“How many scores of Mahler 5 do you own?” came the beleaugered inquirey from one who knows the ins and outs of our library budget here at Vftp International Headquarters.

The reluctant answer is “over four.” I’ve got the Dover (useless for performance, but interesting for comparison as it is almost the earliest version of the symphony, without most of Mahler’s later changes), a pocket score of the Erwin Ratz edition from 1962, the octavo score of the 1999 printing of that same score corrected and updated by Fussl, and am now working from the nearly brand-new critical edition edited by Rheinhold Kubik. On top of these, I have reductions for piano and two-pianos and a score of just the Adagietto. So, over four….

There are many interesting differences between the Kubik and the Fussl editions, but the most dramatic change may be at the end of the Adagietto. In the Ratz/Fussl edition, Mahler has marked “Drangend,’ or “pushing forward,” over the last eight bars, while in Kubik, the same eight bars are marked Sehr Zuruckhaltend, or “very held back.”

I’d been aware of this discrepancy for some time because when I first did the Adagietto on a conducting recital in Cincinnati, the parts had the Sehr Zuruckhaltend, not the more commonly seen Drangend. Kubik’s research clarifies things-

“”Sehr zurückhaltend” {“Molto ritenuto”} in Aut, StV, EA-Stp, St1 and St2, “Drängend” {“pressing on”} only in EA-Dp. In Korr2 “Sehr zurückhaltend” is struck out, in its place is written “Vorw.[ärts]” {“onwards”} and in bars 99 and 101 “rit”; in W-Stp “Sehr zurückhaltend” is struck out and in bar 100 restored.

That both sets of parts — the one which was demonstrably played from several times in 1905 – 1907, and the one which contains Mahler’s last Revision realised in New York — stick to “sehr zurückhaltend”, is ample reason to adopt this reading in the present edition.

If an interpreter decides on “Drängend”, he or she should remember that there are sources according to which the last four or five bars should then still be slowed down. “

What I find interesting about this is the implications of this discrepancy in understanding Mahler’s evolving thinking about other places in this and other works. One of the central tenets of Mahler scholarship seems to be that Mahler never really changed his initial concept of a piece or it’s details once he’d written it, and that his many revisions are primarily practical in nature, not changes of concept. In essence, we’re told that as Mahler performed his works, he discovered more and more what did and didn’t work in the orchestration, and that all those markings in the parts reflect his practical sense of what players needed to do in order to make his music sound in a hall the way it sounded in his head.

I’ve written many times before about Mahler’s use of notation, and the fact that because of this approach, his notation is more a representation of how is music should be performed than how it should sound. Conductor and Mahler scholar David Pickett takes this one step further- if the final versions of Mahler’s scores tell us most accurately how the music should be performed, perhaps the original versions tell us more accurately how they should sound.

“While it might be expected to confer great benefits in the case of a composer whose watchword was clarity, the adoption of extra microphones has often resulted in grossly distorted balances. Mahler’s post-publication changes to his scores reflect his experience in the concert hall; and in recordings, when microphones are placed close to woodwind, harps and percussion instruments, it is sometimes helpful to know what his first thoughts were.”

You can hear an example of this here on his website. I’m also one to argue in favor of ditching some of those extra mics.

Along these lines, a controversy has come up in recent years regarding the last note of the first movement of the Fifth Symphony. In 2002, the New York Times published this long article by Gilbert Kaplan examining the question of just how loud that pizzicato note should be. For many decades, almost all conductors interpreted the sf on that final note to mean a loud thwack- some even used a snap pizz for extra volume and violence. Kaplan reads that passage more literally, noting that the last dynamic in the parts is pianissimo 13 bars earlier. Kubik joined Kaplan in looking into this, and comparing the different sources and found that Mahler’s revisions all seemed to point to a soft ending-

“For one thing, he marked the note to be played softly from the beginning. In his original handwritten score, he wrote pp on the note itself, and pp also appeared in the first published edition.

After conducting the premiere in Cologne in 1904 and a second performance in Hamburg in 1905, Mahler changed the pp to p, a modest boost in dynamics. He also added an accent.

Some months later, Mahler passed these changes on to Mengelberg, who, as resident conductor, was preparing the Concertgebouw Orchestra in Amsterdam for a performance of the Fifth to be led by Mahler. After arriving in Amsterdam, Mahler wrote to his wife, Alma, that the orchestra had been ”splendidly rehearsed in advance” and that Mengelberg was the only conductor to whom he could entrust a work of his ”with complete confidence.”

Mengelberg’s score reveals that Mahler, in red ink, crossed out pp, substituted p and added an accent… The minor change to sf showed up first in the orchestral parts used for a concert Mahler conducted in Strasbourg in 1905, and it remained in all later editions of the score and parts.”

Perhaps most compelling, however, is the fact that Mahler himself recorded this movement on the piano in 1905, and does, indeed, play the last note very softly.

Kubik sounds completely convinced-

“”What he wanted was a soft last pizzicato,” writes Reinhold Kubik, the chief editor of the critical edition of Mahler’s music, who has just completed work on a new edition of the Fifth Symphony”

However, if that’s what Mahler wanted, it seems that the earlier versions make this MUCH clearer than the later ones- in the first version, Mahler had marked the note itself pp, in the second version he’d marked it piano with an accent. Surely either of those are significantly safer ways of asking for a soft final note than a sf on a pizzicato when the last dynamic was 13 bars ago, when the strings were playing col legno, not pizzicato.  Why replace something so clear with something so ambiguous?

Kubik seems convinced that the original version is the key to understanding what Mahler wanted here-

“Mr. Kubik writes: ”Investigating the many, many revisions Mahler did in his symphonies, you find that he almost never changed his original intentions. He just changed the means by which they could be made audible.””

And, perhaps most convincingly in addition to the piano roll recording from 1905 is a review of Mahler’s performance of the piece in Vienna a few months later, after the change to sf had been made which described the end of the movement as follows-

“Max Kalbeck, in his review for the Neues Wiener Tageblatt, wrote that the first movement had ”an ending that fades away quietly” and concluded with ”a final muffled pizzicato from the lower strings.””

Still, we know from Mahler’s letters that around the time he wrote the Fifth he was looking intently for ever more clear and specific ways of notating his intentions. Why would he then end up with such an ambigious way of notating this note, when there were so many obvious ways to write it clearly. Remember, not only did he write sf, he REMOVED both the piano and pianissimo that had been there before. Also, note that all of the revisions are in the direction  of that note being louder, from pp, to piano with accent, to sf. The piano roll recording is from 1905, but we know he was working on this piece up to his death. Is it possible that his concept was actually evolving, not just his way of specifying the execution? Is it possible he felt diferently about this spot in 1902, 1905 and 1910? Could those views have not simply evolved, but fundamentally changed? The end of the Adagietto, with its changes between “speed up” and “slow down” doesnt’ tell us this IS the case, but it tells us this COULD be the case- he could have changed his concept from soft to loud, even to very loud.

Kubik’s point about Mahler’s revision process being one of clarifying details, not changing intentions seems to argue against any change in concept, but the end of the Adagietto tells us that he in so important a moment as that, he could go back and forth between OPPOSITE concepts. Of course, on one level his concept remained the same- the music is getting more intense and the tempo modification is an expression of intensification- but the nature of the tempo modification changed as radically as it possibly could have. (His indecision about this point seems to indicate that he wanted a LOT of intensification and wasn’t convinced either version was generating enough), But the same could be true for a loud final note- either version could express abrupt, desolate finality, in which case, the concept would have stayed the same in the broader sense.

All this tells is that linear, deductive reasoning is only of limited value in determining Mahler’s intentions. We can’t always reduce all the evidence down to whether or not the butler did it. By the logic of the Kaplan’s and Kubik’s argument for the last note of the Trauermarsch, it should be impossible for us to accept Mahler’s change in the order of the inner movements of the 6th Symphony from Scherzo-Andante to Andante-Scherzo. Surely we should accept his original concept as definitive? However, Kaplan and Kubik have consistently been among the strongest advocates for Andante-Scherzo (there’s another Kaplan/New York Times piece on that issue here).

I personally am thinking that Mahler was trying to elicit something more complex than just loud or soft from that last pizzicato. Mengelberg’s score has the word “dof” or dull. That seems good- like a muffled funeral drum. This coda is full of seemingly contradictory information, like the combination of pp and schwer (heavy) at figure 19 a few bars earlier. However, that’s part of the richness of this music- trying to resolve and integrate complex and contradictory ideas, not to water them down into things one dimensional and simplistic. Mahler abandoned a perfectly clear way of asking for a soft note in favor of something that’s much more ambiguous, in a musical context that is full of complex emotions. I think we’ve got to come to terms with that.

Now- hear this. The last few bars of the Trauermarsch as recorded by Bruno Walter, replete with loud final thwack. (It’s worth noting that Walter’s 7’ 35’’ Adagietto is often cited by scholars as a prime piece of evidence for a faster tempo in that movement, but Kaplan dismisses Walter’s reading of this note in his times article: “Yet other conductors, including Otto Klemperer and Willem Mengelberg, were also close to Mahler and heard him perform, and many of their interpretations are quite different from Walter’s. One should also bear in mind that Walter recorded the Fifth more than 35 years after Mahler’s death.

Let me make this clear- I’m not arguing that Kaplan and Kubik are wrong about this pizzicato. What I am saying is that their richly researched scholarship seems to point to more nuanced and complex readings of this and other points than resolving along simple lines of binary oppositions between loud/soft, fast/slow, Andante/Scherzo-or-Scherzo/Andante. We’re being forced to reduce everything down to whether or not the butler did it. 

Maybe soon we can talk about the Adagietto, and not just the last 8 bars. It’s been so widely debated and discussed these last few years in a way that seems to, again, direct the discussion towards very simplistic, binary readings of the piece, changing “it’s about death” to “it’s not about death, it’s about love.” Again, it seems liket the essence of Mahler is the co-existence of contradictory ideas- can’t the Adagietto be about death and love? What could be more Mahlerian?