Ludwig van Beethoven (1770-1827)

[calyptorhynchus]

I've always been taken by this theory about the metronome markings

http://www.hansgal.com/storage/writings/tempo.pdf

[Gurn Blanston]

Well, most entertaining... and most intriguing, because he draws on an immense knowledge of the political and economic realities of the time. And he's a darn good writer. His Mozart book is much better, still... but/and it's also a treatise on the political history of Austria / Europe at the time -- and how that influenced everything that went on in the word of culture. It's a hyperbole, claiming it would be "the best" book, of course, but it's very insightful... except it's a little hard to separate fact from assumption, which annoys the German within me.

What is this book you are talking about, Jens? Is it in English, for one thing? And possibly available in USA? Political history of Austria in Mozart's time is something I must have, although I don't care a damn about Beethoven's metronome. I always thought Maelzel made Beethoven's metronome wrong just to mess with him over the Panharmonicon / Battle of Wellington incident... 

[Wakefield]

What is this book you are talking about, Jens? Is it in English, for one thing? And possibly available in USA? Political history of Austria in Mozart's time is something I must have, although I don't care a damn about Beethoven's metronome. I always thought Maelzel made Beethoven's metronome wrong just to mess with him over the Panharmonicon / Battle of Wellington incident... 

It sounds compelling, indeed.

I think Jens is talking about Harke de Roos, Gurn.

These are his books: http://www.katharos-verlag.de/buch.html. Apparently, they haven't been translated into English.

[asin]B00CHBD2BY[/asin]

[(poco) Sforzando]

It sounds compelling, indeed.

I think Jens is talking about Harke de Roos, Gurn.

These are his books: http://www.katharos-verlag.de/buch.html. Apparently, they haven't been translated into English.

[asin]B00CHBD2BY[/asin]

How entertaining. Harke has other theories besides:

Die Dritte Symphonie, die ,,Eroica", wurde zum Andenken Mozarts komponiert.
Beethovens ,,Neffe" war in Wirklichkeit sein leiblicher Sohn.

The Eroica was composed in memory of Mozart.
Beethoven's "nephew" was really his own biological child.

[Gurn Blanston]

It sounds compelling, indeed.

I think Jens is talking about Harke de Roos, Gurn.

These are his books: http://www.katharos-verlag.de/buch.html. Apparently, they haven't been translated into English.

[asin]B00CHBD2BY[/asin]

Ah, well I would need to have Harry come and read it to me, I guess...    Thanks, Gordo.

Oh, good luck tonight at the football. If the rain ever stops. Chile v Argentina should be entertaining.

[Wakefield]

Oh, good luck tonight at the football. If the rain ever stops. Chile v Argentina should be entertaining.

Thanks! It was a rainy night in Chicago.

Now the final vs. Argentina. 

[Jo498]

Norrington attempted to do just that in his LCP recordings. The 2nd Symphony in particular, right at Beethoven's speeds, just zips along bubbly and lightly, and is for me the outstanding performance on the set. The fact that Harke can't play the violin runs cleanly on his piano only shows that Harke is not a very good pianist.

For the outer movements there are very fast interpretations of the 2nd by e.g. Toscanini and Reiner from the 1940s. The first movement has its shapes and phrases mostly in half and whole notes, so the tempo of 100 for half notes is only crazily fast if one thinks of it as 200 for quarters. Sure, a few flourishes are in very fast notes, but they are flourishes, turns etc. not real melodic/motivic shapes.

I don't have more time right now, but the two biggest metronome problems IMO are in the 9th, the trio to the scherzo and the tenor solo. And possibly the first movement to the Hammerklavier.

Some in the 9th have long been recognized as probable errors. The tenor solo might be a very fast dotted half=84 (which ist often reached in the instrumental fugato following the march) instead of the funereal dotted quarter=84) and the trio is also probably a copying mistake. Again, one reading, bar =116 (as before, only 2/2 instead of 3/4) seems too fast, the other one half note=116 too slow and neither makes a lot of sense with the accelerando from the 3/4 main section. Because accelerando to either the same or exactly double speed does not make a lot of sense. I once read that this could have been a dictation or copy error and Beethoven meant 160 for the half notes.

performance. Tchaikovsky is another composer whose metronome markings are routinely ignored. Set your metronome at 104 and tell me if you normally hear the first movement of #5 that fast.

This is fast. It's as fast as the (often played but sometimes a little more relaxed) 6/8 first movement of Beethoven's 7th. Which is "traditionally" a considerably more "dancing" and vivacious movement than the darker one in Tchaik 5th

I had not been aware of the fact until recently but e.g. Berlioz also uses Metronome markings in many works and the few I looked at (Fantastique and Harold) are also on the fast side and roughly consistent with what Beethoven considered fast, e.g. 150 for a fast allegro con brio or so.
(Schumann is apparently a somewhat problematic composer in this respects with diverse markings in different edition, the suspicion of Clara fiddling with some of them later on, but there are apparently also a few deemed way too fast or too slow and hardly ever played in practice.)

Historically, I think an important point is also that there are dances like the waltz that have a continuous tradition since Beethoven's time. And a viennese waltz is usually around 60 3/4-bars/minute. This agrees with metronome markings for waltz-like movements from early/mid 19th century works, e.g. "un bal" from the Fantastique has 63, I think.
Of course such parallels do not give exact tempi. But they would certainly show if the tempo markings were actually 50 or 100% off the intended tempo.

[Jo498]

Regarding the Metronome marks, it seems to me it would be worth while if someone would adhere to them strictly in a recording so we can judge for ourselves what we think of them.

There are by now many recordings of the symphonies with tempi at or very close to the metronome markings. Norrington has been mentioned (the first issues had the MM numbers printed on the backcover after each movement) but there are some problems in the 9th.
On modern instruments several of Scherchen's, Leibowitz', later Gielen's and Zinman's symphony recordings are also close. (Zinman is actually too fast in most of the scherzi I have heard... probably because those would seem too slow otherwise with his zippy first movements)

For the string quartets, try the Smithsonian for op.18 and the Emerson for the middle period works (including the notorious finale of op.59/3) But while in these pieces the tempi are often relaxed a little compared to the markings there is hardly a tradition of playing many movements considerably slower (like e.g the first movements of the 3rd, 6th or 9th symphonies). (There are no original markings for the late quartets which is a pity, especially in movements with many tempo changes like the variations in op.127 and 131.)

For the notorious op. 106, amw listed somewhere fast recordings. The fugue is usually played pretty close to the mark, often a little slower (say 136 instead 144), same for the scherzo. Fast first movements are by Schnabel and Korstick. The fastest slow movement by a major well known pianist (and quite close to the marking) is by Gulda.

There are no original markings for the other piano sonatas or the concertos, but we have some by e.g. Beethoven's pupil Czerny and many of them are roughly followed not that rarely although some, especially slow movements are often also played considerably slower. Overall, Gulda/Amadeo might be the one closest to these tempi in the sonatas. Also Kovacevich's EMI recordings.

[jlaurson]

What is this book you are talking about, Jens? Is it in English, for one thing? And possibly available in USA? Political history of Austria in Mozart's time is something I must have, although I don't care a damn about Beethoven's metronome. I always thought Maelzel made Beethoven's metronome wrong just to mess with him over the Panharmonicon / Battle of Wellington incident... 

No, it's not in English yet... although he has been looking into the possibility of getting it translated. (I'd prefer his Mozart book to get translated, first, which I find works better as a book... and is totally fascinating reading. But I'd like to have endnotes included, to make sure everyone can separate fact from 'imagination', which isn't easily done at this point.) In any case, his theory (and mind you that any abbreviation of his theory will make it sound obviously less credible) is that Ludwig effed with the Metronome numbers to eff with Maelzel over the Panharmonicon / Battle of Wellington incident. Well, there's more to it, but that was the part that I bought or enjoyed contemplating. Beethoven trying to screw w/Maelzel to make his metronome look ridiculous... and people eventually taking the metronome markings seriously rather than noticing that some speeds absolutely don't work. (Well, people do still notice that, but simply do what they think is right. And it is not just too fast... Beethoven's markings get even more ignored where they strike us as too slow.

[Jo498]

Which ones are too slow according to de Roos? Except for the already mentioned probably erroneous ones in the 9th symphony.
(I think the ones for the 9th should ALL be considered with suspicion because of difficult relationships in the finale and generally deafness etc.)

To my knowledge most of the Scherzi are often taken slightly faster but only within a few points, e.g. 126 instead of 116 for bars in the 9th symphony scherzo. The finales of the 7th and 5th are often taken at the marked tempo or a little faster, say 80 instead of 72 for bars in the 7th finale. But all these deviations are "in the ballpark" and not at all comparable to the 30-40% slowing down in the first movement of the Eroica or op.106 or half the indicated tempo for some sections of the adagio of the 9th.

I think op.106,i is also problematic (too fast, but not by as much as typical tempi around 100-112 would indicate, about 120-126 seems reasonable).
I didn't have the time but I wanted to compare pianists in this and other similarly indicated (allegro alla breve) Beethoven movements, namely the first movements of op.2/1, op.31/2 and op.81a. I think op.106,i should be roughly in the same tempo as the last two (op.2/1 somewhat slower).
Because in addition to the same tempo indication these also have thematic shapes in 8th notes (that is these are not mainly/merely fast notes accompaniment) and a generally somewhat similar character (I think). My anecdotal impression/recollection is that most pianists play op.106,i (considerably) slower than e.g op.31/2i. Obvious for those with a slow and majestic op.106 like Gilels but it seems also true for fast ones like Gulda.

[calyptorhynchus]

"Beethoven trying to screw w/Maelzel to make his metronome look ridiculous... and people eventually taking the metronome markings seriously rather than noticing that some speeds absolutely don't work. "

Sounds like a very roundabout way of getting back at Maelzel, especially as the danger would be people would distort Beethoven's works in the process.

[(poco) Sforzando]

Some in the 9th have long been recognized as probable errors. The tenor solo might be a very fast dotted half=84 (which ist often reached in the instrumental fugato following the march) instead of the funereal dotted quarter=84) and the trio is also probably a copying mistake. Again, one reading, bar =116 (as before, only 2/2 instead of 3/4) seems too fast, the other one half note=116 too slow and neither makes a lot of sense with the accelerando from the 3/4 main section. Because accelerando to either the same or exactly double speed does not make a lot of sense. I once read that this could have been a dictation or copy error and Beethoven meant 160 for the half notes.

The best explanation I have heard for the trio (and I forget who made it) theorizes that the tempo should remain constant for the quarter note. This means that if the main tempo is dotted half = 116 (quarter = 348), the trio becomes half = 174, which is obviously peppier than half = 116 but not so manic as whole = that same mark. I'm not as disturbed by the accelerando as you; one can accelerate and then return to an original tempo.

As for the tenor solo, Norrington takes the funereal 84 and Gardiner goes to the other extreme with dotted half = 84. Both sound wrong to these ears, though Gardiner preserves the pace of the original theme and therefore doesn't have to accelerate wildly in the fugato with an unmarked Più mosso. Most conductors I've heard take a tempo between these two, but after the solo concludes it's off to the races.

[jlaurson]

"Beethoven trying to screw w/Maelzel to make his metronome look ridiculous... and people eventually taking the metronome markings seriously rather than noticing that some speeds absolutely don't work. "

Sounds like a very roundabout way of getting back at Maelzel, especially as the danger would be people would distort Beethoven's works in the process.

Indeed, it does. It gets convoluted from there on. I don't buy it, but I'm fascinated by it and he raises so many questions in the process that seem very legitimate.

Which ones are too slow according to de Roos?

Will try to get back to you on that.

[Gurn Blanston]

Indeed, it does. It gets convoluted from there on. I don't buy it, but I'm fascinated by it and he raises so many questions in the process that seem very legitimate.

Will try to get back to you on that.

Thanks for your reply above, Jens. Interesting. I have to admit that I had no idea that friction between Beethoven and Maelzel was even contemplated by anyone else, I sort of made it up on the spot as I was writing because the idea was funny to me and fit with Beethoven's personality, especially by the mid 18'teens.       Oh well, just goes to show, there is nothing new under the sun.



PS - I completely agree about the Mozart book needing translating first. I need stuff from the 1780's most of all. 

[jlaurson]

Thanks for your reply above, Jens. Interesting. I have to admit that I had no idea that friction between Beethoven and Maelzel was even contemplated by anyone else, I sort of made it up on the spot as I was writing because the idea was funny to me and fit with Beethoven's personality, especially by the mid 18'teens.       Oh well, just goes to show, there is nothing new under the sun.

Beethoven tried to sue the *#( out of Maelzel and he wrote about him, in no uncertain words, to others. Those are some of the neat things that de Roos does: He snoops in every letter or historical document about references to OTHER things... and then puts them in context. That's why in the end, even a prima facie "crackpot" theory like: "Beethoven messed up his own metronome markings on purpose" becomes... well, if not plausible so at least worthy for giving some thought to.

[Gurn Blanston]

Beethoven tried to sue the *#( out of Maelzel and he wrote about him, in no uncertain words, to others. Those are some of the neat things that de Roos does: He snoops in every letter or historical document about references to OTHER things... and then puts them in context. That's why in the end, even a prima facie "crackpot" theory like: "Beethoven messed up his own metronome markings on purpose" becomes... well, if not plausible so at least worthy for giving some thought to.

Ah yes, I have some vague memory of that. The only thing I really recall was a sort of copyright fight over Wellington's Victory. But snooping in historical documents often turns up the most interesting things imaginable, so more power to him. 

[Mandryka]

Having heard this recording, may I just say that it is without doubt the worst performance of the Second to ever reach my personal eardrums.

I listened to the first movement only and I thought it was very bad. But not primarily because of tempi but because of the stiffness, the plodding stiffness of it. Does he not believe in rubato in classical style or something?

[Jo498]

The best explanation I have heard for the trio (and I forget who made it) theorizes that the tempo should remain constant for the quarter note. This means that if the main tempo is dotted half = 116 (quarter = 348), the trio becomes half = 174, which is obviously peppier than half = 116 but not so manic as whole = that same mark. I'm not as disturbed by the accelerando as you; one can accelerate and then return to an original tempo.

This does make sense, certainly more than the extremes. Although I'd probably prefer an even more relaxed trio but I would have to check some recordings how different choices actually sound in practice.

As for the tenor solo, Norrington takes the funereal 84 and Gardiner goes to the other extreme with dotted half = 84. Both sound wrong to these ears, though Gardiner preserves the pace of the original theme and therefore doesn't have to accelerate wildly in the fugato with an unmarked Più mosso. Most conductors I've heard take a tempo between these two, but after the solo concludes it's off to the races.

I could imagine the deaf composer imagining that fast tempo for the instrumental fugato that concludes the section and not taking into account that this is too fast for the beginning. As we discussed elsewhere the 6/8 fugue section in op.133 is usually played (no original metronome marking) in ca. 80-84 for the dotted half which is not a problem for a string quartet. But they don't have to articulate sung text...

In any case I think the 9th as the last work for which MM markings were provided and when Beethoven was definitely deaf and maybe further away from practical considerations, he certainly could not test things in practice anymore should be to some extent "bracketed" for the metronome discussion.

Even in the first movement the marking, while playable is very fast in the 32nds passages and apparently in sketches or letters Beethoven corrected it to quarter note = 88 from 108 at which the 32nds would be a mere scramble, so I wonder if this is another mistake. It is fairly strange because quarter 88 is slow for an allegro ma non troppo but if the piece sounds de facto like in 8th notes 176 is more like presto.

[Karl Henning]

I could imagine the deaf composer imagining that fast tempo for the instrumental fugato that concludes the section and not taking into account that this is too fast for the beginning.

I realize it may be chronologically problematic to think of what LvB "would not have trouble with" based on my experience, centuries later . . . but in fact I should be surprised if LvB would need physical hearing to judge that sort of thing.  I cannot feel that LvB had less capable an "inner ear" than do I.

[Jo498]

There is ample documentation that composers corrected tempo prescriptions during rehearsals. From what I have read, almost always towards the slower side. I think even Rosen has a similar anecdote in one of his essays about a performance of a 20th century piece he participated in. They played in rehearsal and the composer said: that's too fast whereupon the musicians replied that they were already 20% slower than the composer's metronome marking...

And note that I don't even claim that his inner ear was faulty. It was basically right for the instrumental section (and the fact that a similar section in a piece without metronome markings is played usually at the fast reading of the controversial 84 seems to point to that as well) but not taking into account that a singer could hardly articulate his part at that frantic speed. I do not have Gardiner's recording and I am not sure if I have one that is as fast as his. But the section is usually played "in between" the controversial (literal: very slow and according to a probably error that confused dotted quarters and dotted halfs very fast) markings with considerable speeding up at/during the instrumental fugato.