Slashdot Mirror
Do Music and Language Obey the Same Rules?
Emre Sevinc writes "Ever felt as though a piece of music is speaking to you? You could be right: musical notes are strung together in the same patterns as words in a piece of literature, according to an Argentinian physicist. This article in Nature states that Damián H. Zanette's analysis also reveals a key difference between tonal compositions, which are written in a particular key, and atonal ones, which are not. This sheds light on why many people find it so hard to make sense of atonal works. In both written text and speech, the frequency with which different words are used follows a striking pattern. In the 1930s, American social scientist George Kingsley Zipf discovered that if he ranked words in literary texts according to the number of times they appeared, a word's rank was roughly proportional to the inverse of the its frequency squared. Herbert Simon later offered an explanation for this mathematical relationship. He argued that as a text progresses, it creates a meaningful context within which words that have been used already are more likely to appear than other, random words. For example, it is more likely that the rest of this article will contain the word 'music' than the word 'sausage'. Physicist Damian Zanette of the Balseiro Institute in Bariloche, Argentina, used this idea to test whether different types of music create a semantic context in a similar fashion."
I see how music could have some content in the way of emotion, and I guess that would count as a semantic composition, but whether individual phrases can translate to words, I'm not so sure about. Perhaps it has more to do with some sort of innate appeal to aesthetics, and as we listen to and formulate speech, it starts to conform to some aesthetic pattern. This isn't too far out. Some languages are considered more beautiful than others.
Many musicians already know this. Have you ever heard a soloist described as "lyrical"? (grep for "lyrical")
Have you ever heard a musician compare improvising a solo to "telling a story"(grep for "telling a story")
Ever heard a short musical idea described as a "phrase"?
Listening to a good jazz solo is a lot like listening to a conversation: There are main points, and there are variations on that point. It should be grounded but not to repetative
What is the soloist doing when he attempts to "build"? Actually the ideal process hardly ever takes place--that is, it is hardly ever the case that a conscientious soloist plays a thinking solo for a hard-listening hearer--but when this does happen, the key process is memory. The soloist has to establish for the listener what the important POINT, the motif if you like, is, and then show as much as he can of what it is that he sees in that motif, extending the relationships of it to the basic while never giving the feeling that he has forgotten it. In other words, I believe that it should be a basic principle to use repetition, rather than variety--but not too much. The listener is constatnly making predictions; actual infinitesimal predictions as to whether the next event will be a repetition of something, or something different. The player is constantly either confimring or denying these predictions in the listener's mind. As nearly as we can tell (Kraehenbuehl at Yale and I), the listener must come out right about 50% of the time--if he is too successful in predicting, he will be bored; if he is too unsuccessful, he will give up and call the music "disoganized."
Thus if the player starts a repetitive pattern, the listener's attention drops away as soon as he has successfully predicted that it is going to continue. Then, if the thing keeps going, the attention curve comes back up, and the listener becomes interested in just how long the pattern is going to continue. Similarly, if the player never repeats anything, no matter how tremendous an imagnation he has, the listener will decide that the game is not worth playing, that he is not going to be able to make any predections right, and also stops litening. Too much difference is sameness: boring. Too much sameness is boring--but also different once in a while.
-Richmond Browne
WARNING: If accidentally read, induce vomiting.
It is the nuance and the timing that the performer put into it that make it speak,
Exactly. Nuance and timing. Pattern and frequency. Just like language.
the notes on the page are almost secondary as far as expression goes.
So, if your favorite song or composition was done entirely in 2 notes, but the timing was the same, it wouldn't seem that different to you?
So magic really isn't some transcendental hokus-pokus, it's really a description of the abilities of those who have studied and mastered the art of predicting the next set of actions based upon previous vocal intonations.
+++ATHZ 99:5:80
The exact same thing could be said about spoken/written language. The nuance of spoken language is at least as important as the content of what's being said. Lets look at your two points, sheet music and midi. Sheet music (the written form of music) is unreadable by most people, at least in the way that we would read a book. If you consider music as a language, then most people who read sheet music must translate as they read. Sheet music is also informationaly dense. In adition to multiple notes played overlaping eachother, it contains information about tempo, volume, ect. It's the diffrence between reading a play and seeing it proformed. While both have meaning, seeing the play is more enjoyable because it has the nuance inherent too it, not noted in the stage directions. As for midi, try being moved by a synthisized speech of any good written work. You get just as much feeling out of an answering machine message as you do a computer reading Hamlet.
Apple has never claimed not to be evil, they're just very stylish about it.
All kinds of nonrandom data follows the Zipf distribution, not just written texts. But the relationship between music and language is interesting nonetheless, especially when you consider the psychological aspects, for instance language learning versus music learning.
I appreciate the mathematical analysis of music, but it's important to note that Western music originated out of liturgical chanting (Organum) and folk music. Since the composers of both were generally writing to texts, they naturally placed musical phrase-endings (cadences) at the end of phrases. Therefore, music naturally followed our preconceived ideas of language. Furthermore, since musical understanding is primarily a learned phenomena (compare South-East Asian music with Western; both cultures appreciate their own music first but can learn the other's), it is natural that our learned conception of melody would continue in its textual beginnnings simply through continual, generational reinforcement of the format of melodic conception.
I'm not so sure that music is really a language. It is definitely not a simple question and people have been arguing about it for a long, long time. I've heard people say it is, but I've read some authors who are convinced that it isn't. Here's what I see as the main difficulty. (Hold on, let me try to remember my Derrida from 11th grade English.) The problem is that language consists of two things: The words (signifiers) and the actual objects they represent (signifieds) (linky).They have a concrete relationship. However, does music really have the ability to act as signifiers? Depends on who you ask. You can't exactly look up in a dictionary what an instrumental piece of music means. On the other hand, it's very obvious that some pieces of music are meant to evoke emotions and/or thoughts. The problem is that the meaning isn't set. Different people will get completely different things out of the same music. And if the signifier/signified relationship is different for everyone, is it really a language?
WARNING: If accidentally read, induce vomiting.
Well how about music from spam?
The nuance of spoken language is at least as important as the content of what's being said.
And anyone who doesn't get this should try to find a recording of Robert Morely or Peter Ustinov reading something. Fan-bloody-tastic.
For those not willing to take the effort, or who cannot find such a recording, you can at least rent the movie Arthur and just listen to John Gielgud, or Ghandi and listen to John Gielgud and Ben Kingsley, or Lawrence of Arabia and listen to Peter O'Toole, Alec Guiness, Anthony Quinn, Omar Sharif, Jack Hawkins, Jose Ferrer, Arthur Kennedy, Anthony Quayle and Claude Raines.
Turn off the picture and just listen to the music in the voices of that one.
KFG
Sheesh, I'd swear people down there are capable of holding complete and intricate conversations using solely that word
and you'd be right!
uses of the word
This is pretty trivial. Zipf's Law is regarded in linguistics as a curiosity rather than a deep result. It doesn't really explain anything interesting about language. Music and language are both more and less similar than both following Zipf's Law suggests. On the one hand, as a previous poster has pointed out, language is meaningful. Music may have an emotional impact, but it isn't meaningful in the sense in which language is. On the other hand, there are deeper similarities in the formal structure, pointed out by Fred Lerdahl and Ray Jackendoff in their 1983 book A Generative Theory of Tonal Music.
hmmm....Generally, you are correct, but it is all about context. I compose music for a living. We often imitate the temporary (scratch) music an editor laid in. I copy the mood of the piece, the style and the tempo, but nothing else if I can help it. Sometimes though, the editor is hell-bent on a certain sound. I can get away with 4 or five notes, often more, as long as it is not a blatant ripoff and they are the liable party. It is all subjective though An example with words:
"Oh Romeo, doth thy name and for thy name which is no part of thee, take all of myself."
I might change it: "Romeo: drop that last name of yours and come fuck me."
I could maybe get away with: "Oh Tyrome, deny your family; declare yourself free, and come fuck me."
As a musician, it is hard not to copy, not to realize that I have just dreged up a Led Zepplin riff from the back of my mind. Often, it is impossible not to copy to some degree. There are only so many ways to play 'something in D minor that sounds scary'.
I guess my point is: It is horribly subjective. The current standard is: If a judge/jury can discern that a riff came from a specific source (like the Simpson's theme or Close Encounters) you are screwed. I am all for letting small riffs be considered the words of music, but the issue is, where does one draw the line?
Sig (appended to the end of comments you post, 120 chars)
Besides, if a 'word' is a motif of five to eight notes, a symphony would read like this: "Dmitri Shostakovich wrote this. Stalin was an overbearing ass. Stalin is dead now, and I'm still alive. Dmitri Shostakovich wrote this symphony. Suck it, Stalin."
Then again, works which repeat motifs tend to be more effective than works that go on without reiterating anything. Sort of like Martin Luther King's "I have a dream" speech, which uses that phrase over and over again to slam the point home.
*****
Dear Mary,
I yearn for you tragically,
A.T. Tappman, Chaplain, U.S. Army.
my thought is that in 2nd grade, the teacher told us to make a topic sentance, and then for the rest of the paragraph, DESCRIBE THAT SENTANCE, not some other completely random and arbitrary idea. so, is it mathamatical, or the opression of the 2nd grade teacher?
For the ancient Greeks, music and language were inseparable. 'Mousike' meant choral songs, solo songs with or without instrumentation, and poetic recitations. They did have instrumental music - on stringed and reed instruments mostly - but that wasn't in the same class.
'Mousike' was the art of the 'mousai', Muses. 'Mousa' could be a common noun as well as a goddess, meaning "metrical speech". The word is a derivative of 'mna-', "to remember out loud" - same root as "mental" and "memory", which we get from Latin cognates.
You find a similar thing in Vedic Sanskrit. 'Sangita' means "song-and-movement"; it might include instrumental accompaniment, but purely instrumental music was something altogether. Many Greek musical terms also implicitly include the element of dance: Classical Greeks would have found a 'khoros', "chorus" that didn't move to be a contradiction in terms.
In addition to Zanette's work on music and language, there's also some interesting work being done on language and movement (e.g. George Lakoff). Hooking all of these together and getting a picture of how music, cognition and motor function work together is going to be very interesting.
Google confirms: Ruby is the world's most beloved programm
The old ethnomusicologist in me is tempted to dismiss this as a poorly designed study -- jazz and classical music alone does not make for a representative sample, and people in different parts of the world like all kinds of music that other people find unpalatable. Furthermore, you can't apply his method directly to West African drumming, which is a very popular and exciting music, but you could to the cultural crime that is Britney Spears. ;-)
.5x would come next (octaves), followed by 3x/2 (the dominant) and 4x / 3 (subdominant)... until you get to that nasty tritone.
But looking over the linked study, it's actually quite an elegant look at European and American music. It's neat that the frequency of frequencies (har har) in song parallels the frequency of words in novels. That doesn't mean that "Zipf music" inherently speaks to its listeners, just that people are attracted to this kind of basic math in the world. It's like finding a Golden Ratio -- pretty frickin' cool.
I wish I could see which notes were which on the diagrams. My suspicion is that the relative uses of each note corresponds to the mathematical relationship of the frequency to the tonic. So if x is the tonic, 2x /
Atonal music intentionally avoids emphasizing the mathematically strong relationships, liberating the composer from maintaining that pesky context to a tonic. So it makes sense that Zipf's law won't apply. But before we conclude that people dislike atonal music because it deviates from Zipf, we must answer whether we might also dislike it because we have been indoctrinated into tonality at an early age. And that's where cross-cultural studies are most valuable.
Why did I leave academia to work on websites? This stuff is fun!
rouftop
QAExpress: Solid bug tracking for you. Graphs and reports for your PHB.
From the book Art and Fear (published in '01), which I highly recommend:
The artwork's potential is never higher than in that magic moment when the first brushstroke is applied, the first chord struck. But as the piece grows, technique and craft take over, and imagination becomes a less useful tool. A piece grows by becoming specific. The moment Herman Melville penned the opening line. "Call me Ishmael", one actual story - Moby Dick--began to separate itself from a multitude of imaginable others. And so on through the following five hundred-odd pages, each successive sentence in some way had to acknowledge and relate to all that preceded. Joan Didion nailed this issue squarely (and with trademark pessimism) when she said, "What's so hard about that first sentence is that you're stuck with it. Everything else is going to flow out of that sentence. And by the time you've laid down the first two sentences, your options are all gone."
It's the same for all media: the first few brushstrokes to the blank canvas satisfy the requirements of many possible paintings, while the last few fit only that painting - they could go nowhere else. The development of an imagined piece into an actual piece is a progression of decreasing possibilities, as each step in execution reduces future options by converting one - and only one - possibility into a reality. Finally, at some point or another, the piece could not be other than it is, and it is done.
I belong to the ______ generation.
"(Don't call me Kid)" Jonny Lang and B.B. King were playing a show together at some state fair. B.B calls Jonny up during B.B.'s set to do a song together.
The song gets to the solo part and B.B. motions for Jonny to take the lead.
Jonny kicks out all the stops and plays a blistering solo that shows he's at the top of his game, he's out of his mind - he's damn good. He's doing bends, he's sliding all over, he's sweating with exertion and feeling.
Now it's B.B.'s turn.
B.B. closes his eyes, leans back -
And plays one note. And keeps playing it. With every bit of blues that ever happened to anyone all in that one note.
The crowd goes mad screaming.
Jonny got schooled. :)
Although I still listen to music, I think that music is better off for the most part without words altogether. Look at Beethoven's Moonlight Sonata. Could you imagine it with words? I cant. Music expresses emotion. We shouldn't use words to express emotions, we should use notes.
That's my opinion, at least. Im sure everyone else disagrees.
Ask a good jazz pianist to play a solo. During the solo, try and engage her in conversation. Either she will continue soloing, or she'll talk to you -- but she won't be able to do both.
I've tried this several times while sitting at the keys. The same part of my brain that strings together sentences is busy creating musical phrases -- it stubbornly refuses to multitask.
That this relation exists has been known to jazzists for some time: pianist Bill Evans is revered for his 'conversational' improv style. A master of tone color, Bill could say something humorous or profound with each cluster of notes.
Musicology - long history of research, tons of papers, stuff on score analysis and psychoacoustics thats pretty incredible. However, especially for psycho acoustics, theres little research outside of western cultures.
Computer science - Frequently individuals with a hobby that have brought formidible computing skills and analysis techniques from other fields, but are largely ignorant of the works within music departments (see pretty much any IEEE paper on music for examples). Biggest problem is typically lack of statistically valid experiments (like the test sample of 4 pieces in this article).
Ethnomusicology - The first 20 years of the fields existance (~55-~75) was dedicated to this kind of research. Now that computers are powerful enough to do more meaningful analysis, the backlash against this analysis is fading, but any ethnomusicologist can tell you all the pitfalls - especially how critical cultural context is to the analysis.
As for the relationship between music and language, both music and language are tightly tied to culture. In that sense, they are similar. Anything more profound requires an accurate, precise definition of the two terms. This is extremely difficult and a good way to start a fight with ethnomusicologists if your so inclined.
The number you have dialed is imaginary, please rotate your phone 90 degrees and try again.
By the way, the Spamradio guys are in need of Icecast relays to keep them running. (I have no connection to Spamradio - I just listen to it from time to time and think they run a good website).
"There are already a million monkeys on a million typewriters, and Usenet is NOTHING like Shakespeare." - Blair Houghton
There even exists a _real_ musical language, in which musical patterns represent actual words. This language was developed in the 19th century, by Sudre, and was called Solresol. He even wrote dictionaries and such. It never really cought on.
More info on Solresol
Z
Having had a quick RTFA, it's clear that there's plenty of substance in this research. On the other hand, I'm a perl geek, and I wanna hear what Larry has to say on the subject! He is *the* man where languages and linguistics are concerned after all, and there's probably More Than One Way To Do It In Music!
I find your ideas intriguing and I wish to subscribe to your newsletter.
coming from the book that tells the meaning of life, shouldnt the babelfish be able to translate music to speech assuming it is a language?
Exactly. Because the word is so frequently used, its meaning grows more and more dependent on its intonation (== music).
Besides, if a 'word' is a motif of five to eight notes, a symphony would read like this: "Dmitri Shostakovich wrote this. Stalin was an overbearing ass. Stalin is dead now, and I'm still alive. Dmitri Shostakovich wrote this symphony. Suck it, Stalin."
Please don't say this so conclusively. The idea that Shostakovich was making some kind of a rebellious political statement in his music is highly debated, and there is a great deal of evidence to suggest he truly supported the Communist ideal and strove to make his music conform to socialist realism. Sure, there is evidence in the opposite direction, which is why one should qualify such a statement with something like "His symphony might read like this..."
The purpose of language is to convey meaning. If meaning is conveyed through the use of a sound, then the purpose of language has been fulfilled. Thus, that sound is a form of language.
Eh? The purpose of language is to convey specific meaning. That's how come words have specific meanings (yes, even in English). What would the point of language be if I told you to pass the sugar and you thought I said "You fucking bitch you ruined my life!"
Music has no specific meaning that can be determined in its expression. Certainly I mean something when I write a passage of music, but you probably won't understand what I meant when you listen to it. You'll still get the emotional overtones and so forth, and you might get some visualization along those lines, but...
Well, with language the language carries it's meaning with it. You have a store of your own that will closely match mine, so when I write these words you will understand each word. You still have to piece them together, analyze their relationships, spell-check, and then hopefully the understanding you will achieve is the meaning I'm trying to convey.
Music is different than that. It lacks a store of definitions, for one thing. It also varies from person to person. Take a regular rock song. Two guitarists play the same riff, but neither one of them plays it exactly the same. The bass player plays the same riff, but it's completely unlike either guitarists. The drummer plays the same riff, and it's also completely unlike the rest of the band. The singer's worthless, they always are. What you bring to the song when you listen to it is your own life experiences. What I bring to the song when I write it is my own life experiences. Since two people can't experience the same life, then it's not possible for two people to pull the exact same meaning from a piece of music. Many will come close, and lyrics help a lot (no matter how worthless singers are), but no one will be the same in this regard. Music is more personal, while language is social.
Am I making sense? I'm pretty sleepy...
Like what I said? You might like my music
With language you can say things so general and abstract. You can also be very, very specific if you take greater care.
IMO, one of the big downfalls of language (English, anyway) is that it is much too easy to be imprecise and ambiguous. Even legal text which strives to be precise can be interpreted in different ways. This is a huge problem because years down the road after text is written and meant to capture a certain meaning, it can be re-interpreted years later to mean something else.
Is this a problem with every language? It seems like more of a problem these days, maybe just because I am noticing it more, but what can be done? Better education? English 2.0?
On an interesting and related note (ahem): The Clangers. A BBC tv children's programme in which the characters communicated entirely by whistles.
The whistles were blown by actors, using a script. When they aired the show, they found people writing in saying "my child insists the characters said X, Y and Z" - is he mad?
The thing is, the kids usually got it spot on.
Lucky they took out the swearing in the original script, then. Also of note is the final paragrah in that link, which says:
I took an episode of The Clangers to the 1984 E.B.U.
conference in Germany and showed it to the participants without my voice-
over. Afterwards I asked them whether they had been able to understand
what the Clangers were saying.
"But of course." they replied. "They are speaking perfect German."
"But no." said Gerd, "That is not so. They spoke only Swedish,"
'No rational religion claims "supernatural" exists, that's an atheist slander.' - seen on slashdot.
This is hogwash. You can write atonal music with a great deal of recurring context. It isn't atonality that drops context it's the composer's decision not to repeat himself.
That description really only seems to work for painting, where each brush stroke is a near permanent piece of the finished product. Even if covered up with another colour, the original will still show through in some minute way. As another poster has pointed out, Melville very well may have written the second chapter first, then added the first later on to bring more depth to the story. I've written many songs before that come out nothing like they were originally imagined after stumbling on a guitar chord that sounds better than what I had in mind, or because rearranging a few pieces made them more interesting. A piece does grow by becoming specific, but it very well may change entirely from the first concept, and may not always be growing into the best piece that it could be, only the best that the artist could imagine at that time.
do not read this line twice.
There is a lot of theory in this area that does provide a more solid linking that the article suggests. The research in this area has been going on for a long time and has gone far beyond the ideas mentioned in the article. For anyone seriously interested in the subject, a good place to start is with "A Generative Theory of Tonal Music" by Fred Lerdahl and Ray Jackendoff. Originally published in 1983 and reprinted in 1996, this book has been the foundation for a lot of subsequent work. It's not an easy read, but
much of it should be comprehensible for those who have experience with computational linguistics and a reasonable familiarity with Western music.
I would add, however, that one should not confuse the fact that music contains structure that is amenable to many of the same analytical methods as natural languages with the rather vague notions of how it "speaks to us". It doesn't appear that music has symbolic and semantic systems that are anything at all like natural language.
Often, a piece passes through a stage, or several, where it grows by becoming more specific. When an author takes a rough draft, and trys to cut it down to a required size, for example, that's all about becoming more specific. Some authors concentrate whole writing sessions on finding just the right word over and over, while others enter that mode sporadically, for a few minutes as they continue to work on on other parts of the book.
Other changes simply aren't about distilation. What happens when an author introduces a spear carrier character such as a coach driver, and polishes the character a bit, and suddenly realizes there's the possibility to work a romance between that character and another into the story and that in turn will even let her comment on social issues in her fictitious society that she will otherwise not get to address?
Who is John Cabal?
I agree with you.
I consider writing, painting, and drawing to work in the same way. It never is created in a straight progression from beginning to end as the perceiver reads/views it. There is almost always the initial layout phase, which then continues into fleshing out the concept and then into working the details out as they should be. You can see evidence of this in every area of human design. Buildings aren't built by placing a stick of wood in the ground and then adding more on, regardless of how LEGOs work.
There are only a few people who actually create in the order that the viewer/reader will perceives their art, and those people are rarely ever confined by such nonsense as meaning. These people also will not fall into the category of losing all options as the work progresses.
It is a romantic, idealistic idea that holds little truth.
Hmm, has anyone used this technique on spam texts? Spammers are adding a lot of irrelevant and even nonsense words to try and defeat Baysian filters, it seems to me this would make spam distinctivly less coherent.
"Music is powerless to express anything at all"... or something like that, I think from it comes from his (in)famous Poetics Of Music... there we go with that poetry schtick again... most readers agree that Igor was probably being somewhat facetious, but his point was (as I take it) that music doesn't "express" or "have meaning", it just "sounds", and we go ahead and stick any number of beliefs and ideas on the experience... see Morton Feldman's commentary on this sort of thing... we can't have music that just sounds, now can we ?... also see Copland's remarks on audiences and what they think happens in a complex piece of music...
"Give them a jig and tale of bawdry, else they sleep." [William Shakespeare on his audience...]
In Douglas Adams' book Dirk Gently's Holistic Detective Agency the main character made his name with software that translates business figures into music. Apparently now someone is trying to get a PhD based on it. Douglas Adams is such a visionary.
move along, nothing to
Well, the problem is that most people write songs by repeating one word over and over again. Naturally, if you then use that word in your song, you've stolen their entire work.
Now if they made a song with 500 words, you could rearrange those 500 words and make a completly different song.
Or maybe it's just a poor analogy.
That only was true pre-1700, when the piano was based on harmonics rather than math. Pianos (well, harpsichords) were tuned in a certain key, and had to be re-tuned halfway through the concert to play in something different. Therefore, a harpsichord tuned in C would sound great in C, and sound okay in G or F, but if you tried F# the relationships between the major notes would be wrong, and the whole thing would sound awful.
While Bach was composing, a new idea of "equal temperment" gained popularity, where the difference in frequencies between notes on the keyboard were all the same (1/2^12). You could play in any key you wanted to and it would sound the same as any other key. Bach liked this so much he wrote a prelude and a fugue in each key, just because he could, and then he did it again. These are some of the most popular of Bach's preludes.
"Physicist Damian Zanette of the Balseiro Institute in Bariloche, Argentina"
NOT
Linguist Damian Zanette
"Cornflakes are not the innocent critters they seem"- Sterling Morrison
This is not at all true. What most people think of as "tonal" is the predominant 12 tone system in the western world. We grow up hearing it because every tune we hear is based on the 12 notes and every instrument is tuned to them. We are also used to the standard system of scales, i.e. if a piece is written in A minor, then chances are it will end in A minor and the preceding chord will be E major. Anything that departs from these traditions can be considered "atonal" to most of us.
But other cultures have other tonal systems that may sound totally foreign to our ear, because they are not based on the 12 tone system. An example may be the Byzantine chant which is modal music based on several different (not 12 tone) scales and which also does not follow our traditional understanding of rythm, but there are countless others. I seriously doubt that it is "sometimes not pleasing to listen to".
So tonality is a relative term, which to me makes it appear that effort of trying to tie it mathematics is futile (as mathematics is not at all "relative")...