the time, the composer himself is unsure exactly how he wants every aspect and detail of the work to sound until the piece is actually played with a specific group and with very particular instrumentation. As Benson highlights, Mozart would at times perform different versions of the same piece to a group of friends in order to seek their input as to which version they thought best. Having performed in several jazz orchestras and dabbled in jazz composition myself, I find this claim rather convincing. It was often the case that our director, who was an accomplished composer and arranger, would present us with his scores and then during the rehearsal time, numerous changes would be made-changes that he could not foresee until the actual music appeared. Clearly, he had a definite intention of how he wanted the piece to sound, yet the various intricacies of tempo, dynamics, and so forth were not solidified.
But what about after all these things are made more precise, is it the case that at that point the work is finished? This leads us to the next issue, viz. what counts as the correct interpretation of a piece? To illustrate, Benson cites Edward Cone who comments on the difficulties performers face in playing Chopin’s music,
The performer’s first obligation, then, is to the score-but to what score? The autograph or the first printed edition? The composer’s hasty manuscript or the presumably more careful copy by a trusted amanuensis? The composer’s initial version or his later emendation? [and so on].
To be sure one might give good reasons for choosing and preferring one version over another. But still we must recognize that performers, conductors and arrangers play a role in the process of composing. That is, the performers, conductors and arrangers in some genuine sense continue to compose a work that is already as it were “finished.” Yet, as we stated earlier, composers certainly have some definite intentions, but how extensive those intentions are is another issue. Also, the fact that composers may not even be cognizant of places of indeterminacy in their own compositions until the music is actually performed suggests that a determinate intention, though having some definiteness to it, may also “contain” what we might call a kind of built-in-flexibility that does not destroy its identity.
Bruce Ellis Benson, The Improvisation of Musical Dialogue: A Phenomenology of Music
. (Cambridge: Cambridge University Press, 2003), p. 67.
 Edward T. Cone, “The Pianist as Critic,” in The Practice of Performance Studies in Musical Interpretation, p. 244, as cited in Benson, Improvisation of Musical Dialogue, p. 70.
 This idea of on-going composition strikes me as having something in common with Gadamer’s hermeneutical insight that texts always exhibit an “excess of meaning” upon which tradition builds. Elucidating his position, Gadamer writes: “Every age has to understand a transmitted text in its own way, for the text belongs to the whole tradition whose content interests the age and in which it seeks to understand itself. The real meaning of a text, as it speaks to the interpreter, does not depend on the contingencies of the author and his original audience. It certainly is not identical with them, for it is always co-determined also by the historical situation of the interpreter and hence by the totality of the objective course of history. […] Not just occasionally but always, the meaning of a text goes beyond its author. That is why understanding is not merely a reproductive but always a productive activity as well.“(Hans-Georg Gadamer, Truth and Method, 2nd ed. trans. and revised Joel Weinsheimer and Donald G. Marshall (New York: Continuum, 2004), p. 296).