Poster session 4

Quantitative Analysis of the Musical Style of Mozart: How Many Persons Should Play Eine kleine Nachtmusik?

  • Michiru Hirano (Tokyo Institute of Technology)
  • Hajime Murai (Tokyo Institute of Technology)
  • Takehiro Inohara (Tokyo Institute of Technology)

Eine kleine Nachtmusik K.525, one of Mozart's most popular works, holds many mysteries, one of which has to do with how many players are intended to play it. The work indicates five string parts, but there is no instruction on whether it should be played like chamber music (wherein each string part is played by only one person) or as orchestral music (i.e., with each string part played by more than one person). For most of his other instrumental works, Mozart specified a form that would indicate how many players were intended. For example, works titled “string quartet” are always played as chamber music, and works titled “symphony” are always played in orchestral form. Therefore, it seems likely that Mozart would also imagine Eine kleine Nachtmusik as played with a certain form. Unfortunately, there is little information about this work: the purpose of composition is unknown, and no document remains from the time of its composition concerning its intended form. Four other Mozart works—Divertimentos K.113, 136, 137, 138—have the same problem.

Lacking infallible evidence, an analysis and investigation of the musical style of these works might reveal valuable clues for solving the problem. That is, if Mozart composed chamber music differently from orchestral music, and the clear difference between the two forms could be objectively recognized, then a work lacking a clear form might be assessed by looking for these differences and considering its similarity with other form-clear works. This study uses a quantitative approach to measure features of musical works in the scores to avoid the subjectivity of analysts on which conventional musical analysis is usually based.

It is often said that in chamber works, each string part tends to have more independence, more of a solo quality, than in orchestral works[1]. Therefore, at the beginning of this study, three types of “uniqueness” are identified and calculated for each part in the strings. “Rhythmic uniqueness” is an index of the frequency at which a note given to a part is unique in terms of its duration among notes of all parts placed at the same time. For example, when a half note is given to the first violin part and eighth notes are simultaneously assigned to the other parts, only the first violin part is unique; thus, a value corresponding to the duration of a half note is added to the “uniqueness” of the first violin part. The concepts of “melodic uniqueness” and “harmonic uniqueness” are similar to that of “rhythmic uniqueness,” but they focus on the direction of pitch changes (upward, downward, or horizontally) and on the kind of steps used to construct the harmony, respectively.

As subjects of this research, 23 string quartets (chamber form) and 39 symphonies (orchestral form) are under investigation. The process of the analysis is as follows.

  1. Data from the score (the New Mozart Edition by Bärenreiter-Verlag) are imported into the computer and exported into MusicXML format.
  2. The features are measured using the program designed for this research.
  3. Cluster analysis is performed in order to confirm whether the features are suited for distinguishing between chamber form and orchestral form.

The progress of the research to date shows that the rhythmic uniqueness of the first violin part of early symphonies is clearly lower than that of string quartets.

music, Mozart, quantitative analysis, instrumentation, the number of string players


  1. Milliot, Sylvette: Le quatuor, Presses Universitaires de France (1986).