Musical Villain Series: Did Stravinsky Ruin Music?

Igor Stravinsky (ca.1920s-1930)

Nicholas D. Lewis

I am not one to cast aspersions, but we should be honest with ourselves: today’s performance culture has been sterilized.  Seventy or eighty years ago, the performance culture was defined by liberal interpretations and individualism.  Now, the creativity involved in the music process has largely been stifled; what appears on the page has a higher value than the ideas of the performer.  As a result, it is difficult to tell performers apart as they often strive to sound the same. Somebody or something must be the catalyst for this decline, and the arguments I’ve heard are large-scale conflations and wholly conjecture.  I am therefore writing a set of five short and blunt articles that offer a candid look at the cause of this sepulchral situation.

The first person to be held culpable is Igor Stravinsky.  Don’t get me wrong, I love the guy’s compositions. In fact, I spent a whole summer
compiling a catalog of his works. But his skill as a composer does not negate his blameworthiness as a musician. It is not a dubious exaggeration to claim that Stravinsky may be one of the most influential musicians of the 20th century in terms of both composition and musicality. In the late 1930s, Stravinsky was a frequent lecturer at Harvard University, and he was also publishing a fair number of essays on music in various magazines and journals. A significant portion of these essays and lectures dealt with the relationship between the composer and the performer, and the position that Stravinsky held was that it was a ‘crime against the composer’ for performers to take the interpretation of a piece of music into their own hands. In other words, every instrumentalist must obey exactly what is on the printed page; anything else represents an impurification of the composer’s intentions. Stravinsky drove this dogma home so potently and forcefully that many academics were won to his side. Since those same academics were the ones who would most directly influence student musicians through teaching, this mindset over time became institutionalized.

20 thoughts on “Musical Villain Series: Did Stravinsky Ruin Music?

  1. I’d like to take issue with your premise: in order to understand Stravinsky the way you do you have to suppose that he–or any composer–could exactly write down the way the music sounds in his mind. He (or she!) can’t. Think about it: you are playing your violin (my instrument) and reading the music. You meet a series of eighths with dots on top. The dot means: the note is lasting less long than its full value. How much less: the composer didn’t write it in. You can do the dot in may different ways: staccato, spiccato, martellato, also depending on the tempo. He would still be busy with his manuscript if he wanted to be that precise all the way through. Same for dynamics: a hair pin opens and closes again: how strong is the loudest note: do you go from p to mf and back? Or all the way to f? Or maybe just a tiny rise to mp? How about metronome figures: quite a few composers gave them and almost all of them are obviously unrealistic: most of the time you have to play slower than the composer thinks for the audience to be able to follow (they can be useful for relative tempi though). It is your decision as the interpreter. You have to take into account your instinct but also the style (classic, romantic, neobaroque..) and the overall architecture of your piece and the role that the passage in question has in the context of the piece.

    Personally I don’t think either that Stravinsky was the first composer who wanted his music to sound the way he heard it in his head before writing it down. But Bach or Haydn conducted the performances of their works themselves: they made sure it sounded the way they wanted it (or tried to make sure anyway). From Bach for example we have colorful laments about the inability of his musicians to play the way he imagined they ought to.

    I can understand that a composer wants to be respected; he puts a lot of work into his music. The role of the interpreter (in classical music) is to try and understand the composers intention–beyond what is written in the score–and come up with technical and musical means to bring the music to life in a manner that seems to him (if not to everyone else) to reflect the composer’s intention.

    I am never quite sure what “creativity” is: I would say it is the ability to select a solution from an array of possibilities and be able to explain this choice either in words or–in music–by executing it properly (or in words as well as in musical practice). Defined this way my idea of the interpreter’s role does indeed involve creativity.

  2. “The role of the interpreter (in classical music) is to try and understand the composers intention–beyond what is written in the score–and come up with technical and musical means to bring the music to life in a manner that seems to him (if not to everyone else) to reflect the composer’s intention.”

    And this is a case in point of the problem about which I am writing.

    Thank you for commenting, I will address your concerns. Stravinsky was not the first composer to want his works to sound the way that he wanted them to sound. However, he was the first “major” composer to impose that musicians have to follow what they think are the composer’s intentions.

    It turns out that up until the 1960’s, music was performed in such a way that the performer’s intentions were held at a higher value than the composer’s intentions. The written notes, rhythms, and articulations were really seen as approximations. In other words, they were all viewed as molds for the performer; they were not viewed as the “be all and end all”. If a composer writes “forte and smooth” and the performer played it “piano and disconnected” then that was simply seen as an artistic modification.

    Stravinsky argued that it was disrespectful to change what the composer wrote in favor of your own ideas. However, this is in part what led to a highly individualized performance culture. Now the performance culture is largely… Sterile.

  3. I would agree that using the word “crime” as Stravinsky did is way over the top. However I am not sure if–more modestly expressed–lots of people and certainly many composers would agree wit the point he was making. If a composers markings are just proposals why bother composing? Why spend weeks and months figuring out the best music you can come up with if every two bit pianist will alter it according to the mood of the day when he/she plays the music?

    On another plane: you would postulate that dynamic or articulation markings are optional. How about grace notes? Are we allowed/expected to add or omit any? Or can we change the harmony, play a triad if we don’t like the seventh chord in some place? Which part is optional, which is not? Or is there even such a distinction? Is there a core to musical compositions that is “untouchable” as it were?

    Let me postulate that the score as written by the composer is the primary inspiration for every performance; it comes before the interpreters musical talent and taste, before his/her knowledge of the style and the historical period.

    Now this truth to the score is not always easy and straightforward to realize. Examples: Bach’s sonatas for violin with obbligato “Clavier”. Which instrument do you choose for the “Clavier”? It means “anything with a keyboard” in the historical context, harpsichord, clavichord, organ etc. Obviously Bach had no access to a Steinway or Bösendorfer. However, looking at the score I think (not everybody agrees) modern pianos are the most suited instrument: they are great at producing the polyphonic texture of most of the movements, they can be played to be in balance with the violin sound, they do not blend with the violin sound, i.e. the texture is clear and easier to hear for listeners. All choices that were available to Bach were suboptimal for these (highly innovative!) sonatas: harpsichords are too soft and they tend to blend with the violin, forcing the violin to stay in piano, organs are too loud, having maybe one single registration that produces acceptable balance, clavichords are out of the question; you won’t even hear them if the violin plays mezzoforte. So I prefer to use a “non-period” instrument even though Bach could not have intended it this way because he did not know about this option.

    Second example: tempi: It is very useful to pay close attention to the time signature: it is 2/4 the counting unit is the quarter note, if it is 4/8 the counting unit is the eighth note. One takes then the tempo marking and comes up with a tempo in the range for that marking counted in the indicated units. This will often result in faster tempi in slow movements than are customary. Haydn has many andante or andante quasi allegretto movements which are usually played in slowly plodding eighths and last forever. If one plays them as written, i.e. counting in quarters they cease to be plodding, all of a sudden they have a pretty lilt, the melodies have individual charm and loose their generic “Papa Haydn” character, the movements are interesting from start to finish.

    This sort of thing is what I like to call “taking the score seriously”. It is not slavish obedience: Exact and detailed observation of the score yields insights that help bringing the music to life.

    You are likely going to respond to this and I would like to ask you to expand on your definition of “sterility”. And if it is a problem (I think it is though not because of Stravinsky); what do you propose will help us getting over it?

  4. There are a lot of reasons to compose even if many of the markings are just proposals. Firstly, in the 19th century it was largely expected that the performers would change things to suit their personalities. Yet composers still composed, and we have the music to prove it. It is actually a fallacy to make the conclusion that if composers’ markings are not taken seriously, then there must be no reason to compose. The composers are still fundamentally creating a useful ‘template’ or ‘mold’, and very few people would change enough details to actually change the piece in a substantial enough way for it to be unrecognizable.

    Why can’t performers add or subtract grace notes? If you listen to Kreisler (one of the great violinists) recordings from the 1940’s, you will find that he often added grace notes and slides as to compliment and accentuate his personality. Glenn Gould often added and subtracted ‘rolls’ in the Beethoven and Mozart sonatas recordings. He also added grace notes.

    With regards to tempi, and I’ll still use the character Glenn Gould, listen to his recording of the first Brahms Piano Concerto, where he nearly halved the tempo. Cortot, arguably the greatest musician of the 20th century, also modified tempi in order to suit his artistic ideas.

    To put the composer before the performer implies that the composer knows best. It turns out that composers sometimes do not know best, and it turns out that a lot of compositions can be improved. Performers, in my view, have no responsibility to a composer. Rather, they have a responsibility to themselves and to the music. If a performer can – at least in his or her own mind – improve something, then they should.

    I do not want to hear the same performance of a piece over and over again, and I do not want to hear things played the same way. It is much more interesting when individualization is imbued in the atmosphere of a piece through modification suited to the performer’s interest.

    We have a dichotomy here, and it is the source of our disagreement: who comes first, the composer or the performer. I happen to stand by the performer. If he wants to subtract a seventh chord and make it a triad, he or she should have the liberty to do so.

    My definition of “sterility”, as applied to this argument, is the result of overall standardization and uniformity in music. If we follow the composers wishes (especially blindly and without questioning them), then people are likely to sound very similar. When comparing recordings from the 1940’s with recordings from today, there difference among the way performers played the pieces was much more salient than the differences today.

    As for a proposal, it is the purpose of my article series to expose the cause of the problem. The solution is easy: don’t assume that the composer is right just because he is the composer. Put yourself ahead of the composer – especially ones who have been dead for 200 years – and make something interesting out of the music. When you try to play things the same way as everyone else, classical music becomes boring and academic. It should be spontaneous, and exciting as a result of that spontaneity.

  5. I don’t think we disagree as much as you seem to imply. So a few things to clarify:

    Composer / performer: what this is really about is this: quod licet Iovi non licet bovi. I put the word two-bit-pianist in there on purpose: Fritz Kreisler had the authority and genius to do things that you and I should rather avoid–or at any rate I should avoid. (I used to have a recording of his of the Beethoven spring and Kreutzer sonatas–on LP, so no longer useful due to a lack of hardware–which I liked very much except for this: the pianist was nowhere close to his level–here is something that has gotten better in recent years; first rate pianists like Barenboim or Argerich have no problem being “accompanists”. Personally I also object to even Kreisler taking so much liberty with tempi–they change in some movements with every change of mood, but we do not know and will never know how Beethoven himself handled tempo, so this is my taste and has nothing to do with our topic.)

    My main point is this: there are numerous possible versions of any decent composition; there is no need to disregard the text of the score to be original. I go one step further and say: read and analyze and feel the score carefully and take your time over it. You will inevitably arrive at your own original recreation of the work. And after this sort of preparation you are even in a position to disregard some of the composer’s instructions because you know what overall result you are aiming at. What I object to is decisions made out of some mood or–even worse–out of a desire to be “original”.

    Tempi are not really Metronome figures they are character indications (similar to your distinction of “pitch” and “tone” or whatever your terminology is): if Gould can play Brahms at “half tempo” (i.e. half the metronome figure) without losing the character of the movement and without boring the audience to tears: more power to him. He would not, just by being slow, counteract Brahms’ intention (let alone commit a crime against him). It would be a mannerism (and Glenn Gould showing off Glenn Gould rather than Glenn Gould showing off Brahms) that nobody else should repeat even in the unlikely event that he could pull it off, but Brahms must have been aware that by writing a concerto he wrote a piece whose idea it is to show off the pianist.

    In summary: I am not advocating a brainless literalism, I am advocating taking the composition seriously, understanding it seriously, putting in the work required and represent it in performance to the best of the performer’s ability. As opposed to spending one’s time figuring out how it could be done differently and in an”original” manner.

    A short word on sterility: The ability to record music has brought a chance for performers to listen to themselves as their own “audience”–earlier generations did not have that. Everybody who has ever recorded himself must remember being shocked at how he sounds. So this has had a disciplining effect, which for example accounts for much more perfect intonation in modern vs. old recordings. It also dampens the enthusiasm for flourishes made out of the moment: more often than not one does not like them when one hears them played back. So far so good. But with the rising sound quality there was also an emphasis on “beautiful” sound, especially for string players. So instead of playing the music as alive as possible and taking some risks in order to achieve this, the goal became to be accurate throughout and to sound “beautiful” throughout (my favorite example is Anne Sophie Mutter). Now this is the very essence of sterility.

    One has to acknowledge though that there are examples of performers that do not follow this trend, Nadia Salerno Sonnenberg comes to mind or Gidon Kremer.

    Footnote: read up on how Rossini treated his singers and you have your precursor to Stravinsky.

  6. 1. “Fritz Kreisler had the authority and genius to do things that you and I should rather avoid–or at any rate I should avoid.”

    Authority or genius has nothing to do with the principle of changing a piece of music in order to fit individual personalities. Kreisler changed things long before he had any authority and long before his genius was recognized. Other less authoritative figures (often obscure ones) from the time period would frequently change things. It is not correct to attribute this attitude to just one person who has “authority”; rather, this attitude was taken by most musicians during the early part of the 20th century.

    2. “there is no need to disregard the text of the score to be original. I go one step further and say: read and analyze and feel the score carefully and take your time over it. You will inevitably arrive at your own original recreation of the work.”

    While there is not a need to disregard text in order to render original conceptions, why does the text “need” to be followed? You still have not directly answered the question. Why should decisions not be made from a certain mood? Often times they yield positive results; even if they yield negative results, it is at least usually interesting. And if it is not interesting, oh well. But what is “not” interesting is hearing roughly the same “interpretation” over and over again.

    3. “Tempi are not really Metronome figures they are character indications (similar to your distinction of “pitch” and “tone” or whatever your terminology is)”

    Pitch and note, and the term “tempo” is not similar. Tempo refers to the speed of the pulse – a metronome marking. Often ‘qualifiers’ are added to indicate the character of a piece. Incorrectly, terms such as ‘allegro’ are mistaken by ‘charlatan pedantics’ to mean “happy”. This is true, but it historically does have a relatively absolute range for metronome markings.

    4. You are quite right about the dangers of recording, and I think I even wrote an article about that a while back. If not, it will be in the musical villain series. It turns out that recording has ironically contributed to the insularity of classical music, and to the arguably negative state where it is currently stuck.

  7. Starting with your point 2: here is my answer: the mood of the performer is irrelevant to the audience; the mood of the performer (or indeed the composer) should not influence the performance any more than the mood of a baker should influence the quality of his bread.

    It is true that performers of the early 20th century felt free to do things we generally don’t do any more (one of those things is–to be blunt–lousy intonation). This does not obviously mean that they were right. And since they could not/did not double check by listening to recordings it does not even follow they intended it to sound the way it did. Here is an incident from his early career Louis Spohr reports in his autobiography: some critic had criticized about Spohr’s performance that he took way too many liberties with the tempi. Spohr got angry, but then he went and double checked himself in practice. He ended up agreeing with the critic and strove to be more disciplined in his tempi going forward. And I think this is very much what this is about.

    Her is another way to look at this: I have this from Armin Schibler, a Swiss composer who was one of my music teachers at school (his bread and butter job; I remember playing 2nd violin in the student orchestra in his concerto for 5 percussionists, 2 pianos and strings). He built his classes around the history of music and spent quite some time on the 20th century, or the part of it that had already taken place (that was in the late sixties). Back then the traditional rules of the craft were challenged. A general preoccupation with the technique of composing and a dread of doing anything like it used to be done in the nineteenth century had taken over the field; more traditionally minded composers (Vaughan Williams, Shostakovich, Hindemith in his later years) were considered less than stellar just because they insisted on using some sort of harmony in their works.

    Schibler told us more than once that the difficulty for creative work in this environment was the fact that everything you wanted to do was allowed. If you are allowed to do anything, nothing has any meaning; only if you work against the “resistance” (Schibler’s word) of some system of rules can you be truly, meaningfully creative.

    I believe that, inasmuch as performance is a creative activity this same principle applies: the resistance you are up against are the composer’s instructions (which of course originate in his intentions).

    As to the dangers of recording: I believe that here lies the cause of what you deplore much more than in one person’s statements–even if that person is of Stravinsky’s stature. It is worth nonetheless to point out that the invention of recording is a good thing: it brought about means for performers to self assess their work. It also created a situation where “not only lascivious monarchs have the privilege to enjoy chamber music” (Adorno).

  8. There is quite a difference between mood and spontaneous decisions. Funny that you seem to say “lousy intonation” in such a way where that is a undesirable attribute. I don’t think it matters too much, although more often than not the intonation was great. The performers much more frequently changed notes consciously. This is a good thing in many cases: there is no reason to assume that the composer’s intentions are actually best. Performers can make artistic decisions over the composer if those decisions are warranted in their own mind. Sometimes the changes make things better.

    Let us take the case of Alfred Cortot, who actually rewrote some of Chopin’s music and changed notes in order to fit his personality. Like me, he didn’t think the notes too much mattered, and did whatever he felt like a lot of the times. Well, Cortot is arguably the greatest pianist of the 20th century. Not for his technique – which was nonetheless fantastic – but rather for his musical insight. The “wrong notes” and “rewriting” does not affect that; indeed, perhaps it adds to the musicality.

    Again, I ask the question you still have not answered. WHY are the composer’s intentions more important than the intention’s of the performer? What makes the composer any more important?

    Recording is indeed a double edged sword, which I will be writing about this week. The performers assessing their work has contributed to the deplorable state in which musical performance is currently stuck.

  9. Your question: composers spend time on their work. Everything they put into the score is based on a considered decision. It is no more than simple respect if a performer (second hand creator if you will) takes the score seriously. I opt for respect. And I submit that you are not answering the counter question either except by assertion.

    More importantly: this question is not really the core of what you postulate: that the performers’ freedoms are essential to a performance culture that you would deem acceptable. And there is my real disagreement: you can produce exciting interpretations while still following the composer’s recorded intentions. Moreover: changing the music does not guarantee a good performance as you seem at least to imply (are you sure by the way if Cortot’s freedom was not in fact a memory lapse?–there are important and less important notes in a composition, especially for the piano; I myself could never memorize a series of Chopin pieces note perfectly and I never knew why performers are being forced into this masochistic playing from memory).

    The intonation question does not really belong here: but there is such a thing as lousy intonation (and in old recordings of for example the Budapest quartet) though you make some good points.

  10. It is great that composers spend time on their work, but did you know that often the articulations and expressive markings in a piece of music are often not even added by the composer? Rather, they are added by some arranger or editor.

    Joachim and Auer actually modified bowing and chords in order to better fit the instrument and their interpretations of the phrase. Cortot changed notes in Chopin’s music in order to bring out certain colors that might not otherwise be seen.

    My belief is that composers have a responsibility to the music, not to the composer. If a composer spends 100 hours on a piece, and a performer believes the piece can be improved, then the performer can and should improve it. There is no good reason to respect the intentions of the composers, especially of composers who have been dead for hundreds of years.

    But remember, the “sterility” that exists in today’s performance atmosphere is not all Stravinsky’s fault. While he certainly had a huge influence, there are four other main factors – perhaps even more important – that we will examine later this week.

    I am sure that I will be hearing from you.

  11. My belief is that you intended to write: performers have a responsibility to the music not to the composer at the opening of your third paragraph… Sorry, can’t help it.

    I’d agree and say that everybody in this business, even the audience, has a responsibility to the music. But the composer and his music are not entirely separate entities. And classical music has a tradition of being written down and entering tradition in that written “format”.

    You are touching on topics which are a bit peripheral to this debate, such as editors adding markings. Which is a problem: if a piece does need to be amended I want to do that myself–usually the editor is not Joachim, but someone nobody has ever heard of. Look at the Peters edition of Haydn’s quartets: you find lots of markings, mostly dynamic, that are not in the Urtext (usually edited from autographs and first editions); in some cases, especially op. 20, the whole music is altered. If you compare these passages you will find Haydn’s version better and the Peters version easier to play. Now lots of quartets play the Peters version, not even aware how many markings they obey which Haydn did not put there… On the other hand will you find glissandi in Haydn (op. 33, e flat major, trio of the scherzo), indicated by fingerings that are missing in Peters; maybe that sort of “folksy” trick was not genteel enough. So this editor thing opens up a whole new can of worms.

    I start to think that we can not entirely convince each other (and we don’t entirely disagree anyway) and maybe should move on to the next villain (villains being more fun than good people). I had been hoping someone else would jump in with different view points.

  12. Well, here you are – a different viewpoint! I’ve only just seen the debate. I speak from the point of view of both composer and performer, and I’d like to add a few questions:

    1. A conductor once emailed me to say he wanted to replace the Amen at the end of one of my pieces with a cadence from earlier in the piece. I intended the Amen to be a questioning, ambiguous finish to the piece, after some punchy and dynamic writing. It was a carefully thought-out decision when I wrote it. It was what i wanted. How should I react to the conductor’s request, in each of your views?

    2. Does a knowledge of the conventions of the time the piece was written matter? Take early choral writing (say up to the Renaissance), where there are never indications of tempo, and there is musica ficta and suchlike where one edition says this note should be sharpened, and another says it shouldn’t. At the time, I don’t doubt that the composer and performers knew the conventions – the written music wasn’t ambiguous to them, and the tempi would be generally understood.

    3. Coming to the Baroque, there were conventions of ornamentation and embellishment. When I was a child, these conventions were generally ignored – even, for example, written trills in the B minor Mass and Messiah and so on. Was this though ignorance, inability on the part of performers to be creative (and express their individuality), or because they couldn’t do them? Did it matter? Even now, mostly choirs never attempt written trills, and that I know is often because it is hard to get a choir to do such things unanimously and cleanly, and there is a tradition that you just ignore them as some kind of aberration. Is this interpretative freedom, or laziness and technical ineptitude?

    4. Given that no composer can precisely notate what they mean, a performer will inevitably have to make interpretative decisions. Albrecht raises the question of how far this can reasonably go. Clearly nuance and rubato are always in the performers’ hands, degrees and linearity of crescendo and decrescendo likewise. Clearly too, the acoustic of the performance area affects tempo, and general markings like Allegro are pretty vague anyway. But how far can performers go and still claim they are performing John Smith’s Symphony in A major, for example? Personally I would say that performing it in A minor is more than interpretation. And if the second violins decide they’ll play it at a third of the speed of everyone else… might be interesting, but can you really allow such interpretative freedom and still claim it’s the same piece?

    I think these are very important questions, and it is not a matter of which is more important, the composer’s vision, the written score, or the performers’ interpretative freedom. It is a fusion of all of these. To selectively ignore what the composer writes is a gross and arrogant insult, to ignore the conventions and deficiencies that surround the notation of music and to treat it as literally exact is ignorant, and to try to stifle the interpretative freedom of the performer is to kill the music. I remember learning Poulenc’s Flute Sonata for an exam when young, and when I played it through with one of the music teachers in school playing the piano part, for the first time I realised what it – and music generally – was actually about. He illuminated the notes, he understood the music, and that caught me up and my whole sensitivity to performance changed from that moment. I had been playing the notes, and observing the stage directions before – now I was playing the music. It was still the same notes and dynamics and tempi and everything else, but now I added the something that made it sing and speak to the soul – the umami of music, perhaps. That to me is what interpretation is all about. It is all sensitivity and understanding – but it all originates in the composer’s vision. There is still room for an infinite number of different viewpoints without ignoring what the composer wanted, or what is on the written page, or denying the interpreter freedom. That is what makes music live. If current performances sound like performers’ creativity is stifled, I think the cause lies elsewhere…

  13. Or is it arrogant for the composer to assume his ideas are better than the ideas of a performer? Taking ideas and modifying them to fit personality (like Cortot) can often improve the composition.

    You say that that not following the composers’ markings is disrespectful to the composer. That inherently places the composer above the performer as though there is some sort of hierarchy. It is equally disrespectful for the composer to complain about a performer changing something – why should the performer’s ideas be stifled by the will of the composer?

  14. This is a great way to say what I have been trying to get at, Oliver. To your points:

    1. Your anecdote recalls something I left out of earlier posts to save space: there is a privilege in working with composers who are alive: you can ask them to explain their vision; you can even challenge them on their decisions. This conductor is suggesting a very major change to your piece; the entire trajectory of your music is altered (personally often wince when “amen” comes along as an exhilarating fugue undergirded by trumpets and timpani, though that happens quite often–even the words “dona nobis pacem” at the end have this upbeat-ending-tone in so many masses. So–though not knowing the piece–I’d risk to say I’d prefer your version of it.) If you want to accept his suggestion or argue against it is of course your decision. There is the story about the Violin sonata by Cesar Franck: Ysaye wanted to play it slower than the composer had envisioned it and Franck actually changed the tempo marking of the first movement to sanction the change (this is of course a much smaller change than the one in your anecdote). So it is possible for an interpreter to see things differently from the composer and be “right” about it. But we can not go and ask Mozart about it and this puts more responsibility on the performer.

    2. & 3. The conventions of the time of the composition do matter in my opinion (though it is also important to remember that we are playing Vivaldi to people who have heard Shostakovich; Vivaldi’s vision is possibly better served by using more modern conventions such as “well tempered” tuning than by painstaking imitation of baroque practice). But the fact is that our understanding of these conventions keeps changing and is not always accurate. Example: Mozart- and Haydn-symphonies are typically played nowadays (not in my youth) with 6 first violins, maybe 30 musicians total, because few venues were able to mobilize more than those back in Mozart’s day. But there is a letter by Mozart to his father where he described a performance of one of his symphonies with 40(!) violins and all winds doubled (I believe in Mannheim). He obviously loved it. It seems we have set ourselves a rule that Mozart would not have supported. For Baroque: I still remember “Terassendynamik”, the idea that everybody handled dynamics the way a harpsichord has to (due to technical limitations) and that there were no crescendi or diminuendi, just abrupt changes. Nowadays it seems absurd. Also: The continuo has been played by a basse instrument (Cello, Bassoon) combined with a chord instrument (harpsichord, organ, lute etc.) for quite some time now. But then came Andrew Manze and claimed that new evidence suggested that the correct execution is with harpsichord alone…
    The fact is that we don’t have all that many sources (Leopold Mozart’s “Violinschule” and Quantz’s book about the flute are the most often quoted sources and they deal primarily with technical problems; besides they come very close to the end of the baroque) about this stuff. There is an edition of the Corelli sonatas (from those days) with embellishments which the editor claims were the ones Corelli himself used in performance. If you look at them closely you see that they often cut against the phrasing of the original (i.e. the un-embellished text of Corelli); I can’t imagine that a composer would embellish his works like that.

    And Bach of course (also Handel in some works) wrote out the embellishments (almost always); no point in adding a second layer. Depending on the character and function of the movement in question there is more and less embellishing going on, a sarabande has less of it than a slow movement in a sonata etc.

    Choral trills are technically demanding and eat lots of rehearsal time. I can understand if amateur performers sing the works without the trills. And I think that this compromise is better than a trill that is just a “cluster” of people trying supervibrato. Not even all soloists can sing proper trills.

    4. On this just a question: repeats are often skipped (though not as often in the age of the CD as they were when the LP was the recording medium…). As a composer what do you think about that? In many works there are prima volta / seconda volta endings that strongly suggest that the composer envisioned the repeat. Also separate markings for the first and second time (Haydn, emperor quartet, first movement, “seconda volta piu presto” for the coda).

    I’d like to reformulate what you say at the end from the composer’s point of view to reflect the performer’s task: The source for the composer’s vision is the score (in vocal works also the text); it is the primary guide. The task then is to discover, re-envision the composer’s vision through study of the score and to find means to make this vision sound. My point in all this has been that there is plenty to do which does not involve changes to the text of the score; you can follow every marking conscientiously and still produce live (in the literal sense of the word) performances. Which would mean that more freedom for performers is not a necessary recipe against “sterility”.

    NLewis, you seem to me to resemble those opera fans who are convinced that the true art of singing has been lost and nobody can sing nowadays like Maria Callas. When Callas was alive and active of course the same people said nobody could sing like Malibran…

  15. I actually think a lot of opera singers can sing these days – even the ones who are not so famous. A couple of days ago I was with Kara Shay Thomson, and she is a fantastic singer. Good musicians do exist these days, they are just hard to come by. Also, I do not particularly like Maria Callas. But that aside, it turns out that opera singers are usually the most musical of the batch. As much as I often dislike jazz, I think that jazz musicians are superior in this day and age to classical musicians for the reasons I’ve been hashing. Classical music used to be approached much more like jazz is approached now.

    When I conduct I sometimes use an idea that was often utilized by Stokowski called “free bowing”. The composers would likely dislike this idea as we ignore the bowing written on the page in order to achieve a musically lush tone. And I have to say, although “appeal to authority” is a fallacy, I come to the following ultimatum: good enough for Stokowski, good enough for me.

    And what do you think about Bernstein conducting Shostakovich’s fifth symphony? Did you know that he changed the tempo of the final part of the finale to about twice the written speed?

  16. Nicholas, I think you’re rather subverting the very important issues you raised in your original article, which, if I read it correctly, are:
    1. Has the modern performer lost interpretative freedoms that they used legitimately to have to alter what has been prescribed by a composer?
    2. Does a composer disempower interpretative performers by trying to dictate to precisely to them what they must do in every respect, and thus sterilises the music?

    And that collectively, these have sterilised musical performance.

    Let me try to address these.

    1. I never heard Cortot live, and probably neither you nor Albrecht did. We have read about him and listened to his recordings – and both of you have qualms about what recordings have done to musical performance. His live performances were probably inspirational, and if we had been there, we probably would have forgiven the wrong notes and memory lapses, and never have forgotten the occasion, but we don’t know this first-hand. It is invidious to hold him up as a shining example of performance practice and the power of the interpreter under these circumstances. Witnessed live (if live performance is important), most of us most of the time experience less than brilliant performances by less than brilliant performers. How much license would you like them to have? Where Bernstein can decide to double a composer’s tempo marking, with his background as performer, composer and generally rather flamboyant musical genius, I’m not sure that I’d want Fred Bloggs, plumber by day, who conducts the Smallsville Amateur Symphony Orchestra to feel he has the interpretative authority to do the same. But he can still inspire his players to produce a performance that is as true to the score as the players’ faulty techniques allow, and nevertheless excites and moves the audience, who maybe last heard a live orchestra at their last concert six months ago. Or it may be a rubbish performance and I find I’m bored to tears. But whichever, that’s not sterility in my book.

    I reiterate that there are copious possibilities of interpretative freedom in the scores of all composers, even those who smother their scores with interpretation marks (like Ustvolskaya, for example – how would Cortot have interpreted her music? There’s a question for you!).

    2. A composer, through notation or lack of it, tries to enable performers to recreate what he hears in his head or otherwise imagines. The composer can choose deliberately to leave major decisions up to the performers (like instrumentation, or tempo, or pitches, or indeed everything, in the case of some graphic scores), or to chance (as in some of Cage’s works). With other parameters, the composer will expect certain conventions of his time. For example, in Renaissance and Baroque times, he would expect that performers would add their own ornamentation whether he liked it or not. As you say, Albrecht, Bach might sometimes strive to prevent it by writing the ornamentation out in full, so there was no room for the performer to add more (I bet that didn’t stop some of them, though!), but otherwise it was likely to happen. Puccini could write I think it is an eight note top C in Nessun Dorma, but would have known full well that tenors would dwell there for as long as their technique would allow, ignoring the score and milking the applause for all they’re worth. He would know that. Why didn’t he write a long note with a hefty pause on it? Did he actually wish the tenors would shut up and get on with the plot, or did he happily accept it?

    But whatever you may say, Nicholas, and whatever the composer may write, there is always a lot of interpretative freedom possible while adhering to the score.
    This is not to say that composers are always best judges when it comes to those parameters that they do prescribe – far from it. But who is to decide that they are wrong in major ways? A great artist, like Cortot or Bernstein? A maverick like Glenn Gould? Another composer? Any old Tom, Dick or Harry? Well, yes, of course, any of them! Who’s to stop them? But who’s performance would you want to hear? For my first encounter with a piece, I’d rather like to hear a reasonable stab at what the composer intended, as far as that can be done. If I hear it lots of times, I’d be interested in what other performers can do with it without straying too far from the composer’s markings. And then sometimes, for variety, I like to hear other more extreme interpretations. But first, let’s hear what the composer’s saying.

    I put it to you, as a conductor, Nicholas, what would your feeling be if you were conducting Ravel’s Bolero and the side-drum player decided, as an interpretative act, to perform a gradual and continuous accelerando through the piece – something that would be very hard to ignore and fight against. You’d either have to stop the piece or go with him. I understand that it is a gruelling part for the side-drum on account of its excessive repetitiveness, and an accel would really express his frustration. And it could be enormously exciting! But you’re the conductor, it’s in performance, what do you do? Pat him on the back at the end and say “Great to hear you expressing yourself!” Or do you say “You fired! I make the artistic decisions!” Who’s the performer? Who does the expressing? Remember Ravel didn’t rate it much as a piece of music.

    As a footnote, concerning opera singers, there are those I like and those I dislike from all eras. My opinion is frequently not the same as the critics or popular opinion. Ah, critics! There’s another beef for you Nicholas! What power do they have and what good and what damage have they done over the years?

  17. It seems I have to explain my allusion with the opera singers. There is a sort of opera lovers, who always think that the art of singing is disappearing from the world and that nobody can quite sing like Caruso (or Callas) anymore nowadays. There are also movie lovers who think nobody nowadays matches up to Hitchcock or Laurence Olivier or Charlie Chaplin. What I tried to hint–half seriously–is that Nick belongs to this group and thinks that nobody nowadays plays the piano as well as Cortot, nobody the violin like Kreisler etc.

    And it is true that Cortot and Kreisler (and Callas) died before any of the three of us has a chance to hear them live.

  18. I just noticed that–apart from some examples–my own favorite area, chamber music has not been introduced into this discussion. It offers some different angles.

    Following up on the example of the sidedrummer who could, if he wanted, “sabotage” a performance of “Boléro” by keeping to accelerate (he might well risk his job doing that on purpose; conductors as a rule are more dictatorial than composers): this can happen in chamber music: if any one member of the ensemble keeps accelerating the others can’t do anything but to follow, at least to the next G.P. (If one person drags there is a good chance to “fix” it while playing, but slowing a colleague down is impossible even to the person who is supposed to lead at that particular passage). More importantly though, in the absence of a conductor every decision is up for debate, sometimes heated. So now we have the text of the composer PLUS several versions of the interpreters to somehow fit into a coherent result. Most debate in my experience involves no change to the text; just different yet legitimate readings of it.

    Which brings me to Shostakovich’s piano trio (no. 2), which I once worked on for performance with my friends (I should admit to being an amateur; our coach suggested the choice, but it was one of the highlights of my musical life and we gave a creditable performance for our friends).

    We listened to several recordings to see what other people made of it. It turns out that some people have actually changed the text and added glissandi and much later than Cortot. In the first trio section of the scherzo there are a sort of “inverted sigh” motives, with a crescendo on the long note and with the short noted loudest, played with up bow in the strings. The recording with Yo Yo Ma for example (others also) plays these motives glissando. Gidon Kremer on the Argerich/Maisky/Kremer recording (extremely impressive though very opposite to Shostakovich’s own style) does very ugly things with glissando in the last movement. There is glissando in the score, but not at these places.

    There is a recording around with Shostakovich, Oistrach and a Hungarian cellist (Milos Sadlo?) from 1948 (plus or minus a year or two). There is none of that on it, the whole recording is dry and matter of fact, just giving the music as written. The piano playing is far from perfect technically (I believe Sh.’s had disease had already started; the strings are flawless). The tempi are fast, almost as fast as the metronome markings, there is hardly any rubato, the dynamic range is relatively modest, no screeching fff (though fff occurs at several places in the score; there may be technical reasons for that though).
    Yet the performance is incredibly alive and in large part it comes from Sh.’s piano playing, particularly also when accompanying: the way staccato is differentiated, rhythm is subtly underlined is masterful and utterly devoid of performer-vanity. To me it is the best recording of the piece I have heard (the trio is considerably more impressive live, so don’t miss a chance to hear it).

    This I think is the art of performance at its best: reproducing the score as written and filling it with life by means of phrasing, accenting, articulating the notes and phrases. No gimmick can live up to such an achievement.

  19. I’m sorry, but this article is wrong on some basic facts. Stravinsky was never a “frequent lecturer” at Harvard, nor did he publish many articles. He gave the Norton Lectures in Poetry at Harvard in 1939-40, reading from a manuscript which had largely been prepared by Roland Manuel, delivered in French and published only in 1947 in English as the Poetics of Music. His remaining public writings were his (also ghost-written) 1936 autobiography, Chronicles of My Life (which does not make any argument of the sort described in this item) and, much later, the dialog books with Robert Craft. The scarcity of his writings and their late transmission should be rather strong evidence that Stravinsky was not the rhetorical driver in any dogma of the sort described here and his writings concern performer’s restraint with regard to individual expression convey a far more subtle point.

  20. This is very interesting; I know nothing about it. It means Stravinsky is innocent no matter who is correct in the controversy here.
    However, you end your comment with the “far more subtle point” Stravinsky was making. Do you think you have time to expand on this a little and let us know?

Comments are closed.