Name: Lyapunov: Transcendental Etude Op.11 No.10, Lezghinka @AshishXiangyiKumar
Uploaded: 2026-09-23T08:17:00-05:00
Duration: 8 min 17 s
Description: Lyapunov: Transcendental Etude Op.11 No.10, Lezghinka

Lyapunov: Transcendental Etude Op.11 No.10, Lezghinka

Ashish Xiangyi Kumar 2026-09-23 | Lyapunov: Transcendental Etude Op.11 No.10, Lezghinka

Balakirev: Islamey (Jando, Pogorelich, Kantorow, Gavrilov, Berezovsky)

Ashish Xiangyi Kumar 2023-09-21 | Islamey is one of those works (like Schumann’s Toccata, or Liszt’s better paraphrases) whose very substantial musical merits tend to be obscured by the virtuosity of its writing. It’s not uncommon to hear people complaining that it’s loud, messy, repetitive, or some unpleasant mixture of these qualities. And there’s a certain logic to this: Islamey employs a very unusual series of textures for a virtuoso work, textures that, measured by Romantic standards, have a certain crudity to them (rhythmically unvaried thematic material; prolonged passages in bare thirds; bare octaves used for both RH melodies and LH accompaniment; stark toccata-like passages; ugly pedal points (m.77), unusual harmonic movement – is there a single V-I in the whole work?).

But Islamey’s weirdness is also its virtue – it tends to abhor middle-of-the-road interpretations, so that extremity in performance works very well: extremity of volume (very loud or soft), extremity of tempo (very fast or slow), extremity of clarity (or abandon). You can’t try to be too conventionally musical with this material – you sort of just have to go for it, and if you’re lucky it turns out musical, in its earthy, ungainly, ingenious way. There’s a surprising amount of discipline in how the work is put together, in fact – it’s essentially a series of variations on three folk themes, organised into a clear ABA’-Coda structure. The way Balakirev continuously develops material that doesn’t sound promising at first blush is very clever, and some of the thematic transformations are so smoothly pulled off you might not even realise you’re listening to the same material – see for instance the transformation of Theme 2 (0:26) into the smoky haze of 1:29, or the reduction of Theme 1 into a five-note motif (5:05).

Here are five performances that really nail it:

00:00 – Jando. Played in a completely unshowy way, but still totally effective owing to its clarity and sense of control. Proof that anti-virtuoso approaches to virtuoso works can produce great results. Jando never gets enough love.
09:38 – Pogorelich. Slow, like Jando (at least at first), but the sheer mass of sound produced gives the impression of something gnarled, inhuman, volcanic. Melodic intensity, hard, metallic accents. Sometimes all sense of harmony is lost and you’re just left with a torrent of transcendent, terrifying noise (15:41). [Shoutout to @Gazda Mitke II, whose recording of Pogo’s 1990 concert at Carnegie Hall produced this account for the ages.]
18:12 – Kantorow. Conceptually almost the opposite of Pogorelich. Lithe, playful, with superhuman control of colour, voicing, and phrasing even at extremely fast speeds. Of all the performances, this one also best captures the dancelike character of the first two themes.
26:28 – Gavrilov. Occupies a nice middle ground between Pogorelich and Kantorow. What really stands out is his ability to project large-scale structure in a work of such sustained fantasy – the entire last section starting from 30:51 is experienced like a constant acceleration to the finish, with the momentum never letting up.
34:23 – Berezovsky. Of all these performances, Berezovsky makes the most out of the contrasts in the work. It’s surprising how much of the performance is spent in piano-pianissimo—and how delicate and intimate Berezovsky makes these sections sound. In the final pages he also really cuts loose in a really nice way – you can sense he’s riding right on the edge of what’s technically possible for him, and it’s tremendously exciting (41:32).

Beethoven: Choral Fantasy, Op.80 (Barnatan, Shelley)

Ashish Xiangyi Kumar 2023-09-02 | Is this work kitsch as hell? Well, yes. Do its seams show? Well, sometimes. Is it one of the most joyous, life-affirming, indispensable things Beethoven ever wrote? Also yes.

There possibly is no other work (Bolero, perhaps?) that illustrates how silly it can be to think of music as ink on a page, or even ideas in the head, because so much of what is in here that looks ordinary (even crude) is aurally extraordinary. For a work that began life as a jury-rigged concert-closing crowd-pleaser, and whose individual parts (outside the piano opening) don’t especially impress, the Op.80’s got a *lot* going for it.

For a start: structure. For one reason or another, Beethoven always got this right. At first glance, this work is absurd – a 3-minute introduction followed by a 15-minute “Finale”. But the “Finale” is in fact very neatly structured: a theme and variations packaged into sonata form, with the recapitulation featuring added choir and solo voices (a stunt B. repeated in his 9th Symphony).

The work also features consistently clever thematic manipulation, and its most poignant passages (Var.7, the transition into the Finale’s recap) all feature transformations of the main theme so deft and organic they easily pass unnoticed. Added on to this of course is the sheer variety of Beethoven’s writing: there’s the disarming intimacy of the opening variations, which gently guide you through small constellations of instruments, exploring their colours; there follow concertante passages, lyrical explorations in B and A major, a march, and eventually, the ecstatic exuberance of the final variations (with canons in the mix, because why not?).

In the end, though, the loveliest quality of this work is the total, anti-ironic earnestness of Beethoven’s writing. There’s lots of places here that could be mentioned, but my favourite is the climax of this work (19:03) in which Beethoven highlights the word “Kraft” (“strength”) with a unexpected chromatic mediant shift to Eb. It’s a great moment, and Beethoven evidently likes it so much that, instead of ending the work right after, he quite deliberately circles back and repeats a good bunch of preceding material so that he can present the moment to the audience again. (More evidence: when Beethoven wrote to Breitkopf & Härtel to suggest obtaining a new text for the libretto, he asked that the word “Kraft” or something similar had to be kept in its place.)

The two performances here have accompanied me on the way to work for quite a few years now – few works lift my mood this well. Barnatan (accompanied by the ASMF) has a unassuming, natural style, so that it only gradually creeps up on you how excellent this recording is. The opening, in fairness, is hair-raisingly dramatic – Barnatan summons a huge sound, and his rubato is cunningly judged (that little pause on the first RH chord, and the drastic slowdown on the octave descent in each bar). There are also nice details in the introduction, such as the clearly projected RH inner line at 3:07. But if I were forced to generalise, I’d say that the focus here is on generating an unaffected, direct line to the music. There are some very nice results: the A major variation is extraordinarily tender, and in the finale at 18:08 the sudden shift in metric gear is very audible.

Shelley (accompanied by the Orchestra of Opera North, which really should be much more well-known) is vibrant, alert, improvisatory. I’ve not heard a recording of the Op.80 that exceeds this one for sheer happiness. Right from the opening the contrast with Barnatan is clear; Shelley accelerates on the rising semiquavers, and entire Adagio is faster, laden with tension. The sense of improvisation persists even after the entrance of the orchestra – in Var.1, for instance, the tempo shifts to emphasise the contour of the semiquaver line. Shelley also tends to highlight interesting detail, such as the piano’s staccato/slur-staccato articulation in Var.8, or the punchy martial accents in Var.9. In the final variations, the voices are more extroverted and present than in Barnatan’s account, and feature liberal use of vibrato – this makes for the most fun Var.11 I’ve heard, and the entry of the entire choir in Var.12 is breathtaking.

Barnatan:
00:00 – Adagio
03:49 – Finale, Part 1 (Vars. 1-5)
08:53 – Finale, Part 2 (Vars. 6-9)
15:19 – Finale, Part 3 (Vars. 10-12)

Shelley:
19:43 – Adagio
23:00 – Finale, Part 1 (Vars. 1-5)
27:45 – Finale, Part 2 (Vars. 6-9)
33:49 – Finale, Part 3 (Vars. 10-12)

Liszt: Grandes études de Paganini, S.141 (Guarrera, Raekallio)

Ashish Xiangyi Kumar 2023-07-15 | Sometimes, you come across a recording of a well-known work by a relatively unknown pianist that turns out surprisingly good. If you’re lucky, the recording might go down an untrodden interpretive path that makes you think you’ve maybe underappreciated the work a bit. And if you’re insanely lucky, the recording just flat-out turns out to be the best recording of the work you’ve heard.

So it is with Giuseppe Guarrera’s breathtaking (live!) account. Nothing in the set is played even close to literally, everything sounds so playful and improvised – and yet there’s not the slightest hint of wilfulness or perversity. The tempo variations in each work are huge, the phrasing intimate and hyper-flexible, the articulation/dynamic control superhuman, and the pedal used masterfully for colour. So many features of these works that you tend to overlook – a straightforward scale, a dominant-tonic resolution – become extraordinarily expressive in Guarrera’s hands. And a warhorse encore piece like La Campanella actually sounds like the masterpiece of texture and variation that it is. (If you’re like me, you’ll stupid-grin through a lot of this.)

I’d praise Raekaillo’s recording almost as much as Guarrera’s but for that fact that he’s actually pretty well-known (in particular, for an astounding set of Prokofiev recordings). It’s a recording that tacks in the opposite direction from Guarrera’s – while Guarrera plays so that you never quite know what’s coming next, there’s a sense of irresistible propulsion and inevitability to Raekallio’s account. It’s not that there’s no detail (there are many nice things going on – separated voices in the middle section of No.2 [32:35], desynced RH chords in the B section of No.5 [42:34], the ripped-out chords that open No.6 [44:20], the archly clipped phrases at Var.7 [46:29]), but all of it is subject to an overriding priority to maintain the underlying pulse. Phrases sound longer, hypermeter is clarified, climaxes relentlessly built up to. I have a special fondness for “antipretty” recordings such as this one, which essentially go – here’s the music, take it or leave it. (We used to have more of these pianists around – Kocsis, Weissenberg, Kovacevich – but they’re pretty scarce these days, unforch.)

Guarrera

00:00 – Etude 1 in G minor (Caprice 6, intro/coda from Caprice 5)
05:04 – Etude 2 in Eb major (Caprice 19)
10:29 – Etude 3 in G# minor, “La Campanella” (Violin Concerto Nos.2+1)
15:33 – Etude 4 in E major (after Caprice No.1)
17:46 – Etude 5 in E major, “La Chasse” (after Caprice No.9)
20:33 – Etude 6 in A minor (after Caprice No.24)

Raekallio

26:17 – Etude 1 in G minor (Caprice 6, intro/coda from Caprice 5)
30:53 – Etude 2 in Eb major (Caprice 19)
35:43 – Etude 3 in G# minor, “La Campanella” (Violin Concerto Nos.2+1)
40:02 – Etude 4 in E major (after Caprice No.1)
41:57 – Etude 5 in E major, “La Chasse” (after Caprice No.9)
44:20 – Etude 6 in A minor (after Caprice No.24)

Beethoven: Sonata No.2 in A Major, Op.2 No.2 (Korstick, Levit, Pletnev)

Ashish Xiangyi Kumar 2023-06-10 | The general impression of the Op.2 No.2 is that it is a lithe, graceful thing. This belies, however, its radical (even parodic) tendencies.

For a start, the sonata is fiendishly difficult. Stock classical-era devices (arpeggios, scales) are pushed here to silly extremes (and lovely effect). The most infamous example is the impossibly-fingered RH descents at mm.84 and 89, but you also have the treacherously exposed triplet semiquaver runs in the first movement and the ever-expanding - and increasingly ridiculous - arpeggio that opens the last movement.

The sonata also modulates in unexpected and extremely modern-sounding ways: see the second theme group at m.58, where it moves in intervals of a minor third (an interpolated omnibus sequence!), the use of C and (extended) F major in the development of the 1st movement, Bb in the 2nd, G# min in the 3rd, and F + Bb major at the end of the 4th. It’s pretty striking that something that sounds so graceful modulates with such aggressiveness.

There’s also the textures B. deploys here, which were more or less unheard of in his time. In the opening movement, there’s a spectacular triple stretto of the rising scale motif elaborating a circle of 5ths (m.182); in the second, gorgeous Brahmsian textures (m.17) and pizzicato plucking in the lowest registers against sustained middle tones; and in the last, an opening featuring an absurd operatic downward leap, alongside deliberately overinvolved mimicry of the first movement’s rising A arpeggio motif.

MVT I, Allegro vivace
EXPOSITION
00:00 [m.1] – Theme Group 1, Theme 1. Note Motif (A), the opening descending 4th, the related long octave descent starting in m.4 (A*), and Motif (B), the quick downward run in m.1.
00:07 [m.9] – Theme Group 1, Theme 2. A rising scale (C), followed by a lovely three-part chamber texture. Has a ritornello function.
00:26 [m.32] – Theme Group 1, Theme 3. Containing a much faster version of (C), and slower descent (D).
00:34 [m.42] – Dominant preparation, leading deceptively to
00:48 [m.58] – E minor. Theme Group 2, Theme 1. An incredibly modern-sounding series of enharmonic modulations, each raising the tonality by a minor 3rd.
01:03 [m.76] – (B) returns. (Also (A)’s rhythm interspersed). At m.84, cadence-theme 1.
01:18 [m.92] – (C)+(D) return (ritornello!)
01:28 [m.104] – Cadence-theme 2
DEVELOPMENT
03:23 [m.122] – After closing chords in E min, TG1, T1 in C
03:30 [m.131] – In Ab, (A)+(B)+(A*), moving into dominant of Fm
03:50 [m.162] – TG1, T2 in F. Repeats itself, led by LH. Moves into Dm.
04:13 [m.182] – Motif (C) developed in close 3-part imitation moving around the circle of fifths, borrowing the arpeggio that opens TG1, T2. At m.189 (4:19) (A) appears, together with fragment of (C). Process repeated twice.
04:30 [m.203] – Dominant preparation, extensively using (C) and TG1, T2’s opening arpeggio (in LH).
RECAPITULATION
04:50 [m.225] – TG1. The counter-statement of TG1, T1 is absent. Instead, T2’s closing bars are repeatet in D (m.245) and A.
05:34 [m.278] – TG2 in Am, leading straight into closing cadence without coda.

MVT II, Largo appassionato
09:38 [m.1] – Theme, in D. The 20 seconds after 11:14 (m.16) contain some of the most moving music B. ever wrote.
11:37 [m.19] – Episode 1, in Bm.
12:59 [m.32] – Theme
14:53 [m.50] – Episode 2, beginning like a coda (alternatively, the start of a genuine and massive coda). At m.58 theme returns suddenly in Dm in ff, moving into Bb and home dominant
16:39 [m.68] – Theme, in higher register, with delicately articulated accompaniment
17:11 [m.75] – Coda

MVT III, Scherzo
18:08 [m.1] – Scherzo. At m.19, second melody in G#m(!)
19:33 [m.45] – Trio, in A min
20:18 – Scherzo

MVT IV, Rondo
21:04 [m.1] – Theme (Note the opening arpeggio extends the arpeggio figure from TG1 T2)
21:35 [m.16] – Transition, moving to E
21:53 [m.26] – Episode 1
22:22 [m.41] – Theme
22:54 [m.57] – Episode 2, in Am. A shocking, even ugly contrast.
24:40 [m.100] – Theme
25:12 [m.115] – Transition, staying in A
25:29 [m.124] – Episode 1, in D
25:51 [m.135] – Theme/Coda (can also be read as Theme + Episode 2 + Theme, giving the movement an ABACABACA structure). At m.140 – sudden diversion into F, where the theme stays for 5 bars. At m.148 (16:17) the theme is developed into the bass into tonic-dominant swing. At m.154 (26:28) the portamento from the theme is deployed in what looks like a final cadence that suddenly moves using an enharmonic switch (D#=Eb) into Bb.
26:41 [m.161] – Episode 2’s theme recurs in Bb before the theme returns one last time, richly ornamented and diverted at m.180 into the final tonic-dominant swing.

Liszt: Bénédiction de Dieu dans la Solitude (Hough, Korstick)

Ashish Xiangyi Kumar 2023-05-27 | Liszt’s Benediction of God in Solitude is a marvel: a tender, luminous testament to Liszt's underappreciated lyric abilities. For flat-out beauty few things match it. It’s also a masterclass in textural writing – just count at all the ways the main theme is decorated: An inner-voice arpeggio underneath a measured trill (right at the beginning); a semi-measured arpeggiated chord with emphasis on the lowest voice (m.86); a bigger semi-measured arpeggiation (m.105); crystalline 2-voice chains (m.156 – and from m.158, the shape actually borrows from the dominant-tonic-supertonic-mediant pattern of the melody’s first 4 notes); a straighforward LH rising/falling arpeggio (m.253) that gradually expands (m.272); and long strings of rippling RH wavelets (m.308). It’s entirely possible to hear these textures as lushly spiritual, but in all honesty I hear waves – which places this work in my head alongside Liszt’s other water-related masterworks (St Francis, Au bord d'une source, The Fountains of the Villa d’Este).

Structurally the Benediction is a kindo-of-rondo, with the main theme (refrain) getting a fair bit of development. It also contains an inspired moment (which Liszt was to replicate in the B minor Sonata) when a completely new theme (Andante semplice) enters in the midst of extensive and recapitulatory coda. The tonal layout of the work is very precise – each section’s tonality drops by a major 3rd, while the opening section in F# anticipates much of what is to come by prominently featuring D and Bb. And in the coda of the work, the 2nd and 3rd themes are stated in F#, thus resolving the tonal tension between the sections much as the recap in a sonata would.

00:00 – A section, in F#. Theme 1. A long, lyrical melody in two parts. Prominently features the key of D in its second half, then modulating into A# minor before entering darker, almost improvisatory territory (m.43).
03:16 – A’ section. Theme 1 developed, and gets a new, exultant tail (4:47) in Bb. Modulates back into F#, and we get a codetta heavily borrowing from the shape of m.9 in the LH, and from the melody’s first four notes in the RH.
06:40 – B section, in D (a key anticipated in the earlier section). Theme 2. A simple melody based around chains of falling thirds, harmonised fauxbourdon-style (note the preponderance on fourths in the RH). The only time the piece ventures into tragic territory comes at m.211 when we move into B min – before being gently steered back into D.
09:38 – C section, in Bb. Theme 3. The apparently new melody here subtly references the A section at m.47-49 (similarly melodic shape). Theme 3 is diminuted, before slipping into Db (m. 234, 10:26), and modulating through several keys to reach G min. Theme 3’s first phrase in then canonically treated starting from m.244, rising in tension to lead back into the
11:39 – A’’ section. Remarkable how a return to a simple treatment of Theme 1 – simple arpeggiated chords over simple arpeggios – suddenly sounds so moving. Builds slowly and inexorably into a massive climax.
13:27 – The codetta returns, this time with the RH spinning out a single delicate line of arpeggios. The shift into 3/2 time is nearly unnoticeable, but gives the melody a lot more breath.
[Coda]
14:37 – C’ section, in F#. Now transformed into something almost like a chorale. Moves into E (as it did earlier) and pauses in a repeated B, which gives way to
15:45 – Theme 4, in F#. A magical moment. Hard to say what exactly gives this melody its consoling, tender quality. The newness is part of it, certainly. But also the fact that this melody contrasts Theme 3 in its shape – descending stepwise to the tonic, rather than ascending – and the simplicity of the accompaniment.
16:26 – Theme 2, in F#. An unexpected recollection.
16:58 – Theme 3 is stated in B (implied by the E natural) and D Lydian (same harmonic progression as in the codetta!), before we get the final cadential 6/4.

Schumann: Humoreske, Op.20 (Fejérvári, Endres)

Ashish Xiangyi Kumar 2023-05-01 | The Humoreske is one of the most extraordinary and underappreciated works in the Romantic literature. Despite its diminutive title, it really is a large single-movement work, almost 30 minutes in length, that binds together a staggering variety of contrasting materials within an intricate and innovative framework. Broadly speaking, the Op.20 has a sort of telescoped structure: you can break it into 5 broad movements, further subdivide it into 7, and then into 29 distinct sections each with their own distinct tempo directions (some of which are repeats). Completely absent are your usual Schumanisms – the descriptive titles, the Eusebius/Florestan stuff, the place-setting. Instead only austere performance directions loosely subdivide the work, although the music itself contains some of the most stunning (/stunningly quirky) ideas Schumann ever generated. The title “Humoreske” is hence better read as “a study in the humours” (in the sense of emotions), rather than the more conventional denotation of a short, fun bit of music.

1. Einfach (Simple), in Bb. ABCBA. The first section of this movement features one of Schumann’s loveliest melodies, beginning far away from the tonic on an F#, as if we’ve stumbled across the music mid-stream. But this F# foreshadows the melody’s development, as is modulates to Gb major (from which LH enters into a delicate canon with the RH). The Sehr rasch (Very fast) section which follows contains a lively melody whose first bar regularly leaps back into the fray (in forte) in order to provide some funny interjections. And the third section (Noch rascher, even faster) features some glorious fanfares, one of which features a tonic-dominant swing underpinned by a daringly sustained pedal.

2. Hastig (Hasty), in G minor. ABA. The star of the set. The middle stave contains an unplayed “inner voice” indicating a line implied by the RH semiquavers. One of the rare instances where a composer seems to acknowledge that a part of the score is really there to be heard in the mind, not in the sound. The middle section of the work features the spiciest texture of the entire work, in which the LH is instructed to play in tempo, while the RH must play “as though out of tempo”. The effect is one of the most startling, aurally confusing syncopations ever written for the piano (6:10; 33:04). And the surprises don’t stop there: the following passages feature ecstatic swirls of notes over static harmony, canonic marches, a dramatically dissonant diminished seventh descent over a lower *and* middle-voice D pedal, and a touching reprise of the opening material in which it arrives transfixed as a series of static long-held chords (7:51), before the original texture returns exactly where the chords left off, at the melody’s ninth bar. Most composers which spread this much invention over a couple of large-scale works: Schumann packs it all into under 5 minutes of music.

3. Einfach un zart (Simple and delicate), in G minor. ABA. The outer sections feature a plaintive melody over a dolefully rocking accompaniment. The middle section’s Intermezzo is a wondrous thing: relentlessly descending semiquavers punctuated by a upper-register Bb bell. The counterpoint of this section grows in complexity, eventually reaching a point where the semiquavers have become treacherous octaves, and the LH features a fully distinct countermelody. Eventually the Bb bell slowly fades, as if the semiquaver torrent is bearing us away from wherever it’s coming from.

4. Inning (Heartfelt), in Bb. ABA. A return to the mood of the opening, but with denser counterpoint. (The trick here is to make sure that the octaves *don’t* sound like octaves, but two melodies moving in parallel.) This movement is often grouped together with the next two to form a larger movement.

5. Sehr lebhaft (Very lively), in G minor. Running semiquavers out of which emerges a cunningly dotted octave-displaced melody (17:04; 44:10). Ends with a rather insectoid stretta that eventually grows into brilliance. The most energetic and possibly the funniest movement of the work.

6. Mit einigem Pomp (With some pomp), in C minor. The only movement not in Bb or its relative minor, a 4/4 polonaise that is equal parts grand, menacing, and ironic. The middle section features a canon between the LH and RH – but in a sharp contrast to the canon in the first moment, this one is bumbling and obtrusive, as if overeager to justify its place.

7. Zum Beschluss (To the resolution), in Bb. The only movement that Schumann marks out as such, with its own distinct title. A recitative is gradually built up into a densely chromatic, contrapuntal structure. After a gentle closing phrase is repeated thrice, each time in a higher register, the coda suddenly bursts in with a finale that could not be more different than the preceding material – it is both gloriously triumphant and also a little silly (think boozy trombones sloshing around the LH).

Liszt: Années de pèlerinage, Year 2 - Italy, S.161 (Piemontesi, Gorus)

Ashish Xiangyi Kumar 2022-05-22 | Compared to Year 1, the second book of the Années de pèlerinage is less focused on textural intricacy, more lyrically concentrated – the three Petrarch sonnets, the centerpiece of the set, all contain stunning melodies which integrate harmonic colour in really fun ways. Book II’s works are also considerably more expansive: only one of them can be called a miniature, and this book’s counterpart to Year 1’s Vallee de Obermann, the Dante Sonata, is also correspondingly more ambitious and sweeping in scope (it gets its separate video, so don’t worry that it isn’t here!). Even the opening work, Sposalizio, features a good deal of thematic transformation/combination, deployed to generate an air of (sometimes intimate, sometimes joyous) reverence.

Piemontesi is warm, melody-focused (especially in the sonnets), and exceptionally well-controlled. Gorus is neurotic in the best Horowitzian sense – so much unexpected detail leaps to the fore, so much playful & expressive extremity. A good example of the different approaches is the B section of Sposalizio (4:12; 40:00) – Piemontesi produces a lovely RH dolce and keeps the LH fairly measured, while Gorus gives the LH a fair bit of rubato (generally speeding up slowing down within the space of a bar). Gorus also lets the LH at the climax at Sonnet 104 (1:01:27) dramatically surge and retreat, while Piemontesi more clearly subordinates it to the melody. Another neat contrast is between their renditions of the Canzonetta – Piemontesi is lithe, rhythmically precise, and even a little funny, while Gorus is faster, looser with the dotted rhythms, more festive.

Both Piemontesi and Gorus have some spectacular voicings – at 1:02:00 in Sonnet 104 there’s the emergence of a beautiful lower line in the RH (you’ll miss this now whenever you hear a recording that doesn’t pull it out); and at 26:52 in the ending of the same work Piemontesi goes from emphasising the top to the bottom voice of the chain of descending thirds. In general, though, Gorus is much freer. Sometimes it’s in the way he lets the music come to nearly total halt (1:09:12 in Sonnet 123), anti-metrical, parlando phrasing (the first statement of Sonnet 104’s theme at 58:48), or even in small things like variation in the way arpeggios are placed (51:23 – the first three arpeggios three start on the beat, the fourth before).

Piemontesi
00:00 – Sposalizio
07:20 – Il Penseroso
12:15 – Canzonetta del Salvator Rosa
15:19 – Sonetto 47 del Petrarca
21:21 – Sonetto 104 del Petrarca
27:37 – Sonetto 123 del Petrarca

Gorus
34:54 – Sposalizio
43:30 – Il Penseroso
48:12 – Canzonetta del Salvator Rosa
51:08 – Sonetto 47 del Petrarca
58:11 – Sonetto 104 del Petrarca
1:05:38 – Sonetto 123 del Petrarca

Chopin: Fantasy in F Minor, Op.49 (Tacchino, Rubinova, Avdeeva)

Ashish Xiangyi Kumar 2022-05-06 | Possibly the most un-fantasy-like single-movement fantasy ever written. So much of what makes this masterwork tick is essentially anti-spontaneous: large-scale tonal organisation, structured repetition, subtle motivic anticipation. (That said, it’s worth noting Chopin rather made a habit out of gainsaying genre expectations: his Barcarolle isn’t a barcarolle, the Scherzi aren’t even remotely funny or even lighthearted, the Waltzes are completely undanceable, the Preludes aren’t prelude to anything, the Nocturnes render Field unlistenable…)

To give an idea of how perverse this work is, consider the only texture in the work that actually sounds improvised (everything else is too harmonically dense): the “preluding” transition at 3:13. This little bit of music – texturally wonderful though it is – contains in ultra-compressed form the whole tonal layout of the work’s main five themes (Fm, Ab, Cm, Eb). The unification of momentary texture with large-scale tonal progression is the sort of grand trick associated with Beethoven and Schubert, and it’s nice to see it pop up in Chopin too. Plus, this preluding transition is used to generate all the transitional material in the entire work – see 6:16, 7:24 (and similar), plus the ending flourish.

There’s lots of other structural stuff going on here too – the most important is probably the radiant hymn in B that takes up the middle of the work. The hymn – unlike the middle section of the Polonaise-Fantasy, for instance – has *absolutely no thematic link* to the rest of the work: it’s there as a magic, otherworldly release, and does not pretend to be anything else. But there’s tonal ingenuity happening here too: the key relation of this section to the whole work (B to F minor) is anticipated by the two themes which come immediately before the hymn, which have been deliberately rearranged to be in the keys of C minor and Gb. And the way the piece finds one last pause the hymn at 12:28 – ecstatically gory and tender at the same time – is a fantastic touch.

It goes without saying that there is much more to this work than structure. Much as the Polonaise-Fantasy is a rather abstract meditation on the idea of the polonaise, this is a study of the march. All three (and a half, if you count Theme 4: 5:22) have distinct characters. The first is tragic, the second lilting and a little hopeful, the last vacillating between parodic and celebratory.

00:00 – Tacchino. I can’t tell you how happy I was to find this recording – it’s so very, very, very hard to find performances that actually sound raw and fresh and fun. It’s a fantasy, it ought to sound, you know, fantastic! The main thing here is the sense of spontaneity: listen to the way he accelerates through the rising figuration at 4:50, the actual audible slurs at 5:09, the (hilarious) tempo whiplash with the entrance of March 3 at 5:49, and the incredibly flexible phrasing of the Lento sostenuto middle section – especially the way rubato is used to highlight important rest points.

14:03 – Rubinova. This one is all about texture. Just a profusion of very intelligent little things done well: taking the very first note staccato, the consistent overdotting of the introductory marches, playing up sequential or simultaneous contrasts (a lyrical phrase giving way to the rigid march at 14:46, LH vs RH at 16:06), the pedal-less Theme 1 letting the LH chromatic voicing shine (18:09) before giving way to the most delicate rendition of Theme 2 I’ve heard, the leonine octaves of Theme 4 leading to jagged (un-pedaled) climaxes (19:15). Even the slurs which Tacchino puts into Theme 3 are there, though in a much more subtle form.

27:58 – Avdeeva. A very different approach to either Tacchino or Rubinova – in short, it’s pretty focused on large-scale considerations of form and balance, and on keeping the work as unified as possible. So there’s very little rubato throughout, as the focus is on maintaining the pulse – but this makes moments when the leash is loosened truly epiphanic (e.g., the sudden slowdown at the lovely modulation at 29:11; or the agogic pauses during the climatic 32:33). Another nice thing is the way Avdeeva builds up momentum in large chunks: the transition at 30:58 is taken faster than you expect, but that’s because the whole section is treated as one large preparation for the entrance of Theme 1 – it builds and builds relentlessly. Similarly the transition into Theme 4 (32:40), which starts at a healthy clip and ends up almost precipitously fast.

Alec Sievern: Prelude No.1

Ashish Xiangyi Kumar 2022-05-02 | Once in a while you come across a work that’s such a absolute and immediate joy to listen that it’d be criminal not to share it: this is one of them. (When I finally got around to listening to the file Alec sent me I think I was only 20 seconds in when I told myself this had to go on the channel.)

What’s so great about this prelude? In short: it manages to stuff a huge variety of ideas into just a tiny space in a way that sounds totally convincing. Consider just what you’ve got here: motoric washes of Lydian sound in 6/8, sudden strait-laced (and gently parodic) diatonic runs in 4/4 (m.9), Japanese-style harmony & (tresillo) rhythm (m.10), polymetric descents (mm.11-13) backed by butter-smooth harmony, quartal sequential ascents (mm.14-15).

There are two things, I think, which make all this harmonic and rhythmic variety sit so neatly inside such a small work. The first is detail: there a Ravel-like exactness and rigor to a lot of the writing here. The voice-leading is consistently great (see the descending outer/inner LH voice at mm.3 and 6) and even the smallest motifs have something fun going on (the fact that the RH oscillating figure at mm.16-17 comprises two voices moving in different rhythmic contours, or that deliciously textured closing run in 7ths). And there even more subtle things you only hear the second or third time: the way the tresillo rhythm at m.10 actually sets up the polymetric m.11 by introducing the dotted quaver as a melodic unit (and how m.11 surprises us when the dotted quaver becomes its own pulse, rather than a subset of the tresillo), or how quartal harmony entering from m.14 actually breaks its sequential pattern after the first beat to better reach its target key.

The second thing – which is probably more important – is structure: the way this work balances repetition with variation. The work is in ABAB form (m.1, m.10, m.23, m.31). Broadly, this means that many things which we encounter as pleasant surprises the first time are given an opportunity to sound natural the second. But the repetition is far from mechanical, and many things are recontextualised the second time to sound fresh. So the opening motoric figure gets a chantlike countermelody when it returns at m.24, but is also lengthened through the insertion of a new 5/4 bar in each phrase. The quartal passage at m.14, whose harmony is basically unanalysable vertically, returns with long bass notes at m.33 which suddenly cast it as a A Lydian/F# Lydian/C# Lydian modulation leading to the G Dorian/Bb Lydian of m.37. And the descent at mm.18-19 takes on a much more dramatic character at m.40-41, interrupting Bb Lydian to confirm Eb Lydian as the closing modality, and stripping the arpeggios of their earlier figural complexity .

A brief comment on the performance: this is a Keyscape midi produced by Alec, but it’s a beautiful one. The decision to keep a steady underlying pulse means there’s always sense of forward momentum despite the work’s rhythmic variegation. There’s a lot of (Schiff-like) careful dynamic tiering and shading, and there isn’t a single chord which isn’t gorgeously voiced (in a way, I should point out, that’s quite hard to do on a traditional instrument). And even in non-contrapuntal parts, there is a pretty deliberate delineation of melodic and textural parts, so that nothing in the performance feels even slightly “notey”.

Liszt: Vallée dObermann (Pace, Korstick)

Ashish Xiangyi Kumar 2022-04-22 | A masterclass in thematic transformation from the first book of Liszt's Années de pèlerinage. Texturally, lyrically, and harmonically, there’s so much wonderful stuff going on in here it’s easy to overlook how taut and ingeniously put-together VdO is. In short: a single theme (in minor) gets major-ised, and then its major and minor versions become treated as different themes in a four-part form roughly analogous to sonata form (Theme 1, Theme 2, unstable development, steady build-up to climax). Over the course of the work a three-note motif at the head of the theme becomes more and more prominent, until at points the material is reduced into a descending scale (a trick Liszt also pulls off – albeit via a different transformational procedure, and in an more radical fashion – in the Dante Sonata). The miracle is that something so monomaniacally focussed on such limited material manages to traverse such vast musical terrain – dead-eyed listlessness in the opening bars, lyricism in the second section, blind fury in the third, and rhapsodic ecstasy in the close.

“EXPOSITION” – Theme Group 1
00:00 – Theme 1 (Phrase 1). The first three notes in the LH constitute the main motif (M*); this gets lots of independent development. (Note the similarity of T1 to the opening bars of Liszt’s B minor sonata – same rhythm & harmonic restlessness.) There’s nice modulatory scheme here too: the key up in minor thirds to Gm and Bbm before a common-tone shift to F#m, which is followed by a cunning tritone sub (m.13) that leads to the dominant of Cm.
00:44 – Theme 1, (Phrase 2). At first it appears to take a different direction from Phrase 1, but eventually becomes dominated by M* (which appears as soon as m.10).
02:00 – Transition 1. The dominant of Em is spelled as Cb aug 6, preparing for a move into Ebm in which the first phrase comes to a rest on an gorgeous suspension that spells a Gb augmented maj7. The process is repeated a tritone lower, setting the stage for a return to Em via a B7 b9 chord.
02:38 – Theme 1 returns.
03:56 – Transformation 1. Theme 1 suddenly slips into Eb (and B), losing its tail.
04:32 – Transition 1. At 5:07 this is diverted into a statement of T1 in E Neapolitan minor (= A Hungarian minor’s 5th mode).
“EXPOSITION” – Theme Group 2
05:47 – Theme 2 (Transformation 2). Now in luminous C, with an emphasis on M* in augmentation (both its melodic and rhythmic contours). It’s a cunning touch that the augmented M* has the same rhythmic profile as T1’s second bar (m.2). Moves into A (6:35).
06:40 – Theme 2 further developed (Transformation 3).
“DEVELOPMENT” – dramatic, rhetorical, tense, lots of tremolos
07:36 – Part 1. M* developed as a dramatic recitative over LH tremolos (Transformation 4).
07:59 – Theme 1 appears in the RH, before being transformed into long chains of octave descents (Transformation 5).
08:11 – Part 2. M* developed in the LH in two new and different forms: a stabbing three-note descent in the bass, as well as a chain of descending thirds in the upper registers. Yet again, the theme is distilled into a series of descending scales (8:41; 8:55 – note the prevalence of short descending three-note cells, recalling M*). A climax arrives in the form of RH parallel tritones (9:13). The tension then ebbs via plaintive statements of M* (m.161) in the RH, mirrored by similar descents in the bass (mms.163-164, 167-168).
“RECAPITULATION”
10:10 – Theme 2 (Transformation 6), in E, over rolling triplet semiquaver accompaniment.
11:09 – Theme 2 in inversion, quiet but radiant (Transformation 7). The continuation at 11:34 (m.184) is especially clever – combining an inversion of the 4-note descent of T1 (see the LH of m.1) with the two-note (non-inverted) motif that comes right after. (This two-note sigh is also a truncated M*.)
11:52 – Theme 2 (Transformation 8) now with a quivering, repeated chord accompaniment, growing in strength.
12:21 – Theme 2 in inversion (Transformation 9), with dramatic LH descents and ascents. The second phrase gets some nice modal colour, using the minor iv to move into G.
12:36 – M* developed first in the LH, before leaping into the RH.
12:50 – Theme 1, a rhapsodic octave descent (Transformation 10).
“CODA”
13:03 – M* developed again – but now in the RH, then the LH (the latter with double harmonic colour).
13:16 – M* repeats 4 times in the bass, emphasising the b7 and b6. Passing C aug 6 harmonies in the RH, before a E arpeggio rises up the keyboard.
13:27 – Closing phrase, reintroducing the darkness from the beginning. T1 returns in E double harmonic (= major-ised Neapolitan minor), with a stark emphasis on the augmented sound of the C appoggiatura (recalling m.201 and similar).

Liszt: Dante Sonata (Nakamatsu, Khozyainov, Korstick)

Ashish Xiangyi Kumar 2022-04-09 | A stunner of a work. The Dante Sonata is basically a massive set of thematic transformations on two (linked but very different) themes, interleaved within a sonata form. The efficiency of the musical engineering here is insane – all the music here is derived, more or less, from just a chromatic scale and a descending tritone motif, and yet the result is 16 minutes of music which traverses huge emotional terrain. Within this already impressive large-scale framework, Liszt also inserts many moments of harmonic/textural/structural magic. For instance – the whole-tone harmonic descents you get at 10:49 and 15:37 are actually extensions of the modal colour of the second theme at m.103 (take T2’s opening I-bVII mixolydian progression and iterate it). And that heart-stopping moment at 8:31 when the key slips, with improbable tenderness, from F# to C? Well, it’s just an elaboration of the tritone motif. As indeed are the octatonic scale (3:38) and a long passage built around a chain of chords a minor third apart (11:05).

A brief word on Liszt’s thematic transformation of Theme 1 here, which is unusually ingenious even by his standards. T1 is essentially a descending and rising chromatic scale (1:52). It never really goes away in this basic form – it recurs in in a bunch of transitions – but over the course of the work Liszt gradually de-chromatises it and emphasises its descending opening phrase (5:13), so that it eventually turns into a simple major descending scale. And yet it sounds utterly gorgeous! (See 8:36, 15:11). It’s such a radical transformation it feels like reality-warping.

Interpretively, the Dante Sonata is one of those works with a high floor and a high ceiling. So much of what makes it effective is structure and harmony – and those you can’t really mess up, they’re just there in the score. Where a recording can go from merely good to great is in the textures Liszt writes, which give lots of space for an imaginative or sensitive pianist to do amazing things. The three recordings here do a pretty stellar job:

00:00 – Nakamatsu. A goldilocks recording – not extreme in any direction, but featuring robust, intelligent musicianship and filled with many lovely touches. The phrasing is especially nice – Nakamatsu has a wonderful rhetorical style. See for instance the broadening at 1:11 (m.20), the agogic separation of phrases at 2:10, the entrance T2 at 4:01, the staccato octaves at 5:50, or the lovely rubato at 6:54. There are some pretty creative things that Nakamatsu throws in the mix too. For instance, at 10:18 he directly contradicts Liszt’s piu mosso marking to slow the tempo down quite drastically, but it really works – Motif A opens up into a sudden, sharp grandeur. And at 9:20 he emphases the first of every 4 LH chords so as to create a macabre march.

16:24 – Khozyainov. A performance that excels in the fuzzy places – in everything quiet, dreamy, penumbral. Comes across as almost anti-virtuosic, even though there is lots of virtuosity on display here. Khozyainov has a fantastic gift for texture – I don’t think the primordial, muted entrance of T1 or those despairing accents Liszt places at 18:24 have ever been captured better. The care lavished on T2 is remarkable – at 23:42 for instance (m.147) Khozyainov captures the rhythmic tension between the RH triplets and LH dotted rhythm very well (most pianists tripletise the dotted rhythm). And in the quiet moments the playing is just out of this world – just listen to how the upper voice is separated from everything else at 24:11 (and how the percussive sound of the hammer hitting the string becomes almost as important as the actual pitches). And from 25:07 (m.167) onwards, how the gossamer-thin RH builds before busting into glory at m.175.

33:30 – Korstick. My go-to recording most days of the week, and a real monument in the history of Liszt playing. Epic in scope and conception, huge contrasts in dynamic range and colour, taut tempi. No prettiness for prettiness’ sake : consider T1’s first entry at 35:05 – the whole thing is played as one huge phrase, building from a lament into a blizzard of pure terror. Similarly the apocalyptic intensity of f T1/Motif A at 43:29, or the nervous intensity of the tremolo at 42:15. When the music calls for it, though, Korstick can be extraordinarily sensitive. At 39:10 (m.136), T2 is given lovely cantabile treatment; at 40:40 (m.157), a gauzy harmonic haze gently buoys T1; and at 45:43, T2 starts outs in non-measured tremolos that almost unnoticeably slip into measured ones when the melody gains momentum – fantastically subtle but very effective stuff.

Chopin: Barcarolle in F# Major, Op.60 (Volodin, Dong-Hyek Lim, Zayas)

Ashish Xiangyi Kumar 2022-03-26 | This isn’t a barcarolle. I mean, sure, it’s in the name, but if you’ve actually listened to this and thought, “What a lovely Venetian gondola-song,” you’re insane. The Op.60 is really Chopin’s 5th Ballade – and it probably ranks alongside the Op.52. (I know only one water-based description of this work that doesn’t reduce it to kitsch, from Iwaszkiewicz: “We walk on water into nothingness.”)

The work is fairly self-explanatory, but a couple of notes about what makes it stand out. First: the sheer consistency of its lyricism. The Barcarolle is so relentlessly lovely that it contains significantly less contrast than all of Chopin’s large-scale works. There are only two points where lyricism takes a backseat (about which see below).

Second: the use of ostinato and pedal points. My first real memory of the Barcarolle was hearing its opening LH figuration and thinking it was the most beautiful thing in the world. Throughout the work, the LH has a gentle insistency – both in the rising/falling line of the outer sections, or the pulsing warmth of the middle. As for pedal points – the two non-lyrical sections of the are marked by startling LH pedal points on E (4:50) and F# (7:00). Both moments are extraordinary. The first features one a kind of stasis knit out of chromatic voice-leading, which suddenly opens into the famous sfogato (Ravel: “From the depths, a quick luminous trail rises…). The second is sublime in the Burkean sense: something so intense it passes through beauty into something like terror (or awe). Over that grinding F#, the music modulates into wild territory (A# major, anyone?) before spinning out of control into a Neapolitan augmented 6th (/tritone sub) resolving to the tonic.

Third: the prevalence of unusually dense textures (even by the standards of late Chopin). Nearly every bar features some striking counterpoint, which is never repeated in exactly the same form. The gentle opening LH figure thickens into tectonic octaves at 5:50, while the coda obsessively develops what first appears to be a passing outer/inner voice in the B section (compare 3:01 and 7:00).

Fourth: harmony. Can we talk about the augmented 6ths? There’s the (French) A aug 6 at 2:59, gently dropping the passage from A to G# major; the (German) D aug 6 at 3:23, congealing over a C# pedal, the A (German) aug 6 at 4:24, adding in passing C# minor colour, followed almost immediately by a C (French) aug 6 that reverses that shift; and of course that epic, hallucinatory G (German) aug 6 at 7:29. One other moment I especially love is the B# dominant 7th poking up at 6:38. It’s a secondary dominant (leads to E# dom 7, then A# minor – which, by the way, slips back into C# via yet another A aug 6), but also a tritone leap from the immediately preceding F# harmony.

Because the Barcarolle is so pretty, it’s actually quite rare to come across a bad interpretation of the work. That said, these three performances are spectacular.

00:00 – Volodin. Volodin’s got a marvellously energetic approach to Chopin, and his playing here is both free and intense (those arpeggiated chords at 2:13!). That’s not to say there isn’t a lot of careful musicianship: listen to the way the arc is shaped at 1:28, how he clears the air starting from 4:06, or how the lower trill on the A# enters as an unbroken continuation of the inner voice at 5:49. But the defining quality of this performance is ecstasy: it’s one of the few that is not afraid of the pedal F# at 7:00, and actually follows Chopin’s sempre f instruction.

08:16 – Dong-Hyek Lim. A perfect archetype of “modern” pianism (and I don’t mean that pejoratively!) – incredible control of voicing, dynamics, and lots of sparkling detail. At 9:17, for instance, he nails the notoriously tricky staccato/pedal/leggiero sixth sound (you can even hear the slur over the last four notes of the descent). And starting from 9:47 there is some lovely voicing – first the lower voice in the RH, and then (in the second run) the LH top line transfers to the RH. The rising chords at 10:40 are played like a single long inhalation, while the middle section features fine pianissimo playing and very clear separation of primary and subordinate musical elements. The same hierarchisation features in the exceptionally clear coda (15:28).

16:50 – Zayas. Zayas’ playing is marked by a quality that’s become a bit scarce these days – intimacy. The introduction is laden with rubato and dies away in a long overhang of pedal, while the B section’s theme is taken in a rapt, beautifully voiced whisper. There are little improvisatory gestures scattered about, such as the staccato on the RH turn at 17:44 or the LH voiced nudged forward at 22:34. And there is also a surprising sense of large-scale structure: Zayas holds back for much of the piece until the main theme makes its return at 23:11; and in the coda she comes close to matching Volodin’s intensity.

Liszt: Années de pèlerinage, Year 1 - Switzerland, S.160 (Pace, Piemontesi)

Ashish Xiangyi Kumar 2022-03-13 | The Annees de Pelerinage are testament (if it was needed!) to Liszt’s staggering abilities as a manipulator of sonority, texture, and harmony. The first book in particular features some of Liszt’s most concise, evocative writing – it’s remarkable the number of times he deploys textures here that he uses nowhere else, even as the set retains a generally anti-virtuoso character (compare e.g., the brillante extroversion of the Year 2 supplement). There’s a lot of tenderness and warmth in the writing here, embedded in figuration of crystalline intricacy and big washes of harmonic colour (+ approx. a billion chromatic mediant modulations). It’s unfortunate that individual works here (apart from Au bord d'une source) are not programmed more often, as the AdP is now widely recognised as one of the high points of Romantic piano writing.

Both Pace and Piemontesi boast fantastic gifts for articulation, colour and dynamic control, but deploy these to slightly different ends. Pace leans toward the atmospheric and lyrical, and is happy to dwell in the minutest variations in texture or rubato; Piemontesi is fiercer, faster, and gestural. For instance – both of them reduce the LH of Au lac du Wallenstadt to the barest murmur, but Pace phrases the LH non-literally, while Piemontesi keeps it perfectly steady and presents the melody semplice. In Eglogue, Pace takes a dreamy tempo and is at pains to distinguish between staccato and staccatissimo in the second theme 19:09 (m.26); Piemontesi’s faster tempo produces something more rustic and playful, but still laden with wonderful detail, such as the teasing apart of inner voices at 52:10 (m.14) or the melting passage at 53:56 (m.96 onward).

Pace is at his absolute best when performing works which call for the generation of very particular, unusual sonorities. In Chapelle de Guillame Tell he absolutely nails the middle section – the initial horns call at 1:23 (m.21) are given expressive agogic accents, and the echo effects starting from 1:52 (m.33, now without agogic accents) are strikingly audible. The outer sections also feature some very fine playing: the way the melody chords are voiced at 0:51 (m.13), or how at 3:05 (m.61) Pace arpeggiates through both RH/LH chords as if they are a single large chord. The absolute standout of Pace’s set, however, must be Au bord d’une source. The voicing here is exquisite – listen to how clearly the melody emerges at 10:31 (m.5) or 12:13 (m.33) despite dense inner textures. At 11:13 (m.17) the LH transforms into a transfixing harmonic glow supporting the sparking upper sonorities. Inner voices also come through beautifully: see the arpeggiated chords at 11:31 (m.57), or the inner appoggiaturas at 13:16 (m.57).

Piemontesi’s strengths are on full display when the music takes on an improvisatory air or builds in intensity – he’s got a real gift for shaping climaxes and generating ecstatic colour. His rendition of Orage is cataclysmic – see those menacing buildups at 47:49 and 49:54 (m.93), his dangerously fast but perfectly controlled LH octaves at (47:59; m.8), or the throbbing chromatic thirds at 48:32 (m.38). In the long, sustained buildup in the second half of Les cloches de Geneve Piemontesi is spectacular too – the articulation of the “quasi arpa” figuration from 1:02:50) (m.108) onward has a ringing, rhapsodic quality.

Pace
00:00 – 1. Chapelle de Guillaume Tell
05:28 – 2. Au lac de Wallenstadt
08:37 – 3. Pastorale
10:17 – 4. Au bord d’une source
14:17 – 5. Orage
18:23 – 7. Eglogue
22:01 – 8. Le mal du pays
27:52 – 9. Les cloches de Geneve

Piemontesi
33:53 – 1. Chapelle de Guillaume Tell
39:35 – 2. Au lac de Wallenstadt
42:15 – 3. Pastorale
43:47 – 4. Au bord d’une source
47:37 – 5. Orage
51:50 – 7. Eglogue
54:41 – 8. Le mal du pays
59:22 – 9. Les cloches de Geneve

Beethoven: Eroica Variations, Op.35 (Kikuchi, Katsaris)

Ashish Xiangyi Kumar 2022-02-05 | Among all of Beethoven’s variation sets, this is the most funny, good-natured, and consistently extroverted. There’s lots of hand-crossing (the second cantus firmus variation, Vars. 3, 8, 10, 11), pedal points both funny and sublime (Vars. 9, 10, 13, 15, 16), and only occasional forays into minor keys (Vars. 6, 14).

All the surface glitter might obscure that this is a very cleverly structured work – Beethoven was unusually proud of it (as well as the Op.34 variations written at the same time) writing to Breitkopf, “Usually I have to wait for other people to tell me when I have new ideas, because I never know this myself. But this time I myself can assure you that in both these works the method is quite new.”

For a start, Beethoven introduces the bassline of the work before the theme proper, and provides four cantus firmus variations on it before he gives us the theme – that latter of which also serves as a neat structural hinge because it *is* the fourth cantus firmus variation. Once the theme proper is introduced, though, the bassline continues to dominate the work, as fewer than half of the variations work off the melodic material the theme introduces – the rest explore the underlying harmony in different textures.

The way the work closes is also pretty nice – we get two slow double variations (which significantly deepen the work’s emotional terrain) leading to a surprising Coda in C minor. This ends on a huge dominant pedal (G). From this expectant stasis we leap into a fugue – not in C, but in Eb – on the bassline’s head, before another two slow, majestic variations close. It’s a brilliant series of dramatic structural steps coming in quick succession – a winding-down, an intensification (fugue), and then a call-back and unfurling into sublimity – that represents Beethoven at his very best.

It’s worth mentioning that within the main body of the work, there are groupings that naturally arise from how the variations are ordered. The first four cantus firmus variations are a natural grouping – each incrementally thickens the texture around the bassline. Vars. 1-3 slowly dissolve the theme, while Vars. 4-6 bring it back. Vars. 7-13 form a looser grouping, but in general have a fondness for leaps and pedal points (Vars.8-10 also prominently feature Cb as a colouristic device). Vars. 14-15 are both slower and more intense than the rest, and prepare the transition to the fugue.

00:00 – Yusuke Kikuchi. Bright, clear, open, precise. Exceptional control of phrasing/articulation: see the way the lines are tapered in the first and second cantus firmus variations, the sudden crescendo at 5:30, the turns in Var.11, or the beautiful, brass-like voicing in the ending chorale (21:25). Every line in every contrapuntal passage is perfectly shaped and projected – the fugue is wonderful. The risk of such ultra-precise interpretations is that they can sound too controlled, but there is a lot of spontaneity here too (e.g., the way the bass is varied during the repeats in Vars.1-3).

22:49 – Katsaris. Just a huge ball of fun. The work has an improvised air – tempo fluctuations can be pretty wide, and there are some marvellous variations of voicing/dynamics/articulation on the repeats. Like Kikuchi, Katsaris tends to emphasise the bass more in the returning material, but there’s much else: see for instance how quietly Var.4 is played the second time round, the wild LH voicings in Var.9 (32:11), or the way (different) inner voices are teased out in Var.10 (32:43). Perhaps the most shocking (/pleasing) variation comes in the repeat of Var.13, where identical repeated chords are voiced to imply an arpeggio up the keyboard (34:50; similar to Saint-Saens’ Op.52 no.2).

Chopin: Polonaise-Fantaisie Op.61 in Ab Major (Yoshihiro Kondo, Kate Liu, Leonskaja)

Ashish Xiangyi Kumar 2022-01-15 | The Op.61 is the closest Chopin ever came to the austerity (and strange, liminal warmth) of late Beethoven. In any case, there’s little doubt that it’s one of the worst-titled works in the repertoire. “Polonaise-Fantasy” suggests a brillante romp – instead you get this thing that opens with a harmonic meditation, but from then on presents as craggy, even unfinished: full of failed, fragmentary melodies, unstable progressions needlessly prolonged, vertiginous modulations, all arranged inside a disemboweled ABA (/very lopsided sonata) form. It’s very easy to listen to this work with the expectation that some big lyrical melody will come clean, and to be disappointed when this never quite happens. (The main theme does come to a climax, but it’s an odd one. Where Chopin typically puts a pattern of ascending/descending chords in the LH – see the Ballades and the Barcarolle – here the LH gets bare contrapuntal octaves: part celebration, part nervous breakdown.)

The Op.61’s apparent instability actually belies a surprisingly tight-knit structure. The work’s in ABA form, except that B section is punctured – after its appearance, it continuously leaks material into the rest of the work. The transitions between the sections are very substantial, involving two new themes (3 and 5) that at first seem very different – one’s a pleasant nocturne, the other a doleful polonaise – but are in fact closely related (same rhythmic outline).

This work is also a rare instance of Chopin using large scale motivic unification in his writing. Rather spectacularly, that modest little turn in m.4 (0:39) ends up generating the extended transition passage at m.50 (2:59), the lovely LH counterpoint at m.152 (6:56), the RH melodic contour at mm.216-217 (10:26), and the frantic material at m.225 (10:57).

Why then this work come across as so subjective and dissociative?

A couple of things. The most important is Chopin’s insistence in lingering on or broadening interstitial (/unstable) space – a tendency already at work in some of the slow passages in the Barcarolle and the Op.52 Ballade. Transitory aches are leaned into rather than resolved: quite apart from Themes 3 and 5, you have stuff like the agonizing passage in thirds beginning at m.52 or the massive transition beginning from m.128 (5:45). The latter is quite something: you start with a passage which a younger Chopin would almost have certainly written as a sort of cadenza or fioritura, but here is a stark passage in measured semiquavers, flat and aggressively inexpressive. This intensifies into a chromatic two-voice descent in the RH, and finally reduces into a single line that winds listlessly for six bars before arriving at the dominant of B major (though not without landing on one striking augmented 6th dissonance first).

The other big factor is the nature of the material itself: the themes have an open-ended, subjective quality. The main polonaise theme – to take one example – begins with two standard four-bar phrases, but then overspills into a massive, modulating 12-bar phrase that gradually sheds all hint of polonaise-ness. When Chopin repeats the theme (m.45), it’s even more unstable – not only is there a jarring Eb pedal in the bass, the second phrase isn’t even allowed time to finish. The only theme that really has a stable structure is the B section’s touching trio/nocturne, and it’s pretty significant that it’s this which Chopin uses to end the piece.

Given all the above, it’s unsurprisingly very hard to find effective performances of the Op.61. The three here find different solutions to the Op.61’s interpretive problems.

00:00 – Yoshihiro Kondo. A “keep it together” approach. Steady pulse throughout – Kondo allows himself only the slightest rhythmic liberties. Emphasis on contrapuntal clarity (see the entrance of T2 at 6:35, and dense passages like 4:44) and tone production (see the beautifully controlled climax at 11:40).

13:20 – Kate Liu. Rather the opposite of Kondo – leans into the sprawl, emphasizes the distinct characters of each theme and the enigmatic aspects of the work. Lots of intoxicating detail: at 15:34 (m.24), the interaction between the RH melody and the LH polonaise rhythm is underscored; the suspended Ab at 17:29 (m.73) is given unusual prominence; the return of T5 (24:40) is barely audible, a pale shadow.Very expressive phrasing throughout.

27:38 – Leonskaja. All about rhythm; my favourite recording most days of the week. Opens deceptively, with a very slow introduction, but the contrast at 30:11 when T1 enters with march-like staccato (and sharp textural contrast between both hands) is lovely. At m.254 (40:27) the RH triplet and LH semiquaver rhythm is kept distinct, imparting huge momentum to the finale. This continues into the final run at 40:52, where the last note of each triplet is effectively overdotted. Also the handling of the finale and the lead-up to it is wonderful - real biting intensity.

Chopin: 24 Preludes, Op.28 (Pogorelich)

Ashish Xiangyi Kumar 2022-01-05 | Everyone loves Chopin’s Op.28, but I think very few people understand what a staggering work this is, and just how many out-and-out masterpieces are stuffed into this set (apart from the usual well-known preludes – Nos. 4, 15, 16, 20, 24). The craft on display is extraordinary – lots of harmonic weirdness (/clever voice-leading), rhythmic playfulness (especially hemiolas), and textural invention. It’s hard to avoid the impression when going through these works that Chopin was quite consciously going for something deeply unsettling and strange – almost every prelude contains a grain of weirdness, something unexpected and intoxicating. But you’ll have to check out the (admittedly very long) pinned comment below for the gory (& glorious!) details.

Pogorelich’s recording is a fairly dangerous one – after getting familiar with it other recordings almost become unlistenable, and even today I think only 2 (perhaps 3?) renditions that can hold a candle to it. Pogo’s interpretive method is fairly simple – he takes the basic idea or emotion of a prelude, whether it’s lyrical or violent, and intensifies it *almost* to the point of absurdity. The slow preludes are very slow, and the fast ones breathless; textures are enlivened with sparing use of the pedal, craggy accents, and overdotting. The recording speaks for itself, but some brief notes:

00:00 – No.1, C maj. Crisp, sparingly pedalled, with the last melodic note of each bar pleasingly clipped (as Chopin indicates in the score).
00:39 – No.2, A min. LH flat and expressionless, the RH weirdly improvisatory.
02:49 – No.3, G maj. Beautifully detached LH figuration, over which the RH floats with a surprising amount of rubato (notice how the 16th notes enter late, while the LH miraculously appears to stay in tempo).
03:44 – No.4, E min. Taut, restrained – the way Pogo holds back at 5:12 is very moving.
06:15 – No.5, D maj. Taken fast, with little pedal. Chattering, schizophrenic quality.
06:42 – No.6, B min. Pogo puts agogic accents on the first B of each RH pair – lovely touch (and he isn’t didactic about this too, ditching this when the LH needs a bit more propulsion).
09:27 – No.7, A maj. The rhythms here are close to triple-dotted, but the effect is very expressive. At such a slow tempo, the piece is wistful and almost sad.
10:22 – No.8, F# min. Elastic phrasing, hyperfine dynamic control. The climax at 11:45 is so intense it’s painful.
12:34 – No.9, E maj. Modulations beautifully highlighted. And the attention to rhythmic detail!
14:25 – No.10, C# min. Preternaturally light.
14:59 – No.11, B maj. The absence of pedal makes the entrance of the legato melody at 15:11 a real hair-raising moment.
15:37 – No.12, G# min. Biting, angular playing with a lots of of rhythmic drive. Also lovely the way the bass descent at 16:11 is highlighted.
16:41 – No.13, F# maj. Very slow, very tender. Almost a meditation. I think this one perfectly showcases Pogo’s ability to play with lyric intensity – how can the entrance of a simple chord at 20:39 be so beautiful?
21:38 – No.14, Eb min. Guttural, mesmeric. A tiny fragment of an unexpected melody will appear before sinking back into the murk.
22:04 – No.15, Db maj. Achingly tender in the outer sections, but the middle is like a whole other universe – so huge and terrifying it could be Mussorgsky.
29:25 – No.16, Bb min. We all know there’s a tradeoff between speed and control/expressiveness, but apparently Pogo was not told. In all seriousness – this piece has never been recorded better. The whole point of this piece is that it should be played irrationally, right at the margin where control becomes impossible (contrast all those flaccid, well-meaning performances we get keep getting at the Chopin Competition); that madness is perfectly captured here.
30:29 – No.17, Ab maj. Beautifully layered– lots of notes, but the melody is always supreme.
33:52 – No.18, F min. Demented, hyper-rhetorical.
34:48 – No.19, Eb maj. The slickest performance of this monstrously difficult work – it comes off almost like an aural bonbon. Surprising accents in the inner/LH voices – teasing countermelodies that are just barely there – and daubs of pedal for coloristic effect.
35:56 – No.20, C min. More overdotting! And a beautifully quiet final phrase.
38:04 – No.21, Bb maj. The lyrical highlight of the set. The RH sings an unbroken line, while the LH is dry, barely audible (even though its counterpoint is kept very clear). Perfectly contrasts the middle section, where the pedal dramatically enters in and the LH takes a much more prominent role.
41:03 – No.22, G min. Bitingly intense. The LH is phrased very rhetorically, so as to establish a motivic link to the opening bass upbeat.
41:48 – No.23, F maj. Played leggiero/staccato throughout.
42:33 – No.24, D min. Uncompromising and surprisingly tender by turns. The whole piece takes on a sort of mythic character.

Bach: Goldberg Variations, BWV 988 (Tharaud)

Ashish Xiangyi Kumar 2021-01-20 | So there’s 2 big problems you face when playing the Goldberg Variations.

The first has to do with structure. The GV has structure at three levels. On the smallest level, there’s each individual variation – they don’t all internally climax at the same point, and there’s always the issue of whether repeats are taken, and how the music will be varied if repeated. On a second, larger level, the 30 variations here come in 10 groups of 3 – each group contains one virtuoso (“toccata”/“arabesque”) variation, one character piece, and one (fairly) strict canon, with the canons presented in a sequence of increasing intervals. (The virtuoso variation and character piece switch places in the first group of 3, but otherwise the character piece comes first.) But on top of this almost fanatically rigid structure (broken only by the last variation), there is a third, much freer one – in which the GV reaches an obvious climax in the French Overture (Var.16) marking the halfway point of the piece, reaches an inverse peak of emotional intensity in the “black pearl” of Var.25, and then gradually recovers into the joy of Var.30.

The problem is figuring out how these structures – all of which are very (very!) different – mesh together; keeping each variation distinct while also giving a sense of larger narrative shape. I think it helps that the apparently static grouping-in-3s structure that Bach sets up is actually much more flexible than it first appears. The canons are all very different – between the 9 of them we have 8 time signatures, some (Var.3) don’t even sound canonic, and all (except for Var.27) are accompanied by basslines that hide a lot of their strict contrapuntal character. Two variations which aren’t canons (Vars. 2, 23) also have quite a good time pretending to be.

The second problem has got to do with the fact that the GV is not just diverse but *ridiculously* so, to the point that it’s basically impossible a lot of the time to detect the bass line holding 75 minutes of music together (actually, I don’t think a longer set of variations comes around until Reicha’s Op.52). I think the trick to dealing with this problem is just not to fight it at all – the GV is an example of variation as a subterranean form, something you feel rather than detect in the underlying movement. It’s a pretty fun exercise to try to follow the bass line through the variations, and see how often Bach paints himself into a corner and has to engineer an escape by compressing it, filling it in, or treading harmonic water.

Tharaud’s recording of the GV is a triumph. It’s the kind of wonderful music-making that makes an hour-plus of fairly densely argued music sound straightforwardly enjoyable. There’s very little point-making or fretting over profundity – instead: lots of humour, a miniaturist’s devotion to characterisation, heaps of colour and air. If I had to reduce Tharaud’s style to a defining feature I’d probably point of his phrasing. He’s got quite a parlando style, with semiquavers that – as Argerich might say – tend to “lean forward” a little. Some of the things he does are really subtle but incredibly effective. Some examples: the little acceleration at m.29 of Var.4 (9:48), the minute dotting of the note at the top of the phrase at m.19 of Var. 7 (14:26), the surprising rhythmic steadiness of Var.9, the tiny adjustments all through Var.13 (23:45 etc), the slightly sped-up quavers of Var.22, and the improvised, rhetorical manner of the alternating chords in Var.29 (with octaves imitating organ pedals). There are some big stonking flourishes too – most obviously in Var.16, the French Overture, whose opening chord overspills into a lush arpeggio (35:24) followed by a long, taut silence. Some of the textural choices here are also worth pointing out: the melting running lines of Vars. 6 & 19, the delicate and oddly vulnerable Var.18, the staccato in the quodlibet’s second half.

Liszt/Gounod: Waltz from the Opera Faust (Leschenko, Thibaudet)

Ashish Xiangyi Kumar 2021-01-01 | Two barnstorming performances of one of Liszt’s best transcriptions. Both Leschenko and Thibaudet have top-flight hypervirtuoso techniques, and it’s kind of interesting to see how they’re used to such different ends here. Leschenko has a jaggedly white-hot, expressionist playing style – something of a rarity these days, when it actually comes across as quite old-school. This kind of freewheeling style is really dangerous, because you’re always skirting *this close* to total chaos, but it’s amazing how everything is kept on balance here, and how well-judged the musical choices are. For instance: the startling accents in the opening build up tension leading to the triumphant octave storm, the absurdly fast stretta passage (8:20) is still beautifully phrased and dynamically shaded, and the sudden dynamic change starting from 6:48 leads neatly into the recapitulation of the main waltz theme. In general: lots of rubato, textural fun, dynamic variation (it’s surprising, given the overall impression of constant brilliance that her playing creates, how often Leschenko is playing in pianissimo – 5:00 etc).

Thibaudet has a much less free/rapturous style (contrast 1:15 and 10:33), but he plays scrupulous attention to Liszt’s articulations, and his relative rhythmic stability lets him build into these really satisfying paragraphs of sound (see the first entrance of the waltz for instance, which has a real swagger to it). His filigree passagework is preternaturally even (listen to the silky runs at 14:53), and he does some surprisingly funny things too (for instance: the glissandi starting from 17:00, which are taken softly, with accented first notes).

This transcription is well-known enough that it doesn’t really need an introduction, but it’s worth pointing out how much Liszt improves on the original material. The original waltz is rather bouncy and trivial – nice enough, but honestly pretty un-Faustian. Liszt’s addition of the b6 chord + chromaticism in the opening, plus the thickening of the waltz texture (fat chords, that thunderous low A) give a merely fun theme a kind of danger that it otherwise wouldn’t have. Liszt also adds in some lovely chromatic extensions to the melody at 3:26, which are considerably more poignant than the rather too neat wrap-up the melody gets in the original. In a similar way the transitional passage at 4:42 is lengthened considerably from the original (so that it feels a bit Wagnerian, actually).

00:00 – Leschenko
09:11 – Thibaudet

Stravinsky-Agosti - Three Movements from The Firebird (Rana)

Ashish Xiangyi Kumar 2020-08-25 | Agosti’s fearsome transcription of The Firebird is so maniacally colourful that listening to it feels a bit like a Kandinsky painting just punched you in the face. The number of canny decisions that have to made to wrestle all of Stravinsky’s ingenious orchestration into something that coherently fits on a piano is kind of ridiculous, and the fact that this transcription even works, on any level, is a miracle.

The variety of touch, articulation, dynamic control, rhythmic precision, and rubato which Rana brings to bear in just under 12 minutes of music is pretty hard to believe. In the Infernal Dance, for instance, the melodies skittering madly through different registers connect perfectly, and the voicing is never less than inspired (see 0:51, the rubato-laden melody at 1:49). The Lullaby is basically a Ravel-esque study in dynamic tiering, and some of the stuff Rana pulls off at the quieter end of the spectrum is breathtaking (see the top voice at 7:03, the contrast between the light and dark colour at the piano’s top and bottom registers starting at 8:02). The Finale here is the only performance I know of where the tremolos are really played très fondu (meltingly) – you can barely hear the individual notes, and even the scales dissolve into these huge, brilliant ripples of sound. The gradually increasing tempo over the course of this last section is also wonderfully handled (and reminds me a bit of Stravinsky’s own conducting, actually). And that gloriously long last chord – after hearing this one, you’ll think that all the others are too rushed.

0:00 – Infernal Dance
5:05 – Lullaby
8:42 – Finale

Beethoven: Piano Concerto No.3 in Cm, Op.37 (Argerich)

Ashish Xiangyi Kumar 2020-05-30 | Argerich never produces recordings that are anything less than excellent, but even by her atmospheric standards this recording is unbelievably good. The playing is mercurial and fiery – but there’s also a real sense of integrity to it, and in the middle movement the pianism reaches a trancelike level of concentration. It’s a very Argerich thing, actually, to play in a manner that sounds completely spontaneous while also having every musical decision also make total sense – consider how carefully she varies her trills, for instance: 5:41 (quiet, even), 8:30 (percussive), 14:27 (from forceful to wafer-thin pianissimo), and 26:59 (mechanical, rigid, even a bit frightening).

In the first movement, you’ve got the granite-hewn hardness of the piano’s entry put in violent contrast against the expressiveness of the rising chords at 3:28. And immediately after, the melodic lines at 3:37 and 4:17 are presented improvisationally, with dozens of dynamic dips/rises and microscopic changes in tempo. Even a simple thing like the falling arpeggio at 4:07 is given a steely brilliance that lifts it a notch above what you usually hear.

The second movement is taken at a really slow tempo, but this works wonders – the ending of the movement seems to stretch to musical infinity, and 20:09 becomes into a moving meditation between piano and winds. And I’ve not heard a better handling of the LH tremolo at 17:17, which here dissolves into a translucent haze of sound.

The last movement is taut and combative, filled with jabbing accent, and some surprisingly coy moments (e.g., 28:14, where Argerich plays the alternating high and low phrases very differently: the first rhythmically tight and detached, the second drowsily legato and noticeably slower, with a hair’s-breadth pause between the two). My favourite modulation in all the Beethoven concertos is also handled brilliantly here: the piano cuts off the orchestra mid-sentence at 31:17 by coming in very loud, and with these arrogant agogic accents. And needless to say, the coda is taken at a blistering, reckless speed: 34:24.

Beethoven: Piano Concerto No.3 in Cm, Op.37 (Minnaar)

Ashish Xiangyi Kumar 2020-05-30 | Beethoven’s most concentrated and dramatic piano concerto. What’s amazing about it is not just that its themes are so powerfully built, but that the whole structure of the piece is designed to maximise the sense of drama across all three movements. For a start, the Beethoven packs the orchestral exposition with so much material that that the piano isn’t left with any truly new theme to introduce in its own exposition. Instead Beethoven uses this opportunity to enact some very neat thematic transformation – 0:15 becomes 3:47, and 1:05 becomes 4:31 – which gives the movement a sense of continuous development, but also a terse, almost hungry feel. Then there’s the motivic development of the work – the march-like motif in mm.3-4 practically dominates the development section (starting at 7:38), but is nearly completely absent from the otherwise hyper-dramatic cadenza. The reason for this is revealed when (m.march) enters in the timpani at the end of the cadenza – this is the first time Beethoven really tries to turn the cadenza’s ending into something really magical, an effect enhanced here by how quietly the cadenza ends, and how the piano continues playing after the orchestra enters (it’s much more usual for the orchestra to come in with a big tutti, while the piano falls silent). In hindsight, you can almost imagine that Beethoven wrote (m.march) with the timpani in mind, and deliberately withheld us hearing it in its most natural form until just before the end of the movement.

The way keys are used here is also crucial to the narrative of the work – the 2nd movement is in E, which sounds coolly ethereal compared to the Cm of the 1st movement. But it also integrates references to Cm – most obviously the sudden transition to G at 17:53, but also the wonderful C chord which emerges at 18:05, barely kept aloft by the E/C tremolo thrumming below. The transition from the distant E to the Cm of the third movement is also masterfully handled – the 2nd movement ends with a G# at the top of the harmony, which is then interpreted as an Ab in the first note of the rondo. And the rondo itself exploits this common tone – at 30:09 in the rondo, Beethoven first turns a G to an Ab in the orchestra, and then the Ab to a G# in the piano, ushering in a gorgeous developmental episode in E. It’s by far the most fun modulation Beethoven pulls off in all the concertos, and a wonderful example of how key relations can be used *simultaneously* for large-scale dramatic ordering between movements, and also for in-the-moment colour.

Minnaar and the Netherlands SO (led by de Vriend) are excellent here – the NSO’s razzy valveless brass adds a lot colour to a concerto that really benefits from such a fierce texture (see 25:57 and similar). Minnaar’s playing is actually surprisingly expressive for a concerto that’s gained a (reasonable) reputation for being rather merciless in its outer movements. At 4:31, for instance, the transition theme is taken with a detached LH and a very free RH – the way the legato line slinks across the beats is just beautiful. Minnaar also takes the descending passage at 30:54 very softly when it’s usually played forte, which makes the dominant prolongation actually feel a lot more tense than it usually does. Plus the phrasing in the concerto’s hair-raising quiet moments is wonderful (the 40 seconds or so starting from 7:52), there are some nice colour changes (the LH at 10:59, just barely texturing the RH), and sometimes a lovely bit of startling articulation pops to the fore (21:26). The tempo choices here are also work pretty well – the 2nd mvt is taken at a rather classical clip, making it unusually warm and vigorous, and the 3rd is slower than usual, which impacts a real sense of menace.

Beethoven: Piano Concerto No.2 in Bb. Op.19 (Helmchen)

Ashish Xiangyi Kumar 2020-05-23 | The Op.19 is the only one of Beethoven’s piano concertos that can properly be called a work in the High Classical tradition – it was mostly written between 1787 and 1789, when Beethoven was in his teens and well before he started work on what is now known as the 1st Piano Concerto, deploys a small orchestra without any timpani, trumpets, or clarinets, and uses (especially in the first movement) a lot of figuration that could pretty reasonably be called Mozart- or Haydn-like. But it’s also remarkable that even at this early stage, Beethoven was producing moments which really chafed against the classical idiom. The most obvious of this is the rapt “con gran espressione” passage in the slow movement (20:37), which seems to draw a line straight to late Beethoven, but there’s also the lyrical flight at 17:39, and all those accents in the last movement aggressively misplaced in the middle of the beat (22:42; 23:58, 24:55). Even the first movement, which has a typically neat Mozartean structure and a Haydnish profusion of thematic material, features a bunch of unusually intense gestures, such as extended chromatic colour (4:30), the ominous bass oscillation at the end of the development (7:49), and the obsessive diminished 7th harmony at 7:16 – let alone the whole cadenza, which was written down long after most of this was composed.

As usual, Helmchen delivers a superb performance – he really is just incapable of generating an uninteresting texture, and his ability to pull off even the most fleeting and subtle variations of articulation is unparalled among the Beethoven concerto recordings I’ve come across. Some highlights: the rarely-observed but lovely pianissimo at 5:17, the burbling LH at 7:06, the gorgeous slurs at 17:26, the agogic accent on the first note of the syncopated phrase at 22:42, and the way that intensely dramatic LH counterpoint leaps into the foreground at 25:28.

Beethoven: Piano Concerto No.2 in Bb, Op.19 (Schwizgebel)

Ashish Xiangyi Kumar 2020-05-23 | A wonderfully spiky and exuberant performance, practically thrumming with energy from start to finish. Part of this is the wonderful playing of the LPO under Thierry Fischer – its attacks (very noticeable in the first movement) are unusually aggressive, with a good deal of string buzzing caught up in the sound picture, and the strings and woodwinds are exceptionally perky and responsive (the woodwinds at times can even be a little astringent). But the other part of this really is Schwizgebel’s extraordinary playing – he has the very unusual and very Argerich-ian gift of being able to make the most ordinary figuration, be it a trill (23:26; 24:56) or a scale (22:45; 22:59), sound absolutely thrilling. His approach in the slow movement is pretty nice too – not a lot of rubato (with the startling exception of the very free recitative) but every line is shaped very cleanly, and the music preserves a lot of its inward momentum.

Beethoven: Piano Concerto No.4 in G, Op.58 (Minnaar)

Ashish Xiangyi Kumar 2020-05-16 | One of those recordings you just fall in love with a dozen bars in – everything about it is just so vital & joyful & expressive. Start with The Netherlands SO under de Vriend – it’s not a period orchestra by any means, but boasts some inspired HIP touches. The horns and trumpets are all valveless period instruments, and give the tuttis an exhilaratingly razzy sound – see the soaring texture at 2:00 (I swear to god, this was so unexpected the first time I heard it I actually teared up a little), and similarly 12:53 and 28:03. The timpani here, too, is a period instrument (those sharp attacks at 22:42 & 25:17!). What results is a much more textured and responsive orchestral sound than usual, with some lines that are often hidden (e.g., the horn + timpani gesture at 29:07) gaining a lot more presence. The strings also use very little if any vibrato – this trades away a bit of body in the sound, but the purity of the resulting tone creates a wonderful yearning quality in many passages (4:33; 16:36; 23:35; 29:34) and makes for some nice dialogues with the piano (the cello at 22:22 & 22:35).

Minnaar’s playing here is extraordinary, equal parts tender and brilliant – he’s just at home in sparkling triplet runs and brilliant trills (4:17; 22:25), coyly dissonant passages (4:43), the sfumato outlines of the development (7:36), and the jagged shapes in the cadenza (13:53 – he plays B.’s second cadenza, which is more fun than the commonly played first, what with those notes dancing around the trills at the end). The articulation in many passages (3:16; 5:38; 23:50; 26:02) is wonderfully (and subtly) handled, and at particularly climactic points Minnar also lets some rubato assert itself (10:00). One real highlight of the performance is the second movement, where the orchestra is considerably more clipped and brutal than you’ll usually hear, and Minnaar responds with heartbreakingly tender pianissimos (19:30 and similar).

Beethoven: Piano Concerto No.4 in G, Op.58 (Lewis)

Ashish Xiangyi Kumar 2020-05-16 | You’re probably not allowed to say this is the best concerto ever written, but if you did, who would actually bother to seriously contest that claim? I mean, really – the Op.58 is such a miraculous work. It’s kind of an impossible combination of the Waldstein and the last two piano sonatas – it has all the brightness and generosity of the former (semiquaver triplets!), and the lyrical, confessional qualities of the latter (the opening of this concerto, for instance, really seems to occupy the same emotional space as the opening of the Op.110, and the orchestra also gets some lovely contrapuntal writing). Plus the trills in here (and there are lots of them) are amazing – people say the trill reached its apotheosis in Beethoven’s late sonatas, but they make a pretty good case for themselves here.

There have been volumes written about the opening of this concerto, but it’s just one of the most moving things ever written. It’s not just that the piano opens with that intimate, improvised meditation (is there a more terrifying opening to a piano concerto? None really springs to mind) – it’s also how the orchestra enters immediately after in the key of B, not so much having modulated as colouring the piano melody with a kind of otherworldly harmonic brightness (you can even hear the orchestral entry as implying a Gmaj7♯5 kind of sound, which is wild). The whole first movement is packed full of incredible passages – the soaring lyricism of the 3rd theme, the magical entry of the piano, the rapt excursions of the development (7:39; 9:09), the hair-raising trill that ends the 1st movement cadenza (16:39). Structurally, it’s also very efficiently put together (another late-Beethoven trait); the opening’s 4-note pulse bookends the 1st movement, which is littered with almost unnoticeable callbacks to earlier material (compare 7:39 and 3:12). And then, of course, you have lots of fun textural stuff (the dissonances at 4:45) and harmonic innovation (the theme at 1:11 deceives at every turn).

The 2nd movement is the probably most emotionally concentrated movement in all Beethoven’s piano concerti – only 3 pages long, but a tight knot of despair tightly wedged between two luminous movements. It doesn’t really have a structure to speak of – instead, it’s really a series of dialogues between piano and strings (the rest of the orchestra remains silent throughout), with the piano and orchestra gradually converging on a kind of common language of grief. The agonizing trills in this movement will be re-purposed in the last movement to exhilarating effect.

The last movement opens with a deft trick – the theme enters with a repeated C chord that’s quickly established (when the F♯ arrives) to really be the IV of G in m.6. But when the piano enters, it introduces an F♮, confirming that we really *are* in C, before immediately modulating to G. This use of IV colour plays a big part in giving this movement its sunny lyricism – there’s also something a little funny in how often Beethoven builds into these massive G7 passages (25:17; 32:47) in order to prepare for the theme’s entry in C, only to immediately modulate away once C is reached. The first episode in this movement (24:29) is also pretty striking – it’s in two-part counterpoint, but the voices are spaced very far apart (another late-Beethovenism), and float above a cello pedal on D. There is also some lovely colouration with a ♭6, giving the passage a mystic breadth – plus the whole thing is followed up by some beautifully warm counterpoint in the orchestra, especially the woodwinds.

Lewis, accompanied by the BBCSO (under Belohlávek) puts in one of my favourite performances of this work. Because the Op.58 has become over time so wrapped up in associations of profundity/spirituality etc., many performers feel pressured into finding some kind of magic dust in the work (sometimes to great effect), but Lewis shows here that playing in an unassuming, expansive manner can work wonders too. I think this kind of playing is especially well-suited to the Op.58 because it has such a personal character – sure, the opening chords are notoriously hard to interpret, and you can spend hours figuring out how to voice and articulate them, but at the end of the day the most important thing is that they sound sincere, artifice-free, full of breath – and they are, after all that, just G major chords. It’s hard to pick out specific moments in Lewis’ playing that stand out – his eye is really on the big arc here, and although he doesn’t at all ignore Beethoven’s markings he doesn’t go out of his way to accentuate them either – this is very much in the “the music’s good enough to speak for itself” school of interpretation. Nonetheless, several things do stand out – the mystery in the first movement’s development, the violence of the trills in the second movement and stillness of the piano’s final entries, and the beautifully shaped episodes and fiery cadenza in the last.

Beethoven: Piano Concerto No.1 in C, Op.15 (Anderszewski)

Ashish Xiangyi Kumar 2020-05-09 | This is far from a typical recording, but a superb one nonetheless. In short, it’s pretty fierce, and even a little brutal – about as anti-classical as a recording of this concerto can get.

First off, there’s the glorious sound of the Deutschen Kammerphilharmonie Bremen – the brass is more prominent than usual, but what’s perhaps most striking is the raw, hammer-blow power of the timpani (using period-appropriate hard sticks rather than the usual mallets). The tuttis are as a result considerably fiercer than usual (12:19; 35:02), and the last movement has a deranged, earthy intensity (29:07).

The woodwinds are also spectacular. The bassoon bassoons away like there’s no tomorrow (1:57; 4:50; 6:03), the flutes and oboes are lovely in the development (7:51), and that clarinet basically steals the show in the second movement. (In fact, the whole performance of the second movement is striking – the strings use nearly no vibrato and the clarinet just the merest smidgen of rubato, but everything comes together incredibly expressively. I think it’s a function of how the piano’s relative freedom, especially in the development, sets off everything else. Have yet to hear a performance of this movement that’s been bettered.)

Anderszewski is to Beethoven (kind of) what Cortot was to Chopin – he has a way of finding and highlighting intoxicating detail. In the first movement, lots of overlooked LH material gets nudged to the fore, and there are some really surprising but effective pianissimos (5:05, and 13:34 in the cadenza). The second movement is taken at a faster clip than usual, but this has the effect of giving the melody a lift it usually doesn’t quite have. In the last movement, the second episode’s accents are foregrounded in a pretty fun way (31:22) as are the sharp jabs in the coda’s scales (36:20); the tempo of the lovely final modulation (35:42) is also cleverly held back to build up tension for the final tutti.

Beethoven: Piano Concerto No.1 in C, Op.15 (Brendel)

Ashish Xiangyi Kumar 2020-05-09 | The most annoying thing about the Op.15 is how often it gets compared to the concertos of Haydn and Mozart – in its scope and vigour, it is pretty far ahead of what those two produced, and calling it a concerto in the “classical mould” (whatever that means) always comes across as a tiny bit disparaging. For a start, the Op.15 is huge; it is the second-longest of Beethoven’s concertos, falling just shy of the Emperor’s 38+ minutes. It’s also misnamed, as it’s the second concerto proper Beethoven wrote (the Concerto No.2 is the first). But what really stands out about the Op.15 is its sheer energy – there is something approaching a kind of jolly savagery in the outer movements, and the slow movement (the longest among Beethoven’s concertos) is one of Beethoven's most beautiful lyrical outpourings.

The first movement features not just motivic trickery and a bunch of fun modulations, but an *absurdly* long cadenza that clocks in at just under 5 minutes (Beethoven actually wrote 3 cadenzas for this work, and it’s the longest one that’s usually played). It’s probably no surprise that the cadenza for this was written quite some time after the Op.15 was actually completed, from around the period he was writing the Emperor. The second movement, structured in a kind of compressed sonata form, is an extended showcase of the lyrical possibilities of the piano and clarinet, which engage in several lovely duets. (The clarinet was a relatively new instrument at the time, so it’s easy to see why Beethoven was so temped to show it off – in fact, all the other woodwinds, plus the trumpets/timp, don’t make a single sound in this movement, giving it a rather dark colour despite its sweetness.) The movement also ends with a coda whose massive size is fully justified by all the melodic beauty it’s laden with. The third movement is a full 600 measures of rhythmic joy – from the slippery slurs of the main theme, to the off-kilter accents of the first episode, the Style Hongrois trappings of the second episode, and the little woodwind jabs colouring the whole work – it’s consistently boisterous and propulsive and a little bit drunk. Really not the kind of stuff Mozart would have written.

Brendel (accompanied superbly by the Vienna Phil under Rattle) puts in a performance here so thrillingly buoyant, arresting and natural that it’s actually kind of dangerous – if you listen to only this, you become convinced this is the only way the piece can be performed. Perhaps the nicest thing about this recording is how Brendel understands that the work is supposed to be funny – the slurs of the rondo theme in the 3rd movement are exaggerated so that the first semiquaver almost becomes an acciaccatura (29:31), the upbeat that opens the movement is made to sound like a downbeat (to give you rhythmic whiplash later on), and the first beats in each bar of the 2nd episode get a subtle agogic accent (32:08) to give the music a swagger that’s equal parts dramatic and tongue-in-cheek. And even though Brendel has acquired a reputation as an anti-virtuoso pianist, the playing, especially in the first movement, is lithe and powerful. It's hard not to hear the epic octave glissando at 8:55 and wish that every other pianist opted for the same (rather than breaking the run in the LH to hit the low G). The VPO is superb too – the tuttis are wonderfully balanced, and the strings in particular are very responsive (listen to those jabbing Abs in the rondo – 29:56 and similar).

Beethoven: Piano Concerto No.5 in Eb, Op.73 (Helmchen)

Ashish Xiangyi Kumar 2020-05-03 | So this is – for the time being, at least – my favourite Beethoven concerto recording of all time, and it’s all really down to one word: texture. Helmchen is scrupulous in observing nearly all of Beethoven’s fun little accents and dynamic markings, and is freakishly good at differentiating along the staccato/non-legato/slurred-staccato/legato (and the gentle-slur/slur/exaggerated slur) continuum. And the Deutsches Symphonie-Orchester Berlin under Manze puts in some phenomenal playing – the brass is absolutely stellar, the rapport with the soloist wonderfully spontaneous, and the tuttis are some of the best I’ve ever heard. Had a stupid grin on for nearly the whole duration of this performance the first time I heard it, and it’s not become any less impressive since then.

0:00 I. Allegro
20:22 II. Adagio un poco moto
27:25 III. Rondo, Allegro ma non troppo

Beethoven: Piano Concerto No.5 in Eb, Op.73 (Grimaud)

Ashish Xiangyi Kumar 2020-05-03 | The most exultant and thrilling of Beethoven’s piano concertos. It’s actually pretty rare in the so-called classical canon that a work that is just so unreservedly *happy* for most of its length gets acknowledged as a full-blown masterpiece, but the Emperor has deservedly made it. (As an aside, the title “Emperor” has always struck me as really annoying, mostly because this piece really does not have much by way of bombast or imperiousness. Instead it’s consistently generous and open, even as its tone shifts from celebratory in the first to confessional in the second and joltingly dancelike in the third.) Every movement is perfectly put together – the first movement is ingeniously built up from a diverse but closely-related set of themes, the second movement is a study in how a major key can be so tender as to be almost sad, the third has a lot of fun with juddering syncopation (pretending to be in ¾) + colourful modulations, and even the first movement’s orchestral exposition (which in many other concertos is a bit of a plod) is almost unbearably exciting. It’s also remarkable how much of the musical connective tissue in the work is beautiful in its own right – lots of passing material in the first movement, for instance, is really nice.

Grimaud and the Staatskapelle Dresden put in an inspired performance here – Grimaud has a Brahmsian way with the piano part: alternately meltingly beautiful and intensely dramatic, with some truly majestic soundscapes (see 10:26 and the octave passage after), while the SD has some really nice playing in the brass/woodwinds (all those lovely closeups of secondary lines!). Some highlights:

0:03 – Not that uncommon these days, actually, but Grimaud takes the opening arpeggio with measured tuplets. It's a nice effect, turning a standard-issue arpeggio into a much more interesting texture.
5:34 – T2 played in such a hushed/glowing manner (nobody else takes it this softly), and the subtle dynamic shading (and rubato – barely there but definitely there) of the major-mode variant which comes after.
29:30 – The RH arpeggiated chord is taken in interesting way (both here and elsewhere) – Grimaud jumps back to the lowest note of the chord after reaching the top, making the chord sound much more angular.
32:05 – The wafer-thin pianissimo at the top of the phrase beginning here (and similar).

Debussy: Estampes, L.100 (Perianes, Goerner)

Ashish Xiangyi Kumar 2020-04-19 | Two drop-dead awesome recordings of Debussy’s famous triptych, with the first on some days taking its place as my favourite Debussy recording of all time. Perianes takes the work at a slower tempo than you’ll usually find, and there’s a general anti-virtuoso character to his playing. But the colour, melodic shaping and *obscenely* fine dynamic tiering he generates is so much more than worth the tempo tradeoff. In Pagodes, I don’t think any other recording has managed to make the sections at 0:41 and 1:20 sound completely improvised; and in Jardins a whole bunch of overlooked textures are caught – the alternate-bar bass accents starting from 13:42, or the differentiation between the semiquavers and the grace notes at mm.148/150 (14:43). Goerner’s recording has a brighter hue than Perianes – quicker, the fast runs more sharply etched, a little more generous with primary colours and a little less insistent on pastel shades. Where Perianes is sphinxlike in La soirée, Goerner has heady exultancy, and if Perianes’ Jardin opens with a fine mist of water, Goerner opens with a mechanically precise, insistent patter of rainfall, building rapidly into a full-size rainstorm (with superb articulation throughout: see 28:12 for a perfect demonstration of how to pull off a non-legato-quick-semiquaver-against-staccato+tenuto-quaver-planing-chord texture.)

Pagodes (Pagodas) – A surprisingly effective imitation of gamelan music (2:16 and similar), especially since that gamelan tuning is really very different from Western systems. Written in ABA-Coda form (sections beginning at mm.1, 32, 53, 78). Debussy asks for the first theme to be played “almost without nuance”, suggesting an image of a pagoda frozen in a painting (the set’s called “Prints, so) or in a distant mist. The A section is knit together, essentially, from a mixture of pentatonic scales and a rising G#-C#-D# motif (sometimes expanded [m.11, 0:41], sometimes reduced to a C#/D# oscillation [m.15, 0:54], sometimes deployed polyphonically [m.23, 1:20], sometimes augmented [m.37, 2:04]). The LH melody which enters at m.7 (0:27) is especially clever: it not only reinforces the very deceptive syncopation established from the opening (listening to this section with a mental emphasis on the downbeats is basically impossible), but also prominently features the E and A#, the two notes missing from the B pentatonic scale from which the RH is built. The B section introduces some nice Lydian colour, while the coda beautifully “corrects” the syncopation heard at the opening (the rhythm hasn’t actually changed, but you hear the main motif’s rhythmic placement the way it’s actually written for the first time at m.80 [4:26]).

La soirée dans Grenade (Evening in Granada) – One of the most striking things Debussy ever wrote. Basically a habanera (its rhythm pervades the whole work), but one made to venture into some very anti-habanera soundscapes – stark, doleful, mystical, wry, and only occasionally joyous. The structure is a sort of interrupted arch form, Intro-ABCB*A*-Coda (sections beginning at mm.1, 7, 38, 67, 98, 122, 130), but there are so many distinct themes in this one speaking of structure isn’t terribly helpful. The most interesting large-scale features of the work are (a) the way the Arabic-scale theme bookends the piece in a kind of rapt stillness, wound around a high octave C# with the D nagging against it, and (b) how the work’s themes recur in an increasingly compressed form after they are first introduced. Harmonically, there are lots of treats here – planing (including the use of chromatic parallel dominant 7th chords for some textural raspiness at 6:12), whole-tone passages, and the use of an A+E pedal point (over which a C# chord ends up sounding like a A maj 7th chord with a b6 thrown in [m.52, 7:31] – a point Debussy cunningly acknowledges when he actually uses the b6 in the minor iv chord a moment later [7:40]).

Jardins sous la pluie (Gardens in the Rain) – ABCD form (sections beginning at mm.1, 75, 100, 126). A taut little toccata study in various rainlike textures – precise and torrential in the first section, tender in the second, rich and swelling in the third, and exultant in the last. Several motifs recur throughout this piece: for instance, the melody from the A section returns in the C Section at m.112 (13:54), while the B section’s melody gets transformed into a little upward flourish in the D section at m.128 (14:14), and then into a big joyful (éclatant, Debussy writes in the score – “brilliant”) leap up from the bass at 14:21 (interspersed with a chromatically planed version of the A section melody).

Liszt: Ballade No.1 in Db Major, S.170 (Fischer, Pierdomenico)

Ashish Xiangyi Kumar 2020-03-30 | Liszt’s first Ballade is a pretty sadly underrated work, probably because its sibling in B minor, written around 10 years later, has a much more obviously ambitious narrative scope. But the first ballade too is a lovely work of great craft. It’s essentially a series of variations on two themes – a cantabile melody in thirds, and a contrasting gently humorous march. The variations on the first theme are especially striking – all of them heavily emphasise the dominant and are built around untethering the movement of the melodic and non-melodic hands (hence all that hand-crossing). It’s the kind of ingenious textural exploration that Liszt would also deploy La Campanella, and generates some wonderful results here. [A little note: the first theme is actually in two parts, and in the analysis below I’ve separated the two parts when numbering variations to better take into account all of Liszt’s changes in texture.] The march theme, which constitutes the middle section of the work, rather sneakily integrates a motif first heard in the work’s introduction, and undergoes some pretty striking transformations to eventually turn, at its climax, into an anguished cry. (The transformation of Motif A into a heroic theme at 4:36 is also pretty neat.) After the middle section, the variations on the first theme resume, and a heart-stoppingly delicate, glittering final variation (7:04) leads into a final statement of the march theme, which closes the work.

INTRODUCTION
0:00 – An improvisatory line ascends out of the bass (it’s nice entertaining the thought that this is a homage to Chopin’s Op.23), followed by Motif A descending. This pattern is repeated, the second time with diminished 7th harmony.

A SECTION, in Db
0:43 – Theme 1a, a lush melody built around tonic/dominant oscillation, is introduced.
1:11 – T1a, Var.1. Languorous hand-crossing, with Ab pedals in RH.
1:39 – Theme 1b, beginning in the relative minor.
2:05 – T1b, Var.1. Trills in the upper register, then chromatic mediant colour at 2:17.
2:38 – T1a, Var.2. Gentle cross-rhythms strongly emphasising the dominant, plus lots of free-hand-crossing.
2:59 – T1a, Var.3. Broken Ab octaves gently daubing the entire upper half of the keyboard. Still more hand-crossing.

B SECTION, in A
3:24 – Theme 2, an initially hopeful, even self-deprecating march. Modulates into dark C# min at its tail.
3:46 – T2, Var.1. T2 moves up one octave, while the LH now places widely-spaced but nearly empty chords. The modulating tail now takes on a dramatic character and leads into
4:07 – T2, Var.2. Triumphant, in E, with an implied B pedal.
4:15 – Motif A makes a surprise return from the introduction, retaining its slightly tipsy character. But at 4:36, it’s suddenly diverted from its usual downward scurry and attains a rather triumphant sound.
4:48 – T2, Var.3, in F#. Broad and powerful, with lush chords in the LH. But the tail of the theme at 5:02 launches the melody into a sequence of increasingly dramatic modulations over a descending/ascending octave line in the LH, culminating in
5:19 – T2, Var.4. A climactic statement of the march in F# min, with agonising leaps in the LH. But then a zipper-like scale leads back to
5:31 – Motif A, now with alternating chords in both hands. Motif A is developed as before, taking on a more dramatic character, until it culminates in
6:00 – Two strepitoso octave descents, leading back to the

A SECTION, in Db
6:15 – T1a, Var.4. The theme now forceful, while Abs leap from the bass into the top registers of the piano.
6:31 – T1a, Var.5 Leaping trills on Ab in the RH (hand-crossing inevitable), with occasional chromatic runs. (La Campanella vibes on this one)
6:48 – T1b, Var.2. Triplet runs in the RH, with decorative harmonic neighbour tones.
7:04 – T1b, Var.3. The most drastic transformation in this work. The melody dissolves into a series of downward arpeggiated runs that gradually descend the keyboard and grow in strength, reaching the
CODA
7:38 – T2 returns in ff, rapidly ascending the piano until m.188, when Motif A brings us back down to the final tonic-submediant oscillation at m.190 (this is the harmonic progression on which the march is built; see the LH at m.63).

Liszt: Ballade No.2 in B minor, S.171 (Goerner)

Ashish Xiangyi Kumar 2020-03-30 | Written at around the same time as the Bm Sonata (and sharing some marked similarities with it – key, use of thematic transformation, a quiet ending), Liszt’s second Ballade is one of his most important and moving works. It’s built around violent contrasts – the first theme is a kind of black, formless void, while the second is intimate and the third, waywardly melodic. There are three main climaxes (5:34, 9:25, 13:22), the first and last of which feature the first theme, but in radically different forms. The structure of this piece is hard to parse, but it’s essentially in 4 sections which can be loosely analogised to sonata form (but then, any ternary structure can be loosely analogised to sonata form). Across these four sections, the first theme recurs 7 times, growing in intensity each time (until it drops into B maj), while the second is constantly given a new colour whenever it reappears. The use of pure texture here (the opening chromatic line) as a structural feature which is independently developed is remarkable, and when it recurs in split-octave form to accompany the harmonised first theme the effect is pretty amazing (8:45). Similarly too the use of harmony and colour – the opening growl, the Mixolydian colour that slips into the second theme, the magisterial breadth of the climaxes.

Section 1 [“Exposition”]
00:00 – Theme 1 in B Aeolian. A desolate chant over chromatic rumbling & darkly syncopated pulses of vague harmony, climaxing at m.17 (0:39) in a G13(#11) chord. Irregular phrase lengths (6+4+4+3). The chromatic accompaniment becomes its own distinct motif (Motif A) and will receive textural development over the course of the work.
01:00 – Transition 1, with a mystic, Phrygian sound. The bass line uses the first four notes of the T1 melody (F#-G-A-B).
01:22 – Theme 2 in F#. A sharp contrast to T1. Tender and a little yearning, comprised of chords in open voicings.
02:07 – T1/transition 1/T2 repeated a semitone lower(!), almost like an exposition repeat.

Section 2 [“Development”]
04:12 – A march is introduced, integrating a scale and repeated chord motif (Motif B).
04:42 – Transition 2, employing a grating C# pedal
05:09 – Motif A is developed in the RH by being expanded into a split-octave line, while the LH plays (though it’s very hard to hear this) an inverted, chromaticised version of the T1 melody. At m.105 the hands switch roles
05:34 – T1 returns in F# min, in epic form. The first climax of the work.
06:07 – Transition 3, using a tritone substitution at m.134 (6:22) to get to
06:27 – Theme 3, in D. This short, beautiful theme only recurs twice more, always functioning to destabilize whatever tonal centre came before it. It manages to deploy some really striking harmonies while sounding completely natural (the shift from Gmin to A9 from mm.140/1, with the melodic Bb turning into an appoggiatura resolving into B). The turns in this theme into a sort of aural echo of the turn in T2 (m.32, 1:52).
07:01 – T2 in D/G (Mixolydian). Now with closed voicings, giving a more hushed sound, over a D/A ostinato. The use of the C natural gives T2 a bit of warm/dark Mixo sound. T2 moves into G, then abruptly into Eb (with b6 colour/Phrygian dominant sound), which sets up
08:10 – T1 in G# min. Now mf, missing its syncopated chords, and with fuller harmony.
08:37 – Short transition, employing Motif A.
08:45 – T1 in C min, in full chords, now sweeping.
09:09 – Motif A is developed texturally, morphing into a precipitous downward cascade of alternating octaves. This leads to
09:25 – Climax 2. A macabre recollection of the march from 4:12, built around Motif B.
09:42 – A return of Transition 3.
10:17 – T3, in B. Remarkably, the downward run starting on m.230 (10:41) is an elaboration of the LH recitative line at m.18 (0:43).
10:54 – T2, in B (Mixolydian). A new colour, with the chords lower down and alternating hands (the trick is to play these chords so that you can hear the hand-switching), while bells sound in the bass. Wanders into Eb and back to B.

Section 3 ["Recapitulation"]
12:01 – T1, in B. The emotional centre of the work. Not just because T1 is really transformed here for the first time, with full harmony and 4+4 phrases (as opposed to the previous variations, which only intensified it), but because this is the first time in the work any theme takes on a really different character. From this moment on, T1 will only be heard in this luminous variant (call it T1*).
12:28 – T1*, Var.1
12:51 – The final return of T3 in F#, now ecstatic and spanning a much wider range on the keyboard. The tail is lengthened with a long melodic descent that leads into a F# pentatonic octave storm (13:16).
13:22 – T1*, Var.2. This variation and the next constitute the last, glorious climax of the work. The polar counterpart to 5:34.
13:47 – T1*, Var. 3.

Section 4 [Coda]
14:15 – T.2, beginning a long decrescendo/diminuendo over an F# pedal.
14:52 – A slow statement of T.2 (note the link with T.3) closes.

Liszt: Harmonies Poétiques et Religieuses, S.173 (Guy)

Ashish Xiangyi Kumar 2020-03-30 | One of Liszt’s most visionary sets: epic but inwardly so, full of radiant space where you might expect virtuosity, and filled with with gorgeous experiments in colour and harmony.

1. Invocation. An ecstatic meditation built from almost-constant development of 4 motifs (A: the rising 4-note motif starting from m.5; B: the rising/descending octave figure in the LH of m.19, C: the LH descent in m.22; D: the melodic RH figure in m.25). The harmony here is striking: in m.6 you get an almost-fully voiced B13 chord (heard as Amaj7/Bsus2), for instance. The hair-raising passage from mm.77-112 is one of the best things Liszt wrote: it’s built from the tail of Motif A, and bereft of any sense of tonal centre. At m.89 you dwell on an F augmented maj7 chord, and the F7-Eb-Abm harmonic cell that repeats from m.94 is voiced so as to grind in the A in the RH against the Ab in the bass. (This harmonic cell is also deployed in reverse at m.185.)

2. Ave Maria. After 17 measures of introduction beneath an upper F pedal, a simple melody is introduced and developed. The modulations in this one are especially lovely.

3. Bénédiction de Dieu dans la solitude. The interval of the 3rd provides both large-scale tonal structure, in-the-moment harmonic colour, and melodic shape to the work. ABCA form, which each section’s key dropping a maj 3rd. A: built from a gorgeous soaring tenor line and coloured by use of the b6 chord, which anticipates the tonality of the next section; B: m.179, with a theme built from a chain of descending 3rds harmonised fauxbourdon-style; C: m.223, actually a prelude which Liszt had written in 1845. Its melody is like the A section’s – built from a rise/fall pattern that repeats itself higher each time. The C melody is presented in diminution at m.227. At m.253 the A section returns, more sweepingly than before. The ending of this piece quotes the C section (m.330) before magically introducing a theme untethered from the surrounding material at m.340. The B theme is quoted, the C theme reemerges (with a foray into D lydian), and the piece closes.

The different ways Liszt decorates the A theme constitute some of the prettiest stuff ever written for piano -- note the use of water/wave-like patterns.

4. Pensée des morts. An opening 19 measures of tonal and rhythmic ambiguity (containing Motif A, a 3rd descent/rise), followed by a terrifying lament (note Motif A at m.20 and similar, in the bass from mm.33-35, and the melody starting in m.39). A huge recitative on Psalm 130 enters at m.58 (those modulations at m.65!), followed by a recap/combination of material from the first 2 sections. At m.85, a tranquil echo of the Moonlight Sonata (the modulation at m.94/5!), which integrates Motif A at m.143 and similar. The ending (m.164 onward) beautifully integrates Motif A (m.166,) and other preceding motifs.

5. Pater Noster. Recalls 4-part cantus firmus writing from the middle ages, but Liszt goes far beyond this in creating a kind of harmonic dreamscape where otherworldly modulations drift in and out of focus.

6. Hymne de l’enfant à son réveil. AB form. Liszt wrote 5 versions of this. It’s not hard to see why – it’s a very nice (set of) melody(-ies), and doesn’t need to pretend to be more.

7. Funérailles. Lots of astonishing stuff in this one. Opens with one of the best introductions ever written – a growling bass ostinato (a minor 9th that agonisingly expands), while the RH wails out a rising chromatic line (harmonised with diminished 7th chords). The low notes are probably tolling bells – Liszt syncopates the RH so that it never obscures them. The rest of the work is in ABCAB*C* form, and is notable for exploiting the #5/b6 sound – it’s in the stark fanfare at m.19, the E of the A theme at m.24 + the Fb of the B theme at m.57), and colours the final recollection of the C theme in the work’s final bars. The introduction + A&C themes are marches, and it’s nice how Liszt gives them all such a distinct character.

8. Miserere, d’après Palestrina. A stylistic exercise in Renaissance sonority, massively expanded by the possibilities afforded on the piano. (No link to Palestrina that we know of, though.)

9. (Andante lagrimoso). Sounds like late Faure! ABA form. The A section is based on a 4-bar theme (mm.1-4) built around a 2-note lamenting slur motif (in the 3rd beat of each bar, and also in diminution in m.2), with a walking bass below. The A theme is developed until m.22, when the B theme (a loose inversion of the A theme, with its 2nd bar a retrograde inversion) is introduced and developed. At m.66 the A theme returns in reduced form (3 repeated notes + slur) in glowing Ab maj, before reverting to its original mournful posture. The piece ends with 5 bars of extraordinarily modern harmony, which lead directly into the

10. Cantique d’amour. In the key of E, ending the cycle as it began. ABA form; the outer sections variations on a lush 14-bar melody, and the middle section (m.46) chromatic and wistful.

Ravel: Le Tombeau de Couperin (Milstein, Lortie)

Ashish Xiangyi Kumar 2020-01-20 | Is there any composer whose music expends more craft, on a bar-by-bar basis, than Ravel?

Well, ok, so that isn’t really a sensible or meaningful question, but it’s hard to escape the feeling when you’re actually listening to Ravel that his music is put together with a level of care and precision that no other composer quite matches. Dedicated to 6 friends who died in WWI, Le Tombeau de Couperin is the last set of solo piano works Ravel wrote, as well as the last works he wrote in any “antique” form. Le Tombeau represents a fusion of classical discipline & economy with colouristic imagination of the highest order – each of these pieces could be the subject of a longish essay, and despite my best efforts any description will bust YT’s limit, so I’ve put it in the comments.

Bach: Cello Suite No.1 in G Major, BWV 1007 (Lipkind, Yamashita, Dieltiens)

Ashish Xiangyi Kumar 2019-10-12 | Given the foundational place Bach’s Cello Suites occupy in the cello repertoire, it’s kind of amazing to think that they languished in almost-total obscurity until a young Pablo Casals found the score on sale along a Barcelona street in 1890, and that at the time Bach wrote the suites – in 1720 or so – the cello was a relatively new and experimental instrument (we still don’t know what kind of cello Bach wrote these for, since so many variants proliferated during the period). Until today the autograph manuscript of the suites has not been found (there is a remote possibility – not kidding – that it is stashed a trunk lost in Illinois around 1950), and the copy which Anna Magdalena Bach made (the best source we have) might have been from a manuscript which was still a work in progress, or even one that was never completed.

In any case, the cello suites are some of Bach’s profoundest works, and fully deserve their reputation – even when only a single melodic line is being played, they are replete with little miracles of craft and harmony and counterpoint, and it’s remarkable how Bach can resolve one note into another very far away in time and register. They are arranged in increasing order of difficulty (though 2/3 and 4/5 are kind of close), and have a common structure: they open and close with a prelude and gigue, with three old Baroque dances (allemande, sarabande, courante) in between, plus one newer dance (minuet/bourrée/gavotte).

1. Prelude – Composed almost exclusively of broken chords, with scales only entering at three important points. The use of texture here is very purposeful – the prelude opens with a G major chord in its most generous, relaxed spacing, but as tension builds the spacing contracts, reaching its tightest at m.11, when D major is yanked away from beneath your feet, and in the beautiful bariolage passage using the A string at mm.31-6. The harmonic schema is pretty clever too – the Em harmony starting from m.5 sets up an obvious move to D major via an A dominant 7th, but the big landing on D (m.10) is almost immediately cancelled out by a diminished 7th chord which nudges us into Am. And then after the opening formula is repeated, we get a scale (m.19) which leads to a low C# which we now think *must* resolve up to D in the way we were earlier denied, but it instead slips down to C natural (m.22), giving us a D dominant 7th in its most tense inversion. (And since that C can't resolve down on a cello as it wants to, it has no choice but to find closure in the B two octaves above in the middle of m.25.) Also – is there a more moving 6/4 chord than the one in m.39?

2. Allemande – Where the prelude self-consciously avoided scales, the allemande is all about them. And where the prelude played harmonic tricks, the allemande gives us what we expect, with the E in the bass of m.6 dutifully guiding us into D major, where we stay (see the C# m.10). The second strain starts off cabined in a narrow range, but grows into some strikingly jagged textures (see especially mm.19-23).

3. Courante – The allemande was built from small, scalar intervals, and this courante celebrates the big ones. After a perky 8 bars of dance rhythm, it joyfully wanders off into irregular phrases, but note how often the figure introduced at mm.5-6 occurs, either in its original form or in inversion, to build up momentum (see m.36, where it interrupts the structure by being inserted before the cadence). The tension peaks at the D# of m.26 – note also that augmented 4th that leads into that bar, and the augmented 5th leading out of it.

4. Sarabande – Gentle and uncomplicated. As each strain progresses the basic note values are shortened. Note also the lovely war Bach leans on the 2nd beat of each bar by placing triple/quadruple stops there, as well as the technically rhythmically incorrect notation of the demisemiquavers at m.3, which implies the freedom of the gesture.

5. Menuets I/II – Menuet I begins with the G/D/B gesture that also opens the Prelude, Allemande, and Sarabande, and is a sharp contrast to the Courante – while still recognisably dancelike, its sense of movement is offset by a relatively relaxed basic tempo, and it comes across as more heartfelt or lyrical than anything else. Menuet II is built around the Andalusian cadence, and its single line implies a surprising amount of harmony (the 7-6 suspension in the first two bars, for instance, or the 9th chords in mm.9 and 11).

6. Gigue – Lots of nice rhythmic work going on here. The first strain sets up a series of clear 4-bar phrases, but this structure breaks down dramatically in the second strain with the entrance of that looping figure in the second half of m.16 and the use of a 7-bar phrase at the close. Note also how Bach builds into climaxes by alternating between two different rhythmic patterns, and then suddenly losing one to focus on the other (mm. 28-32). Another nice feature is the use of G minor near the ends of strains for harmonic colour.

Stravinsky: Three Movements from Petrushka (Won Kim, Ullman)

Ashish Xiangyi Kumar 2019-09-27 | Stravinsky’s joyous & psychedelically colourful transcription of 3 Movements from Petrushka. I’d heard some of Stravinsky’s other piano music before I got around to this, & didn't get the impression he was a terribly good composer for the instrument – but by the time I finished with this, I was like: Yeah, alright, he's a genius at this too. Just think of the number of textures here that you find basically nowhere else – the rapidfire planing chords at the beginning of the Russian Dance (weird to use the term planing, since the technique is used in such a drastically different way from Debussy), the bassoon line that peeps out from the middle of the texture at 0:55, the muted chordal tremolos at the beginning of The Shovetide Fair, the shy oboe at 8:44, the exuberant canon over the E pedal at 14:01, & all those wild passages of bright, obsessively folkish counterpoint (the 5 main melodies in The Shovetide Fair are derived from Russian folk songs, it turns out). And’s it’s not just about these moments – the whole piece just reverberates with such an unusual & compelling style – percussive & anti-lyrical yet intensely melodic, with long passages constructed from the repetition of tiny motivic cells. Plus there’s the lovely harmony too – from the lydian/dorian colour at the beginning (which then slips right to the opposite end of the dark/bright spectrum by deploying the Locrian #2 at 0:20 – the E seems tonicized in the RH, but there’s also that Bb in the LH), the dirty chords punctuating the end of the Russian Dance, the ecstatic 7th chords at 9:13 (when the violins let the melody loose), the tritone-ized folk tune at 9:50 – all great stuff.

Rachmaninoff: Moments Musicaux, Op.16 (Litvintseva, Giltburg)

Ashish Xiangyi Kumar 2019-05-05 | The first great set of miniatures Rachmaninoff wrote – less acerbic and harmonically dense than the later preludes and etudes-tableaux, more generous and exultant. Nonetheless, many of the traits that would come to dominate the later works – expressive counterpoint, chromaticism, long melodies, huge bell-like sounds – are fully present here for the first time. The set is interestingly organised; it alternates slow and fast pieces, with the slow pieces exploring various romantic genres (Nos. 1, 3, and 5 are a nocturne/variation set, elegy/funeral march, and nocturne/barcarolle respectively), and the fast taking the form of concert etudes that match Chopin’s in their meticulous exploration of pianistic texture.

No.1 – Andantino in Bb min. Five variations on a nocturne theme (0:00, 1:18, 2:18, 3:49, 5:17), with the climax coming late in Var.4, the only variation that forsakes counterpoint for a single sinuous melodic line. Var.3 breaks from the other variations in transforming the theme quite radically (putting it in Gb and 7/4 time), superimposing an overall ABA form on the work.

No.2 – Allegro in Eb min. An intoxicating thing. A syncopated, chromatic line rises out of the keyboard in octaves, while both hands each play their own triplet semiquaver counterpoint. Unexpected shifts in harmony + sly, sudden changes in volume. If this were an etude, it’d be a study in voicing a melody over rapid counterpoint and dynamic control. ABA form, with the middle section beginning at 7:53 built around the two-note rising motif that closed the A section.

No.3 – Andante cantabile in B min. Starts out as an elegy, with the entrance of thirds in the middle section adding a sense of longing. When the theme returns, it is now accompanied by the iron rasp of an octave accompaniment deep in the bass, and it comes clear this is a funeral march. Not too surprisingly, the bass line traces out the Dies Irae (12:48; this is just one of the many, many places in R.’s oeuvre where the Dies Irae is referenced). Also notable is the lovely use of chord extensions at key moments for expressive effect (see the 13th at 16:25 – sounds a bit to stable to just be a passing harmony, to my ear).

No.4 – Presto in E min. The famous one, for good reason. As an etude, it’s indebted to Chopin’s Revolutionary (all that fierce, delicate work in the LH), and also the Op.25 No.11. Essentially, this work inverts the 25.11’s technique – just as the 25.11, rapid figuration is built from alternating chromatic and harmonic notes, but R. puts the figuration in the LH (at least at first), sometimes doubles it an adds octave displacement, and alternates one harmonic note with two chromatic notes (rather than Chopin’s one) to create a more pungent texture.

No.5 – Adagio sostenuto in Db maj. A beautiful and understated piece, where a lot of work is done by small stepwise shifts in harmony. Pretty remarkable how much emotional effect R. can generate with the most modest countermelody or secondary voice. Reminiscent of the middle section of the Op.9 No.1, and Op.27 No.1 (especially the coda).

No.6 – Maestoso in C maj. A masterpiece. Begins in glory and ends in glory, with nothing but glory in between. The ostinato which runs through both outer sections is motoric and almost Prokofiev-like, but sets off the majestically ascending melody (syncopated, aggressively triple-dotted) perfectly – no other composer I know of even comes close to deploying such a texture. The middle section hums with light, and the coda contains another one of R.’s infinite melodies.

Liszt: Two Legends, S.175 (Jando, Pierdomenico)

Ashish Xiangyi Kumar 2019-05-01 | This is how impressionism began. (Well, perhaps even earlier, with Harmonies du Soir and Chasse-Neige, but the basic point stands.) The Two Legends, written in 1863, represent unprecedented attempts at musical naturalism and religious expression – Liszt’s total mastery of pianistic colour and texture only rarely found vehicles of expression as pure as these. The first work narrates the tale of St Francis of Assisi stopping along his travels to deliver a sermon to the first in the trees; as he speaks, the come down and surround him until he blesses them. The second depicts St Francis of Paola, having been denied use of the ferry to cross the Straits of Messina, laying his cloak over the waters and crossing the straits on it. Both pieces are works of genius: the first adopts an unusual call-and-response structure and spins out a plethora of shimmering ornithological effects; it was by far the most sophisticated imagistic treatment of birds in music when it was written (you have to wait for Messiaen, really, to better it, or maybe Ravel’s Oiseaux Tristes, although the listless, depressed birds in the latter seem hard to compare to this). The second has one of the most beautiful themes ever written (somehow Liszt, when in his religious mode, came up with some fantastic melodies: see also Benediction of God in Solitude), underpinned by rippling harmonies and some ingenious deployment of the piano’s deep bass register. The surging waters eventually build up into a storm that nearly overwhelms the theme, until it reasserts itself in massive and glorious fashion.

Both performances here feature some truly exalted sound-painting. Jando is magisterial, with the ability to build up with perfect control into some huge soundscapes (see the long buildup from 11:23 or 18:16 in St Francis of Assisi, for instance). Striking textures abound: see the crystalline 2:09, the eerie LH legato at 11:57, the ragged clusters at 13:54, or the dissonance which leaps out at 17:19. Pierdomenico’s stunning recording is altogether more intimate, and perhaps of all the Liszt keyboard recordings I have comes closest to evoking real spiritual feeling in the manner of late Beethoven. There are too many nice touches to count: the smoky disintegration of the chromatic line starting from 20:52; the careful delineation of inner voices at 29:54; the microscopic pulling-back at 31:45, the resplendent refraction of the right hand through the left at 34:06.

(Notes on structure in the comments section)

Respighi: 6 Pieces for Piano (Scherbakov)

Ashish Xiangyi Kumar 2019-04-14 | Resphigi is famed primarily as a master orchestrator, but it seems that the same vivid awareness of timbral colour which served him so well in the Roman Trilogy, Ancient Airs and Dances, and The Birds led him to write some remarkably effective piano music. The 6 pieces for piano, written relatively early in Respighi’s career, show him deftly deploying a wide range of pianistic textures to magical effect. Scherbakov delivers a wonderful performance – perfect dynamic control, crystalline melodic lines, and some very intelligent handling of multi-layered textures (notice how he slyly modifies the big rolled chords in the nocturne, for instance).

00:00 – Valse Caressante. Lilting and quintessentially salon-esque.
03:26 – Canone. A canon at the octave. Rhythmically playful, as the trailing voice is two beats behind the leading. Starts off rather inward and reticent, before attaining a kind of luminous trepidation when it moves into Eb min. Then a swift move into D, triumphant, before returning into G min. Interesting to note how the voices are placed – at first the leading voice is in the RH, then both voices are in the RH, then they switch places in the D section.
06:10 – Notturno. The emotional centrepiece of the set, and a minor masterpiece. A neverending melody poised above delicately shifting thirds. Impressionistic, with occasional drifts of aeolian colour.
10:14 – Minuetto. A sly minuet whose strong third-beat accents make it sound like it’s in 3/2 time. At 10:21, a passage in pleasingly archaic Renaissance-era fauxbourdon (melody harmonized with notes a 6th and 4th below), contrasting with the dorian-to-the-point-of-sounding-like-the-whole-tone-scale colour of the middle section. The chromatic LH voice in the middle section supplies an unexpected undertone of grating menace.
13:42 – Studio. Shimmering, combining distinctly Romantic harmony with impressionist figuration. Some echoes of Chopin and Lyapunov (maybe even Alkan).
15:16 – Intermezzo – Serenata. A gorgeous melody, occasionally gently nudged aside by occasional bits of filigree.

Schumann: Waldszenen, Op.82 (Hadland, Uchida)

Ashish Xiangyi Kumar 2019-02-09 | Schumann’s Waldszenen are often described as a late call-back to his simple early style, which produced works like Kinderszenen and Papillons. In fact, the Op.82 represents some of the most sophisticated piano writing Schumann ever produced. The set displays a level of cyclical integration that goes beyond what Schumann achieved in any of his other collections. This is apparent at the most general level in the overall key structure of the work: it is arch-like, with Bb major occupying the outer pieces (appropriately entitled “Entry” and “Farewell”) as well as the middle movement. The series of 7 inner movements is also symmetrical, beginning and ending with titles involving hunters (“Hunter in Ambush”; “Hunting Song”). And the movements are bound together through plenty of motivic interrelations. The second last measure of Verrufene Stelle (8:36 – note the chromatic lower neighbour G#) is the origin for the rapid arpeggio flourishes in Vogel als Prophet, and the opening motif of Herberge (10:10, m.1) is embedded in Abschied (18:52, m.3). The figuration of Freundliche Landschaft (8:57) is recalled in m.39 of Herberge (11:31), as is the opening 4-note figure of Einsame Blumen (3:15; cf m.19, 10:49). The same 4-note figure also fleetingly appears in Verrufene Stelle in the LH at m.7 (5:53). Most conspicuously both the first and last pieces in this set have similar endings – a descending arpeggio embellished with a G over sustained bass.

The craft that goes into each of these pieces is also extraordinary. Take the opening of Eintritt: it begins mid-phrase, as an accompaniment missing its melody; but when the melody arrives, it serves only to close the phrase. The phrase structure here is also remarkable, and Schumann plays some neat tricks. For a start, the opening phrase is an odd 3 and a half bars long. But the return of the opening phrase features something much weirder – the repeat telegraphs the opening phrase too early (the second half of m.8) so that it is repeated when you jump back to the opening, as if there has been a memory lapse or the printer slid the plates into the wrong position. In Jäger auf der Lauer you have an ABA structure where the return of A is compressed to a mere 4 bars (2:38), after which there comes a dramatic and lengthy coda which aggressively develops material already introduced. Einsame Blumen features some delicate counterpoint and some wonderfully expressive dissonances (3:31) woven into a concise rondo form. Verrufene Stelle and Herberge contain a wealth of disparate musical material that requires careful treatment to integrate, while Freundliche Landschaft’s quick triplets imply more counterpoint than you’d expect. Vogel als Prophet contains some truly striking impressionistic writing (all that dwelling on chromatic neighbour tones – this is not Messiaen, but this bird is nonetheless exotically dissonant) and craggy phrasing, while Jagdlied’s middle section features some playful syncopation. Much like Eintritt, Abschied closes with a long coda that contains much beautiful new material (more symmetry here between the opening and closing movements)

Hadland’s playing leans toward the unconventional, but – not to put too fine a point on it – it produces the best recording of this set I’ve come across. The tone is consistently beautiful, the legato silken (see especially Einsame Blumen), and the melodic lines are shaped and carried perfectly. The control of colour and dynamics is also stunning – the whole of Jäger auf der Lauer is a study in touch (see the semi-detached (!) legato at 1:52 and similar, the crescendo at 2:08, how the tone magically shifts from loud to barely audible at 2:17, and the voicing at 2:52). While Hadland seems to always have an eye on what’s happening 4 bars downstream, Uchida dwells a lot more in the moment, with a more intimate, detail-oriented approach to the set. She has an especial gift for integrating Schumann’s shifting textures (see Herberge and Verrufene Stelle), and is one of the few pianists who observe Schumann’s very precise instructions re articulation: in the opening of Eintritt, for instance, the staccato-legato pattern of the RH chords can actually be heard.

Hadland:
00:00 – 1. Eintritt (Entry)
01:50 – 2. Jäger auf der Lauer (Hunters on the Lookout)
03:15 – 3. Einsame Blumen (Lonely Flowers)
05:15 – 4. Verrufene Stelle (Haunted Place)
08:57 – 5. Freundliche Landschaft (Friendly Landscape)
10:11 – 6. Herberge (Wayside Inn)
12:23 – 7. Vogel als Prophet (Bird as Prophet)
16:00 – 8. Jagdlied (Hunting Song)
18:43 – 9. Abschied (Farewell)

Uchida:
22:38 – 1. Eintritt (Entry)
25:04 – 2. Jäger auf der Lauer (Hunters on the Lookout)
26:26 – 3. Einsame Blumen (Lonely Flowers)
29:06 – 4. Verrufene Stelle (Haunted Place)
33:01 – 5. Freundliche Landschaft (Friendly Landscape)
34:03 – 6. Herberge (Wayside Inn)
36:07 – 7. Vogel als Prophet (Bird as Prophet)
39:45 – 8. Jagdlied (Hunting Song)
42:42 – 9. Abschied (Farewell)

Schubert: 3 Klavierstücke, D.946 (Sokolov, Avdeeva)

Ashish Xiangyi Kumar 2019-01-26 | If you’re at all like me, you might at some point have had the impression that Schubert was an efficient generator of nice melodies and not a lot else. The D.946 Piano Pieces, written just months before Schubert’s death in 1828, are an uncannily good demonstration of how false that impression is – for while Schubert could certainly write a gorgeous melody, there is something cipherlike and subterranean about much of the music here, even when it seems to be at its most good-natured. In No.1, for instance, the agitation established in the A section never quite goes away, even when the music shifts into the major (0:44), by virtue of the boiling textures in the LH and the urgent dotted rhythms in the RH. Even in the B major trio there is something uneasy – for a start, the opening figure of the trio and the A section is virtually identical, and the trills & tremolos in the trio retain a nervous quality. The second trio in Ab also opens with the same dotted upbeat that opens the A and B sections, and despite some beautiful writing in its first half moves into more twilit music terrain in its second half. [Aside: Schubert crossed out this section in his manuscript (tinyurl.com/y7y635ml), though this decision completely baffles me.] No.2 opens with a melody of Mendelsohn-ian sweetness, but its trios contain some of the bleakest music in the whole set. The first (in C minor) contains thirds muttering darkly over a menacing bass; even when it moves into the major it never remains completely comfortable (see the gently grating F# and Ab in the bass at 17:50, for instance). The second trio (in Ab min) has an intense, paralytic sadness, breaking into a kind of desperate anger in its B minor(!) middle section. No.3. is a study in contrast: the flanking sections are dancelike and helter-skelter, with their deceptive syncopations and 5-bar phrases, but the middle section hitches up the tonality by a rapt semitone to suddenly deliver a reverential chorale. Schubert’s casual mastery of modulation is on display throughout No.3, generating some ecstatic colours: see the section that begins at 28:16 and culminates at 28:30.

Sokolov’s playing here is – well, frankly, weird, although I mean that in the most positive sense. This is a performance that makes explicit the emotional complexity of the D.946. The rhythmic manipulation in No.1 serves to emphasize the disturbing aspects of the music – see for instance the way the Bb-Gb leap in bar 5 is stretched out to an agonizing length, the extension of notes at 9:03, or the treatment of the dotted rhythms in the second trio (the semiquavers are drastically shortened, and the dotted quavers stretched out). The tremolos in the first trio are also given dry, anti-lyrical treatment. In No.2’s first trio (16:36) the thirds are played staccato rather than in the indicated legato, giving the phrase an especially martial character, while the bass dissonances are accented at 18:38 (during the repeat). The second trio’s melodic line is sustained with incredible intensity despite the slow tempo. No.3’s natural contrasts are blown up by the big difference in tempo between the A and B sections – the A section is punchy and even fantasy-like at points, while the B section is hypnotic, with each strange, quicksand modulation allowed to slowly uncoil.

Avdeeva has a much more intimate, natural way with this music, and the performance has a profoundly touching narrative quality (see the trios of No.1 & No.3, which receive some of the best renditions I’ve ever heard). The dynamic control and legato touch on display here is extraordinary, and there is a lot of attention paid to small details in the score – see 34:30 and similar, where she carefully clips the phrase after that long chord after just 2 notes, and the two-note slurs observed at 45:38 and all through the trio of No.3. This is not to say there aren’t many intelligent personal touches scattered throughout – the detached phrasing at 34:40, following by a wonderfully liquid touch on the 5 repeated chords; the emphasis on the hemiola at 51:19 through a sudden drop in tempo; the rubato at 57:00.

No.1 in Eb min, Allegro Assai
00:00 – A section, Eb min
02:42 – B section, B
07:57 – A section
09:24 – C section, Ab
12:32 – A section

No.2 in Eb, Allegretto
14:10 – A section, Eb
16:34 – B section, C min
19:04 – A section
20:25 – C section, Ab min (with its own middle section at 22:12)
25:57 – A section

No.3 in C, Allegro
27:44 – A section, C
28:43 – B section, Db
32:53 – A section

Liszt: Mephisto Waltz No.1, S.514 (Montero, Grynyuk, Goerner)

Ashish Xiangyi Kumar 2019-01-11 | Liszt’s first Mephisto Waltz has become regarded as just a (admittedly dazzling) showpiece, when it’s a wildly creative and technically extremely sophisticated work. (It ought to be celebrated as much as Chopin’s 1st or 4th ballades, it really is that good.) Start with the work’s structure – it’s extremely tightly knit, and follows an A-B-A’-Coda form in which the B section is essentially a theme and variations, and almost every bar in the A’ section engages in some drastic thematic transformation of material already heard before. Most of the work, therefore, consists of continuous development. The efficiency with which material is used is remarkable: the middle section melody (3:08), for instance, generates the passages at m.676 (7:41), m.737 (8:22), m.764 (8:43), m.792 (9:05), and so on. On a more granular level, the colours/harmonies/textures Liszt deploys here are extraordinary. Most famously, the work begins with a pileup of 5ths that sneakily introduces the rhythmic shape of the work’s first real theme. But there’s lots of other wonderful things too: the care Liszt takes to write out the sparkling 32nd notes at m.149 (1:17) & similar, instead of using staccato chords or grace notes; the syncopation & yearning non-harmonic tones (especially the flattened 6th, which becomes its own motif and helps to unsettle major-key passages) which give the middle section an untethered, seductive effect; the way the flattened Bb comes to life in the C section, taking over the role of the dominant & giving the section its double harmonic major colour; the birds at 4:58 and the languorous nightingale in the Coda.

A SECTION
00:00 – Introduction/tuning section. Quintal (and in the unison passages, implied quartal) harmonies. It feels to me like we’re in E dorian at m.14, when Liszt avoids the G# (so stopping this from becoming a usual dominant prolongation), with a Bm7/E harmony entering with the D. At 0:33 rising sequence of fourths, in octaves. Culminates in a huge C# min climax, which then pauses (0:46) as if it has made a mistake, before correcting itself into an implied Esus4. (Note that the bass like of the whole introduction outlines an E-C#-A descent, which explains the rather surprising C# min chord).
00:57 – Theme 1 = T.1, basically just an elaborated A major chord. Note how the opening rhythm of T.1’s first 3 notes is already heard at m.17 + 93.
01:11 – Transition
01:15 – T.2
01:24 – T.3
01:46 – T.1/2/3 presented again, before being capped off at 2:35 by a restatement of T.1

B SECTION
03:08 – T.4, in Db, built from 3 important motifs. Motif 4.1 = (4.1) is the first 4 melody notes you hear starting from m.339. (4.2) is the next 4 notes you hear the melody – a single note (flattened 6th) repeated thrice (or four times), before dipping by a semitone. You first hear it at 3:14, but it’s everywhere, even in the next section of the melody: 3:42, 3:47, etc. (4.3) is the sighing RH followed by an upward run at 4:06. (Note also how the melody starting from m.389 mirrors the shape of T.1.)
04:58 – T.5. Birds, perhaps? Mehistopheles’ laughter? A nice recurring motif, anyway.
05:10 – T.4, Variation 1, in E. (4.1) in syncopated duplets, (4.2) in LH decorated with upward run in RH.
05:42 – T.5
06:03 – T.4, Variation 2, in Db.

C SECTION
07:19 – T.1, sotto voce, in Bb.
07:41 – T.4 developed. (4.1) + (4.2) in bass, beneath arpeggios
08:00 – T.1, now in a frantic, desperate variation.
08:21 – (4.3) developed.
08:43 – T.4 developed. (4.1) in faintly absurd octave leaps, (4.2) transformed into a spasm.
09:04 – (4.2) turns into a kind of heady, glittering cataclysm. Cadenza in double harmonic major at 9:25, the Bb and A clashing violently, before diminishing into a trembling in the middle of the keyboard over which there enters
09:42 – T.5

CODA
09:59 – The Nightingale. A recitative variant of T.4, repeated thrice in ever-higher registers. Even the B-C-Eb-B-C-F# figure at m. 852 (10:39) is based on (4.1).
11:08 – Final flourish, but even this is based on (4.2), grinding away in the bass. The flattened 6ths so prominent in the rest of the work reappear, as does a reference to Bb harmony. The final octave run is based on the double harmonic major scale (flattened 2nd and 6th scale degrees).

Schumann: Ghost Variations, WoO 24 (Anderszewski, Levit)

Ashish Xiangyi Kumar 2018-12-27 | On the night of 17 February 1854, Schumann, suffering from severe aural hallucinations, claimed that he heard angels dictating a theme to him. If Clara Schumann’s diary entries are to be believed, Schumann immediately wrote down the theme, and on either 22 or 23 February started writing variations on it. (All that survives of this first draft is a single page of music, and so we cannot know if at this stage Schumann completed work on the variations, though is likely he had not). At 2 in the afternoon of 27 February Schumann tried to drown himself in the icy Rhine; he was rescued by bargemen who dragged him ashore. The next day he returned to these variations and (it seems) completed them. He sent the work to Clara, but by then she had already left to stay with a friend at the advice of a doctor. On 4 March Schumann voluntarily committed himself to an asylum in Endenich, where he would die just a little over 2 years later.

The Geistervariationen (“Ghost Variations”) are Schumann’s last work. He did not seem to realise that the lovely chorale theme that he wrote down was one he had used several times before: in the 2nd mvt of his Violin Concerto in D min (in a fragmentary form), the 2nd mvt of his 2nd String Quartet, and the Lieder-Album für die Jugend (No.19, Frühlings Ankunft, with a different harmonic colour). Clara forbade the publication of the work (we don’t know why – possibly they were too personal, possibly she thought it was not musically up to par with Schumann’s earlier work), and it was only until 1939 that the work saw print, although Brahms wrote a set of 4-hand variations on Schumann’s theme in 1861.

The Ghost Variations are, like most of Schumann's late work, extraordinarily intimate. All the variations cleave closely to the original theme, never quite departing its soundscape, and the original melody is always present. Rather than dissect the theme, the variations eavesdrop on it: Var. 1 adds triplet counterpoint in a middle voice; Var.2 unfolds as a touching canon; Var.3 opens the theme up just slightly by placing it in the LH and giving the RH gossamer-light triplet figuration; and Var.4 drains the theme of some of its warmth with intricate note placement, glacial and clear as ice. Var.5 represents something of a break from the earlier variations; it follows the harmony of the original theme exactly, but at first blush can be hard to recognise as related to the original theme. There is for the first time something disturbing here: both upper melody and middle-voice accompaniment are awash with chromatic grace notes, with the middle voice chromatic notes given to the LH in a way that just about suggests they have a separate life of their own. If emphasised, as Levit does (19:29), these LH nonharmonic notes create a gently dissonant haze in which the melody is nearly lost – kaleidoscopically beautiful and broken at the same time. Given what we know about the circumstances of this work’s composition, it is hard to be musically objective, but it seems that there is no more appropriate ending to a work that also bookended Schumann’s life. (We’re not even entirely certain if the work is complete – Schumann might have ended it where it did because he was unable to write more, and on purely intuitive grounds I'm inclined to believe this is the case.)

Schubert: Piano Trio No.1 in Bb, D.898 (Trio Dali)

Ashish Xiangyi Kumar 2018-12-15 | Schubert’s Bb piano trio is often treated a lesser (if still important) cousin to the Eb, mostly because – unlike the latter, which changes moods drastically throughout – it’s a mostly sunny, good-natured work. In fact the structural wizardry on display here is pretty astonishing. In Mvt 1, for instance, the recapitulation revs itself up twice in the wrong key before it gets properly going – and even then the true recapitulation slips in almost unnoticed. There's also the way Mvt 3’s trio borrows from the melody of Mvt 2. Mvt 4, however, is where the real spectacular stuff is. It’s labelled a rondo (and you can indeed read it as having an ABCD/ABCD/+ structure, albeit with funky section lengths), but really is in a sonata form where, crazily, the development and recapitulation *overlap*. The section at 36:04 develops Theme 1, but also functions as the beginning of the recap, as the latter opens with Theme 2 and entirely leaves out Theme 1– something Chopin would also do. There’s also some brilliant motivic manipulation here, with what sounds like a brand-new developmental theme at 35:01 being comprised of two motifs already introduced earlier. (One of the motifs is also peppered through the movement as an all-purpose modulatory device.)

MVT I, Allegro moderato
EXPOSITION
00:00 – Theme 1, in Bb. Dotted motif (a) in piano LH.
01:10 – Transition using (a) and incorporating T1’s melody, which is developed into rising figure.
01:52 – T2
02:29 – Cadence theme/T3 (octave leap, followed by scale – loosely linked to T2). The closing bars at 3:17 develop T2’s melody (note similarities in rhythm/melodic contour).
DEVELOPMENT
07:29 – T1, in Bb min. Moves through Db, Eb min, Eb, F min, then into
08:23 – Ab. T2 in cello, with T1 as accompaniment. Moves magically into E, C, and through a beautiful sequence into (what you think must the) dominant preparation at 9:07. The head of T2 is extensively deployed in the strings. Tension builds with the addition of a minor 9th & the diminution of T2 until sudden relaxation at 9:23.
09:35 – Forceful development of T2, switching between minor/major. At 9:39, melody from codetta introduced in cello.
09:52 – A moment of genius. Return to end of the codetta, with T1 in dialogue in strings. Harmony shifts momentarily to F dominant 7th, heralding recapitulation, except that we move, nonchalantly and slightly ridiculously, into
10:01 – Gb! We get an absurd “recapitulation” in this distant key, with strings having additional scalar figuration and melody diverted to drop downward at its tail. Repeats in Db. Then – with no fanfare, we slip into Bb (10:48), arriving at
RECAPITULATION
10:51 – T1, hushed
11:11 – Transition
11:53 – T2
12:30 – Codetta. Moves into true coda at 13:37, developing T1. Huge climax in Ab (13:55). Transition theme closes.

MVT II, Andante un poco mosso
14:33 – A section. Barcarolle. Taken by cello, violin and piano, and eventually blooms with lovely counterpoint.
19:23 – B section. Dramatic. Style Hongrois at first, turning sweeter later.
21:25 – A section, in Ab. Broken chords in piano. Melody moves through gorgeous modulations: Ab, E, C, and back to Eb via touching statement of the melody in pure octaves in the piano (22:41).

MVT III, Scherzo – allegro
24:51 – Scherzo. Puckish & slightly absentminded. Lots of pretty counterpoint.
28:18 – Trio. A waltz missing its downbeats. Melody recalls Mvt 2, A section.
30:18 – Scherzo.

MVT IV, Rondo – allegro vivace
EXPOSITION
32:05 – T1, Eb. The long-short-short rhythmic motif (r) in the first 2 measures will become important, as will the faux-dramatic unison motif (u) at 32:41.
32:43 – T2, G min. at 32:58, orchestral texture introduced, with canon based on (u) in strings. This acts as a transition into
33:06 – T3, Ab. Bell-like, floating over deep water.
33:27 – T2, F min.
33:52 – T3, F
34:10 – (u) is extensively developed. At 34:37, (r) enters prominently in strings, at one point even turning into a rising figure that mimics exactly the opening of T1. At 34:49 we hear (u) enter twice, moving into Db.
DEVELOPMENT
35:01 – An apparently new developmental theme (TD) is introduced, in 3/2 with polonaise-like rhythm and bass drone. In fact, it’s a merger of (u) and (r), with (u) now fitting within a single bar. At 35:18 the connection to (u) is made explicit in the violin.
36:04 – T1, Eb. At 36:32, (r) is developed.
37:06 – (u) chained together in the violin, modulating. Dotted rhythm in cello recalls T2. At 37:22, modified (u) in canon.
RECAPITULATION
37:36 – (u), in its first form, leading into T2
38:02 – T3
38:23 – T2
38:48 – T3
39:06 – Development of (u), as in the exposition. Surprisingly, we don’t get any sense of resolution here, and it looks like we’re heading back into another development. And indeed
39:57 – TD returns, in Gb, as a kind of farewell.
CODA
40:53 – A humorous sped-up version of the development of (u) we’ve already heard twice. (r) enters nearing the end.

Schubert: Piano Trio No.2 in Eb, D.929 (Trio Wanderer)

Ashish Xiangyi Kumar 2018-12-03 | Schumann hailed Schubert’s 9th Symphony for its “heavenly length”, but that phrase could be just as appropriately used to describe this trio, a work of miraculous diversity whose vast swells and deflations have a boundlessness about them. The ingenuity on display here is extraordinary – the implausible modulations throughout (Mvt 1’s first theme is in Eb, the second in B min), the way a tiny motif briefly wedged between the first 2 themes of Mvt 1 expands into an ocean of sound, the gorgeous melody of Mvt 2, the baffling intricacy of Mvt 4’s sonata-rondo form.

Choosing a performance of this work to showcase was annoyingly difficult; loads of recordings were just weirdly lopsided – Schiff/Perenyi is too piano-ey, while the recent Mutter/Trifonov is far, far too violin-centric. This recording is close to perfect – lyrical without gilding this work’s undercurrents of pain, confiding in the slow sections, gripping in the climaxes, deft in those many wonderful modulations. And when the music really should be left to speak for itself – Mvt 2, for instance – it’s left just as it is, perfect.

MVT I, Allegro
EXPOSITION
00:00 – Theme 1 (T1), Eb
00:17 – Motif (a), in the cello. Note semitone descent/ rise pattern (Bb-B-Bb, C-B-C, and so on)
00:53 – T2, B min(!). Developed as soon as it is introduced.
01:51 – (a)
02:10 – (a) blooms into T3, in Bb. Rhythmic/textural change at 2:38 = T3*
02:58 – T4/Codetta, beginning with hemiola & incorporating (a).
DEVELOPMENT
03:31 – Transition
03:41 – T3*, in B. At 3:51, (a) introduced in bass & developed. This structure (T3*, then Motif 1) is repeated twice in F# & Db, in a long-drawn sequence replete with gorgeous modulations.
06:27 – Long dominant preparation
RECAPITULATION
07:20 – T1
08:16 – T2, E min
09:34 – T3, Eb
10:22 – T4
10:48 – Extended coda. (a) returns with a vengeance at 10:54. At 11:11, T2 returns. At 11:54 T1 enters, displaced into rising figure, before T2’s rhythm closes.

MVT II, Andante con moto
12:15 – A section. Melody with some lovely modal colour. At 13:04, in its second strain, a falling motif (b) is introduced (itself from the falling Gs at 12:58).
14:11 – B section. Melody introduced in violin, developing (b). At 15:19, climax of this section is reached.
16:02 – A section. Melody in piano, counterpoint in violin/cello. At 16:59, melody developed further in Ab min and C# min. At 17:30, outburst in F# min. (b) erupts over the Neapolitan at 17:44, which becomes dominant of C.
18:02 – B section. At 19:14, an exclamation, cycling through A, F, and C twice.
20:50 – A section. Coda. Melody introduced momentarily in the major, before returning to minor. In unison, violin/cello state a modified version of the melody a final time, before small cadenza in piano closes.

MVT III, Scherzo (Allegro moderato)
22:01 – Scherzo. Canonic imitation between piano & strings. Magical modulations throughout. (Ab to E, 22:44. Eb to C, 23:13, etc.)
24:53 – Trio
27:17 – Scherzo

MVT IV, Allegro moderato
EXPOSITION
28:48 – A section. Eb.
30:06 – B section. C min.
30:53 – C section. Eb. Lovely modulations at 31:18.
31:39 – B section, preceded by diminished 7th chord. Later presented in the major.
31:10 – A section, vigorously developed. The dramatic descending broken chords in unison recall (b) from Mvt 2 (19:14).
DEVELOPMENT
32:47 – Shift to B min, recalling Mvt 1. Continues in vein of the previous section, developing the A theme. At 33:38, a beautiful syncopated hemiola arpeggio is introduced, under which there suddenly returns
33:42 – the main theme from Mvt 2. Transcendent.
34:27 – B theme developed. At 34:43, the texture/rhythm is transformed: piano plays triplet counterpoint, cello and violin sustain theme in canon. Shortly after, B theme returns to B min & original texture, before surprising burst of F major at 35:24, exploiting ambiguity of preceding diminished 7th. At 35:35, return to 6/8, with dramatic/Brahmsian statement of B theme in Eb min follow by sudden shift to B min.
36:05 – Motif from A theme introduced, immediately followed by its inversion (which also recalls the marchlike texture of Mvt 2). Shifts into Eb min.
RECAPITULATION
36:32 – A section.
37:57 – B section. F min.
38:28 – C section.
39:15 – B section. F min, then major.
39:46 – A section. At 40:11 things appear to be wrapping up. Instead, we get
CODA
40:25 – Sudden shift to E min, where the A theme is developed & moves (through an ingenious set of modulations) into Eb min.
41:01 – The syncopated hemiola broken chords return, with the melody from Mvt 2 slipping in beneath as before. At 41:33, a burst of joy, as we move into the major. The only happy statements of the Mvt 2 theme are heard, suddenly celebratory and generous, in unison in the strings. They reenter twice at successively higher points – a Bb and Eb – before the strings take up the hemiola accompaniment from the piano. A truncated statement of A theme closes.

Debussy: The Complete Preludes (Jumppanen)

Ashish Xiangyi Kumar 2018-11-24 | One of the great peaks of the prelude literature for piano, alongside Chopin’s, Rachmaninoff’s and Scriabin’s (late) prelude sets. These works are consistently ingenious in the deployment of those devices that are now consistently associated with impressionism – modal colour, exotic scales (whole-tone, pentatonic, double harmonic, Hungarian), non-functional harmony (planing chords, purely colouristic intervals, hanging and extended chords), pedal points, timbral figuration. Interestingly, Book 2 is very different from Book 1: it’s far more pessimistic and abstract, so much so that it was greeted upon publication with some disappointment. Book 1’s fantasy has been replaced by a certain weary or ominous realism, and you get the feeling that Book 1 was more explicitly imagistic in nature, while Book 2 is symbolic – it’s going for what certain images and objects *mean*, rather than trying to depict them per se. Book 2 is also completely written on 3 staves – its sound is a bit more expansive and less propulsive, more delicately tiered.

Book 1
1: Dancers of Delphi – A languorous melody embedded in a chordal texture.
2: Sails/Veils – Whole-tone scale, then the pentatonic, then whole-tone scales again. Sailing-boats stuck to a pedal-point
3: The wind in the plain – A displaced trill trembling like grass on a plain. Flashes of lightning in the middle.
4: "The sounds and fragrances swirl through the evening air" – Harmonically rich, disfigured waltz
5: The hills of Anacapri – Luminous bells introduce an ecstatic tarantella
6: Footsteps in the snow – Possibly the bleakest soundscape in the preludes. Trudging footsteps form an ostinato in the LH. Ends in emptiness.
7: What the West Wind has seen – The frozen landscape descends into a maelstrom.
8: The girl with the flaxen hair – The most tonal prelude, recalling Debussy’s early style.
9: Interrupted serenade – A guitarist tries to serenade a potential lover, but is interrupted.
10: The submerged cathedral – First, bells stifled by the depths of the sea (given mixolydian, lydian, and then phrygian colour); as the cathedral begins to rise, full chiming bells, a sustained plainchant, then a booming organ pedal-point.
11: Puck’s dance – The mischief-maker from A Midsummer Night’s Dream. A loose gigue (in sonata form?) stuck through with dorian colour.
12: Minstrels – Banjo imitations interrupted by surprisingly blatant dominant-tonic pedals in G. A warmly sentimental lyric is introduced, as well as a drum-like theme.

Book 2
1: Mists – Fleeting shadows in the RH over a mystic, stunned chorale in the LH. The second theme is played in octaves separated by a desolate gulf in the middle of the keyboard.
2: Dead leaves – Dejected, inbent.
3: Wine door – Inspired a postcard Debussy received from De Falla of a Moorish wine-gate in Grenada. A playful evocation of an archaic habanera, underlaid with supressed violence.
4: The fairies are exquisite dancers – A fluttering dance alternatives with lyric passages.
5: Bruyères – Pastoral, with improvisatory figures that mimic birdcalls. Nostalgic and a bit lonely.
6: General Lavine – eccentric – Alternately comic and sentimental, crude and refined.
7: The terraces of moonlit audiences – Aching, rapt. Possibly inspired by a newspaper article about the coronation of George V as King of India, which concluded with the words: “…the audience was observing the events in a shimmering moonlight.”
8: Ondine – Ravel’s water nymph was hushed and candescent; Debussy’s is evasive, even, at points, slightly threatening.
9: Homage to S. Pickwick – Opening with a forceful but absurd parody of ‘God Save the Queen’, this prelude presents a sequence of contrasting musical styles, all linked only by their humour. Debussy was, for reasons only he will know, awfully fond of the British (see the little jig tune that the character whistles at 1:14:43 while walking into the distance).
10: Canopic jar – Debussy’s desk was occupied by two Egyptian canopic jars: this prelude is a meditation on death, but death held at a great distance, still and somehow hermetic. Even the sudden tonal shift in the fourth bar sounds inevitable, and the recurrent chromatic phrase centred on C# has the character of an incantation.
11: Alternating thirds – An etude, bone-dry.
12: Fireworks – A dazzle through which a deformed recollection of the Marsellaise is heard. Virtuosic and brilliant, but in a brutal way, like light too bright to look at directly.

Prokofiev: Sonata No.6 in A Major, Op.82 (Chiu, Lugansky)

Ashish Xiangyi Kumar 2018-09-22 | The most epic and varied of all the War Sonatas. Prokofiev’s typical inventiveness is present here, along with his favourite textures (bells! toccatas!), but here they are allied with a Beethovenian ambition and mastery of structure. The first movement is stuffed full of ingenious ideas, from the grinding march that opens it, to a displaced chromatic line and a faux-naïve white-key melody. But it’s not the wealth of ideas that astonishes as much as the way in which these ideas are manipulated. Themes will shed one character for another, a tiny idea will grow into a motif of staggering force (see (D) in the structural analysis below, for instance), and in a very Prokofiev-ian way the structure *itself* has expressive function – the return of material invariably serves a narrative or dramatic purpose (4:17 and 6:31 are perfect examples of this). The second movement is a jaunty scherzo, richly spiced with playful harmonies. The third movement is a lush waltz whose theme keeps appearing in different keys, refusing to fully settle. It’s here that evidence of this sonata’s cyclic form creeps in in the form of allusions to themes from the first movement. It’s in the turbulent, coiled-spring fourth movement, however, that the sonata’s cyclic structure becomes fully evident: references to the first movement abound. The structure here is beautifully put together: it’s a kind of sonata-rondo (opening with an ABACADAE!) with a highly truncated recapitulation that, short as it is, contains all the intensity of the entire opening exposition.

Chiu’s performance is clipped and bone-dry, but is deliciously textured and biting. As you’d come to expect from him, there are some nice innovative touches: the slower tempo in the first movement’s recapitulation to emphasises the more gnarly, full-throated timbre of the lower register, the harried and harebrained tempo of the second movement, and the uber-dramatic clipped final note of the sonata. The general lack of pedal also greatly augments the impact of those moments when Chiu finally lets it loose: the bell passages in the first movement or the coda of the last, for instance. Lugansky’s is probably the model “typical” performance. Compared to Chiu, this is a more sonorous and sweeping rendition, although it certainly has its hard edges. In particular, the final movement is taken at a blistering tempo, without an iota of sacrificed precision, clarity, or expressiveness. While Chiu speeds up the second movement and slows the third for contast, Lugansky opts for more balanced tempi that lend the third movement more warmth/shapeliness (cf Chiu’s icy coldness, especially in the B section) and give the sonata a sense of overall unity – something really difficult to do in a work this diverse.

Chiu:
00:00 – Mvt 1
07:16 – Mvt 2
11:24 – Mvt 3
19:30 – Mvt 4

Lugansky:
26:09 – Mvt 1
34:31 – Mvt 2
39:00 – Mvt 3
46:15 – Mvt 4

Bach: Keyboard Concerto in D Minor, BWV 1052 (Bahrami, Dinnerstein)

Ashish Xiangyi Kumar 2018-09-08 | When you mention Bach’s masterpieces, it’s usually the very big (St Matthew Passion; Bm Mass) or very contrapuntal (WTC; AoF) works that spring to mind. But this concerto – all its 20 minutes of not-particularly-contrapuntal music – also numbers among Bach’s best works.

The BWV 1052 is one of the finest examples of ritornello form. In it, a repeated section of music, the ritornello (literally “a little thing which returns”) alternates with freer episodes. The ritornello recurs in various keys, often in highly compressed or fragmented form, and its restatement in the home key usually heralds the close of the piece. The first and last movements of the BWV 1052 are in this form – the ritornello is the material played in union at the beginning of both movements. Bach fully exploits the opportunities of the form, using the ritornelli to anchor movements awash with key changes (if you ignore a bit of episodic messiness, the first movement goes Dm-Am-F-Am-Em-Am-C-Gm-Dm-Bb-Dm, and the last isn't too different). Bach also superimposes an A-B-A’ structure on both outer movements by introducing some gorgeous toccata-like contrasting material in the middle sections (2:11 and 14:36). He also toys cleverly with ritornello form: at 6:02, what you'd expect to be the final ritornello appears, only to be interrupted by yet another episode of perfidia. Similarly, in the last movement we return to the ritornello in the right key (17:48), but are quickly cut off by a massive episode culminating in a cadenza before we finally hear the closing ritornello. It’s exactly the kind of bold structural delaying device we’ve come to associate with Beethoven (and others after him), and it’s interesting that Bach’s musical intuition led him to similar ideas much earlier.

Apart from structure, however, it’s really just the sheer expressive power of the work that hits you. The writing has a raw, even elemental quality; the keyboard entrance throbs with menace, and all throughout the writing for the soloist is both unusually intense and free. An interesting point to note is the widespread use of passages of perfidia in the work – repeated semiquaver figuration that’s meant to imitate (or is derived from) improvisation (see 2:51, 5:08, 6:06, 15:37, 17:04, 17:15). These passages are basically devoid of counterpoint, but represent some of the most powerful and idiomatic writing in the whole work, inhaling/exhaling tension, underpinning gorgeous modulations, scaffolding structure. They’re reveal Bach as master of instrumental texture and melodic line, not Bach as contrapuntalist. The 6-bar opening ritornello is also masterfully constructed, with that painful (syncopated) ever-expanding leap in the first 3 bars and a descent/ascent/descent arch in the last 3. (To explain some of the more interesting features of the keyboard writing here, it’s been suggested that this concerto was based on a lost violin concerto. It’s an interesting idea, especially as some of the keyboard writing – the more toccata-like stuff – imitates bariolage on the open strings, with 2:11 as an obvious example.)

Bach: Goldberg Variations, BWV 988 (Rana)

Ashish Xiangyi Kumar 2018-09-02 | ...

Rachmaninoff: Variations on a Theme of Corelli, Op.42 (Kern)

Ashish Xiangyi Kumar 2018-07-22 | From 1917 to 1930, Rachmaninoff composed no original works for piano. Then, in the summer of 1931, he wrote his last piece for solo piano: the Corelli Variations. It’s a masterwork that’s a bit hard to grasp, because unlike the Op.22 Chopin Variations or the Op.43 Paganini Rhapsody, it inhabits a very constricted emotional space. In short: with the possible exception of one variation, it’s very, very sad. Every variation contains something that’s disturbing or off-kilter – chromaticism (even if disguised: Var.20), icebergs of wayward harmony, unsettling changes in meter, and biting motoric movement. In the Paganini or Chopin variations you build up into joyous perorations, but there there isn’t a single triumphant note.

The work is compact (it’s a lot shorter than the other variations R. wrote) and very tightly organised: three clear movements, with a similar tonal structure to the first movement of R.’s D minor sonata. Liszt’s description of the Moonlight’s 2nd movement (“a flower between two chasms”) pretty accurately describes the middle movement here: it’s also in Db, and is a tiny two variations long. R. rigorously applies structural unifying devices in this work: he prepares the Db tonality of the middle movement by introducing Db harmony in Vars. 9, and 11, and a little rhythmic motif he introduces in Var.13 suddenly blossoms into huge proportions in Vars. 18-20 before a magical coda curls the piece inward and folds it closed.

[MVT I]
00:00 – Theme, austere & pure, desolate.
01:03 – Var.1, Poco piu mosso. Lush initially, but with disturbing accents in bass.
01:33 – Var.2, L’istesso tempo. Motoric, built off theme’s melodic shape.
02:01 – Var.3, Tempo di menuetto. Eerie, with rich harmonies.
02:44 – Var.4, Andante. Similar to 13th variation of Op.22. Modal colour.
03:58 – Var.5, Allegro (ma non tanto). Vacant and primitive, with open fifths/octaves providing most of the colour.
04:19 – Var.6, L’istesso tempo. A development of Var.5, with fistfuls of chromatic movement.
04:41 – Var.7, Vivace. Motoric, imitating crossed strings on a violin.
05:10 – Var.8, Adagio misterioso. Oscillating lines, irregular phrases, pungent harmonies.
06:07 – Var.9, Un poco piu mosso. Extraordinary, almost Debussy-like colours prowling in the harmonic movement of the LH.
07:11 – Var.10, Allegro scherzando. Return to harmonic stability. Mendelssohn-like.
07:45 – Var.11, Allegro vivace. Motoric, brutish toccata.
08:10 – Var.12, L’istesso tempo. A compact and dark march, with wrongfooted rhythms.
08:45 – Var.13, Agitato. Convulsive. Frequent metrical changes, with almost all chords 7ths/9ths/11ths. Introduces a galloping rhythm developed in Vars.18-20.

[INTERMEZZO]
09:19 – Var.13+1. An ingenious construction, sounding like a concerto cadenza that leads into a thematic restatement (compare it to Var.11 of the Paganini Rhapsody). Built around the double harmonic minor scale (flattened 2nd and 6th), which gives it an eastern flavour. Washes of coloristic harmony and dense embellishment loosen the piece’s grip on D minor, preparing for next variation.

[MVT II]
10:34 – Var.14, Andante (come prima). Theme restatement. Played alone, this variation might sound warm – but it the context of this set it is world-weary and resigned. The register is lower than the first statement of the theme, and the harmonies are lusher and darker: the theme seems have aged somehow. This might be a function of the Db tonality, a semitone down from D min, but in the second-last bar a D minor chord appears as a reminder of where we are. (NB: R. also exploits the Dmin/Db tension in his first sonata.)
11:43 – Var.15, L’istesso tempo. A radiant nocturne. The only happy moment of the piece

[MVT III]
13:13 – Var.16, Allegro vivace. Sudden shift back to D min. Open fourths/fifths, with Gypsy scale runs for colour.
13:44 – Var.17, Meno mosso. Open fifths remain. Metrical changes, chantlike melody floating over ostinato.
14:52 – Var.18, Allegro con brio. The beginning of the closing triptych. The little rhythmic seed planted in Var.13 sprouts. Schumannesque.
15:26 – Var.19. Rhythm further developed and made more jagged with the addition of snare drum-like figuration (cf: the dramatic vivace passage at the end of R.’s 3rd PC). The end of this variation dissolves into a chromatic whirlwind (note the chromatic line that also emerges in the LH).
15:55 – Var.20. Climax. A wall of sound echoing from all registers, with lots of implied chromatic movement in the outer + inner voices. Some biting harmonic writing (15:21), and an ending like Chopin’s last Op.28 prelude.

CODA
17:04 – Var.20+1, Coda. Smokily chromatic. Melody unfurling in neverending fashion over a D pedal. A kind of weary, improvised extension of the low Ds ending Var.20. (Cf. the Meno mosso in the Op.22 coda).