|
12
dischi a 78 rèm - (p) 1953
|
|
1 CD -
CD 379 - (c) 2004 |
|
2 CDs -
SU 4213-2 - (p) 1953 (c) 2016 |
|
BRANDENBURG
CONCERTOS, BWV 1046-1051
|
|
|
|
|
|
|
|
Johann Sebastian
BACH (1685-1750) |
Concerto
No. 1 in F major, BWV 1046 |
|
23' 28" |
|
|
-
(without tempo indication)
|
4' 30" |
|
1-1 |
|
-
Adagio
|
3' 46" |
|
1-2 |
|
-
Allegro
|
5' 09" |
|
1-2 |
|
-
Menuetto · Trio I · Polacca · Trio II
|
9' 54" |
|
1-4 |
|
Concerto
No. 2 in F major, BWV 1047 |
|
12' 03" |
|
|
-
(without tempo indication) |
5' 16" |
|
1-5 |
|
-
Andante |
3' 41" |
|
1-6 |
|
-
Allegro assai
|
3' 00" |
|
1-7 |
|
Concerto
No. 3 in G major, BWV 1048 |
|
12' 46" |
|
|
-
(without tempo indication) |
6' 40" |
|
1-8 |
|
-
Allegro |
6' 07" |
|
1-9 |
|
Concerto
No. 4 in G major, BWV 1049 |
|
19' 02" |
|
|
-
Allegro |
8' 05" |
|
2-1 |
|
-
Andante |
4' 40" |
|
2-2 |
|
-
Presto |
6' 08" |
|
2-3 |
|
Concerto
No. 5 in D major, BWV 1050 |
|
22' 43" |
|
|
-
Allegro |
11' 09" |
|
2-4 |
|
-
Affettuoso |
5' 21" |
|
2-5 |
|
-
Allegro |
6' 03" |
|
2-6 |
|
Concerto
No. 6 in B flat major, BWV 1051 |
|
17' 19" |
|
|
-
(without tempo indication) |
6' 36" |
|
2-7 |
|
-
Adagio, ma non tanto |
4' 33" |
|
2-8 |
|
-
Allegro |
6' 02" |
|
2-9 |
|
|
|
|
|
Members of
the WIENER KAMMERORCHESTER and guests
- Edith Steinbauer, violin,
leader (solo in No. 4), viola (2nd viola
in No. 6)
- Alfred Altenburger, violin
- Alice Hoffelner (Harnoncourt), violin
- Eduard Melkus, viola (1st viola in No.
6)
- Nikolaus Harnoncourt, violoncello,
viola da gamba?
- Frida (Krause) Litschauer, violoncello
- Gustav Leonhardt, viola da gamba
- Elisabeth Schaeftlein, recorder
- Jürg Schaeftlein, recorder, oboe
- Camillo Wanausek, flute
- Helmut Wobisch, trumpet
- Bruno Seidlhofer, harpsichord (solo in
No. 5)
- and others
Josef MERTIN, conductor
|
|
|
|
|
|
Luogo
e data di registrazione |
|
Casino Baumgarten,
Linzer Strasse, Vienna (Austria) -
1950 |
|
|
Registrazione:
live / studio
|
|
studio |
|
|
Executive
producer |
|
Matouš Vlčinský
|
|
|
Sounf engineer |
|
Karl Wolleitner
|
|
|
Prima Edizione LP |
|
Supraphon |
23291/23302 - (12 dischi, 24
facciate, 78 rpm) | durata 108'
05" | (p) 1953 | Mono |
|
|
Edizione CD |
|
ORF "Alte Musik | CD
379 | 1 CD - durata 48' 43" | (p)
& (c) 2004 | AAD Mono | BWV
1048, 1049 & 1051
Supraphon | SU 4213-2 | 2 CDs -
durata 48' 41" - 59' 27" | (p)
1953 (c) 2016 | AAD Mono | BWV
1046-1051
|
|
|
Original Cover
|
|
-
|
|
|
Note |
|
The First
recording with period instruments.
Special thanks to Alice
Harnoncourt, Ingomar
Rainer and Robert Wolf
for the information on the
recording and the members of the
esnsemble.
|
|
|
|
|
On
Josef Mertin's
recordings of the
Brandenburg Concertos
(CDs Supraphon SU
4213-2)
When,
in 1950, post-war Europe,
whose political and cultural
scene had been mercilessly
divided by the Iron Curtain,
was commemorating the 200th
anniversary of the death of
Johann Sebastian Bach
(1685-1750), the festivities
were borne in the spirit of
numerous symbolic
connotations. They were
aimed at re-embracing the
most valuable assets of
conflict-debased German art,
as well as celebrating
Bach’s universally
comprehensible musical
language. The festivities
were particularly vigorous
in war-ravaged, divided
Germany and neighbouring
Austria, whose music centres
of Salzburg and Vienna saw
the Bach anniversary as am
opportunity to hold numerous
commemorative events and
concerts. The Bach jubilee
was not overlooked by the
Czechoslovak music publisher
Supraphon, which duly
implemented a project that
would have no parallel in
the country in the years to
follow. Making the best of
the composer’s anniversary,
the company utilised the
repertoire in the record
catalogue, the post-war
availability of fledgling
Viennese artists and its
contacts with the musical
circles in the Austrian
capital. In the 1950s.
Supraphon produced a host of
remarkable recordings,
primarily featuring core
Czech 19th- and 20th-century
music, with many of them
catching the attention of
critics and discerning
listeners abroad. From the
late 1950s, Supraphon’s
success was increased in
part owing to the engagement
of renowned foreign
conductors and
instrumentalists, including
those hailing from beyond
the Eastern bloc (John
Barbirolli, Jean Fournet,
Antonio Pedrotti, and
others). Nevertheless, at
the time, none of the state
publisher’s projects came
into being outside the
country, and without the
participation of
Czechoslovak artists, as had
been the case of the
recording of Bach’s
Brandenburg Concertos, BWV
1046-1051 (1718-1721), made
in Vienna in 1950.
The history of the album
started in the heart of the
Broumov Promontory in the
northeast of Bohemia by the
Czech-Polish (until 1945,
Czech-German) border. In
March 1904, Josef Mertin (b.
21.3.1904 Broumov, d.
16.2.1998 Vienna) was born
into a German family living
next to the Benedictine
Monastery. Following his
graduation from the local
grammar school, where he had
received thorough training
in singing, the violin,
piano and organ, and a brief
spell as a music teacher in
his remote native town
(1922-1925), in 1925 he
received a scholarship from
the company Benedikt
Schroll’s Sohn and moved to
Vienna in order to study
voice and sacred music at
the Wiener Musikakademie.
While in the Austrian
capital, in 1927 and 1928 he
formed a chamber orchestra
and passed exams in church
music and pedagogy, and in
1928 he graduated as
Kapellmeister from the Neues
Wiener Konservatorium.
Mertin concurrently attended
musicology seminars at the
Universität Wien. In 1928,
at the age of 24, he made
his debut with the Wiener
Kammerorchestervereinigung;
from 1932 he conducted Hans
Gál’s Madrigalchor
(1890-1987); and in 1933 he
founded his own instrumental
ensemble, Collegium musicum
Wiener Musikakademie.
At the end ofthe 1920s and
the beginning of the 1930s,
in addition to new
contemporary music and Bach
pieces (the Saint Matthew
Passion, on period
instruments), Mertin also
performed compositions by
Guillaume de Machaut (c.
1300-1377), whose
moderntiine premieres in
Vienna caused quite a stir,
as well as by his beloved
Heinrich Schütz. He taught
at the Kapellmeisterschule
and the Neues Wiener
Konservatorium (1928-1938),
the Wiener Volskhochschule
(1932-1938), and at the
Musikakademie (1937-1938).
In 1950 he left the
Konservatorium der Stadt
Wien so as to continue
teaching at the
Musikakademie (1946-1978)
and organ restoring (from
1931 he worked at the
Federal Monuments Office),
to carry out research into
the building of historical
musical instruments and put
together a collection of
early instruments at the
Kunsthistorisches Museum in
Vienna. Evidently the most
intriguing of his activities
were Mertin’s experimental
early music performances at
the Hofburgkapelle and the
Albertina gallery (from
1934, he held his own
concert series at the
Festsaal der Graphischen
Sammlung der Albertina). His
achievements in the domains
of music, education,
organisation. restoration
and collecting earned him
the title of Professor, the
Cross of Honour for Science
and Art (1960), the Gold
Medal of Merit for the
University of Music and
Performing Arts in Vienna
(1989), and the Silver Medal
of Merit of the Republic of
Austria (1994).
Before Josef Mertin died at
the age of 93, he could not
only look back fondly at his
long life filled with music
and pioneering work focused
on its early stylistic
periods, he was also able to
observe with pride the
progress of his numerous
pupils (Claudio Abbado,
Mariss Jansons, Zubin Mehta,
and others), many of whom
had been enticed by his
unconventional teaching
methods and imbued with a
passionate ardour for early
music. The New Testament’s
"For many are called, but
few are chosen" (Matthew
22:14) also applied to
Mertin’s students, the
majority of whom could only
put up with his not overly
systematic educational
methods for a few lessons.
Yet those who did remain
faithful to Mertin’s
apostolic verve and
rccondite pedagogic
techniques embraced
performance of early music
on period instruments and
copies in the post-war
decades so fiercely that
they almost condemned their
teacher's name to becoming a
mere encyclopaedia entry.
The most gifted of Mertin’s
pupils in the late 1940s
included the Austrian
violinist Eduard Melkus (b.
1.9.1928 in Baden an Wien),
who in 1946 assumed the post
of concert master of
Mertin’s Collegium musicum
and served his teacher as a
faithful and practical guide
through the vieissitudes of
the music scene in Vienna
(from 1951 to 1953, he
studied in Switzerland with
the Vienna-born violinist of
Czech origin Petr Rybář, a
friend of Bohuslav Martinů).
Melkus also followed in
Mertin’s footsteps by
founding early-music
ensembles, Schola antiqua
Wien (1952) and Capella
academica Wien (1965), and
finally, as a professor of
the violin, viola, Baroque
violin and historically
informed early-music
performance at the
Universität für Musik und
darstellende Kunst Wien
(1958-1996).
When in the autumn of 1950,
following years spent at the
Schola cantoruin basiliensis
in Basel (1947-1950), the
gifted Dutch organist and
harpsichordist Gustav
Leonhardt (1928-2012)
arrived in Vienna to study
musicology, he immediately
joined Mertin’s early-music
seminar attended by a number
of antagonistic talents.
Mertin’s students also
included the gifted recorder
player Elisabeth-Liesel
Schaeftlein (1927-1993), a
Graz native and sister of
Jürg Schaeftlein
(1929-1986), the legendary
oboist of Concentus musicus
Wien (1953). Probably in
1948, she introduced to
Mertin and his disciple
Melkus her gangly compatriot
Nikolaus Harnoncourt
(1929-2016), who from 1948
studied the cello with
Emanuel Brabec at the
Musikakademie. Had
Elisabeth-Liesel Schaeftlein
not done so, the young
Harnoncourt would most
likely have pursued the path
of a solo instrumentalist,
or “just” a player of the
Wiener Philharmoniker,
performing Dvořák’s and
Strauss’s music, instead of
becoming one of the major
figures of historically
informed early music
performance of the second
half of the 20th century.
Had it not been for
Elisabeth-Liesel
Schaeftlein, in 1950
Mertin’s student team,
extended for the sake of the
imminent recording of the
Brandenburg Concertos with a
group of players of the
Kammerorchester of the
Wiener
Konzerthausgesellschaft, who
for its time and in
comparison with other
Viennese orchestras had an
unusually high proportion of
female members, would have
had to do without the
cellist Harnoncourt.
The talented recorder
student Elisabeth-Liesel
Schaeftlein, the rising
violin star Eduard Melkus,
the hitherto unknown cellist
Nikolaus Harnoncourt and the
subtle Gustav Leonhardt, in
the company of members of
the Wiener Kammerorchester
(l 946), got together in the
studio to make under the
guidance of Melkus a
groundbreaking album of the
Brandenburg Concertos. The
project had been preceded by
the complete recordings made
by Alfred Cortot (1932,
Orchestre de l’Ecole Normale
de Musique) and Adolf Busch
(1936, Adolf Busch Chamber
Players), as well as
accounts of individual
pieces, including, for
instance, Wilhelm
Furtwängler’s live
recordings of Brandenburg
Concertos Nos. 3 and 5 with
the Wiener Philharrnoniker
at the Salzburger Festspiele
in 1950. Yet, some two
centuries after Bach’s
death, Josef Mertin decided
to take a revolutionary step
and perform the flagship
work - a "showcase of the
composer's instrumental
mastery" (N. Harnoncourt) -
in a chamber formation,
making use of the
instruments and applying the
performance canon of Bach’s
own time. "within the
Baroque concerto genre, the
concertos represent an
ultimate apex; with regard
to the instrumentation, they
are true chamber music,
unveiling their value in the
more intimate milieu in
which they were formerly
performed too. Your
gramophone recording (with
its most natural use being
for personal listening in a
private circle) thus
complies with the essential
trait of this music," Mertin
wrote to Supraphon after
listening to the recordings
that were being completed.
A number of the period
instmments employed in
Mertin`s recording of the
Brandenburg Concertos were
from his own collection,
which was also made use of
by the members of the Wiener
Gamben-Quartett (1949):
Melkus, Harnoncourt, Alfred
Altenburger (1927-2015) and
Alice Hoffelner (b.
26.9.1930), who would marry
Harnoncourt in 1953.
Instruments from the
collection were also used by
Gustav Leonhardt, who in
Mertin’s recording of the
Brandenburg Concertos played
the viola da gamba
(Brandenburg Concerto No. 6
in B flat major). By the
way, the collection and an
organ built by Mertin
himself (organo di legno)
were indispensable in the
making of the generally
better-known 1954 radio
recording by Paul Hindemith
of Claudio Monteverdi’s
opera L'Orfeo, performed by
Melkus and the oldest
generation of the then not
yet named historical
instruments ensemble
Concentus musicus Wien,
helmed by Harnoncourt.
In addition to the
minimalist configuration,
made up of students of
Mertin’s early music
performance class and the
members of die Wiener
Kammerorchester, headed by
the concert master Edith
Steinbauer (1901-1996) and
the cellist Frieda
Litschauer-Krause
(1903-1992), the wife of the
orchestra’s founder, another
natural facet of the
pioneering 1950 recording
was the adherence to the
original instrumental
structure of Bach`s Kothen
orchestra, including two
recorders in the fourth
concerto, which up until the
1970s were commonly replaced
with traverse flutes.
Specific information about
the instruments played in
the individual Brandenburg
Concertos has not been
preserved, nor has the date
on which the album was made.
Yet Mertin’s studio
recording is more than a
mere sonic document of a
revolutionary moment in the
history of performing early
music on modern and period
instruments. Compared to the
later projects of Viennese
provenience - Jascha
Horenstein’s, implemented in
September 1954, using
historical instruments
(performed by Nikolaus
Harnoncourt and members of
Concentus musicus Wien), and
the 1957 recording of Felix
Prohaska conducting the
members of the
Kammerorehester der Wiener
Staatsoper - Mertin’s
account stands out owing to
his endeavour for the utmost
sonic transparency and
precise leading of the
instrument parts. Mertin
also gave great thought to
the tempos. Even though
Horenstein opted for
markedly faster tempos.
Mertin’s account is
strikingly akin to the first
of the series of
Harnoncourt’s recordings of
the Brandenburg Concertos
(1964, 1981/1983, 1982).
Mertin’s dismissive attitude
to the romanticising
conception of Bach’s
orchestral concertos is
boldly audible in comparison
with Furtwängler’s 1950
album: Whereas Furtwängler
himself played the piano on
Brandenburg Concerto No. 5
in D major (29:23), Mertin
invited to perform on the
harpsichord the technically
superlative Bruno Seidlhofer
(1905-1982) (22:38).
Furtwängler turned Bach’s
work into a
Classicist-Romantic piano
concerto, while Mertin
returned to the dialogical
character of the Baroque
concerto.
"]osef Mertin was the father
of all the endeavours to
purge Romantic and Baroque
music of romantic deposits
and comfortable tradition,"
the conductor Milan
Turković, bassoonist of
Concentus musicus Wien,
wrote years later. And
bearing cogent witness to
this is even the oldest of
Mertin’s (precious few
preserved) studio
recordings, surprisingly
made by the Czechoslovak
label Supraphon. For the
first time since its lirst
release, on 12 shellac
discs, in 1953, Mertin’s
complete account (to whose
final recording the (self-)
critical Mertin took
exception and, following the
recording’s completion, he
even offered to make for
Supraphon new recordings,
this time only with
Collegium musicum) is now
being presented to listeners
on CD (in 2004, the year
marking the centenary of
Josef Mertin’s birth and the
75th birthday of Nikolaus
Harnoncourt, Austria’s ORF
radio station
released a selection of the
Brandenburg Concertos Nos.
3, 4 and 6). Thus, after an
interval of 66 years, music
lovers are offered Mertin’s
historically first recording
of the Brandenburg Concertos
in their entirety, as
performed by his ensemble on
modern and period
instruments. The project
serves to pay tribute to
Mertin's visionary approach
and express admiration for
his followers, headed by
Nikolaus Harnoncourt. The
present unique album is also
a proud reminder of their
Czech connections.
Martin
Jemelka
Translation:
Hilda Hearne
Major birthdays are often
very welcome affairs. In
this case, they provide a
suitable occasion to
celebrate two anniversaries,
Josef Mertin's 100th and
Nikolaus Harnoncourt 75th,
whose first gramophone
recording appears on the
present edition.
Of course I realise theat
Josef Mertin (one of the
most modest peole I have
known) would not have
considered this a real
reason to re-issue an
historic recording, least of
all one of his own.
Maybe his description of the
circumstances of this 1950
recording as a "scating over
thin ice" is somewhat
exaggerated, yet it does
represent a memorable step
in the early music revival.
It was prompted by a search
for musical authenticity in
the 1950 Bach year.
Matters that seem
self-evident to us today,
such as the use of recorders
in the 4th concerto (until
the 1970's the use of flutes
was still customary), the
two viola da ganbas in the
6th or the chamber.music
scoring of Bach's "Six
Concerts Avec plusieurs
Instruments" were real
pioneer events in 1950.
The appearance of Eduard
Melkus and of Gustav
Leonhardt, who taught
harpsichord in the 50's at
the Vienna Music Academy,
as gambist in the 6th
concerto only add to the
artistic value of the
production.
As with the restoration of
historic instruments,
greatest care was taken with
the production of this
re-issue. The goal was not
to reproduce the original
sound (almost impossible
anyway) but, in favour of a
wider sound spectrum, to
document the condition of
the original shellac discs
in 2004. Sound filters were
threfore used only seldom
and then extremely
sparingly, and it was
decided not to put movements
together (akthough this
would have been quite
possible) which had been
split up due to the limited
playing time of the discs.
I hope that this recording
from a time far.removed from
ours may not only remind us
that musical interpretations
should always be heard and
judged in the context of
their times, but far more
serve to commemorate a
full-blooded musician who,
until late in life,
tirelessly trained and
influenced several
generations of students
(including myself), a "homo
faber" archetype who
contributedvsignificantly to
the burgeoning of early
music, a warm-hearted,
caring, modest and very
humane person, Josef
Mertin!
Althofen, December 6th, 2004
Bernhard
Trebuch
|
|
On
Josef Mertin's
recordings of the
Brandenburg Concertos
(CD ORF 379)
This edition
consists of a transfer onto CD of
12 shellac 78 rpm discs made by
Supraphon in Prague containing
three of six Brandenburg concertos
by Johann Sebastian Bach. The
label on the discs attributes the
recording to "Members of the
Chamber Orchestra of Vienna"
conducted by Josef Mertin and
recorded in 1950. A few
explanations, amendments and
corrections are required
concerning both the time and
circumstances of the recording and
also those taking part.
The early music and authentic
sound movement in Austria in the
twentieth century is closely bound
up with the name of Josef Mertin
(1904-1998) who, until the sixties
and seventies, was considered one
of its most important instigators
and mentors. Born in Braunau in
Bohemia (now part of the Czech
Republic), Mertin arrived in
Vienna in the twenties to study
music. He completed his studies
within a very short period
(1925-28) with diplomas in church
music, voice and conducting from
the then State Academy and the
so-called New Conservatory. In
addition, he attended lectures on
musicology given by Guido Adler
and Rudolf von Ficker at the
university. His experience of
Ficker's combination of
scholarship and musical practice,
togheter with the knowledge and
skill he soon acquired in
instrument-making, did much to
mark Mertin's later work as
musician, teacher and maker.
Numerous concerts and church music
activities during the thirties in
the field of youth music-making
and organ playing (with the
associated rediscovery pf pre-Bach
music which saw the beginning of
Mertin's lifelong dedication to
Heinrich Schütz) were interspersed
with the first occasional attempts
to use historic instruments. This
was to be continued after the war
at the Collegium musicum of the
Vienna Music Academy, where Mertin
confronted an international body
og highly-qualified students with
questions concernin the
interpretation of early music,
inspiring them to their own
exploration, as he liked to call
it, which was to spread his
exemplary impetus throughout the
world.
These recordings date from this
period shortly after the war. An
(unfortunately undated) copy of a
letter (Mertin-Archiv, Universität
für Musik und darstellende Kunst
in Wien) from the professor to the
"management of the Supraphon
Record Company, Prague" provides
valuable documentation about the
circumstances as well as Mertin's
own particular views of the
recording, a "personal statement
and assessment of the
recentlz-completed recording of
Bachßs Brandenburg concertos".
Mertin assumes that the
necessarily "different nature",
even "apparent lack of uniformity"
in the sound world of each
pieces is inherent to their
differing instrumentation and
design (he discusses only
concertos III, IV and VI which he
apparently received as test
pressings) and thus found the very
different sounds of the individual
recordings acceptable. However
there were other serious problems,
also concerning the editing and
other technical aspects of the
recording, about which Mertin did
notwithhold his criticism: "The
test tape... sounded much more
faithful than the finisched disc.
In my opinion some important
elements in the sound have
disappeared in the cutting
(Frequency range relationships and
balance altered)."
The producer was Mertin's friend
and colleague Karl Wolleitner
(1919-2004) working, according to
our research, in the so-called
"Casino Baumgarten" on the Linzer
Strasse in Vienna's Penzing
district. The processing of the
material however took place
entirely at the Prague factory.
The criticism continue in concrete
detail: "The 3rd concerto has the
most satisfactory 'sound', since
the performers all belong to a
group used to playing together and
the concerto iteself presents the
least problems in terms of sound.
As a recording it is a technical
success, since the composition's
design can be clearly heard. The
concerto has only one dynamic
distorsion... the violins are
unreasonably favoured by their
closer position to the microphone.
The record is good."
"Recorders are added to the
strings in the 4th concerto. These
are real historic instruments, and
on top of that, in the hands of
wind players with particular
stylistic experience. This puts
the quasi-modern string sound at a
disadvantage, lending it an
unflatteringly penetrating
quality." However, all in all the
recording is "well-worth listening
to, and of a higher quality than
other records of this piece up
till now."
Matters start to worsen with the
assessment of the final concerto,
apparently recorded in winter
(February 1950?): "The recording
[of Concerto VI Ed.] suffered from
the heasting failure and contains
more faulty notes than acceptable
even under the circumstances."
(sic!)
In this context, Mertin addresses
a foundamental problem and
handicap to the whole production:
"The Wiener Kammerorchester' was
booked for the recording... this
orchestra is not an ensemble
specialised in early music,
although highly-regarded in Vienna
and working with care and
devotion. The orchestra semply
plays in the same standard and
exemplary way as the Philharmonic
etc. are used to playing. But they
are not early music 'specialiss',
and as a result certain stulistic
wishes cannot be fulfilled with
this ensemble. The 6th concerto
suffers particularly in this
respect... with these players... a
new recording would probably not
produce significantly better
results."
A possible alternative was
offered: "With my Collegium
musicusm as the Vienna State
Academy (where the recorders come
from!) I have built up an ensemble
that plays on actual historic
instruments." String instruments
restored to their original form
are meant here, subsequently
referred to as "short-necked
violins", which proved more
suitable to the demands, since "a
whole host of problems which
otherwise hinder the performance
of early music disappear with the
use of instruments in their
original state."
This stylistic approach of the
whole performance is very
reminiscent today of Paul
Hindemithìs surviving recordings
of his own works such as the
Concerto for Orchestra op.38 and
similar pieces fron the same
period. Mertin's Bach
interpretations also owe something
to the neo-baroque and new
objectivity in the result of his
effort to cleanse and
de-romanticise Bach, to reveal the
compositional steucture. Indeed
Hindemith and Mertin worked
closely together on the 1954
Vienna performance of Monteverdi's
"Orfeo" which so impressed the
young Harnoncourt and for which
Mertin provided a specially-made
"organo di legno".
Meanwhile, the stimulos and
occasions to become involved with
so-called period instruments in
Vienna came most of all from
Othmar Steinbauer (1895-1962),
himself a violin student of Sevcik
who, rejecting the excessive,
highly-individual romantic string
sound as understood by the Hauer
circle, preferred "old"
instruments (even including the
pseudo-Middle Age vielle
newly-designed from iconographic
models but with modern tuning in
fifrhs). The Vienna Gamba Quartet
also made its mark in this field
of activity during the 1950 Bach
year with a sensational
arrangement of the Art of Fugue
(including a completion of the
closing fugue by Eduard Melkus
which remains exemplary today).
Its four members, Alfred
Altenburger, Alice Hoffelner,
Eduard Melkus and Nikolaus
Harnoncourt, who were all closey
associated with Mertin's
Collegium, played on adapted viola
d'amore instruments tuned in
fifths, with the bass gamba being
the only instrument we would
consider historic today.
All this led to the suggestion in
a letter for a re-recording of the
sixth concerto with "new soloist:
1a viola: my best pupil at the
Academy with an original Quinton,
2nd viola: Prof. Steimbauer on an
original old Viennese master
viola; both instruments played
with historic viol bows! 1st
gamba, ny best Academy
gamba-player on an original
instrument using historic bowing
style. 2nd gamba: ditto: also an
outstanding pupil."
Further names included the cellist
Frieda Krause-Litschauer, Bruno
Seidlhofer on the harpsichord and
an unnamed double-bass player from
the Philharmonic, also with an
original instrument. Mertin
requested that the additional
recording sessions be organised
quickly: "The students with whom I
could make this
stylistically faithful recording
have already graduated and are
only available until May."
He mentions that one has got a job
in a "top-rank Swiss orchestra...
one gentleman is going back to
France, another to Holland". This
sets definite time limits as well
as giving some indications about
the partecipants in the recording.
Nikolaus Harnoncourt was among
them, wherther on the gamba as
mentioned above or wherther on the
cello instead of Frieda Litschauer
is no longer certain, as was
Gustav Leonhardt, an excellent
gamba-player in addition to being
a performer on historic keyboard
instruments. He had made his
Vienna debut in 1950 as
harpsichordist and taught at the
Music academy from 1952 before
taking up his position at the
Amsterdam conservatory in 1954. No
doubt he was the gentleman
referred to in Mertinìs letter who
must return to Holland. In the
same year Eduard Melkus ("my best
pupil at the Academy" almost
certainly refers to "the baroquest
violinist" according to
Hindemith's famous dictum) took up
a solo viola position in the
Zurich Tonhalle orchestra, the
"top-rank Swiss orchestra"
mentioned above. Thus the earliest
recording date for the 6th
concerto was to be in the spring
of 1954, and it includes the
"youthful work" of a few players
who were later to become some of
the best performers on the scene!
We may now reconstruct the
definitive list of those taking
part as follows: Edith Steinbauer
(1901-1996), leader as well as
soloist in no. 4 and second viola
in no.6. Eduard Melkus, viola,
also as soloist in no. 6. Nikolaus
Harnoncourt possibly as cellist in
no. 6, but perhaps on one of the
gambas. (That Harnoncourt and very
probably also his future wife,
Alice Hoffelner, were members of
the orchestra can also be
confirmed by the listing of the
complete recording as quasi opus 1
in his official discography.) The
stylistically expert recorder
players in the 4th concerto were
brother and sister Jürg
(1929-1986) and Elisabeth
(1930-1993) Schaeftlein, the
former soon to become a
long-serving member and leading
light of Concentus Musicus Wien as
baroque oboist. It goes without
saying that the recorders they
used also had nothing in common
with authentic instruments in the
strictest sense. Finally, Bruno
Seidlhofer (1905-1982), later a
renowned piano teacher, is listed
as playing harpsicord continuo.
"The ensemble mentioned here would
also have been better for the
remaining 5th concerto..."
During conversations in later
years Mertin candidly described a
recording of the 5th Brandenburg
(with Bruno Seidlhofer as
soloist), clearly made at the same
time, as "entirely unusuable"; it
never gor beyond the test
pressing, which is also why we
have chosen to ignore it in the
context of this re-issue.
Apparently Hindemith attempted to
direct a recording of the second
concerto with the same team (with
Helmut Wonisch playing trumpet
alongside Elisabeth Schaeftlein on
the recorder). A recording of the
first concerto never seems to have
been attempted.
In an introductory text to a
production of all six concertos
(and thus not directly for this
edition) containing much other
useful information Josef Mertin
also expresses an interesting
thought about his own, carefully
considered relationship to the
recording medium: "The concertos
represent an absolute pinnacle of
achievement in the genre of the
baroque concerto; their
instrumentation in like true
chamber music, whose value is best
revealed in intimate surroundings
such as those in which they were
forst performed. Hnece their
appearance on record (assuming the
most natural use of the record for
personal purposes in intimate
surroundings) corresponds to an
important characteristic of
this music."
May this commemorative re-issue of
his production of the Brandenburg
concertos be granted a suitable
affectionate treatment "for
personal purposes in intimate
surroundings"!
Ingomar Raine
|
|
|
|