This is String Quartet nr. 18 by Gaetano Donizetti with some
corrections of mine.  This edition originated from a derived version
of the original score I used to create MIDI files, which corrected
some mistakes in Donizetti's autograph.  I slowly improved this new
score by adding some missing articulations and correcting other subtle
errors.

An interesting feature of this edition which makes it -- I think --
rather unique, is the presence of the original autograph together with
the edited score.  In this way everybody can compare the derived
version with the music the composer actually wrote.

The rules I followed in preparing this edition are listed here:


1. Percent repeats and tremoli are always avoided: notes are fully
   written in the staves.  It might be useful for those who study
   scores by playing them on the piano.


2. Full measure rests are used when possible (e.g. fourth movement,
   bars 89-91).


3. Tuplet numbers and cautionary accidentals are added to improve the
   readability of the text (note that the autograph does not use
   tuplet numbers at all, with the exception of the second movement,
   bars 112-114, Violin I part).


4. Articulations and dynamics which are written once in the four
   staves are duplicated in the other three parts only when such a
   correction is obvious.

   Example: In the first movement, bar 90 has a crescendo in the first
            and the last staff which is obviously to be performed by
            the other two parts as well.


5. Bars with wrong durations are always corrected.  Of course these
   passages can sometimes be interpreted in more than one way, and
   some additional criterion must be taken into account.

   Examples:

   a. In the second movement, bar 70, cello plays a quarter note (Cis)
      which is obviously an eighth one (see bar 69).  In this case no
      ambiguities arise.

   b. Bars 110-111 of the same movement raise a difficult question.
      Since Violin I was the only part with wrong durations, my first
      tought was to halve every note in the first staff.  But it is
      clear that Violin I is playing here the same rythmic pattern
      seen in bars 106 ff., so it is better to keep Violin I unchanged
      and correct the other three staves.


6. Dissonances that arise by probably misspelled alterations are
   corrected.  No corrections are made in those cases where the
   dissonance could have some "raison d'être".

   Example: Fourth movement, bar 175, Violin I part.  Here there is an
            Ais which is dissonant with the Viola part.  Note however
            that Donizetti wrote the sharp sign before this A and he
            placed a natural sign before the A which follows shortly.
            In my opinion this is a strong evidence that Ais should be
            played here.


7. Some passages sound uncertain but are harmonically correct.  I
   chose not to change them, unless some precise evidence of a mistake
   could be found.

   Examples:

   a. In the first movement (bars 27-28) cello plays the sequence
      A-E-C (ignoring the grace notes), in contrast with the previous
      bars, where the last note is always equal to the first (also
      note that bars 21-22 have A-E-A).  The C however avoids an
      unison between Viola and Cello and Violin I, so there is the
      doubt that the mistake -- if there is any mistake at all -- is
      in bar 22.  Therefore I chose to left this passage unchanged.

   b. In the same movement (bar 132) Violin I plays a dorian scale.
      This is probably a mistake, since bar 129 has a similar passage
      with a melodic minor scale.  Moreover, in the first half of the
      XIX century church modes were not widely used to my knowledge.
      In this case I turned G into Gis.


8. The Trio (third movement) has a wrong repetition mark at its end,
   since the final closing repeat (bar 162) has no corresponding
   opening repeat.  In bar 162 Violin I is playing a scale which leads
   naturally to Dis, so I chose to start it from bar 131.  This
   solution has the advantage that the scale in bar 162 is matched by
   the scale in bar 132.


7. The .ly source files have comments beginning with [E].  These
   comments indicate the passages I chose to correct.  A [R] means
   that there is something doubtful there, but I did not correct it.


Maurizio Tomasi, July 2003
