Several complete or nearly complete recordings of Mozart’s piano concertos are listed as available—though whether you will be able to find them at your local record store, or even order them, is another question. What you are more likely to find are CD reissues of famous old Mozart performances: performances by Schnabel, Casadesus, Lipatti, Michelangeli, Clara Haskil, Myra Hess, Clifford Curzon, and others, none of them much improved by digital remastering. The first to issue a complete Mozart set was Alfred Brendel, whose ten CDs span the transition from the analogue to the digital eras (1970–1984). He has been followed by Murray Perahia, Vladimir Ashkenazy, and, most recently, Malcolm Bilson.1

The Bilson series has stimulated occasional gusts of interest ever since the first disc, featuring Concerto No. 9 in E flat, K. 271 (“Jeunehomme”), was issued in 1983. The impending completion of the series allows for a comprehensive appraisal of Bilson’s enterprise. As a Mozart player Bilson stands up very well, I believe, not only to his more renowned contemporaries, but also to the venerable and revered remastered masters. His achievement has a good deal to tell us about the current scene in musical performance, as I shall try to suggest later.

Most of Mozart’s thirty-odd piano concertos and concerto rondos were written in the 1780s, the last decade of his life, after he had broken away from his traditional position as provincial court servant and set himself up as a freelance composer and performer in Josephine Vienna. Brillant successes were matched by depressing failures; his final success—Die Zauberflöte—came just too late and he died young, of course, in straitened circumstances. Among his earlier successes were a run of Akademien, or subscription concerts, which he was able to set up, most of them featuring concertos.

With these concertos, Mozart attempted to make the scene in music’s capital city. On one level, the inner drama of concerto relationships invites a metaphorical reading as one person’s effort to gain acceptance from the group. On another level, the concerto is the genre which more than any other (even opera) was fine-tuned by its composer to dazzle and delight his contemporaries.

That the Mostly Mozart audiences of today are pleased by exactly what the composer wrote to please his own paying public must count as a mysterious circumstance, possibly even an ominous one. Our special empathy for these works is much less easy to understand than the lack of interest in them during the nineteenth century. Only one concerto survived in the repertory for very long after Mozart’s death, the turbulent Concerto No. 20 in D minor, K. 466, and it has had to pay a price for its solitariness, as Charles Rosen observed:

The D minor Concerto is almost as much myth as work of art: when listening to it, as to Beethoven’s Fifth Symphony, it is difficult at times to say whether we are hearing the work or its reputation, our collective image of it.2

The canonization of Mozart’s other concertos is attributed primarily to the activity of two great pianists, Artur Schnabel and Edwin Fischer, in the 1930s and 1940s. Connoisseurs have been known to prefer the interpretations of Fischer, who specialized in Bach and Mozart—Schnabel favored Beethoven and Schubert—and who actually founded his own chamber orchestra to present the concertos in a “historical” fashion. But Schnabel was the more important international figure, and his recordings have never been allowed to remain out of circulation for very long. A new CD series, Mozart and Schnabel, transmits six of his concerto recordings, along with his famous performance of the Rondo in A minor, K. 511, and several sonatas.

A CD changer with repeat function makes it a simple matter to compare Schnabel and Bilson in, let us say, the opening piano solos in movements of the D minor Concerto. (This kind of comparison doesn’t speak to the nature or quality of their interpretations as a whole, of course, but at least it speaks with some specificity.) In the first movement, the piano enters some time after the orchestra, as always happens in classical concertos; here—and this happens rather rarely—the piano enters with a completely new, notably expressive theme. Schnabel does a number of little things with tempo and dynamics in order to make this entry theme into something assertive and particularly characteristic.

There are six tiny phrases—two plus two plus two—in the entry theme (see score on facing page). Schnabel is renowned for his impeccable molding of melodic lines; by speeding up the tempo a little at the start, he tints the line with urgency and disquiet. Playing phrases 1–3 more and more intensely and freely, with increasing volume and rubato, results in a precipitous emotional climax at the end of phrase 3. But then Schnabel is more or less forced to play phrase 4—the highest so far—more quietly, and it trails away.


The solo persona projected by Bilson seems to be thinking less about the presence it is going to establish than about the theme it has just heard in the orchestra (as well it might: it is an astonishingly tormented theme). Slowing the tempo slightly, Bilson plays phrases 1–2 and 3–4 in a straightforward soft/loud (piano/mezzo forte) pattern. Besides reflecting the inherent paired structure of the phrases, and their register, which is low/high in each case, this brings the music out at a mezzo forte so that phrases 5–6, which are the highest of all, can soar beautifully. Notice how Bilson glides through the grace note in the last measure. Certainly he underplays phrase 3—and gains a quiet pinch of pathos without having to work for it. Distinguish this from Schnabel’s underplaying of phrase 4, which has to be called sentimental.

All the phrases are linked by repetitions of a single four-note figure in the left hand. Bilson (like most other pianists) plays these calmly and consistently—a reasonable reading, since the very neutrality of the figure serves to anchor the wide-ranging, emotional span of the right-hand melody. By accenting the figure in various different ways, Schnabel destroys this anchor; his concept of this entire entry passage as basically disruptive is easier to accept than the arbitrary measures he takes to achieve the effect.

But if Schnabel is more idiosyncratic and Bilson more restrained in the first-movement theme, attitudes are reversed in the finale. Within the angry, impacted rondo theme a number of little fragments repeat themselves almost obsessively, and Schnabel makes no effort to distinguish these one from another. However stormy, his playing seems ultimately bland compared to Bilson’s, where the entire theme mounts in intensity, fragment by fragment, driving from a percussive trajectory at the beginning to a furious climax at the very end.

On paper, this may seem like a very “unclassical” strategy for a Mozart theme. But it sounds thrilling, and it suits both the aberrant theme itself and also the deeply unclassical movement which that theme launches. Usually, once a rondo theme has been played by the soloist, the orchestra repeats it and so appropriates it; not so in the D minor Concerto. After the first offer of the piano theme, the orchestra can’t get even the music right for more than three measures, and when it attempts to develop the theme rather than repeat it, the orchestra simply gets lost. The second offer of the theme throws it into a virtual panic. After the last offer, which now begins to sound uncharacteristically like a challenge, the orchestra declines to deal with the initial theme at all—it literally changes the subject. Bilson’s explosive interpretation helps to rationalize the orchestra’s utter failure to cope, both technically and emotionally.

Exhausting analyses of this kind can be carried out with other discs slipped into the CD changer; and the comparison is not always so clearly to Bilson’s advantage. The pendant to the D minor Concerto, Concerto No. 21 in C, K. 467, is the vehicle for some of Schnabel’s finest playing. Bilson’s performance may be preferable as a whole—it is certainly less erratic—but he does not quite capture the verve with which Schnabel projects the opening piano entry in the first movement, or the piano entry after the main theme in the finale, or many of the bravura passages. Also in the finale Schnabel finds a perfect combination of wariness and charm for the delightful passage of mock uncertainty that starts the development section. Phrases in the celebrated slow movement are molded with a pensive grace that is almost unbelievable, especially in view of the slow tempo (more than half again as slow as Bilson).

Bilson does, however, have an in-built advantage in playing Mozart concertos on an eighteenth-century instrument—the so-called “Mozart piano,” or fortepiano—together with an orchestra of period or period-style instruments. The fortepiano he uses is a beautiful reproduction by the American maker Philip Belt, and the orchestra is conducted, very well indeed, by John Eliot Gardiner.

The Bilson-Gardiner enterprise represents, in fact, another major foray by the early-music movement into the later reaches of the canonic concert repertory: first Bach and Handel, now Mozart and Beethoven. Moreover, Roger Norrington has already conducted period-instrument concerts monstres of music by Berlioz, and Bilson will soon record the Schumann Piano Concerto on a piano of 1838. Small wonder that the old dispute among musicians about so-called “historically authentic” performance has flared up once again.

The action may be viewed at a safe distance in an anthology called Authenticity in Early Music, edited by Nicholas Kenyon.3 Not much of intellectual substance remains to the idea of “authentic” performance after Richard Taruskin, in an explosive essay, has challenged its historical plausibility, denied its desirability, and mocked the pretensions implicit in the loaded term itself. His own term, “authentistic,” is handily backloaded, “authentistic being to authentic as Hellenistic was to Hellenic.” At times Taruskin writes as though the whole early-music movement were a scam in aid of the recording companies, similar to the hoax that Joseph Horowitz claimed the mass-culture industry perpetrated around “authentic” versions of symphonic classics by Toscanini.4 This is a lot less convincing than Taruskin’s argument that the true ideological basis of “authentistic” performance is not really historical reconstruction at all.


Is not—or was not. For some time after World War I, it is certainly true that both historical performance and “traditional” performance tended toward the impersonal, the objective, and what T.E. Hulme called the “geometrical,” as opposed to the “vitalistic” in nineteenth-century art. Indeed, the original impetus behind the early-music movement was the revulsion against romantic emotionality endemic to early-twentieth-century modernism, the modernism of Pound, Eliot, Hulme (who admired “the dry hardness which you get in the classics”), Ortega, and Stravinsky. Historicism became important only later. Stravinsky is a key figure, with his astringent music of the neoclassical period, his influential polemics against musical expression in the Harvard lectures The Poetics of Music, and his likewise influential piano performances (as in an almost unbelievably mechanical Mozart recording of 1938).

A common ideal animated both historical and traditional performers, or at any rate many traditional performers, not to speak of poets and composers. The ideal was to find poetry in geometry. And while it is not entirely clear where musical performance is supposed to be right now, fifty years after Stravinsky’s days as a pianist, and about thirty after Eliot’s as a culture hero, an authentically modern style appears to have surfaced: “authenticism” sans cant and stripped of geometry. Taruskin mentions a late recording by Gustav Leonhardt with approval:

What a lilt due attention to meter can impart!… Leonhardt’s recording also demonstrates the joyful results of thoroughly passionate and committed experiment with old instruments. His players have truly understood what I see as the inestimable and indispensable heuristic value of the old instruments in freeing minds and hands to experience old music newly.5

What, then, does experiment with old instruments free us to hear newly in the Mozart concertos?

Mainly, and most simply, we hear new proportions. The use of old instruments tips the balance between soloist and orchestra that grounds Mozart’s concerto aesthetic, and thus sharpens up the inherent contrast between them. Paradoxically, although the fortepiano is quieter than the modern piano, it comes out ahead in the relationship between solo and orchestra.

The relationship depends less than one might think on volume of sound and more on sound quality. In spite of its thin tone and low volume, the fortepiano “speaks” more clearly than a modern piano because of its characteristic attack—a sharper attack than a modern Steinway, not to speak of a Bösendorfer or a Baldwin. Early orchestral instruments, on the other hand, attack more softly than modern ones (the timpani excepted). Donald Tovey, whose pioneering accolade to the Mozart concertos is still well worth reading, made much of the way the soloist’s agility and expressiveness is pitted against the orchestra’s sheer mass and range of instrumental color. With old instruments, distinctness of voice is added to the soloist’s side of this calculus.

The orchestra used by John Eliot Gardiner in the Bilson recordings is much smaller than a modern symphony orchestra, even a reduced modern orchestra. (In some performances of Mozart’s time the orchestral strings were cut down to one on a part when the piano was playing. This might not have been necessary at performances in theaters, where the piano was presumably on stage and the orchestra in the pit.)6 Even so, there are places in nearly every concerto where the piano writing is so thin one has to strain to hear the soloist. On the other hand, where the piano writing is full the fortepiano becomes a real contender. It does this most vividly in the fairly long bravura passages, such as those at the end of the exposition and recapitulation sections of the first movements.

The fortepiano goes all out in these passages; the fragile instrument fairly rattles with exertion. Scales, octaves, and such in the low register cut through with special vigor. A Mozart concerto is and should be played as a virtuoso vehicle, not just as an elegant dialogue, and in a fortepiano performance the sweat is all but palpable, up to and including that notorious concerto trademark, the cadential trill. On a modern piano, this long trill can sound blank and merely formal, a gratuitous pirouette appended to fast runs which the soloist has negotiated with suspicious ease. On the fortepiano the trill sounds like the last triumphant gasp of an athlete completing a particularly exhilarating turn. When the orchestra roars its approval—a very loud orchestral ritornello follows the trill—we can only endorse the applause: the soloist has earned it.

One should certainly acknowledge the skill with which modern pianists have learned to fake these bravura passages—to somehow give the impression that they are achieving the same kind of prodigies with the passage-work of Mozart’s concertos as with that of the Liszt Totentanz, Tchaikovsky, Prokofiev, and the Ravel Concerto in G. I have already mentioned Schnabel in this connection. But the exercise must be a rather dispiriting one, and pianists are sometimes tempted to try something more creative. Murray Perahia’s recording of Concerto No. 26 in D, K. 537 (the “Coronation”), is an interesting case in point. Perahia shades the first-movement bravura sections with forte and mezzo forte articulations (undreamed of by Mozart, by the way)—a subtle and delicate reading, but one which I am afraid confirms the current reputation of this work as rather slight and mannered. Bilson’s extra-strenuous performance of the piece gives it a fighting chance.

What has been said about the bravura passages applies a fortiori to the cadenzas. Bilson plays excellent cadenzas, but I would not be ready to say that they are better than other pianists’ or that he plays them better. I can only say that played on a fortepiano they convince in a way they never have before as supreme demonstrations of improvisational virtù, outdoing even the bravura passages and shaming the orchestra into what is, with Mozart, always a somewhat subdued roar in the final ritornello.

True, the cadenza is not a feature of the concerto that has fared well with critics. The cadenza cuts a spontaneous, incorrigible swath through formalist critical constructions, like a runaway power-mower in a formal garden. “A bad cadenza is the very appendicitis of concerto form,” grumbled Tovey, and Rosen wrote an entire chapter on the concertos in The Classical Style barely mentioning the cadenza at all. With Bilson’s exciting fortepiano performances in our ears, perhaps we can begin to appreciate freshly the sawdust-ring atmosphere that is essential to Mozart’s as to all other good concertos.

The fact that pianists have had to hold back when they play Mozart on a modern instrument has surely contributed to the image of the composer as a contained, even muted spirit—witty rather than intense, precise rather than exuberant. To be sure, this image is not unique to the piano concertos. Mozart’s installation in the canonic musical repertory starting in the 1920s was another outcome of the same anti-romantic movement that led to “geometrical” performance styles, and to the music of Stravinsky, Hindemith, Poulenc, and early Copland. The times demanded a brisk, clear, restrained Mozart, and that is what Edwin Fischer with his chamber orchestra (not Schnabel!) provided.

It is not what is demanded today. Listen to Bilson’s use of rubato, the expressive stretching of music’s steady beat always associated with Romantic interpretation, in the adagio middle movement of Concerto No. 23 in A, K. 488 (see above). Most pianists—Brendel is an exception—shade the rhythmic values at least a little bit in measure 7 of the theme, where the fast-note figures can be made to seem more emotional than ornamental. Bilson leans so hard on the first of the six notes in each figure that measure 7 becomes a passionate parenthesis within the overriding muted lament. In addition, he takes an astounding Luftpause before measure 2 (without, however, a Schnabelesque reduction of volume) and another after measure 4, playing the next half-measure in each case slightly slower. As a result measure 2 hovers in a kind of strained distress, which cedes to resignation at the half-cadence in measure 4. The full cadence at the end of the theme is treated analogously.

Bilson may have risked this highly inflected reading because the theme looms so large within the slow movement as a whole: most of its other material closely resembles the theme,7 and its uncomplicated, austere form highlights the theme in the two solo presentations. Though these presentations of the theme are identical in the score, they should not be played identically, especially in so distinctive an interpretation. Bilson adds (improvises?) small but telling ornaments at the repetition, in measures 2–3. On the whole he tends to err on the side of caution when it comes to the embellishment of Mozart’s slow melodies—perhaps in conscious reaction (overreaction, in my view) to some rather frilly efforts that have been turning up on recent records. It so happens that this Adagio is the one Mozart concerto movement to survive in a fully ornamented contemporary copy. 8 The atrocious chromatic runs in this source should give pause to even the most hardened historical reconstructor.

Nearly all other players count as geometricians compared to Bilson when they come to this adagio theme. It would be otiose to go into detail about recordings by Brendel, Malcolm Frager, Richard Goode, Steven Lubin, Perahia, Christian Zacharias, and more. Make what you like of certain similarities of detail between Bilson and Horowitz, in his extraordinary recording of 1987; Horowitz is something, as they say, else again—something calculated and eccentric. Ashkenazy is a special case. Although he applies only the most discreet rubato, he makes emphatic use of his own proper resource—the modern piano, with its rich and subtle range of tone quality far exceeding that of any eighteenth-century instrument. He molds the theme almost entirely out of sonority, rather than rubato, dynamics, or articulation.

Thus while the high A at the beginning of measure 2 comes on time (with Bilson it comes very late), it has in Ashkenazy’s recording a quiet ringing quality that seems to well up out of the previous measure and that makes the melodic climax radiant. The specially sonorous B in measure 2 flows like liquid to the cadence, and so on. Ashkenazy lingers a little on some of the single ornamental sixteenth-notes (in the dotted-eighth-sixteenth motif) so as to make them sing. The result may seem more limpid than distressful, but it is certainly very beautiful indeed.

Bilson cannot even try for this effect because the fortepiano lacks the necessary range of sonority; this was a resource that was developed by piano makers only in the nineteenth century. I spoke confidently above about Bilson’s advantages in playing an eighteenth-century instrument. Ashkenanzy reminds us (in this concerto and others) of the advantages of the modern piano—advantages that pianists traditionally leave aside when they play Mozart.

What comes very naturally to the fortepiano are sentimental, delicate effects, which Mozart indulged in the slow movements of several of his earlier Viennese concertos, such as No. 10 in F, K. 414, No. 12 in C, K. 415, and No. 14 in B flat, K. 456. (Gardiner too does very well by these, with his deep-sighing low-vibrato strings.) Browsing through the complete set of Bilson recordings, one can only admire his responsiveness to the shades of expression called for in this amazing and amazingly diverse repertory. Charm, serenity, Empfindsamkeit, melancholy, pathos, majesty, power, high comedy, low humor—he is always there. Bilson and Gardner can also project impressive long-range effects, as for example in their performance of Concerto No. 25 in C, K. 503. This slightly forbidding work depends to a large extent on the manipulation of large rhythmic proportions. Though the finale starts lightly, as usual, it develops unusual grandeur; the balance achieved with the weighty first movement is projected here as in no other recording I know.

What would you rather hear—a great performance on modern instruments, or a fair-to-middling performance on historical instruments? People in the “traditional” camp would always trot out this mindless question as an argument-stopper when the pros and cons of historical performance were debated. The question was not only mindless but rather offensive. Nobody ever asked whether you would rather hear a great performance on the wrong instrument or a great performance on the right one. The answer to that question, of course, is “both”…

Still, after one has had some experience with the fortepiano, the use of a Steinway for a Mozart concerto can begin to feel like overkill, like using your twelve-cylinder BMW 750 to tool around the back streets of Salzburg. (Even in the bravura sections, Mozart’s Autobahn, the thing only seems to be going at about 80 kilometers.) Listening to Murray Perahia or Peter Serkin, one senses a certain reserve in their performances that has nothing to do with their feelings about Mozart, only with the consciousness that they are underplaying their instrument.9 Bilson, a player of comparable stature, deploys the resources of his instrument to the full, resources that he has discovered in a process of passionate, committed experiment over many years. He has to be the exemplary Mozart pianist for our time.

This Issue

May 18, 1989