Performance Practice Issues Koussevitzky

University of Nebraska - Lincoln
DigitalCommons@University of Nebraska - Lincoln
Student Research, Creative Activity, and Music, School of

Performance - School of Music
April 2006
Performance Practice Issues in the Short Solo Works of Serge

Koussevitzky
Maurice Kelley
University of Nebraska-Lincoln, [email protected]
Follow this and additional works at: https://digitalcommons.unl.edu/musicstudent
Part of the Music Commons
Kelley, Maurice, "Performance Practice Issues in the Short Solo Works of Serge Koussevitzky" (2006).
Student Research, Creative Activity, and Performance - School of Music. 2.
https://digitalcommons.unl.edu/musicstudent/2
This Article is brought to you for free and open access by the Music, School of at DigitalCommons@University of
Nebraska - Lincoln. It has been accepted for inclusion in Student Research, Creative Activity, and Performance -
School of Music by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln.
Performance Practice Issues in the Short Solo Works of
Serge Koussevitzky
by
Maurice E. Kelley II
A DISSERTATION
Presented to the Faculty of
The Graduate College at the University of Nebraska
In Partial Fulfillment of Requirements
For the Degree of Doctor of Musical Arts
Major: Music
Under the Supervision of Professor Russell White
Lincoln, Nebraska
May, 2006
Performance Practice Issues in the
Short Solo Works of Serge Koussevitzky
Maurice E. Kelley II, D.M.A.
University of Nebraska, 2006
Advisor: Professor Russell White
This document deals with performance practice issues in the four original short
solo double bass compositions of Serge Koussevitzky. There has been minimal literature
published concerning performance practice issues for the double bass as a solo
instrument. By analyzing the 1929 recording Koussevitzky made of two of his pieces for
double bass and the 1988 recordings modern bass virtuoso Gary Karr made of
Koussevitzky’s four short pieces a direct comparison of performance practice issues can
be made. The four short pieces, Andante, Valse Miniature, Chanson Triste, and
Humoresque were written by Koussevitzky around 1900. His recording of Chanson
Triste and Valse Miniature provides the opportunity to compare the first known recording
of a double bass virtuoso performing his own compositions to that of a modern virtuoso.
By comparing Koussevitzky’s recordings to those of Karr performance practice issues
such as vibrato, rubato and portamento can be analyzed and changes in their use can be
determined. The outcome of this comparison will provide performers with the tools
necessary to produce a historically informed performance of all four pieces.
This study examines the most striking differences between the Koussevitzky and
Karr recordings in regards to performance practice, focusing on tempo manipulation
(tempo rubato), vibrato, and portamento. Chapter Two analyzes tempo manipulation as it
applies to recordings of Valse Miniature and Chanson Triste. Chapter Three focuses on
the use of string instrument vibrato during the late nineteenth century and includes an
analysis of Koussevitzky’s use of vibrato. The use of portamento by Koussevitzky’s and
Karr’s recording of Chanson Triste is examined in Chapter Four. Chapter Five brings
together the material collected in previous chapters and presents a detailed method for
creating an historically informed performance of the two pieces Koussevitzky did not
record, Andante and Humoresque. Chapter Six concludes the study by addressing the
issue of authenticity and balancing the primary source material (the recordings of
Koussevitzky performing his own compositions) with the expectations of a twenty-first
century audience.
ii
Table of Contents
List of Figures…………………………………………………………………………….iii
List of Musical Examples………………………………………………………………...iv
Acknowledgements……………………………………………………………………....vii
Chapter
1. Introduction……………….…………………………………...…………...….1
2. Tempo Manipulation in Valse Miniature and Chanson Triste...........................9
3. Vibrato in Valse Miniature and Chanson Triste…………...…………………34
4. Portamento in Chanson Triste………………………………………………. 54
5. Suggestions for an Historically Informed Performance of……………..……68

Andante and Humoresque
6. Conclusions and Suggestions for Further Research………………………….83
Bibliography……………………………………………………………………………..86
iii
List of Figures
1.1 Photograph of Koussevitzky’s bass strings……………………………..………..5
1.2 Letter from Gary Karr to Daniel Larson…………………………………………6

iv
List of Examples
2.1 Valse Miniature, Antecedent/Consequent Pattern……………………….11
2.2 Valse Miniature, Measures 34-38………………………………………..16
2.5 Valse Miniature, Meaures 55-58………………………………………....19
2.8 Valse Miniature, Measures 71-74…………………………………….….21
2.11 Valse Miniature, Measures 103-end……………………………………..24
2.12 Chanson Triste, Measures 1-3…………………………………………...26
2.13 Chanson Triste, Measures 4-7…………………………………………...27
2.14 Chanson Triste, Rhythmic Pattern Measures 10, 13, 16, 17……………..28
2.15 Chanson Triste, Measures 27-31…………………………………….…..29
2.16 Chanson Triste, Measures 32-35………………………………………...30
3.1 Chanson Triste, Measures 4-7……………….……………………....…..37
3.2 Chanson Triste, Measures 8-11………………………………………….38

v
3.5 Chanson Triste, Transitional Material……...…………………………....40
3.6 Valse Miniature, Antecedent/Consequent Pattern……………………41-42
3.9 Chanson Triste, Measures 20-35……………………………………...…48
3.10 Chanson Triste, Cadenza………………………………………………...49
4.1 Three types of portamento…………………………………………….…54
4.2 Chanson Triste, Measures 4-7…………………………………...………56
4.3 Chanson Triste, Measures 8-11………………………………………….57
4.4 Chanson Triste, Measures 12-15…………………………………..…….58
4.7 Chanson Triste, Measures 24-27……………………………………...…61
4.8 Chanson Triste, Measures 28-35…………………………………...……63
4.9 Chanson Triste, Cadenza……………………………………………...…64
4.10 Chanson Triste, Measure 52……………………………………………..66
5.1 Andante, Measures 1-22………………………………………………….70
5.2 Andante, Measures 53-59………………………………………………...74
5.3 Andante, Measures 23-39………………………………………………...76

vi
5.4 Humoresque, Measures 7-23…………………………………………….77
5.5 Humoresque, Measures 52-68…………………………………………...78
5.6 Humoresque, Measures 24-34………………………………………...…80
5.7 Humoresque, Measures 37-52…………………………………………...81

vii
Acknowledgements
I wish to thank the members of my Committee for their encouragement and help
in writing this document. I gratefully acknowledge my advisor, Professor Russell White,
for his continuous support. I also express my gratitude to Dr. Peter Lefferts for his
assistance, enthusiasm, and guidance concerning this topic. In addition, I express my
gratitude to Dr. Karen Becker and Professor Clark Potter for reading this document and
offering advice on its improvement.
Finally, I wish to thank my wife and my parents, who have continually supported
all of my endeavors.
Chapter One
Introduction
A great deal has been written about the technical and mechanical development of
the double bass. In contrast, there is a relatively small portion of literature concerning the
history of the bass as a solo instrument from a performance practice perspective. The
lives of the few early double bass virtuosi have been documented in a variety of books
and articles. It is a simple matter to access information about bass virtuosi such as
Dragonetti and Bottesini. Biographical data, such as when and where they performed,
their compositions for solo double bass, and documents written by their contemporaries,
are readily available and provide some understanding of their technical ability. With the
assistance of modern scholarship, we can even speculate what their performances might
have sounded like. In 1929 Serge Koussevitzky recorded his Chanson Triste, Op. 2,
Valse Miniature, Op. 1, and the second movement of his Concerto for Double Bass, Op.
3. It is not until this time that we have indisputable evidence of a composers’ intent.1
With these early recordings, Koussevitzky presents a unique opportunity to hear him
perform his own compositions and compare them to more recent renditions. By making
this comparison, an historically informed performance of his four short solo pieces for
double bass can be achieved.
1
Serge Koussevitzky, The Complete Double-Bass Recordings, Biddulph recordings WHL 019, 1994,
Compact Disc.
2
Koussevitzky was born in Russia in 1874 and is best known for his work as
conductor of the Boston Symphony Orchestra, a post which he held from 1924 until
1949.2 His career as a bassist has been largely overlooked outside the bass community. At
the age of 14 he enrolled in the Moscow Philharmonic School as a recipient of a double
bass scholarship. He completed the program, for which five years was usually allowed, in
five months. In 1894 he received his degree and began his professional career as a
bassist. By the end of the century he had established himself as a virtuoso
instrumentalist.3 It was during this period that he made his most lasting contributions to
the bass repertoire. He transcribed several works to be performed on his recitals,
including Bruch’s Kol Nidrei,4 Mozart’s Concerto for Bassoon, K. 191, and Eccles’
Sonata in G Minor among others.5 Koussevitzky also wrote four original short solo pieces
for double bass around this time. Andante and Valse Miniature, were originally published
under the title Deux morceaux, Op. 1. These two works, along with the Humoresque, Op.
4, were originally published in Moscow during the 1890s. The Chanson Triste, Op. 2,
was also published in Moscow in 1896.6 All four original pieces were published by the
firm P. Jurgenson and were probably composed during the 1890s. 7
In comparing Koussevitzky’s recordings to more recent renditions it makes sense
to examine the modern double bass virtuoso Gary Karr as a representative of the modern
2
Paul Brun, A History of the Double Bass, trans. by Lynn Morel and Paul Brun, (Paris: by the author,
1989), 170.
3
Arthur Lourie, Sergie Koussevitzky and His Epoch, trans. S.W. Spring (Freeport: Books for Library Press,
Repr. Ed. 1969), 32-26.
4
Max Bruch, Kol Nidrei. Transcribed for double bass by Serge Koussevitzky. Manuscript (photocopy),
Koussevitzky Collection, Library of Congress, Washington, DC.
5
Lourie, 37.
6
Serge Koussevitzky, Chanson Triste, Opus 2. Reverse image copyist’s score (photocopy), Koussevitzky
Collection, Library of Congress, Washington, DC. Moscow: P. Jurgenson, 1896.
7
Kevin LaVine, Reference Librarian, Music Division, Library of Congress, electronic mail inquiry by
author, 21 January 2005, Koussevitzky Collection, Library of Congress, Washington, DC.
3
style of double bass performance. The relationship between Karr and Koussevitzky is
filled with irony and coincidence. Koussevitzky’s 1611 Amati double bass was given to
Karr by Olga Koussevitzky, Serge’s wife.8 According to Karr, in 1962 Olga heard him
perform Koussevitzky’s Concerto for Double Bass in recital. The next day she invited
him to her apartment where she offered Koussevitzky’s double bass to him as a gift. She
did this because she felt he was the one to carry on her late husband’s legacy. She told a
mutual friend that during Karr’s performance she saw the ghost of her husband with his
arms around Karr, and this vision influenced her decision to give Karr the instrument.
This has produced a significant amount of speculation concerning a psychic or
paranormal link between Koussevitzky and Karr. Patrons often refer to the “aura” or
ghost of Koussevitzky looking over Karr during recital performances.9
In 1988 Gary Karr recorded The Spirit of Koussevitzky using the 1611 Amati
double bass given to him by Koussevitzky’s widow.10 This recording includes both short
pieces that Koussevitzky recorded as well as the Andante, Op. 1, and the Humoresque,
Op. 4, in addition to compositions by contemporaries of Koussevitzky. The four short
solo pieces, Andante, Valse Miniature, Chanson Triste, and Humoresque, were written by
Koussevitzky prior to 1900.11 These recordings provide the unique opportunity to
compare the first known recording of a double bass virtuoso performing his own
compositions to that of a modern virtuoso. By comparing the recordings of Chanson and
Valse of Koussevitzky to that of Karr, one may hear the transition in performance
8
Gary Karr, American Symphony Orchestra, Dialogues and Extensions (www.americansymphony.org/
dialogues_extensions/98_99season/4th_concert/koussevitzky).
9
Anthony Angarano, CD liner notes to The Spirit of Koussevitzky, performed by Gary Karr, Vogt Quality
Recordings VQR 0231, 1989, Compact Disc.
10
Gary Karr, The Spirit of Koussevitzky, Vogt Quality Recordings VQR 0231, 1989, Compact Disc.
11
Lourie, 37.
4
practice between the early and later twentieth century. By further analyzing Karr’s
recording of the Andante and Humoresque in comparison to Koussevitzky’s performance,
an historically informed performance of all four pieces can be achieved. The fact that
both recordings, made 59 years apart [Koussevitzky in 1929 and Karr in 1988], are
performed using the same instrument is an added bonus.12
Before a direct comparison can be made between the recordings there are several
factors that should be considered. Almost thirty years passed between the time that
Koussevitzky wrote and first performed his compositions13 and the date he recorded
them. It is impossible to know for certain whether or not Koussevitzky’s interpretation of
his own compositions changed during this time. Also, there is very little documentation
concerning the physical adaptations made to this particular instrument and how that may
have influenced Koussevitzky’s performance of his compositions. In addition, there have
been significant advances in recording technology during the twentieth century.
It is important to note that it was during the 1930s when the first complete set of
aluminum-covered strings were developed for the double bass.14 These strings were
developed by Pierre Delescluse and were admired by Koussevitzky. Delescluse was hired
as a bassist by the Boston Symphony Orchestra under Koussevitzky15 and sometime after
this the section adopted the strings. It is unclear when Koussevitzky became aware of
these strings, although development of metal strings for the double bass had been under
way for some time.
12
Angarano.
13
Ibid. Koussevitzky is thought to have first performed his compositions in a Berlin recital (1903) using his
newly acquired Amati double bass.
14
Brun, 16.
15
Ibid, 170.
5
A silent auction was held during the 2005 International Society of Bassists
convention in Kalamazoo, Michigan. Among the items for bid were some of the contents
of the case which held Koussevitzky’s double bass at the time is was given to Karr.
Two of Koussevitzky’s gut strings probably dating from the 1920’s were included
(Figure 1.1).
Figure 1.1
Photograph of Koussevitzky’s gut strings, taken at the 2005 International Society of

Double Bass Convention by the author.
It is impossible to know if these were the actual strings that were on the bass when
Koussevitzky made his recordings; however, the fact that they were in his case and date
from the 1920’s makes it likely that Koussevitzky was using gut strings at the time of the
recording in 1929 (Figure 1.2).16
16
Gary Karr to Daniel Larson, not dated, in reference to items purchased at silent auction during the 2005
International Society of Bassists convention in Kalamazoo, Michigan, June 6-11, 2005. Signed by Gary
Karr.
6
Figure 1.2
7
The choice of metal versus gut strings is important and is directly tied to the issue
of tuning. There are a variety of tuning schemes for the double bass. The most common,
referred to as orchestral tuning, has the bass tuned E-A-D-G. However, bassists often use
the alternate tuning of F#-B-E-A, transposing the instrument from the key of C to the key
of D for solo performance. This is generally referred to as solo tuning, since many solo
works are written for bass in D and it is not used for playing in ensembles. Koussevitzky
clearly intended his pieces to be played using solo tuning. He used this tuning when he
made his recordings in 192917 and most of the older editions of his compositions call for
it.18 In addition, a reverse image copyist’s score of Chanson Triste dated 1896 is written
with both the bass and piano parts in the key of e minor.19 Before the advent of metal
strings, bassists often used the solo tuning mentioned above to create more tension on the
gut strings, therefore producing more sound and a more incisive, soloistic quality. One
current argument is that with the technological advances in metal strings, the solo tuned
bass is no longer necessary since the metal strings render a less noticeable difference
between the tunings than their gut predecessors.20 An historical summary of the various
changes and inconsistencies in tuning the double bass is outside of the scope of this
document. It is sufficient to say that even today there are many players who do not use
the standard tuning of the double bass, choosing their tuning scheme based more on
individual preference than any other factor. Since publishers now have the capability to
change the key of an accompaniment part with the push of a button, many solo works are
17
Koussevitzky, Complete Double Bass Recordings.
18
Serge Koussevitzky, Four Pieces for Double Bass and Piano, ed. by David Walter, (Cincinnati: Liben
Music Publishers, 2001).
19
Koussevitzky, Reverse image copyist’s score, Chanson Triste.
20
Gary Karr, “Karr Talk: Name That Tuning!” International Society of Bassists, Vol. XX, No. 3, Winter
1996, 4.
8
published with piano parts in two keys, one in orchestral tuning and one in solo tuning.
This allows the bassist to finger the solo the same way regardless of the tuning scheme.
However, it was not until David Walter’s 2001 edition that the four short Koussevitzky
compositions were accompanied by piano and bass parts written in the same key,
indicating it should be played using orchestral tuning.21
Significant advances in recording technology have also been made during the
twentieth century. How the limited technology available to Koussevitzky in 1929
effected his recording is impossible to determine for certain.
This study examines the most striking differences between the Koussevitzky and
Karr recordings in regards to performance practice, focusing on tempo manipulation
(tempo rubato), vibrato, and portamento. Chapter Two analyzes tempo manipulation as it
applies to recordings of Valse Miniature and Chanson Triste. Chapter Three focuses on
the use of string instrument vibrato during the late nineteenth-century and includes an
analysis of Koussevitzky’s use of vibrato. The use of portamento by the Koussevitzky
and Karr recordings of Chanson Triste is examined in Chapter Four. Chapter Five brings
together the material collected in previous chapters and presents a detailed method for
creating an historically informed performance of the two pieces Koussevitzky did not
record, Andante and Humoresque. Chapter Six concludes the study by addressing the
issue of authenticity and balancing the primary source material (the recordings of
Koussevitzky performing his own compositions) with the expectations of a twenty first-
century audience.
21
Koussevitzky, Four Pieces, ed. Walter.
Chapter Two
Tempo Manipulation in Valse Miniature and Chanson Triste
Tempo Rubato
A great deal of research has been dedicated to the practice of tempo modification
during the later Romantic period and early twentieth century. The rubato of Chopin is
well documented22 and several books contain information or are entirely dedicated to the
history of rubato. When it comes to manipulating the tempo where no indication is made
by the composer, there are essentially two types of rubato. The first deals with changing
note values or rhythms without altering the beat. For instance, the accompaniment stays
steady and the musician is free to stretch or push individual notes or rests, or to
manipulate melodic passages throughout a phrase.23 This type of rubato will be referred
to as ‘type A’ throughout this document. The second form of rubato, hereafter referred to
as ‘type B’, deals with altering the beat. Wherein the basic pulse of the music changes
either for a short time or for an extended passage for structural, dramatic or expressive
purposes.24 While there are examples of both types of rubato in the recordings of
22
Richard Hudson, Stolen Time: A History of the Tempo Rubato, (New York: Oxford University Press, Inc.
1994), 175-237.
23
Clive Brown, Classical and Romantic Performance Practice: 1750-1900, (New York: Oxford University
Press, Inc. 1999), 378.
24
Ibid, 377-378.
10
Koussevitzky and Karr, the type B rubato is the most common type associated with later
Romantic performance practice.25
Valse Miniature
By examining Valse Miniature in regards to tempo modification, the first
noticeable difference is the slower overall tempo that Karr chooses. The performance
length of Karr’s rendition is three minutes and twenty-three seconds26 compared to
Koussevitzky’s two minutes and thirty-four seconds.27 The introduction in each
performance foreshadows each artist’s interpretation. Koussevitzky’s accompanist, Pierre
Luboshutz28, maintains a steady beat during the four bar introduction, using a modest
amount of the type A rubato, but generally maintaining a consistent downbeat (dotted
half note equal to 54 beats per minute). It is not until Koussevitzky enters in the fifth bar
that there is any large sense of tempo manipulation. Even though Koussevitzky utilizes
both the type A and B rubato throughout his performance he reserves significant changes
in tempo in order to emphasize the transitions from section to section. Within the first A
section (mm. 5-20) the downbeat stays relatively consistent. This section is divided into
four phrases, four measures each, that play out as an antecedent-consequent, antecedent-
consequent pattern which is repeated. Two-measure groupings are evident within each
phrase. Example 2.1 shows the antecedent-consequent relationship.
25
Hudson, 300.
26
Karr, Spirit of Koussevitzky.
27
Koussevitzky, Complete Double Bass Recordings.
28
Ibid.
11
Example 2.1 (a)-(d). Antecedent-consequent pattern, section A, Valse Miniature.
(a) Antecedent 1: mm. 5-8 (also mm. 43-46, 91-94).
(b) Consequent 1: mm. 9-12 (also mm. 47-50, 95-98).
(c) Antecedent 2: mm. 13-16 (also mm. 51-54, 99-102).
(d) Consequent 2: mm. 17-20.

12
By using the “Tap” feature on the Dr. Beat metronome it easy to measure the
changes in tempo throughout this section.29 As mentioned above, the four measure piano
introduction is consistent with a dotted half note equal to 54 beats per minute. When the
soloist enters, the tempo slows slightly for the first antecedent, but only to approximately
48 beats per minute. The consequent increases to a tempo slightly faster than the original
tempo established during the introduction to 60 beats per minute, only to give way to the
restatement of the antecedent at the original tempo. He stays at 54 beats per minute until
the middle of the next consequent where there is a slight acceleration into the first ending.
Koussevitzky uses a slight, momentary rubato in mm. 5-6 as well as in subsequent
statements of the antecedent. This manipulation gives the listener the impression of
accelerating slightly at the beginning of m. 5 and holding back for the last three eighth
notes of m. 6. This is achieved while generally maintaining a steady downbeat from
measure to measure. The first consequent is steady and is followed by a restatement of
the first antecedent. This statement of the antecedent mirrors the first statement. The
following consequent seems to accelerate during mm. 17-18; however, the overall tempo
only increases slightly, with the piano slowing in m. 20 before the repeat.
The repeat of the A section, while varying to a slight degree from the pattern
established above, follows the same general contour throughout until it reaches the
second ending. At this point the pianist does not slow down, giving the impression of
moving forward into the new section. There is a beat of indeterminate silence before the
anacrusis to the B section. Overall the A section has a distinct sense of push and pull. The
tempo seems to sway back and forth using the 54 beats per minute, established during the
29
The “Tap” feature of the Dr. Beat DB-88 metronome indicates in beats per minute how fast an individual
is tapping a pad on the control panel. By tapping the beat along with the recording one can come very close
to establishing the actual tempo of a section, measure or individual beat of music.
13
introduction, as point of departure. While Koussevitzky does move the tempo faster and
slower, he never wanders very far in either direction. It is in the upcoming B section that
the first dramatic change in overall tempo is heard.
Gary Karr’s rendition of the A section is significantly different than
Koussevitzky’s, but does maintain some traits of late Romantic performance practice
techniques in regard to tempo fluctuation. As mentioned above, the first noticeable
difference is the tempo established by the accompanist. The introduction to Karr’s
version gives the impression of a slower tempo, since the pianist does slow down in m. 4.
The established tempo is very close to the tempo established by Koussevitzky’s
accompanist, but this is where the similarities end. By placing a slight rallentando on the
second and third beats of m. 4 and changing the placement of the following downbeat
Karr’s accompanist has foreshadowed his approach to rubato throughout the section. The
significant difference between Karr and Koussevitzky’s rendition of the A section are
evident in the statement of the first antecedent. Karr starts the antecedent at tempo and
accelerates through mm. 5-6, with only a slight hesitation near the end of m. 6. The
tempo then slows significantly in m. 8. The consequent follows the same pattern; there is
a significant acceleration during the first two measures, followed by a retardation of the
tempo in m. 12. The restatement of the antecedent mirrors the first phrase, but the
consequent, after a slight push forward in m. 17, slows throughout mm. 18-20. This is a
dramatic departure from Koussevitzky, who placed a slight accelerando in m. 18.
Throughout the A section Karr pushes the tempo during the measures that have running
eighth notes only to pull back during the measures that contain sustained notes. With the
exception of the first couple of notes in each statement of the antecedent, Karr’s eighth
14
notes are relatively equal in length, albeit faster than the established tempo. Koussevitzky
alters the length of his eighth notes during the first two measures of each phrase.
Applying the “Tap” feature to Karr’s performance and establishing an overall
tempo for the A section is somewhat more challenging than applying it to Koussevitzky’s
rendition. As mentioned above, the tempos established by the accompanists are very
similar (54 beats per minute), but where Koussevitzky leaned more toward the type A
rubato, only moving slightly on each side of the established tempo and maintaining a
consistent downbeat, Karr moves significantly and utilizes the type B rubato to a much
greater degree. Overall, Karr’s tempo averages somewhere between 40 and 50 beats per
minute with the first two measures being faster than the second pair of measures in each
phrase. The slower average tempo is due to a significant slowing of the tempo at the end
of each phrase. The fast measures accelerate to as quick as 63 beats per minute, and the
slower measures decelerate to as slow as 36 beats per minute.30 There is not a sense of
maintaining a steady downbeat, and the time each measure occupies changes significantly
throughout Karr’s A section.
A mathematical analysis of the overall tempo of each performer’s A section
further indicates his use of the different types of rubato. This section, including the
introduction and repeat is 36 measures in length and takes Koussevitzky 40 seconds to
complete. The result is an average tempo of 54 beats per minute for his A section, which
is consistent with the starting tempo determined by using the Dr. Beat metronome. This
supports the conclusion that Koussevitzky utilizes the type A rubato, since his overall
average tempo and measure to measure tempo are the same. Karr’s A section takes 49
30
This still uses a dotted half note as the beat, although to establish this it was necessary to “tap” quarter
notes and divide the tempo by three.
15
seconds to complete even though he begins at the same pace as Koussevitzky. The result
is an average tempo of 44 beats per minute, which is somewhat slower than the tempo
established at the beginning, and supports the conclusion that he utilizes the type B
rubato to a greater degree since his starting tempo and overall average tempo are
significantly different.
The B section, beginning with the pickup to m. 23 and ending at m. 42, is a
section that each performer interprets differently in terms of tempo. As has been
established, Karr’s rendition of this piece is 49 seconds longer that Koussevitzky’s, and
the B section is the first section to contribute significantly to the difference.
Koussevitzky takes 28 seconds to perform this short section and Karr takes 40 seconds.
As mentioned earlier, when Koussevitzky gets to the second ending of the A
section, there is no slowing of the tempo. After a rest in the second measure of the second
ending (m. 22) Koussevitzky establishes a new tempo with his pickup note that is steady
for several measures. This increased tempo, now approximately 74 beats per minute,
drives this section forward until a somewhat sudden change to 66 beats per minute in
m. 35. From this point the tempo slows slightly until a very deliberate ritardando in
mm. 39-42. When Koussevitzky begins to slow the overall tempo in m. 35 he still
projects a sense of moving forward until m. 39. He does this by slightly increasing the
tempo of the eighth notes in mm. 35 and 37 while still maintaining a gradual slowing of
the overall beat. Measures 35 and 37 begin slightly slower than they end, but
Koussevitzky still allows the sequence to slow during both iterations (Example 2.2).
16
Example 2.2. mm. 34-38. Valse Miniature.
Koussevitzky’s ritardando in mm. 39-42 is very dramatic. He uses the octave leap
in m. 40 to slow the beat from a dotted half note being equal to approximately 65 beats
per minute to a quarter note being equal to approximately 54 beats per minute with a
continued ritardando until the end of the section at m. 42 (Example 2.3).
Karr’s approach to the B section is different. The sound never stops between the
A and B sections as it does during the second ending of Koussevitzky’s version. This
interpretation causes the listener to perceive the sustained note in the second ending as a
fermata instead of a set rhythmic duration. Where Koussevitzky establishes faster tempo
with the anacrusis, Karr maintains the slower tempo but shifts to more of a type A rubato
for mm. 23-31. During this portion of the B section Karr’s performance mimics the type
A rubato used by Koussevitzky in the A section. The B section contains two measure
groupings with a forward push of the tempo during mm. 23, 25, 27, and 29 and a
hesitation after each half note in mm. 24, 26, 28, and 30. Unlike Koussevitzky, who
17
maintained a steady tempo during this section, Karr alternates the length of every other
measure. Each two measure group takes up the same amount of time, but the odd
numbered measures are faster, and therefore shorter, than the even numbered measures
(Example 2.4). This projects a sense of swaying back and forth around the established
tempo. The overall tempo during these measures stays relatively consistent at around 48
beats per minute when the two measure groupings are considered, which is very close to
the average tempo Karr used in the A section.
Both performers elongate the last note of m. 30 to emphasize the octave leap and
the end of the phrase (Example 2.4, last note). Instead of simply establishing a slower
tempo during the second half of the B section, as Koussevitzky did, Karr places a
ritardando in m. 32 which is continued by the accompanist in mm. 33-34 under Karr’s
sustained note. The sustained note is held through the rest in m. 34 and the pick up to
m. 35 is held out of tempo. The pattern, increasing the tempo during the measures with
faster rhythmic motion and slowing during measures that contain sustained notes,
continues until mm. 39-40 where there is a significant ritardando. This ritardando is taken
18
even further by the accompanist under the sustained note in m. 41, then in m. 42 the
chromatic descent slows freely and leads back to a restatement of the A section (Example
2.3).
While this document deals expressly with the recorded performances of Karr and
Koussevitzky it should be noted that in both the International31 and Liben editions32 of
this work the only indication of tempo fluctuation occurs at the end of the B section. Fred
Zimmerman’s 1949 International edition indicates rit. beginning in m. 39 and tempo
when the A section returns in m. 43. David Walters’ 2001 Liben edition indicates poco a
poco rall. under the second beat of m. 40, and a tempo as the A section returns at m. 43.
The A section that follows is identical to the first statement, with the exception of
the last four bars of the second consequent and the omission of the repeat (Example 2.1,
a-c). During this A1 section both performers stay very close to where they started with
regard to overall tempo and the use of rubato. The only noticeable difference occurs in
the second consequent at the point where the music changes. In m. 54 there is a slurred
leap up from a half note C# to the quarter note G which occurs in the first statement. The
downbeat of m. 55 is another articulation of the G, but this time it is a quarter note and is
followed by another G an octave lower (Example 2.5). Koussevitzky uses this as a point
of momentary hesitation before moving to the end of the phrase. In the middle of m. 56
he accelerates to the end of the phrase moving the tempo forward as a foreshadow to the
upcoming variation of the A section. Karr’s treatment of the altered consequent is similar
to Koussevitzky’s in that he uses the C# to G in m. 54 as a point of hesitation. Karr,
however, does not increase the tempo at the end of the phrase. He concludes this phrase
31
Serge Koussevitzky, Valse Miniature for String Bass and Piano, Opus 1, No. 2, ed. by Fred Zimmerman
(New York: International Music Company, 1949).
32
19
in a very relaxed way that is similar to his conclusion of the first A section. As
Koussevitzky did, Karr foreshadows his interpretation of the following section with his
treatment of this consequent (Example 2.5).
The ensuing section, mm. 59-90, is a conversational variation of the A section.
The accompanist plays the antecedent that has previously been played by the soloist to
mark the beginning of each A section. The bassist takes on an accompanying role for two
measures. Instead of playing straight quarter notes, as was the case when the pianist
accompanied the bassist, the bassist now plays a syncopated pattern, then echoes the
motive heard in the first measure of this section by the pianist (Example 2.6).
bass
piano
20
This interplay between bassist and accompanist continues throughout the section. The
music maintains the same overall shape as the A section, but the motives are passed back
and forth during the antecedent phrases.
It is during this variation on the A section that Koussevitzky makes his most
dramatic changes in tempo. The syncopated effect is somewhat ambiguous throughout
the section due to his use of both types of rubato, coupled with the way he connects the
syncopated notes. The connectivity and fluid tempo sometimes obscures the
syncopation. Another reason the syncopation sometimes seems unclear could have to do
with the recording technology of the time. The bass is the most prominent sound on the
recording, even when it should be in an accompanying role. When the motive first
appears in the piano it is overpowered by the bass. This, coupled with the use of rubato
throughout the performance, makes the beat difficult to find when this section begins and
therefore further obscuring the syncopation. The listener is unsure whether there is a
rhythmic error occurring or if the passage is being played accurately. It should be noted,
however, that rhythmic accuracy did not take precedence over musical effect during the
early twentieth century, and Koussevitzky’s performance of the syncopated patterns is
more likely a convention of that period. Koussevitzky’s tempo, once established, is very
similar to the opening and utilizes the type A rubato for the first several measures.
However, he adds a significant stretch on the held notes at the end of each phrase, thus
changing the beat and incorporating the type B rubato. This elongation of what are
generally the highest notes in the phrase occurs in m. 62 (Example 2.6) and again in mm.
65-66 (Example 2.7).

21
Measures 71-74 mark the point of most dramatic tempo manipulation by Koussevitzky in
the piece. At this point the quarter note slows to 66 beats per minute, effectively changing
the pulse from a dotted half note to a quarter note (Example 2.8).
After a point of indeterminate hesitation in m. 74, Koussevitzky returns to a tempo only
marginally faster than what he established at the beginning of this section and moves
forward. He continues to elongate the higher sustained notes that occur at the end of each
phrase in mm. 78, 82 and 86. The final consequent of this A section is a slightly
compressed version of the final consequent of A1 (Example 2.5) discussed earlier, and is
handled in much the same way (Example 2.9).

22
Karr’s interpretation of this section does not vary significantly from
Koussevitzky’s. The most prevalent difference is, again, the overall slower tempo he
chooses. Many of the same manipulations of tempo occur in Karr’s rendition; however,
they do not appear to have the same effect due to Karr’s slower tempo of approximately
48 beats per minute. The syncopation is clear from the beginning of the section, and the
balance between bassist and pianist is appropriate. As mentioned above, this could be due
to advances in recording technology. The first place Karr deviates from Koussevitzky’s
example occurs in mm. 65-66 (Example 2.7). Karr employs a more significant slowing of
the beat and holds the sustained note in m. 66 longer than Koussevitzky. Karr also slows
down in mm. 71-74 as Koussevitzky did (Example 2.8); however, relative to the tempo
he has chosen, this segment of Karr’s type B rubato does not appear to be as dramatic as
Koussevitzky’s. The peak note for Karr is the high sustained half note in m. 82. He sets
this up by employing the type B rubato in the preceding phrases, specifically in
mm. 77-78 and 81 (Example 2.10).

23
As related above, Koussevitzky drives forward between mm. 75-81. Following suit with
his conclusion to the A1 section, Karr chooses a much more relaxed approach to the final
antecedent-consequent of this section.
Measure 91 begins the final statement of the A theme before the end of the work.
Measures 91-102 are identical to both the A and A1 statements (Example 2.1, a-c). Both
performers handle this iteration as they did previously, although Koussevitzky increases
his tempo marginally to 60 beats per minute. The transition to the closing material occurs
at m. 103 where the second consequent phrase should be. This is the same point in the
antecedent-consequent relationship that section A1 deviated from what had been
established in A (Example 2.5). The closing material takes the place of the second
consequent.
Measure 103 marks the beginning of the closing section with a descending
diminished arpeggio. Koussevitzky uses this as a vehicle to begin his race to the end. He
slurs the arpeggio in groups of two and lands in m. 104 at an increased tempo of 71 beats
per minute, which is maintained until the end of the piece. The next six measures consist
of tonic and dominant arpeggios, first presented as eighth notes, then triplets. He
maintains a consistent tempo, but the written part creates the sense of acceleration. The
last six measures of his performance are simply a means of ending the work. Measure
110 consists of the same arpeggio heard in m. 103, only this time it is ascending. While it
is written as straight eighth notes, Koussevitzky takes liberty with his own composition
and maintains a triplet feel, significantly changing the rhythm and employing rubato to
the end.33 The final flourish of the tonic arpeggio in triplets with a lower neighbor in mm.
112-113 is altered as well. By the time Koussevitzky gets to m. 113 he has abandoned the
33
Koussevitzky, Valse Miniature, ed. Zimmerman.
24
music and is moving forward, playing a harmonic arpeggio until he arrives at the
dominant-tonic final cadence. Koussevitzky takes advantage of the early twentieth-
century notion that the motion and overall intent of the music is more important than each
individual note. This use of the type A rubato and the exciting concluding flourish creates
an exciting and flashy end to the piece (Example 2.11).
Example 2.11. Closing Material, mm. 103-end. Valse Miniature.
Karr’s ending is equally as exciting, although he employs the type B rubato,
demonstrating a different interpretation. As it has been throughout the piece, his tempo is
slower than Koussevitzky’s. Another significant difference is that every note is clear,
articulated and audible. Although the tempo only increases marginally, there is a distinct
sense of acceleration to the end. After elongating the descending arpeggio in m. 103 as
25
Koussevitzky did, Karr incorporates a slight acceleration in tempo from m. 104, which is
approximately 48 beats per minute, to m. 110, which is approximately 63 beats per
minute. Measure 110 slows marginally while ascending to m. 111. After the downbeat of
m. 111 he prolongs the rest therefore creating a moment of hesitation to set up the final
flourish from m. 112 to the end. Karr takes advantage of the increase in written rhythmic
intensity, conveying a sense of ending flourish to the listener without sacrificing any of
the cleanliness expected from modern performance practice. Karr sets up the final
flourish by stretching the rest in m. 111 (Example 2.11).
Chanson Triste
As with the Valse Miniature, the title of the Chanson Triste establishes the overall
tone for the performance of the piece. Triste is defined by the New Harvard Dictionary
of Music as:
Triste. (1) [Fr., Sp.] A love song of sad character found in Peru and
Argentina, closely related to the Andean yaraví. The triste is slow, varied
in meter, and often with an alternating or concluding section in faster
tempo.34
The tempo indication supplied by Koussevitzky at the beginning of the piece is con
tristezza or “with sadness”.35 Both Koussevitzky and Karr use rubato as one of the tools
to relate sadness to the listener.
As in the Valse Miniature, the first noticeable difference when comparing
Koussevitzky and Karr’s recordings of the Chanson Triste is the tempo taken during the
three measure introduction. Koussevitzky’s accompanist establishes a slow tempo of
34
Don Michael Randel, ed., The New Harvard Dictionary of Music (Cambridge: Harvard University Press,
1986. Repr. Ed., 2001), 874.
35
26
approximately 46 beats per minute, and as was the case in Valse Miniature, does not
deviate until Koussevitzky enters on the down beat of the A section (Example 2.12, a).
Karr’s accompanist begins the piece at a slightly faster pace, establishing a tempo of 54
beats per minute. Unlike Koussevitzky’s accompanist, Karr’s includes a metered
ritardando by adding a fourth measure to the introduction. The introduction is written as
three measures. However, Karr’s accompanist establishes the beat and performs four
measures, placing what is written on beat three of m. 3 on the downbeat of what would be
m. 4 and making it a dotted half note (Example 2.12, b).36
Example 2.12. mm. 1-3. Chanson Triste.
(a) The introduction as written and played by Koussevitzky’s accompanist:
(b) The introduction as played by Karr’s accompanist:
36
The International Edition (1949) and the copyist’s score (1896) both have a three measure introduction.
The David Walter edition (2001) indicates a four measure introduction.
27
Koussevitzky enters after the introduction and any sense of a steady beat is lost
for several measures. The first two phrases of the A section make abundant use of the
type B rubato. The tempo immediately slows during mm. 4-5 to approximately 33 beats
per minute, but to say this is an established tempo would not be accurate. On beat three of
m. 5 Koussevitzky elongates the triplet pattern, slowing the tempo even more until he
arrives at m. 6. Measure 7 is not metered. Koussevitzky ignores his own rests in m. 7 and
enters with the downbeat of m. 8 as the notes in the previous bar decay (Example 2.13).
Example 2.13. mm. 4-7. Chanson Triste.37
Koussevitzky enters early after a held note or during a rest several places
throughout the A section, and the rest of the piece. This obscures the beat and maintains
a sense of free or non-metered time. There are instances when the loss of time is almost
metric. In m. 14 there is a half note followed by a quarter note; Koussevitzky moves to
the quarter note almost exactly a half a beat early, making this particular measure sound
as if it is written in 5/8 time.
Just as there are several occasions at which Koussevitzky loses time, there are as
many at which he gains time. This usually comes when there is a moving pattern, such as
37
Examples of the bass part for Chanson Triste are written as it sounds (solo tuning).
28
the triplets in m. 5 (Example 2.13), or when three eighth notes follow a dotted quarter
note, as in mm. 10, 13, 16, and 17 (Example 2.14).
Example 2.14. rhythmic pattern mm. 10, 13, 16, and 17. Chanson Triste.
Usually this stretching of the rhythm is accomplished by prolonging the first eighth note
with the following two eighth notes in the previously established tempo. This is often
emphasized by the use of portamento, which will be discussed in Chapter Four.
Karr’s treatment of the A section is similar to Koussevitzky’s in some aspects but
differs significantly in others. Both performers enter at a slower tempo than their
accompanists established during the introduction; however, Karr’s beat is easier to
discern than Koussevitzky’s. A tempo of approximately 43 beats per minute is
established by Karr in m. 4, and his accompanist and is steady until the triplet pattern in
m. 5. As Koussevitzky did, Karr slows the tempo during the triplets making m. 5 longer.
The sustained note that follows in m. 6, however, is given its full value and is in tempo.
Where Koussevitzky ignored the rests and entered early in m. 7, Karr enters late,
stretching the rests to create a slightly longer pause before his entrance in m. 8 (Example
2.13). Karr follows this pattern throughout the A section.
At the points where Koussevitzky entered early, Karr tends to hesitate or stay very
close to his established tempo. Both performers use occurances of faster rhythmic
29
motion to manipulate the tempo. Karr, however, slows down during the eighth notes as if
there were a rallentando indicated (Example 2.14). Koussevitzky increases and decreases
the tempo within the eighth note patterns, creating an ambiguous sense of time. This
concept of temporal ambiguity is something that permeates Koussevitzky’s performance
of the A section. Karr’s performance, while using a significant amount of rubato with
equally dramatic results, maintains an overall sense of the beat that does not exist in
Koussevitzky’s performance.
The B section begins at m. 20 after a repeat of the A section and ends before a
short cadenza at m. 35. Più mosso is indicated in both the accompaniment and solo parts
at m. 20.38 Koussevitzky increases the tempo in m. 20 to approximately 75 beats per
minute, only to relax slightly in m. 21. This increase and relaxation of the tempo is
repeated in mm. 22-23 as well. The accompaniment mirrors the added motion with
sixteenth notes on the first two beats of each or these measures. Koussevitzky stays very
close to the 75 beats per minute established at m. 20 for the first half of the B section
holding back slightly at the end of mm. 24, 25, and 26. At m. 27 there is a descending
eighth note pattern that leads to m. 28. Koussevitzky increases the overall tempo again,
accelerating as the pitch descends and slowing slightly as the pitch ascends through m. 31
(Example 2.15).
38
30
At m. 32 an accelerando is indicated.39 Koussevitzky, who has accelerated the
tempo from m. 28, stays steady, waiting until m. 33 to deliberately increase the tempo
until the fermata at m. 35 that marks the end of the B section. Throughout the B section
Koussevitzky increases and decreases the tempo, but in mm. 28-35 there is an overall
acceleration that culminates by reaching a tempo of 138 beats per minute in m. 34
(Example 2.16).
Throughout the B section Karr’s tempo becomes more ambiguous. He also
creates a sense of more motion, as indicated, when he gets to m. 20; however, he sets this
up in a different way than Koussevitzky. In the repeat of the A section he slows
significantly, to less than 38 beats per minute, before the sustained note and subsequent
silence that precedes the B section. This ritardando helps Karr convey forward motion at
m. 20, even though his tempo does not increase significantly. For Karr, it is the long
notes in the first half of the B section that are elongated and the shorter notes that
accelerate. The dotted quarter note in mm. 20 and 22, as well as the quarter notes in
m. 21 are at a tempo of approximately 54 beats per minute, while the eighth notes in the
same measures are significantly faster. The increasing and decreasing tempo that
Koussevitzky used in his rendition is also used by Karr, but in a different way. Karr’s
39
Ibid.
31
tempo fluctuations occur within the measure and exaggerate the rhythm. The short notes
(eighth notes) are faster and the long notes (dotted quarter notes and quarter notes) are
slower (Example 2.17).
Karr also increases the tempo at m. 27. The tempo becomes relatively steady at
approximately 73 beats per minute, although he does stretch the third beat of each
measure (Example 2.15). As he approaches the accelerando indicated in m. 32 he slows
the tempo. Instead of increasing the speed from mm. 32-35 Karr stays close to 70 beats
per minute and adds a ritardando at m. 34 which leads to the fermata in m. 35 (Example
2.16).
Both performers treat the cadenza in a slightly different manner. Both of them are
playing the same notes, but the speed at which they play is different. Koussevitzky holds
the fermata at the beginning of the cadenza for only a second, and then takes a total of 17
seconds to get from the end of the sustained note at the beginning of the cadenza to the
last fermata before the transitional material that begins with a pick up to m. 36. Karr
holds the fermata at the beginning of the cadenza for four seconds, and 25 seconds pass
before he releases the last fermata and begins the pickup to m. 36. The length of time it
takes each performer to play the cadenza reflects his overall approach to the work. Karr
32
tends to be more concerned with the clarity of each individual note and is more deliberate
in many ways than Koussevitzky.
A three-measure transition separates the end of the cadenza from the final
statement of the A section, which begins at m. 39. This statement of the A section is
identical to the first statement, with the exception of the last measure. A dotted half note
is written to fill the last measure of the final statement, where a quarter note followed by
two quarter rests was written at end of the first statement. The only indication of tempo
given in the score is Tempo I.40
Koussevitzky treats the final statement of the A section in much the same way he
treated the first statement, although he maintains a comparatively steady pace when not
deliberately altering the written rhythm. The tempo is consistent when he enters in m. 39
at 43 beats per minute. He still uses a significant amount of rubato; however, throughout
this final statement of A he stays very close to his established tempo. The rubato during
this section is reserved for specific rhythmic events such as the triplet figure in m. 40
(which corresponds to m. 5, Example 2.13). He still ignores the rests that happen after
the first phrase, entering as the last note of the phrase decays, but the majority of the
second phrase is steady. He adds a ritardando at the end of the piece and holds the last
note until it dies away.
Karr performs the last statement of the A section in a manner similar to his first
statement. His tempo is close to Koussevitzky’s at m. 39; however, he does increase and
decrease the tempo more than Koussevitzky did. He slightly exaggerates the rubato used
at the beginning of the piece, elongating the faster rhythmic events and adding slight
40
Ibid.
33
hesitations between phrases. As Koussevitzky did, he adds a ritardando at the end and
lets the final note decay.
Karr’s performance of Chanson Triste takes five minutes and thirty-four seconds;
almost a full minute longer than Koussevitzky’s which takes four minutes and thirty-five
seconds. As mentioned in Chapter One, the advances in recording technology could have
an influence on the differences in performance time. The playback speed of
Koussevitzky’s recordings may be faster than his actual performance. The fact that his
pitch in Chanson Triste is approximately fifteen cents higher than Karr’s supports this
hypothesis. However, this conclusion assumes that both performers used the same exact
tuning pitch, which is impossible to determine. In addition, the tempo at the beginning of
both the Karr and Koussevitzky recordings is almost identical. Regardless of whether or
not the playback speed of Koussevitzky’s recording has been altered, the relative changes
in tempo would be proportional. The exact tempo he used may not be what is indicated
in this chapter, but the relative flexibility of tempo, and the comparison to Karr’s tempo
manipulation, is still accurate.

Chapter Three
Vibrato in Valse Miniature and Chanson Triste
Vibrato
One of the most significant changes in performance practice that occurred during
Koussevitzky’s life was the use of vibrato. There is not much written concerning the use
of vibrato as it applies to the solo bassist during the late nineteenth and early twentieth
centuries, but there are several sources that discuss its use by the solo violinist. Between
1750 and 1900 vibrato, in a variety of forms, was used almost entirely as an ornament.41
The accepted ruling tone of the period was steady and without oscillation. The use of
vibrato at the end of the nineteenth century was controversial and its acceptance can be
summed up by Cecil Forsyth (1914):
The Strings had to fight for centuries before they were allowed to use
vibrato without question. It was regarded as a disreputable way that
bad fiddlers had of making music sound emotional. Everybody protested
against it, tooth and nail, as a circus-clown’s way of playing. Leopold
Mozart, Spohr, and Courvoisier all took this view…Clear bowing was the
thing they wanted, and a nice variety in dynamics. Vibrato was taboo.42
It should be noted that Forsyth’s Orchestration does not include a discussion of vibrato in
the chapter on string technique. Rather, it is included in the addenda near the end of the
41
Brown, 521.
42
Cecil Forsyth, Orchestration (London: Macmillan and Co., 1914; reprint, New York: Dover
Publications, Inc., 1982), 494.
35
treatise.43 It was during the 1920s that vibrato became more widely accepted, although
there were still teachers who regarded it as an effect, such as Leopold Auer (1921):
violinists…those who are convinced that an eternal vibrato is the secret

of soulful playing, of piquancy in performance-are pitifully misguided
in their belief…No, the vibrato is an effect, an embellishment…44
During the same decade there were others who began to argue that the absence of vibrato
had become the embellishment instead of the norm. It was during the 1920s and 30s that
the continuous vibrato to which modern audiences are accustomed to became
fundamental to the technique of string playing.45
The effect this had on Koussevitzky’s use of vibrato and his interpretation of his
own works is impossible to surmise. Koussevitzky was a double bass student of the late
nineteenth century, and one can assume that he was taught to use vibrato very sparingly
and only as an embellishment. When Koussevitzky made the recordings in 1929, the
accepted use of vibrato had clearly changed from the time he composed the pieces during
the 1890s. As a conductor and player, Koussevitzky had to have been aware of these
changes. This is demonstrated by his frequent use of vibrato during his performance of
both Chanson Triste and Valse Miniature. Vibrato had moved from being an
embellishment of the tone to an expected attribute of string playing.
To explore performance practice in terms of vibrato, the first comparison made
will be Koussevitzky’s use of vibrato in the slow, lyrical Chanson Triste to his use of the
technique in his performance of the faster, more technical Valse Miniature. Then a direct
43
Ibid, 493.
44
Leopold Auer, Violin Playing As I Teach It (New York: 1921), 22-3.
45
Robert Philip, Early Recordings and Musical Style: Changing Tastes in Instrumental Performance,
1900-1950, (New York: Cambridge University Press, 1992, Repr. Ed. 1994), 101.
36
comparison can be made between how Koussevitzky employed the technique in 1929 to
Karr’s use of the technique in 1989.
Koussevitzky and Chanson Triste
Koussevitzky’s approach to vibrato during his recording of Chanson Triste is
evident during the A section. Throughout the A section, and its repeat, he employs an
almost constant vibrato on notes that are a beat or more in length. In general, oscillations
are the same from note to note with regard to the speed (how fast the oscillations occur)
and depth (the amount of deviation from the original pitch). The speed of the oscillations
is consistently fast and does not alter based on the length of the note. The depth does not
move a great deal from the primary pitch; however, it does vary enough for the listener to
notice the vibrato. Two factors that effect Koussevitzky’s choice to apply vibrato to a
given note are: 1) the length of the note in real time (i.e. how long does the note sound
regardless of its’ rhythmic value); 2) the application of portamento. Koussevitzky
generally applies vibrato to those notes which are longer and omits vibrato from those
notes on which he employs portamento.
This application of vibrato can be heard throughout the A section (mm. 4-19), and
is evident during the first note he plays on beat one of m. 4 (Example 3.1). The tone
begins soft and steady without vibrato, as the volume increases during the beat
Koussevitzky adds vibrato. This quarter note is approximately 2.7 seconds in length with
the vibrato beginning almost a full second after the note sounds. Koussevitzky does not
gradually increase the speed or depth of the oscillation to introduce vibrato to the note,
rather the vibrato begins after the pitch has been given enough time to solidify in the
37
listener’s ear. The next note, which occurs on beat three of the same measure, also
includes vibrato, but it is not as prominent. This quarter note is shorter in length
(approximately 1.5 seconds long) and not as loud. Beat one of the following measure is a
quarter note that is similar to beat one of m. 4. This quarter note is approximately 2.1
seconds in length. During the triplet pattern that follows in m. 5 there is an absence of
vibrato. As Chapters Two and Four discuss, there is a very prominent use of portamento
and rubato during this pattern. While the first eighth note triplet of the pattern is
rhythmically shorter than the preceding quarter notes, Koussevitzky’s use of rubato
makes it long enough (approximately .8 seconds) to use vibrato. He chooses not to
though, instead employing portamento to emphasize the note. He does not use vibrato
again until the dotted half note in m. 6 (tied to a quarter note in m. 7). The vibrato on this
pitch begins almost as the pitch sounds and is similar in speed and depth to the previous
quarter notes. This note, the longest in terms of rhythmic value (four beats) and real time
(3.7 seconds) to this point, demonstrates the even manner in which Koussevitzky utilizes
this technique. Even though there is ample time to vary the speed of the oscillations,
Koussevitzky does not.

38
The consequent to example 3.1 occurs in mm. 8-11 (Example 3.2) and
Koussevitzky’s use of vibrato is very similar to the antecedent. The first quarter note is
approximately 2.3 seconds long (it sounds through the notated rest) and the vibrato is
similar, at this point, to that of the antecedent. Throughout the consequent, as well as the
following antecedent-consequent phrase (mm. 12-19), vibrato occurs on the longer notes,
as well as some of the eighth notes, with a consistent speed and depth.
It should be noted that in the A section there are instances of vibrato that occur at
the beginning of notes which are later embellished by portamento. This does not happen
often and occurs in situations such as m. 14 (Example 3.3). The half note in m. 14 has a
steady vibrato which ceases as Koussevitzky uses portamento to ascend to the next note.
The B section of Koussevitzky’s rendition also is somewhat predictable when it
comes to his use of vibrato (mm. 20-35). As in the A section, it is the longer notes on
39
which Koussevitzky vibrates. However, this section increases in tempo and has a faster
written rhythmic pattern, thus decreasing the number of notes to which the technique is
applied. Koussevitzky also vibrates the highest notes in most phrases regardless of their
rhythmic value. The notes which are circled in example 3.4 indicate where Koussevitzky
applies vibrato during the B section.
Example 3.4. mm. 20-35. B section. Chanson Triste.
It is notable that all of the dotted quarter notes are emphasized with vibrato as
well as most of the highest notes in each passage. Koussevitzky creates time to apply
vibrato to these higher notes (the circled notes in mm. 28, 30, 32, and 33 in example 3.4)
by combining the technique with rubato.

40
The cadenza is performed without vibrato with the exception of the three half
notes which occur in the transitional material near the end (Example 3.5). This is notable
since Koussevitzky has used vibrato throughout the performance on longer notes and
elects not to apply it to the longer notes in the cadenza.
Example 3.5. Transitional material at the end of the cadenza. Chanson Triste.
Koussevitzky used vibrato during the last statement of the A section as he did in
the first. The speed and depth of the vibrato remain consistent, even throughout the
ritardando that leads to the final note.
Robert Philip notes in his book, Early Recordings and Musical Style, that the
limited use of vibrato in early recordings by string players gives an impression of
inexpressive playing. He cites that, of the string players who lived through the transition
of vibrato from its use exclusively as an ornament (the traditional approach) to that of an
almost constant expressive tool (the modern approach), cellist Pablo Casals succeeded the
most in combining both styles. He indicates that Casals’ approach was to use vibrato
most of the time, but to vary its speed and depth based on the musical situation.46
Koussevitzky’s performance of Chanson Triste can be classified as transitional, based on
46
Philip, 105.
41
that statement. He does use vibrato frequently during the recording, but does not vary its
speed or depth.
Koussevitzky and Valse Miniature
Many similarities exist between Koussevitzky’s recordings of Chanson Triste and
Valse Miniature when considering his use of vibrato. While both pieces are different in
overall style, Koussevitzky’s approach to vibrato is the same. The B section of the
Chanson can be compared directly to the A section of the Valse (mm. 5-20). Both
sections have a fast rhythmic motion and Koussevitzky reserves his use of vibrato for
notes that are of longer rhythmic value or that are emphasized with tempo manipulation.
As presented in Chapter Two, the Valse begins with a four measure introduction, which is
followed by two antecedent-consequent phrases. Koussevitzky uses vibrato, often in
conjunction with a slight rubato, on most of the sustained notes that occur in the third and
fourth measures of each antecedent and consequent (Example 3.6). Vibrato is absent
from the final note of the first consequent, possibly because Koussevitzky fingered this
note as a harmonic instead of a stopped note (Example 3.6, b). This is particularly
noticeable during both the first iteration of the section, as well as the repeat, because the
previous dotted quarter note was embellished with vibrato.
Example 3.6 (a)-(d). Antecedent-consequent pattern, section A, Valse Miniature.
(a) Antecedent 1: mm. 5-8 (also mm. 43-46, 91-94).

42
(b) Consequent 1: mm. 9-12 (also mm. 47-50, 95-98).
(c) Antecedent 2: mm. 13-16 (also mm. 51-54, 99-102).
(d) Consequent 2: mm. 17-20.
Koussevtzky’s use of vibrato in the following section (mm. 23-42) is less
frequent, due to the increased tempo and faster rhythmic motion. Based on his previous
use of vibrato, the listener would expect to hear the half notes in mm. 24, 26, and 28
performed with this embellishment; however, this is not the case. Koussevitzky does not
use vibrato in this section until he reaches m. 30. The half note in this measure is
embellished not only with vibrato, but rubato and a portamento to the following quarter
note (Example 3.7).

43
Example 3.7. mm. 22-30. Valse Miniature
Vibrato is used throughout the remainder of this section on most sustained notes,
such as the dotted quarter note that is tied across the bar line in m. 33. There are notes of
shorter rhythmic value in this section which are embellished with vibrato as well, such as
the quarter notes on beat three of mm. 34, 36, and 38. Koussevitzky emphasizes the
eighth note on beat two of measure 40 by using vibrato and rubato. Circles around the
notes in example 3.8 indicate instances of vibrato during the last phrase of the section,
mm. 39-42.
Koussevitzky uses vibrato in the restatement of the A section, mm 43-58, as he
did during the first statement. A syncopated pattern begins in m. 59 with repeated pitches
(Example 2.6). This pattern returns again in mm. 63, 67, 75, 79, and 83. As mentioned
44
in Chapter Two, the rhythm during this syncopation is obscured; however, Koussevitzky
does use vibrato constantly during these repeated notes. While he uses his bow to convey
the rhythm, the vibrato produced by his left arm is the same, as if there were only one
long sustained note for two measures. Also discussed in Chapter Two is the significant
slowing of the tempo in mm. 71-72 (Example 2.8). These notes are also embellished
with vibrato to emphasize their importance.
Measures 91 through the end of the piece consist of another statement of the A
section and closing material. As with previous statements of the A section, Koussevitzky
does not change his use of vibrato. The rhythmic intensity and acceleration during the
closing material does not leave time for vibrato, and the technique is absent after measure
104 for the remainder of the piece.
Koussevitzky’s use of vibrato during performances of both Chanson Triste and
Valse Miniature is frequent but not constant. In addition, it does not vary significantly in
speed and depth. These attributes mark Koussevitzky’s use of vibrato as transitional
between the old style and the modern style. In 1924 Carl Flesch discusses the use of
vibrato during this time in his method, The Art of Violin Playing.
There are harsh differences of opinion as to whether vibrato should

be used in a continuous or intermittent manner. Purely theoretically,
the vibrato being a means of heightened expression should be used
only when the musical, expressive feeling justifies it. If we take a
survey of well-known violinsts of our time, we realize however that
practically all of them use vibrato constantly.47
He also discusses the depth and speed of vibrato.
The opposite [of a satisfying vibrato] is a vibrato that appears grafted

onto the sound and gives the impression that sound and vibrato don’t
belong together. In the last few years, with the increasing
47
Flesch, 20.
45
“industrialization” of musical life, this shallow manner has found its

home in places of entertainment of a lower order, so that one can, with
good cause, speak of a “coffee-house” or “movie-house” vibrato. The
tone becomes syrupy mush which through its continuous, unvaried
use of vibrato lacks all diversity and deeper, meaningful expressive
ability. This type of vibrato is too slow and too wide, and becomes
quite unbearable within a short time. In a general sense a rapid vibrato
is preferable to a slow one because it produces less deviation from the
original pitch and therefore the tone seems firmer and more stable.48
This quote supports Koussevitzky’s use of a fast vibrato that does not deviate far from the
original pitch. While any performance in any era can be classified as transitional in one
aspect or another, Koussevitzky’s use of vibrato is indicative of the trend away from the
old style and moving toward what is now thought of as the modern style.
Karr and Chanson Triste
The Oxford Companion to Music defines vibrato as:
A wavering of pitch used to enrich and intensify the tone of a voice

or instrument; it is practiced in particular by wind players, string players,
and singers. At present the technique is used frequently, and is commonly
held to be an important constituent of a competent player’s or singer’s tone.
In this form (often described as ‘continuous vibrato’) the device has been in
currency since the beginning of the 20th century, having been made popular
by such players as Kreisler. In consequence, the senza vibrato tone has
become a special effect in contemporary music.49
Karr’s use of vibrato during his 1989 recording of Chanson Triste is
representative of the modern style. It is constant and varies in intensity, depending on the
expressiveness of the music. The oscillations are slightly slower, which allows greater
deviation from the original pitch. There are instances in which Karr begins a note with a
steady tone then adds vibrato gradually. Throughout the performance there are very few
notes during the recording that do not have vibrato. When analyzing Karr’s performance
48
Ibid, 19-20.
49
Alison Latham, ed., The Oxford Companion to Music (New York: Oxford University Press, 2002), 1335.
46
it is more prudent to point out those instances that Karr does not use vibrato and to
highlight the differences in particular phrases and sections between his recording and
Koussevitzky’s.
Both performers make frequent use of vibrato throughout the A section of
Chanson Triste. Unlike Koussevitzky, Karr’s vibrato is not effected as much by the
length of the notes in real time. The first note Karr plays on beat one of m. 4 (Example
3.1) is approximately 1.7 seconds in length, and vibrato is in use from the initial attack of
the note. Koussevitzky’s first note was a full second longer than Karr’s and the vibrato
was not used until a full second after the pitch began. The next two quarter notes (beat
three of m. 4 and beat one of m. 5) are treated equally by Karr with the same intense
vibrato he used on the first note. Karr’s note on beat three of m. 4 is approximately 1.1
seconds in length, and there is no break between that pitch and the following quarter note
on beat one of m. 5. His vibrato does not stop, but sounds consistently throughout both
notes. Koussevitzky’s vibrato on beat three of m. 4 was considerably less intense than
the surrounding notes. The triplet pattern that follows is another point of significant
difference between the two performers use of vibrato. Koussevitzky omits vibrato
entirely until the dotted quarter note in m. 6. Karr, on the other hand, omits vibrato for
only the first three notes of the triplet pattern. The remainder of the pattern includes
vibrato. This demonstrates early in the performance that Karr will use vibrato on notes of
shorter rhythmic value.
The first time through the A section Karr follows the pattern established in the
first phrase. Generally, the vibrato is constant; however, there are some pitches on which
Karr does not vibrate. Beat three of m. 8 is one example. This pitch (concert E) can be
47
played as a harmonic instead of a stopped note. Generally, bassists do not use vibrato
when they play harmonics. The omission of vibrato on this particular note stands out
because m. 8 functions as the consequent to m. 4 and Karr used vibrato at this point
during the antecedent.
Unlike Koussevitzky, the repeat of Karr’s A section is significantly different than
his first iteration. Both performers change to a softer dynamic, but Karr takes this farther
by using flautando.50 He adjusts his use of vibrato to compliment the effect during the
first two phrases of the section, which results in a much softer, transparent sound. Karr
does use vibrato throughout the repeated section, but it is on the sustained note in m. 11
that he first changes the speed of the oscillations during the note (Example 3.2). This
note marks the point of transition from the flautando to his original bowing style and
begins without vibrato. Karr gradually adds vibrato; the first oscillations are slow and
speed up along with the dynamic level. This is a style of vibrato that Koussevitzky did
not use in either his recording of Chanson Triste or Valse Miniature.
The B section of Chanson Triste demonstrates Karr’s use of vibrato on notes of
shorter rhythmic value. As shown in example 3.4, Koussevitzky used vibrato during this
section very sparingly. Karr used the technique continuously throughout the section with
the exception of the eighth notes in m. 22 and the faster notes in mm. 28-30. The circled
notes in example 3.9 indicate those notes on which Karr does not use vibrato during the B
section.
50
Flautando is a flute-like effect usually produce by bowing over the fingerboard. It is generally used in
soft, sustained passages.
48
Example 3.9. mm. 20-35. Chanson Triste
The omission of vibrato in mm. 28-29 is related to the accelerando Karr uses at
this point in the performance.51 The tempo does not allow every note to last long enough
to use vibrato.
During the cadenza that follows, Karr omits vibrato at certain points, as an effect
(Example 3.10). Karr gradually speeds up during the first few quarter notes of the
cadenza and uses vibrato on all of them. As he reaches the bottom octave he slows down
(even though the written rhythm indicates eighth notes and sixteenth notes) but omits
vibrato. He does use vibrato on the lowest note of the passage (the whole note), but the
oscillations are slower than has been previously heard. Vibrato is absent during the rest
of the cadenza until the final note, which is sustained and marked with a fermata.
51
See Chapter Two, page 28-31.
49
Example 3.10. mm. 35, cadenza. Chanson Triste
The following three measures are transitional material, followed by the final
statement of the A section. This section is treated in almost an identical manner to the
first statement of the A section. Karr’s vibrato is almost continuous throughout. There
are a few instances where it is left out that were discussed earlier in this chapter. For the
final two notes of the piece (the leading tone resolving to tonic), Karr varies the speed of
the vibrato. The oscillations slow on the leading tone and vibrato is added gradually to
the resolution. This functions with the concluding ritardando to create a sense of finality
and ending to the piece.
It should be noted that in the most recent edition of Chanson Triste the indication
sempre e senza vibrato occurs at the beginning of the final statement of the A section.52
While neither Koussevitzky or Karr omit vibrato during this section, the fact that the
indication exists in a modern edition further supports the changing role of vibrato. It is
now the absence of vibrato that is the embellishment or special effect, as opposed to the
beginning of the twentieth century when the presence of vibrato was the embellishment.
52
50
Karr and Valse Miniature
Valse Miniature, with its faster tempo and greater rhythmic motion than the
Chanson, leaves little opportunity for either performer to use vibrato. Both performers
use vibrato consistently on sustained notes, which in turn highlights moments of rhythmic
flexibility.53 This is evident in the A sections of both performances. Both Karr and
Koussevitzky use vibrato during the last two measures of each antecedent and consequent
throughout the A sections (Example 3.6, a-d). The most significant difference in their
use of this technique during this section happens at the initial attack of each of the long
notes (such as mm. 7-8, 11-12, 15-16, and 19-20). Koussevitzky’s vibrato, when used,
begins almost immediately as the note sounds. Karr often chooses to let the note sound
for an instance before the vibrato begins. As a result, when Karr’s vibrato starts it is
slightly slower near the beginning of the note than at the end. This variation in speed is
not something that was apparent in Koussevitzky’s performance and is a trait of the
modern style. There are instances in which both performers allow the sustained notes to
sound briefly without vibrato, but in Koussevitzky’s rendition the ensuing vibrato begins
and remains at a constant speed throughout the note. There is one point during the A
section where Koussevitzky uses vibrato and Karr does not. This occurs at the end of the
second antecedent which is the highest note of the section (written G, Example 3.6, c,
m. 16). This note is generally played as a harmonic rather than a stopped note when it is
approached by intervallic leap. Since harmonics usually do not respond well to vibrato,
this is most likely the reason Karr chose not to vibrate. He plays the same note in m. 40,
only this time as a stopped note with vibrato.
53
See Chapter Two
51
A significant difference in the following section (mm. 23-42) is each performer’s
decision to apply vibrato to certain notes. As mentioned earlier, Koussevitzky does not
use vibrato in this section until m. 30. Karr chooses to apply vibrato to all notes that have
greater rhythmic value than an eighth note (Example 3.7 and 3.8). He uses vibrato
consistently throughout the section and it does not vary greatly in speed and depth. There
are two instances where the performer is required to make a leap of an octave, once in
m. 30 and again in m. 40. Karr does not use vibrato on the high quarter note on beat
three of m. 30. As he did in the A section, he plays this note as a harmonic. In m. 40,
however, he uses a portamento to get to the top note, which is performed as a stopped
note. The note is held due to the tempo slowing during this measure, and vibrato is used.
The combination of rubato, portamento and vibrato at this point in the piece creates a
very dramatic effect.
After a restatement of the A section, the syncopated portion discussed earlier in
this chapter begins in m. 59. Again, Karr uses vibrato throughout this section as well on
notes that are greater than an eighth note in rhythmic value, with the exception of the
anacrusis to each phrase. In each instance that an upbeat is followed by a measure of
syncopation, Karr plays the note senza vibrato. By doing this the emphasis is removed
from the upbeat and placed on the syncopation. Circles around the notes in example 3.11
indicate the upbeats that are played without vibrato.

52
Measure 91 marks the final statement of the A section before the closing material.
The sustained notes in the last two measures of the second antecedent (mm. 101-102) are
53
the last time during the piece that vibrato is used by Karr. This is followed by closing
material during which the rhythmic motion is too fast for vibrato.
The comparison of each performer’s use of vibrato allows us to follow the change
in attitude toward vibrato throughout the twentieth century. Koussevitzky, a student of
the late nineteenth century, employs vibrato too frequently for it to be classified as an
occasional embellishment. The nineteenth-century attitude of restraint and selective use
is not evident in Koussevitzky’s performance; however, the idea of the continuous
vibrato heard in Karr’s performances is not present either. Koussevitzky’s use of vibrato
is transitional; it bridges the gap between vibrato as an ornament and continuous vibrato.
Karr’s use of continuous vibrato is an excellent representation of the late twentieth-
century’s evolution of vibrato.

Chapter Four
Portamento in Chanson Triste
Portamento
One of the most striking differences, when comparing the recordings of
Koussevitzky and Karr, is their use of portamento. In regards to bowed stringed
instruments, this audible sliding between notes probably began in the early eighteenth
century when violinists started shifting on a regular basis.54 It is generally agreed that
there are three types of portamento: 1) a continuous slide on one finger from note to note
with all intermediate pitches being audible; 2) a slide in which one finger slides from the
starting note to an intermediate note, followed by the final note being played by a
different finger; 3) one finger plays the starting note, another finger is placed on an
intermediate note and then slides to the final note (Example 4.1).55
Example 4.1. Three types of portamento.
54
Brown, 560.
55
Philip, 144.
55
The New Harvard Dictionary of Music defines portamento as:
A continuous movement from one pitch to another through all of the

intervening pitches, without, however, sounding these discretely. It is
principally an effect in singing and string playing, though for the latter and
for other instruments capable of such an effect, the term glissando is often
used.56
For purposes of this document the term portamento will refer to any audible sliding
between two notes.
Chanson Triste: A Section
A comparison of Koussevitzky’s and Karr’s Chanson Triste recordings
demonstrate many examples of portamento, although they often use this technique in
different ways. Koussevitzky uses a very obvious portamento, often combining it with
rubato to further intensify the effect. The bass enters in m. 4 after a three bar
introduction.57 Koussevitzky uses three portamentos in the first phrase, with one of them
manipulating the tempo. The first occurs at a point where a shift is needed, between the B
on beat three of m. 4 down to the G on beat one of m. 5. This portamento is somewhat
restrained; however, the portamento that happens between the E and F sharp in beat two
of m. 5 is very obvious and is emphasized by a dramatic rubato. The third is extremely
subtle and leads up to the final note, a sustained C in m. 6. Koussevitzky’s use of
portamento is indicated in example 4.2 with a line above the staff.
Karr’s use of portamento in the opening A section of this piece is less frequent
and obvious than that of Koussevitzky. Within the first phrase there are two very subtle
56
Randel, ed., 649.
57
The International Edition (1949) and the copyist’s score (1896) both have a three measure introduction.
The David Walter edition (2001) indicates a four measure introduction. Measure numbers for this
discussion are taken from the International edition (see example 2.12).
56
instances: the first occurs at the same place as Koussevitzky’s first portamento, between
the B in m. 4 and the G in m. 5, and the second occurs in m. 5 during beat three between
the G and A. Both instances by Karr are barely perceptible. These are also shown in
example 4.2, indicated by a line below the staff. The line enclosed in parenthesis
indicates a very restrained use of portamento.
Koussevitzky’s use of portamento is indicated by a line above the staff. Karr’s

use of portamento is indicated by a line below the staff. Parenthesis indicate a
very subtle use of portamento.
During the second phrase of the A section, mm. 8-11, both performers make use
of noticeable portamentos. Koussevitzky’s first use happens on beat one of m. 9
descending from F-sharp to E. In mm. 10-11 there are two in a row, the first between E
and G and the next crossing the bar line to the final sustained note of the phrase (Example
4.3). This phrase shows Karr’s first clear use of portamento. There is an obvious slide
between beats two and three in m. 9. This deliberate slide emphasizes the descent in
pitch to the lowest note of the phrase. There is another between the eighth notes on beat
three of m. 10 which is combined with rubato. This combination assists with the
expressive nature of the phrase ending (Example 4.3).

57
Koussevitzky approaches the opening note of the third phrase with a portamento.
This is a vehicle not only to recover the tempo rubato from the end of the previous
phrase, but a means to connect the last note of the second phrase to the first note of the
third. There are three more instances of portamento in this phrase and each becomes
more pronounced, with the most dramatic use in m. 14. The slide between the F-natural
and C emphasizes the perfect fifth and could have been written by Koussevitzky for the
purpose of using a very pronounced portamento. He stays on the half note F-natural long
enough for the listener to realize that this is outside the key, and then moves slowly to the
C. This emphasis takes advantage of the physical distance between the two notes on the
fingerboard of the double bass (Example 4.4).
Karr uses three instances of portamento in this phrase. While not as prominent as
Koussevitzky’s, the first matches the transition from phrase to phrase by connecting the
sustained final note of phrase two to the first note of phrase three with a subtle
portamento. The second is also not very obvious and occurs between beat three of m. 12
and beat one of m. 13. In m. 14 Karr emphasizes the same interval as Koussevitzky;
however, Karr adds rubato to the arrival note on beat three for added aesthetic effect
(Example 4.4).
58
Koussevitzky again takes advantage of the physical distance between notes on the
fingerboard of the double bass during the final phrase of the A section, although in a
different manner than he did in m. 14. Since the distance between half steps is
comparatively greater on the double bass than it is on other stringed instruments,
Koussevitzky can stress passages that contain consecutive half steps. He does this in
m. 16. The portamentos that occur in this measure convey a sense of falling away and
are more dramatic because of the close intervallic relationship of the notes, as well as the
time taken during the slide. This chromatic passage begins on the B-flat in m. 16 and
continues to the downbeat of m. 17. He further emphasizes the end of the phrase by
using portamento as a means to slow the tempo. In m. 17 Koussevitzky takes time to slide
between the F-sharp and E as well as between the C-sharp and the final note of the phrase
on the downbeat of m. 18 (Example 4.5).
Karr’s rendition of the final phrase of section A differs from Koussevitzky’s. He
does not highlight the descending chromatic passages with prominent portamento. The
only portamento used in the phrase occurs in m. 17 between the F-sharp and E. Karr
does slow the tempo at the end of the phrase, but does not use portamento as a vehicle to
do this (Example 4.5).

59
Both performers utilize portamento throughout the A section, although those used
by Karr are much more subtle than those of Koussevitzky, and not as frequent.
Koussevitzky uses a total of fifteen portamentos during the sixteen-measure A section
where Karr uses only eight; and of those eight, only three are obvious. In addition, the
portamentos that Karr uses are shorter in length and do not manipulate the beat as much
as those Koussevitzky employs. While there is some slight variation in the use of
portamento during the repeat of the A section, both Karr and Koussevitzky are consistent
in their overall approach.
Chanson Triste: B Section
Koussevitzky continues to use a more prominent and frequent portamento than
Karr throughout the B section. As mentioned in Chapter Two, this section (mm. 20-35),
is a section in which the tempo changes significantly. Koussevitzky uses the portamento
as a tool to emphasize the changes in tempo.
As seen in example 4.6, the first use of portamento in the B section by
Koussevitzky occurs in m. 20. After playing the dotted quarter note, he slides to the
following F-sharp. The next portamento occurs at the beginning of m. 22. The last note
of m. 21 is a B and the downbeat of m. 22 is an F-natural. The portamento engages the

60
listener by bringing added attention to the tritone leap in the melody as well as
emphasizing the dissonance between soloist and accompanist. The F-natural on the
downbeat is sustained over an arpeggiated E major seven chord. Measure 23 contains
two instances: the first occurring from D down to C between the two eighth notes on the
first beat, and the second returning to C on beat three from the B on beat two. In this
measure Koussevitzky brackets the chord tone C with nonharmonic tones, and uses
portamentos to travel from each of those nonharmonic tones to the chord tone (Example
4.6).
During the first four measures of the B section Karr only utilizes the portamento
twice. Both happen at places as Koussevitzky used it and have a similar effect. The first
occurs when moving from the B on beat three of m. 21 to the F-natural on beat one of
m. 22. As Koussevitzky did, Karr uses portamento in this instance to emphasize the
dissonance and the melodic leap of a tritone. The second happens in m. 23. Karr slides
from the nonharmonic D on the downbeat to the third of the chord (Example 4.6).
Koussevitzky’s interpretation of mm. 24-27 conveys a deceptive sense of calm
before the upcoming accelerando that begins in m. 28. There is a portamento that
connects the last note of m. 23 to the first note of m. 24, and another that connects the
61
dotted quarter note in m. 24 to the following eighth note. Measure 25 is a repeat of m. 24
and he uses the same portamento again to slide from the dotted quarter note to the
following eighth note. As mentioned in Chapter Two, Koussevitzky holds back the
tempo at the end of each of these measures, which along with the connectivity created by
the use of portamento, creates a very smooth and continuous melodic line (Example 4.7).
Example 4.7 shows that Karr also uses portamento in mm. 24-25 at the same
place as Koussevitzky. That said, Karr does not connect the dotted quarter note to the
eighth note in the same manner. Karr crescendos slightly during the dotted quarter note
in m. 24, then clearly separates it from the following eighth note. The first eighth note in
m. 24 has a negligible portamento attached to it. This, combined with the accented
articulation placed on this note, exhibits a different interpretation than that of
Koussevitzky. In m. 25 Karr again uses portamento, but this time the dotted quarter note
is connected to the following eighth note. There is another instance of portamento on the
downbeat of m. 26. This portamento is also very slight in comparison to Koussevitkzy’s
method. Even though Karr uses portamento in two of the same places as Koussevitzky
during these four measures, the effect is notably different. Koussevitzky deceives the
listener by creating a false sense of calm, but Karr foreshadows the upcoming increase in
intensity by using some of the same tools as Koussevitzky in a considerably different
way.

62
Each performer’s use of portamento throughout the remainder of the B section is
an excellent representation of the change in this aspect of performance practice between
the early and late twentieth century. Koussevitzky uses portamento in several instances,
from m. 28 until the fermata at the downbeat of m. 35, to embellish the melody (Example
4.8). This begins with a very obvious portamento on the C-sharp downbeat of m. 28. To
place added emphasis Koussevitzky does not simply slide down from the E in m. 27
(Example 4.7) to the C-sharp. Instead he leaves the E, lands on the C-natural below the
C-sharp, and scoops up to the downbeat. An argument could be made that Koussevitzky
missed the shift in hand position from the E to C-sharp and that the slide from below the
C-sharp was included for pitch correction. The musical effect and emphasis this slide
places on the C-sharp indicates otherwise. It appears to be a deliberate embellishment
included to draw extra attention to the note and to signify that something different is
about to happen.
With the exception of one instance, Kousseivitzky’s use of portamento during the
rest of the B section is used to accentuate intervallic leaps in the melody. There are no
examples of portamento in mm. 28-29 other than the downbeat of m. 28. During
mm. 30-31, which are a sequence of mm. 28-29, Koussevitzky uses portamento twice. In
m. 30 he slides the distance of a minor third between D and F-natural. This is followed
by an E, which is tied over the bar line to m. 31. The portamento that occurs after the tie,
between the E and D, highlights the brief syncopation that happens during this iteration
of the sequence. Koussevitzky varies the second instance of this sequence by including
the portamentos. Measures 33-34 are sequences of m. 32. A portamento is included

63
before each intervallic leap during these measures, as well as into the fermata at m. 35
(Example 4.8).
As shown in example 4.8 and in Chapter Two, Karr takes a different approach
than Koussevitzky. Not only does he slow the tempo where Koussevitzky increases it,
but this passage is performed very cleanly and deliberately. Measures 28-31 are devoid
of portamento. During these four measures Karr plays each note in a very precise manner.
The few instances of portamento that occur in mm. 32-35 are not as obvious as those of
Koussevitzky’s. Karr does use portamento in the same places as Koussevitzky as a
means to accentuate the sequence, but he does not stretch the tempo as Koussevitzky did.
Karr’s alteration of the tempo is not related to his use of portamento during these
measures.
64
Chanson Triste: Cadenza and Transition
Both performers play the short cadenza as it is written.58 As shown during the last
eight measures of the B section, the cadenza is also a good representation of the change
in the use of portamento during the twentieth century. Koussevitzky performs the
cadenza, making frequent use of portamento. This is especially evident as the note speed
increases near the end of the cadenza. Example 4.9 shows the second half of the cadenza
which includes a series of minor thirds in three-note groupings (some written as
augmented seconds). Koussevitzky uses portamento as a means of transitioning from
group to group, highlighting the overall direction of the passage. All of the notes are
audible, but by using portamento in this way some are obscured. Koussevitzky sacrifices
clarity to emphasize the effect of the ascending minor thirds.
Example 4.9. Second half of the cadenza, mm. 35. Chanson Triste.
Karr, on the other hand, only uses portamento once, leading to the final sustained
note. His rendition of the cadenza is precise and each note is audible and deliberately
placed. This is a good example of the later twentieth-century concept that each note must
be clear, in tune and clean. Where Koussevitzky lost some clarity to emphasize the
musical effect, all of the notes Karr plays are exactly where they should be according to
58
65
the music. Both performances are equally effective, but demonstrate a significant change
in performance practice.
Measures 36-38 make up a three measure transition between the cadenza and the
final A section. Both performers use this as a brief interlude to set up the final A section.
Koussevitzky uses portamento and Karr does not.
Chanson Triste: Final A Section
Measures 39-54 are the final statement of the A section. With the exception of
the accompaniment in mm. 53-54 and the length of the final note in the solo part, this is
an exact repeat of the first A section. Koussevitzky uses portamento in much the same
way during this iteration of the A section as he did previously, although he does change
where he uses it in some instances. One case of portamento by Koussevitzky does
warrant notice. Example 4.10 shows how m. 52 is written compared to Koussevitzky’s
interpretation. In this measure he re-articulates the G briefly after he changes bow
directions instead of moving directly to the F-sharp on the second half of beat two. This
results in the eighth rest being omitted and a portamento occurring after the change in
bow direction. While it is impossible to know for sure which direction Koussevitzky’s
bow was traveling when he played this measure, one can assume that he used a down
bow on beat one and an up bow for the eighth notes.

66
Example 4.10. mm. 52. Chanson Triste.
a) as written b) as played by Koussevitzky
Karr almost eliminates the use of portamento completely during his performance
of the final A section, using it in only two instances, both of which are restrained.
Example 4.11 shows each performer’s use of portamento in the final A section.
Koussevitzky’s frequent use of portamento is a trait of the nineteenth century.
While the use of portamento was in a state of transition at the time Koussevitzky
67
recorded Chanson Triste, the prominence and variation of the portamentos in his
performance lean toward the accepted practice of the nineteenth century. The selective
use of portamento by Karr is indicative of the late twentieth century.

Chapter Five
Suggestions for an Historically Informed Performance of Andante and Humoresque
The information gained from the comparison of the Karr and Koussevitzky
recordings of the Valse and Chanson can be applied to performance of the Andante and
Humoresque. The purpose of this comparison is not produce an historically authentic
performance of these pieces, but rather as a guide to an historically informed
performance. What aspects of Koussevitzky’s performance are acceptable to the twenty-
first century listener? The application of portamento, rubato, and vibrato mentioned in
this chapter are suggestions based on a synthesis of modern and early twentieth-century
performance practice. They are hypothetical scenarios that could be applied completely,
or in part, to performances of these pieces. Potential performers should not implement
these suggestions without applying their own interpretations and modifications.
It is evident, when comparing this recording of Karr to his other recordings, that
he is attempting to include enough Koussevitzky era performance practice to present an
historically informed rendition of the composer’s works without straying too far away
from acceptable modern performance practice. This, along with the ghost (discussed in
Chapter One), could have influenced the title of his album, The Spirit of Koussevitzky. It
must be mentioned that a comparison of the two artists performing the second movement
of Koussevitzky’s Concerto for Double Bass indicates an even wider gap in the use of
69
vibrato, portamento and rubato. On his 1979 recording of the concerto, Karr uses very
little portamento, and when it is used it is very light. Koussevitzky’s 1929 recording
employs the same style of vibrato discussed in Chapter Three and a wide use of both
rubato and portamento that has been observed in the Valse and Chanson.59
When deciding what aspects of Koussevitzky’s style to apply to a modern
performance, the first consideration is vibrato. It is impossible to know for certain how
Koussevitzky’s use of vibrato changed between the time he wrote the four pieces in the
1890s and his recording of Chanson and Valse in 1929. However, it is possible to
speculate, based on the methods and recordings of that era, that the use of vibrato was
changing significantly. As discussed in Chapter Three this was a transitional time for
vibrato. It had changed from an ornament, used very sparingly, to an expected aspect of
a string player’s tone. Even during this time the use of frequent vibrato was
controversial; it had not yet evolved into the continuous vibrato that is characteristic of
the modern string player. Karr’s recordings of all four pieces employ the continuous
vibrato that has become familiar to modern audiences. Today’s audience expects to hear
a continuous vibrato that varies in speed and depth; and because of this, employing the
Koussevitzky era vibrato in a modern performance would not be acceptable.
Andante, Op. 1
Several places in Andante seem to ask for either or both the use of portamento and
rubato. The challenge to the modern performer is to add enough of these devices to
respect the composer’s intent without alienating him or herself from the modern
59
Koussevitzky, Dragonetti and Paganini, Virtuoso Double-Bass Concertos, performed by Gary Karr,
Koch-Schwann Misica Mundi CD 311 044 H1, 1980, Compact Disc.
70
audience. Karr’s recording of Andante includes the use of both rubato and portamento,
but not to the extent that would be expected of Koussevitzky. During the A section
(Example 5.1) Karr uses portamento and rubato to varying degrees.
Example 5.1. mm. 1-22. Andante.
Karr uses portamento four times during the A section. The first is slight, and
occurs between the first two notes he plays in m. 2. The second occurs in m. 5 and is
considerably more obvious. Karr uses portamento to travel between the last two eighth
notes of the measure which, are an interval of an octave. This is combined with rubato to
add emphasis to the interval, and happens again in m. 19. In m. 21 Karr slows the tempo
71
considerably between the quarter note and following eighth note and includes a slight
portamento. Koussevitzky often used rubato to end a phrase or to set up a sustained note
in this manner. Karr uses a great deal of rubato throughout the section. The anacrusis to
m. 3 is out of tempo, and it is slower than the tempo established by the accompanist
during the brief introduction. From this point forward, during the first phrase, Karr plays
the first half of each measure slightly faster than the established tempo and the second
half of each measure slower. This movement around the established tempo is similar to
Karr’s use of rubato in Valse Miniature.
Portamento could be applied to other portions of the A section beyond Karr’s use.
His use of rubato during the A section is constant; he manipulates the tempo in almost
every measure the way one would expect of Koussevitzky. Based on Koussevitzky’s
recording of Chanson Triste he may have used both rubato and portamento almost
anywhere throughout this section. He certainly would have applied portamento not only
more frequently, but with more emphasis than Karr. The choice modern performers have
to make is how often to use these devices. Using portamento in the manner that
Koussevitzky did would not be acceptable by twenty-first century standards, but Karr’s
use of this device during the A section is well within what is acceptable to the modern
audience.
Other points in the A section to which portamento could be applied are the
remaining upbeats to mm. 11 and 17. Karr combined rubato and portamento the first
time the anacrusis is heard in m. 2; however, it could be applied all three times the pattern
is heard or just the first and last time. By playing the same pattern with varying degrees
of portamento, rubato or vibrato the performance becomes less predictable. For instance,
72
the opening anacrusis could be played very softly, with a slower vibrato and out of tempo
(as Karr did). Instead of putting a portamento between the first two notes of the piece, a
slight portamento could be used to approach the sustained note on beat one of m. 3. This
is similar to the way Koussevitzky handles the opening of the first movement of Eccles’
Sonata in G Minor, which he also recorded.60 During this piece Koussevitzky uses
portamento and rubato to get from the anacrusis to the first downbeat of the solo line,
which is a sustained note. The accompanist could then reestablish the original tempo as
the soloist applied a crescendo to the sustained note, increasing the speed of the vibrato,
along with the volume. By applying these techniques to the first three notes of the piece
the performer has used attributes of both the modern and Koussevitzky era performance
practice during the opening statement. The performer must be careful at this point not to
use these devices in exactly the same manner every time the statement occurs. Both Karr
and Koussevitzky varied their use of these techniques throughout their performances,
with Karr employing a wider variation of these techniques.
The pattern is heard for the second time in mm. 10-11. To add variety the
dynamic level could be increased for this statement since the highest point of the section
occurs in m. 12. Portamento could be omitted completely, vibrato applied in a more
forceful and slightly faster manner than the opening statement, and the tempo could push
slightly forward until the sixteenth notes in m. 12, where the tempo should slow to
emphasize the approach to the highest note of the section. It should be noted that the
high G in m. 12 is marked with a fermata in the David Walter edition.61
60
Koussevitzky, The Complete Recordings. Track 1 and 2.
61
Walter, Four Pieces.
73
The third statement of the opening motive occurs in mm. 16-17. Since the
previous statement was performed at a loud volume this statement should be soft to create
contrast. There are two approaches to this statement that are appropriate. The first
approach is to play the anacrusis as harmonics without vibrato or portamento, but slightly
slower than the tempo of the previous phrase. The sustained note that follows in m. 17
could crescendo as it did during the opening statement and vibrato could be added along
with the increase in volume. The second approach is to add a very deliberate portamento
between the two eighth notes in the second half of m. 16.
Another place in the A section to apply rubato and portamento is the octave leap
that occurs in mm. 5 and 19. During Karr’s recording, these are the points at which his
portamento is the most deliberate. As mentioned earlier, Koussevitzky often used rubato
to prepare or embellish a sustained note. The last three eighth notes of m. 4 set up the C
that follows in m. 5 and would be an appropriate place for rubato. The octave leap that
happens in m. 5, and again in m. 19, seems to have been written for the express purpose
of utilizing portamento. Again, as both Koussevitzky and Karr have demonstrated, it is
not necessary to play the phrases with the same application of these techniques each time
they appear. By employing these techniques in varying degrees, the performer is allowed
to utilize a greater mix between modern performance practice and that of the
Koussevitzky era.
As an example, the octave leaps in mm. 5 and 19 also occur in mm. 42 and 56, a
total of four times throughout the piece. The intensity of the portamento and the tempo
manipulation should change depending on what is happening in the structure of the piece.
The second time this passage is heard is mm. 18-19, which leads to the B section. If the
74
phrase preceding the B section is played soft, then a stretching of the time is appropriate
in m. 18. This enables the octave leap in m. 19 to be more drawn, allowing the
portamento to be emphasized more than it was the first time this passage was played.
Furthermore, this gives the illusion of even more contrast between the A and B sections.
The third iteration of this passage happens after a short cadenza in mm. 41-42. Since this
is a restatement of A after a contrasting section and cadenza, it should mirror the
beginning. The final time this statement is heard occurs at the end of the work in
mm. 55-56. When the rubato happens this time it becomes the ritardando that continues
to the end of the piece. To further emphasize the end of the work, an obvious portamento
should be placed in m. 58 when the A descends to G (Example 5.2).
During the 2005 International Society of Bassists convention in Kalamazoo,
Michigan, Dennis Trembly performed Koussevitzky’s Andante as a tribute to David
Walter. For this performance he used the Amati double bass used by both Koussevitzky
and Karr.62 His interpretation of Andante includes limited use of portamento; however,
he does use rubato at the beginning and end of most phrases. One technique he uses to
add variety occurs in the last phrase of the piece (Example 5.2). He performs this
62
International Society of Bassists, 2005 Convention Documentary, prod. and dir. John Knific, 102 min.,
ISB, 2005. Digital Video Disc.
75
statement of the phrase entirely on the D string (E string in solo tuning). This changes
the timbre of the instrument making this final statement much darker than previous
statements of the same material, and is a significant contrast to the preceding phrase
which was performed loud and on the G string (A string in solo tuning). He does use
portamento in m. 56 as the point to begin the final ritardando of the piece.
Although short, the B section (mm. 23-39) of Andante offers several opportunities
to merge the conventions of the Koussevitzky era with those of the modern era,
especially when considering rubato. As seen in example 5.3 on the following page there
are two measures which include repeated chromatic passages. These occur in m. 25 and
29. Karr does not use portamento for these measures, but he does hesitate slightly before
the second half of each measure, putting an emphasis on the fourth eighth note of the
measure. Measures 31, 33, 34, and 35 also include chromatic eighth notes for which Karr
does not use portamento. One could speculate that Koussevitzky might have used
portamento during the chromatic measures, based on his use of the device during the
chromatic eighth notes in Chanson Triste. An appropriate place to include portamento in
mm. 31 and 33 would be between the fifth and sixth eighth notes. The tendency at this
point of the music is to speed up, and the David Walter edition includes markings to that
effect.63 Overall, the acceleration is necessary; however, a brief slowing of the tempo
within the overall acceleration between the fifth and sixth eighth notes in mm. 31 and 33
would be appropriate. This, accompanied by the portamento mentioned above, would
eliminate the predictability of the accelerando.
63
Walter, Four Pieces.
76
Humoresque, Op. 4
The Oxford Companion to Music defines humoresque as:
A name used as a title in the 19th century for a lively instrumental

composition, often ‘good-humoured’ rather than ‘humorous’…
Humoresques are generally short and in one movement…64
The tempo and the arpeggiated A section melody of the Humoresque do not allow
the use of portamento; however, the use of a push-and-pull rubato is easily applied. The
two-measure statements of the A section can all be approached by starting the three
sixteenth note pick-ups slowly, accelerating to a quick tempo for the sixteenth notes of
the first beat and slowing during the eighth notes in the second beat (Example 5.4). Each
phrase in the A section can follow this procedure. Since the A section is repeated often
throughout the piece, variations on this approach are needed to avoid predictability.
64
Latham, 596.
77
Example 5.4. mm. 7-23. Humoresque.
Karr’s recording of Humoresque offers several ideas to avoid repetition. In
addition to varying the rubato in this section, the arpeggiated nature of the melody is
easily transposed from octave to octave. The International edition of this piece shows
the A section placed in the octave seen in example 5.4.65 The Walter edition notates m. 7
through the down beat of m. 15 in the bass clef, an octave below example 5.4.66 In the
second half of m. 15 it returns to the octave notated in example 5.4. Karr chooses to play
the first statement of the A section as it is printed in the International edition, without
displacing the octave.
The final statement of the A section, beginning in m. 52, also includes this
discrepancy. For this statement of the A section Karr uses the instrument’s natural
harmonics to add variety. He plays each two measure statement in a different octave,
alternating between the octave indicated in example 5.5 and an octave higher. When
playing the upper octave he uses harmonics. All of the notes of the A section are
65
Serge Koussevitzky, Humoresque (New York: International Music Co. )
66
78
playable using the harmonics in the third octave of the D and G strings (E and A strings
in solo tuning). He begins in the written octave, and moves up an octave after the
sixteenth rest in m. 53, then back to the written octave after the sixteenth rest in m. 56.
This pattern continues until the end of the piece, when he plays in the upper octave
beginning after the sixteenth rest in m. 62 until the sixteenth rest in m. 66. The final two-
and-a-half measures are played as written. Example 5.5 shows Karr’s approach to
alternating octaves as it would be written for double bass.
Since the title of this piece is Humoresque, the octave displacement approach to
the A sections is ideal. The performer can relate a sense of irony to the audience by
moving from the usual tessitura of the double bass to the extreme high range. The visual
effect of moving the left hand approximately three feet, from the nut to the end of the
fingerboard and back again in quick succession, is also an entertaining benefit of this
approach. Several variations to the octave displacement approach are possible. As

79
mentioned earlier, the first half of each A section is written in the bass clef, an octave
lower than examples 5.4 and 5.5. If this were combined with Karr’s use of octave
displacement in the final A section, a third octave could be used. It is also possible to add
a fourth octave by playing one portion of the section in the lowest octave of the
instrument. It should also be noted that there are discrepancies between the International
edition, the Walter edition and Karr’s performance for certain notes. These occur on the
down beats of mm. 13 and 58 in which the quarter note is the next higher B instead of the
D indicated. These variations in octave displacement enable the performer to create a
great deal of variety by moving throughout the full range of the instrument. This
approach coincides with the humor intended for a piece titled Humoresque. In addition,
there is precedent for this in the recordings of Koussevitzky. In his recording of the
second movement of his concerto, Koussevitzky performs the last statement of the theme
an octave higher that written and uses harmonics.67
The repeated section in mm. 24-34 provides a significant contrast to the opening
A section (Example 5.6). Karr’s interpretation is very legato and he plays the entire
section an octave higher than indicated in both the International and Walter editions. He
also continues to use rubato during this section. The section begins with an upbeat
sixteenth note to m. 24, on which Karr takes extra time to set up the new section. He then
accelerates through the sixteenth notes in mm. 24-25, only to slow again in the second
half of m. 25. This tempo pattern repeats again in mm. 26-27. Karr plays mm. 28-29
with a somewhat steady beat after slowing as the melody descends in the second half of
m. 27. There is a moment of hesitation between mm. 29-30 and a more pronounced
rallentando during mm. 30-31. The second iteration of this passage is performed in the
67
Koussevitzky, Complete Recordings.
80
same manner; however, when he reaches the second ending he gradually slows the tempo
throughout mm. 32-34.
The use of portamento would be appropriate at are several instances in this
section. Koussevitzky often used portamento to emphasize a change in harmonic or
rhythmic motion. In this section, mm. 28-29 as well as mm. 33-34 contain rhythmic
motion that is new to the composition. A portamento between the last two sixteenth
notes of m. 27, combined with a slight rubato, would highlight the upcoming rhythmic
alteration. In addition, a portamento could be included after any of the eighth notes in the
syncopated measures. It would not be necessary to add a portamento after every eighth
note, but a selective use of this device would add nineteenth-century characteristics to the
passage. Measures 32-34 require a more exaggerated use of both rubato and portamento.
These measures provide an opportunity to again remind the audience that it is supposed
to be a humorous performance. A steady and significant ritardando throughout these

81
measures, in conjunction with an obvious portamento after each eighth note, is one
vehicle to accomplish this. The portamentos that occur should increase in length each
time. That is, the portamento from the D to B in m. 32 should be restrained with each of
the ensuing portamentos becoming longer and more blatant until the final slide between
the F-sharp and C-sharp in m. 34. It should take a significant amount of time to travel
between these notes, with the resolution to D being delayed as well.
After a two-measure interlude by the accompanist the rhythmic motion slows
significantly in mm. 37-52 (Example 5.7). Koussevitzky creates a lyrical passage as a
final point of contrasting humor, before returning to the A section and ending the piece.
This section should be performed as legato as possible with a rich, modern
vibrato. The Koussevitkzy era vibrato would not be acceptable to the twenty-first century
audience. While there are several places in the music to which portamento could be
applied, there are two points that are reminiscent of Koussevitzky’s performance of
Chanson Triste. Throughout Chanson, he applied portamento not only to large intervals,
but to short chromatic intervals. The minor second from mm. 39-40 as well as
82
mm. 43-44 are points in the music which a modern performer might not consider adding
a portamento, but that would be acceptable based on Koussevitzky’s recording of
Chanson.
Chapter Six
Conclusion and Suggestions for Further Research
Conclusion
Many facets of Koussevitzky’s four short pieces for double bass could be
discussed with regard to performance practice and the use of portamento, vibrato and
rubato. The examples mentioned are vehicles chosen to help bassists achieve a more
historically informed performance. Many of the characteristics of late nineteenth and
early twentieth-century performance practice are acceptable to the modern audience,
however, not to the extent to which they were used during Koussevitzky’s life.
Nevertheless, the most overwhelming question when considering these works is the
question of authenticity. Gary Karr did not ignore Koussevitzky’s rendition, but he did
not try to precisely imitate Koussevitzky either. One could surmise he included enough
of the techniques and practices from the Koussevitzky era to let the listener know that he
was aware of, and had considered the primary source--the Koussevitzky recordings. His
performance appears to be a synthesis of both modern and Koussevitzky era performance
practice.
The wide use of portamento and rubato, evident in Koussevitzky’s recordings, has
been toned down significantly since the beginning of the twentieth century. This is
understandable when the work is from a different era. Koussevitzky employs the same
84
practice during his performance of Eccles’ Sonata in G Minor as well as Beethoven’s
Minuet in G. The difference between these pieces and the Koussevitzky compositions is
that for the first time in the history of the double bass as a solo instrument there is
undisputable primary source material that conveys without a doubt the composer’s intent.
These are period pieces, written and recorded by the composer himself. It is difficult to
debate Koussevitzky’s intent in face of this evidence.
There are other issues that must be considered when considering how much
Koussevitzky’s recording should influence modern performances of his works. The early
twentieth century was a transitional time for performance practice. As discussed in
Chapter Three, Koussevitzky himself probably changed his use of vibrato significantly
between the time he composed these works in the 1890s and recorded them in 1929. This
would indicate that the composer himself was not opposed to altering the way his music
was performed based on current trends in performance practice. Another consideration is
recording technology. The twenty-first century audience is used to hearing every note
played to perfection due to the ease of recording and mass production. This was not the
case during Koussevitzky’s life.
As this document has demonstrated, it is possible to present a performance that is
somewhat true to the Koussevitzky era without completely offending the twenty-first
century audience. The main consideration is what aspects of Koussevitzky era
performance practice to use, and the frequency of their use. The modern continuous
vibrato has become so inherent in the expected sound of string instruments that the
absence of it is considered an embellishment. For this reason, using a Koussevitzky-style
vibrato would not be acceptable to the modern audience. In considering portamento and
85
rubato, however, it is frequency that would determine acceptance by the modern listener.
These devices are in use in modern performance practice, but not to the extent that
Koussevitzky used them. Because of this it is possible, as Gary Karr did, to create a
synthesis between the two styles. This compromise respects the intent of the composer
and the expectations of the modern audience. The goal of an historically informed
performance is to incorporate enough of the idiosyncrasies of the era to tell the educated
listener that one’s performance was influenced by that period.
Suggestions for Further Research
This topic has presented many avenues for further research surrounding
performance practice issues for the double bass, as well as topics concerning
Koussevitzky himself.
A new biography of Koussevitzky is overdue. There is a considerable quantity of
information contained in the Koussevitzky Collection housed in the Library of Congress
that could contribute to this topic. Much of the information published about
Koussevitzky centers around his life as a conductor; however, there is comparatively
small amount of information about Koussevitzky the bassist. A biography that placed
some emphasis on his contribution to the double bass, with regard to composition,
transcription and performance, would be a significant addition to literature surrounding
the instrument.
Pertaining to performance practice, a survey of various recordings of
Koussevitzky’s concerto would be valuable. His recording of the second movement of
his concerto would be a good place to start. By comparing it to subsequent recordings,

86
the subtle changes in performance practice throughout the twentieth century could be
traced.
Bibliography
American Symphony Orchestra. On-line at [http://www.americansymphony.org].
Angarano, Anthony. CD liner notes to The Spirit of Koussevitzky. Performed by Gary

Karr. Vogt Quality Recordings VQR 0231. 1989. Compact Disc.
Auer, Leopold. Violin Playing As I Teach It. New York: Frederick A. Stokes Co., 1921.
Reprint, New York: Dover Publications, Inc., 1980.
Berger, Arthur. Reflections of an American Composer. Los Angeles: University of

California Press, 2002.
Bowen, Jose. “Koussevitzky, Sergey.” In The New Grove Dictionary of Music and
Musicians. 2nd ed. Vol. 13, pp. 844-845. Stanley Sadie, ed. New York: Grove’s
Dictionaries, Inc., 2001.
Brown, Clive. Classical and Romantic Performance Practice: 1750-1900. New York:
Oxford University Press, Inc., 1999.
Brown, Howard Mayer and Stanley Sadie, eds. Performance Practice after 1600. New
York: W.W. Norton & Company, 1990.
Bruch, Max. Kol Nidrei. Transcribed for double bass by Serge Koussevitzky. Manuscript
(photocopy), Koussevitzky Collection, Library of Congress, Washington, DC.
Brun, Paul. A History of the Double Bass. Translated by Lynn Morel and Paul
Brun. Paris: by the author, 1989.
Butt, John. Playing With History: The Historical Approach to Musical Performance.
New York: Cambridge University Press, 2002.
Ciorata, Lucian. Contrabbassissimo. Homokord. 2000. Compact Disc.
Drew, Lucas. Double Bass Sampler II. Wirl Music Group B0216CD. 1977-82.
Compact Disc.
Elgar, Raymond. Introduction to the Double Bass. Princeton: Stephen W. Fillo, 1960.
________. Looking at the Double Bass. Princeton: Stephen W. Fillo, 1967.
________. More About the Double Bass. Princeton: Stephen W. Fillo, 1963.
88
Flesch, Carl. The Art of Violin Playing. Vol. 1. Translated and Edited by Eric
Rosenblith. New York: Carl Fischer, 1924-30. Reprint, New York: Carl
Fischer, 2000.
Forsyth, Cecil. Orchestration. London: Macmillan and Company, 1914. Reprint, New
York: Dover Publications, Inc., 1982.
Fuller, Jerry. The Double Bass and Violone Internet Archive.

[http://www.earlybass.com/].
Goilav, Yoan. The Virtuoso Romantic Double Bass. Tuxedo Music TUX CD 1090.
1991. Compact Disc.
Grodner, Murray. Comprehensive Catalog of Music, Books, Recordings and Videos for
the Double Bass. 4th ed. Vol. 1. Littleton: Grodner Publications, 2000.
Halfpenney, E. “A Note on the Genealogy of the Double Bass.” Galpin Society Journal
1 (1948): 41-45.
Hudson, Richard. Stolen Time: A History of Tempo Rubato. New York: Oxford
University Press, Inc., 1994.
International Society of Bassists. “Karrmati Gift Makes ISB History.” Bass World Vol.
28, No. 2 (Winter 2004): 21-22.
________. International Society of Bassists 2005 Convention Documentary. Produced

and directed by John Knific. 102 min. International Society of Bassists, 2005.
DVD.
Karr, Gary. “Karr Talk: Name That Tuning!” International Society of Bassists Vol. XX,
No. 3 (Winter 1996): 4.
________. The Spirit of Koussevitzky. Vogt Quality Recordings VQR 0231. 1989.
Compact Disc.
________. Virtuose Kontrababkonzerte, Koussevitzky, Dragonetti, Paganini. Koch-

Schwann Musica Mundi CD 311 044 H1. 1980. Compact Disc.
Koussevitzky, Serge. Andante, Opus 1, No. 1. David Walter, ed. Cincinatti: Liben Music
Publishers, 1989.
________. Chanson Triste, Opus 2. Reverse image copyist’s score (photocopy),

Koussevitzky Collection, Library of Congress, Wahsington, DC. Moscow: P.
Jurgenson, 1896.
89
________. Chanson Triste, Opus 2. Fred Zimmerman, ed. New York: International
Music Company, 1949.
________. The Complete Double Bass Recordings, Early Recordings with the
Boston Symphony. Biddulph Recordings WHL 019. 1994. Compact Disc.
________. Concerto for String Bass and Piano, Opus 3. Fred Zimmerman, ed. New
York: International Music Company, 1948.
________. “Etudes”. Manuscript (photocopy), Koussevitzky Collection, Library of

Congress, Washington, DC.
_________. Four
Pieces for Double Bass and Piano. David Walter, ed. Cincinnati: Liben
Music Publishers, 2001.
________. Humoresque, Opus 4. New York: International Music Company, Unknown.
________. Valse Miniature, Opus 1, No. 2. Fred Zimmerman, ed. New York:
International Music Company, 1949.
Latham, Alison, ed. The Oxford Companion to Music. New York: Oxford University
Press, 2002.
LaVine, Kevin. Reference Librarian, Music Division, Library of Congress, electronic

mail inquiry by author, 21 January 2005, Koussevitzky Collection, Library of
Congress, Washington, DC.
Leichtentritt, Hugo. Serge Koussevitzky: The Boston Symphony Orchestra and the New
American Music. Cambridge: Harvard University Press, 1946.
Levinson, Eugene. New York Legends. Cala Records LTD CACD0507. 1996. Compact
Disc.
Lourie, Arthur. Serge Koussevitzky and His Epoch. Translated by S.W. Pring. Repr. Ed.
Freeport: Books for Libraries Press, 1969.
Marillier, Daniel. Quintette des solistes de l’Orchestre d’ Auvergne. Polymnie POL 580
216. 2002. Compact Disc.
Morton, Joelle. “Evolution of String Instruments.” On-line at

[http://www.earlybass.com/evolution.htm]. 2001.
________. Joelle Morton’s Historical Bass Website. [http://www.greatbassviol.com/].

2001.
O’Connell, Charles. The Other Side of the Record. New York: Alfred A. Knopt, 1947.
90
Philip, Robert. Early Recordings and Musical Style: Changing Tastes in Instrumental
Performance, 1900-1950., New York: Cambridge University Press, 1992. Repr.
Ed. 1994.
Planyavky, Alfred. The Baroque Double Bass Violone. Translated by James Barket.
Lanham: Scarecrow Press, Inc., 1998.
________. Die Geschichte des Kontrabasses. 2nd ed. Tutzing: Hans Schreiber, 1984.
Quantz, Johann Joachim. On Playing the Flute. 2nd ed. Translated by Edward R. Reilly.
New York: Schirmer Books, 1985.
Randel, Don Michael, ed., The New Harvard Dictionary of Music. Cambridge: Harvard
University Press, 1986. Repr. Ed., 2001.
Reiley, Steven, ed. “The Karr-Koussevitzky Double Bass.” Bass World Vol. 29, No. 1
(June 2005): 35-37.
Slatford, Rodney, and Alyn Shipton. “Double Bass.” In The New Grove Dictionary of
Music and Musicians. 2nd ed. Vol. 7, pp. 519-525. Stanley Sadie, ed. New York:
Grove’s Dictionaries, Inc., 2001.
Smith, Moses. Koussevitzky. New York: Allen, Towne and Heath, 1947.
Stowell, Robin. Violin Technique and Performance Practice in the Late Eighteenth and
Early Nineteenth Centuries. New York: Cambridge University Press, 1985.
Vayspapir, Roma. Russian Snow Fantasy. RGM Productions. 1993. Compact Disc.

Performance Practice Issues Koussevitzky

Uploaded by

Copyright:

Available Formats

Performance Practice Issues Koussevitzky

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Performance Practice Issues Koussevitzky

Uploaded by

Copyright:

Available Formats

University of Nebraska - Lincoln

DigitalCommons@University of Nebraska - Lincoln

Student Research, Creative Activity, and Music, School of

Performance Practice Issues in the Short Solo Works of Serge

Follow this and additional works at: https://digitalcommons.unl.edu/musicstudent

Part of the Music Commons

Presented to the Faculty of

The Graduate College at the University of Nebraska

In Partial Fulfillment of Requirements

For the Degree of Doctor of Musical Arts

Under the Supervision of Professor Russell White

Short Solo Works of Serge Koussevitzky

Maurice E. Kelley II, D.M.A.

University of Nebraska, 2006

Advisor: Professor Russell White

Humoresque were written by Koussevitzky around 1900. His recording of Chanson

By comparing Koussevitzky’s recordings to those of Karr performance practice issues

necessary to produce a historically informed performance of all four pieces.

Karr recordings in regards to performance practice, focusing on tempo manipulation

analysis of Koussevitzky’s use of vibrato. The use of portamento by Koussevitzky’s and

Koussevitzky performing his own compositions) with the expectations of a twenty-first

List of Musical Examples………………………………………………………………...iv

2. Tempo Manipulation in Valse Miniature and Chanson Triste...........................9

3. Vibrato in Valse Miniature and Chanson Triste…………...…………………34

4. Portamento in Chanson Triste………………………………………………. 54

5. Suggestions for an Historically Informed Performance of……………..……68

6. Conclusions and Suggestions for Further Research………………………….83

1.1 Photograph of Koussevitzky’s bass strings……………………………..………..5

1.2 Letter from Gary Karr to Daniel Larson…………………………………………6

2.1 Valse Miniature, Antecedent/Consequent Pattern……………………….11

2.2 Valse Miniature, Measures 34-38………………………………………..16

2.3 Valse Miniature, Measures 39-42………………………………………..16

2.4 Valse Miniature, Measures 22-30………………………………………..17

2.5 Valse Miniature, Meaures 55-58………………………………………....19

2.6 Valse Miniature, Measures 59-62………………………………………..19

2.7 Valse Miniature, Measures 63-66………………………………………..21

2.8 Valse Miniature, Measures 71-74…………………………………….….21

2.9 Valse Miniature, Measures 87-90………………………………………..21

2.10 Valse Miniature, Measures 75-82………………………………………..22

2.11 Valse Miniature, Measures 103-end……………………………………..24

2.12 Chanson Triste, Measures 1-3…………………………………………...26

2.13 Chanson Triste, Measures 4-7…………………………………………...27

2.15 Chanson Triste, Measures 27-31…………………………………….…..29

2.16 Chanson Triste, Measures 32-35………………………………………...30

2.17 Chanson Triste, Measures 20-23………………………………………...31

3.1 Chanson Triste, Measures 4-7……………….……………………....…..37

3.2 Chanson Triste, Measures 8-11………………………………………….38

3.3 Chanson Triste, Measures 12-15………………………………………...38

3.4 Chanson Triste, Measures 20-35………………………………………...39

3.5 Chanson Triste, Transitional Material……...…………………………....40

3.6 Valse Miniature, Antecedent/Consequent Pattern……………………41-42

3.7 Valse Miniature, Measures 23-30………………………………………..43

3.8 Valse Miniature, Measures 39-42…………………………………….….43

3.9 Chanson Triste, Measures 20-35……………………………………...…48

3.10 Chanson Triste, Cadenza………………………………………………...49

3.11 Valse Miniature, Measures 58-90…………………………………….….52

4.1 Three types of portamento…………………………………………….…54

4.2 Chanson Triste, Measures 4-7…………………………………...………56

4.3 Chanson Triste, Measures 8-11………………………………………….57