Comments

Well, you need to know the rules before you can break them. So I think the historical practice should come first. Perfect that, understand the piece thoroughly in the context of when it was played (political, social, artistic etc.) and only then maybe can you can start to emphasise particular elements that appeal or start to personalise it. The problem can come sometimes when the artist can overcome a piece - we no longer hear the composer but the interpretation. I suppose there should be a precedence:- notes; context; individuality. The later should definately come last!
j I do what my teacher says to do. Sometimes that's the score, sometimes it's historical practice. We've had discussions about both Bach and Mozart, where my initial interpretation hasn't been historical. She's told me to sit with it a while, try it both ways, think about it. She clearly has an agenda but she also doesn't want to make me feel steamrolled.

I admit, though, that through no fault of hers, I sometimes do feel rolled over by the weight of historical precedent, and I'm grateful for my teacher's approach as I work through it. One example is the Bach cello suites. I started out playing them (on viola) with too much rubato. I was taking time, speeding up and slowing down, and enjoying doing that. I found the recordings a little stiff. My teacher told me it was too much. She said I should keep strict time, the first 16th note of a group should always land exactly on the beat. Sometimes she might, within a beat, move ahead and then make up a little time, but it was so subtle when she did it I found it frustrating and almost impossible to hear. I did it her way, reasoning that she's the teacher and she knows what she's doing, but grumbled mentally.

Then, a few weeks or months later, the cellist busker who plays at Harvard Square in the mornings, was playing the same piece and I heard him on the way to work. He's got this long-haired romantic aura about him, and he plays with a lot of passion. And he was applying this heavy, rubato-and vibrato-filled, romantic style to the Bach cello suites. And I had an aha moment. On one hand, I kind of liked it; he was clearly getting into the music, and, in fact, people put $ into his case when he finished (with a flourish, of course). But on the other hand, it was a little like the bad accent Laurie is talking about. It was too much. I don't think I ever took it that far myself, but I suddenly understood much better where my teacher was coming from. Bach didn't intend the cello suites to be played like that.

I would go with my best interpretation based on my understanding of the score and the historical practice. Score is the source of both truth and confusion when it comes to a clash in interpretations. If the score is self-evident with respect to the composer intention, we wouldn’t have the interpretation problem to start with. Historical practice on the other hand varies from individual to individual so it’ll still require a personal choice as which master’s interpretation I would choose to follow. Interpretation is a relatively new invention, I have been told. Before the nineteenth century, musicians were expected to play with good craftsmanship and good taste, and that was all.
In one of the rooms at the Groningen Conservatory there was a cartoon on the wall with a wigged eighteenth-century music teacher telling his pupil: "Du sollst dieses Stückchen authentisch spielen, mein Junge!" ("You should play this piece authentically, my boy!") I am still learning how to play Baroque pieces (both the Bach Double and Handel's Sonatas), and listening to recordings of both, I am all too painfully aware of how little I know about how to play them! I'm not talking about technical facility; that is coming along. But as someone else has said here, you have to know the rules before you can break them. And as I admire the crisp fluency of the musicians in the recordings (one of whom is Heifetz), I know that before I can even begin to attempt my own interpretations, I have to know the language much better. So, I will stick with the score, following my teacher's suggestions, and learn as much as I can about historical practice. Only then will I attempt my own interpretation (within in the limits of good practice, of course). Steal from the best, learn from the rest. Where that gets interesting is "the best" seems to keep changing and evolving with time... I follow the score as closely as possible. But sometimes there is something in the score that doesn't totally make sense. Then, one has to use one's judgement about what the intent of the composer may have been. One's judgement/interpretation can be based on historical performance practice, other pieces of music by the same composer, possibly with different instrumentation, and how a similar motif was handled.

Also, in a string quartet, which is where I do most of my playing in, one has to resolve one's opinion with that of who one is playing with. Sometimes one can have a differing opinion and it still works within a piece, sometimes it is important that the motif is played the same way by all members.

Liszt believed that written music was an imperfect representation of an abstract sonic material.
I think that this thought is important to consider when deliberating what a composer wanted, because I think that our search for what the composer wanted relies to heavily on literally interpreting the score. For instance, I played in an orchestra with Neville Marriner, and he had half of the second violins play spiccato in the opening. It's very clearly marked slurred, but in his opinion, it added a clarity that would actually approach what Mozart would have likely wanted. At the same time, I don't change notes or rhythms or rearrange articulartions (gee, I think that I should slur the last movement of the 2nd Wieniawski concerto). EDIT: Opening of Mozart Sym. 41, last movement.