From Eigeldinger, Chopin: Pianist and Teacher
(an indispensable book for the Chopin-lover; keep in mind this is only a small tidbit from the book). I will also give the footnotes since they are so instructive.
Streicher via Niecks wrote:
[Chopin] required adherence to the strictest rhythm, hated all lingering and dragging, misplaced rubatos, as well as exaggerated ritardandos. 'Je vous prie de vous asseoir' [Pray do take a seat] he said on such an occasion with gentle mockery.
In keeping time Chopin was inexorable, and some readers will be surprised to learn that the metronome never left his piano. Even in his much maligned tempo rubato, the hand responsible for the accompaniment would keep strict time, while the other hand, singing the melody, would free the essence of the musical thought from all rhythmic fetters, either by lingering hesitantly or by eagerly anticipating the movement with a certain impatient vehemence akin to passionate speech.
Note 92, Eigeldinger wrote:
Of those who heard Chopin play, rarely did any criticize his rubato. Those who did were non-Poles: Berlioz (see p. 272) (below) who was little of a pianist and whose symphonic aesthetic was quite contrary to Chopin's, and Mendelssohn (see p. 267) (below) who expressly declared allegiance to a more metrical conception. Moscheles's sudden change of mind (see note 97) is significant; he recognized, though, that Chopin's music in the hands of other players could well degenerate into mere lack of rhythm [Taktlosigkeit]. We know how much any such accusation irritated Chopin if directed against himself (see Peruzzi/Niecks, p 339). The quotations reproduced here concerning Chopin's rubato all reject this accusation and come from the most reliable pupils.
In fact, Mikuli is referring here to generations of pianists subsequent to Chopin's; victims of a pseudo-tradition, they submitted his music to agogic distortions in the name of the vague and convenient term 'rubato'. This practice was in vogue from before the second half of the nineteenth century up to the 1930s. The statements where Chopin's pupils denounce such abuses - which Chopin was accused at the time of having fathered - all date from that period: Streicher (around 1879), Mikuli (1879), Mathias (1897), Viardot/Saint-Saëns (1910). Kleczyński - a pupil of some of Chopin's pupils - reacted similarly in 1880. It was only with the gradual return to respecting the letter of scores (which carries its own dangers), from the years 1930–1940 or so, that this type of pseudo-rubato disappeared from most playing.
Berlioz via Eigeldinger p272 wrote:
Chopin was impatient with the constraints of meter; in my opinion he pushed rhythmic independence much too far [...] Chopin could not play in time.
Mendelssohn via Eigeldinger p267 wrote:
...as a pianist Chopin is now one of the greatest of all - doing things as original as Paganini does on the violin, and bringing about miracles that one would never have believed possible. Hiller too is a remarkable player, vigorous with a touch of coquetry. Both, however, labor somewhat under the Parisian tendency of overdoing passion and despair, and too often lose sigh of calm, discretion and the purely musical; I on the other hand perhaps do this too little - and so we all three supplemented and, I believe, learned from each other... (Letter to his mother, Düsseldorf, 23 May 1834)
Viardot via Saint-Saëns wrote:
Through Mme Viardot [...] I learned the true secret of tempo rubato [... where] the accompaniment holds its rhythm undisturbed while the melody wavers capriciously, rushes or lingers, sooner or later to fall back upon its axis. This way of playing is very difficult since it requires complete independence of the two hands; and those lacking this give both themselves and others the illusion of it by playing the melody in time and dislocating the accompaniment so that it falls beside the beat; or else - worst of all - content themselves with simply playing one hand after the other. It would be a hundred times better just to play in time, with both hands together.
Pauline Viardot was a famous singer, Chopin's favorite, and also a pianist and sometimes student of his. As for this playing one hand after the other thing - I listened to Lang Lang playing Chopin 27/2
recently (a different performance than the one I was looking for), and he does this sometimes. Very annoying. Also, for those who think this type of rubato is specific to only one composer (Chopin), it appears as though Saint-Saëns at least related to it quite well from the above quote. See note 95 for a quote from Mozart also supporting it, and Chopin recommended playing Weber's music in this way in the next quote (after the footnotes).
Note 93, Eigeldinger wrote:
This independence of the two hands, aiming at a complex complementarity, is one of Chopin's characteristic traits. It cannot be fortuitous that he chose to be represented in the Fétis-Moscheles Méthode des méthodes  by three Etudes of which two are based on polymetric principles: the first (F minor) with threes against fours and the third (A-flat) with threes against twos. The [Fantasie-]Impromptu Op. 66 offers a juxtaposition of these two systems through its constituent sections; similarly the Etude Op. 25/2 is based entirely on 'rhythmic exchanges'. Similar ideas occur at bars 249-72 and 849-75 of the Scherzo op. 54, as also in the Waltz op. 42, which Lenz called 'the most typical embodiment of Chopin's rubato style'.
I hadn't previously read this footnote. ^^ It's nice to see an echo of my own thoughts here.
Note 94, Eigeldinger wrote:
This practice, criticized by Saint-Saëns, is clearly recognizable in the recordings of 'renowned' Chopin players of the time, notably Leschetizky, Pugno, Pachmann, Friedman and, to a lesser extent, Paderewski and Maurycy [Moritz] Rosenthal.
Hey, that's our great-great grandteacher there! (Many of us, if I recall.)
Everyone knows that rubato is an indication often encountered in old music; its essence is fluctuation of movement, one of the two principal means of expression in music, namely the modification of tone and of tempo, as in the art or oration, whereby the speaker, moved by this or that emotion, raises or lowers his voice, and accelerates or draws out his diction. Thus rubato is a nuance of movement, involving anticipation and delay, anxiety and indolence, agitation and calm; but what moderation is needed in its use, and how all too often it is abused! [...] There was another aspect: Chopin, as Mme Camille Dubois explains so well, often required simultaneously that the left hand, playing the accompaniment, should maintain strict time, while the melodic line should enjoy freedom of expression with fluctuations of speed. This is quite feasible: you can be early, you can be late, the two hands are not in phase; then you make a compensation which reestablishes the ensemble. In Weber's music, for example, Chopin recommended this way of playing. He often told me to use it, it's as though I still hear him: in the Sonata in A flat [op. 39], in the A flat passage of the agitato in the Concertstück [Op. 79, first movement bars 57ff]...
Note 95, Eigeldinger wrote:
Tempo rubato: stolen time. Although this expression first appears in 1723 in the treatise by Tosi (Bolognese theoreticial of bel canto), the musical reality which it reflects can be traced back at least to the beginnings of accompanied monody in Italian humanist circles. The following postulates emerge from Tosi's seminal writings for the intelligent use of the singer in particularly expressive passages [mostly in slow tempi] in various pieces [recitatives, arias, ariosi], rubato is a system of compensation whereby the value of a note may be prolonged or shortened to the detriment or gain of the succeeding note. This metric 'larceny' is best applied to improvised ornaments [taking the sense of the words into account as much as the music] over the imperturbable movement of the bass (underlined by Tosi). It results from counterpoint between the solo line and the bass line and is characterized, vertically speaking, by moments of metric displacement between the two parts; it is left to the singer's discretion to use it with moderation, according to the rules of good taste. Here we have the pure tradition of Italian Baroque bel canto, linked with the art of improvising suitable ornaments, and deriving from theory of affetti.
Bel canto, dominant in Europe at the end of the seventeenth and above all the eighteenth century, was transposed together with the art of rubato into the domain of instrumental music in its chamber, concertante and solo genres. Thus it came to be codified fairly accurately in the important instrumental treatises of the period: C.P.E. Bach for the keyboard, Leopold Mozart for the violin, and Quantz for the transverse flute. Wolfgang Mozart, who had been well schooled, proudly related to his father (Augsburg, 24 October 1777): 'They all are amazed that I play accurately in time. They can't grasp that in tempo rubato in an Adagio the left hand goes on unperturbed; with them the left hand follows suit'. If this independence of the hands is applied to some places in the B minor Adagio (K540), the adagio sections of the Fantasies in D minor (K397) and C minor (K 475), the various reprises in the A minor Rondo (K 511) or even to some slow movements in the sonatas and concertos, one can feel how closely Mozart anticipated Chopin!
This tradition was maintained in the instrumental field well into the Classical era, and codified once again by Türk in 1789. Tempo rubato, still very much alive in Romantic bel canto (and, exceptionally, in Paganini's Concertos - Chopin heard him in 1829 when the latter gave ten concerts in Warsaw) was, by the beginning of the nineteenth century, gradually supplanted in instrumental music by larger-scale tempo fluctuations. The Frenchman Louis Adam, educated in the old tradition, observed: 'Some people have tried to start a trend of playing out of time, playing all genres of music like a fantasy, prelude or capriccio. It is thought to enhance the expression of a piece, while serving in effect to distort it beyond recognition. Naturally, expressivity requires certain notes of the melody to be slowed or quickened; however, these fluctuations must not be used continually throughout the piece, but only in places where the expression of a languorous melody or the passion of an agitated one demands a slower or a more animated pace. In this case it is the melody that should be altered, while the bass should strictly maintain the beat.' The notion of rubato is then confused with that of tempo ad libitum in pieces written in the free style. According to Schindler, Beethoven finally adopted the term rubato in this new and incorrect sense designating fluctuations of tempo. In 1828 Hummel denounced this latest practice: 'Lately several artists have been trying to replace natural feeling with manufactured feeling; as for instance [...] by slowing down the beat (tempo rubato) at every possible opportunity to the point of satiation.' Hummel, who was a pupil of Mozart, continues to recommend the use of traditional rubato in Adagios, but no longer designates it by this name.
Thus Chopin practiced and taught rubato in its traditional and original meaning, at a time when that practice was on the decline, if not already abolished, in other piano music. His attachment to the Baroque aesthetic may be explained by two factors: first, his training from Żynwy and Elsner, both products of the pre-Classical era and raised within Italianized circles (Prague and Vienna respectively); second, Chopin's own taste for bel canto, evident from in his adolescence on - we have seen how assiduously he frequented the Warsaw National Theater, where Italianism dominated as much through Rossini as through the operas of Kamieński and Kurpiński. One might add that the singing class at the Warsaw Conservatory was then directed by the Piedmontese teacher Carlo Soliva. Faithful to the aesthetic of his education, Chopin was to transmit it through his own teaching (whence the continual appearance of indications along these lines in the annotated scores of Mme Dubois).
For more information on Chopin's connections with Baroque and Classical rubato, see (a bunch of references).
Note 96, Eigeldinger wrote:
This assertion of course applies equally to the inverse case, when the melody is in the left hand and the accompaniment in the right. Amongst many examples can be mentioned the Etude op. 25/7, Prelude op. 28/6, Mazurka op. 7/3 (bars 56-73), Polonaise op. 26/1 (bars 66-82), Waltz op. 34/2 (principal motif and bars 169-88), etc.
What characterized Chopin's playing was his rubato, in which the totality of the rhythm was constantly respected. 'The left hand,' I often heard him say, 'is the choir master [Kapellmeister]: it mustn't relent or bend. It's a clock. Do with the right hand what you want and can.' He would say, 'A piece lasts for, say, five minutes, only in that it occupies this time for its overall performance; internal details [of pace within the piece] are another matter. And there you have rubato.'
Note 97, Eigeldinger wrote:
Chopin was fond of this metaphor and used it often. It appears with small variants or commentaries in the following texts:
—Franchomme, Potocka, Czartoryska/Planté
Mme Peruzzi recalls Chopin 'calling his left hand his maître de chapelle and allowing his right to wander about ad libitum.' This corresponds precisely to the impression received by Moscheles (who had a more fundamentally metric conception) on first hearing Chopin: 'His ad libitum playing, which with other interpreters of his music tends to degenerate into a mere lack of rhythm [Taktlosigkeit], in his hands is the most graceful and original feature of the discourse [...] one feels drawn as by a singer who, unpreoccupied with the accompaniment, completely follows his or her feelings'.
Note 98, Eigeldinger wrote:
This text and the preceding one (Mathias) are of prime importance. Originating from Chopin's students, they are the only ones that let us assume that his rubato took two different forms, by no means mutually exclusive. Kleczyński was of this opinion (see above), followed recently by Higgins. I share this view, with the added nuance that a third component of Chopinian rubato is derived from the mobile rhythm of the Mazur. The first type of rubato, descended from the Italian Baroque tradition, has been discussed in note 95; it occurs principally in works with broad cantilenas. The second, more common type consists of fleeting changes of pace relative to the basic tempo; these agogic modifications may affect a whole section, period or phrase, slowing down or accelerating the flow depending on the direction of the music. This rubato is to be applied not arbitrarily but as a function of the musical texture and the basic laws of declamation. These agogic fluctuations are called rubato by extension only, since they affect the musical structure from top to bottom, not merely the melodic line. It is not unusual for these nuances of tempo to be specified in Chopin's music. Thus the section sof the Waltz op 64/2 are differentiated by the indications tempo giusto - più mosso - più lento - più mosso - tempo I - più mosso. Within a section the tempo is also to be progressively accelerated, then slowed down, by the indications agitato - sempre più mosso - calando - smorz. - riten. (Ballade op. 23, bars 40-67) - similarly within a musical period by the complementary copuled indications stretto - riten. (Etude op. 10.3 bars 7-8, 15-16) or poco riten. - accel. (Polonaise op. 26/2 bars 1-6 and similar). The coda of the Mazurka op. 24/4 is a remarkable example of progressive rallentando specified by riten. - calando - mancando - sempre rallent. - smorzando (bar 129 to end). (On the subject of his musical editing, we may note the growing scarcity of agogic and other expressive markings from op. 25 onwards, as discussed in note 99 below.)
The above considerations merely distinguish the two basic types of rubato employed by Chopin; they cannot by any means convey all the subtle flexibility of movement in his playing, of which we know only that it was conditioned by an acute awareness of the length of the piece and by an internal logic commanding the tempo nuances in relation to the basic pulse.
Koczalski's explanation of rubato - although he himself mastered it to perfection - is unconvincing, which is why it is not quoted here.
Mikuli via Michałowski wrote:
How did Chopin understand rubato? Was it synonymous with complete freedom and arbitrariness of rhythm, or was it just the expression of a living undulation of tempo which avoided exact coincidence with the strict metric framework [...] Mikuli, on the basis of his personal reminiscences, answered as follows:
Chopin was far from being a partisan to metric rigor and frequently used rubato in his playing, accelerating or slowing down this or that theme. But Chopin's rubato possessed an unshakeable emotional logic. It always justified itself by a strengthening or weakening of the melodic line, by harmonic details, by the figurative structure. It was fluid, natural; it never degenerated into exaggeration or affectation.
[...] rubato is never a defect in the time; the idea of rhythm, and consequently of the relative value of the notes, must never be lost, apparent changes and momentary incongruities notwithstanding. I shall now give the result of my own reflections on the rubato of Chopin:
1. Precise rules for it cannot be given, because a good execution of the rubato requires a certain musical intuition, that is to say, a certain particular talent.
2. Every rubato has for its foundation the following idea: each musical thought contains moments in which the voice should be raised or lowered, moments in which the tendency is to retardation or acceleration. The rubato is only the exaggeration or bringing into prominence [of] these different parts of the thought: the shadings of the voice make themselves more marked, the differences in the value of notes more apparent. Hence there arises in the mind an image of the musical thought more full of vitality and of poetry, but always in accordance with law and order [...] We in all cases borrow the time from notes of smaller importance for the purpose of giving it to the principal notes.
In his playing the great artist [Chopin] rendered most exquisitely that kind of agitated trepidation, timid or breathless [...] He always made the melody undulate like a skiff borne on the breast of a powerful wave; or sometimes he made it hover like an airy apparition suddenly sprung up in this tangible and palpable world. In his writings, he at first indicated this way of playing - which gave such an individual stamp to his virtuosity - by the term 'tempo rubato': stolen, broken time - a rhythm simultaneously supple, abrupt and languid, vacillating like the flame under the breath that agitates it, like the corn in the field waving under the soft pressures of the warm air, like the tops of trees bent hither and thither by a strong breeze.
But as the term taught nothing to whoever already knew, and said nothing to those who did not know, understand, and feel, Chopin later ceased to add this explanation to his music, persuaded that if one had the sense of the music, it would be impossible not also to divine this rule of irregularity. Also, all his compositions must be played with that kind of speech-like, accented lilt, that softness [morbidezza], the secret of which it was difficult to grasp if one had not often heard him play in person. He seemed to wish to teach this style of playing to his numerous pupils, especially his compatriots to whom, more than to any others, he wanted to communicate the breath of his inspiration.
Liszt via Niecks wrote:
'Look at these trees!' [Liszt] said, 'the wind plays in the leaves, stirs up life among them, the tree remains the same, that is Chopinesque rubato.'
Note 99, Eigeldinger wrote:
Chopin ceases in effect to mark the word 'rubato' from op. 24 onwards. (This goes with a parallel progressive decrease in indications of mood or character and metronome markings, very frequent and diversified in his early works, but thereafter tending towards an increasing sobriety visible at all levels of Chopin's musical editing.) Liszt's explanation is convincing: doubtless Chopin realized that the word was insufficient to convey his intentions and could be misleading to his contemporaries - who did indeed criticize his attempts to notate 'to a certain extent' some aspects of 'his rubato' (see Le Pianiste, 1834-1835, pp. 78-9, on the subject of his op. 15).
But what meaning (or meanings) does this word have in the thirteen compositions in which it occurs? Does it refer to the Italian vocal tradition, as Kamieński maintains? - or, as Kreutz believes, to both the types described in note 98 above, according to the context? Or rather, as Jadwiga and Marian Sobiescy believe, does it emphasize the mobile agogic rhythm derived from Polish folk melodies? To attempt an answer to this thorny question we have to examine the musical contexts and genres in which this notation is used.
With the exception of the G# minor Polonaise [without op. no., c. 1823], where the term is used improperly in the final cadences of bars 12 and 27 to indicate an approaching senza rigore, the term 'rubato' occurs in two broad types of context:
1a - At the beginning of a piece (opp. 15/3; 24/1; 67/3 - in the last case according to the Fontana edition, the manuscript being lost).
1b - At the beginning of a new motif which is to direct the piece towards the final cadence (op. 9/2 bar 26).
2a - At the repetition of a phrase or half-phrase (op. 6/1 bar 9; op 6/2 bar 65; op. 7/1 bar 49; op. 7/3 bars 17 and 93; op. 21, finale, bar 173; op. 24/2 bar 29).
2b - In the second half - last four bars - of a phrase (op. 8, first movement, bars 22-4 and 159-61; op. 16 bar 132; op 21, finale, bar 157).
As for the genre of compositions featuring this notation, a good three-quarters of these works are genres connected with Polish folk music. Concerning tempo, all the above-mentioned pieces are in a quick tempo with the exception of the Nocturnes op. 9/2 and 15/3 and Mazurkas op. 24/1 and 67/3. These last three pieces are marked rubato at the first bar; thus placed, the indication applies to the entire piece or at least to its first section. It therefore concerns agogic fluctuations, the second type of rubato described in note 98. As for op. 9/2, a perfect example of bel canto adapted to the piano, it arises out of the Italian tradition: even if the rubato here is applicable to various other points in the same piece, it belongs essentially to one particular phrase of a more pathétique character - to use Tosi's own words. This definition also applies logically to the above-mentioned passages of the Trio op. 8, the Rondo op. 16 and the Concerto op. 21, even though all of these are in a quick tempo: underneath the piano melody, the violin and 'cello parts, or the orchestral parts, bear no mention of rubato but keep the beat. At the same time, the passages marked rubato in op. 21 derive directly from the Mazur and so relate also to categories 2a and 2b, of pieces inspired by the mobile rhythm of Polish folk music. In each case Chopin took the trouble to notate a 'rhythmic rubato' in the melodic line; thus the term 'rubato' serves there merely to underline the precise flexibility required for these subtle nuances. This type of 'national' rubato, the third component of Chopinian rubato, is by no means incompatible with that derived from the Italian Baroque: the best Polish folk musicians, in monodic chants, employ the compensatory system (lengthening or shortening one note value to the detriment or gain of the next), while stamping a strict triple meter with the foot. This brings one final point: of the twelve compositions examined, nine are in triple time (op. 8 is in 4/4, op. 9/2 in 12/8 and op. 16 in 2/4); moreover, the nocturne op. 15/3 features many folkloric characteristics (see p. 153, note 187). This supports an argument in favor of this 'national' rubato having been instinctively applied by Chopin, harmoniously combined in his music with the other two types, each in its context.
If, then, Chopin's rubato may be seen to take diverse meanings, when he marks it explicitly it seems to be the 'national' element that takes precedence. As for the Italian tradition, it evidently applies to works with a broad cantilena, as much in slow tempi as in more restless fiery passages. It is easy to conclude from this that Chopin, after op. 24, renounced the use of a term which he would have had to employ constantly without the slightest assurance of his intentions being correctly understood.
Note 100, Eigeldinger wrote:
Altogether Liszt's poetic evocation alludes to the Italian vocal tradition adapted to the piano by Chopin. This is confirmed in an excerpt from Lachmund's diary (p. 62, on the subject of the sixth of Liszt's Consolations): 'On this occasion [Liszt gave us] an important insight into the Lisztian rubato, consisting of subtle variations of tempo and expression within a free declamation, entirely different from Chopin's give-and-take [Eilen und Zögern]. Liszt's rubato is more a sudden, light suspension of the rhythm on this or that significant note, so that the phrasing will above all be clearly and convincingly brought out. While playing, Liszt seemed barely preoccupied with keeping in time, and yet neither the aesthetic symmetry nor the rhythm was affected.'
Towards the end of his description Liszt singles out the Polish students to whom Chopin devoted the greatest care; this might also suggest that Liszt was equally aware of the mobile rhythm of Polish folk music as a component of Chopin's rubato. In fact Chopin readily affirmed that the purely national aspects of his playing and his music tended to escape foreigners: 'When one of his French pupils played his works to the approval of the listeners, Chopin would often remark that the performance had indeed been good but that the Polish element and the Polish inspiration were lacking' (Karasowski). This is corroborated by Marie Roubaud: 'He often said that French did not understand his Mazurkas, and that one had to be Polish to feel the subtleties of the national rhythm, and to render the proper local color' (Ganche). This is vividly illustrated, too, by Chopin's dispute with Meyerbeer, and by Hallé's and Moscheles's astonishment at the rhythm Chopin imparted to the Mazurkas.
Some references are omitted or truncated.
One conclusion is that we must be familiar with Polish music to play Chopin. Could it be it that to understand him better we need to listen to Szymanowski, Gorecki and Penderecki than to Schumann and Mendelssohn? To a musical world where two traditions prevailed, the German and the French, the Polish one must have been difficult to grasp. I now wonder, is my dislike of Chopin related not to his music, but by the way he is misplayed, as if he were no Pole, but a German? How would Rachmaninoff sound if we were to pay him as if he were Richard Strauss? How does Handel sound when played as if he were Bach?