Prokofiev: Piano Concerto No.2, op.16

2014-09-08 — Original posting (on Blogger)
2014-09-13 — Added reference to review for a concert on 2014-09-11
2014-11-13 — Re-posting as is (WordPress)
2014-11-14 — Expanded discussion of blind comparisons in the bottom of the posting
2015-07-04 — Added links for music score
2016-07-25 — Brushed up for better readability

My First Encounters with “Prok 2”

This posting is about the Piano Concerto No.2 in g minor, op.16 by Sergei Prokofiev (1891 – 1953). I wasn’t really familiar with this composition, but through YouTube I watched Yuja Wang playing it at the Verbier Festival (2012, probably), with Charles Dutoit conducting. Then, I saw the release announcement for her CD with this composition. This opened a good opportunity to experience that astounding pianist more directly than through the “YouTube filter” with its sound limitations. I felt that listening to a CD should also be more objective without the visual distraction that concerts and videos always entail.

A Failure in a Radio Comparison

Interestingly, soon after that CD arrived (I hadn’t listened to the CD yet), there was a blind comparison of six recordings of this composition on France Musique, in “La Tribune des critiques de disques“. It was quite obvious that they had to include Yuja Wang‘s new CD (a live recording with Gustavo Dudamel). Unexpectedly, and to my disappointment, Yuja Wang’s recording fell prey to outright rejection already in the first round. Evgeny Kissin‘s interpretation shared the same fate. In the case of Yuja Wang, one of the experts did not even care to give arguments for his rejection. However, based on the excerpts that were played, I actually had to agree with the judgement of the three experienced, seasoned critics.

The winner in that comparison was Yundi Li, accompanied by the Berliner Philharmoniker under Seiji Ozawa.

It is quite possible that a good interpretation falls through the cracks in such comparisons: an unfavorable selection of the first excerpt (a part of the first movement in this case) can lead to the elimination of a good recording in the first round. Plus, Yuja’s CD is a live recording, and in a concert a pianist may first need some time to adjust to the acoustics and the audience, which makes “bad picks” more likely.

I felt that I wanted to do justice to this recording by doing my own, full comparison with the winner from that radio comparison. Hence, I’m comparing the following two recordings (in chronological order):

  • Yundi Li, Seiji Ozawa, Berliner Philharmoniker (live, 2007)
  • Yuja Wang, Gustavo Dudamel, Simón Bolívar Symphony Orchestra of Venezuela (live, 2013)

Background, About the Composition:

Sergei Prokofiev (1891 – 1953) wrote his Piano Concerto No.2 in g minor, op.16 in 1912/1913. The composer dedicated the concerto is dedicated to a former friend from the St.Petersburg conservatory, Maximilian Schmidthof. In April 1913, Schmidthof had committed suicide after writing a farewell letter to Prokofiev. The composition premiered in August of the same year, with the composer at the piano. The orchestral score apparently burned in a fire during the turmoil of the Russian Revolution, the solo part remained with the composer and survived. Prokofiev restored the score in 1923, adding numerous revisions to the composition. The changes were incisive enough for the composer almost to consider it his fourth piano concerto (he had completed and premiered his Piano Concerto No.3 in C major, op.26, in 1921). The new version premiered in 1924, again with the composer playing, this time with Serge Koussevitzky (1874 – 1951) conducting.

Initial Performances

This concerto is one of the most demanding pieces in the entire piano literature. Its first premiere must have appeared so revolutionary and provocative that according to some accounts, the audience left the hall in protest. We have no recording that documents how the concerto sounded at that event, nor is there any recording (i.e., direct, first-hand information) of the composer interpreting his op.16 altogether. All we have of Prokofiev as pianist is a recording with him playing his Piano Concerto No.3, op.26. In 1935, he also recorded some of his solo piano music.

Alterations in Perception over the Years?

In other words: all we have is what performing history made of this concerto over the past 90 years. That’s a time long enough for distortions and mis-perceptions to creep in. Many may associate Prokofiev with the “brutalist” aspects of compositions from his later years. These works also reflected what Stalinism and the Second World War did to people. Prokofiev’s “war sonatas” from 1939 – 1944 (the Piano Sonata No.6, op.82, Sonata No.7, op.83, and Sonata No.8, op.84) are quite popular in concerts today. Of course, it would be incorrect to “back-project” the character of these sonatas onto the concerto op.16. Further, the composition of op.16 pre-dates the revolution in 1917, as well as the First World War.

Back-Projections in Style?

What is legitimate, though, is, to have a look at other compositions by Prokofiev from that same time, such as his second piano sonata in d minor, op.14. Here, we find both the very virtuosic aspects (Prokofiev must have been at the height of his pianistic abilities in those years), along with lyrical passages. And of course, one can find Prokofiev’s personal (piano) style in most these compositions. Though, what we now see as his personal style in early compositions can also be an illegitimate back-projection from later works. It’s a tricky topic!

Finally, the fact that the composition is dedicated to his late friend who committed suicide in 1913 one can hardly conclude that op.16 should be dominated by sad, mournful moods. That suicide occurred after the first version of the composition was completed. He might have selected a “fitting” composition to honor the tragic death of his colleague. However, I believe that his tribute consisted in dedicating the work that he regarded his most valuable / precious at the time, and I’m sure this could only be this concerto.

The Movements:

The concerto has four movements of very uneven duration: the outer movements are about 11 minutes each, the second movement, a Scherzo, is very short (around 2 minutes), movement #3 (Intermezzo) is around 6 minutes:

I. Andantino — Allegretto — Poco meno mosso — Tempo I

The first movement opens quietly, with the strings con sordino and pizzicato. The piano follows in bar #3, initially p, the right hand octaves are annotated narrante (telling), then very lightly rising the volume (pochissimo cresc.) up to f. At [2] in the score, a solo, caloroso con gran espressione (warm, with great expression), starts a section, over which the music is gradually escalating, both in intensity, as well as in the density and complexity of the piano part, up to the Allegretto (at [7] in the score).

The Allegretto introduces new material, and even with the accompaniment of the orchestra, the solo part is clearly dominating. A 1-bar poco ritardando leads to section [17], which is labeled Poco meno mosso: merely a transition section into the development part at [20]. This is a giant solo cadenza of horrific difficulty, taking up almost 40% of the entire movement (over 4 minutes). Parts of the cadenza are written on three systems, requiring the pianist to play huge jumps. Even just physically this is extremely demanding. The cadenza starts with the initial theme, so is essentially to be taken as Andantino (even though that’s not labeled in the score). The annotations in the cadenza give an indication about its nature:

dolce — espressivo — (crescendo up to fff) — precipitato — pesante — con affetto — colossale — accelerando — con tutta forza

The last 3 bars of the cadenza are accelerando. But then, at [21], the orchestra joins in again, and the final part bears the annotation Tempo I (Andantino). The movement ends with a pp reminiscence / quote from the very beginning, with the narrante section in the solo.

II. Scherzo: Vivace

The Andantino leaves the listener with the expectation of a slow movement following. But that second movement is not slow, but a strenuous Scherzo that offers no relief for the pianist: at a very fast pace (Vivace, in quarter notes) and without any break, the soloist is required to play continuous, relentless, semiquavers with both hands in octave parallels, throughout the entire movement. Just the very last note is a quaver. This requires the utmost concentration in order to achieve perfectly coordination, both between the hands, as well as between soloist and orchestra.

III. Intermezzo: Allegro moderato

Again: not a slow movement! It’s an Allegro moderato, initially dominated by a stomping 4/4 beat, annotated ff and pesante (heavy). In a way, an anticipation of works from his later period. I can’t resist thinking of music to Sergei Eisenstein’s films when I hear the orchestral introduction. The movement turns more moody and ironic when the piano enters the scene.

IV. Finale: Allegro tempestoso — Meno mosso — Più mosso (Allegro) — Meno mosso (Moderato) — Allegro tempestoso

The annotation Allegro tempestoso says it all: a stormy / storming Allegro, very fast, yet rhythmically very intricate. In a way, it resumes the tone of the Allegretto part in the first movement, but with substantially more drive / tempo. With the Meno mosso, the musical flow suddenly stops and gives way to a more vocal motif that the orchestra introduces. The piano takes this up in modified form and evolves it in a longer solo. The orchestra then re-joins for a dialog. Here, the piano part gradually gets more vivid, up to the Più mosso (Allegro), where orchestra and solo go through a climax. After this, the tempo slows down to the Meno mosso (Moderato). The music comes to an abrupt halt, and this movement’s solo cadenza follows.

That cadenza starts with an impressive & interesting sequence of dissonant chords, growing into a fermata. A p restart then gradually softens the harmonies but rapidly builds up in several waves, to towering, extremely virtuosic culminations. Then, the orchestra joins in again. A ritardando and diminuendo down to pp is leading into a sudden return of the concluding Allegro tempestoso.

Metronome Values?

The score does not include metronome annotations. There is only a statement “Duration: 23 minutes“. That’s value which appears hardly achievable, based on the two recordings discussed here.

The Interpretations, Overview:

In order to provide a rating overview, as well as an idea about tempo relations both within an interpretation, as well as between the two recordings, I have prepared the little table below. The color coding for the tempo (blue = slower, green = faster) refers to the average between the recordings (as mentioned, there are no metronome numbers in the score):Prokofiev: Piano concerto No.2 op.16: M.M./rating comparison table

The Interpretations, Detail:

My own comparison did not reach quite the same conclusions as “La Tribune des critiques de disques“. I think that the experts dismissed Yuja Wang’s interpretation / recording  too soon. Plus, the reviewers were “locked in” on the traditional interpretation style for this composition. Based on the given excerpts, I agree with the reviewers that the four additional recordings in their comparison (Evgeny Kissin / Vladimir Ashkenazy, Horacio Gutierrez / Klaus Tennstedt, Anna Vinnitskaya / Gilbert Varga, Jean-Efflam Bavouzet / Gianandrea Noseda) deserve a rating below that of Yundi Li. The experts dismissed Kissin in round #1, based on the cadenza in the first movement (along with Yuja Wang, see above), next in turn (after the Scherzo) was Bavouzet, Gutierrez fell out after the Intermezzo.

Yundi Li, Seiji Ozawa

Prokofiev: Piano concerto #2; Ravel: Piano concerto in G — Li/Ozawa; CD coverProkofiev: Piano concerto Nr.2 in g minor, op.16;
Ravel: Piano concerto in G major

Yundi Li, Seiji Ozawa, Berliner Philharmoniker

DG 477 6593 (CD, stereo); ℗ / © 2007
Booklet: 22 pp. e/d/f
Prokofiev: Piano concerto #2; Ravel: Piano concerto in G — Li/Ozawa; CD, UPC-A barcode
—Find CD(s) on—

spacerThe Artists

In the International Chopin Competition in Warsaw 2000, the Chinese pianist Yundi Li (*1982) won the first prize, at the age of 18. This was after an intermission of 15 years, as no first prize was awarded in 1990 and 1995. So far, he has mainly recorded works by Chopin and Liszt. The one notable exception is the above recording with piano concerti by Prokofiev and Ravel. Up to and including this recording, Yundi Li worked with Deutsche Grammophon. After that, he signed a contract with EMI Classics, with the goal to record Chopin’s complete oeuvre for solo piano. Yundi Li was born in Chongqing and began piano studies at the age of seven. He subsequently received training from the Shenzhen Arts School and later from the Hochschule für Musik und Theater in Hannover, Germany.

The above CD is made from live recordings during concerts with the Berliner Philharmoniker, directed by the Japanese maestro Seiji Ozawa (*1935). The recording was made in the Berliner Philharmonie, the orchestra’s own concert hall. One definitive advantage of this setting is not just the orchestra and the excellent conductor, but (probably at least as important) the recording team’s familiarity with the location and the acoustics.

Notes on the Interpretation

The piano sound is excellent, the orchestra recording differentiated, detailed and transparent, yet well-balanced and coherent.

I. Andantino

Duration: 11’12”

  • Introduction, up to [2]: For an Andantino, the tempo feels rather heavy. Also, one notices a tendency of the soloist to exaggerate the dynamics (this appears to persist throughout the interpretation): f almost by principle is already ff, and p often is mf. Quite obviously, the artist is aiming for a forceful, “brutalist” type of interpretation. To me, so much “steam” already at the beginning gives away the chances for further development.
  • caloroso con gran espressione etc., up to [7]: see also my comments for Yuja Wang below. The p at [5] fairly dense and rather mf.
  • Allegretto, up to [17]: Prior to [7] (Allegretto) the orchestra makes a ritardando (not in the score), the Allegretto itself is faster than with Wang/Dudamel, but with an unrelated tempo. Strangely, four bars later (con eleganza), Yundi Li starts another, slightly faster tempo: I feel pushed. None of these tempo alterations / relations make sense to me. Yundi Li’s interpretation is of utter conciseness, very strongly articulated and very virtuosic. But it’s mostly a level too loud and forceful, I think.
  • Poco meno mosso, up to [20]: The ritardando to this section is not “poco“, but rather dramatic (in line with the rather dramatic character of this interpretation!), and similarly, [17] to me feels Molto meno mosso. I can only guess that the artists felt the need for a stark contrast, given the otherwise forceful nature of their interpretation in this movement.

Development Part to End

  • Development part / cadenza, up to [21]: yes, it’s technically excellent, astounding, and certainly colossale / pesante / con tutta forza, to a degree that may make the listener feel overwhelmed. It’s impressive, no doubt, and an entirely different (maybe more traditional) approach than Yuja Wang’s. Many (most) people will be very impressed / pleased with this never-ending climax. I think that a little more differentiation (rather than the permanent fff that makes the listener feel run over by a giant machine) might have made it even better.
  • Final part, through [24]: The bars that follow the cadenza are ff in the orchestra, not fff. Ozawa probably wanted to make this sound as colossale as most of the cadenza!

II. Scherzo: Vivace

Duration: 2’17”
Yundi Li won this part of the radio comparison (see above) because, as the experts stated, “who can do it fastest wins”. This wasn’t the only argument they used. Nevertheless, I claim that it’s not as simple as that. Yes, this was the fastest version among the four that were left in the radio discussion, and Yundi Li’s technical skills are astounding, but

  • it takes the artists about 8 bars to achieve coordination between the orchestra and the piano, and there are a couple missed keys (mostly high peak notes in the first part). Maybe the tempo chosen was a little too fast? After all, the annotation is Vivace, not Presto!
  • the dynamics are compressed. Mostly it’s played f and ff, many of the p are mf at least;
  • yes, the movement ought to have Scherzo character, and it ought to be fast, as fast as possible. But apart from the coordination issues at the beginning, I feel pushed constantly by the soloist, preventing me from enjoying the music;
  • from the point-of-view of recording technique, the piano is too much in foreground, the orchestra is essentially covered by the soloist and remains dull (in the listener’s impression).

III. Intermezzo: Allegro moderato

Duration: 5’41”
This is considerably faster than with Wang / Dudamel. Too fast, in my opinion. The movement is mostly in 4/4 time (with some 3/4 and 2/4 interjections), but here I hear 2/2. Even though it is fast, it sounds heavier and slower, just like 2/2 time with 1/2 = 56. The recording puts the piano into the center again, and with Yundi Li’s tendency to play rather (too) loud, the orchestral accompaniment lacks differentiation, sometimes hardly audible. Is the orchestra really always meant to be accompaniment only? Also, with the way in which Yundi Li plays this movement, I fail to see the irony in the music, as well as the interplay with the orchestra.

IV. Finale: Allegro tempestoso

Duration: 10’36”

  • Allegro tempestoso, up to [97]: Yundi Li appears to point out the similarities with the fast part of the first movement, and maybe with the preceding Intermezzo. Here, I see the tempestoso mainly in the (very virtuosic) machine-like / “industrial” playing. These attributes are not meant to be derogatory.
  • Meno mosso, up to [108]: In the solo ([100]), the solo appears to lose momentum. Certainly, the slow-down is intended for the following build-up to be more dramatic. However, to me, it’s a bit too obvious, as an effect.
  • Più mosso (Allegro), up to [117]: Played well, technically clean. But all a bit (if not a loot) too loud. I don’t hear a real p, let alone pp.
  • Meno mosso (Moderato), up to [127]: Yundi Li doesn’t skip any opportunity to “enhance” a p or mf to f, and f to ff. This is in the spirit of how he started this movement.
  • Allegro tempestoso: (see above)


Overall Duration: 30’22” (including applause)
Rating (see above for details): 4.0 — If you are looking for a traditional interpretation (presumably in the spirit of Jorge Bolet’s reference recording from 1953), this may be your interpretation: a very good interpretation (certainly technically). But you may be missing out on aspects that Yuja Wang brings to light…

Yuja Wang, Gustavo Dudamel

Rachmaninoff: Piano concerto #3; Prokofiev: Piano concerto #2 — Wang/Dudamel; CD coverRachmaninoff: Piano concerto Nr.3 in d minor, op.30;
Prokofiev: Piano concerto Nr.2 in g minor, op.16

Yuja Wang, Gustavo Dudamel, Simón Bolívar Symphony Orchestra of Venezuela

DG 479 1304 (CD, stereo); ℗ / © 2013
Booklet: 22 pp. e/d/f/Spanish
Rachmaninoff: Piano concerto #3; Prokofiev: Piano concerto #2 — Wang/Dudamel; CD, UPC-A barcode
—Find CD(s) on—


The Artists

Yuja Wang (*1987) doesn’t require an introduction. She is running the world on concert tours, and she is omnipresent on the on-line channels such as YouTube. That’s not just because of her (I think) refreshing (and sometimes daring) fashion statements. It’s also also (primarily, I believe) because the stupendous agility of her fingers. Yuja was born in Beijing. She started playing the piano at the age of 6 and spent 3 years at Beijing’s Central Conservatory of Music, where people quickly recognized her talent.

At age 15, she started a 5-year study period with Gary Graffman (*1928) at the Curtis Instititute of Music in Philadelphia, where she graduated in 2008. By the time of her graduation, she had already won several prizes, such as the 1st prize at the Aspen Music Festival‘s concerto competition in 2002. In 2003 she made her European debut with the Tonhalle Orchestra in Zurich, under David Zinman. Her North American debut was in 2005/6 in Ottawa. There, she replaced Radu Lupu in a concert with Pinchas Zukerman conducting. Her breakthrough was in 2007 in Boston, where she replaced Martha Argerich in four subscription concerts with the Boston Symphony Orchestra under Charles Dutoit. Yuja Wang now lives in New York. For the season 2014/15, she was artist in residence at the Tonhalle in Zurich. So, she has returned to the place where she made her European debut.

Recording, Orchestra

The above CD was recorded live at the Sala Simón Bolívar, Centro de Acción Social por la Musica in Caracas, Venezuela. The Simón Bolívar Symphony Orchestra of Venezuela is directed by one of the young shooting stars among the conductors, Gustavo Dudamel (*1981), who is also making a very fast & splendid career. He now is Music director of the Los Angeles Philharmonic, after having conducted the Gothenburg Symphony Orchestra between 2006 and 2012.

Notes on the Interpretation

I. Andantino

Duration: 11’02”

  • Introduction, up to [2]: This interpretation starts with a more fluent tempo, more appropriate for an Andantino (lighter than Andante!), I think.
  • caloroso con gran espressione etc., up to [7]: This reveals the specific qualities of Yuja’s interpretation. She is not primarily forceful, let alone brutalist, but listening into the piano part. She keeps it transparent and allows for the lighter, lyrical aspects to come through. One example: between [2] and [3], Yundi Li focuses on the melody in the right hand, one can barely hear the chords in the same hand. With Yuja, that section is very balanced and transparent.
  • Allegretto, up to [17]: An excellent idea: at [7], the tempo is exactly doubled, the listener does not feel a change in the beat, yet all of a sudden senses the spirit of an Allegretto.
  • Poco meno mosso, up to [20]: This is even slower than Yundi Li’s interpretation. And yet, it doesn’t feel this way. It is wonderfully light and feels “suspended”: an excellent transition to the cadenza that follows (rather than an intermission, as with Yundi Li!).

Development part to End

  • Development part / cadenza, up to [21]: This starts at pretty exactly the initial tempo (Andantino). This is the segment that failed in the “radio competition”. Parts of this may be caused by the sound management, which may have attempted to give a “realistic view” into the concert hall, putting the piano less in the foreground than in Yundi Li’s interpretation. And yes, the colossale part sounds less forceful, less colossal than with the contender, but unlike the radio judges, I don’t see this as technically insufficient. Yuja may use (and have?) less physical power, but she does also observe the lyrical aspects (dolce / espressivo / con affetto) in which “keyboard thunderers” just storm through! It’s perhaps a flaw of a live concert recording that sound management may have made this sound softer than it really was?
  • Final part, through [24]: An indication for the suspicion just mentioned is that at the beginning of the final part, the orchestra almost covers the piano. That’s realistic, probably, but perhaps also to Yuja’s disadvantage in a direct comparison? At the same time, Dudamel adjusts the force applied by the orchestra to Yuja Wang’s interpretation. The ff in this part is less overwhelming than with Seiji Ozawa. After all, this is not the end of the concerto!

II. Scherzo: Vivace

Duration: 2’22”
The tempo here is a tad slower thanYundi Li’s (1/4 = 168 vs. 174), but the result is equally astounding, in my opinion even better:

  • I can’t make out coordination issues, nor missed keys in the entire piece. Yuja’s playing is controlled throughout (without ever being boring), yet retains the Scherzo character (lighter and more subtle than Yundi Li’s, though)
  • dynamically, Yuja Wang’s playing is vastly more differentiated;
  • one could almost say that the pace is relaxed. Certainly, I don’t feel pushed all the time, like in Yundi’s interpretation;
  • in this movement at least, the acoustic balance is better, the orchestra sound is more detailed, the overall recording more transparent.

III. Intermezzo: Allegro moderato

Duration: 6’35”
Here, the movement is played properly in quarters. It’s slower, but with four beats per bar. Therefore, with 1/4 = 92 this is still Allegro moderato and pesante. Yuja Wang brings out the irony in this movement. She allows for p and intermediate tones, even lyrical passages, and she does not try to overpower the orchestra. To the contrary, she maintains a dialog with the orchestra. For example, at [69], the woodwinds remain partners with the soloist, one can follow them properly. In my view, it’s an excellent interpretation on both parts, the orchestra and the soloist. Maybe some people prefer a more uniform interpretation, such as Yundi Li’s. But to me, this provides a richer, more multi-faceted listening experience.

IV. Finale: Allegro tempestoso

Duration: 10’28”
In all parts,Yuja Wang plays faster than Yundi Li. If “faster” means “better” (as one of the radio critics stated), doesn’t that make her the winner here??

  • Allegro tempestoso, up to [97]: Not just faster than Yundi Li, but with more agility, less brutal, less “black & white”, i.e., more differentiated. The recording is far more balanced, with the orchestra and the solo piano being partners talking at the same level.
  • Meno mosso, up to [108]: Yuja’s solo is much more differentiated than Yundi Li’s, with decent agogics, keeping an eye on the polyphony in this section. Yuja also plays with very differentiated dynamics. The following dialog with the orchestra is very subtle initially, the build-up impressive and harmonic. One may miss the “brutalist” aspect in the latter part, but technically, the playing is better than Yundi Li’s.
  • Più mosso (Allegro), up to [117]: Faster also here, and definitely more accurate in the dynamics. The chord sequences after [115] may be softer in the articulation (not as much hammered as Yundi Li’s). But, after all, these are p and pp!
  • Meno mosso (Moderato), up to [127]: Once again, more differentiated than Yundi Li. Yuja Wang is properly observing the dynamic annotations, allowing for lyrical, even calm moments. The second, virtuosic part is brilliant! And the orchestra remains clear and audible, transparent, is not covered by the solo part.
  • Allegro tempestoso: The recapitulation / reprise of the first section is even faster than the beginning. Astounding & brilliant, the frenetic applause is justified!


Overall Duration: 30’56” (including applause)
Rating (see above for details): 4.8To me, clearly the better interpretation overall. I don’t care about the “Jorge Bolet standard / reference”, but how much the artists are reading from the score. If you prefer the “brutalist approach”, though, this may not be the interpretation for you…


There is no “one size fits all” in this (or any?) piece. If you like this concerto to sound like Prokofiev’s “war sonata” / post-war compositions, then Yundi Li’s recording is a solid choice. However, if you are looking for a more differentiated interpretation that also addresses the lyrical aspects, Yuja Wang’s recording (in my opinion) offers far more. The recording / sound engineering with Yundi Li / Ozawa favors the piano. This is at the expense of the audibility & clarity of the orchestra. With Yuja Wang / Dudamel, the sound engineers were taking a more balanced and transparent approach. The result is probably also closer to the “live experience” in a concert hall. A minor disadvantage may be that the “brutalist” / harsh aspects of Prokofiev’s music are softened. But this actually fits Yuja Wang’s more differentiated playing. My personal preference clearly goes towards Yuja Wang / Gustavo Dudamel.

A Comment on “Blind” Comparisons

Some words of caution about the above-mentioned radio comparison “La Tribune des critiques de disques” and the dismissal of Yuja Wang’s interpretation (the CD below) in that contest:

  • This is meant to be a blind comparison. The experts are supposed not to know which interpretation they are talking about. They should not know anything about the selection of recordings for the comparison. Occasionally, there is a “CD du jour”, but not in that case.
  • However, the comparison almost always includes one or several recent recordings that are currently in the news, promoted by music industry. The three experts are experienced, seasoned critics. So, they must have known (about) Yuja Wang’s interpretation, as well as probably Yundi Li’s. Yuja’s performance is so much different from all or most other interpretations on the market (it certainly struck me as being very different). Therefore, they (some) must have recognized it instantly. This may not apply to Yundi Li’s recording.
  • With music, there is hardly such a thing as objectivity & neutrality. The fact that the three “judges” were experienced critics does not guarantee that their judgement is neutral: they are all human. Despite the fact that they tend to rely upon musicological arguments, they still have their personal preferences and possible bias from what is “en vogue” or traditional perception. Plus, they may shy away from recommending something that the majority of the listeners may disagree with.

Potential Flaws

  • One flaw in such comparisons is that in the first round, two or three of the six recordings are eliminated, based on the pre-selected excerpt. That segment may appear particularly unfavorable for one or several of the interpretations. These won’t get a second chance for a more balanced view. In the case referred to here, the cadenza and the ending of the first movement were played. It is of course speculative whether a different excerpt would have changed the outcome.
  • The critics are not “young rebels”, so there is likely some tendency to favor traditional style recordings. One can gather this from their personal recommendations for “other recordings that were not part of the comparison” in the closing remarks. That said, they are certainly aware of HIP recordings. For music up to the early 19th century, these stand a fair, if not excellent chance of winning such comparisons.
  • In favor of the experts in such comparisons one should keep in mind that they typically have one single “shot” at each excerpt. A person listening to CDs at home can repeat a comparison as many times as he/she likes. Quite often, repeat passes reveal additional details that go unnoticed in the first auditi0n.


Such radio comparisons ought to be taken with a grain of salt. Generally speaking, “La Tribune des critiques de disques” is the best one that I have heard so far. Of course, it is perfectly OK, even desirable, to have independent personal preferences. In that sense, I have no problem with people rejecting a recording even where I strongly disagree. However, in such radio discussions, dismissals deserve a fair argument & justification. Worse, upon dismissing Yuja Wang, the experts could not resist making a remark about her shoes. A remark that was not justified: after all, what she is wearing in a concert has nothing to do with her interpretation. On that particular CD, Yuja depicts herself as stylish, feminine, i.e., she is not wearing one of her sexy, provocative outfits.

Follow-up: Feedback from a Reader

In response to the above notes on blind comparisons, a reader submitted the following comment:

A typically thoughtful and detailed review. Particularly valuable is the analysis of why ‘blind’ comparisons can be flawed, the inherent biases of experienced reviewers, and the fallacy of judging the whole from the parts. To me, the use of excerpts and extending that judgement to the entire performance is the least tenable. The impact on any excerpt depends greatly on the relationship of that section to the whole. To be able to react to, or describe, in an ideal form, the performance of a section of a composition does not enable one to accurately describe the whole. If a scientific laboratory, the isolation of genetic fragments and their proper sequencing, while valuable, does not by itself ensure that the entire research study will be fruitful.

My response to this added some additional ideas on the above points:


It might help a little if the sequence of the recordings was randomized / unknown. Even the discussion leader should not know about the identity of the recordings. Ideally, a third party or some software could pick, say, 6 out of a larger selection of 10 – 12 recordings. But even then, experts may recognize some recordings and jump to conclusions.

As stated, I still think that these emissions in general have value to the listener (especially as long as one keeps in mind the above limitations). The one in France Musique still was the best one I know. Unfortunately, they have just switched to a new scheme. They shortened the overall duration to 90 minutes, and they use a larger pool of experts. Plus, they play very short excerpts only, and in the argumentation, “out of the stomach” arguments dominate. It’s not an enhancement at all. But I still prefer it over the one on BBC R3, which pretends to cover / include all available recordings of a given piece.

At least some instances of the Swiss equivalent (Diskothek, on SRF Kultur) are worse than the other two. Here, there are only two experts, sometimes with questionable qualification for a given work. Also, they play rather short excerpts, and only too frequently there is an active bias from the discussion leader.

It can’t hurt if the discussion leader is a knowledgeable person (e.g., a musicologist or a musician). However, especially given that he/she knows what is being played, the discussion leader should not exert too much — if any — guidance, other than keeping the schedule. The leader should certainly not instruct the experts what to look for.

Also, such comparisons are not the right place for lengthy musicological presentations, neither by the experts, nor by the discussion leader.

Addendum 1:

Prokofiev’s Piano Concerto No.2 in g minor, op.16 is available as study score (10.25″ x 7.25″ / 26 x 18.4 cm) —Find score on—

Addendum 2:

The opening concerts for the 2014/15 season at the Tonhalle in Zurich featured Prokofiev’s Piano Concerto No.2 in g minor, op.16. Yuja Wang played with the Tonhalle-Orchestra Zurich and its new conductor, Lionel Bringuier. See my concert review in the posting “Bringuier & Wang in Zurich, 2014-09-11“.

Share:Share on Facebook0Tweet about this on TwitterShare on Tumblr0Share on Google+0Share on LinkedIn0Share on StumbleUpon0Pin on Pinterest0Digg thisShare on Reddit0Email this to someone

3 thoughts on “Prokofiev: Piano Concerto No.2, op.16

Leave a Reply

Translate »
%d bloggers like this: