Chapter 2

A Comparison and Analysis of Feuermann's

Dvorak Concerto Recordings with those of

Casals, Piatigorsky, Rostropovich and Yo-Yo Ma


The most direct and obvious evidence of Feuermann's playing comes from the recordings which he left. No other evidence could relay as clearly his style and abilities as a cellist. The goal of this analysis is a comparison of all aspects of Feuermann's playing -including tempos, use of glissandi, vibrato, 'timing' for musical effect, the use of accents and other forms of emphasis, and specific phrasing at fixed points- to those of the other four most famous and influential cellists of this century, Casals, Piatigorsky, Rostropovich and Yo-Yo Ma. The intention is to record only what is clearly observable, i.e. that X produces a slower vibrato than Y, or that Z makes a large ritard in a certain measure, and in this way come to a better understanding of what made Feuermann unique, without raising the argument of what is 'good' or 'beautiful' versus what is not. Reviews of recordings tend to focus on what the reviewer finds either compelling or distasteful. Although not conceding the view that all of what is aesthetically beautiful in music is 'relative' to the tastes of the listener, this is not a forum for that argument.

All of us, when we listen to music performed, form an opinion as to whether it is beautifully rendered or not. Professionally trained musicians form their opinions with a background of many years of interpretation, performance and analysis, allowing them more insight into the mechanisms of performance and interpretation with which to make their judgment. Still, it is shocking to realize how few of the thousands of minute details that together comprise an interpretation are never consciously comprehended.

To take the example of a cellist. There are only a few factors which a cellist may control in producing any given note: from the right hand -bow speed, bow pressure, the placement of the bow between the bridge and fingerboard, and the angle of the bow, which controls how much hair comes in contact with the string- and, from the left hand, vibrato - both width and speed. Furthermore, between notes, by controlling the coordination of the two hands, he can control the timing between the notes and he can control whether any glissando and what type is made between notes (In addition, if the coordination is not precise and the left hand does not stop the string completely, noise can also be produced, but this is not intentional.) This is the sum total of all the physical variables that we can control and from which we must create everything that happens musically on the cello.

In musical terms this means that on one note we can control the length, various types of accent or emphasis ( >, — ), diminuendo or crescendo during the course of the note, vibrato, general dynamic and the type of sound. Between notes we also control the length of time until the next note sounds -allowing us to establish a feeling of tempo, accelerando and ritard over the course of a several notes- the dynamic relative to the previous note -allowing us to establish a sense of crescendo or decrescendo- and glissandi. Over the course of a phrase, these simple variables combine together, on and between each note, to create all the nuances we describe as 'phrasing' . The complexity and density of these nuances is greater even than that found in speech. Herein lies a factor in music's powerful communicative ability perhaps, but also a situation that is very difficult to break down in analysis without becoming hopelessly lost in minor details.

The intention of this chapter is to provide a comparison of the playing of these five cellists, not in the sense of what is good and what is 'bad' or who is 'better', but only of what is clearly recognizably different in their approach, specifically, in their recordings of the Dvorak concerto. While this restriction might seem to create an artificially limited comparison, the patterns noted here can be seen throughout the recorded works of these artists. The Dvorak merely provides a convenient point of reference. Some generalizations can be made from the recordings as a whole, but for other factors it is necessary to compare very specific points, in order to be able to truly understand the details of each performance. It is striking how many of these details the listener, even a trained musician, is unaware of during casual listening. One gets a general impression from the sum total of these effects, but unless truly trained to be aware of them, one is never really clear as to how that general impression is formed. It is similar to listening to spoken dialogue, where we are aware of the meaning the speaker conveys by inflecting a sentence without being aware of the mechanisms of the many subtle articulatory gestures that combine to convey that meaning.

To those who find this an overly technical treatment of what is essentially an inspirational art, consider that Form was developed to describe what was the result of inspiration to the composers, and so we also must find a way to understand and discuss musical performance if we truly wish to understand and improve our art . We should not be satisfied to say 'This is beautiful ' or 'This is not' without making an effort to understand what specifically we are reacting to.

For this comparison, six recordings were used; two versions of Feuermann, one recorded commercially and one taken from a live performance, and one each of Casals, Piatigorsky, Rostropovich and Yo-Yo Ma. These performers were selected because they are the best known cellists of this century and have had the most widespread influence as performers. The details of these recordings are as follows:

Cellist Conductor Orchestra Year Age

Feuermann Michael Taube Berlin State Opera House 1928-29 [1] 25-26

Casals George Szell Czech Philharmonic 1937 60

Feuermann Leon Barzin National Orchestral Assoc. 1940 (Live) 37

Piatigorsky Charles Munch Boston Symphony 1960 56

Rostropovich Seiji Ozawa Boston Symphony 1985 58

Ma Lorin Maazel Berlin Philharmonic 1986 30


The first striking feature one notes in comparing these recordings is the wide range of timings for the recordings. While this might seem to be unimportant at first, these timings are the result of the performers tempos and use of rubato and, while small differences may be insignificant, huge variations cannot be ignored. The variation from the shortest recording, the 1928-29 Feuermann version at 32'49", to the longest, Yo-Yo Ma at 42'08" is stunning. The Ma version runs almost ten minutes longer, almost a third again the length of the Feuermann version. The others fall in between, with Casals at 35'31", Feuermann live also at 35'31", Rostropovich at 38'58" and Piatigorsky at 41'55".

Some might further argue that timings are irrelevant to an analysis of the performers' playing, due to the influence of the conductor and the sizable tuttis, but a clear and direct correlation between the performers' chosen tempos and their overall timings was found, as will be seen in the next section. While avoiding an argument about whether it is better to play either faster or slower, a comparison of the Casals and Feuermann tempos and timings to those of the later performers illuminates a clear difference in approach.

The differences in these five recordings raise a broader question as to whether these recordings might be an indication of a wider trend. While it was not feasible to compare the specific tempos of each performer on every available recording of the Dvorak, it is a fairly simple matter to check the timings for each movement, which correlate fairly consistently and meaningfully with the performers choice of tempos. The following is a list of commercial recordings with timings, put in chronological order by recording date. Unfortunately, there are very few recordings dating from before 1960, besides the aforementioned Casals and Feuermann, so any generalizations about trends over time must assume them to be representative of their respective generations.

Performer Conductor Orchestra Mvts Total Year

Feuermann Taube Berlin State Opera House 11'58" 32'49" 1928-29


Casals Szell Czech Philharmonic 13'27" 35'31" 1937

Feuermann Barzin National Orch. Assoc. 13'09" 35'31" 1940

Piatigorsky Munch Boston Symphony 15'45" 41'55" 1960

Fournier Szell Berlin Philharmonic 14'44" 38'26" 1962

Starker Dorati London Symphony 15'08" 38'06" 1962

Performer Conductor Orchestra Mvts Total Year

Rose Ormandy Philadelphia Orchestra 14'54" 38'48" 1963

Gendron Haitink London Philharmonic 14'33" 39'12" 1967

Rostropovich Karajan Berlin Philharmonic 15'42" 41'21" 1968

Du Pré Barenboim Chicago Symphony 15'19" 42'00" 1971

Nelsova Susskind St. Louis Symphony 14'08" 36'49" 1975

Harell Levine London Symphony 14'50" 41'59" 1975

Rostropovich Giulini London Philharmonic 16'28" 43'06" 1978

Tortelier Previn London Symphony 15'20" 39'19" 1979

Helmerson Jarvi Gothenburg Symphony 14'55" 38'51" 1983

Rostropovich Ozawa Boston Symphony 14'44" 38'58" 1985

Ma Maazel Berlin Philharmonic 16'04" 42'08" 1986

Maisky Bernstein Israel Philharmonic 16'33" 43'28" 1989

Kliegel Halasz Royal Philharmonic 16'13" 42'19" 1991

Performer Conductor Orchestra Mvts Total Year

Gutman Swallisch Philadelphia Orchestra 14'39" 39'01" 1991

Starker Slatkin St. Louis Symphony 14'38" 38'27" 1991

Schiff Previn Vienna Philharmonic 14'17" 37'20" 1992

Ma Masur New York Philharmonic 15'04" 40'28" 1995

This list is, of course, only partial. Although no type of underlying trend seems immediately apparent from this list, what these timings clearly show is the extraordinary difference between Feuermann and his successors. One need only look at Feuermann's tempos for the third movement, which yield a timing of 10'29" (10'36" in the live version.) The closest recording, Casals', is a full minute longer (11'36") and most modern versions are a full two to three minutes longer, with one Rostropovich version at 13'42". It is difficult to believe that they were playing the same music! What can clearly be seen from this is that Feuermann played at tempos which were, on the average, much faster than those of all of the leading cellists of the last 40 years. In other words, either no cellists choose tempos as fast as Feuermann's, or they indulged in a tremendous deal more rubato, or some combination of these factors. In the analysis of the six chosen recordings, both of these factors were observed.

The objective here is to understand how it was that Feuermann played, and his tempos are a significant ingredient of that, not only for bravura technical display in fast passages but also, much more significantly, in his choice of faster tempos for lyrical passages, which greatly change the listeners perception of the melodic line.

The earlier table of recording timings lacks any sense of generational flow since it is organized based on the date the recordings were made rather than the age of the cellist. In order to observe better any possible generational groupings, consider the same list organized by timing, shortest to longest, rather than by recording date:

Cellist Timing[2] Date

Feuermann 34'10" 1928/40
Casals 35'31" 1937
Nelsova 36'49" 1975
Schiff 37'20" 1992
Starker 38'16" 1962/91
Fournier 38'26" 1962
Rose 38'48" 1963
Helmerson 38'51" 1983
Gutman 39'01" 1991
Gendron 39'12" 1967
Tortelier 39'19" 1979
Rostropovich 41'08" 1968/78/85
Ma 41'18" 1986/95
Piatigorsky 41'55" 1960
Harell 41'59" 1975
Du Pré 42'00" 1971
Kliegel 42'19" 1991
Maisky 43'28" 1989

From this a pattern can be seen to emerge. If you except Schiff and Piatigorsky, both of whom diverge from the trends of their respective generations, there is a grouping of 'older' generation cellists -Feuermann through Rose- with faster timings (and thus either faster tempos, less rubato or both), a middle ground -Helmerson through Tortelier- containing cellists of both generations, and the slower group which, with the exception of Piatigorsky, are all players considered to be of the 'modern' school. This provides some support for the idea that rubato has become more liberal and tempos have become slower in this half of the century, as many have claimed. While this is interesting, we should also not assume too much merely from record timings. Conductors and tuttis do have an impact, and while timings correlate with general tempos, small differences aren't necessarily significant. Only the larger, striking differences, such as seen in comparing Feuermann's performances with those of the modern cellists, can truly be considered meaningful.

This generational difference is not limited to cellists. Consider Harvey Sachs' discussion of Rubinstein's 1929 recording of the Brahms Second Piano Concerto:

"... this version is worth hearing, inasmuch as most of its tempos are shockingly fast compared to those that have become the norm in our day. Brahms did not often put metronome numbers in his works, but in this case he took special care to do so. If one were to follow his indications inflexibly-which, of course, was not his intention-the first movement would last about 16:30, the second and third about 7:30 each (including the repeat in the second and allowing a great deal of leeway for the slower parts of the third), and the finale about 8:50. Here are the timings for two thoughtfully and beautifully played modern recordings: Alfred Brendel with Claudio Abbado and the Berlin Philharmonic (Deutsche Grammophon)-17:52, 9:20, 12:15, 9:21; Vladimir Ashkenazy with Bernard Haitink and the Vienna Philharmonic (Philips)-18:40, 9:26, 13:07, 9:29. The timings for every movement of Rubinstein's 1958 recording are also slower (16:53, 9:05, 12:39, 9:02) than Brahms indicated but not as broad as Brendel's or Ashkenazy's-except the third movement, which is slightly shorter in Brendel's version. If one listens to the well-known Horowitz-Toscanini-NBC Symphony recording of 1940, one hears a performance that is closer still to Brahms indications (the timings are 16:15, 8:06, 11:05, 8:25) and-in the case of the first and fourth movements-a jot faster than the composers guidelines suggested. But to contemporary ears, Rubinstein's 1929 recording (timings: 14:35, 8:10, 9:09, 7:54) sounds rushed in every movement except the second, and absurdly so in the third, which, however, is not played nearly as quickly as Brahms indicated.
...When one listens, today, to a recordings of the Brahms symphonies conducted by Felix Weingartner or Toscanini, one hears tempos that are considerably faster, on the average, than those to which we have become accustomed in recent decades. And yet Weingartner had conducted Brahms for Brahms (and had received the composer's praise) and Toscanini had modeled his approach on that of Fritz Steinbach, who was one of Brahms's favorite conductors. Barth, Rubinstein's teacher, had played Brahms's music in the composer's presence and the young Rubinstein had heard Brahms played by many other proto-Brahmsians. I would be willing to bet that, despite its defects, Rubinstein's 1929 recording of the concerto was closer to what Brahms had in mind when he wrote the work than is Rubinstein's 1958 recording or any of the later ones by other pianists." [3]


Returning to consideration of the six selected recordings, a specific comparison of tempos in the first movement further reinforces the differences suggested by the overall timings, in more meaningful detail. No performer plays in a strict metronomic fashion, so it is difficult to assign metronome markings for any section that are entirely accurate. While Dvorak indicates only two different tempos for the entire movement, all of the cellists in these recordings make additional tempo changes, which are traditional but not specifically indicated in the score. To get a practical map of each performers' tempos, the movement can be divided into 13 general tempo sections, though some of these sections contain accelerandi or other subtle tempo changes within. The starting points for each of these sections are as follows:

From these we can construct a basic 'map' of each performers tempos throughout the movement as follows:

Again, it should be stressed that no performance was metronomic, and that many subtleties could not be accounted for in such a format, so the above results are only an approximate guideline. This chart does clearly indicate the great difference between both of Feuermann's performances and the other cellists', however. The similarity between Feuermann's tempos on his recorded version, which he made when he was 25-26, and the live version, made when he was 37 is striking. While a few of the 'live' tempos were incrementally slower, specifically towards the end, where Feuermann slightly moderated the extremely fast tempos of his earlier version in the last sections, the overall agreement is so striking as to provide a clear rebuttal to the argument, which is occasionally put forth, that Feuermann chose his tempos in order to make the sections fit onto the sides of a 78 RPM records. Clearly these are the tempos Feuermann felt belonged to the music. His consistency in this provides an interesting contrast to the three Rostropovich recordings mentioned in the earlier section on timings, in which Rostropovich's tempos greatly varied from recording to recording.

The opening section finds all the cellists in more or less the same tempo range, with Casals producing the quickest and most straightforward version and only Ma significantly slower (and also somewhat more erratic rhythmically) than the rest. At the following 'Vivo' section, Feuermann's tempo is near that of the others, who are all significantly faster than Dvorak's indication. The difference comes with Feuermann's gradual accelerando during the technically challenging passage work in the measures prior to 5, which leaves him at a significantly faster tempo than the others. Piatigorsky, in contrast, ritards in this difficult passage. This illustrates a tendency seen frequently with Feuermann. In climactic passages which are also technically demanding, while other cellists often slow slightly to accommodate the passage work, Feuermann tends to accelerate, giving the impression of breathless excitement and daredevil risk taking.

It is in the second theme, and other slower, more lyrical themes, that the contrast between Feuermann and Casals versus the latter cellists becomes more apparent. In the second theme, which Dvorak indicates at q =100, Feuermann plays (both times) at q =96 and Casals plays at q =104. Rostropovich plays much slower at q =84, Piatigorsky even slower at q =72-76 and Ma begins at only q =66, only reaching q =84 after the accelerando at the end of the passage. When we examine this passage in detail later in this chapter, an interesting connection is seen. The cellists who choose the slower tempos are also the ones who engage in the largest amounts of rubato.

Without arguing whether playing more slowly is more or less desirable, we should consider the difference that a quicker or slower tempo makes in the listener's perception. For the purpose of demonstration, consider the analogy of a sentence. There is a natural speed at which the spoken sentence is easily understandable. If the words are spoken too quickly, the details blur. If spoken too slowly, the overall meaning of the sentence becomes difficult to focus on, as each individual word and inflection takes on an exaggerated importance. This phenomenon is described in the science of Information Theory and applies to any transmission- whether audible, visual or electronic- which contains 'information'. Thus the same principles can be seen in a musical phrase. Playing too quickly for the brain's ability to analyze information would cause important details to be missed. Playing too slowly would cause the focus to shift away from the larger phrase and towards the details of each small segment or even each note. This is intuitively understood by many musicians who teach that 'playing faster helps to bring out the long line'. Thus, when choosing to play the second theme at much slower tempos, Rostropovich, Piatigorsky and Ma are, perhaps unwittingly, shifting the listeners perception from the larger structure towards the individual notes and small 'sub-phrases'.

In the Tempo Io section following the second theme, Feuermann plays the spiccato sextuplet figures at a much more rapid tempo than any of the other cellists. All the cellists make a ritard at the end of this figuration into the next section, after which they accelerate to the end of the exposition. Here again, although they all follow a similar framework, Feuermann is distinctly faster than the others. The entrance of the cello at 10 shows again how exaggerated the differences between the cellists becomes in the slower passages. Dvorak's marking is q =100. Casals and both Feuermann versions are at that tempo. Rostropovich is 20 points lower at q =80, Ma even slower at q =76 and Piatigorsky playing only at q =63. In the Animato that follows, all the cellists make an unwritten but traditional accelerando into 12. Again Feuermann is much faster, and again we see Feuermann's tendency to accelerate in sections where both the musical tension and the technical difficulty are increasing. The differing impact of Feuermann's tempo upon arrival at 12 at q =152 and Ma's at q =116 must be heard to be truly understood.

The ensuing restatement of the second theme and subsequent material in B major follows much the same pattern as before, though all the cellists play the second theme a bit faster this time except Casals, who starts slightly slower this time and then accelerates. The tempos of the spiccato sextuplet section were similar to the tempos of its first occurrence for all the cellists, but in the accelerando in the following section, Casals and Feuermann made even more accelerando than they did in the earlier statement, while the other cellists played much as they had the first time. In the coda and ensuing Più mosso we again see Feuermann clearly faster than the others, and again accelerating through the technical difficulties towards the climax, while the others held steady or slowed. Here Feuermann is much faster than Dvorak's indication, ending his recorded version at a stunning q =176 where the marking in the score is q =132, which the other cellists approximately adhere to. Feuermann's spectacular tempo is certainly compelling to the listener, but Feuermann later moderated it somewhat when he ended the live version, twelve years later, at q =160, still much faster than any of the other cellists but slightly slower than his tempo for the recorded version.

From this discussion of timings and tempos, it is obvious that Feuermann performed at tempos much faster than those of the other cellists, with only Casals' tempos anywhere close. Clearly tempos have a large effect on how the music is perceived by the listener, and Feuermann's choice of tempos clearly contributes to the difference between his style of playing and that of other cellists. But it is also obviously only one aspect of what distinguishes Feuermann's playing .

Another thing that becomes clear, when trying to attribute metronome indications to these recordings, is how much the listener's perception of rhythm differs from the reality of the metronome. While Feuermann's performances seemed to provide the clearest 'feel' of the beat -meaning that to a listener, the rhythm and tempo seemed the most clear and compelling- when trying to set a metronome, one found a slightly changing tempo throughout almost every measure -a constant rhythmic 'push and pull'- making metronome indications sometimes recordable only as a range between two or three adjacent markings or as an average. At the same time, other performers, particularly Rostropovich and Ma, whose performances did not yield to the ear as strong a sense of tempo or rhythm, fit more easily within a specific metronome marking. From this, it is clear that the feeling and perception of rhythm are conveyed much more by the performers choice of emphasis or 'pulse' than by strict adherence to any absolute metronomic rhythm.

Listening to Piatigorsky, one gets the feeling that he is frequently ritarding, seemingly in almost every phrase. Listening to Rostropovich and Ma, one gets the impression of much slowing down and speeding up, sometimes seemingly capriciously. Casals, in contrast, seems to be the 'straightest' of the group. With Feuermann, while one might sense a small 'pushing and pulling' of the rhythm in the course of the phrasing, one is only really aware of a tempo change during the major tempo changes. Perhaps this is the result of more subtle use of rubato or by his emphasis of key notes (and avoidance of emphasis on notes which are only transitory and rhythmically weak) to create a clear sense of pulse, even if the rhythm is not metronomic. Perhaps he is also aided by a quicker tempo, which shifts the listener's focus towards the larger outline. In any case, it seems that for Feuermann, the sense of rhythm is created by the way he phrases.

In a 1987 review in The Strad, Harris Goldsmith makes similar observations about the performers' tempos and use of rubato in comparing three of the same recordings of the Dvorak (Feuermann, 1940, Casals, 1937 and Ma, 1986) that we have considered:

"...In 1940-41 when these performances took place, Feuermann's style doubtless did sound more 'modern' than Casals' - leaner, more brilliant, less rhetorical and effusively emotional. What I find instructive listening to the playing forty-five years later is how similar the Feuermann manner was to Casals' in its essentials - and how far removed both these masters are from today's norm. The biggest change is in our attitude towards forward direction: one has but to juxtapose either this 1940 Feuermann/Barzin recording of the Dvorak Concerto, or the 1937 version of same by Casals and Szell, with a fine modern performance such as the newly issued Yo Yo Ma/Maazel/Berlin Philharmonic(CBS...) to appreciate how much the pendulum has swung to the other side. (The contrast is even more noteworthy since Maazel is generally regarded as being a "precisionist" maestro in the same "objective" and technically-oriented tradition as Barzin and Szell.) Whereas both Feuermann and Casals project the music's tensile qualities with impelling thrust, Ma, and, more crucially, his conductor, take slower tempos and seize upon every excuse for a tenuto as if forward motion was to be equated with insensitivity.
...(in the 1940 Feuermann recording) much of the rapid high position passagework is cleaner, more assertive, than on the Casals record, but, as with Casals, there is an ardour, a willingness to take chances...And present, too, is a

marvellous sense for the long line...It (the 1940 Feuermann recording) is, everything considered, a very great interpretation (which will, one hopes, gain adherants and help reverse the current fondness for lugubriousness.)" [4]


While each cellist's usage of vibrato varies from note to note, their overall patterns and differences are consistent throughout the recordings. It is difficult to discuss vibrato in any kind of absolute terminology. The factors which may vary are the width and speed of the oscillation, and the change in these throughout the duration of a note. Even if it were possible to make a quantitative representation of these factors for each cellist, it would not necessarily be enlightening. It is more practical to pick one recording as a point of reference, in this case the earlier Feuermann recorded version, and compare the others to it. Since it is clearly easier to notice the differences between vibrati on slower, sustained passages, the second theme of the first movement and the opening of the second movement were two of the major sections used for comparison.

Compared to Feuermann, Casals' vibrato is sometimes slightly wider, though generally the same approximate width. It is almost always slower, however, producing a less 'active' sound. Casals' vibrato was not applied to all notes as consistently as Feuermann's, as some notes had an initial impulse of vibrato which then tailed off, while others sometimes stood out in contrast because they had no vibrato at all. Casals also sometimes began the vibrato just slightly after the note had begun to sound. This effect is definitely noticeable in terms of the type of sound it produces, but its cause is not easily identified unless one is specifically listening for it, as the delay is a very subtle.

Piatigorsky uses a vibrato that sounds much sparser and more constrained than Feuermann's. In comparison to Piatigorsky, Feuermann's vibrato sounds as if it has more 'beats' -i.e. vibrato impulses- per note, and that these impulses come more quickly. Furthermore, Feuermann's vibrato carries through the note, while Piatigorsky's tends to trail off. The actual physical dimensions of Piatigorsky's vibrato are narrower and slower than Feuermann's, and in this recording, his use of vibrato is also much sparser than Feuermann's. With Piatigorsky's vibrato, the effect resulting from beginning the vibrato slightly after the note has begun to sound is a little more pronounced than with Casals.

Rostropovich's vibrato is much wider than Feuermann's, and also slower. It is, in fact, the widest vibrato of the five cellists. Rostropovich also, to a greater extent than Piatigorsky and Casals, plays many notes with no vibrato, which therefore stand out in the phrase because of their different sound quality. The 'delayed vibrato' effect is also more noticeable with Rostropovich than with the either Casals or Piatigorsky.

Ma's vibrato is much slower than Feuermann's and is, in fact, slower than the vibrato of any of the other cellists. It is also almost as wide as Rostropovich's, but, in contrast to the other cellists, whose vibrato tends to be centered, varying evenly above and below the core pitch, Ma's vibrato occasionally becomes uneven, leaning towards the higher pitch and producing a somewhat irregular oscillation. What is strikingly different about Ma's vibrato, however, is the extent to which he delays the production of vibrato on some notes. Whereas the effect with the other cellists was relatively small and sometimes difficult to identify, with Ma the delay is much larger and the vibrato furthermore tends to increase throughout the note, producing a type of 'swell' which is very noticeable. Ma also produces the largest number of notes without vibrato. In addition, his vibrato speed is more variable from note to note than that of the other cellists.

Comparing Feuermann's vibrato on the live 1940 recording with that on his earlier 1928-29 recording (which was the standard used for the comparison to the other cellists) it seems that, though Feuermann's vibrato in the later version is still the same width and consistency, and still begins precisely with the sounding of the note and sustains through it as in the earlier version, the speed of the vibrato has now increased. This is the opposite of what one might expect from the now older Feuermann, as many cellists' vibrato speed slows with age, but the effect of this faster vibrato makes the sound quality more focused and adds a greater underlying element of tension to the sound. One can not help but wonder whether this change in his vibrato might have been influenced by Heifetz, with whom he was now collaborating regularly. Certainly the character of his vibrato at this time sounds very similar to that of Heifetz's.

Besides the greater speed, the major difference between Feuermann's vibrato and that of the other cellists was the initial impulse. His students have sometimes described the initial sounding of a note with the Feuermann vibrato as an 'impulse' from the left hand. If this impulse actually results in more vibrato at the beginning of the note and then a slight relaxation, the effect is almost too subtle to detect. What is clearly noticeable, however, is that the vibrato begins exactly with the sounding of the note, whereas for the other cellists, the vibrato begins, to varying degrees, after the note has begun to sound. Since adding vibrato to a note that has already sounded without vibrato intensifies the character of the sound, the vibrato technique of the other cellists produces at least a slight swell after the beginning of most notes. Feuermann's notes, in contrast, began with full vibrato, and thus maximum intensity, at the start, from which point it is either sustained or relaxed. While this might seem a relatively minor technical point, the difference made by having the maximum intensity of sound at the beginning of the note rather than having an intensification during the course of the note makes an enormous difference, and was integral to Feuermann's sound and manner of phrasing. Considering its overall effect, it is a striking difference between Feuermann's playing and that of the other cellists.

The second major difference could be described as consistency of application of vibrato. All of the other cellists studied tended, to varying degrees, to let the vibrato trail off during the course of a note, and also to produce notes with no vibrato at all in the middle of otherwise heavily vibrated passages, drawing attention to these notes in a manner that was presumably not deliberate. For Feuermann, it seems that the vibrato was as integral a part of producing the note as stopping the string, and that he would rarely do one without the other, unless in a passage so rapid as to make it impossible. Feuermann also vibrates on notes which many other cellists do not or can not. One example of this is the octave passage at the end of the first movement of the Dvorak:

Here Feuermann vibrates on the top note of the octave which most cellists do not, producing a vibrant, 'alive' sound almost akin to a violin. Furthermore, only rarely do you hear a swell or diminuendo effect produced from Feuermann's changing the speed of vibrato during the course of a note, and only when it is intentionally produced to coincide with a crescendo or diminuendo made by the bow. The swells produced by Ma's delayed application and changing intensity of vibrato produce the effect of many small crescendi, sometimes on each successive note, and it seems possible that it is done more from habitual reflex than deliberate choice.

Though Feuermann's approach to vibrato seems unique among cellists, it is not dissimilar to that of Heifetz and other violinists of the era, and the type of focused, intense sound that it produced may have contributed to the impression given by so many of his contemporaries that he was 'playing the cello like a violin'.


While it is difficult to make generalizations about the cellists' use of glissandi without referring to specific examples, it is very clear to anyone who has ever heard a Feuermann recording that he employs them to a much greater extent than most other cellists. As he explained to David Soyer [5], his use of glissandi related to his philosophy that the cello is "not a clarinet. You don't just cover the holes with your fingers." He believed that the cello should have the same fluidity as the human voice and that the glissandi he employed helped to make it sound so. But there were major differences in the way Feuermann produced his glissandi in comparison to the other cellists.

Consider the factors a cellist may control in the production of a glissando: the speed of the glissando, and the volume, including crescendi and decrescendi, with which it is produced (which depends on the pressure, speed and placement of the bow.) In spite of all the complications that arise when one considers the many different techniques of producing glissandi that are employed, these are the only significant factors which can vary. It is important to clarify here the difference between a glissando and a change of position, both of which are often referred to as 'shifts'. It is generally agreed that it was Casals' innovation to shift positions without necessarily producing an accompanying glissando, as his predecessors had done, often in a manner that many considered vulgar or even grotesque. It is clear from Feuermann's teaching that he was very much against the incidental production of glissandi as a byproduct of changing positions, and from the recordings it is also clear that Feuermann produced glissandi only where he chose to, and that he conscientiously avoided producing glissandi otherwise.

What is immediately obvious in examining Feuermann's glissandi is that, despite the fact that he employed them to a great extent, the glissandi do not receive a great deal of emphasis, either from the amount of time taken in producing them or from their dynamic. They seem rather more like a type of 'connective tissue' between notes than an event in their own right. Whatever the degree of emphasis that Feuermann gave a glissando, the arrival and departure notes received more.

Feuermann's slides tend to fall into certain distinct types: when making a descending [6] glissando, Feuermann almost always decrescendos, with the important emphasis going to the top (departure) note and uppermost part of the glissando. Frequently, Feuermann performs the glissando by traveling back on the finger which played the top note, only switching to the lower finger when the arrival position has been reached. This produces a glissando that does not sound in the very last portion prior to the arrival note, which helps to emphasize the upper portion of the glissando and matches the decrescendo. For ascending glissandi, Feuermann employed both crescendi and decrescendi. In either case, it is proportionally the upper part of the glissando which is heard, and it is the arrival note which takes precedence as the musical 'event'. In fact, some Feuermann glissandi are so subtle that they require repeated listening simply to become clearly aware of them.

Casals, in contrast, produces glissandi that tend to be slower and more evenly emphasized throughout the course of the glissando. When he makes a glissando, one is much more aware of it. In contrast to a Feuermann glissando, Casals would tend to depart more quickly from the departure note, travel more slowly through the course of the glissando, and sustain the same dynamic throughout the course of the glissando. This created a much larger, more obvious effect, sometimes overshadowing the departure and arrival notes. In addition, Casals sometimes emphasizes the lower part of the glissando on downwards glissandi, where Feuermann would do the opposite. Casals employed glissandi less frequently but, because of his method of producing them, more obviously than Feuermann.

The other three cellists make much less use of glissandi than either Casals or Feuermann. In the case of Piatigorsky, many of his few obvious slides are short slides of a whole or half step. While some of Piatigorsky's glissandi employ a release of tension similar to Feuermann's, others, particularly the shorter slides, are produced in an even, heavily emphasized style and slower speed, more similar to Casals.

Rostropovich also makes few glissandi. One which is particularly noticeable, and unique to him in this comparison, was a type of heavy downward glissando over a large distance, fast and heavily emphasized, and with the major emphasis on the lower part of the glissando. An example of this can be seen at 1 m. before 5 :

The few other types of glissandi Rostropovich makes are fairly fast, but with a constant dynamic throughout,. They sometimes seem almost to be incidentally produced by shifting, and produce a sound that has been rather vividly described as 'slurpy'.

Ma employs more glissandi than Piatigorsky or Rostropovich, though they similarly tend to concern the change of position in small intervals. His glissandi are produced at a slow speed with heavy emphasis throughout, producing a very noticeable, and more extremely emphasized effect than any the other cellists. An example of this can be seen 3 mm. before 6 :

Here Ma makes a heavy glissando from the E to the A and then another heavy glissando from the A back down to the F#. These heavy, slow glissandi tend to draw the listeners focus away from the melodic line.

In Feuermann's 1940 live recording he employs a greater number of glissandi than in the 1928-29 version, but the glissandi are of lighter character. The glissandi on the earlier recording are sometimes slightly slower and more emphasized. In the later recording, the glissandi are more numerous but less obvious.

Besides his greater use of glissandi, what is clearly different about Feuermann's glissandi, in comparison to those of the other cellists, is his technique of producing them with only enough emphasis to affect a connection between notes, but not so much as to distract from the musical line. More specifically, it seems as if most of his glissandi are accompanied by a feeling of relaxation or release. On the descending glissandi, the emphasis of only the upper part of the glissando ensures that it does not disturb the natural release of tension that is usually involved in descending melodically. When other cellists occasionally produce descending glissandi with more emphasis on the lower part of the glissando than the upper, the effect can sound somewhat distorted. In the ascending glissandi, Feuermann also emphasizes the upper part of the glissando, again following the natural musical line. It is often only the very last part of the glissando, right before the arrival note, that the listener is aware of. Regardless of the degree of emphasis a glissando received, all were produced with the same basic underlying principles.

It is also interesting to note how cleanly and completely without a hint of glissando Feuermann played in those places where he changed position but chose not to make a glissando. These instances clearly show that he did not create glissandi inadvertently. But it was Feuermann's frequent use of subtle glissandi to connect notes without distracting from the overall phrase that was one of the unique and immediately recognizable aspects of his playing.

A cautionary note: it is almost impossible to accurately describe all the subtle differences between glissandi in words. Certain details that are obvious to the ear defy verbal description. General differences in approach have been outlined here, but to completely understand the differences between the cellists glissandi, one must also hear the recordings.


Many of the details of phrasing can only be seen through detailed examination of an interpreter's performance. To better understand each performer's use of accents, emphasis, dynamics and rhythm, we must focus more closely on specific sections. Rather than attempting to discuss with each cellist every different technical factor that helps creates an emphasis on a given note, it is simpler and probably more meaningful to approach the discussion from the listener's perspective, noting simply on which notes the performers create emphasis. Typically this emphasis results from stronger bow pressure or speed, but other factors may also cause or contribute to a note's feeling of emphasis, including accents, vibrato, slides, the relative dynamic level, etc... For instance, a crescendo during the course of a note (which will also be described as a swell) would cause the normal emphasis, on the first part of a note, to be overshadowed by the latter part of the note. Similarly a slide into a note, a greater or lesser amount of vibrato on a note, or any other factor which distinguishes a note from the surrounding notes could contribute to the amount of emphasis it receives in the ears of the listener. In short, the 'emphasized' notes are those to which the listeners attention is drawn. Typically these are the climactic points of phrases. The term 'focus' of a measure or phrase is used to refer to the point within the measure or phrase with the greatest amount of emphasis.

We will consider two excerpts: the initial entrance of the cello in the first movement and the second theme, which provide an interesting contrast in which to observe the performers. For simplicity of reference, these excerpts are reprinted before the discussion:

The opening motif (mm.1-2) provides immediate contrast. Feuermann extends the rhythmic value of the initial B and shortens the 16th notes, while placing the important emphasis on the B's of the first and third beats. Casals similarly lengthens the opening note and shortens the 16ths, but makes a swell on the second B of each measure, moving the emphasis towards the very end of the measure, away from the second B.

Piatigorsky also lengthens the opening B, but does not correspondingly shorten the following 16ths, and also puts an accent on the first 16th -the C#- moving the focal point of the measure to that note. Piatigorsky also makes a crescendo on the B in the second half of the measure. In the second measure, unlike the first measure, Piatigorsky does not lengthen the B, but does shorten the 16ths this time, again adding an element of rhythmical ambiguity. Whereas with Feuermann and Casals the emphasis for both measures was on the first (and for Feuermann third) beat B's, with Piatigorsky, the emphasis was on the C# 16th in the first measure and the B on the first beat of the second.

Rostropovich plays the first measure fairly rhythmically, also crescendoing through the second B, and playing the 16ths in a much more legato fashion. In the second measure, however, he elongates the 16th notes, moving the focal point of the measure towards the 16ths.

Ma shortens the 16th notes greatly and also plays them somewhat unevenly, making a slight accelerando. He also places heavy accents on the opening B, the first 16th C#, and swell on the third beat B, which he shortens more than the other cellists, making it almost a quarter note. In the second measure, he places emphasis on the opening B and the first 16th A, though the 16ths are this time at full value, making a rhythmic contrast with the first measure. The third beat B is shortened and swells as in the first measure. Only Feuermann and Casals play the rhythm and accents of both the first two bars similarly, and in the Feuermann version the focal point of each measure is clearly on the first and third beat both times.

In the following measures (mm.3-4) all the cellists increase the tempo somewhat, with all the cellists placing the greatest emphasis on the third beat D#. For Feuermann, the tempo is immediately faster and the rhythm is proportional. Clear emphasis is also heard on the first two beats, though the third is clearly the arrival point. In the following chords, Feuermann is unique in this group of cellists in his creating a slight decrescendo during the descending chord sequence, though still giving each chord strong emphasis.

Casals places a heavy emphasis on the first beat, as well as the arrival at the third, and plays the following chords in a very short detached manner. Piatigorsky also places emphasis on all the first three beats, but his accelerando is noticeably different in that the first 16th note seems to fall too quickly, even for the new tempo, giving an element of rhythmical ambiguity to the new tempo.

Rostropovich plays the down-beat of m.3 slightly earlier than would be expected based on the tempo of the previous measure, and shortens the first 16th note to approximately a 32nd and the second 16th to a value about halfway between a 16th and 32nd. The tempo of the ensuing chords is immediately slower, though there is a slight accelerando and ritard during the course of the chords.

Ma lengthens the first B on the down-beat of m.3, which he follows with a very short 16th and an accelerando through the second beat to the accented third beat. This third beat emphasis, however, is detracted from somewhat by a crescendo during that note. The tempo of the ensuing chords seems to have little relation to the earlier tempo in the measure, being much slower, though he does make a small accelerando during the course of the chords.

The ensuing four measures (mm. 5-8) follow similar patterns to the first four measures except that Casals now lengthens the 16ths in m.6, increasing their proportional importance in the measure.

In mm.9-12 Feuermann plays in almost the identical tempo to the previous measures, making only a slight accelerando from the end of m. 11 to the fourth beat of m.12, where he pulls back. He accents each beat of the first two measures, in accordance with Dvorak's fz indications, though the largest emphasis is given to the third beat. In mm.11-12, Feuermann places the major emphasis on the down-beat of measure 11, and then no major emphasis until the four 16ths on the fourth beat of m.12, marked pesante , which Feuermann accents.

Casals takes a tempo at m.9 which is immediately quite a bit faster than that of the preceding measure, and goes faster yet at m.11. The main accents in mm.9-10 are on the first beats of the measures, and mm.11-12 are played quite 'straight' with relatively little accent or accelerando, even on the pesante 16ths.

Piatigorsky starts m.9 a little faster than the previous measure, but the tempo of the dotted eighth/sixteenth figures on beats three and four seems not to match the tempo of the 16ths in the first part of the measure. The main emphasis in the first two measures is on the third beat, but there is also an accent on the third 16th of the first beats of both measures, preceded by a crescendo over the first two 16ths. In m.12 Piatigorsky makes an accelerando and then a ritard over the course of the last two beats. This ritard starts earlier and slows more, in comparison to Feuermann's at the same section.

Rostropovich begins m.9 in approximately the same tempo as the previous measure, and plays the next two measures in this tempo, though he shortens the 16ths of the third beats to 32nds. At the down-beat of m.11, Rostropovich goes immediately much faster, and makes a tremendous accelerando over the next few beats. At the down-beat of m.12, he holds the down-beat C and then makes a very big ritard into m.13

Ma accents each beat in mm. 9-10. His tempo is suddenly faster than in the previous measure, but the last 16th of the first two beats is too short relative to the previous three, producing a slightly arrhythmic 'limping' effect. Ma also accelerates through the third and fourth beats of both these measures. In m.11, Ma makes a small ritard over the first three notes and then an accelerando until the pesante , where he ritards.

In mm.13-16, Feuermann places a strong emphasis on the down-beat F#. The tempo in m.13 is established by the orchestra, but when Feuermann again has moving notes in m.14, he takes a tempo which is slightly faster than that of the orchestra. A slide into the D# on the fourth beat of m.14 gives that note a certain amount of emphasis, and Feuermann makes a small tenuto there. This does not create the effect of a ritard, however, as the following G is shortened and the down-beat of the next measure falls where it would have in the original tempo. Feuermann makes a small ritard on the last beat of m.16 and slides into the final G, and the consequent emphasis makes it the focal point of the measure.

Casals also begins m.14 slightly faster than the orchestra's tempo, and slides to the G, the last note of m.14. This slide, combined with a ritard, puts the main emphasis on this note. In m.16 Casals makes a slight accelerando and then a ritard, giving a slightly uneven feel to the overall rhythm. Casals also slides and ritards into the final G in m.16, again making it the 'arrival' point. Piatigorsky similarly ritards in the last beats of m. 14 and 16, and makes a similar slide into the final G of m.16.

Rostropovich begins m.14 at a much faster tempo than the orchestra, and again in m.16 starts much faster than the orchestra's tempo in the preceding measure. In m.16 he accelerates and then ritards, with the ritard becoming extreme on the fourth beat. The overall effect is somewhat chaotic rhythmically.

Ma begins m.13 with a large accent, but no vibrato, which produces an unusual sounding accent, with a harsh attack and a quick decay. He accelerates through m.14 and makes a similar type of accent on m.15, except that this time he adds vibrato as the note continues. He accelerates through m.16 until the last beat, and makes a slow and heavily emphasized slide to the final G.

In the trills (mm.17-23) Feuermann is immediately somewhat faster. He follows the crescendo to m.22 and this down-beat is given the greatest emphasis, though the C# on the fourth beat of this measure, and the B and A# on the first and third beats of the following measure also have significant emphasis. Feuermann makes a ritard in m.23 and then begins the Vivo at a new faster tempo.

Casals places the emphasis similarly to Feuermann except that he makes a swell on the A# and B of m.21. This means that instead of the main emphasis going to the beginning of each of these notes, it goes to the latter part, and the beginning of the following note is a slight dynamic drop. Casals plays m. 23 going into the Vivo with very little ritard.

Piatigorsky holds the base note of each trill before beginning the trill, making it heavily emphasized and making the trill sound relatively less emphasized. He emphasizes m.22 and 23, and ritards before m.24, but the last three trills before the Vivo have a very noticeable decay in volume, giving the effect of a small decrescendo on each note.

Rostropovich's version is most noticeably different in that the tempo he takes for these measures is actually faster than his tempo for the 'Vivo' section, so that when the 'Vivo' begins, he actually slows down a bit from his previous tempo. Otherwise, the general line that he follows is similar to that of the other cellists.

Ma's version is noticeable for its extreme accent and subsequent decay on the first trill. He follows Dvorak's fz, decrescendo marking in this way, and then plays the following trills very much in a piano dynamic. This approach is unique among the cellists studied here, but the success of this effect will have to be judged by the listener. Ma makes swells on both the trills in m. 21, shifting the emphasis towards the ends of each of those notes, and then makes a large accent on the down-beat of m.22. This accent is followed by a diminuendo through the course of the note, and then there is another swell on the following C#, and again on the final A# before the Vivo.

The second section considered is the second theme of the first movement:

In mm.1-4, Feuermann places the major emphasis on the down beats of m.1 and m.2, the third beat of m.3 and the first and fourth beats of m.4. He elongates the first F# of the first measure slightly, and then moves forward on the following eighths to compensate. He also moves forward through the first three beats of the second measure, and pulls back on the fourth beats of m.3 and m.4. Casals interpretation is much the same, except that he decrescendos on the fourth beat of m.4, and hence does not emphasize it as Feuermann does.

Piatigorsky 's choice of emphasis is similar, but he makes much larger ritards. He takes a small pause before the fourth beat of m.2 and then makes a large ritard through m.3, and another again in m.4. Besides the larger proportion of these ritards, what is different from the ritards of Feuermann and Casals is that there is less sense of a resumption of the original tempo following the ritard. While the ritards of Casals and Feuermann could be considered 'pushing and pulling' the phrase, Piatigorsky seems only to be slowing down.

In Rostropovich's version, a swell on the first F# in the first measure moves the emphasis towards the latter half of the note. The second eighth note, the D, also receives a strong emphasis. The down-beat of m.2 is emphasized, and the next three beats have a small accelerando, but the open A string on the third beat is so loud as to give it a (presumably) unintentional accent and the fourth beat D is also emphasized. There is also a slight pulling back going into the down-beat of m.3, and emphasis is also given to that down-beat. Rostropovich also differs from the previous cellists by decrescendoing to the top A on the third beat of m.3. He makes a slight ritard on the last two beats of that measure, and also a ritard on the last beat of the fourth measure.

Ma begins the whole passage at a very soft dynamic, but immediately makes a swell during the course of the first F# in m.1. While he accelerates to the third beat of m.3, he makes a decrescendo on the fourth beat, drawing the focus of the measure away from the third beat. Ma makes a very large ritard on the fourth beat of m.4. What is most notable, however, are the numerous small swells, as many notes tend to begin very softly and than increase in intensity during the course of the note.

In mm.5-8, Feuermann emphasizes the down-beat and the first eighth note of m.5, the down-beat and the fourth beat of m.6, and the third beat of m.7. He ritards slightly through the three eighth notes in m.5, elongates the F# on the third beat of m.7, and pulls back the tempo slightly on the following triplets. He also pulls back on the fourth beat of m.8 which is, of course, actually the beginning of the next phrase.

Casals emphasizes the first beat of m. 5, but lengthens the second eighth note, the D, making the last two eighth notes sound almost like a dotted eighth/sixteenth pairing. This naturally adds emphasis to that D. While Casals similarly emphasizes the A on the fourth beat of m.6, he crescendos during the note, adding more emphasis towards the latter part of that note. Casals slides into the third beat of m.7 and lengthens this F#, but doesn't make much ritard on the following triplets. Casals also pulls back the fourth beat of m.8, and plays these up-beats at a much softer dynamic.

Piatigorsky continues his ritard from m.4 over the first three beats of m.5 but, though he never really regains the tempo, the A on the third beat of m.6 falls earlier than would be expected. Piatigorsky likewise gives the most emphasis to the F# on the third beat of m.7, and makes a large ritard over the course of the final two beat of this measure. Like Feuermann and Casals, he also ritards in the last beat of m.8, but slightly more so.

For Rostropovich, the third beat of m.6 similarly comes earlier than would be expected. The striking difference, however, is that Rostropovich decrescendos into the third beat of m.7, which the previous cellists had made the arrival point of their phrases. This decrescendo also makes the following E, which begins the triplets, seem to stand out with special emphasis. Rostropovich ritards in the last beat of m.8 and drops down to a pianissimo dynamic, which he maintains throughout the next phrase.

Ma again makes a swell on the first note of m.5, and on the first note of m.6. He makes the A on the fourth beat of m.6 the high point of his phrase, and decrescendos from there into m.8. He ritards through the triplets in m.8 and makes a very large ritard into m.9

In the next two measures (mm.9-10) Feuermann places the emphasis on the down-beats of each measure, and makes possibly a slight accelerando through three eighths in m.9. Casals places the emphasis similarly, but makes a swell on the opening B of m.9 and also shortens the third beat of m.10, presumably in preparation for the position change. Piatigorsky, too, makes the same emphases and similarly accelerates through the eight notes in m.9. Rostropovich and Ma are similar, though Rostropovich plays this section pianissimo.

Over the course of the next seven measures (mm.11-17), Feuermann begins a gradual accelerando, which becomes even more apparent at the Animato at m.15 and carries through the end of the section. While Feuermann makes a ritard in the last two beats of m.17 in the earlier recording, he does not in the later live version. The emphases in this section are quite straightforward: In the measures with two groups of two notes (i.e. mm.11, 13 and 15-17) the emphasis is clearly on the first and third beats, that is the first note of each grouping of two. There is no decrescendo to the second note of these groups however, which are also strongly present. In the other two measures (mm.12 and 14) the emphasis is on the down-beat only. The secondary emphasis on the latter note of the groupings of two continues to increase during the section, however, so that eventually the first and second notes begin to have almost the same amount of emphasis. Feuermann makes one significant change to the score at this point, playing the final climactic phrase (mm.15-17) an octave higher than written, yielding a very intense effect. In the live version, Feuermann also effects a small slide into the down-beat of m.13. While the largest emphasis in mm.11-12 is on the F# on the down-beat of m.12, but this slide helps to move the focus of mm.13-14 to the E on the down-beat of m.13, and thus, in both cases, agree with Dvorak's indicated crescendi and decrescendi.

Casals slides into the first E on the down-beat of m.11, and it receives corresponding emphasis. It is the second and fourth beats of this measure, however, which receive the greatest emphasis, effecting a kind of crescendo through each of the two note groupings. While only the down-beat of m.12 is emphasized, the previous pattern is again apparent in m.13, and a descending slide into the fourth beat adds further emphasis to it still. In m. 14, the down-beat F# is emphasized, but there is a crescendo during the following E. In the final three measures each of the quarters notes receives approximately equal emphasis, though a slide into the second beat of 16 gives it further emphasis. Casals also ritards in the final two beats of m.17. Casals' accelerando throughout the section is much less evident than Feuermann's.

Piatigorsky makes a slow slide into the second beat of m.11, causing it to have a strong emphasis. He ritards in the last beat of m.11 and plays the first two notes of m.12 much slower than expected, making them sound more like even valued notes than a quarter followed by a half note. This rhythmic distortion may also be the cause of the fourth beat of the m.12 seeming to sound too early. Piatigorsky then repeats the same slide, ritard and rhythm in the next two measures. From mm.15 to the end, after a small tenuto on the down-beat A of m.15, his phrasing is fairly straightforward, with the traditional ritard on the final two beats of m.17.

With Rostropovich, the emphasis on the latter note in the groups of two is even more pronounced, giving the effect of a crescendo through each group of two. In m. 12 and m.14, though he does not change the rhythm of the first three beats, the fourth beat still seems to come too quickly, as with Piatigorsky. In mm.15-17, the emphasis of the second note of each grouping becomes even more pronounced, as he does not vibrate the second and fourth beats of mm.15 and 16. His accelerando begins only at the Animato at m.15

Ma accelerates through m.11, but pulls back at m.12. He accelerates again through m.13 and then somewhat more at the Animato. He also emphasizes the second note of the groupings of two to an extent, mainly because certain notes are without vibrato and hence stand out to the listener, receiving undue emphasis. Ma also makes a very slow, heavy slide into the fourth beat of m.13, and similar slides both ascending to and descending from the A on the down-beat of m.15, creating a situation where the listener is more aware of the two slides than of the notes that they connected.

While some patterns do become clear from this method of describing these excepts, there are also details which are impossible to fully describe. All the cellists studied here accelerate and ritard, slide, accent, crescendo and decrescendo, and otherwise shape the phrase. The difference with Feuermann is really seen is in the timing, degree and extent to which these effects are used. Consider, for example, the performers use of rhythm. The performers accelerandi and ritardandi were measured here by beating as many subdivisions of each note as possible, to get an accurate sense of the motion of the beat. With Feuermann, the beat accelerated and ritarded more from note to note than with any of the other cellists, making it the most difficult to follow. And yet the magnitude of these variations, particularly the ritardandi, was nowhere near as great as with Piatigorsky, Rostropovich and Ma. The other cellists were much stricter rhythmically on a microscopic scale, but on a larger scale, they were more variable. It is difficult to completely appreciate this difference in this general analysis.

Other issues which were difficult to fully describe include the degree of emphasis on notes, and the subtle use of crescendi and decrescendi. The discussion of crescendi and decrescendi here focuses on those either of broad scope or in highly exposed positions. But all the cellists make subtle natural decrescendos as they relax tension in a phrase and similarly subtle crescendos as the tension increases. What is different about Feuermann is the greater extent to which he does this, producing a more clearly depicted rising and falling line. Also noticeable is the fact that those notes which Feuermann emphasizes receive a clearer emphasis than the emphasized notes of the other cellists. Partly this stems from his immediate use of vibrato and precisely coordinated attack between the two hands, but this also comes from the fact that he does seem to give proportionally more emphasis to those notes which he chose to emphasize, and less to the notes where he did not intend emphasis, than the other cellists. Again, this small but crucial difference is almost impossible to fully account in this analysis.

Despite these limitations, certain patterns do become clear. The major difference between Feuermann and the other cellists, particularly the latter three, seems to be a certain simplicity of effect. It is not that he does not accelerate and ritard, crescendo and decrescendo, slide and accent as the other cellists do, but with Feuermann, it is never done in such a way as to detract from the clear emphasis of the important notes of the phrase. One can make the argument that the other cellists created what appears to be extraneous emphasis in seemingly unimportant places deliberately, because they personally saw these places as important and deserving emphasis. Certainly this must be true in at least some cases. But it is also clear from microscopic examination that all these cellists follow a certain underlying structure in their phrasing, and their general adherence to this framework makes at least some of the unexpected emphases which detract from this structure, seem likely to be unintentional

In this general framework, which they all seem to follow to an extent, the relative importance of each note is influenced by: a) the relative melodic pitch, with higher notes receiving more stress than lower notes, b) the rhythmic placement of the note, with the first beat receiving the most importance, the third beat next (in four), etc.... c) the composers dynamic indications, and d) the harmonic importance of the note. With a combination of these factors a natural musical outline is formed. For example, consider the opening of the second theme:

(Dvorak makes only a general dynamic indication, so it isn't a factor in this case)

Combining these factors the notes that would have the greatest amount of emphasis are:

Whether one agrees with these principles or not, it is a framework that all five of these cellists follow to some degree. Though none of them follows it exactly, the fact that they all stay within the broad outlines of this principle does make it hard to believe that an isolated disruptive emphasis, from a crescendo, slide, unexpected rhythmic placement or other source, is necessarily deliberate. It would be reassuring to believe that all the cellists discussed here controlled and created every detail of their performance exactly as they chose to, but it is very difficult to believe that bow swells, notes without vibrato and irregular rhythms are performed by choice, rather than by a lack of awareness, particularly when they do not recur consistently. While the specific impact of these extraneous emphases may not be consciously realized, their overall effect is to make the direction and focus of the phrase less clear. The difference with Feuermann is that he makes his intended phrasing much clearer, and conscientiously avoids those things which would disrupt it.

Some of what we have described as 'disruptive' emphases and rhythmic alterations are, however, clearly deliberate, particularly in the cases of Rostropovich and Ma, and are the result of a conscious decisions made by the cellists. These choices relate to what is sometimes described as a 'sense of indulgence.' It is clear that adding additional emphases and causing rhythmic ambiguity creates a greater level of complexity in the rhythm and direction of the phrase, and hence a greater level of 'complication' (in the Information Theory definition of the word.) To what extent this complication is considered aesthetically pleasing is a matter of taste, but clearly a greater degree of complication can not necessarily be equated with a greater degree of artistry. Consider an actor reciting the following line of Shakespeare: "To be or not to be. That is the question". A reading in a flat monotone would convey little meaning or emotion. A good actor learns to emphasize certain words. Perhaps "To be or not to be. That is the question" or "To be or not to be. That is the question." or other such emphases and inflections. However if too much indulgence is taken it could become "To be ...or...not to be. That the question" Such a reading would clearly sound ridiculous and overly dramatic, to the point of losing it's artistic effect, but it demonstrates that there is a limit to the extent to which greater complication can make a sentence or, similarly, a phrase more meaningful. In the theater they teach that 'not every word can be a revelation'. So also, not every note can be an expressive climax unto itself.


The goal of this comparison is not to say who is 'better' or 'worse' among the five cellists, or, in fact, to draw any subjective conclusions at all. Instead, the goal is to describe everything that can be clearly differentiated that Feuermann did differently from the other cellists. Although the description of small musical details with words is sometimes impossible, several factors are clearly shown.

Feuermann definitely chose tempos much faster than his colleagues, both for faster and for slower sections (though often Casals' tempos on the slow sections were similar.) Feuermann produced a clear beginning for each note, with the vibrato beginning exactly with the start of the note and not slightly afterwards as with the other cellists. The precise coordination of the left and right hands also contributed this extraordinarily clear sounding of each note. Feuermann's vibrato was faster than that of the other cellists, of an average width and more consistent in application. Feuermann used glissandi frequently to produce a sense of connection between notes, but they were generally fast and lightly emphasized. Whatever the degree of emphasis of these glissandi, it was secondary to (though it also, in fact, helped to create) the emphasis of the arrival and departure note. Often only the highest part of the glissando was heard.

Feuermann made many small accelerandi and ritardandi during the course of a phrase, but these effects were never extreme and the overall rhythmic flow was rarely disturbed. Feuermann played with a clear and strong emphasis on those notes which he wished emphasis, and very little on those he did not. He relaxed the tension in between emphasized notes, creating a clearer sense of rising and falling line. Feuermann interpreted similarly to the other cellists, but avoided technical carelessness and interpretive extremes which could distract from his musical line. One of his colleagues summarized it thus "With the other cellists, I hear every note-every swell and gesture. With Feuermann, I am only conscious of the line"


This comparison first appeared as if it ought to be simple. As trained musicians we assume that we are aware of, and can explain everything that we are hear and react to, but it is amazing how difficult it is to find symbolism or terminology to describe all the factors that comprise what is broadly labeled as 'interpretation'. It is more shocking still to realize how little time or thought is generally put into the basic factors of phrasing -what we control, how we control it and whether it is desirable or not. By and large, teachers leave it to the student to provide whatever phrasing they are able to produce naturally, only correcting, after the fact, what strikes them as undesirable. We too rarely discuss how to read a piece of music and understand the natural phrasing structure, or the technique of producing what is beautiful and avoiding what is undesirable. Too often we are satisfied only with producing the correct notes and rhythm, with most of the interpretive responsibility left to what comes 'naturally'. We allow ourselves to emote without turning a critical and analytical ear towards the results that are produced.

We entertain a mistaken impression that physically reacting to whatever we feel internally at any given moment will automatically produce something which is beautiful to a listener. While phrasing will always be instinctual and to a degree improvisational, that intuition must be guided by an understanding of the principles that create musical beauty, an ear trained to be aware of these details, and a command of the instrument that will allow a player achieve these goals. There is a shocking belief in some circles that one can only worry about playing the 'correct notes and rhythm with a beautiful sound' since everything else is 'just a matter of taste'. The fact that musical tastes may vary is no excuse for leaving a student unequipped to control or discern the details of phrasing.

You could no more expect a beautifully interpreted performance from a student who is unaware of the details of interpretation than you could expect a moving reading of Shakespeare from a foreigner who knows how to pronounce the phonemes of the English language but has no idea of the meaning of what he is saying. The unrecognizable rhythm, swells and accentuation that such a reading would yield are, in fact, analogous to the distorted rhythms and disruptive accents and swells that are produced by musicians who are unaware of the principles of and unable to control the specifics of phrasing. The results of this comparison point to an alarming direction in cello playing. Many of the trends heard in today's 'modern' cellists, represented here by Rostropovich and Ma, are foreign to the approaches of Casals and Feuermann. We are moving in an alarming direction, and the tragedy is that it is by ignorance and carelessness rather by choice.

Regardless of the readers opinion of Feuermann, Rostropovich, Casals or any of the others, I hope that one will gain from this reading this chapter (as I have in writing it ) not only the details of Feuermann's playing, but also a renewed awareness that meaningful interpretation requires acute detailed awareness of everything that we produce on our instruments, and that interpretation is not only a means of self expression, but also a duty, to the composer and to the audience, to produce what is correct to the intentions of the music and most beautiful.


1 This recording was masde in 12 takes in two separate sessions. The first in May, 1928 and the second in October, 1929.
2 In case of multiple recordings, timings were averaged.
3 Harvey Sachs, Rubinstein, A Life (New York: Grove Press, 1955)
4 Harris goldsmith, record Review 1998 (February 1987): 143
5 P. 141
6 For this discussion, the term “descending glissandi” will refer to those that descend melodically and “assending glissandi” to those that ascend melodically.