Rhythm is the pulse in music. Rhythm marks the beating of its heart, proves its vitality, attests its very existence. Rhythm is order. But this order in music cannot progress with the cosmic regularity of a planet, nor with the automatic uniformity of a clock. It reflects life, organic human life, with all its attributes, therefore it is subject to moods and emotions, to rapture and depression.
There is in music no absolute rate of movement. The tempo, as we usually call it, depends on physiological and physical conditions. It is influenced by interior or exterior temperature, by surroundings, instruments, acoustics.
There is no absolute rhythm. In the course of the dramatic developments of a musical composition, the initial themes change their character, consequently rhythm changes also, and, in conformity with that character, it has to be energetic or languishing, crisp or elastic, steady or capricious. Rhythm is life.
According to a current story, Chopin used to say to his pupils: "Play freely with the right hand, but the left one act as your conductor and keep time." We do not know whether the story should be afforded the benefit of the doubt. Even if it be exact, the great composer contradicted it most energetically in such wonderful compositions as the Etude in C-sharp minor, preludes No. 6 and No. 22, the Polonaise in C minor, and in so many fragments of others of his masterpieces, where the left hand does not play the part of a conductor, but most distinctly that of a prima donna. Another contradiction of this theory, or rather of the way Chopin put it into practice, is the testimony of some of his contemporaries. Berlioz affirms most emphatically that Chopin could not play in time, and Sir Charles Hallé pretends to have proved to Chopin, by counting, that he played some Mazurkas 4/4 instead of 3/4 time. In replying to Charles Hallé, Chopin is said to have observed, humorously, that this was quite in the national character. Both Berlioz and Hallé evidently intended to testify against Chopin. Berlioz, although extremely sensitive to the picturesque and the characteristic, was not emotional at all; besides, the instrument which he played best, the instrument on which he even tried to perform before some of his friends is Symphonie fantastique, the sonorous and expressive guitar, could not reveal to the great man all the possibilities of musical interpretation. As for Sir Charles Hallé, a distinguished, but rather too scholastic pianist, this estimable gentleman, who knew so many things, ought to have known better here. Our human metronome, the heart, under the influence of emotion, ceases to beat regularly—physiology calls it arythmia.  Chopin played from his heart. His playing was not national; it was emotional. To be emotional in musical interpretation, yet obedient to the initial tempo and true to the metronome, means about as much as being sentimental in engineering. Mechanical execution and emotion are incompatible. To play Chopin's G major Nocturne with rhythmic rigidity and pious respect for the indicated rate of movement would be as intolerably monotonous, as absurdly pedantic, as to recite Gray's famous Elegy to the beating of a metronome. The tempo as a general indication of character in a composition is undoubtedly of great importance; the metronome may be useful; Melzel's ingenious device, though far from being perfect, is quite particularly helpful to students not endowed by nature with a keen sense of rhythm; but a composer's imagination and an interpreter's emotion are not found to be humble slaves of either metronome or tempo.
Our Olympian predecessors, the classics, although living under different conditions, and on a plane above that of our present-day nervousness and excitement, seemed to realize the impossibility and containing some of their ideas within the limits of the indicated time and rate of movement. In Bach's works we sometimes see Adagio and Allegro, Animato and Lento in the same bar. Haydn and Mozart frequently use expressions such as quasi cadenza, ad libit., leaving thus to the interpreter the entire freedom as to the rhythm and rate of movement. The most human of them, the most passionate, the only composer who knew almost exactly how to express what he wanted, Beethoven, took quite particular care of tempi and dynamic indications. When we look at the first movements of the D minor Sonata, of the op. 57, of the op. 111, at the Largo in the op. 106, and especially at the beginning of the Adagio in the op. 110, we see the embarrassment and discomfort to which all the tempo-sticklers and metronome-believers are exposed when attempting to play or to teach these works. And yet, in spite of his stupendous, almost abnormal, sense of precision, in spite of his vast knowledge of italian terminology—a quality in which nearly all modern, non-Italian composers are positively deficient—Beethoven could not always be precise. Why? Because there are in musical expression certain things which are vague and consequently cannot be defined; because they are according to individuals, voices or instruments; because a musical composition, printed or written, is, after all, a form, a mould: the performer infuses life into it, and, whatever the strength of that life may be, he must be given a reasonable amount of liberty, he must endowed with some discretional power. In modern meaning discretional power is Tempo Rubato.
Tempo Rubato, this irreconcilable foe of the metronome, is one of music's oldest friends. It is older than the romantic school, it is older than Mozart, it is older than Bach. Girolamo Frescobaldi, in the begging of the seventeenth century, made ample use of it. Why is called rubato we do not really know. All lexicons give the literal translation of it as robbed, stolen time. Now, the most common, the most frequent, the simplest form of Tempo Rubato is obtained by ritenuto or a ritardando which, as every one knows, serve to increase the value of respective notes. Where there is an increase, there can have been no robbery. Addition cannot be called subtraction. Although we protest against the use of the words: robbed, stolen time, we recognize that the very essence of Tempo Rubato is a certain disregard of the established properties of rhythm and rate of movement. The French translation of Tempo Rubato: movement derobe, while not giving the full, modern meaning of it, is the best of all. It implies the idea of fleeing away from the strict value of the notes, evading metric discipline. We should be inclined to call it evasive movement.
It would be wrong to pretend that Tempo Rubato is the exclusive privilege of the higher artistic form in music. Popular instinct evolved it probably long before the first sonata was written. Expressed although nameless, it has always been in all national music. It is Tempo Rubato which makes the Hungarian dances so fantastic, fascinating, capricious; which so often makes the Viennese waltz sound like 2/4 instead of 3/4 time; which gives to the mazurka that peculiar accent on the third beat, resulting sometimes in 3/4 + 1/16: 
Example 1 ( see a larger image).
The literature concerning Tempo Rubato is not particularly rich. Apart from short notes to be found in lexicons, we can only quote a few really authoritative opinions, always admitting that there may be some others, and very valuable ones, unfortunately unknown to us. Liszt, in his beautiful though rather bombastic volume, Frederic Chopin, devotes to the subject a few interesting passages; Ehlert and Hanslick, as far as we can remember, seem to pay little attention to it; on the contrary, Nieck, Kleczyński, and especially Huneker, treat it more extensively.
Peculiarly enough, all the above-mentioned authors speak about the matter incidentally and in conjunction with Chopin, as if Tempo Rubato were an exclusive attribute of Chopin's music; all of them say excellent things without solving the question, which is still and will be open to further investigation.
We do not pretend to have anything to say upon the subject; our desire is to remove the stigma of morbidness which seem to be attached to it. Tempo Rubato is not pathological, it is physiological, as it is a normal function of interpretative art. In our opinion it is not so much Tempo Rubato, as the romance of Chopin's life and his premature end, which are responsible for the silly superstition that Chopin should be played in a soft, sentimental, sickly manner. Tempo Rubato is a potent factor in musical oratory, and every interpreter should be able to use it skillfully and judiciously, as it emphasizes the expression, introduces variety, infuses life into mechanical execution. It softens the sharpness of lines, blunts the structural angles without ruining them, because its action is not destructive: it intensifies, subtilizes, idealizes the rhythm. As stated above, it converts energy into languor, crispness into elasticity, steadiness into capriciousness. It gives music, already possessed of the metric and rhythmic accents, a third accent, emotional, individual, that which Mathis Lussy, in his excellent book on musical expression, calls l'accent pathètique.
The technical side of Tempo Rubato consists, as is generally admitted, of a more or less important slackening or quickening of the time or rate of movement. Some people, evidently led by laudable principle of equity, while insisting on the fact of stolen time, pretend that what is stolen ought to be restored. We dully acknowledge the highly moral motives of this theory, but we humbly confess that our ethics do not reach such a high level. The making up of what has been lost is natural in the case of playing with orchestra, where, for the security of the whole, in spite of the fractional alterations of the movement, the metric integrity should be rigorously preserved. With soloists is quite different. The value of notes diminished in one period through accelerando, cannot always be restored in another by ritardando. What is lost is lost. For any lawlessness there is, after a certain term, proscription.
As we have already said, Tempo Rubato appears frequently in popular music, especially in dances, consequently it ought to be used in the works of Chopin, Schubert, Schummann (Papillons, Carnival), Brahms, Liszt, Grieg, and in all compositions which have folk music as a foundation. Practically, it can be used anywhere—save, perhaps in some ancestral music, where there is room for no passion, where the serene purity of architecture, a majestic dignity and repose, led to spheres of almost immaterial and unearthly beauty.
It would be unthinkable to play Chopin without using Tempo Rubato; but neither would anyone do justice to such works as Schumann's Fantasia, Fantasiestucke, Carnival, Humoreske, the sonata in F sharp minor, etc., without wisely applying that means of expression. However strong and peculiar was the Mendelssohn's dislike of Tempo Rubato, we cannot recommend too unconditional a respect for the great composer's personal feeling in this matter. Some of his Songs Without Words, of predominantly lyric character, music be played freely, because whatever is lyric defies the rigidity of metric and rhythmic lines. Curiously enough, one of the most striking examples of Tempo Rubato is to be found in Mendelssohn's Violin Concerto, in the short Intermezzo leading from the Andante to the Finale. We will remember the playing of this by the great Joachim—in our opinion the greatest exponent of classical music; it was most distinctly rubato. 
I.A.F.M., in his concise, but excellent description of Tempo Rubato, published in Grove's dictionary, expresses doubt whether rubato should be used in Beethoven. To this we answer without hesitation in the affirmative. Rubato was Rubinstein's playing of the opening bars and the Andante of the G major concerto; Rubato was Joachim's rendering of the middle part in the finale of the violin concerto; and Bülow, whom we by no means pretend to put on the same level as the two artists just mentioned, but who was a great authority in Germany, indulged in Tempo Rubato very frequently, when playing Beethoven. The Largo in the C minor, the Andante in the G major, the Adagio in the E-flat concertos call imperatively for Tempo Rubato. And what would a pianist with a grain of common sense do in passage such as [the one from] opus 111, without Tempo Rubato?
Example 2 ( see a larger image).
In fact, every composer, when using such words as espressivo, con molto sentimento, con passione, teneramente, etc., demands from the exponent, according to the term indicated, a certain amount of emotion, and emotion excludes regularity. Tempo Rubato then becomes an indispensible assistant, but with it, unfortunately, appears also the danger of exaggeration. Real knowledge of different styles, a cultural musical taste, and a well-balanced sense of vivid rhythm should guard the interpreter against any abuse. Excess of freedom is often more pernicious than the severity of the law.
Richard Hudson, Stolen Time: The History of Tempo Rubato (Oxford: Oxford University Press, 1997); Ewa Dobrzynski, "Le tempo rubato selon Paderewski" [Tempo rubato according to Paderewski], in L'interpretation de Chopin en France (Paris: Champion, 1990), 143-152; Albert W. Zak, Paderewski's "Theme Varié" op. 16 no. 3: An Eclectic Analysis (D.M.A. thesis, New York University, 1999); Paul A. Carlson, Early Interpretation of Debussy's Piano Music, (D.M.A. thesis, Boston University, 1998); William Hunter Heiles, Rhythmic nuance in Chopin performances recorded by Moriz Rosenthal, Ignaz Friedman, and Ignaz Jan Paderewski, (D.M.A. thesis, University of Illinois, 1964).
Henry T. Finck was a steadfast Paderewski supporter, he also wrote popular books about Chopin and other composers (1892), and a brief volume about Paderewski, Paderewski and His Art (New York, Looker-on Pub. Co., 1896). Back
Hudson discusses Paderewski's contribution to the "modern tempo rubato" on pp. 318-321. J.A. Fuller-Maitland (1856-1936) published books on music including: The Age of Bach and Handel (Oxford: Clarendon Press, 1902); English Music in the XIXth Century (London: G. Richards, 1902); Brahms (New York, J. Lane, 1911). He was the editor of the second version of the Grove's Dictionary of Music and Musicians (New York: Macmillan, 1907, 1908, 1909, 1910, 1911). Back
For Claude Debussy's views see Three Classics in the Aesthetic of Music: Monsieur Croche the Dilettante Hater, (New York, Dover Publications, 1962; reprint). For Messiaen's theory of rhythm see his Traité de rythme, de couleur, et d'ornithologie: (1949-1992): en sept tomes (Paris: A. Leduc, since 1994); also see Conversations with Olivier Messiaen by Claude Samuel (London: Stainer & Bell, 1976); English translation of Entretiens avec Olivier Messiaen. (Paris: Paris, Belfond, 1967). Back
The best source to Paderewski's concert repertoire is Małgorzata Perkowska, Diariusz koncertowy Ignacego Jana Paderewskiego [Paderewski's Concert Diary] (Kraków: PWM, 1990). Perkowska lists his concerts with details by year and his repertoire by each composer and work.Back
For Paderewski's Debussy repertoire see Perkowska, op. cit., p. 219.Back
Paderewski's Bach repertoire is listed in Perkowska, op. cit., p. 201-202.Back
See Perkowska, op. cit., pp. 220, 226.Back
For Beethoven see Perkowska, op. cit., pp. 201-205.Back
For Mendelssohn see Perkowska, p. 226.Back
Thomas Gray (1716 - 1771), British poet, author of odes and sonnets. Entry from Volume 109: Eighteenth-Century British Poets, John Sitter, ed. (The Gale Group, 1991), pp. 168-182].Back
Hector Berlioz (1803-1869), French composer also working as a music critic, frequently wrote about Chopin, reviewing his Paris performances for Le Pianiste and Gazette musicale. For more information see Jean-Jacques Eigeldinger, Chopin Pianist and Teacher as Seen by His Pupils (London: Cambridge University Press, 1986); Jean-Jacques Eigeldinger, L'univers musical de Chopin (Paris: Fayard, 2000). For Chopin's own remarks about performance and interpretation see Jean-Jacques Eigeldinger, ed. Frédéric Chopin. Esquisses pour une méthode de piano [F.C. Sketches for a method of piano performance] (Paris: Flammarion, 1993). Back
Charles Hallé (1819-1895) was a British pianist. His remarks about Chopin are found in his letters, Life and Letters of Sir Charles Hallé Being an Autobiography (1819-1860) with Correspondence and Diaries, ed. C. e. et Marie Halle (London, 1896). Back
Paderewski also uses the concept of "arytmia" ["a-rhythm" or "absence of rhythm"] as a national trait in his article on Chopin of 1910 (speech delivered during Chopin Centennial in Lwów, Austrian Poland), published in a translation by Laurence Alma Tadema in 1911. A reprint of this text will appear in Polish Music Journal vol. 4 no. 2 (Winter 2001). Back
Girolamo Frescobaldi (1584-1643), was an Italian composer of the transitional period from the Renaissance to the Baroque. His music deeply impressed Johann Sebastian Bach who copied Frescobaldi's choral collection by hand, Fiori musicali (Musical Flowers, 1635). Back
This definition appears for instance in the entry on "rubato" in the third edition of Grove's Dictionary of Music and Musicians, ed. H.C. Colles (New York: The Macmillan Co., 1942). The entry begins: "The word means 'robbed' (time), and the thing may be defined as the effect of pitch upon duration, a definition which holds also if we extend pitch to mean a cluster of pitches, in counterpoint, harmony or orchestration." For a comprehensive study of tempo rubato see Richard Hudson, op. cit. Back
First two measures of Chopin's Mazurka Op. 68 No. 1 in C major; the tempo is Vivace and the editions (including the one supervised by Paderewski himself) do not include any pauses. The mazurka's meter is 3/4 and there is a sforzato accent on the last note of each measure.Back
Franz Liszt (1811-1886), Chopin (Paris, Escudier, 1852; reprints - Paris: Club français du livre, 1950; Leipzig: Breitkopf & Hartel, 1923). Back
Louis Ehlert (1825-1884), Frederic Chopin: An Essay (New York: Charles F. Tretbar, 1884). Eduard Hanslick (1825-1904) did not write books about Chopin; his remarks might be included in his concert reviews; see Hanslick, Sämtliche Schriften (Wien: Böhlau, 1993).Back
Frederick Niecks, Frederick Chopin as a man and musician (London: Novello, Ewer & Co., 2 ed. 1890); Jan Kleczyński (1837-1895), The Works of Frederic Chopin, and their Proper Interpretation: Three Lectures Delivered at Varsovia (London: W. Reeves, 2nd ed. 1880, 1889, reprinted in 1913); James Huneker (1857-1921), Chopin: The Man and his Music (London: William Reeves, 1903). Back
Mathis Lussy (1828-1910), Musical Expression, Accents, Nuances, and Tempo, in Vocal and Instrumental Music. (London, New York, Novello, Ewer, 1884). Back
Joseph Joachim (1831-1907) was a Viennese violinist, a friend of Brahms, and the foremost virtuoso of his era. Paderewski met Joachim in 1882 and performed some of his pieces; his paths then crossed during concert tours, for instance in Wiesbaden in 1900 and in Bonn, in 1901. Back
Paderewski used the second edition of the Grove's Dictionary of Music and Musicians issued in 1907 and reprinted in 1908; the third edition of the Grove's Dictionary, published in 1928 and reprinted almost yearly between 1935 and 1942, includes an entry on rubato, penned by "A. H. F. S." - A. H. Fox-Strangways. The list of contributors to that volume includes J. A. Fuller Maitland, the editor of the second edition of the Grove's Dictionary, who penned the entry cited by Paderewski. The dictionary was published by Macmillan in London.Back
Joachim, see note 21; Anton Rubinstein, Russian pianist-composer of Jewish descent (1830-1894), frequently appeared on European concert tours and preceded Paderewski to the U.S.; Hans von Bülow (1830-1894) was a German conductor, the champion of modern music, including Wagner and Brahms. Back
Ignacy Jan Paderewski