Tonal implications and desire for resolution in Scriabin’s late works

The two piano pieces op. 57 by Alexander Scriabin (composed 1908) stand on the cusp between common-practice tonality and the post tonality of his late works.

In spite of several almost “Classical” cadences – especially in the second piece “Caresse dansée” – much of the music barely sounds tonal. Nevertheless, the harmonic language is driven by a desire for resolution that had been developed for over two centuries. This blog post discusses the beginning of the first piece “Désir” (desire) and tries to demonstrate how the feeling of harmonic progression can be analysed using the two opening chords of the piece. I should point out, however, that these tendencies for resolution are always ambiguous: the opening chord can be resolved in several directions, but not in every thinkable one. The voice-leading partly supports these resolutions, but needs to be understood in conjunction with the progression of the root of each respective chord.

Example 1 shows the beginning of “Désir” (LISTEN HERE), with a reduction below. A, B, C and D represent harmonic interpretations of the same chords. If the first bass not C is understood as the root (Example 1A), then E is the 3rd, F sharp the lowered 5th (G flat) and B flat the (dominant) 7th. Equally though, the F sharp could be the root (Example 1B), A sharp the 3rd, C the lowered 5th and E the 7th. Both chords are built identically, only the distribution of notes is different.

Example 1: Implied resolutions at the beginning of Scriabin’s “Désir” op. 57 no. 1

As Kenneth Smith highlights, most of Scriabin’s chords at that time can be understood as 7th-chords, more specifically dominant 7ths (Smith 2010). For this reason, the most likely resolution is always that of a falling 5th between the respective roots (“authentic” resolution), even if the chord that follows rarely functions as a (consonant) tonic. This “authentic” resolution occurs in Example 1A from chord 2 to chord 3 only, while in 1B the upwards step between the roots of chords 1 and 2 can be understood as a deceptive/interrupted motion.

There are, however, two alternatives (Examples 1C and 1D). These may initially seem odd, as the respective roots D and A flat don’t even sound in the chord. This is not an unknown occurrence though: in modern harmonic theory, the diminished 7th chord on the leading note of a major or minor scale is, for example, usually understood to function as a dominant (without the root). This interpretation originates from Jean-Philippe Rameau’s concept of a basse fondamentale, which began to be accepted more widely in the late 18th century, but it can be applied here as well. In Example 1C, F sharp would be the 3rd, A sharp the raised (!) 5th, C the (dominant) 7th and E the major 9th. Here, too, the A flat can be read as the root – in a kind of “tritone substitution” – without the structure of the chord needing to be altered.

Alternative 1C has the advantage, that we’d have a falling 5th from chord 1 to chord 2, while 1D has a deceptive/interrupted resolution from chord 2 to chord 3. In principle, of course, one could combine the various harmonic interpretations: chord 1 to 2 as falling 5th (1C) or deceptive/interrupted motion (1B), chord 2 to 3 as falling 5th (1A) or deceptive/interrupted motion (1D). Of course it’s not about cherry-picking; instead, all of these possibilities are implied in the beginning, at least in principle.

But what does the voice-leading suggest? In general, Scriabin seems to have ascribed some importance to chromatic voice-leading and a relatively “Classical” resolution of chordal dissonances. Nonetheless, we’re left with a kind of attempt at squaring the circle: If we favour the falling 5th in the beginning (Example 1C), the raised 5th of the chord resolves “correctly” to the 3rd of the following chord. The B natural can even be seen as part of the chord here, instead of the A sharp, namely as a so-called “Chopin chord” with a 6th replacing the 5th (D-F#-B-C). This interpretation is taken up again in the final cadence of the second piece (Example 2; LISTEN HERE). However, the 7th C (in the bass) leaps downwards, rather than resolving by step. Similar problems are found not just in Example 1C, but in all of the variant readings.

Example 2: “Chopin chord” in the final cadence of Scriabin’s “Caresse dansée” op. 57 no. 2

In my next post, I will be discussing the beginning of the second piece. This starts with the same chord as the first, but proceeds differently. In the meantime, I’d be very interested in knowing if you find this analysis convincing and which harmonic progressions you hear. Or are the harmonic implications so veiled that one only hears sonorities which are somehow (randomly?) connected?


Kenneth Smith: “‘A Science of Tonal Love’? Drive and Desire in Twentieth-Century Harmony: the Erotics of Skryabin”, Music Analysis, 29 (2010), 234– 63.


Let’s stay in touch

If you find this blog interesting and would like to be notified when a new article has been published, please enter your name and e-mail address here:
* We don't share your data. See our Privacy Policy
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Dr Stephan Schönlau next to piano smilingDr Stephan Schönlau next to piano smiling

Send me a message to arrange a free consultation.