Originally published at: https://boingboing.net/2021/01/01/the-mystery-of-beethovens-metronome.html
…
4 Likes
OK, this has to be a logical fallacy of some kind:
This slowing down follows, on average, a systematic deviation, so it is not random, but conductors tend to play consistently below Beethoven’s marks.
It will obviously be biased towards slower (relatively) since it is considered fast (absolutely). In other words, if Beethoven wanted his music to be played fast indeed, but everyone else thinks it’s too fast, there’s literally no chance that anybody would ever play it too fast.
2 Likes
Or, as AuthenticSound’s Wim Winters has been advocating for long time, in Beethoven’s era they counted the whole beat instead of half beat as we count today.
I seriously recommend the channel, because even if you don’t agree with the “Whole beat theory” at least he’s not pulling it out his ass: he offers full documentation, cites sources and patiently answers most questions and criticism.
He’s currently rerecording lot’s of pieces in whole beat and I have been enjoying his music videos for a long time 
EDIT ADDENDA: Just for reference, in this video he talks about precisely why the “famously broken metronome” was not only not broken, but that modern musicians simply ignore the indications the same Beethoven left us:
6 Likes
half beat … whole beat …
The past is a foreign country - they do things differently there.
2 Likes
“Consistently” means that orchestras all turn the tempo down about the same percentage, no matter what piece is played.
3 Likes
the link to the actual article is buried…
4 Likes
I find this interesting. This tells us that Beethoven was using state of the art music technology back then. this shows that he was somebody embracing technological progess in the arts. In our time he would probably use modern computer technology to compose music.
4 Likes
This makes more sense than any other explanation I have met. They still have his metronome, and anyone who has ever tested it says it is pretty accurate. The metronome in the picture looks similar to the one my grandmother had. If you stick it on 120 beats per minute, the visible beat is 60 beats per minute with an audible click on the left and right extremes. If you were deaf you might look at the metronome arm as you might a conductor’s baton. The metronome works, but Beethoven was using it wrong because he was deaf.
It is hard to be sure what Beethoven wanted but his figures are madly fast as given, and dividing them by two gets us to a sensible, if sightly slow speed by modern tastes, and compatible with other composers of his era. If the opening of the 5th Symphony is a knock on the door, then the half-speed version is an ominous thumping from a policeman or a bailiff (I think this is great), while the full-speed one just wants to borrow your toilet.
9 Likes
Well, according to the WB teorists, is not that he used it wrong, but that they counted it differently than we do it today. As it was a new invention it took a while until they standardized how to count it.
Beethoven was not the only one to publish in whole beat but also Haydn, and Czerny (which was a Beethoven pupil so makes sense he used the same method). The latter is quite important because he helped republish several Bach and Mozart compositions with tempo anotations, which we use now as reference.
5 Likes
Thanks for the follow-up.
Modern notation uses an equation such as [crotchet] = 120, so there is no ambiguity in what ‘beat’ we are counting.
I saw the reference to the alternative definition of a beat. This, as I understand it, is like counting heartbeats (Lup-dup) or breathing (in-out) as two rather than one one events. This was a new idea to me, but I buy it.
There are two things here: a double definition of what a ‘beat’, and the visual ambiguity of Maelzel’s metronome, which ‘conducts’ at half the click rate. Beethoven knew what 120 clicks a minute was - but if his beat is one and two and three rather than one two three then his numbers make sense. Beethoven knew Maelzel, and was a very early user of the metronome.
It’s not proof, but I am convinced.
5 Likes
Yeah, I have to say I’m no expert in music but subscribed to the channel because his performances are quite good.
I know that in the video I posted in the first reply he sounds a bit like a conspiracy theorist but If you’re interested I recommend going to older videos where he was not as convinced and discusses music publications of that time in context; it’s quite a fascinating trip to the past.
2 Likes
Sure. I don’t see how that answers my point, though.
Let me give an example: I open a clothing store. I decide all items I sell will be burning neon pink. It’s most likely many potential customers will tell me “I’d buy in your store if only you’d also sell something a bit more sombre. How about some nice purple?” It’s very unlikely that someone comes in and says instead: “I’d buy from you if only your burning neon pink was even more burning and neon and pink!”.
If Beethoven had purposefully decided to be on the extreme of the tempo scale, he’d certainly get consistently slower interpretations.
But the point here is that evidence suggests a specific offset; if we were talking about a general preference for slower tempi, you would expect that to produce a whole range of (normally-distributed) coefficients. Some people would prefer 90% speed, fewer would prefer 80%, a very few would prefer 50% etc.
The argument, as I understand it, is that Beethoven’s written tempi make more sense if you just subtract a certain constant from whatever number he wrote (as opposed to multiplying by a certain percentage), which would be consistent with him reading the metronome scale from the wrong mark, i.e. not the standard mark.
Of course there’s no way to be sure, which is why standards are so important in metrology. But if you know some standard was used, you can make strong inferences about its value from math alone. Architectural historians do a lot of this to figure out ancient units of length. The rooms in a Prussian castle won’t all measure an exact number of feet because there will be offsets for half brick widths and wall linings and so on, but if you measure a number of different-sized rooms of similar construction, you can see if they’re all multiples of the same number plus-or-minus a constant, and thus infer the length of a Prussian foot, even where no explicit reference survives.
6 Likes
“Mälzel went on to invent more automatons, like the famous Mechanical Turk who played chess . . .”
C’mon, Smithsonian. ICYMI, The Mechanical Turk wasn’t an automaton at all. It was a gaff.
5 Likes
That is a very compelling idea (at least to me, who has zero knowledge of musical history). Maybe it’s YouTube trash that has already been debunked for all I know.
If true, it would be the musical equivalent of how old filmstrips all look frantic because the frame rate of the recording was slightly lower than today.
4 Likes
That’s what it does mean, grammatically. But the writer surely meant to say “consistently play” (not “play consistently”).
Your interpretation would suggest that they all play the same ‘amount’ slower than Beethoven intended - which is not the case. I’ve heard the same Beethoven piece played at different tempos by different orchestras - probably all slower than Beethoven intended.
2 Likes
He seems rather passionate.
Anecdote alert: years ago I was taught to measure distances by calibrating my pace. Everyone is a little different, so it’s a knack you have to learn by measurement and repitition, but once you get it, you do remember it, including your own offsets for uphill, downhill rough ground, soft ground, pack on, pack off, etc. It’s not super accurate - +/-5% or thereabouts - but good enough for up to abot 200m in detail, and to get a general sense of kms travelled.
Because its a personal thing, you tend to do it in your head and with counting beads in your pocket, so it wasn’t till some years later that I discovered I was doing the technique differently to most of my peers. My calibration was - is - 120 paces (unladen, flat ground, good going) for 100m. But that’s counting both left and right steps. Most folks only count one or the other, so their numbers tend to be in the range of 50-60 paces.
Years later still, even knowing I’m doing it “wrong” I haven’t been able to unlearn my first formed habits, and still double count. Which I suppose, makes me a Beethoven of walking 
10 Likes
I remember this moment in time! At the beginning, most guys who came to discuss Winter’s opinions on the channel were from an academic perspective, so it was one of those unusual channels where you could enjoy the discussions.
The moment he got popular (around the beginning of 2019, when he finally got the new pianoforte and started recording pieces), the classic internet trollies started to appear. So basically for each scholar trying to “hey, but have you read on cherny’s 4th reedition of blablabla” (you get the idea), we had two or three guys saying “this is bullshit and you’re an idiot”. So eventually started moderating the content because the trollies were kicking away the scholars.
Just look at the tone of the reddit thread (which, summarizes as “winters is wrong because he is”) and imagine that as a reply on a mostly academical video, nevermind he’s recording a lot of period pieces in WB (I’ll be honest here, I follow his channel for the music, and I leave the discussions to scholars), not “cherrypicking” the fastests ones precisely.
(for example, this recording of the moonlight sonata, which is not an specially fast piece)
3 Likes
