In more than one case he has restored harmonies of priceless value from incomplete texts, by means of research and reasoning which he sums up in a modest footnote that reads as something self-evident. His prefaces to the Bach-Gesellschaft volumes are perhaps the most valuable contributions to the criticism of 18th-century music ever written, Spitta's great biography not excepted.

Bach's importance in the history of music cannot be exaggerated. His art, neglected as old-fashioned and crabbed by his younger contemporaries, survived only in certain limited aspects as the subject of a desultory and unintelligent academic study, until its re-discovery by Mendelssohn. And yet, whatever disguise may have been foisted on it by corrupt traditions and ignorance of its idioms, whenever any fragment of it gained the inner ear of a true composer the effect on the history of music was immediate and profound. Indeed his influence is by no means chiefly manifested in the time when his work became known in its larger aspects, though the Bach-revival is very obviously connected with certain tendencies in the "Romantic" movement in music. But, however clear we may consider Bach's claim to the title of "the first of Romanticists," the full influence of his whole work has hardly yet begun to show itself. Schumann died before even such enthusiasts as the editors of the Bach-Gesellschaft began to find more beauty than extravagance in Bach's ordinary musical language (see, for example, Hauptmann's letters passim, The Letters of a Leipzig Cantor, trans. by A. D. Coleridge, London, Novello, Ewer, 1892), or, indeed, to grasp the main features of his designs.[3] The labours of the Bach-Gesellschaft have occupied more than fifty years, during which about four-fifths of Bach's choral works have been published for the first time; and it would be surprising if another fifty years sufficed to make these adequately known to the world at large.

It is difficult to make an anthology of such bulky works as church-cantatas, nor does an anthology meet the purpose where the whole work so constantly attains that excellence for which the anthologist seeks. Except for practical difficulties (as when Bach writes for obsolete instruments) the only reason why some cantatas are better known than others is that a beginning must be made somewhere. Indeed, a cantata was recently selected, on the ground of its popularity, for a choral competition in a small English country town the year before it was performed as a novelty in Berlin!

It is clear, then, that the influence of Bach's art as an understood whole is still undeveloped. In the past history of music his part was hardly suspected except by the great composers themselves; and, to any one contemplating the art of the generation after him, it might have seemed that both he and Handel had worked in vain. Yet his was the most subtle and universal force in the development of music, even when his musical language seemed hopelessly forgotten. Mozart, when rapidly advancing to the height of his mastery, had but to read the Baron von Swieten's manuscript copies of the motets and of the Wohltemperirtes Klavier, and his style, quite apart from his immediate essays in the old art-forms, and apart also from the influence of his study of Handel, developed a new polyphonic richness and depth of harmony which steadily increased until his untimely death. Beethoven studied all the accessible works of Bach profoundly, and frequently quoted them in his sketch-books, often with a direct bearing on his own works.

His rendering of the Wohltemperirtes Klavier is said to be recorded in the marks of expression and tempo given in Czerny's edition; and if that record is true, Beethoven must have been completely in the dark as to Bach's meaning in many important respects; but art is full of such illustrations of the way in which great minds influence each other in spite of every barrier which diversity of language and time can set. Beethoven's great Thirty-three Variations on a Waltz by Diabelli were actually described in the publisher's puff as worthy of their kinship with the "Goldberg Variations" of Bach; and that kinship is revealed in its truest light by a comparison between Beethoven's 31st variation and Bach's 25th; for here, just where the resemblance is most obvious, each composer utters his most intimate expression of feeling.

In the same way, Chopin is nowhere more characteristic than where he shows his love of the Wohltemperirtes Klavier in his études and Preludes; and so subtle is the influence of polyphonic style even over a writer so little apt to make direct use of it as Chopin, that one of Schumann's few plagiarisms occurs in his use of a phrase from Chopin's F minor étude (written for the Méthode des méthodes) as the subject of a fugue (Op. 72, No. 3). And, apart from fugues, which Schumann cultivated assiduously at a late stage in his career, the influence of Bach pervades the texture and rhythm of his work in more ways than can easily be followed.

In a more external, but not less significant way, the Passion according to St Matthew made its mark on Mendelssohn from the time when he discovered it at the age of twelve, and suggested to him many features in the general design of oratorios, by means of which he rescued that branch of art from the operatic influences that ruined Beethoven's Mount of Olives. Without the example of Bach, Wagner's schemes of Leitmotif would never in his lifetime have become woven into that close polyphonic texture which secures for his music a flow as continuous as that of drama itself: - and intimately connected with this is the whole subject of Wagner's harmonization, which in many of its boldest characteristics was foreshadowed by Bach. A close study of the texture of Brahms's work shows that he develops Bach's and Beethoven's artistic devices pari passu, and that the result is a complete unification of that opposition between polyphony and form which in the infancy of the sonata (as in every transitional stage in musical history) threatened to wreck the art as a false antithesis wrecks a philosophy.