|
1 CD -
4509-90842-2 - (p) 1994
|
|
Wolfgang
Amadeus Mozart (1756-1791) |
|
|
|
|
|
|
|
Serenade No. 4 in D major |
|
|
|
Marcia, KV
237 (189c) |
|
3' 49" |
1
|
Serenade,
KV 203 (189b) |
|
46' 40" |
|
- Andante
maestoso |
8' 56" |
|
2
|
- Andante |
5' 21" |
|
3
|
- Menuetto
- Trio |
3' 55" |
|
4
|
- Allegro |
5' 18" |
|
5
|
- Menuetto - Trio |
3' 54" |
|
6
|
- Andante |
9' 28" |
|
7
|
- Menuetto
- Trio
|
4' 27" |
|
8
|
- Prestissimo - Coda |
5' 21" |
|
9
|
Symphony No. 23 in D
major, KV 181 (162b) |
|
9' 31" |
|
- Allegro spiritoso
- Andantino grazioso - Presto assai |
9' 31" |
|
10
|
|
|
|
|
CONCENTUS MUSICUS
WIEN (mit
Originalinstrumenten)
|
|
-
Erich Höbarth, Violine
|
-
Dorle Sommer, Viola |
|
-
Alice Harnoncourt, Violine |
-
Gerold Klaus, Viola |
|
-
Anita Mitterer, Violine |
-
Ursula Kortschak, Viola |
|
-
Andrea Bischof, Violine |
-
Herwig Tachezi, Violoncello |
|
-
Helmut Mitter, Violine |
-
Max Engel, Violoncello |
|
-
Peter Schoberwalter, Violine |
-
Eduard Hruza, Violone |
|
-
Karl Höffinger, Violine |
-
Andrew Ackerman, Violone |
|
-
Maighread McCrann, Violine |
-
Robert Wolf, Querflöte |
|
-
Christina Busch, Violine |
-
Reinhard Czasch, Querflöte |
|
-
Editha Fetz, Violine |
-
Hans Peter Westermann, Oboe |
|
-
Maria Kubizek, Violine |
-
Marie Wolf, Oboe |
|
-
Irene Troi, Violine |
-
Christian Beuse, Fagott |
|
-
Herlinde Schaller, Violine |
-
Andrew Joy, Naturhorn |
|
-
Christian Tachezi, Violine |
-
Rainer Jurkiewiez, Naturhorn |
|
-
Thomas Feodoroff, Violine |
-
Andreas Lackner, Naturtrompete |
|
-
Peter Schoberwalter junior, Violine |
-
Martin Rabl, Naturtrompete |
|
-
Lynn Pascher, Viola |
-
Dieter Seiler, Pauken |
|
|
|
Nikolaus
Harnoncourt, Leitung |
|
Luogo
e data di registrazione
|
Casino
Zögernitz, Vienna (Austria) - dicembre 1992
|
Registrazione
live / studio
|
studio |
Producer
/ Engineer
|
Wolfgang
Mohr / Helmut Mühle / Michael Brammann /
Stefan Witzel
|
Prima Edizione CD
|
Teldec
"Das Alte Werk" - 4509-90842-2 - (1 cd)
- 60' 37" - (p) 1994 - DDD |
Prima
Edizione LP
|
-
|
|
Notes
|
It was not until 1831
that Friedrich Rochlitz, writing in
the Allgemeine Musikalische Zeitung,
first signalled the existence of what
he described as "three volumes of
original manuscripts by
W. A. Mozart"
in the possession of the Hamburg
music-publisher August Cranz. All
three volumes were small in size and
bound in blue-grey
cloth. The first of them contained
nine symphonies, the second a
concertone (K. 190) and
three serenades (K. 203, 204 and 250),
and the third only a single serenade
(K. 185) and a march (K.
189). Although the dates
entered at the start of each work have
all been rendered illegible by an
unknown hand, we know that it was only
after his return from his third visit
to Italy, in March 1773, that Mozart
began using this small landscape
format. In other words, the works in question
must have been written either in 1773
or later and were presumably, for the
most part, the result of commissions
from ltalian, Milanese
or Lombardic
patrons.
The D major Symphony K. 181 is written
in the manner of an italian-style
overture and might have been intended
for an opera, an idea that receives
support not only from the fact that
the three movements are joined
together without a break
but more especially from the musical
language of the work, a language
which, to a degree that even today
continues to elicit amazement, seems
to suggest a kind of action of
altogether theatrical immediacy. It
is surely significant in this context
that the not yet seventeen-year-old
composer had been in Milan at the end
of 1772 to prepare for the first
performance of his opera Lucio
Silla. Equally "theatrical" are
the instrumental forces deployed, with
two oboes, two horns
and two trumpets in addition to the
usual complement of strings.
Motivically, tonally and dynamically,
too, the listener is inevitably
reminded of an opera when
following the three movements of this
overture-like
symphony. Or was Mozart simply in an
impish mood? Was he wanting, above all
else, to set his listeners thinking?
Perhaps there is irony waiting to be
discovered here, too, an irony that
has ensured that so idiosyncratically
distinctive a symphony has been
practically ignored in the world of
music.
Certainly, one cannot help being
struck by the way that tremolo
passages in the strings over triadic
figures in the bass can create the
feeling of chords, that G minor chords
and the interplay between major and
minor still have the power to shock,
and that there is a notable absence of
any thematic writing in the sense of a
clear melodic line. The diminished
seventh makes as though to announce
something that subsequently fails to
materialize. What we hear is a number
of syncopated effects, octave leaps
and imitative entries creating a sense
of momentum, but nothing else. Nor is
there a ”second” subject, lout,
instead, a dialogue between strings
and wind, a feature that attests to
Mozart's unique ability to operate
beyond strict rules, to avoid all that
is expected and, as the composer
Wolfgang Rihm once said of him, ”never
to do anything really ’properly':
everything is somehow wrong, which is
why this music still lives for us
today".
It was for precisely this reason,
however, that Mozart's music was able
to give rise to misunderstandings,
even in 1826. Thus, for example, we
read in Hans Georg Nägeli's Vorlesungen über Musik mit Berücksichtigung
des Dillettanten (Lectures on Music
with Due Regard to the Amateur)
of this year: "Tender
melodies frequently alternate with
harsh, abrasive sounds, charm of
movement with violence. Great was Mozart’s
genius, but equally great was his
erroneous wish to create his effects
through contrast, a failing typical of
genius in general and all the more
deplorable in that he constantly
contrasted the non-instrumental with
the instrumental, cantabile writing
with a freer style. It
was inartistic, as it is in all the
arts when something has to achieve its
effect through its polar opposite. It
had a distorting effect, first and
foremost on himself, because as soon
as perpetual contrast is raised to the
level of a general principle, one
loses sight of the beautiful sense of
proportion that exists between the
individual parts of any work of art."
Mozart was clearly
experimenting in the opening movement
of K. 181, just as, thanks to his use
of a characteristic timbral device
involving a trill, he hesitates,
rhythmically and gesturally, between a
"quick march” and what Saint-Foix
termed a "quick step” in the final
movement. In the middle movement he
gives the solo oboe an opportunity to
perform a regular operatic aria.
Timbre is briefly emphasized in the
final movement, when Mozart suddenly
demands divided violas for eight bars.
Like the D major Symphony K.
181, the Serenade in D major K.
203 has clear operatic associations,
as Alfred Einstein has already pointed
out: ”The built-in violin concerto,
this time in B flat major and in three
instead of only two movements, is now
fully formed, a true work within a
work, not a mere episode.
One thinks involuntarily of a similar
process in the history of opera: the
inclusion of an opera buffa or
intermezzo between the acts of an opera
seria. In the opening movement
of the concerto there is another
remarkable pointer to the future: the
contrast between the main melody and
the garrulous interjections on oboes
and violas, and that between Donna
Anna and Don Ottavio on the one hand
and Donna Elvira and Don Giovanni on
the other in the B flat major quartet
- even the key is the same.” In short,
we have further proof here of the
theatrical aspect of Mozart's
thinking as an instrumental composer.
Mozart added a
manuscript note to the autograph of
the Serenade: "Serenata del Sgr.
Wolfgango Amadeo Mozart, nel mese
d’agosto 1774”. Here, too, the year
was later deleted, although there is
no reason to doubt the accuracy of the
attribution. Mozart's
first biographer, Franz Xaver
Niemetschek, argued that it was
written for the name day of the
Prince-Archbishop of Salzburg
Hieronymus Colloredo but although the
work continues to this day to bear the
nickname ”Colloredo Serenade”, it
seems unlikely that the work was in
fact written on this occasion. To
judge by the seriousness of its
extended movements and the strictness
of the thematico-motivic
writing, one is tempted to think that
Mozart must have had a more important
reason for composing it. Here, too,
there are contrasts of a dramatic
nature. A weighty slow introductory
section of symphonic significance is
contrasted with sparkling fast
movements unclouded
in character. An Andante reveals
itself as another evocation of an
operatic aria, but we also hear
distinctively Mozartian rninuet-like
rhythms and fleeting notturno-like
musings. Since Mozart was an
outstandingly good violinist, it seems
likely that it was with himself in
mind that he wrote the violin concerto
incorporated in this serenade.
At the time that
Mozart wrote the D major Serenade, it
was usual practice, at least in the
case of open-air performances, for the
players to march into position and,
at the end of the performance, to
march away again. Mozart
is believed to have written such a
march to introduce and round off the D
major Serenade, too. Since there are
thematic links
between serenade and march, it has
been assumed that both works were
intended as a single entity, but the
fact that the march is not scored for
violas and that it demands two bassoons,
instead of the single bassoon required
in the serenade, might
tend to favour the argument of those
scholars who insist that the march is
not, after all, part of the serenade.
Be that as it may, the march might
still function as a prologue and
epilogue and, given its compositional
density, would also
ensure the right note of
portentousness.
Wolf-Eberhard
von Lewinski
Translation:
Stewart Spencer
|
|
Nikolaus
Harnoncourt (1929-2016)
|
|
|
|