The Dice Game That Lets Anyone Be A Composer

The Dice Game That Lets Anyone Be A Composer

this video is sponsored by Dashlane! hey, welcome to 12tone! in the late 18th century,
a new craze began to sweep across Europe, a revolutionary toy that promised, with a
few simple rules, to turn anyone into a composer. it required no songwriting experience, no
practice, no talent whatsoever: all you needed was a pair of dice. this is the story of the
rise and fall of the Musikalisches Würfelspiel, or musical dice game, and it’s a much stranger
story than you might expect. (tick, tick, tick, tick, tock) so what were these dice games, and how did
they work? well, to understand that, we have to understand
a bit about the world they were born into. the 18th century was the Enlightenment, the
Age of Reason, and it was marked by a deep, philosophical belief that all things could
be explained with science and mathematics. we can see this, for instance, in the work
of renowned music theorist Jean Phillipe Rameau, whose Traité De L’Harmonie attempted to lay
down the foundation for a purely rational conception of musical structure. more broadly, there was a huge amount of public
interest in mathematical ideas, with legendary figures like Leonhard Euler just, like, inventing
new kinds of math all over the place. Euler himself even dabbled in music, because
of course he did: we talked about his consonance formula a while back, and I’m sure I’ll talk
about him again at some point. the other important thing about the 18th century,
though, is that musically speaking, it was the Classical Period. this is a bit confusing
because we often use the term “Classical music” to describe pretty much anything Europeans
made before around the 1920s and a lot of stuff they made afterwards too, but historically
speaking, the Classical Period was the middle section of the Common Practice Era, starting
around the mid 1700s and running until the early 1800s. music of the Classical Period tended to be
cleaner and less complex than Baroque music before it or Romantic music after, relying
more on clear melodies, strong harmonic structures, and strict musical forms. some composers you might have heard of include
Haydn, Salieri, and Mozart, to give you an idea of what the music sounded like. this is the cultural backdrop into which Johann
Philipp Kirnberger released the first known musical dice game. Kirnberger had been a student of Johann Sebastian
Bach, and according to some sources he may be a large part of why you know that name:
Kirnberger believed that Bach was one of the greatest composers of all time, and spent
much of his life trying to keep his legacy alive. but that’s not relevant to our story. what we care about is that, in 1757, Kirnberger
released Der allezeitfertige Menuetten und Polonoisenkomponist, which according to my
German-speaking friend Sheryl from the Roving Naturalist, link to her channel in the description,
roughly translates to something like The Ever-Ready Minuet and Polonaise Composer. this was effectively a system that, if followed
correctly, would allow you to compose your own short polonaise or minuet and trio, which
were two popular styles of the time, with nothing more than some basic knowledge of
musical notation and a couple dice. the rules worked like this: you’d start by rolling 2
dice. then you’d take the total value and check it on the number table for your intended
form: let’s say we’re writing a polonaise. we rolled a 7, so we check the 7s column on
the Polonaise chart, which points us to measure 72. then we go to the list of musical bars,
find number 72, and write it down. that’s bar 1. then we do the same thing but
this time we check our result in the row for bar 2, and we keep going until we’ve got 6
bars for our first section, after which we do the same thing for the second section on
another 8-bar chart. then we can play it and, because there’s about
379 trillion possible results here, we can be fairly confident that no one has ever written
the exact Polonaise we just did. the Minuet and Trio worked roughly the same
way, although they only used one die per roll, but since it produced a longer piece, it also
has an absurd number of possibilities. musicologist Leonard Ratner even observed
that the entire population of Europe at the time could have spent their whole lives playing
Kirnberger’s game and still not come close to writing all the possible pieces. so how did they sound? well… pretty good. not great, mind you, but perfectly fine. no matter what you did, however well or poorly
you rolled, you wound up with a decent, enjoyable composition, and releasing something like
that to a crowd primed to be interested in weird, mathematical phenomena was bound to
be a hit. the idea that, through what seemed like pure
chance, you could be guaranteed to produce a piece of music that not only made sense
but sounded good was fascinating, and while I can’t find any record of sales figures so
I don’t know if it was a commercial success, the wave of musical dice games that followed
indicate that, at the very least, it had a significant cultural impact. the next dice game, or at least the next one
we still know about, was pretty different, though. it was called Ludus Melothedicus, which according
to my Latin-speaking friend Mark from The Endless Knot, link to his channel in the description,
roughly translates as, uh… we’re not sure. Ludus is game, but Melothedicus is not actually
a Latin word. in fact, Latin doesn’t even have a “th” sound,
so it might be borrowing elements from Greek, possibly the word “Methodos”, meaning method,
which would make this the Game of Melody Methods. given the context, that makes a fair amount
of sense, but we couldn’t quite figure it out for sure so if you have another guess,
let me know in the comments. but the name is just the beginning of what
makes this game weird. for starters, unlike Kirnberger’s game, Ludus Melothedicus was
published anonymously. we know it came out in Paris around 1758,
a year after Kirnberger’s, and its instructions are written in French so we can guess that
it’s by a French composer, but beyond that we have no idea. it could be anyone. it could even be Kirnberger again, although
that seems… unlikely. but let’s get to the rules, because my French-speaking
sibling and I spent about an hour trying to decipher them and I don’t want that effort
to have been for nothing. the reason this was harder than with Kirnberger’s
is because Ludus doesn’t just give you complete bars: it generates each note individually. how? well, it starts the same way: you roll
2 dice and add them together. then, if the result is bigger than 9, you
throw it away and reroll, because whoever made this was too lazy to write all the possibilities. once you have a number, you take this grid,
start on the outside of the first line with the number you rolled, then count up to 9
from there, so if we rolled a 4, this first box would be 5, then 6, 7, 8, and 9. wherever the 9 lands gives you your first
note, which you go look up on the note chart. in this case, it’s a quarter note D in the
bass clef. then, from there, you count up to 9 along the boxes again, but, and this
is very important, every other line goes backwards. why? ’cause why not, ya know? there’s even a drawing of a snake in the rules
to help explain it. anyway, you keep going like this, marking
down every 9th number, until you get to the end, which gives you a complete bar of music. also, sometimes there are 0s in the grid,
which you just ignore. oh, and the grid’s not completely filled out.
and you’re collecting notes for two separate parts simultaneously, one in each clef. and
once you finish this box you have to do it all over again 15 more times. but hey, it writes a minuet! now, if you’re thinking to yourself “wait,
if after the first result you just take every 9th number, aren’t there only 9 possible options
for each measure?” then yes, you’re absolutely correct, and if you’re now thinking “in that
case, wouldn’t this be a lot simpler if they just, you know, gave you the 9 options instead
of making you do a bunch of weird calculations with unintuitive counting rules?” then yes,
you’re absolutely correct, but we can’t do that, because an anonymous French dude apparently
thought it sounded too easy, so instead we get this drawing of a snake that’s supposed
to help somehow. but hey. it writes a minuet. after Ludus, a bunch of other dice games were
published, including one by Bach’s son Carl Philipp Emanuel Bach called Einfall, or Incidence. these new games mostly used either Kirnberger’s
or Ludus’s methods, and they wound up being a pretty popular pastime, which brings me
to the controversy section, because if you didn’t think this story about fun little dice
games that let Enlightenment nerds pretend they were composers would have a controversy
section then you have vastly underestimated how weird history is. there’s two main stories here, and I’ll start
with the simpler one. think of it as like a narrative appetizer. in 1793, well into the musical dice games
craze, a publisher in Naples named Marescalchi released Gioco Filharmonico, or the Game of
Harmony, written by none other than famed Austrian composer Joseph Haydn. or, at least,
so Marescalchi claimed, but there’s a bit of a problem. you see, 13 years earlier, in 1780, a composer
named Maximilian Stadler had published a game called Table Pour Composer, or Table For Composing,
which was preeeetty similar to the one allegedly written by Haydn. not only did it use the exact same number
table, but the parts were almost identical: the only difference was that Haydn’s had been
reorchestrated for a different set of instruments. now, it is possible that Haydn plagiarized
Stadler’s game, but more likely is that Marescalchi did the reorchestration himself and then published
it under the name of one of the great masters in order to boost sales. there’s no hard evidence that Haydn wasn’t
involved, but most scholars seem to agree that it’s unlikely. more interesting, though, are the Mozart games. there’s two of these, both published after
his death. the first, initially cited as being published
in 1796 by Nikolaus Simrock, wrote contrandances, and the second, in 1800 by Johann Goetz, wrote
waltzes. both claimed to be posthumous publications
written by Mozart. so what’s the problem? well, for starters,
the two are extremely similar: the instructions are both written in the same four languages,
and the German versions are copied word-for-word except for one additional sentence in Goetz’s
version about song form. Goetz had previously been sued for plagiarism
in an unrelated case, so this should already raise a few eyebrows. but the next point is that, in fact, neither
of these are the original publications: both versions were first printed in 1793 by Johann
Hummel, and Simrock also published the waltz game before Goetz did, which means the most
likely scenario is that Goetz stole his version from Simrock, who had already stolen it from
Hummel, which makes this a rare case of double-plagiarism or, as some youtubers call it, fair use. (rimshot) but there’s even Hummel’s version is fishy.
for starters, it was also posthumous: Mozart died in 1791, 2 years before the first of
his alleged dice games was published, so he wouldn’t have been around to contest any claim
of authorship. but more importantly, the number tables in
both games look… familiar. in fact, they’re the exact same tables from
our old pal Maximilian Stadler. again. unlike Haydn’s version, the music here is
different, so it’s possible that Mozart just didn’t feel like making his own table and
decided to work off an existing one instead, but this is another red flag. sadly, there’s no definitive answer here:
it’s possible Hummel wrote them himself or had a composer friend do it and then slapped
Mozart’s name on them just like Marescalchi did, in order to drive sales. on the other hand, Mozart did enjoy musical
games, and given that much of his career overlapped with the dice games craze, it’s not hard to
believe that he might have made his own. we don’t know for sure, and we probably never
will, so while there’s definitely an asterisk next to his name, these are still generally
viewed as Mozart’s dice games. from there, musical dice games started to
die out: the last one that we can definitively date was Giovanni Catrufo’s Bareme Musical
in 1811, but as the 19th century progressed, they pretty much disappeared. no one was interested in them anymore, probably
in part because music had moved on: we’d transitioned from the Classical Period, with its formal
structures and clear, iconic melodies, to the Romantic Period, which cared more about
emotional expression, elaborate harmonies, and new song forms with longer, more complex
phrases. the music just wasn’t predictable enough to
leave to chance, so chance went away. but that doesn’t mean these are gone forever. you can still make your own. it’s not even that hard, because it turns
out that, in truth, luck had very little to do with making a good dice game. despite what it looks like you aren’t actually
just throwing random bars together to see what happens, and you don’t have to try trillions
of combinations to make sure they all work, either. no, the craft of making a good dice game relies
entirely on having a strong sense of musical and melodic structure. let’s say we wanted a short, four-bar game. the first thing we’d do is pick a chord progression. (bang) that’ll do. next, we just start in the first bar and write
a melodic fragment that a) feels like the start of a phrase, and b) outlines an E major
triad. (bang) sure. then we do that for the other three bars,
making sure each one fits with the intended chord and its place in the phrase, and bang:
(bang) we’ve got our first variation. from there we just do that 10 more times,
making sure the phrases in each bar all start and end at roughly the same places so we don’t
wind up with awkward leaps, and then break these up by bar onto a number chart. oh, and for appearances we’ll jumble them
around a bit so it looks more like a random collection of musical fragments. and that’s
it! we made a dice game! or, I made a dice game, but you can too! it’s a fun exercise,
and you can do it for as many instruments as you want, in whatever style you want. if you do make one, I’d love to check it out,
so send me a note on twitter and I’ll try to share the ones I get. anyway, that’s how you use randomness to make
art, but random numbers also play a pretty important role in security, and Dashlane’s
a great example of that. staying safe online is more important than
ever these days, but one of the biggest vulnerabilities for most people is just having bad passwords. we all know we’re supposed to use good, long,
complicated passwords, but they can be hard to remember, so using a password manager like
Dashlane that can not only generate incredibly strong, unique passwords for every site but
also remember those passwords so you don’t have to is really valuable. plus, it also works as a VPN to keep your
internet browsing safe, anonymous, and private so no one can track you. and it monitors the
deep web for data breaches so if your password does get leaked it’ll alert you and you can
change it before anything bad happens. if you’d like to try Dashlane, you can try
it free for 30 days with the link in the description, and if you like it and want to keep going,
don’t forget to use the coupon code 12tone for an additional 10% off! and hey, thanks for watching, thanks to our
Patreon patrons for making these videos possible, and extra special thanks to this video’s Featured
Patron, Duck. if you want to help out, and get some sweet perks like sneak peeks of upcoming
episodes, there’s a link to our Patreon on screen now. you can also join our mailing list to find
out about new episodes, like, share, comment, subscribe, and above all, keep on rockin’.

Comments (85)

  1. Some additional thoughts/corrections:

    1) I have a merch store now! Check it out at!

    2) If you want to play around with existing games, has a lot of the famous ones, including Kirnberger's and Mozart's. If you'd like to download mine, it's at but full disclosure Mozart's is better.

  2. This channel is why I'm glad I don't have a class at 10:00 AM local time on Friday.

  3. First thing I thought of was Common Practice Era Bach….

  4. Hi dear, awesome and interesting as usual 🙂 the word Einfall means idea though, not incident

  5. Very interesting. Another thing to try

  6. So, were these the first examples of procedural music generation?

  7. Your pronunciation of those not-quite-sure-what-country-they-
    are names is just amazing!

  8. Literally just realized he writes with his left hand, GO SOUTHPAWS!

  9. 6:03 I get the feeling you and your sibling might have been a little miffed at this anonymous French dude, lol

    [furiously circles snake]

  10. But hey, it writes a minuet.

  11. 0:35 the age of Propellerheads 😀 (the music software studio formerly known as)

  12. Did u rolled a d20? Kkkk

  13. Surprised there was no mention of creating a tone row with dice as showing that dice didn't fall away completely.

  14. Fair use badumm tss… .

  15. I've been intrigued by musical dice games for years, but had a difficult time finding anything online that didn't do it for me (e.g. via RNG on a web page). I was only aware of Mozart's games, so this was a great dive for me. On the subject of Euler, I had heard that he did dabble in music and found he wasn't too good at it. There's a book by Eli Maor called "Music by the Numbers" which talks about it. I have to admit that I started the book and then got distracted by another squirrel, so I never finished it…

  16. Considering how some self-entitled "Producers" throw together loops and call it a day, we already have a modernized version of the musikalisches Würfelspiel; ohne Würfel…:)

  17. Melethodius is propably coming from the greeck word μελωδία (melodía) that means melody .

  18. Yay for the Reason logo at 0:37

  19. Btw when I read the title of this video I thought it was about assigning scale degrees to numbers and then using those scale degrees the dice suggested for a melody. The truth is a lot more colorful so to speak 😉 and I was wondering if that is again the general mood of the era at play here. I see some kids who are devouring Jules Verne and thinking of some crazy complicated contraptions right in front of my mental eye 😀

  20. I love that you draw Battletoads whenever you says "hard". It is a great definition of "difficult", indeed.

  21. This reminds me of my favorite aleatory compositions:
    It's a D&D podcast where the players write a song as part of the story each episode by rolling dice to choose chords and kits.

  22. 1:48 i 'member nethack

  23. Someone should make this into an AI program.

  24. I wonder if they knew that 7 was the most common roll as a sum of two die, and put their favorite there

  25. It truly is unbelievable how much Euler has had an effect on

  26. Feel like I'm gonna get weirdly a lot of use out of "but hey, it writes a minuet."

  27. Didn't John Cage experiment with randomly generated music, like using irregularities in the texture of the paper he was writing on to determine which note to use? Serialism seems like a return to the strict forms of classical music (different forms though) so dice driven composition could have had a resurgence, if only Serialism caught on with the general public. Also, Beethoven not Classical? Not romantic enough to be Romantic, too romantic to be Classical, so transitional?

  28. Can we just take a moment to appreciate the elegance, sleekness, and beauty in your logo?
    Its. So. Good.

  29. Great video! I really love all of your videos and it’s been cool to see you progress in your video quality. Even your early vids were good but now there absolutely amazing! Nice job

  30. The word Melothedicus would sort of make sense as a Latinized Greek compound of melos (a musical 'strain'/song/melody) and the- from 'to put/set/arrange'. What does not make sense is the letter d. dicus would rather seem to point to the stem deik (to show) or Latin (and etymologically related) dic- (to say). On the other hand -icus clearly points to the Greek adjectival suffix -ikos. If the d were actually a t, it would make more sense. Ludus Melotheticus would mean something like "game of melody arrangement". But that's not our word…

  31. So…. musical dice game on PC and mobile when? 🙂

  32. Michael New sent me a set of these a while back: , very cool take on the dice/random music assistance thing. I use them when I get stuck or when I just want to start something different. Not sure what happened to his channel, always like his vids.

  33. You probably should have mentioned that Hummel was a student of Mozart (and according to Mozart, his best student). So while that's not proof of anything, if anyone had an authentic Mozart dice game, it would have been him.

  34. Is "Johann Hummel" here the same as "Johann Nepomuk Hummel"? Student of Mozart and composer of a well-known trumpet concerto?

  35. This is an amazing concept and I love that it exists…and it causes FIGHTS.

  36. So if I roll a natural 20, does that mean I’m like the Steve Vai of this game?

  37. I see what you did there. "You feel that eating the kitten was a bad idea."

  38. The idea of using the dice to get a random value between 1 and 12 kinda falls apart when you learn about standard deviation and the proportion of 7’s to every other number

  39. I have no talent AND I have a pair of dice! Roll over Beethoven and tell Tchaikovsky the news. P.S. Kirnberger was correct. BACH was one of the greats!

  40. Loved this video and LOVE your channel!! However, one small point of disagreement: the peerless brilliance of JS Bach is ALWAYS relevant. 🙂

  41. Ludicrous Melodies???

  42. There ought to be a generator website that follows these rules and generates a piece of music for you. Having to keep track of all these numbers sounds like a bit of a headache.

  43. I was already laughing at the absurdly tedious minuet composition process, and your frustration about the anonymous French dude's drawing of a snake that's supposed to help somehow provided the best chortle this week by far! Effectively eliminating any of the joy of artistic creation and replacing it with total clerical tedium didn't catch on for some reason. Now we have AI to do that, right?

  44. 6:28: You're so angry🤣🤣🤣

  45. My favorite thing will always be your slightly obscure drawings for certain words. Like your atom for "dating," referencing carbon dating.

  46. I bet an army of monkeys could have written a musical dice game for the Romantic period XD

  47. How many takes until you got that nat 1 at 3:34?

  48. 3:34 how many tries did that take you? 😛

  49. Very interesting video!

  50. Put "But hey it writes a minuet" on a shirt please!

  51. Excellent video! I can offer my two cents as a dance historian. All the examples of these games are for popular dances of the era. Most of these dances were danced for quite a long time like half an hour or more. And while the dancers are having fun, musicians usually get somewhat bored repeating the same 16-32-whatever bars for more than a few minutes. So it’s not inconceivable that the musicians varied the music somewhat during the dance. But these variations are to be done within very ordered and predictable structure of the dance music as dancers rely on it. This is where these interchangeable musical parts may be from. They are there to add variety to some 30 minutes minuets or polonaises. For the sake of musicians and their audience. And while such long and repetitive dances are not unique to this period by any means, such games seem to be limited to the classical music only.
    N. B. This is my educated guess, not definite opinion.

  52. That German name sounded VERY Russian, the way you said it ^^

  53. Today would be a great day to release a video on Earth, Wind & Fire’s September. Maybe next year

  54. "Euler himself even dabbled in music, because, of course he did." Haha love it.

  55. Hi, I can not find any trace of Johann Goetz musical generator. Could you, please, help me to find it?

  56. Maybe I missed this, but who are the scholars you're thinking of when you talk about the authenticity of the "Haydn" dice game?

  57. In light of this, what do you think of A.I. composition assistants? The concept seems similar, except that there are theoretically infinite possibilities (or at least close to infinite), and the artist has more control of exactly what comes out.

  58. 18th century: musical dice

    21st century: Guitar Hero

    The musical gaming scene has evolved a lot over 300 years…

  59. We need an app for this!

  60. Can you imagine what would happen if someone programmed an AI to write music using these formulas?

  61. i love your sassiness

  62. Can you do an analysis of Firth of Fifth by Genesis

  63. All of that and you don’t even play the minuet? What’s the deal?

  64. As of about 20 years ago, someone was still publishing a mozart dice game…. I don't know where mine ended up, but I had fun with it. Thanks for bringing back good memories.

  65. Here's an article published last year on AI producing music. It's over a year old and aimed at potential investors, so may be a bit out of date now, but I guess this is the 21st century version of the dice game

  66. I don't think you circled that snake enough.

  67. The fair use pun was great.

  68. 1:50 ASCII rogue-like ?

  69. Can you analyze Panic Attack by Dream Theater?

  70. 0:36 Is that… is that the Propellerhead Reason logo

  71. SInce nobody has done it yet… "Game of Tones"

  72. Sometime in the far future, some archeologist is going to find these sheets and mistake it for some kind of weird hieroglyphics or something.

  73. So what about Crazy Bus

  74. The segue way to the ad bahahaha

  75. Maybe the snake meant you were supposed to go down the table in a back-and-forth pattern as the shape of the snake suggests and the inconsistent blank spaces on the right sides of the table helped "randomize" the 9 count?

  76. I love how angry you are, this is one of your funniest videos yet 😂😂. But hey, it writes a minuet

  77. A dice game except it’s always Megalovania

  78. I love the nethack reference.

  79. but what do they sound like? vid?

  80. I’d say that the “melothedicus” is a latinised Greek-inspired neologism which can be best interpreted as “melotheticus” from melos (song, melody, music) and thesis (position, setting); so literally meaning the setting into music.

  81. "This dice game really makes you FEEL like a composer"

  82. I've created my own system of composing by means of a deck of playing cards! Aces are the root note, numbers are steps from the root, and K is a rest. J & Q are the interesting ones though. Normal cards are quarter-notes, but J & Q = 1/2 length. So if you pull a J, the next 2 cards immediately following it will be eighth-notes. And this stacks: so 2 J/Q = the following 4 cards as sixteenth-notes! There's some other rules as well, but I think it's a pretty neat setup I've come up with, and I've actually gotten some interesting chunks of melodies out of this before!

Comment here