CARL ORFFS

CARMINA
BURANA
An Analysis of the Works Performance
Practice on Commercial Recordings
Jonathan Babcock

Jonathan Babcock is the associate director of

choral activities at Texas State University-San
Marcos. He holds a DMA in choral conducting
from the Hartt School, University of Hartford
and an MM and BM in music education from
the Crane School of Music, SUNY Potsdam.
[email protected]

48

CHORAL JOURNAL Volume 51 Number 4

Carl Orffs Carmina Burana is

one of the most frequently performed and recorded twentieth
- century works in the choral
repertoire. Nearly every major
orchestral conductor has recorded the work at least once: Seiji
Ozawa has five different recordings to his credit. The recording
technology has created a great
resource for musicians: it has
documented performance practice and performance tradition
dating back to the early 1900s.
In his book, Early Recordings and Musical Style, Robert
Philip states that [R]ecordings
shed light on the limitations of
documentary evidence in any
period.1 Although the time span
of Carmina Burana recordings is
not extensive, there is still a great
deal of interpretive information
to be gained in an analysis of
these sound documents. Sixteen
separate recordings were chosen
for this study.2

CARL ORFF'S CARMINA BURANA

The conductors represented are considered to be among the most esteemed of
their time. In addition, three of the recordings in the study were prepared with Carl
Orff present at rehearsals and at the actual
recording session.3 These recordings are
quite informative of what may have been
Orff s original intentions. This analysis hopes
to guide artistic decisions for future performances of the work.
The sixteen recordings were selected
according to issue date and nationality of
conductor and performers. The recordings
were analyzed based upon three principal
categories:

Tempo (tempos of the recordings were

compared to Orff s tempo indications);
Text (pronunciation, and articulation);
Musical interpretation (rubato, stringendo,
portamento, phrasing, articulation, and
elision).

Tempo
Each movement of every recording in
the study was timed with a Tempowatch,
both at the beginning of the movement, and
at various tempo changes indicated in the
score.These timings were then compared to

Outstanding Performance
Opportunities Designed for
Undergraduates
Yearly fully-staged undergraduate
opera and opera scenes
Recital series and master classes
Choral tours and major works for
choir and orchestra

Vocal Faculty includes:

Robert Bareeld
Edward Bolkovac
Cherie Caluda
Kevin Jones
Doris Lang Kosloff
Michael Kutner
Joanna Levy
Michele McBride
Marjorie Melnick
Korby Myrick
Wayne Rivera
Kyle Swann
Stuart Younse

Audition Dates
for Fall 2011
Entrance
December 11
(early notication),
January 28 & 30,
February 18 & 19

www.hartford.edu/hartt
The Hartt School | University of Hartford | 200 Bloomeld Avenue | West Hartford, CT 06117 | 860.768.4465 | [email protected]

50

CHORAL JOURNAL Volume 51 Number 4

Orff s metronome markings in the published

score. In total, fifty-one timings were taken
from each recording.The studied recordings
were also compared and contrasted with
each other.
Of all sixteen recordings, the Deutschen
Oper Berlin recording, with Eugen Jochum
conducting, was the closest to Orff s indicated tempo markings. Twenty-three of the
fifty-one-timed sections were within plus or
minus five clicks of the metronome markings given in the score. Of these twentythree, ten of the timed sections on the
Jochum recording were exactly the tempos
indicated in Orff s score. Second closest to
Orff s tempo indications was the Hartford
Symphony performance, conducted by Fritz
Mahler. Twenty-two of the fifty-one timings
taken were within five metronome markings of Orff s. Of these twenty-two, nine
were exactly the tempos indicated in Orff s
score. Orff was in attendance at both these
recording sessions. The Jochum recording is
marked as authorized by Orff.The fact that
Orff was in attendance at the two recordings
most resembling his tempo markings creates
a compelling argument that Orff s markings
were to be taken literally and followed as
closely as possible.
The recording that deviated most
radically from Orff s markings was the Berlin
Philharmonic recording, with Seiji Ozawa
conducting. Only five of Ozawas tempos
fell within five clicks of Orff s metronome
markings. The one section in which most
recordings came closest to Orff s markings was the first four bars of the opening
movement. Of the sixteen recordings, twelve
were within five clicks of Orff s marking.The
movements in which all conductors were
furthest from Orff s markings were the
scatenato section at the end of movement
14 (In Taberna); the opening of movement
15 (Amor volat undique); movement 17 (Stetit
puella); the ancora piu mosso in movement
18 (Circa mea pectora); and movement 24
(Ave formosissima). None of the conductors
was within five clicks of Orff s metronome
markings in these movements or sections.
The recording with the overall fastest
tempos was the Michael Tilson Thomas
recording with the Cleveland Orchestra.
Twelve of the selected timed sections were

Table 1 - Tempo Study

Movement
1
O fortuna
mm 1- 4

Movement Movement
14
11
In Taberna
Estuans
scatenato
Interius

Movement
15
Amor Volat
undique
m. 1

Movement
17
Stetit Puella
m. 1

Movement
18
Circa mea
pectora
anocora
piu mosso

Movement
24
Ave
formosissima

Orff

= 60

= 152

= 160

= 96

= 84

na

= 72

Tilson Thomas

= 57

= 157

= 136

= 64

= 61

= 106

= 50

De Burgos

= 52

= 158

= 190

= 72

= 56

= 92

= 62

Ozawa

= 72

= 162

= 144

= 74

= 70

= 94

= 62

Jochum

= 60

= 170

= 190

= 80

= 61

= 78

= 66

Mahler

= 60

= 154

= 180

= 74

= 58

= 90

= 60

Sawallisch

= 61

= 154

= 180

= 74

= 58

= 90

= 60

Slatkin

= 57

= 162

= 144

= 64

= 70

= 170

= 51

Thiellemann

= 53

= 156

= 168

= 70

= 55

= 84

= 51

faster than any other recording. In contrast,

the recording with the overall slowest
tempos was the Rafael Frbeck de Burgos
recording, with the New Philharmonia
Orchestra. Eighteen of the selected timed
sections were slower than any other recording (see table 1 for specific tempo timings).
When the recordings were compared
by their dates of performance, no noticeable trend or tendency was found. Neither
was there anything of significance found
when the gathered data were compared by
nationality of the conductor.

Text Pronunciation
One of the greatest concerns in Orff s
Carmina Burana should be how to approach
the obscure text. Should the Latin be pro-

nounced with standard ecclesiastical diction

or should German-Latin pronunciation be
used? Should standard German and French
pronunciation be used, or is there a particular pronunciation appropriate for the time
period in which it was written?
In analyzing the pronunciation of the
Carmina Burana text the author observed
three principal trends whether the:
Latin text was performed with a Germanic pronunciation, an Ecclesiastical/
Italianate pronunciation, or some sort
of hybrid pronunciation;
German text was performed with modern
German pronunciation or Middle High
German pronunciation, representing
the time period of the texts inception; and

French diction in movement 16 (Dies, nox

et omnia) was performed with modern
French pronunciation, Old French, or
some sort of hybrid pronunciation.
Two sources were used in the analysis of
diction for this study: Harold Copemans
Singing in Latin, and the collection of essays
titled Singing Early MusicThe pronunciation
of European languages in the Late Middle Ages
and Renaissance, edited by Timothy J. McGee,
A.G. Rigg and David N. Klausner.

Latin Text
Of the sixteen recordings in this study,
only four used Ecclesiastical/Italianate Latin
pronunciation (Muti, Ormandy, Shaw, and
Stokowskis). The other twelve followed
Germanic diction rules.

CHORAL JOURNAL Volume 51 Number 4

51

CARL ORFF'S CARMINA BURANA

Specific to the Carmina Burana text,
there are four principal differences in pronunciation between Ecclesiastical Latin and
Germanic pronunciation:
The treatment of the internal consonant
combination sc, (for example crescis).
In ecclesiastical pronunciation, crescis is
pronounced [crESis]. In other words,
the sc sounds like the English sh,
and the final syllable i is pronounced
closed, as in the English word quiche.
In Germanic pronunciation, crescis
is pronounced [crEtsIs]. The sc
becomes a ts sound and the final
syllable i is pronounced open, as in
the English word it.

The ecclesiastical pronunciation of

egestatem uses a soft g sound
[EdZEstAtEm]. The g sounds like an
English j as in judge. The surrounding e vowels are both open as in
bed.
The Germanic pronunciation of egestatem uses the hard g sound [EgEst
AtEm]. The g sounds like the beginning of the word get. The surrounding es sound like the a in ate.
The treatment of c surrounded by two
vowels (for example glaciem).
In Ecclesiastical pronunciation, glaciem
is pronounced [glAtSiEm]. The c like
ch in the work cheese.

In Germanic pronunciation, glaciem

is pronounced [glAtsiEm]. The c is
pronounced like the t and the s in
the word pots.
The treatment of the ch in michi.
In Ecclesiastical pronunciation, michi is
pronounced [miki].
In German pronunciation, michi is
pronounced [miCi]. Here the ch
is similar to the sound of h in the
English word hue.4
There were several inconsistencies in those
recordings using Germanic Latin pronunciation.
1. In movement 19 (Si puer cum puellula), the
word Felix (fortunate) in most recordings

The treatment of the consonant g, for

example egestatem.

An ISO 9001:2000-compliant manufacturer Toll-free (Canada and USA): 1-877-246-7253 Telephone and fax: +1 604.733.3995

GREAT PERFORMANCES, NO SNAGS.

The

MUSICFOLDER.com

RingBinder

Snag-free rings
never misalign.

Custom foil
stamping adds
elegance and
convenient ID.

successful concert needs a great

performance from every source
including your music folder. Which is why we
designed our RingBinder with special rings
that never snag or misalign so you can
concentrate on singing rather than fumbling.
Its details like these that make our many
folders, with options from clear pockets to
retaining chords to custom imprinting, the
growing choice for choirs around the world.
Order yours from www.musicfolder.com
or your local distributor. Or give us a ring.

ITLL FLIP
FOR YOU
Secure hand
strap lets you
perform with
confidence.

52

A FOLDER FOR INSTRUMENTALISTS!

S!
Our new Concert Band & Orchestra Folder iss
extra-wide for storing and displaying scores, and madee
with the same high quality construction found on all
our folders. Ideal for conductors
and instrumentalists.

CHORAL JOURNAL Volume 51 Number 4

New Jazz Stand sets up with

legs frontward or reversed
then folds flat (right) for easy
storage. Clever! Available
shoulder bag is big enough
for both stand and music.
For singers, instrumentalists and conductors.

The Studied Recordings

Conductor: Gaetano Delogu

Orchestra: Prague Symphony Orchestra
Chorus: Khn Mixed Choir, Pavel Khn, Choir Master

Bambini Di Praga, Bohumil Kulinsky, Choir Master

Soloists: Zdena Kloubova soprano, Vladimir Dolezal
tenor, and Ivan Kusnjer baritone
Recording Date: December 12 and 13, 1995
Recording Place: Dvork Hall of Rudolfinum, Prague
Label:
Supraphon Records. Copyright 1996
Conductor: Antal Dorati
Orchestra: Royal Philharmonic Orchestra
Chorus: Brighton Festival Chorus, Laszlo Heltay,

Chorus Master; Southend Boys, Choir,

Michael Crabbe, Chorus Master
Soloists: Norma Burrowes soprano, Louis Devos tenor,
and John Shirley-Quirk baritone
Recording Date: February, 1976
Recording Place: Kingsway Hall, London
Label: Decca Record Co. Copyright 1976
Conductor: Rafael Frbeck de Burgos
Orchestra: New Philharmonia Orchestra
Chorus: New Philharmonia Chorus Wilhelm Pitz, Chorus

Master; Wandsworth School Boys Choir Russell

Burgess, Chorus Master
Soloists: Lucia Popp soprano, Gerhard Unger tenor,
Raymond Wolansky baritone, and John
Noble - baritone
Recording Date: Not available
Recording Place: Not available
Label: EMI Records Ltd. Copyright 1966, 1987, 1997
Conductor: Stephen Gunzenhauser
Orchestra: CSR Symphony Orchestra (Bratislava)
Chorus: Slovak Philharmonic Chorus

Pavol Prochazka, Choir Master

Soloists: Eva Jenisova soprano, Vladimir Dolezal tenor,

and Ivan Ksunjer baritone

Recording Date: October 28 November 2, 1988
Recording Place: The Studio of the Czechoslovak Radio
(Bratislava)
Label: HNH International Ltd. Copyright 1989

Conductor: Eugen Jochum (Orff present)

Orchestra: Orchester der Deutschen Oper Berlin
Chorus: Chor der Deutschen Oper Berlin,

Walter Hagen-Groll, Choir Master

Schneberger Sngerknaben,
Gerhard Hellwig, Choir Master
Soloists: Gundula Hanowitz soprano
Gerhard Stolze tenor
Dietrich Fischer-Dieskau-baritone
Recording Date: October, 1967
Recording Place: Ufa-Studio, Berlin
(Recording authorized by the composer)
Label: Polydor International. Copyright 1968, 1995
Conductor: Fritz Mahler (Orff present)
Orchestra: Hartford Symphony Orchestra
Chorus: Harford Chorale
Soloists: Sylvia Stahlman soprano

John Ferrante tenor

Morley Meredith baritone
Recording Date: 1957
Recording Place: Hartford, Connecticut
Label: Omega Record Group, copyright 1957, 1999
Conductor: Riccardo Muti
Orchestra: Philharmonia Orchestra
Chorus: Philharmonia Chorus

Southend Boys Choir

Soloists: Arleen Auger soprano

John van Kesteren tenor

Johnathan Summer baritone
Recording Date: Unavailable
Recording Place: Unavailable
Label: EMI Records, Ltd. Copyright 1980, 1997, 1999
Conductor: Eugene Ormandy
Orchestra: The Philadelphia Orchestra
Chorus: The Rutgers University Choir,

F. Austin Walker, Director

Soloists: Janice Harsanyi soprano

Rudolf Petrak tenor

Harve Presnell baritone
Recording Date: April 24, 1960
Recording Place: Broadwood Hotel, Philadelphia, PA
Label: Sony Music, Copyright 1960, 1991

CHORAL JOURNAL Volume 51 Number 4

53

CARL ORFF'S CARMINA BURANA

Conductor: Seiji Ozawa
Orchestra: Berliner Philharmoniker
Chorus: Shinyukai Choir, Shin Sekiya, Chorus Master

Knabenchor Des Staats Und Domchores

Berline, Christian Grube, Chorus Master
Soloists: Edita Gruberova soprano, John Aler Tenor,
and Thomas Hampson baritone
Recording Date: June, 1988
Recording Place: Berlin
Label: Philips Classics Prod. Copyright 1989
Conductor: Krzystof Penderecki
Orchestra: State Krakow Philharmony Orchestra
Chorus: State Cracow Choirs, Piotr Krawczynski,

Bronislawa Wietrzny, Choir masters

Soloists: Venceslava Hruba-Freiberger soprano,

Piotr Kusiewicz - tenor, and Rolf Havenstein baritone

Recording Date: January 27, 28, 1989
Recording Place: Cracow, Poland (live)
Label: Arts Music GMBH, Copyright 1995

Conductor: Wolfgang Sawallisch (Orff present)

Orchestra: Klner Rundfunk-Sinfonie-Orchester
Chorus: Chor des Westdeutschen Rundfunks und Ein

Kinderchor, Bernhard Zimmermann,

Chorus Master
Soloists: Agnes Giebel soprano
Paul Kuen tenor
Marcel Cordes baritone
Recording Date: June 15 20 1956
Recording Place: Sartorysaal, Kln
Performance authorized by Orff
Label: EMI Electrola GMBH. Copyright 1956, 1992
Conductor: Robert Shaw
Orchestra: Atlanta Symphony Orchestra
Chorus: Atlanta Symphony Chorus, Atlanta Boy Choir
Soloists: Judith Blegen soprano, William Brown tenor,

and Haken Hagegard

Recording Date: November 16 18, 1980
Recording Place: Symphony Hall,
Atlanta memorial Arts Center, GA
Label: Telarc International Corp. Copyright 1981

Conductor: Leonard Slatkin

Orchestra: : Saint Louis Symphony Orchestra
Chorus: Saint Louis Symphony Chorus
Soloists: Sylvia McNair soprano

John Aler tenor

Hakan Hagegard baritone
Recording Date: October 9 11, 1992
Recording Place: Powell Symphony Hall, St. Louis, Missouri
Label: BMG Music, Copyright 1994
Conductor: Leopold Stokowski
Orchestra: Houston Symphony Orchestra

Leopold Stokowski Symphony Orchestra

Chorus: The Houston Chorale, Alfred Urbach, Director

Houston Youth Symphony Boys Choir,

Howard Webb, Director
Soloists: Virginia Babikian soprano, Clyde Hager tenor,
and Guy Gardner baritone
Recording Date: April 9, 1113, 1958
December 18, 20, 1957
Recording Place: Jesse Jones Auditorium, Houston, Texas
(1958) Stage Seven of the Samuel Gold
wyn Studios, Hollywood, California
(1957)
Label: Angel Records. Copyright 1959, 2001
Conductor: Christian Thielemann
Orchestra: Orchester Der Deutschen oper
Chorus: Chor Der Deutschen oper, Helmut Sonne,

Chorus Master; Knabenchor Berlin,

Karl-Ludwig Hecht, Chorus Master
Soloists: Christiane Oelze soprano, David Kuebler tenor,
and Simon Keenlyside baritone
Recording Date: October, 1998
Recording Place: Jesus-Christus-Kirche, Berlin
Label: Deutsche Grammophon GMBH. Copyright 1999

Conductor:

Michael Tilson Thomas

Orchestra: Cleveland Orchestra

Chorus: Cleveland Orchestra Chorus and Boys Choir,

Robert Page, Director

Soloists: Judith Blegen soprano, Kenneth Riegel tenor,

and Peter Binder baritone

Recording Date: 1974
Recording Place: Masonic Auditorium, Cleveland, Ohio
30th Street Studio, New York, NY

54

CHORAL JOURNAL Volume 51 Number 4

Latin held a unique position in Western civilization

in the Middle Ages and Renaissance, perhaps similar to
that of present day English in
Africa and India. Although it
was the principal official and
literary language in Western
Europe, it was no ones first
tongue. 9 No one spoke
Harold Copeman best deLatin at home. It was taught
scribes this discrepancy as
in school to a very small
follows:
class of society and it did not
come naturally to anyone.10
Latin vowels all had
The constant use of Latin by
more than one sound,
speakers of vernacular lansometimes three or four.
guages caused it to adapt its
The vowels can all indicate
grammar and pronunciation
different moods. In rejoicing
to vernacular patterns.11 As
e can have a bright sound
vernacular speakers spoke
(Eu, Euge, Euoe, Euare); in
sorrow it can have a more
Latin according to their own
open sound (He, Hei, Heu,
idiom, the pronunciation of
Heiulare).5
Latin divided according to national and regional languages
However, Copeman goes
and dialects.12 There were
on to describe the pronunciaas many pronunciations of
tion of the word Felix: Felix,
Latin across Europe as there
found on coins as FEELIX, the
were languages.The Goliardic
mouth is first half closed, then
poets of the Carmina Burana
open, with lips moderately
and the monks who compiled
forward, with more vigorous
the Burana Codex13 were
breath, letting the tongue
located primarily in the region
touch the front teeth, (a rather
known as Bavaria, or Southopen [e:]).6
ern Germany.14 Therefore,
Taken from Burana Codex Depicting the Fortuna Wheel.
their pronunciation of Latin
2. Another inconsistency bewould reflect the German
tween recordings lies in the
vernacular of the day. Bearpronunciation of the word Ecce (behold)
ciation [EtSe] follows the Ecclesiastical ing this in mind, it is most appropriate for
in those recordings using the Germanic
Latin rules; therefore, is incorrect in a the Latin texts to be pronounced using
pronunciation of Latin. According to
Germanic diction rather than ecclesiastical.
Germanic Latin performance.
Copeman, cc before e should be
Moreover, an ecclesiastical pronunciation
pronounced [kts] according to the rules 3. In Movement 7 (Floret silva), all of the lends itself to a more prayerful hearing
of Middle High German.7 None of the
recordings using Germanic Latin diction of the text, rather than the secular, profane
chosen recordings observed this rule.
pronounced floret (he/she/it flourishes) intentions of the poets.
Instead, Ecce was either pronounced
[flrEt] or [florEt], except for the Eu[EtSe] or [Eke].
gene Ormandy recording with the
According to Copeman, cc before
German Text
Philadelphia Orchestra. In this recording,
front vowels did not become generfloret was pronounced [flore]. AccordMovements 7 10 contain German texts.
ally [ts] until the seventh century, but
ing to Copeman, the proper pronuncia- In the sixteen chosen recordings, two basic
in (northern Europe), the Irish Latin [k]
tion of floret is [florEt]. No justification diction philosophies were followed:
of Columbanus and St. Gall was kept
was found for the pronunciation [flore].
for centuries.8 This explanation justifies
Modern German diction rules were apthe pronunciation as [Eke]. The pronunplied to the texts; or
was pronounced [fEliks]
with an open e sound
as in bed in the first syllable. Christian Thielemann,
Wolfgang Sawallisch and
Eugen Jochum pronounced
the word [filiks] with a
closed e, as in beet.

CHORAL JOURNAL Volume 51 Number 4

55

CARL ORFF'S CARMINA BURANA

A modified form of Middle High German
pronunciation was applied.15

There were no significant differences in

pronunciation among those recordings that
used modern German diction. Among those
using modified Middle High German, the following discrepancies were noticed:

The most obvious differences between

the two styles came in the pronunciation of
v and w. Those recordings that followed
the Modern German diction pronounced The pronunciation of e at the end of a
v as [f ] and w as [v]. Those following
word following a consonant, for exthe modified Middle High German diction
ample umbe, had two different variapronounced v as [v] and w as [w].
tions. Several recordings pronounced
The recordings that used Modern
this as [umb] while others used a
German pronunciation were conducted
more open sounding ending [umbE]
by Ormandy, Muti, Jochum, Tilson Thomas,
or [umbe]. According to Peter Frenzel,
Shaw, Ozawa, Thielemann, Stokowski, Gune is pronounced [e], as in stay, when
zenhauser and Mahler. The recordings using
unstressed in open syllable, but lax
a modified version of Middle High German
and tending to a more open sound.
were by Delogu, Dorati, Sawallisch, PendIt was noted that [] is also a possible
erecki, Slatkin, and Frbeck de Burgos.
ending.16

56

CHORAL JOURNAL Volume 51 Number 4

In movement 10 (Were diu werlt alle min),

measure 13 contains the words ih and
mih. The hs in these words were
interpreted in two different ways.
Some kept the consonant silent [i]
and [mi]. Others interpreted the h
as the allophone [C]. According to
Frenzel, medial h had probably become silent or lightly aspirated by the
fifteenth century in many European
dialects. However, the digraph ch,
pronounced generally [x] in Middle
High German, at some point in the
early modern period split into the allophones [C] (after e and I).17 He
further explains that German spelling
varied by region prior to the 15th
century. Letters were often excluded.

For example, ih and mih meant the

same and were pronounced the same
as ich and mich.18 Following Frenzels
logic, the correct pronunciation of ih
is [iC] and mih [miC]. This would hold
true for conductors following modern
German pronunciation as well.
Because the geographical inception of
the texts has been traced to the German
state of Bavaria, the correct German dialect
to associate with the text is that of Middle
Upper German.19 It is in this dialect group
that most of the songs and stories seem to
have been sung and told during the height
of courtly entertainment, between 1175
and 1250.20

French Text
The French text found in movement
16 (Dies, nox et omnia) contained the
most discrepancies among all the studied
recordings. Several recordings seemed to
completely ignore the
change in text and applied
Latin pronunciation to the
entire movement. The
most notable examples of
this were the Stokowski,
Ormandy, and Penderecki
recordings. It may also
be possible, due to the
archaic French spelling,
that the conductors and
soloists were unaware
that the text was indeed
French, rather than Latin.
Others seemed to mix
diction rules for Latin
and French, using French
consonant sounds, but
Latin vowel sounds. This
is heard most strikingly in
the Ozawa recording.
Robert Taylor states
there is no standard
phonology of Medieval
French, and there were
only the beginnings of
scientific accuracy of
phonetic description by
the sixteenth century.21
Sawallisch, Slatkin, Muti,

Mahler, Jochum, Delogu, Shaw, and Tilson

Thomas conducted the recordings that
obser ved a clear distinction between
French and Latin. In the Frbeck de Burgos
recording, modern French diction seems to
be observed. Although all nine of these recordings obviously recognized the macaronic
text, there were many discrepancies among
all of the interpretations, too numerous to
list here. It would seem that the importance
lies in the recognition of the French text and
an obvious performance of that recognition.

Musical Interpretation
In analyzing the musical decisions of these
sixteen conductors, the author took into
consideration three principal matters:

Individual conductors musical decisions

that are not specifically notated in
the score.

Luftpause
The Luftpause is indicated by a comma,
written above the staff. Orff uses this marking several times in virtually every movement
of Carmina Burana. The dilemma is that the
marking is a rather vague indication with no
quantifiable temporal value. An example of
this conundrum comes in the first four bars
of the piece (Figure 1).
In the surveyed recordings, the particular
markings in the example above were interpreted in one of the following ways:
A complete break in rhythmical time;

The treatment of the Luftpause () Orff

used throughout his score;
The treatment of movements marked
Attacca versus those that are not; and

A quarter rest on the last beat of the measure, while still maintaining a tempo;
A fermata on the last half note of each

CHORAL JOURNAL Volume 51 Number 4

57

CARL ORFF'S CARMINA BURANA

measure and then continuing directly
into the next measure;
An eighth rest on the last beat of the measure while still maintaining a tempo;

The marking was not recognized at all.

The problem lies in the fact that it is unclear
which of the five circumstances Orff is suggesting. Another challenging example comes

in movement 14 (In taberna) at measure 74

(Figure 2).
In this example, the four bars before the
Luftpause are a repetition of the same twobar musical phrases. New musical material
follows the Luftpause.
The conductor could
interpret this marking
in all five of the above
possible scenarios. Of
the surveyed recordings, not one conductor
chose to observe this
marking at all. Unfortunately, it is impossible
to offer a definitive
interpretation of Orff s
intentions from the
results of this study.
Several additional issues
concerning interpretation of the Luftpause are
discussed later.

Attacca Markings
Orff worked within
a concept he titled Theatrum Mundi, in which
music, movement, and
speech were inseparable. 22 There is an
extreme element of
drama inherent in all
of Orff s later works.
With this in mind, Orff
divided the twenty-five
separate movements
of Carmina Burana into
five scenes. To keep
these scenes intact, each
movement within a
scene was marked attacca to the subsequent
movement.
Of the sur veyed
recordings, Mahler, Muti,
Sawallisch, Shaw and
Slatkin offered a strong
sense of scenic structure and integrity. It
is wor th noting that
Shaw is the only con58

CHORAL JOURNAL Volume 51 Number 4

ductor who formatted the recording with

only four tracks (1. Introduction - Fortuna
Imperatrix Mundi; 2. Part 1, Primo vere and
Uf dem Anger; 3. Part II - In taberna; 4. Part
III - Cour damours, Blanziflor et Helena and
the return of Fortuna Imperatrix Mundi). It is
also important to consider that it might have
been the decision of a recording engineer
or other technical factors that prevented a
scenic continuity, rather than the conductors
musical decisions.
Within the surveyed recordings, the
greatest discrepancy in attacca recognition
came between movements 5 (Ecce gratum) and 6 (Tanz). These two movements
occur in the middle of the scene labeled
Primo vere. Movement 6 is the beginning of
Primo veres internal scene Uf dem anger.
There is no attacca marking between the
two movements. Several conductors did
perform what appears to be an attacca
between these two movements (Jochum,
Penderecki, and Thielemann). Perhaps this
was done in order to maintain the structure
of the overall Primo vere scene.

Musical Decisions Other Than

Those Marked in the Score
In any piece of music, whether it be by
Mozart or Bartok, conductors will take
liberties outside of what the composer
has indicated in the score. Such liberties
include ritardandi, accelerandi, grand pauses,
or fermati, to name only a few. However,
taking into account how specific Orff was
in marking his score, such liberties are even
more evident than usual.
The conductor who went farthest from
Orff s indications was Leopold Stokowski
in his 1957 recording with the Houston
Symphony Orchestra. In this age of studied
performance practice, Leopold Stokowski
is now infamous for his romanticizing of
Bach and other tremendous liberties taken
with established repertoire. In Stokowskis
rendition of Orff s work, he actually cuts
the first twelve bars as well as measures
53 and 54 marked feroce from Movement
14. There is no indication in Orff s score
that these cuts are acceptable. In some
cases, conductors will cut portions of the
text to accommodate the performance

abilities of the ensemble. However, bearing

in mind that these cuts are small and, in the
case of movement 14, the material that was
cut is repeated later, it is questionable as to
whether this was Stokowskis motive. The
cuts may also have been made due to time
limitations on the recording medium available. Another decision made by Stokowski
comes in the baritone solo, movement
16 (Dies, nox et omnies). In this recording,
a baritone and tenor are used. The tenor
sings the sections usually performed by the
baritone in falsetto. Although this decision
may have been made due to vocal limitations
of the baritone soloist, artistically it pays no
attention to the text or Orff s conception
of characters playing a role. Throughout this
performance, Stokowski takes numerous
liberties with tempo and articulation that are
not indicated in the score and are not heard
in any other recording.
One other conductor who took extreme
liberties with the score was Penderecki in his
1994 recording of the Karol Szymanowski
State Cracow Philharmony Orchestra
and Choirs. For example, in the first four
measures of the piece, Penderecki makes
extremely large pauses at each Luftpause,
interpreting the marking as a fermata over a
rest, rather than a slight pause.This luftpause
interpretation is also taken at measure 31
(Figure 3) and each subsequent strophe

throughout movement 5.
There also seems to be very little, if any,
recognition of scenic unity on the Penderecki
recording. Attacca markings are sporadically
recognized, creating no cohesion between
each scene.
The single movement in which most
conductors took some sort of artistic license
was in movement 14 (In taberna). Orff s

By
Paul Brandvik

Author of
The Compleet
Madrigal Dinner
Booke
THIRTY SCRIPTS, each containing All
DIALOGUE (including greeting, toasts, festivities, and humorous Renaissance play) necessary
for A SUCCESSFUL MADRIGAL DINNER!
For complete info: VISIT OUR WEB SITE:

www.madrigaldinner.com
knight-shtick press
Box 814, Bemidji, MN 56619-0814
218-586-2270
[email protected]

CHORAL JOURNAL Volume 51 Number 4

59

CARL ORFF'S CARMINA BURANA

Luftpause is particularly
problematic in this movement. For example, in
the first 12 bars of the
movement there are four
Luftpause. No conductor
in this survey made any
observation of these four
markings (Figure 4).
However, in measure
24 a Luftpause is marked
after the last beat of
the measure (Figure 5).
Fritz Mahler and Rafael
Frhbeck de Burgos interpreted this marking
as a grand pause before
the new section of music.
Sawallisch interpreted this
marking as a one beat rest
before the new tempo.
De Burgos also places
a grand pause in measure 36, where there is
no indication for such
an interpretation (Figure
6). One other recorded
discrepancy comes in
measure 70 of Movement
14 (Figure 7).
Here, we see no indication for a break in sound
or tempo before the
subito molto stentato section. However, Frhbeck
de Burgos, Mahler, Sawallisch, and Tilson Thomas
all placed either a break or
grand pause before this new section. When
these two bars are analyzed harmonically,
we see a series of first inversion chords
(perhaps mimicking fauxbourdon) in the
choral part all over an E pedal.The downbeat
of measure 71 is an F-major chord in first
inversion in both the choral and wind parts
with a D pedal in the strings.The indicated
E pedal of the strings and woodwinds is
harmonically leading to the F-major 6-3
chord at the tempo change. In addition, the
chord progression of measure 70 ends with
an E-major 6-3 chord, which also leads the
listener to F major. To place a break after this
progression seems to destroy the harmonic
60

CHORAL JOURNAL Volume 51 Number 4

tension, making this decision questionable. In

addition, Stentato means with difficulty. It is
an expression mark often used by Verdi in
vocal lines at moments of extreme anguish.
In 1703, Sbastien de Brossard wrote it
meant one should sing as though one were
suffering.23 Orff not only indicates Stentato
but also molto and subito. Something must
be done, and immediately! Placing a grand
pause prior to this indication does not seem
to suit Orff s marking.

Summary
Many conductors use commercial recordings
as a resource at some point in their score
preparation. Most of them agree that the
most effective way of listening is to select a
number of divergent recordings to gain different insights on interpretation. Although
this study found the Eugen Jochum recording to be closest to Orff s markings, many
of the recordings in the study are valuable
resources in gaining knowledge of Orff s
desired interpretation.
The 1980 recording of Robert Shaw
conducting the Atlanta Symphony Orchestra

and Chorus is a strong performance of the

work. The chorus is truly presented as the
centerpiece of the work, as Orff intended.

Hkan Hgagard delivers

a convincing performance
of the baritone solos, particularly in the Circa mea
pectoras. Judith Blegen gives
the strongest solo performance, and, perhaps, is the
finest soprano soloist in
all the studied recordings.
Shaw also demonstrates his
understanding of the scenic
structure by not only clearly
observing attacca markings,
but also inserting tracks on
the CD by scene rather than
movement. In the slower
movements Shaw creates
a beautiful, graceful sound
in the chorus. However,
the carnal power and drive
expected in movements
such as Were diu welt alle
min and In taberna quando
sumus seems absent. Shaws
interpretation also falls short
in this study: he uses liturgical
Latin and modern German
diction rather than the more
appropriate medieval forms.
Even though Seiji Ozawas 1988 recording with the Berliner Philharmoniker and the

Its easy to register for the

2011 ACDA National Conference.

Register online at www.acda.org

CHORAL JOURNAL Volume 51 Number 4

61

CARL ORFF'S CARMINA BURANA

Shinyukai Choir is one of the farthest from

Orff s intentions according to the tempo
study, it is still a thrilling recording. The tempos are very fast, yet do not seem rushed
as they do in the Michael Tilson Thomas
62

recording with the Cleveland Orchestra.

Ozawa also employs a Germanic pronunciation of the Latin text, which comes across
quite clearly. The three soloists (Edita Gruberova, John Aler, and Thomas Hampson)

CHORAL JOURNAL Volume 51 Number 4

give striking performances.

John Aler, in particular, also
heard on the Leonard Slatkin recording, gives a magnificent performance of the
roasting swan.
Leonard Slatkins 1994
recording with the Saint
Louis Symphony Orchestra
and Chorus is an excellent
source for pronunciation.
Slatkin employs both the
Germanic-Latin and medieval Middle High German
pronunciations. In this recording, the soloists seem
to be placed at the fore.The
recording quality is superb.
Unlike many recordings of
Carmina Burana, the sound
levels remain consistent.
The 1995 Gaetano
Delogu live recording with
the Prague Symphony Orchestra and the Khn Mixed
Choir exemplifies much of
the diction suggested by
Copeman and Riggs. However, the chorus lacks the
stunning rhythmic alacrity
and unstinting energy that
make others in the study so
exciting and memorable. It
is also interesting to note
that the CD cover has a
picture of a performance
with various actors and actresses in medieval costume
in front of a chorus dressed
in monk habits. This picture
may inspire some scenic
interpretation.
Although the recording
technology used in Fritz
Mahlers 1957 recording
with the Hartford Symphony Orchestra and Chorus
seems archaic compared to
todays standards, it is worth listening to for
Mahlers tempos and interpretations. This is
the first commercial recording of Carmina
Burana and Carl Orff was present at the
rehearsals and performances.

Concluding Remarks
Carl Orff s Carmina Burana has established itself as a cornerstone of the
twentieth-century choral repertoire. Its
combination of medieval and modern elements, along with Orff s distinctive compositional style, creates a thrilling experience
for both the audience and the performer.
Although the elements of strong adherence
to score markings and principles of diction
seem highly restrictive in this study, the freedom of scenic conception allows every new
performance to be as innovative and fresh
as the very first.
Many critics of Carmina Burana find it
vulgar, devoid of the standard techniques
and elaborations that made music an art
and a spiritual force.24 However, the reason for this opinion may be that too many
performers indulge in its excesses rather
than adhere to its strong sense of form and
symmetry. When performed with all Orff s
intentions accounted for, Carmina Burana
becomes a pulsing, visceral, and cathartic
musical experience.
NOTES
1

2
3
4

5
6
7

8
9
10

11
12
13

Philip, Robert. Early Recordings and Musical Style.

(Cambridge University Press: Cambridge
1992) 1.
See appendix for complete list and information
on each recording.
These recordings are indicated as Orff present
in the list of recordings.
Wall, Joan. International Phonetic Alphabet for
Singers. A Manual for English and Foreign
Language Diction. (PSTInc: Dallas, Texas
1989). 218.
Copeman, Harold. Singing in Latin. (Published at
Oxford by the Author, 1996). 97.
Ibid., 98.
Copeman, Harold and Vera U.G. Scherr. German
Latin. Singing Early Music. Ed. T. McGee, A.G.
Rigg, D.N. Klausner. Bloomington: Indiana
University Press, 1996. 261ff.
Copeman. 170.
Copeman. 6.
Rigg, A. G. Singing Early Music The pronunciation
of European Languages in the Late Middle Ages
and Renaissance. (Indiana University Press.
Bloomington) 1996. P. 5
Ibid.
Copeman, p. 8.
Carmina Burana literally means songs from the

14

15

16

17
18

19
20
21

monastery of Benediktbeuern. The source

of these texts is known as the Burana Codex.
This collection of Medieval Latin and Middle
High German and French lyrics, the most
important of its kind, was collected in Austria
between 1220 and 1250.
Parlett, David. Selections from the Carmina
BuranaA Verse Translation. (Viking Penguin
Inc. New York) p. 19ff.
The word modified is used here, because no
recording completely followed the rules of
Middle High German, as summarized in Riggs
Singing Early Music.
Frenzel, Peter. Late Medieval German and Early
New High German, Singing Early Music. Ed.
McGee, T, Rigg, A.G. Klausner, D. (Indiana
university Press: Bloomington 1996) 228.
Ibid.
Frenzel, Peter. Germany and the Low Countries.
Singing Early Music. Ed. McGee, T. Rigg, A.G.,
Klausner. D. (Bloomington: Indiana University
Press: 1996). 244.
Frenzel, p. 220
Ibid.
Taylor, Robert. Old French Singing Early Music
The Pronunciation of European Languages in
the Late Middle Ages and Renaissance. Ed. T.J.

McGee; A. G. Rigg; D. N. Klausner (Indiana

University Press: Bloomington 1996) 65.
22
For more information see Babcock, Jonathan.
Carl Orff s Carmina Burana: A Fresh
Approach to the Works Performance
Practice. Choral Journal 45, no. 11 (May 2006):
26 40.
23
David Fallows. Stentato. In Grove Music
Online. Oxford Music Online, http://www.
oxfordmusiconline.com/subscriber/ar ticle/
grove/music/26681 (accessed July 22, 2010).
24
Lang, Paul Henry. Carl Orff Bungled Fire-works
or Skillful Effects? High Fidelity (January
1982): 52.
For complete tempo timings see: Babcock, Jonathan,
Carl Orff s Carmina Burana: An Analysis
of the works performance practice on
commercial recordings. (D.M.A. dissertation,
University of Hartford 2003) Appendix C.
pp. 99113.

application deadline:
DECEMBER 1

VOICE / OPERA

UNDERGRADUATE DEGREES
GRADUATE DEGREES
PERFORMANCE DIPLOMA
Dr. Karl Paulnack, director, music division
Patty Thom, chair, voice and opera
Dr. William Cutter, director, choral studies
Kirsten Z. Cairns, director, opera studies
FACULTY:
Michelle Alexander, Marilyn Bulli, Kirsten Z. Cairns,
William Cotten, William Cutter, Kerry Deal,
Rebecca Folsom, Sara Goldstein, Thomas Gregg,
Robert Honeysucker, Victor Jannett, James Javore,
Michael McGaghie, Monique Phinney, Mary Saunders,
Merrill Shea, Lisa Sheldon, Melinda Sullivan,
Patty Thom, Craig Wich, Kevin Wilson, Kathryn Wright
8 the fenway, boston, ma 02215 | (617) 912-9153 | admissions.audition.info: www.bostonconservatory.edu

CHORAL JOURNAL Volume 51 Number 4

63