Sing-Speak ? Rubato and Rhetoric

When we think of “music” we think usually of singing melodies. But music does not only sing. It also very often “talks”, and also even shouts, whispers, screams, sighs, pleads, laughs, cries and generally mimics all types of human (and animal) communication including body-language. We can call these declamatory, dramatic (or spoken/conversational) aspects of music, rhetoric. An effective, powerful rhetorical delivery comes partly from the choices of  Articulations and Phrasing/Dynamics but it is also – and perhaps principally – a question of timing (rubato and rhythmic micro-freedom in general). Perhaps we could consider “rubato” as the rhythmic freedom we use in a singing phrase, and “rhetoric” as the rhythmic freedom we use in a more declamatory (or theatrical) style of phrase/music.

Playwrights and screenwriters do not indicate many expressive devices in their written dialogues. There are no indications of tempi, dynamics, phrasing, articulation etc. This gives actors enormous freedom to use their own imagination to deliver the text (speak their lines) in the way they consider most effective. In classical music, on the other hand, composers normally indicate as much as they possibly can. These indications can help us, but should not discourage us from adding more of our own expressive (rhetorical) details. Musical notation is necessary to transmit a composers ideas, but the fact that rhythmic notation is so precise shouldn’t mean that we are not allowed to play with our own commas, pauses and small variations in tempo, as well as being free to add our own additional phrasing, articulations and dynamics that are not specifically indicated in the written score. This is “performance rhetoric”: the art of how to present a certain defined content in a more effective, communicative way. In other words, the art of how to make what we are saying (playing), more meaningful and more powerful.

The opposite of rhetoric is robotic. Imagine an actor reading a text metronomically, without taking any of these subtle, dramatic “rhetorical” liberties. It would be like a child, reading aloud, but not really understanding what they are reading. Very often we musicians are guilty of this type of rhythmically-robotic delivery. This is understandable: not only are we trained to play “exactly what’s on the page” but also, playing without rubato or rhetoric makes fitting together the different instruments/voices of an ensemble definitely a much easier task.

Operatic music is usually played much more freely than instrumental music: the acting and the words certainly encourage this dramatic delivery. Unfortunately however, in instrumental music performance, we are often so frightened at the thought of offending critics, purists, teachers and academics, that we don’t dare to add these little personal, dramatic touches to the score, which becomes like a prison (see Freedom or Obedience?). This is a shame.

Putting words to our musical phrases can often help us to feel freer and to give more meaning to our music, as though we were singing a song or acting a part. Sometimes the “right” words for a phrase might be deep and meaningful. Other word choices might just have the effect of loosening us up with some comic relief:

Bach‘s Solo Suites (and Solo Violin Sonatas and Partitas transcribed for cello) are an excellent starting point for an exploration of (serious) rhetoric in music. They have very few singing melodies, no expressive indications and as we are unaccompanied, we are thus free to not just “sing” them but also to  “talk, recite and declaim” them, exactly as we wish. In the Cello Suites this is especially relevant to the Preludes, but also (to a slightly lesser extent) in the slower, freer dances (Allemandes and Sarabandes). But even many of the faster movements need a lot of these “rhythmic micro freedoms” (in addition to loads of phrasing, dynamics and articulations) in order to avoid them sounding like minimalist sewing machine music. In Bach’s B minor Partita for Solo Violin for example (here transcribed for cello), five of the eight movements run a serious risk of sounding like the dreaded sewing machine because of the uninterrupted flow of equal note values, as shown in the following table:





Double I


378 (semiquavers/16th notes)



474 (quavers/8th notes)

Double II


953 (semiquavers)

Double III


298 (quavers)

Double IV


531 (quavers)


The unwillingness to take (or fear of) rhythmic liberties is probably what leads so many people to play this music very fast. At least that way, the effect of the high-speed virtuosity makes up somewhat for the lack of the performer’s micro-rhythmic courage (or interest).

One good way to develop rhythmic micro-freedom is to play with the metronome beats very far apart. This has nothing to do with the speed of the music, but just means that instead of having the metronome beating each crotchet (quarter-note), we might have it beating each minim, or each bar, or every two bars etc. No matter how fast or slow the music is, we can always choose a slow pulse speed for the metronome. This allows us to take rhetorical micro-liberties with the rhythm while still maintaining a steady basic pulse for the music. This is a very useful skill (see the article by Gary Carr from The Strad issue xxx), but unfortunately most metronomes don’t go slowly enough! The mania for mathematical rhythmic accuracy is such that any metronome will subdivide up to a very high number of beats per minute but won’t go slower than 40/minute which is absolutely the opposite of what we need.

Playing from memory is another good way to liberate ourselves from the “tyranny of the written notation” as we will soon forget what was written, and start playing with spontaneity, personality …………….. and rhetoric (dramatic musicality).