This is in reply to what John Dalmas wrote several months ago (my apologies
for my late comments, I just got my own internet account and e-mail address
a couple of weeks ago) about Josef Hofmann's recording of the Chopin G
minor ballade:

   'Yes, and it should demonstrate to you how NOT to play the Ballade.
   IMHO, Hofmann's mannerisms and distortions misuse the music to show
   off his virtuosity, much as Horowitz was to do later on.  And as it
   was said of Horowitz, "he can sure play the piano, but I am not so
   sure he can play Chopin, et al."'

 From being myself very sceptical for a long time after first hearing
Hofmann's recordings, I have gradually become a very great admirer of his.
Of course many of his electrical recordings have a great many flaws, as
they caught him at a time when not only his technical abilities, but his
musical concentration was progressively deteriorating.  However, I very
much disagree with your attacks on the ethos behind his playing.  I have
come to think that his interpretations can be very satisfying, provided
that you listen to them correctly, so to speak.  Here are a few points
which may make Hofmann's interpretation more convincing:

What you consider distortion will depend on what you think are the
essential elements of a piece.  Many musicians today believe that the
essence of a composer's inspiration extends down to all his performance
markings, and to deviate from them amounts to a recomposition.  Hofmann
on the other hand thought (and I now happen to agree with him) that one
must distinguish between those aspects of a composition whose meaning is
unequivocal, which clearly express the composer's overall intention, and
those many details which can be seen from many different angles without
disturbing the whole.  Taken in this light, much of the composer's markings
represent, in a sense, his own interpretation of the music - certainly not
the whole truth.

The question to ask if we want to do Hofmann justice from his point of
view is therefore, is Chopin's overall intention satisfied, does the
piece make sense as a whole? I believe that when he deviates from Chopin's
prescriptions, H nevertheless understands the structure of the ballade very
convincingly.  For example, when he reaches the first climax at the end of
the exposition of the first theme, he takes the level down to a kind of
ominous muttering, with the 'trumpet calls' of the left hand full of
distant portent.  The mood makes sense, and H leaves himself room to build
more excitement when it is needed later on.  The first ff re-appearance
of the second theme is taken as an impassioned outburst, which the 'piu
animato' section intensifies (you may not like the excessively nervous
speed and heavy accents here, but they are surely appropriate in the
context).  The end of the 'piu animato' comes as the second climax, after
which H.  again very much takes down the level of the second re-appearance
of the second theme (also marked ff in the score).  While one can argue
about the degree of his deviation from the ff, it basically makes sense
in coming after such ferocious intensity and given the new, flowing
accompaniment.  The atmosphere which H.  very movingly creates is of an
exhausted frenzy, of the energy contained in the previous development of
the second theme (representing the yearning vision of a free Poland?) being
completely drained by the 'catastrophe' of the climax of the piu animato.
This marked relaxation of tension also makes sense in terms of the end:
because of it, the fury of the coda is all the more shocking and tragic.
Compare this to, for example, the much admired recording by Murray Perahia,
who takes the re-appearance of the second theme after the 'piu animato' ff
as marked and can then no longer trump it in his coda, which seems like an
afterthought (ditto for Horowitz).

So much for the overall structure.  What about the details? I agree that
by this stage, Hofmann's phrasing had begun to adopt a mannered, 'dotted'
quality.  However, apart from that, his rubato is in fact very precisely
applied.  Onto a base of almost no rubato at all, he applies a kind of
'tenuto' at strategic points where he wants a particular emotional effect.
What I suspect you found particularly artificial were H's voicings and tone
colourings, and his chordal amplifications.  This is where I have found
you really do need to listen differently.  H's voicing seem arbitrary in
contrapunctal terms, but not if they are heard as a coloristic effect, as
'orchestration'.  H emphasises a secondary voice not because he thinks it
is particularly important in itself, but as a 'commentary' on the quieter
main line, which can still be heard, the 'inner voices' surging up below,
giving it greater intensity.  Similarly, H's ever changing tone colourings
are not merely sensual, as are Horowitz's, but express ever changing
emotional states.  To me, H's ability to discover rapidly shifting moods
within overall unity is one of the great pleasures of his playing.  Listen
to the exposition of the main theme:  to me, the improvisatory, ambiguous
nature of H's declamation at this point does far more justice to it than
perfectly balanced phrases would do.  Listen, for example, to the sudden
bright flash of fioriture in this section:  this may be crowd-pleasing
virtuosity, but it is also a short-lived smile in a mood of overall
despondency.  You need to listen with your ears - reacting to the total
aural picture - rather than with your eyes, as it were, checking whether
the playing is an accurate representation of how the notes look on the
page.  I think this is also the reason why H was so loath to record:
a slight deviation from the intended tone colour, or lack of tonal
perspective and dynamic range - and the musical effect is ruined.

The main point is that while H's playing was undoubtedly very free,
indeed improvisatory, and not all of his ideas may have worked completely,
especially not at this stage of his career - although in the case of the
Ballade I think they generally do - this freedom had very solid structural
bounds, and it was a freedom which involved the exploration of the
emotional potential of the music, not blind virtuosity (at least not
when H was still alert and 'on form').

Please excuse this long rant - how to properly evaluate Hofmann's playing
is a difficult problem, given the widely varying quality of his recordings,
which has gone around in my own head for some time.

An afterthought to John Dalmas on Hofmann:  if you want to hear him in
even better vein than in the G minor ballade, listen to his f minor Chopin
concerto and especially the small fragment of the e minor concerto from
1935 on The Complete JH Vol 1, the Moonlight Sonata from 1936, where a few
details in the allegretto don't work, but the effect is exactly the right
one of a graceful vulnerability between two brooding chasms; some of the
numbers, especially the last three, of his 1938 Kreisleriana, a musical
and technical wreck, but a marvellous wreck containing moments of great
eloquence and poetry; ditto for his 1938 4th ballade, of which the last few
minutes have an extraordinary power and vision.  By the way, if you listen
to his earlier, acoustic recordings, you will find that here the playing
does indeed make sense 'on the page' as well as aurally, but they are
rather cooler than his live performances.

Felix Delbruck
[log in to unmask]