I have taken myself time to sit down and compare three recordings of
Liszt's third year of Pilgrimage.  This comparison was motivated by the
rather poor Hyperion-recording by Howard.  Unfortunately Howard's recording
is on the (very) fast side, and this does very much damage to the first
three pieces in particular.

I believe that Kocsis and Berman are the benchmark recordings, and I do
not question their status as reference recordings, in fact they complement
each other.

1.  Angelus

A musical prayer which was inspired by some church bells Liszt heard
from Villa d'Este.  Like so many late Liszt pieces, there is only one
climax which the pianist should aim for.  Exquisite playing from Berman,
but he almost loses momentum because of his slow tempo.  At the climax,
he doesn't playing anything louder than forte.  Kocsis is less magical,
but he is not afraid of fortissimo and his tempo is on average 20% faster
than Berman's.

Here I should also mention the beatiful version for string orchestra
which has been recorded on Hungaroton.

2 and 3.  Threnodie I and II (the Cypresses of Villa d'Este).

Liszt thought that the word 'elegy' was too tender in this context, and
used 'threnody' instead.  Clearly, they are both dark pieces, but there
are also shimmers of hope occasionally (especially in the latter piece).

Again Berman is on the slow and quiet side, and I think he fails to reach
the pain and agony which is inherent in this music.

4.  Les jeux d'eaux a'la (the water fountains of) Villa d'Este

The first impressionistic water piece in history.  The key is F sharp
major, Liszt's major religious key, and the religious connection is
further emphasized by a quotation appearing in the middle of the score.
Howard is again the fastest pianist, but the music does not sound hurried.
All three pianists are quite successful, but Berman takes some liberties
with the score in the coda.

5 and 6.  Sunt lacrimae rerum and Marche funebre

Two more Threnodies/Elegies, if not by name (The original title of Sunt
lacrimae rerum was Hungarian Threnody).  According to Leslie Howard,
Sunt lacrimae rerum is a lament over those who died in Hungarian War of
Independence, whereas Marche Funebre commemorates Emperor Maximillian I
of Mexico, who was executed in 1867.

The introduction to Sunt lacrimae rerum resembles the introduction to
the Faust symphony in that both uses descending Gypsy scales.  In the
final bars Fsharp major chords loses out to A major chords.  The March
funebre moves within a narrow melodic interval, but the coda reaches a
more triumphal conclusion in Fsharp major.  Kocsis is superior in both
pieces, although Howard's later recording of the first version of the
March funebre is very good.  Berman is unfortunately on the slow side
in the Marche Funebre.

7.  Sursum corda (lift up you hearts)

Another religious piece, but this time in E major, which is the other
major religious key of Liszt (both of his oratorios ends in E major).
Again Berman spoils his otherwise excellent performance by chosing a
slow tempo.

To sum up, there is nothing wrong with Berman's recording, I only wish
he could follow the score (and the tempo indications in particular) more
closely.

Kocsis appears to be the more intelligent pianist in that context,
and he also manages to find the more 'Bartokian' aspects of late Liszt,
wherease Berman is more 'Chopinesque'.  It is of course a matter of
taste, but to me Liszt has more interesting things to tell through Kocsis
than through Berman.

Mikael Rasmusson <[log in to unmask]>