As I entered my doctoral degree, I thought I might have the chance to work the study into my program, but as graduate work goes, I got too busy, ended up going another direction in my research and lost the chance to have plenty of student pianists nearby to test my hypothesis. I thought it might be relevant to share the general outline of the study. Maybe someone will one day take it up and test it!
The procedure is simple enough: have two groups of pianists, likely undergraduates though their technical capabilities by no means need be similar. Each group would be given a score of some obscure work, likely from the early classical period, with relatively intermediate technical challenges. The score would make no reference to composer or style. I would recopy the score on notation software myself and include only the essentials: notes, rhythms, tempo indication, and meter. Dynamics, articulation, phrasing, metronome marking would all be absent.
The test group would be given free rein to practice and prepare the score for performance in a given time period. The only stipulation is that they may not consult with any other person in their preparation of the score.
The control group would also be given free rein to practice and prepare for performance in the same time period. They also may not consult with any person in their preparation, but, they are given a recording of the score which they must listen to every day. In the recording, which I would make with an attempt to sound stylistically appropriate, they would hear distinct choices in terms of tempo, articulation, dynamics, phrasing, rubato, etc.
All participants would, after the same amount of preparation, record a final performance. These recordings would be sent to adjudicators. These professional musicians would be aware of the score, plus an edited score representing the distinct choices I made in the recording. Adjudicators would be asked to grade how closely each group adhered to distinct, observable and (relatively) measureable interpretive choices in the recording.
My hypothesis is that the control group would make interpretive choices similar to the recording, more often than the test group would. As my goal in the recording is to not make controversial interpretive choices, I suspect that students in the control group would, without realizing, adopt the logical interpretive choices that I had made. While the test group may also make several interpretive decisions similar, given stylistic conventions, inevitably, something such as exact metronome marking, or articulations in a melody, or dynamics, will vary given complete freedom.
Upon further thought, it may make sense to make one controversial interpretive decision in the recording and see how many of the control group go along with it.
Secondly—What I would include in the score could change. I think it’s important to have as blank a score as possible, so that people’s artistry would be observable on a nearly blank slate. Perhaps I wouldn’t even need a tempo marking, “Allegro” for instance. That would be one way to see who in the control group would resist the pull of recordings enough to question what they were hearing. For instance-imagine having no tempo marking for the opening of Mozart’s Sonata K 545, and hearing it played adagio. One could feasibly, if you never heard this work before, yet intimately understood the style, not question the choice of tempo at all.
Thirdly, it would be interesting to run this study with proficient high schoolers making up both groups, as well as only graduate students, even run it with only professional musicians. Then compare the rate of variance at all 4 levels. What if, on the whole, the control group’s interpretations adhered to the recording at the same rate greater than the test group, whether or not we are dealing with high school musicians, or professional musicians?
I think the results of such a study would be fascinating. None of this is meant to discredit professional musicians, or students. The simple aim is to observe the roots of our artistry, and to find one way of explaining how our general sense of style in interpretation might have a fundamentally different basis than that of artists when the composers of the classical canon were themselves writers of ‘new music’.