Beethoven 9 Pontifications, part II. Decoding those pesky metronome markings

The metronome mark  problem.

If you’ve been reading the classical music press over the last month or so, you would swear up and down that the great Riccardo Chailly had personally discovered the metronome markings  in the Beethoven symphonies.

Not so! It turns out those markings have been in the music for years, that a few other conductors have actually noticed and considered them over the years.

I’m generally a “fast” Beethoven guy. A “do the metronome marks” guy….

Last year, a critic actually accused me of conducting the 2nd Symphony so fast that I must have had a train to catch (I didn’t).

Still, it is a cause for concern when people become so myopic about adhering to a number written by a composer who never had the chance to compare the impression of listening to his music live to what he had heard in his mind’s ear that they don’t listen critically to the musical results that come from adhering to or ignoring those marks. Almost everyone I know (including me) tends to hear music faster in their head than they want to hear it performed live.  My rule of thumb is the Mister Rogers method of metronome use- you’ve got to be “somewhere in the neighborhood” of the metronome marking. How big that neighborhood is varies from piece to piece, and performance to performance. I’ve done performances of the first movement of the Eroica (one of the notoriously fast metronome markings in the Beethoven symphonies) that felt perfectly poised and spacious right at Beethoven’s suggested tempo, but with other, equally skilled orchestras, getting close to the metronome mark always brought a sense of frenzy that didn’t serve the music at all. With time, an orchestra can find more space in faster tempos, but it takes rehearsal, it takes trust, it takes shared experience.

For me, the neighborhood in the 9th is probably a little bigger than in the previous 8 symphonies because the language of the work is more dramatic, dare I say “Romantic,” and seems to call for a more “in the moment” approach, and perhaps because he was another decade or so removed from being able to compare the perception of the outer ear to the imagination of the inner ear. All things considered, I’m likely to be a little freer in realizing the metronome marks in Beethoven 9 than in his previous 8 symphonies.

Still, there’s more to the metronome markings than the exact number of beats per minute (not that you’d get much sense of that from a lot of scholarly writing or conductor manifestos). You can read some of my ideas about reading between the lines of Beethoven’s metronome marks in this blog post, largely about Beethoven 2. (What a coincidence I would write a blog post like that just after being accused of speed-conducting-to-catch-a-train…)

My attitude to the metronome markings in the 9th is slightly more flexible than in the earlier symphonies. There are more obvious mistakes than in the other symphonies, and in the case of the Trio of the Scherzo and the Turkish March in the Finale, I’m not satisfied with any of the simple solutions put forward (generally of the “right number, wrong unit” variety).

Many scholars feel the  “old” dotted-quarter=84 “can’t” be right. It is pretty slow. Therefore, they suggest that Beethoven gave Karl the right number, but the wrong unit- that the tempo Beethoven really mean was dotted-half=84.

The instrumental music of the Turkish March actually sounds rather witty at the dotted-half=84 suggested by Jonathan Del Mar and most HIP conductors (and I agree that dotted-quarter=84 does sound slow). However, Beethoven was far too concerned with text to expect a tenor to sing that many words at that speed- it sounds like a German speed-reading exercise rather than music or poetry, when it should be both. Just getting the notes out at that speed is something only the most agile tenors can do. Guys who can sing that fast are Handel tenors, not Beethoven tenors. Note that the soloists on most recorded fast  versions of the Turkish March are incredibly close-miked.

Also, the fugue that follows and the big 6/8 restatement of the “Joy” theme sound unintentionally comical at dotted-half=84. I take that section at….. wait for it….. the tempo of a rather brisk Turkish March….. About dotted-quarter= 126.

Turning back to the second movement, even Jonathan Del Mar seems to have given up hope of making sense of the metronome marking of the Trio (“thh true metronome marking of the Trio is irretrievably lost”). Printed as Presto, half-note=116 in many editions, this seems perverse because this Presto follows a Molto vivace and a stringendo that starts at, you guessed it, dotted-half=116.  Taken at face value, this would mean we accept that Beethoven was asking the performer to speed up from a Molto vivace to a Presto that was slower than the music that preceded it (remember, if half=116, then dotted-half would be 78).

Some have suggested the correct marking is bar=116 the marking in the old Breitkopf edition, but this is impossibly fast.

I think, hover, that there is a clue in the fact the markings are the same. LvB dictated the metronome marks to his nephew Karl (poor kid). I think he got to the Presto and shouted to Karl, “stringendo to Presto and then the unit speed stays the same,” so Karl wrote down the same metronome marking. What I think Beethoven meant was that the speed of the quarter note should stay the same as one goes through the double bar between ¾ and 2/2. That means my Trio is in one, and pretty darn fast, but not 116.  About 84 to the bar. Del Mar suggests half-equals 160, which is very nearly the same speed, but, in my opinion, absolutely the wrong “feel.” This music cries out to be felt “in one,” not “in two.” Everything about the phrase structure and the harmonic rhythm indicates there is one impulse per bar.

I think there are some interesting threads to explore between the different markings. Rather than using a marking as an absolute measure of speed, it provides a very useful measure of relative speed. One can look at the whole piece and see that Beethoven conceived of some sections or movements as sharing tempi, while others are very closely juxtaposed as slightly faster or slightly slower than ones with which they share thematic or harmonic relationships.

Surely it is no accident that the first and last tempos in the symphony are the same (the beginning is quarter=88, the final Prestissimo is bar=88), especially given that Beethoven uses the same shift from subdivision in 8 to 6 (or 6 to 8 ) in both sections.

Likewise, the tempo of the opening of the 3rd mvt, Adagio molto e cantabile, is the same as the Adagio, ma non troppo, ma divolto, 60 beats/minute.  The Allegro ma non tanto episode (bar 763 in the Finale) is marked 120 bpm. This is interesting, because I and just about every other conductor I know makes a 2-1 tempo relationship at the following poco adagio (bar 810 and again at bar 832). I always did this instinctively, then much later realized that this landed me, in theory, at the same tempo as the other two Adagio’s in the symphony, 60 bpm. This is also the same tempo as the final Maestoso at bar 916.

On the other hand, I probably take the very end of the symphony a little faster than bar=88, maybe bar= 96, and I take the very beginning of the symphony a little slower, maybe 76 or 80.  In spite of the discrepancy, I’m mindful of the connection between the two sections. I hope that comes through in performance.

Also, I’ve always felt the Allegro ma non tanto worked best for me a little slower than the printed 120 bpm, usually settling around 112. I did this for many years on instinct, feeling that it allowed more room to build the tempo in the ensuing Presto and Prestissimo. How pleased am I to find out that this slightly slower allegro tempo means that my 3 Adagios, all settle at the same speed, since I’ve always found 60 bpm for the 3rd mvt Adagio to be a little pressed.

Thus, whenever I see the same metronome mark appear twice, it almost always turns out to be important. The exception for me is the 84 bpm of the Turkish March (bar 331 in the Finale) and the double fugue (Allegro energico e sempre ben marcato, bar 655). As I said above, I’ve got pretty grave reservations about the legitimacy of the 84 bpm marking for the March because of the word setting (unless one accepts the old, slow marking). On the other hand, I’m sure the marking for the fugue (dotted-half=84)  is basically right, even though it’s a good bit slower than most people, especially conductors of a “scholarly” bent, take it. I think when you take it slow-ish lots of good things happen. First, the players can play the notes. The running sextuplets in the strings are usually a sad, totally faked mess, even when played by great orchestra, at the keystone cops tempo for the fugue one so often hears. Second, at a slower tempo the music has genuine grandeur and epic power. I like that. Third, at Beethoven’s tempo, you can hear the moments of dissonance, and follow the contrapuntal interplay far more acutely. Finally, it causes the skies to open, and the holy power of the cosmos to shower down good energy on the audience. Okay, I can’t actually prove that one.

Whatever you make of the tempo of the Turkish March, there is a continuum of tempi in the “80’s” throughout the Finale. The first appearance of the “Joy” theme is 80 bpm. The double fugue is a little faster, 84 bpm, but made of the same musical material (the “Joy” theme, now combined with the “Seid umschlungen” theme), then the final Prestissimo (also made from the “Joy” theme) is 88 bpm; as already mentioned the same tempo as the opening of the symphony. I see this process of tempo intensification as structurally important. The exact numbers might, maybe even should, vary from night to night,  orchestra to orchestra, conductor to conductor, but I do think  each of those three sections needs to get faster. With all that in mind, I also think the logical case for the Turkish March being dotted-quarter=84 becomes stronger.

That which Beethoven says must be the same, that which he says must be faster must be faster, that which he says must be slower must be slower.

Or at least in the neighborhood thereof…..

Related Posts Plugin for WordPress, Blogger...

Spread the word. Share this post!

About the author

American conductor, composer and cellist Kenneth Woods is Principal Conductor of the English Symphony Orchestra, Artistic Director of the Colorado MahlerFest and cellist of the string trio Ensemble Epomeo. He records for the Avie, Somm, Nimbus, Signum, MSR and Toccata labels.

Learn about Kenneth at

All material in these pages is protected by copyright.

6 comments on “Beethoven 9 Pontifications, part II. Decoding those pesky metronome markings”

  1. Evan Tucker

    Even if it’s plainly beyond most performers’ ability, I have to figure that Beethoven meant 84 to the dotted half for the alla marcia. Allowing for the possibility that a bit of slowing down in the fugato is inevitable, it puts the restatement of the Ode-to-Joy theme at the exact same speed as its first statement and allows it to grow to maximum boisterousness from the extreme sparseness of its first statement from the cellos.

    Also, if memory serves correctly, I’m also fairly sure that at least Gardiner and Norrington both take the double fugue around the correct tempo.

    In any event, any number of solutions to Beethoven work. The music can take it. For me, the two greatest conductors of Beethoven in today’s scene are Barenboim and John Eliot Gardiner (whose DC concert I missed to my terrible regret), so IMHO any number of solutions can work musically.

  2. Kenneth Woods

    @Evan Tucker

    Evan- good to have your feedback. Barenboim and JEG! Now that’s broad tastes!

    Let’s say for argument’s sake that one of the two versions of the Turkish March is correct, ie that Beethoven intended it to be 84. Certainly the other markings in the 80’s make this seem very possible. I’ve never heard a tenor, no matter how great,avoid turning the fast version into comical goobledigook, and the slow version feels awkward and stagnant. I’m not sure there’s any amount of agility and ability that could make the tenor aria work at that speed. What to do then?

    I must admit, there’s something to be said for Furtwangler’s willingess to just change the tempo for each subsection to something feels “right” Probably not something we mortals should imitate, however…

  3. Evan Tucker

    In every particular, I completely agree with that assessment. On the other hand, Beethoven was never particularly considerate about performers with his markings (I’m sure you’ve heard that old story about Beethoven and the violinist Schuppanzigh). But I also wonder if that has as much to do with singing technique than as with the tempo itself. I would at least imagine that the tenors of Beethoven’s day, accustomed to Rossini and Salieri, would have far more thorough coloratura training than singers today do. The speed is no faster than conversational German, the problem is, like so many other Beethoven metronome markings, it’s impossibly fast for what the performer is asked to do.

    Beethoven 9’s such a large piece that it can’t be completely controlled – no matter how many details performers go over, there will be some that are missing and have to be left to the moment. My personal favorite conductor of the 9th is probably Herman Abendroth, and tempo wise he often seems to change from bar-to-bar. Yet it works.

  4. liz garnett

    What an interesting discussion – thank you!

    Two things wandered through the back of my mind as I read this:

    1. The idea that a metronome mark controls the entire section/movement until the next one comes along is arguably a more modern idea than 1824. So Evan’s pleasure in tempo flexibility within sections makes sense.

    2. It often strikes me that there is something of an inverse relationship between tempo and timbre – the fuller and more resonant the sound, the more space it needs not to become a mush. I’m sure this instinct is partly conditioned by the correlation between faster tempi and less vibrato that one typically hears in period instrument performances. But there’s also a metaphorical sense of stateliness versus nimbleness too. So the attempt to find the ‘right’ tempo is always conditioned by the actual sounds of the ensemble involved.

    Both of which ideas support your notion of ‘neighbourhood’ of course.

  5. Kenneth Woods

    @liz garnett

    Very perceptive, Liz. You are quite right in no. 1 above- Beethoven explicitly said as much, but so many modern performers are scared to look for any flexibility.

    Also completely right about number 2- at what point does the sound actually suffer or benefit from tempo? Just as an overly quick tempo might make full-bodied music sound anemic, one that is too slow can also make the sound stodgy and unpleasant.

    Thanks so much for the comment

  6. Michael Roháč;

    Hello Kenneth, I am a little confused with your discussion of the Turkish March and subsequent fugue. At the top of your article, you say that you “take that section” at dotted-quarter=126 … which section? The March, the Fugue or the “Freude” restatement?

    I am confused because then lower down you discuss the various possibilities surrounding the 84 mark, whether it refers to a dotted-quarter or dotted-half, even going so far as to say the dotted-half=84 for the fugue is “basically right” … to me that would sound lightning fast, and the following restatement of “Freude” absurd, at that tempo.

    Which brings me to my main point: after the Turkish March at the beginning of the fugue Beethoven writes “sempre l’istesso tempo”, which I believe is the key to this whole issue. From the March through the Fugue to the restatement, the meter (and therefore the basic unit of pulse) remains at 6/8, and Beethoven encourages and reminds us to maintain that tempo (and basic unit of pulse) with his abovementioned “l’istesso” instruction. So in my mind, the challenge for performers is to to find ONE TEMPO at the same unit of pulse to fit all three sections. There can certainly be some variation, but all three sections should be basically the same speed, and in my opinion, neither dotted-quarter or dotted-half at 84 can fit the bill.

    However, dotted-quarter=126, as you stated at the beginning, certainly does work for all 3 sections. I don’t know the history behind the editions that led to the number 84 “getting in”, but to me, the “sempre l’istesso tempo” with no meter change through all three sections is a much more important factor in making sense of this music than the number itself.

Leave a Reply

Your email address will not be published. Required fields are marked *