CHASE

Ferdinand David's revised printed markings and his additional crayon markings in this sonata provide a fascinating example of his propensity constantly to revisit and revise his approach to repertoire that he regularly performed. He played the Kreutzer Sonata throughout his career. It is reported that he and Mendelssohn performed it from memory together, yet he was clearly still modifying his approach to it more than twenty years later.

This and the down-bow sign on the next line, although in pencil, may possibly be David's, they are a similar size and shape as his other down-bow markings (see p. 3).

Not in 4899b.

Not in 4899b.

Not in 4899b.

Not in 4899.

Not in 4899.

The printed Π and V signs are not in 4899b. They were all deleted by David here in connection with the change of slurring in the 2 bars following the fermata, to facilitate the arrival at a down bow in preparation for the slurred staccato execution in the 6th to 8th bars. In 4899b slurred staccato was printed only in the last bar of this passage.

In 4899b, there is a Π and no o. The first of David's blue crayon markings in this copy is to emphasise the V; it was typical of him to duplicate printed markings in crayon, perhaps because of deteriorating eyesight in later years.

There is no o in 4899b. Here and in the following bars David typically breaks up long slurs into smaller units.

A 2 is marked here in 4899b. This version suggests either that he wanted to play the f#' on the G-string, that he regarded the shift to 1st position as obvious, or that he simply omitted a 2 through oversight.

Not in 4899b, where 1 is marked on the following c'''.

In 4899b, ii-iii in this passage are slurred and the rest staccato. David's crayon marking indicates a more complete legato.

There is a printed 3 here in 4899. The purpose of removing it is unclear.

Here David has changed notes, apparently to avoid the awkward diminshed 5th!

In 4899b the 1 is absent. The change to 5th position occurrs with a 2 on the d'''.

In 4899b there are staccato dots here; perhaps David expected a broader execution than these might imply.

David's added line evidently indicates restez in 3rd position rather than keep the 2nd finger down, since it has to play both b flat and e natural.

Not in 4899b.

Not in 4899.

A printed 1, is present in 4899b because there is no return to 1st position three bars earlier.

Not in 4899b.

Not in 4899b.

Not in 4899b. The different bowing in the previous bar is necessary to facilitate the V here.

With staccato marks in 4899b.

With staccato marks in 4899b.

Not in 4899b.

Printed 1 here in 4899b.

Printed 3 on e''' and 2 on b''; no 4 on d''' in 4899b. Here David evidently intended to remain in 7th position for the beginning of this entry and to change to 3rd position on the d'''.

David wrote a crayon 4 under i, but then crossed out both it and the printed 3.

This note is not true to the original text, which has e''' here.

Staccato marks in 4899b.

Staccato marks in 4899b.

Staccato in 4899b.

In 4899b there is a 2 on the c'', indicating a portamento on the A-string, but here David perhaps envisaged remaining in 3rd position, perhaps with a harmonic on the d''; of perhaps a 2 was simply obvious or omitted in error.

David's addition here and in other equivalent places, indicates performance on the A-string, with a slide of the 4th finger from bb'' to a harmonic on a''; this is then followed by a slur incorporating two separate e''s, which are evidently to be performed by a shift from 1st to 2nd finger within the slur; the following trill is almost certainly expected to end with a turn.

Fingered 4 in 4899b. At this stage David apparently wanted to obtain a simpler rendition of the theme here, reserving the portamento until the 7th bar of the stave.

3 in 4899b.

Fingered 2 in 4899b (see next note).

Not in 4899b, where the reviser of David's original text marks 3 on the following grace note.

Not in 4899b, where 3 is marked on the following grace note.

A triplet bracket is present in 4899b. It was omitted here, perhaps to avoid the possibility that it might be mistaken for a portato indication.

Printed 1 in 4899b.

Fingering not in 4899b, where the g#'' is fingered 2.

It is possible that David's new bowing was intended only for the repeat, for the sake of variety.

David's blue crayon 1 has been deleted in the same pencil as the 1 on the preceding note.

David's blue crayon 1 has been deleted in the same pencil as the 1 on the preceding note.

This appears to be a 1 rather than a deletion, probably indicating that David intended to stay in 3rd position with a backwards extension for the g#''.

The slur is probably a reminder that for repeat begins with another up-bow.

A later pencil deletion obscures David's blue crayon emphasising of the printed fingering 3 2 here.

These pencil fingerings and the deletion of the printed 1s on d''', f''' and a''' (which are not in 4899b) are not in Ferdinand David's hand.

It seems that, on at least one occasion, David decided to omit the repeats in this variation. Similar abbreviations for particular performances can be seen in some of his other personal copies. His copy of Cherubini's E flat String Quartet provides several instances.

Fingered 2 in 4899b. The fingering here was probably omitted in error, since an otherwise unnecessary 3rd position fingering is marked in the following bar.

David's blue 2 is partly obscured by a later pencil fingering. David's combination of a slur and the changed fingering (he deleted the 3 at the beginning of the stave) allows him to change fingers on the repeated notes without stopping the bow - a highly favoured effect in the music of members of the Viotti School and his own teacher, Spohr.

David's added cres.

David originally changed this to two tenuto lines, then replaced them with a slur.

David's f.

David's added crescendo.

With staccato marks in 4899b.

Fingered 3 2 in 4899b.

With staccato marks in 4899b.

With staccato marks in 4899b.

With staccato marks on the quavers in 4899b.

Staccato in 4899b.

With staccato marks in 4899b.

Staccato in 4899b.

With staccato marks in 4899b.

With staccato marks in 4899b.

With staccato marks in 4899b.

No 2 on i in 4899b, but 3 on iii. That indicates performance in 1st postiiton with an extension, whereas this fingering indicates 2nd position until IV/2/i.

Not in 4899b.

With staccato marks in 4899b.

Perhaps this line is an indication to use portamento between the bowstrokes when changing position.

Fingering on i and ii not in 4899b, where i is fingered 2.

Fingering on i and ii not in 4899b, where i is fingered 2.

Fingering not in 4899b.

Not in 4899b.

Bars numbered 1 to 4 (to count the repetitions) in 4899b.

The 2 is not in 4899b, which has 4 3 on v and vi.

Not in 4899b.

Not in 4899b.

Not in 4899b.

With staccato marks in 4899b.

This is another characteristic David procedure, employing the harmonic on the G-string to assist in articulating the g' on i and ii.

The 2 on ii and the fingerings vii-ix not in 4899b, where the passage is marked to be played in 1st position with extensions of the 4th finger for the e'.

Fingering not in 4899b.

Not in 4899b, which has 2 on i.

Not in 4899b, which has 2 on i.

David crosses to the A-string for the harmonic thus avoiding a change of position.

With staccato marks in 4899b.

Upper slurs not in 4899b.

Not in 4899b.

With staccato marks in 4899b.

With staccato marks in 4899b.

With staccato marks in 4899b.

With staccato marks in 4899b.

Not in 4899b.