Pathos and Playfulness in Mozart’s Piano Concerto No. 23

Mozart’s twenty-third piano concerto, composed for a subscription concert in 1786, illustrates to stunning effect the composer’s characteristically mercurial sensibilities. Never did he dance more gracefully between the poles of elegant balance and impish amusement, on the one side, and Sturm und Drang dejection and psychological torment on the other. Indeed, affective contrast in central to this work; because the first and last movements are so indefatigably upbeat and buoyant, the wounded core of the concerto, the andante, bleeds all the more red. In this post, I will explore certain aspects of the musical language that Mozart employs in this slow movement to deliver its sublime chill. Moreover, since it’s problematic to address music as a text without referencing its sounding presence, I will briefly examine two recorded performances of the movement, one by Friedrich Gulda on a modern piano and one by Robert Levin on the fortepiano. (Recordings embedded below)

In the rare key of F# minor, the andante employs a range of affective techniques all gesturing towards a universe of sorrow. We begin with an 11-bar piano solo passage that, while starting out with a stable iteration of the tonic, quickly veers into the unexpected. Indeed, the passage is unusual for its wide leaps of strange intervals, a feature particularly evident in m. 2, when we plummet from a G#5 down to an E#2 before stabilizing at B4. There is a sense of profound disjuncture to this wildly vacillating movement between far-flung registers, an effect that is exacerbated by the rupture of the trochee rhythmic pattern established in the first measure. (We’ll return to this opening later.) We thus have a dual breakdown: our tightly-bounded melodic figure lurches unexpectedly into a diminished triad spelled out one note at a time across the range of the instrument, and our rhythmic periodicity is aborted to lend this dramatic gesture additional support. In m. 3, a level of normalcy is asserted once again, as the trochee returns to provide stability to the remainder of the opening. This technique – establishing a framework of normalcy, breaking the rules in a moment of expressive paroxysm, then returning to normal – is not unique to this opening solo. When the orchestra enters in m. 12, it isn’t long before we encounter another rip in the seam of our expectations.

To my ears, the tragic masterstroke of the exposition comes in measures 16-18 (and in the various recapitulations) with a harmonic gesture that exemplifies Mozart’s superlative command over the hermeneutics of despair. We begin (like the opening solo) with a pattern establishing a norm, in this case a grounded harmonic progression that rises from the tonic in m. 12 up the five degrees of the scale to rest on the V7 – cadence position – in m. 16. Unlike the opening (m. 4), where the dominant is sustained through a full measure, however, Mozart jumps the gun here: the tonic arrives on the weak part of the beat. This fleeting rhythmic dissonance is given further destabilizing weight as the ensemble enters subito forte on the tonic to drive home the surprise. Shocking as this disruptive anticipation might be, however, it’s only a harbinger for the wrenching harmonic contortions that follow. In m.17, the basses plunge to C natural, a tritone down from the tonic, and we move into a series of chromatically descending parallel diminished chords. This motion is essentially an unusual reharmonization of the first two bars of the exposition – indeed, the melody in the violins remains the same, although displaced an octave – but more than simply an unorthodox musical procedure, the discombobulating swerve is psychologically potent. Just after sinking through two consecutive diminished chords, when abjection appears complete, we are struck with one more searing diminished harmony on A#. This moment is yet another betrayal of expectations: in the first half of the phrase (m. 14), the melody is harmonized with the tonic in second inversion, but in the second iteration we’re faced with something altogether different. The collapse of expectations is total when the satisfaction of the anticipated tonic is yanked away at the last second for yet another bitter diminished chord, which appears as a tragic fait accompli. After hitting this extreme harmony (Taruskin might call it the FOP, or “far-out-point”), the composer quickly unwinds his position to cadence in time for another piano solo. The process of norm-subversion-norm is complete: unfurling into dark nether regions of the key in a twisted, unpredictable spiral, then pulling himself together for a textbook authentic cadence, Mozart shows us the depth of his control over affect.

Of course, every performance will communicate the pathos of this movement on different terms. I am accustomed to hearing the Mozart concertos played on modern pianos, with their heroic fortissimos, long sustain, and richness of sound. Recently, however, I’ve been drawn to the Christopher Hogwood period instrument recordings with Robert Levin on the fortepiano. At first, the fortepiano struck me as a clumsy substitute for the modern instrument; it decays quickly, lacks the dynamic range, and has a quirky, uneven tone throughout its register. To put it honestly, I despised these recording when I first heard them. The frailty of the instrument simply didn’t seem up to the task of the material (ironic, considering they were all written for the fortepiano) and many passages – especially those involving the lower register at a loud dynamic – sounded almost vaudevillian at times, with rattling, sharp bass notes that conjured images of darting-eyed silent film villains.

Andante (R. Levin – Fortepiano)

In movements such as this andante, however, the structural inferiorities of the fortepiano actually enhance the affective dimensions of the music. Indeed, the very frailty and unevenness of the instrument’s sound serve to highlight the despondent vulnerability of the musical narrative, contributing an additional layer of poignancy to an already taut representation of Weltschmerz. What initially struck me as the distracting imperfections of an inferior, antique instrument have become, through repeated listening, essential interpretative components of the piece itself. Friedrich Gulda’s recording of the work with Nikolaus Harnoncourt, while flawlessly executed, loses some of the movement’s quality of human weakness as a result of its very perfection. The fortepiano, with all of its inherent flaws (to the modern ear, that is), is perhaps better suited, in my opinion, to the subterranean realms explored here. Its sound is a beautiful ugliness.

Andante (F. Gulda – Piano)

To illustrate the difference between piano and fortepiano in this movement, we need look no further than the opening solo. As previously mentioned, part of the rhetorical power of this passage comes in its near-complete (but brief) breakdown of rhetoric in the service of naturalistic expression. When performed on the piano, the sound of each staggered pitch in m. 2 gently bleeds into the next, and the diminished triad materializes out of the mists. Even without the sustain pedal, pianos reverberate differently than their ancestor instruments. But on the fortepiano, this same gesture appears as a distorted, pointillistic helter-skelter of unpredictable sound-dots. The three consecutive notes of the diminished triad are properly contextualized when supported by the superior acoustic projection of the piano; the early instrument, weak in its tonal support and absent in sustain, on the other hand, present these three notes nakedly. They almost sound like mistakes, which in this context is wildly effective. The breakdown of this opening norm is, I think, more pronounced and heartbreaking when rendered on an instrument that equals the affective gesture in abject frailty.

Aside from the choice of instrument used in any recording of the work, one of the major performance practice questions for this and all of Mozart’s concertos (and similar music from that era) has to do with the issue of embellishment and improvisation. Again, on this count, the Gulda and Levin recordings differ profoundly. Gulda sticks quite close to the script, offering occasional embellishments but never venturing into outright improvisation. The repeated sections of the movement’s ternary form are interpreted with exactitude, amounting to the same thing with each repetition. His is a stark and stripped down but ultimately breathtaking interpretation, to be sure. Levin’s recording, on the other hand, takes a number of considerable risks, especially in this day and age when classical performers often shy away from improvisation (or rather, run screaming from it). After the opening, which Levin interprets literally, every other extended passage is embellished to such a florid extent that at times it comes close to pure improvisation. This approach is not without its naysayers, of course, and the pianist makes an explicit point to address these fears in his liner notes, indicating that he would never hope to top what Mozart has written. Rather, improvising so freely throughout is meant to lend a sense of spontaneity to the performance; like a great jazz musician, Levin delights in the radical temporality of the craft. As in any improvised performance, the music on this recording will never be performed the same way ever again.

Listening to the role of improvisation in these two recordings, I’m struck once again by the profound role performer choices make in the delineation of a (composed) musical idea. In different ways, both performances are “authentic”: Gulda follows Mozart’s indications, while Levin engages in a practice that was widespread during the composer’s time (Mozart himself greatly excelled at it, in fact). Both recordings, therefore, are beautiful and truthful in their own ways. However, like the imperfections of the fortepiano’s sound, I find that Levin’s improvisatory flights add profoundly to the dramatic – and indeed, contrastive – character of the music. Improvisation is an art of finitude; it is fleeting, unbounded, and ultimately uncontrollable (at least in the extent that written notes represent “controlled” sounds). The once-in-an-eternity element of Levin’s figures thus signals the transitoriness of the unfolding music, its susceptibility to decay and loss. This interpretation is fitting to the tragic nature of the movement. However, improvisation is also a celebratory embrace of the present moment. As a practice, it stands smiling astride passing time, reveling in each dying second. In this sense, Levin gestures toward the redemptive edification latent in the tragic movement. Thus, through improvisation, Levin boosts the level of both pathos and playfulness.


  1. I also really enjoy Levin’s Mozart concertos, and for the same reasons that you enjoy them. Only one (small) problem: the improvisation, which is so enjoyable on first listening, is frozen by recording, and therefore loses its immediacy on each rehearing. I don’t want to ‘get to know’ the piece as Levin plays it, so I try not to listen to my Levin CDs too much, notwithstanding how brilliant they are.

    1. This is a great point — recording technology has turned improvisation, once purely temporally bound, into a different sort of musical text. There’s really no different between Levin’s recorded improvisations and Mozart’s actual score in fixity: both are frozen, as it were. Limiting how often you listen to them is a good idea; I wonder, though, whether they might remain fresh even if you listen to them often. I listen to my Coltrane records all the time, and I’m continually surprised by his improvisations, even though they are stuck in time on the recording.

  2. This is a very insightful posting. I was particularly encouraged by your growing liking of the fortepiano. I actually first heard these recordings by Malcolm Bilson and the English Concert and have always found the reverse -that contemporary recordings sound a bit sterile in their perfection. I have always loved the SATB qualities of the fortepiano. This is especially true in Beethoven, where when he really pushed what the instrument can do it really comes thru, whereas a modern grand can handle just about anything without breathing hard – its just not the same. It all just takes bit of getting used too…

  3. Thanks, Zach, for the links to the Gulda and Levin performances, which indeed provide a vivid demonstration of the difference that performance can make in bringing out the varying moods of that remarkable Mozart movement.
    I thought I’d say a word about Gulda. He was a very imaginative musician who tried his hand at many things. He even played and recorded a fair amount of jazz, and composed jazz numbers himself.
    A recording he released in ca. 1981, on the Pearl label, of two Mozart concertos (21 in C/K467, 27 in Bb/K595) is the opposite of literal. The jacket claims: “This record offers a tantalizing glimpse of how Mozart may have played his piano concertos…” and, on the front side: “Authentic Mozartean interpretations with extempore embellishments.”
    Much of what Gulda does here seems to me more extreme than what Levin does. Two examples from K467: 1) Gulda plays during the closing ritornello, not in a continuo fashion but melodically, thus drawing the listener’s ear to _him_, instead of blending back in with the orchestral throng. 2) During the first solo of the middle movement (the movement made famous by its use in the Swedish film _Elvira Madigan_), Gulda lets his right hand drag substantially behind the beat–I mean a _lot_, far more than is considered acceptable in Chopin performance, for example. I suspect he was inspired here by performance traditions in jazz.
    Is this an example of Historically Informed Performance? Willful reimagining (a bit like some of the stagings we get nowadays of standard-repertoire operas)? The creative interaction of past and present is something that I normally like to encourage, because it keeps concert (and operatic) life alive. But I always find I want to draw the line somewhere and say: No, this isn’t Mozart anymore (or Verdi or whoever). Still, I also find that an approach that I might consider tasteless or heavy-handed _in theory_ persuades me in practice.
    Anyway, just wanted to put in a word for another, very different approach to embellishing/altering Mozart than Levin’s. And to note that it came from the same pianist as your example of a relatively literal performance!

Leave a Comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s