Hacker News new | comments | show | ask | jobs | submit login
Music Theory for Musicians and Normal People (tobyrush.com)
255 points by Tomte 888 days ago | hide | past | web | 120 comments | favorite

It piqued my interest from the cached thumbnails, and I was able to find a mirror of the complete set:


It's licensed under Creative Commons BY-NC-ND, and Toby W. Rush is the author. Great resources by the way, better and more concise than a lot of theory info for beginners (bordering on intermediate, if you are able to get to some of the more advanced scales, structure and basic part-writing/voice leading at the end).

Hey, thanks for the nice hug, everyone! It's been a fun night. :)

I updated the mirror that ewoodrich referenced; what was up there before was a slightly older version. The new one has a few more pages in it and a few typos fixed here and there.

I've also swapped out the page on my site with a redirect to the mirror, since my web hosting service has rebooted the server twice now. The mirror is on my faculty web space at work, so we'll let them worry about it now! :)

No worries, though. Glad you all are finding the materials useful.

A previous appearance: https://news.ycombinator.com/item?id=4807091.

(Normally we'd point the post at the URL that's working, but that pdf is already downloading so slowly that I fear we would just topple the next server over as well.)

Based on my own experience, I find the No. 1 obstacle in learning music theory is the difficulty, _in the early stage of learning_, to practice the knowledge to gain a real sense of how the stuff is used and why.

Take computer programming as an example. After 15 minutes of HTML lesson, a newbie can already look at her creation in a browser and the html file on hard disk, and have fun tweaking it (<html><body background="green">My home page</body></html>, remember the days?). An hour of Java lesson can result in several variations of "Hello World!" programs, created by the student. Basically, people can put their hands on the knowledge and use it to produce somewhat useful/tangible stuff.

Not so much when learning music theory. After enduring the bombardment of names/concepts/terms/maths and other stuff for quite a while, I don't have the slightest idea where I can practice in using them to analyze or write some music.

I wish there is a lecture in the format of "One hundred songs for one hundred music concepts". The songs should be stupidly simple (children's songs, simple pops, short classical pieces, etc.). Each lesson concentrates on only _one_ concept appeared in the piece, explains what it is, how it is used, why it is used in that way, and how its application makes nice sound, etc. Basically, show me the _effect_ for each concept in a simple real world setting. Then more importantly, give me exercises to write a few bars of music using this concept, no matter how bad my writings are, as long as I am applying the new knowledge.

You're wrong about categories of analogy, but I don't blame you - analogies are a pretty slippery slopes. In particular, you're most likely conflating these because of the implementation detail - you write HTML, so writing music is an analogy. Not really - writing HTML is closer to performing music.

You can teach almost everyone to perform "Twinkle Twinkle Little Star" on piano or "Smoke on the Water" on guitar within 15 minutes, both of which are de facto "Hello world!" equivalents to many aspiring musicians.

EDIT: Removed attempt at drawing a HTML/music analogy which could be closer to truth. It is really fruitless endeavour, makes no sense.

I disagree. The rendering engine of the browser is the performer who looks at my script(HTML/CSS/JS) and produce the final result for the visitors to the web page. Writing HTML is not closer to performing music, but actually closer to composing music, if you will.

Also, I think although knowing theory would help you play a song with an instrument, it's not required, is it? More likely knowing music theory is for understanding/appreciation of the composition/structure of the music. So no, playing "Twinkle Twinkle Little Star" on an instrument is NOT equivalent to writing "Hello world!" program. On the other hand, composing a super simple but complete song is closer to writing a "Hello world!" (or a Fibonacci series) program.

CSS is probably a better analogy for performing music. You generally keep the same content, but presentation is important.

One of the best way to learn music theory (much of which is harmony) is to play jazz, or jazz-inspired styles of music – you'll get the most exposure to the most diverse array of concepts in the least amount of time (not saying it will be a short amount of time – just compared to, say, being a classical musician). The emphasis on improvisation requires players to have comprehensive understanding of the underlying harmonic concepts – you can play jazz completely by ear, but this is not the norm by any means.

As a couple of commenters have said, your HTML analogy just doesn't seem to fit here, music theory is descriptive, not prescriptive. The best way to learn it is through playing music, a lot, and then using concepts from music theory to go back and gain more insight into why certain things sound the way they do, why certain progressions work well, why an F natural sounds ethereal and sublime on top of a B∆9 chord while an E natural sounds awful, and so on.

Again, I believe the optimal way to learn music concepts is to play (not study) music.

If jazz doesn't interest you, I would ask you what your musical goals are, and what genres do interest you, and just encourage you to play, and then play some more. Ideally, play with musicians who are better than you - you'll learn more, faster, just like anything else. It takes thousands of hours to get good at this stuff.

Well I think there are a number of things going on here... If the equivalent set of documents were created for programming you might have all sorts of concepts ranging from methods, classes, pointers, monads, program structure, design, markup, protocols, API, ... and a whole bunch of other stuff. A newbie to programming would be completely overwhelmed and unable to put any of it into context.

I know for me, the stage at which I was doing the very simple HTML examples you listed was when I was 13. Back when you're that young, just playing around with that stuff was SO FUN for some reason. As a 29 year old guy if I were just starting out I'd be much more inclined to try to skip that step and make stuff more real-world.

The equivalent in music theory would be simply playing around with the white keys on the piano (C major). Making interesting melodies (but ultimately kinda boring melodies like mary had a little lamb). To your young ears, this would be very interesting and fun. You'd play around with this for weeks and weeks and make small observations. For example, you might notice B->C has a sense of resolution to end the melody, or internalizing what C->G feels like by just playing it a lot and singing it. You might notice that E->F feels smaller and has more of a "pull" to it than A->B (in the context of C major). It's this stage that's so crucial for later understanding music theory.

Adults want to skip the simple exploration and just dive right into Beethoven. We think we can use our strong analytical sense to just book-learn music theory. What I've learned is that it's much more fruitful to make observations first and then learn what music theory calls those observations.

When you're writing HTML at 13 years old, you don't really know the high level concept of a "mark-up language" and how it might be manifested, you just know that there are parts in angle brackets that don't show up on the page but control the content. Later in high school when you learn the concept of a mark-up language you make that connection to HTML with extreme ease because all those memories of changing background="green" to background="black" flood back to you.

I understand it's important to have plenty of intuition, but as an adult I do want to speed up the learning process. That's one of reasons I set out to learn the theories. My main goal is to be able write songs of quality at least slightly above total crap.

Maybe I have not found the right learning materials or correct way, or maybe because I am the type of persons who need to have lots of hands-on exercises to be able to grasp new concepts.

Sure, programmers also need to know all sorts of things you listed, but armed with only a few concepts, a newbie can start to write simple programs from the very beginning, in the forms of piecemeal topic-focused little exercises. The result of each exercise is a fully functioning program. I wish I can find a similarly structured music theory course that uses the same approach. Also, are there projects in music which are similar to ProjectEuler.net or the likes, where you can do focused practice on specific topic? I would be happy to pay for those services.

> As an adult I do want to speed up the learning process

I understand the desire but it's also very important to realize that these things take a LONG time to develop. You can't really expect yourself to grasp chord progressions until you have a firm understanding in scales and you can't just gloss over it and be familiar with it, you have to internalize it. You have to get scales under your fingers... you have to feel the pull of the leading tone to the tonic... you have to be able to sing them and know them inside and out or your knowledge of chord progressions will be built on a shaky foundation.

The equivalent to project euler would be to just compose your own songs. Seriously, compose then analyze. Since music is subjective there's no way to say, "you're song is correct, move on to the next song". Also, there's no real correlation between song writing and theory knowledge that I know of. You can know about theory a lot but still not be good at writing songs. Also knowing theory doesn't make you a good musician by any stretch of the imagination. Music theory is descriptive, not prescriptive (i.e. it doesn't tell you HOW to write or what you should write, but rather is just a language to describe what you've written)

You can know about programming like if you're good at math you probably have a sense of what programming is about but until you sit down and type out some code, you're likely to not be any good at programming.

> My main goal is to be able write songs of quality at least slightly above total crap.

You will write total crap. The best thing you can do is come to terms that you'll write crap probably for years and that's okay. Here's the first song I ever wrote way long ago:


Bleh... not exactly something I'm super proud of other than the fact that I did indeed finish it and I was honest with myself about what level of song I could write at the time that I wrote it. I'm still not a super great musician, but I've at least improved every song since that one!

A very nice 'Lullaby', I enjoyed it, thanks. Also thank you for the tips.

> Since music is subjective there's no way to say, "you're song is correct, move on to the next song".

Are there some kind of music "compiler" or "linter" for educational purpose out there, which check for a song's obvious and blatant violations of fundamental composing principles? This might helps beginners to hone in on the basics. I know the rules are meant to be broken later, but at the beginning, the 'compiler/linter' is efficient and cost effective in guiding the student and checking their practice work.

Thank you :)

I think a "linter" like that would be kind of like a linter for analyzing poetry or short stories. It's so subjective that the most it could do would be offer very superficial suggestions without really understanding context. Things like "does it have a rhyming scheme and meter". And that's not that useful because some poetry doesn't rhyme, and some doesn't adhere to a meter.

Like, is 50 Shades of Grey bad writing? some might say so... but can you deny the popularity of it? Are the people that enjoyed reading that book wrong for enjoying it? Trying to apply an objective measure to quality of writing or music is futile, I think. Same goes for music. You might write a song that some stuffy old classical music PhD would scoff at but maybe children LOVE it. How could a linter possibly know that?

I say embrace the subjectivity of it. Write a song, ask people what they think of it. Accept that you will not write super great songs right away and just keep writing and experimenting. Use theory as a language to explain what you've done in songs you wrote and songs that you like.

Good luck!!

Analyzing and writing music aren't the only things you can do with music theory. For me the main application at first was playing music by ear. For example, you could learn what a dominant is and how it sounds, then sit down at the piano and figure out the chords to some song that's mostly tonic and dominant, and so on. And once you have that down, you can compose some stuff meaningfully.

Consider learning jazz. The whole approach of many basic kinds of jazz is to take some chords and turn them into music.

Thanks for the tip. Any suggestion on learning materials or where to find beginner friendly resources?

This isn't "music theory" (although almost everyone thinks it is). This is "music jargon and notation systems". A theory is something that explains an observed phenomenon. This jargon and notation stuff doesn't explain music, it just guides you through the jargon and notation.

I have a music degree and learned all this stuff and a lot more. This presentation is an excellent, high-quality overview of this non-theory junk.

It's extremely culturally-biased and, again, offers no real theoretical explanation.

If you want real understanding of music in a useful way for real people, I recommend Bob Snyder's book "Music and Memory: An Introduction" which was actually written to teach creative artistic people about how music actually works rather than how aristocratic European music notation works.

For some other perspectives, check out http://sethares.engr.wisc.edu/htmlRT/contents.html and http://sethares.engr.wisc.edu/ttss.html (normally I wouldn't push these on average folks, although they are superb but the HN crowd should be fine with them)

I'm not sure how parent became the top comment, but "music theory" is absolutely appropriate here. This terminology is commonly accepted by musicians, educators, and practitioners. OP's link could say "western music theory", but that seems unnecessary given the audience.

Secondly, why would you call this junk? This is actually very useful for people learning to play, and for those who want to understand the basics of harmony and rhythm.

What you are talking about is typically called "music perception theory", "music cognition", or "music psychology".

Indeed "music psychology" is a good term but is almost a truism. There is no such thing as music otherwise. Music is only psychology. Everything about patterns of dots on paper and the theories about them is valid when it aligns with psychology and invalid otherwise. A pattern in notation that isn't psychologically experienced is not music.

Teaching the notation by the hypotheses about music that the canon of "music theory" gives us is like teaching chemistry by 100 year old diagrams and names of chemicals.

To be perfectly fair, the post here does do a BETTER job with the Western-notation-scope than most other examples.

Okay, okay, okay! I get it!

Now, explain to me the music theory of the D major section of the Bach Chaconne! Say the first few notes of that section after the opening open D string and the D an octave higher on the A string; or if want to avoid the open string, play the D of the open D string on the G string and the other D on the D string.

Then what is the role of the notes repeated three times, and those triplets repeated several times? These triplets are insistent? Heifetz played them this way, but not all violinists do!

Then what is the connection from the last few bars of the D major section and the first few bars of the following D minor section? Or the last of the D major is the climax of the piece, and the start of the D minor is a catharsis?

For more, at the beginning of the second to last bar of the D major section, there is a chord the first G above middle C, the next G up, and the B above that. The notes are quarter notes except the upper G is an eighth note and changes to the D just below, also an eighth note. Well, what is the role of the change from that G to that D? Some violinists emphasize that change from the G to the D a fourth below, that is, make the D easy to hear and the fall something dramatic, say, as some kind of resignation. What's going on here? What likely did Bach have in mind?

Rostropovich, with his recording with von Karajan of the Dvorak concerto, made something really special at the beginning of the second movement: He has some depiction of remembered pathos. So, how'd he (Dvorak and/or Rostropovich) do that?

From Rostropovich I hear emphasis with vibrato, volume, and tone color and careful timing, phrasing. It's passionate music. What's going on? It sounds a little like some abstraction of some speech with pathos.

Maybe easier is the Prelude of the first Bach piece for solo cello and, there, the rising chromatic passage to what is apparently the climax of that piece. So, what else is going on there just besides that rising chromatic passage?

Then there is the cello solo early in the first act of Die Walkure, maybe representing the girl? How to write a cello solo that sounds like a compassionate girl?

For some more, there is the duet, "Presentation of the Rose" in Der Rosenkavalier with Barbara Bonney and Anne Sophie von Otter as at


Some parts of the duet with the two voices are just marvelous, and I'm not understanding how Strauss did that. What's going on? How'd he do it?

For more, there's too much going on in Ein Heldenleben, but I'd be happy knowing a little about what Strauss did there, even how the girl, that is, the violin solo, worked, or even just how Stauss made the girl so comforting in the last section.

Are such issues what you have in mind as music theory?

How'd I get into such questions? One night on a radio I heard Beethoven's 7th symphony, and it seemed easy to like, e.g., the dance. Occasionally Dad gave me some money, and I bought some recordings, Vivaldi through Rachmaninoff. Then I was a grad student in math at Indiana University in a dorm next to the music building, and a Stern protege put his old Italian violin under my left chin, gave me a lesson, and I took a violin course, starting with how to tune the thing. I learned the music theory of perfect fifths, that is, three times the frequency of the lower open string is the same frequency as two times the frequency of the upper open string, and the beats between those two can be used to tune to perfect fifths. Not long after making it in some form through much of the Chaconne I got too busy to continue on. So, I like the music but don't fully understand just why. Music theory has an explanation?

Your questions are great! They totally fit the scope of actual music theory.

The first thing to understand is that written music notation is not music. Partially, this is like the way that written English is not spoken English. All the notation we've inherited from Bach specifies certain things and leaves other elements unspecified, in much the same manner as this text really fails to carry the tone and delivery you would hear from me in real life.

I can read some English text in different ways and bring out an amazing range of meaning and interpretation in how I express it. Likewise with music. We can speculate about what Bach had in mind, but we can only use various bits of evidence, writing, known practices, etc. to guess about anything not present in the notation.

There are tons of parallels between music and speech, and I recommend the book Music, Language, and the Brain by Ani Patel for more in-depth about this.

The stuff in the book I mentioned originally (Music and Memory) gets into a lot of the issues too. Our processing of all these interacting parts has to do with the various psychoacoustic factors that make things merge together or heard distinctly in our perception. Auditory stream segregation describes the factors in how we follow a line of music and why we relate one note to another instead of in a different way. And some things are really distinct, so everyone hears them the same way. Many others are ambiguous so you and I can hear the exact same sounds and actually parse the music quite differently in our heads.

I actually could sit down with you and make sense of all the subjective experiences given understandings from psychology and explain much of what is going on in each of your examples. It's a bit hard to express in text though, easier in person with the ability to gesture and intone and sing while discussing and experiencing the different examples.

Thanks, I'm learning!

Finally it dawned on me, the G to D change was while in D major and, thus, back to the tonic! So, there, in the next to last bar, right before the top of the climax of the section, Bach touched on the tonic again! So, artistically this tied the motion and emotion of the other notes back to the tonic and not left partly lost out in the ozone -- or some such.

This afternoon, I listened to the Heifetz and Piatigorsky performance of the Andante of the Brahms Double Concerto: Finally it dawned on me: Brahms had an abstraction of some kind of speech, say, of just the sound of the speech, the sounds that conveyed the emotion without knowing the literal words. So, it sounded like speech, maybe like speech in a foreign language where didn't understand the literal meaning of the words but, still, got much of the emotional expression. It appears that you have a book that expands on such a thing.

Edit: Okay, at


I found a book review of

Aniruddh D. Patel. Music, Language, and the Brain.

He's deep into some apparently fairly speculative research and makes me like more my little idea that some of music is like listening to some passionate talking in a foreign language were know none of the words but, still, get the emotional content -- that is, the music and the emotional content of the speech can be similar.

So, back to the opening cello solo of the Brahms Double, we can imagine some passionate speaking where a person says something, just a few words, then says it again a little differently, then again with a little different emphasis, then again using maybe different word order. Well, maybe that's much of what Piatigorsky playing what Brahms wrote is doing.

Or, back to the beginning of the D major section of the Bach Chaconne, we have phrases

F# G E, F# G A D, E F# G F# E E A F#

[this example would be still more clear with the actual music with the note timings -- maybe the harmony would also help!] Here I am inserting commas to separate what I am regarding as the separate phrases repeating the same basic emotional content.

So, an idea is that that music is like a speaker saying much the same thought three times, each time with mostly the same words and in a similar but different way, that is, with variations.

Why three times with variations? The speaker is passionate and, maybe even, pleading, and trying to get their meaning across if not the first way then maybe in one of the two other ways. Besides, the third way is more complete than the other two and, even, with the rise to the A, has a point of high emphasis and emotion. Maybe.

See something similar at the beginning of Silent Night where there are four variations of basically the same emotional expression.

I suspect that Bach and Mozart could "search their feelings" (Star Wars), think of some corresponding speech, hear the corresponding music, and just write it down.

The book review suggested that different languages have different patterns in the speaking and, thus, for each such language, music written in that language will use similar patterns. So from that might argue that the Bach Chaconne was not written by a native speaker of French or Greek. Or Massenet just does not sound like Bach, Mozart, Beethoven, Mendelssohn, Wagner, or R. Strauss! Maybe Massenet sounds more like Debussy or Ravel!

Indeed! You're definitely getting into one aspect of things. The most important thing to know about music is that it is a super-stimulus. Great music activates all sorts of otherwise unrelated parts of the brain all at once! So we can spend infinite amount of time on any one of the cultural references, the language-like aspects, the rhythmic structures, the play with prediction and surprise, the psychoacoustic properties, etc. And on top of all that, lots of music happens in live performance with visual and social elements.

Regarding the language relationship, many people have discussed the ways that music, especially harmonized singing, is like a super-voice, it's triggers our emotional reactions to expressive language but at a higher order of magnitude.

The real interest comes when we realize that all the valid elements of the notation-jargon-style theory actually can be themselves described psychologically. So, the books I've mentioned, don't deny the classical ideas of tonality and chords and voice-leading and all. They just explain these phenomena in ways that are actually more meaningful and more universally applicable than just recognizing the notation patterns.

Knowing how to spell doesn't make you a writer. I think the parent is trying to explain that there's a cultural bias and inscrutability with respect to how music theory is taught. Does learning a notation system make one a theorist? One can be a western musician in, say, a jazz tradition, understand theory, have theories and play music and have never used the system of notation explained in the otherwise awesome comic.

> Knowing how to spell doesn't make you a writer.

Of course. But knowing how to spell is the first step towards learning to write -- my four-year-old will tell you all about it. :-)

Music theory is more than just spelling. It's also grammar -- it puts together rules for harmony and rhythm, and allows you to analyze music and communicate ideas by giving names to patterns (such as scales, chords, cadences, functions, meters, etc.) It allows you to learn for yourself how to break these rules and deviate from these patterns.

You can learn the theory without learning the notation, which is quite common among guitarists. You can also make music without knowing any theory at all, but it is always more helpful to know some theory.

> I think the parent is trying to explain that there's a cultural bias and inscrutability with respect to how music theory is taught.

Maybe, but the parent is also trivializing what is commonly referred to as "music theory" by calling it just "jargon" and "non-theory junk."

The poster above is trying to one-up the OP. Of course a notational cheat sheet is not the same thing as a rigorous intellectual foundation, but the phrase 'music theory' is the most common term to describe the traditional notation and grammar for the western music corpus by a mile. That's what most people want when they search for 'music theory'. It's not a good use of the word 'theory' but the context within which it is embedded is widely familiar.

One can be a western musician in, say, a jazz tradition, understand theory, have theories and play music and have never used the system of notation explained in the otherwise awesome comic.

True, but it's pretty unlikely. Most people who play music pick some up theory (in the sense of this submission) along the way if only so they can have meaningful conversations with other musicians.

> That's what most people want when they search for 'music theory'.

I think most people are actually thinking, "I want to understand music more." when they search for "music theory", and instead of learning deeper understanding about the nature of music, they are taught about notation traditions. Then, most people decide they do not especially care about "music theory" anymore. The folks who continue are those who like studying notation patterns.

We're going to disagreee about this. I do get where you're coming from and I actually don't care for musical notation; I can read it but I can't sight-read to play so studying notes on a staff involves a lot of painfully keyboard pecking (or occasional cheat transpositions into a DAW) followed by a bunch of more fluid experimentation with the musical elements I played so badly, and it's during this experimental phase that I may or may not learn something.

Like you I'd much rather read text about the psychology of musical perception and the why of music. I disliked the notation drills etc. when I had piano lessons as a kid, but as I never had a natural ear for chords or harmony (and thus frequently found myself wanting to improvise things but without a good understanding of how to go about it) I came back to the theoretical stuff as an adult, and ended up getting interested in other musical traditions like Arabic and Indian styles as well as the western one (though I don't want to oversell my knowledge or skills here). Different traditions use different schemes but they all have some way to codify things, because unless you stay within a single tradition or are blessed with an eidetic memory chances are you'll need to note things down at some point.

Where I'm disagreeing with you is on the terminologyl I hold that when people search for 'music theory' it's because they want to be conversant with other people on the basics, even if they just want the bare minimum of how to distinguish between major and minor, construct a few basic chords, or read/write a melody outside of a piano roll. It's unfortunate that the label 'music theory' has become associated with this basic stuff; it's as if we used 'literature' to refer to basic spelling, grammar, and vocabulary and used some other term to describe the actual study of artistic writing - but that's how it is.

It's unfortunate that the label 'music theory' has become associated with this basic stuff

Only in the way that the label mathematics has become associated with arithmetic. The stuff here is what you need to know in order to get started analyzing music of the common practice period (you'd need set theory to talk about the pitch content of most music after 1900). But when people get doctorates in music theory, they are not stil studying the circle of fifths. They use these concepts to describe music in order to come to some understanding of how and why it is structured.

"a jazz tradition"

What system of notation do you think jazz musicians use? I don't think you'd have any luck trying to find employment as a jazz musician since at least the 1930s unless you could at least read music, and be fairly good at sight-reading as well. And as far a composing and arranging? Those guys not only knew tons of theory, but very consciously developed their own extensions to it. They were very familiar with Western classical music, as well as Blues.

> A theory is something that explains an observed phenomenon.

That's true, but if you take the "observed phenomenon" to be "Western, common-practice tonal music (and notation)," then this certainly helps to explain it. It's certainly what gets taught in undergraduate music theory classes (much of it in "fundamentals" courses). Graduate work in music theory involves doing exactly what you say: explaining observed phenomena, in both "classical" music as well as about any other musics you might imagine. That's not to say that all music is Western, common-practice tonal music, but a great deal of music comes out of that lineage; an understanding of these basics are often necessary (or at the very least, useful) before branching out.

You see this sort of thing in all disciplines. I might find an "Introduction to computer science for normal people" that explains big-O notation, or teaches some basic Python, or what have you. Someone might well say "This isn't 'computer science,' but instead 'jargon and programming.'" This person might be right, of course, but for "non-computer scientists and normal people," the distinction is not so clear.

"Western, common-practice tonal music (and notation)," is not anything close to a good definition of the word "music".

Of course it's not. I was just responding to the thought that a theory explains an observed phenomenon, and trying to show how for at least one definition of music that the OP was indeed "music theory."

It never claims to be "a comprehensive guide to all things musical" or "a theory of all music," but it certainly qualifies as music theory.

This is music theory, or at the very least, the bulk of the document is about the theory of harmony. The first few pages may give the impression of solely exposing jargon, but as you progress through the pages, the author moves on to the theory of harmony, or what has been referred to as such for the last century and a half (at least). I'm not saying the expression is semantically perfectly sound, but it's been in use for long enough to be considered correct. It's true that a better expression may have been "Harmonic Aspects of the Practice of Composition".

Schoenberg's classic 1910 Theory of Harmony is a shining example of this. It's pretty much a practical guide for composition, and follows a structure strikingly similar to that of this document.

It's a practical guide for music written before 1910.

Unfortunately quite a lot has happened in music since 1910, and if you're trying to write music for the 2010s Harmonielehre and 'theory' of this kind won't just confuse you, it will actively mislead you.

The problem is that it teaches you nothing about the importance of performance expression, or the many different ways modes get used and abused in pop/rock/electronica, or how to create arrangements that capture and keep listener attention.

Pop has become a massive historical experiment in perceptual psychology. The most popular bands can literally fill a stadium - something common practice music has never done.

While that doesn't mean pop is better in some absolute sense, it does suggest it's doing something right for many listeners.

If your training is too rigidly classical it actively stops you being able to hear and understand what that right thing is, because you're too busy concentrating on a small subset of the many details in the music.

> Pop has become a massive historical experiment in perceptual psychology. The most popular bands can literally fill a stadium - something common practice music has never done.

> While that doesn't mean pop is better in some absolute sense, it does suggest it's doing something right for many listeners.

Not Saying you're necessarily wrong, but you can't say just that without pondering that question : is it due to discoveries in music composition, or is it due to efficiently applying marketing to it ?

I don't know if it's the same everywhere in the world, but here in france, if you turn on the radio and zap through channels, you will hear the same song up to 10 times a day. Then you turn on TV on night and see those same groups as guests on TV show. Then you read in local newspaper than they will play in your town. I can see how it could produce the effect you describe.

Actually, this theory of harmony is largely wrong. Harmony is actually explained in my original post's link: http://sethares.engr.wisc.edu/ttss.html

Just like other fields, people actually do advance in our understanding and figure out what was wrong in the old theories. The music world has just done a very poor job of disseminating the updated understanding. That's probably because academic music institutions are too often about cultural preservation rather than about furthering our understanding.

"music theory" is the widely used and accepted term for "music jargon and notation systems" and their related concepts.

It is not a "scientific theory", as, say, Thomas Kuhn might have used the term.

"Music theory" uses the word "theory" as the colloquial antonym of "practice".

As in the adage, "In theory, theory and practice are the same; in practice, they aren't".

Music practice means strumming an axe, blowing a horn, banging a drum or tickling ivories, and so on, whereas music theory is ... any study related to how to do these things, away from the instruments.

> If you want real understanding of music in a useful way for real people, I recommend Bob Snyder's book "Music and Memory: An Introduction" which was actually written to teach creative artistic people about how music actually works rather than how aristocratic European music notation works.

Thank you! I've long been suspicious of so-called music theory for exactly the reasons you outline. I always wanted to learn more about music and why it "works" as you say, so I assumed what I wanted to learn was music theory, but then when I tried to learn it it didn't seem to explain any of the things I was looking for. I'll have to check out that book.

>Thank you! I've long been suspicious of so-called music theory for exactly the reasons you outline. I always wanted to learn more about music and why it "works" as you say, so I assumed what I wanted to learn was music theory, but then when I tried to learn it it didn't seem to explain any of the things I was looking for. I'll have to check out that book.

The reasons he mentions are bogus -- simply his bias against a term in common use ("music theory") , only because "theory" is used with a different meaning in academic contexts.

That said, there's nothing to be suspicious of in music theory, and what's in TFA is not "so called music theory", it's actual music theory, the way every composer, musician, etc in the western world knows as such.

That it doesn't explain other aspects of music (psychological, physical etc) does not make it any less "music theory", nor does it make those other things "music theory".

I did mention I have a music degree, right? I teach music lessons for a living. So if you were implying that I'm a scientist who doesn't get the value of this notation-system stuff and am just being pedantic about the word "theory", you're totally wrong.

The reason I complain about the term is exactly because of the post above yours: people actually do think they are going to learn explanations about the nature music and they don't get that.

> A theory is something that explains an observed phenomenon.

That's equivocation. Nearly everything in this document is widely considered to be part of music theory.

I wanted to thank you for one of the best comments on HN I've ever read. Having studied music "theory" for years your comment was actually the first words that described what I thought about this topic, and that book recommendation is fantastic. I honestly haven't been this informed on this site in...ever. I'm serious.

Thank you.

James Tenney's thesis "META Meta / Hodos" is another great attempt at constructing a real music theory. https://www.scribd.com/doc/167561563/James-Tenney-Meta-Hodos

>Our first sensory impression as we float in our mother's womb is the rhythmic sound of her heart.



It's pretty dark in there.

>This isn't "music theory" (although almost everyone thinks it is).

Isn't that how language works though? If everyone calls this music theory aside from a small minority of people, then it is music theory. "Music Theory" as a term is a study on it's own, it's not "the theory of music" and you can't really use the terms interchangeably like that.

> Isn't that how language works though? If everyone calls this music theory aside from a small minority of people, then it is music theory.

That's not really how language works in specialist fields though. For example "theory" and "proof" mean different things in science than they do to the general public (which is part of the reason why the general public misunderstand science so much). Or, for a more relevant example, the problems with how the term "hacker" is misunderstood.

Thank you for this comment, seriously. I hear a lot of producers talk about the importance of music theory, but I've never actually seen anything useful. All it is is notation. Notation by itself is meaningless and does not make anybody a better musician.

I would certainly consider species counterpoint and harmonic analysis to be "theory", although the first sections are clearly notation and not theory.

I would love to see a version of this with European (/rest of the world?) terminology. E.g. Crotchet, quaver, minim instead of quarter note, eigth note, half note etc.

I recall once seeing a link on HN for music theory for physicists and engineers, but I didn't bookmark the URI at the time - anyone recall this, and how to find it?

To those of you claiming either that "this is not a theory of music" or "yes, this is exactly what people mean when they say music theory," you are both right and both wrong.

The content of these cheat sheets describes what a music major learns in the first semester or two of music theory classes. But there are people who get PhD's in music theory. What do you think they study for all those years? It's not this.

Saying this is music theory is like saying arithmetic is math. It's not wrong, exactly, but it's leaving out a lot. People spend the first 12 years or more of their math education learning the notation, learning formulas, at least partly because you need to know that stuff in order to do any real math (we'll leave the discussion of whether this is the optimal pedagogy for another day, shall we?). You could say, "but people googling math want to understand why 2+2=4," but if you don't know that 2+2=4, there's not much point in trying to learn why. You could also say that what people learned in high school is what most people call math, but we all know that what people learn in high school math doesn't have much to do with what real mathematicians do.

"Real music theory" uses the contents of these cheat sheets (and many, many other things) to describe pieces of music and to try to understand things like: why did this composer choose the notes and rhythms they did? What can we use to justify the subjective observation that a composer or time period had a unified music style? What are the assumptions of musical language that govern musical composition without necessarily being known to the composers? Why does this collection of notes have meaning? What is that meaning? What does the content of this music tell us about how it should be performed?

The best of my music theory classes tried to keep a clear line from what we were learning to the kinds of questions above. The worst classes we spent a lot of time doing worksheets with the musical equivalent of times tables and algebraic equations, ad nauseam.


I'd also like to note that the things that appear under the aegis of introductory music theory are a strange hybrid of techniques, reflecting the multiple purposes of music theory for theorists, musicologists, composers, and practitioners. So you get things that don't help much with analysis, like species counterpoint (probably a useful exercise for composers, and maybe for musicologists since it is really a historical curiosity, but not really useful for anyone else) or realising figured bass (really only necessary if you're going to be performing baroque music in period style on a keyboard instrument).

You could again say similar things about high school math of course.

My PhD is in musicology, so I've spent a few hundred hours in music theory courses (the differences between music theory, music history, musicology, and ethnomusicology are shrinking, but that's a whole other topic). I can confirm that this is the kind of stuff that we teach in 1st or 2nd year music theory classes. But yeah, this isn't what graduate students study. This is the kind of thing that can help you learn the language of music scholars.

If you want to see a small sample of what music theorists find interesting nowadays, just look a the latest issues of some of the major journals to get an idea. For example:

Journal of Music Theory: https://www.dukeupress.edu/Journal-of-Music-Theory/ Music Theory Spectrum: http://mts.oxfordjournals.org/content/current Perspectives of New Music: http://www.perspectivesofnewmusic.org/ 19th-Century Music: http://www.jstor.org/action/showPublication?journalCode=19th... Journal of the American Musicological Society: http://www.jstor.org/action/showPublication?journalCode=jame...

I appreciate your attempt at balance, but I still fundamentally disagree. Math is universal. The arithmetic and other math basics we learn in grade school is both (A) not very culturally-biased and (B) often not taught well.

There is so much cultural baggage and assumption in the notation-focused "music theory" that is not present in mathematics. When we talk about numbers of things in math, there's no doubt we're describing something objective about reality external to culture. There indeed exists two apples or four apples. The concept of "four notes" is rarely objectively valid because almost everything in music is subjective.

"Music theory" basics are much more analogous to the elementary-school versions of linguistic grammar. And most elementary teachers' understanding of grammar is actually full of claims that are wrong, even though there's something to the gist of it. When we go to "English" class, we expect to learn about English and don't think our lessons apply to all languages. We should similarly not teach one particular language of music and call the lessons "music" without qualification. We should further recognize that grade-school definitions of nouns and verbs do not all hold up under scrutiny.

When "music theory" of the sort here is taught well, it is about as valid and useful as grade-school pedantic, prescriptivist, grammar — the sort that treats minority dialects as "bad grammar". Insisting on music theory being this notation-jargon from classical Europe is like saying that African-American dialects have poor grammar, when actually there are careful and consistent implicit rules in those dialects that sometimes carry useful meaning that standard American English doesn't even have a way to achieve.

I actually don't think we completely disagree here.

First, I definitely didn't intend the parallels with math to extend any further than the problems of perception with regard to how they are taught versus how they are practiced by scholars. Music theory is an art-related discipline and very much culturally inflected.

And this makes sense. Most music theory (in the West at least?) does focus on Western classical music, in large part because Western music has a literate tradition (see Taruskin) that most other musical traditions lack, which allows us to treat the compositions themselves as texts/objects more easily than other musics.

Even within Western music, we tend to use style- (read culture-) specific analytical tools. It doesn't make sense to use harmonic analysis for Stravinsky, and it doesn't make sense to look for twelve-tone rows in Mozart. In general we use a combination of language that composers used and thought with (so for Mozart, think keys and counterpoint) and techniques that arose from observing the style after the fact (in this case, for instance, sonata form, or Schenkerian analysis).

All of these techniques are culturally biased because, as you've pointed out several times, music is psychological, and therefore rooted in culture. Meanwhile, it makes sense to analyze, for instance, Indian classical music using rhythmic modes.

So, coming around to your analogy with elementary school grammar education, I think you're right that there is a prescriptivist parallel here -- learning that parallel fifths are not allowed, or that the second theme in a sonata always modulates to the dominant, or even that there always is a second theme in sonata, all of these do not accurately describe all music, or even all Western music of the common practice period. They are artifacts of the particular techniques being used. The original description of the sonata came into being around the same time as prescriptivist grammar (and has had a great deal of improvement since then -- read Rosen if you haven't already). Species counterpoint and Bach-style chorale harmonizing came about to describe specific styles, and to be used as exercises for composition students in the 18th-19th Centuries. The fact that those two things in particular (which I think are the most grievous examples of what you're talking about) are still included in music theory education is possibly counterproductive, especially since they are treated as sets of rules, rather than as historical examples of the composition pedagogy that the Western classical composers themselves were subjected to. Teaching Fux as anything other than "this is how Mozart and Beethoven learned to write counterpoint" is silly.

So that's the prescriptivist parts. However, where we disagree is that I think there's still a lot here that's useful for descriptive purposes. It's difficult to talk about classical music within its cultural-intellectual context, which I think is necessary, without these tools. You can argue that any interval can be dissonant with the right timbre, but if those timbres are rarely used in the repertoire you're describing, it's not much use is it? The 18th and 19th Century composers thought about dissonance a certain way, and if you're analyzing their work as art objects (cultural objects) rather than physical objects, you have to at least take that into account.

The problem I have with trying to casually learn music theory is that I cannot separate what is "real" and what is "syntax". For example, I can trivially see that "beaming" is entirely syntactic, but the difference between 3/4 time and 6/8 time is far more confusing.

Music notation just seems really really suboptimal for actually understanding music (music software uses totally different notations/interfaces for a reason), and all music theory seems obsessed with it.

In fact, beaming makes the difference between 3/4 and 6/8 clear. But even if it didn't, you can't divorce the cultural object from it's notation. Once you're trained, traditional music notation is pretty optimal for reading music for performance, and you have to remember that that is the purpose of music notation. All alternative notation systems I've seen fail this simple test: once I've learned this, would I be able to play music I've never seen before reasonably well the first time? I've never seen anything that works as well as traditional music notation for that.

Meanwhile, music theorists are "obsessed" with notation only insofar as the music they study is written using that notation. You seem to have skipped my previous comments -- music theory, even at this basic level, has little to do with notation. You read it as notation obsessed only because you're not thoroughly enough steeped in the notation. Reading and writing music notation is as natural to a musician as writing in English is to you. It would be strange and counterproductive to insist on using nontraditional notation when every musician who plays the repertoire you're studying already reads the standard notation, and all the repertoire you're studying is written in it.

I'm curious, what do people who get PhDs in music theory study for all those years?

They analyze lots of music and try to make connections within and between them, using a wide variety of analytical techniques. If you're actually asking and not being sarcastic, the sibling post links to several journals that will give you a good idea what theorists tend to work on.

When I was in school it seemed like most of the theory PhDs would pick a composer to focus on, but there are other ways to go than just that. Certainly, the sorts of questions I listed (asked in much more specific ways) are part of analytical inquiry, but a lot of theory papers are descriptive rather than explanatory.

Edit: I was performance major, but took classes from some pretty great music theorists. A few examples of graduate-level classes I took: the string quartets of Bartok; the music of Stravinsky; the symphony in the 19th Century. They tended to be focused on an individual composer or repertoire. Obviously scholarship is done on a much more granular level than that.

Nobody has mentioned hooktheory.com yet. If you're interested in why some music sounds good, check it out.

I'll second this even though I only checked out the site once a bit back. I'm still quite the newb at this stuff but it got a great response on the EDM production subreddit.

Down for me. Here's a wayback machine cached version: https://web.archive.org/web/20140226025639/http://tobyrush.c...

It appears that the site is somewhat image-heavy, hence the downtime. There are ~50 pages of content, each presented as an image link to a PDF.


This successfully gives you links that fail. I was hopeful that Google caches the PDFs, so that the links might not fail (with a bit of finagling) but apparently the cached PDFs are modified to be unreadable. :(


I'm surprised nobody mentioned that terrible font... Any chance you could fix it?

Hacker News hug of death? I am new to guitar and was excited to see what this was.

You might want to just learn tablature for the time being, unless you're thinking of doing classical guitar.

Not to say don't read them of course; you can read what you like, just that tab is what you'll be using I would have thought.

Guitar professional here.

It's good to keep the limitations of tabs in mind. The're really useful, but they don't represent music nearly half as much as conventional music notation. They represent finger movements instead.

The funny thing is that standard notation is quite poor for the guitar because it has no notion of finger movements- all sorts of annotations are dumped into scores - numbers for fingers, circled numbers for strings, roman numerals for frets, empty noteheads for natural harmonics, and who knows what for artificial harmonics and friends.

I don't think its possible to reproduce a piece from only tabs (without having heard it) while standard notation makes sight reading unnecessarily hard.

I feel like I'm repeating myself too much in this thread, haha, but only because it seems like no one is acknowledging that you can combine both forms very easily and produce very clear results. You just replace the staff notation with tabs, but keep everything else.

I took a quick screenshot to show an example[1]. I think here you can tell very easily what fingerings to play, what the note duration is, and when you should sustain notes over others. The difference between slurs, slides and whammy bar bends (The V symbol you see next to notes like in meas 3), muting notes is an x, yadda, yadda. I'll stop going on about tabs now.

[1] http://i.imgur.com/yog4v5n.png

If you just want to know where you put your fingers to make the same sound as some other guitar made in a recording, tabs are fairly reasonable. If you want to know what's going on musically, staff notation is much better. Also, if you want to be able to play melodies from fake sheets, or arrange music originally written for another instrument, you'll need to know staff notation.

I know countless guitarists, myself included, for which sight reading is not hard at all, because we've practiced this for hundreds of hours or more. All the things you described as making this difficult simply become second nature with practice.

Standard notation can help build a mental representation of how the music sounds. With some training, one can process the standardised fingering annotations into physical movement really fast too.

Tabs are physical movement instructions. With a lot of training, some people can process this into a limited mental representation of how the music sounds.

If you think about it, it's quite analog to the difference between vector and bitmap images.

Tabs don't have to be just hand-scrawled notes or ASCII text anymore. With decent notation software, you can mix the notations. I don't think there's much that can be written in conventional music notation that can't be displayed in Guitar Pro (or Tux Guitar for an OSS alternative). Especially since it allows you to toggle between the two forms, or view them on top of each other.[1] You can also add rhythm and articulation notation to just tabs to make it more clear what they mean as well.[2]

I think it's a great way to begin learning guitar without needing to take a lot of time to learn how to read sheet music first, as that really scares away a lot of new players, or people who are considering learning guitar. Notation for rhythm, slurs, accents, are all pretty clear and can be learned very easily. Unless there's more that I'm not understanding about the differences between the two, I'd like to learn what it is.

There's one last benefit that really makes me prefer digital tabs over conventional notation. There are a lot of songs/riffs/etc that are very dependent on the effects being used. This is more a benefit of software than tabs, because you could do this with any instrument or notation, but support for it is more common in guitar software. Being able to add effects, and mark things such as the rhythm of a wah-pedal can be very helpful as well. An example I can remember is Searching by Joe Satriani. He rocks a whammy pedal back and forth while playing an arpeggio, and I had no idea how to mimic that sound until I saw both the notes written out, and the whammy bends on the same track as well.



edit: I still think it's useful to learn theory, tabs or not. But I don't think sheet music and theory need to go hand-in-hand anymore.

Granted, I'm a bassist, and I appreciate that guitar is its own beast. Bassists also debate the merits of tabs versus notation. I used to be adamant that everybody should learn to read, but have softened my views, especially in light of what you say, that reading tends to discourage beginners.

Instead, my more moderate view is that at the very least, new players should be informed of what the most common notation systems are like, and what they are used for. We can argue all day, but it's often someone else who decides how they want to notate something. Different genres have their own preferences, and there are even regional variations. I played with a bandleader who had moved up from the South, and had to re-write her book because nobody in this region can read Nashville number charts.

I play in a jazz band, and the guitarist and I would both be helpless if we couldn't read standard notation -- dots and chord symbols. But I would add that it's a musical genre where the music is pretty forgiving of the player deciding their own fingerings on the fly.

Tabs imply a certain fingering, which may or may not be optimal for the music - is there any way around that issue with those programs?

You can always switch to notation mode and work out the fingering for yourself, just like you normally would. The programs can also allow you to move notes between strings, and it will shift the tab so the note played is the same, even in alternate tunings, so you can quickly try different positions on the neck. I also know Guitar Pro has a feature where you can enter notation or tabs and it will attempt to work out the best fingering for you. I don't know how well this feature works, though.

If something is tabbed inefficiently, that's a problem with the author of the tabs, not the format. I'd say incorrect tablature is similar to incorrect notation. Just because someone can notate music incorrectly doesn't mean that sheet music is a poor format.

The fact that tabs imply a certain fingering is a larger benefit than an issue. Many riffs are only realistically playable with one specific fingering, and a tab can help you figure this out much more quickly than you could with your ear or notation alone. Especially when it comes to deciphering whether something is using techniques that can produce the same notes, but different sounds or ease of playing. Like sweeping as opposed to tapping, or songs written in alternate tunings.

Not to mention that the SOUND matters too, depending on the fingering. E.g. A high D played on the (plain) 2nd string sounds different than the same pitch played on the wound 3rd string which sounds different than the wound 4th string played up above the 12th fret, etc.

Circle of fifths and scales are very relevant for guitar though, even if using tablature. And it would be very useful to learn basic treble clef reading to better understand chords and scales.

There's not a lot here about the different blues scales you're more likely to have use for though. The document even warns against the tritone.

> The augmentation dot is a dot placed to the right of a notehead. Though small, this dot wields some serious power: it changes the length of the note by 150%. In other words, it makes the note half again as long.


Very nifty and fun though.


I would like to recommend :

Music Theory For Computer Musicians.

I appreciate the style of presentation, but why do we still pretend this is "music" theory? Most of this is completely useless for anyone who is interested in understanding music that wasn't written by people in powdered wigs.

For some actually useful music theory, I can recommend the goldmine that is Ethan Hein's blog: http://www.ethanhein.com/wp/category/music/music-theory/

Listen to techno, house and tribal, and you hear Koyaanisqatsi, 18 Musicians, and the work of Glass, Reich, and Riley.

Listen to Glass, Reich, and Riley - and you hear the deconstruction of symphonic structure.

Listen to Beethoven's symphonies and you hear the struggle of form from Mozart's classical era construction to what would become Beethoven's pure romanticism.

Listen to the interplay of an Alberti Bass and melody in Mozart's Piano Sonatas, and you hear the beauty in the evolution of counterpoint as written by Bach, Scarlatti, and their band of powdered wigs.

Listen to Bach and Scarlatti, and you hear the echoes of Gregorian Chant and the Diabolus in Musica (and why Black Sabbath sounds like Black Sabbath.)

Those who forget the past are doomed to be Miley Cyrus.

There's even plenty of music theory to learn with Miley Cyrus, particularly if you want to know about just intonation and harmonic theory. Even the simplest and most ubiquitous pop chord progressions with just I IV and V require a fair amount of harmonic theory to reason about.

A number of more talented musicians than myself have also remarked that there is actually a lot of weird stuff going in top 40 music. It's so bland and repetitive that they have to do something different for it to be memorable; it's just not the focus of the song.

That's pretty much the definition of pop music, or at least good pop music.

>Listen to techno, house and tribal, and you hear Koyaanisqatsi, 18 Musicians, and the work of Glass, Reich, and Riley.

No, you hear Koyaanisqatsi, etc.

And this is exactly why you should never try to understand a popular genre by crossreferencing its stylings against some other style or genre.

Because other people hear other things - many, many other things - which perhaps you're not aware of.

I get your point, and of course everybody hears something different. That's music!

But this particular combination is like saying, "Listen to Billy Joel and you hear piano." ;) House and techno really are direct lineage from the minimalists in many ways. This is not a blind cross-reference. Any artist learns and borrows from what has come before. Some borrow more heavily than others. (As an aside, check out the 2006 album "Reich Remixed" for some really cool interpretations... )

Suppose I am listening to Les Claypool's bass work. What olden stuff am I hearing?

Listen to Les, and hear the technique and round sound of Jaco Pastorius. (or at least I do.)

Listen to Portrait of Tracy (or any great harmonic jazz) and you hear the same open chord structures of Impressionist works like Debussy's Preludes or Ravel's Valse Nobles.

Listen to the impressionists and you go many places...

Ravel and Debussy's piano music is barely one step removed from jazz. Makes sense, because the early jazz greats lived and played around the same time. There's a famous picture of Ravel at a birthday party when he was in his 50s, with George Gershwin standing beside him. The early 1900s was an amazing time in music... classical was mixing with early jazz was mixing with vaudeville was mixing with the first stirrings of delta blues was mixing with gospel was mixing with african folks songs...

(I can do this all night... ;)

Progressions are old as dirt. Pretty much all solos are melodic arrangements which fit over an underlying chord progression. Bassicaly (ahahhahahaha), if you broke it down enough, you would be able to "strum" a Les Claypool bass solo.

I see what you did there, TOBy Rush!

Seriously, I make techno and am intensely interested in music theory. You might as well complain about mathematical cheat sheets containing arithmetic, geometry, and other outdated concepts that were cooked up by ancient Greeks wrapped in sheets.

If you're a techno composer - check out some of the great minimalist and musique concrete composers, then use the intertubez and read about how they got to their musical vision. Their take on theory and how they warp/deconstruct traditional notions of structure is something I learn from all the time...

Some of my favorites: Steve Reich - Music for 18 Musicians & New York Counterpoint Phillip Glass - North Star Morton Subotnick - A Triumph of Reason Iannis Xenakis - Pleiades, Musique Concrete 1959


I completely disagree.

For one thing, a lot of what one learns studying Common Practice music theory most definitely is applicable to modern styles. Things like dominant function, for example, aren't going anywhere.

That said, I agree with the point that is implied by your post, namely, that "music theory" as the term is usually used is actually a description of a specific period of music (Common Practice) and it is a mistake to view modern popular music as some kind of degenerate case of it. I suggest that people who are interested in learning a theory that applies to more recent music take a look at "What to Listen For in Rock" by theorist Ken Stephenson. http://yalepress.yale.edu/book.asp?isbn=9780300092394

I also don't think the link you sent is going to make things much clearer for a beginner. I think that Desi Serna's guitar theory lessons in podcast form actually do a great job of explaining the basics, even if the guitar isn't one's primary instrument. https://itunes.apple.com/us/podcast/guitar-music-theory-less...

One of the most enlightening things I learned while studying music in college was discovering why classical harmony ended up the way it did. What makes a V7 chord work? Where did augmented 6th chords come from? How did the Neapolitan chord fit into these systems? By looking at the progression of music history from one-voice chants to counterpoint to harmony and then to the explosion of ideas in the Romantic era and beyond, you acquire the skills to know why music — any music — can sound good and express rhetorical ideas.

With a foundation in classical music theory, you can go beyond what anyone is doing in popular music today. I strongly believe that (for example) Beethoven's 5th, from a compositional perspective, has more rhetorical power than any music produced in the last century. I think this is largely due to popular music being very simple structurally. Where is the counterpoint? Where are the modulations? All we get is simple chord progressions repeated ad nauseum. Imagine how much farther we could go if we combined the lyrical and textural variety of popular music with the compositional depth of classical music! I salivate at the thought of an electronic or metal musician[1] delving into this territory...

[1]: Phish is one of the few bands I know of who actually focus on this stuff. Some of their pieces even include bona-fide fugues. I'm not the biggest fan, and it's a simple example, but I love listening to the motifs and variations in "Stash": https://www.youtube.com/watch?v=8OfQCAj2Ppg

>Where is the counterpoint? Where are the modulations?

In the sound design, the rhythms, and the production, where they never trouble the hearing of anyone who has been trained to believe that musical structures are defined almost entirely by pitch patterns.

Even the idea that music exists to 'express rhetorical ideas' is... actually quite odd, outside of classical training, anyway.

I happen to like fugues very, very much. (The fugue in BWV 542 is a favourite - http://www.youtube.com/watch?v=g6Kul3JzKI0.)

But the idea that music should aim for that always? Uh, nope.

I never said "always"! But it would be nice to have sometimes. It's an area that's almost completely unexplored in popular music.

So, doing things that don't scales?

Out of curiosity, have you studied just intonation? I'm currently reading the mostly excellent Haromonic Experience [0]. I'm reasonably well educated on music theory, but I had never looked into just intonation and the way frequency relates to our perception of pitch and harmony. I'm finding this stuff enlightening.

An ideal string (or any oscillator) when disturbed (e.g. plucked) will tend to vibrate at a fundamental frequency of x Hz as well as modes of 2x, 3x, 4x, etc. These higher frequencies at integer multiples of the fundamental frequency are called harmonic partials. This is a physical phenomena unrelated to music and human hearing.

But the basic elements of harmony come from the fact that our auditory system seems to be "tuned" to identify harmonic partials.

If we start with a fundamental frequency x Hz (let's call it C), the first partial is 2x Hz, and it is another C an octave above (multiplying or dividing a pitch by a power of 2 will give you the same pitch in another octave).

The next partial is 3x Hz, and it sounds like a G. This is the interval called a perfect fifth, and is the strongest, most stable sounding interval (other than the octave). The next partial is 4x Hz, which is just another C, two octaves above the C we started with.

The next partial is 5x Hz, which sounds like an E. This is a major third, which is another strong and stable interval which is ubiquitous in most music.

[0] http://www.amazon.com/gp/product/0892815604/ref=oh_aui_detai...

Yeah, a little bit! What's interesting is that we're so used to equal temperament (i.e. a bit of error added to most intervals in exchange for greater flexibility) that pieces played in just intonation sound "off" to us, even though they're more mathematically correct.

Also, I believe instruments with arbitrary pitch (violin, voice) tend to naturally drift closer to just intonation.

I've been trying to train myself to hear the JI resonances (Harmonic Experience has tons of exercises for this). I can hear the sharpness of the ET major third and the flatness of the ET minor third fairly easily.

Barbershop quartet music (and other a capella music) is a great place to hear just intonation. The consonant barbershop seventh chord was a relevation to me when I first learned of it. It's really close to the dissonant dominant 7th interval, but serves a very different purpose harmonically. The minor seventh is also a different interval. Thus there are actually 3 seventh intervals, all of which are approximated in ET by a single note. This blog post has a good summary:


I feel the same way about gaining insight from studying the history of music theory, but I have to disagree with a couple of points, too:

1. Studying the history of Western music theory doesn't give you insight into any music, as you claim. It gives you insight into Western music. Check out music from another part of the world with a completely different notion of tuning, different scales, etc, and your music theory background isn't going to help very much. The only thing that is truly universal is the overtone series, because that's based on physics.

2. I think we could come up with plenty of structurally complex music from recent idioms, particularly jazz and jazz-influenced music. I would say that a bigger difference between something like, I dunno, Steely Dan, and Beethoven's Fifth, is length. Popular musicians don't seem to attempt such longer pieces. But for me personally, longer != more powerful. For that matter, more complex != more powerful, but I happen to share your desire for what we're calling "compositional depth", whatever that may be...

Classical music theory is a rigid, structural, almost mathematical system that emerged over several centuries. In those centuries, composers learned some very powerful techniques to express ideas in their music. Although the specifics of classical music — the rhythms, the chords, the tonalities — may be antiquated, I really do think that the forces that guided their development are just as relevant to modern music. For example: why is a V7 chord so powerful? It's because you have the interval of a 5th — a tense interval with almost universal meaning vis a vis the overtone series — combined with two minor 2nd intervals that are brimming with tension to resolve to the tonic and the third. The study of intervals and of tension in voice leading is, I think, absolutely essential in jazz as well, and can help explain some extremely confusing jazz chords in analysis as almost incidental results of voice leading. This is what I mean by classical music theory offering insight into any genre of music: classical music emerged out of the study of things like intervalic tension, and even though our own century's music has diverged from this origin, I think it's very important to study the lessons learned over those hundreds of years.

(EDIT: I just noticed that you said outside of Western music. Yes, that may be correct. Many systems of music in the world are very different from our own. But I think most people writing music today are writing with Western ears in mind. After all, we've been conditioned to understand this kind of music for hundreds of years!)

As for your second point, I think what I latch on to in classical music more than anything is the use of motif, and Beethoven's 5th of course epitomizes this. Don't get me wrong — I love popular music, and I listen to it more than anything — but most of the "top 100" songs in any popular music ranking feature wandering, improvised music more than any systemic use of motif to make a rhetorical point. And that's not a bad thing! I love Stairway to Heaven. But it just "sounds good" — it doesn't really develop its musical material.

Compare to something like this[1], where every single musical detail is repeated and developed across all the voices. It's so intricate. I've never heard this level of motivic development in popular music, and I've been looking for a few years now.

I am sorry if this is not making any sense, I am having beers.

[1]: https://www.youtube.com/watch?v=hWLotW5AKjg

You're making plenty of sense, beers or no

I would say that the V7 gets it's power from the flatted fifth interval, aka the tritone, aka the most dissonant interval this side of a minor second. And like you say, it can be understood in terms of voice leading, e.g.the 7th (the "leading tone"!) "wants" to resolve up to the tonic.

It's interesting that you mentioned motif because I think that fits well with what I was saying about the length of the pieces. Seems to me that one isn't going to have much room to develop motifs unless writing in an extended form. But for me personally, this doesn't seem like a good requirement for a piece to be satisfying in the way that I think we are talking about. If you are seeking that kind of intricacy from modern music, have you checked out Bela Fleck & the Flecktones? Also, have you explored Zappa at all?

I mean, there are many ways to be satisfying in music. I am very satisfied by the exploration of texture and improvisation in modern music, for instance, and I am very grateful that we live in such an exciting time for musical experimentation. But at the same time, in a way that's hard to describe, the moment where the theme returns in the third movement in Beethoven's 5th[1] just absolutely trumps any musical moment I've heard in popular music. I'm talking Schindler's List level goosebumps. There are plenty of long pieces in popular music, too — prog is big on that. But the techniques of Beethoven and ilk just aren't getting used, which I think is a great shame. I don't think it's "magic" or "genius". I really do think it's a matter of harnessing the medium to its full potential, which might require more education in music theory than (in my experience) many musicians feel comfortable with.

I don't think you need a long piece to develop motif. Look at something like "Vocalise"[2], or, heck, any of Bach's Well-Tempered Klavier.

One of my teachers used to stress the original meaning of the word "sublime". I think that's what's missing in modern music. A feeling of awe that leaves you speechless.

Zappa's been on my to-do list for a while, I'll have to give him a listen soon!

[1]: https://www.youtube.com/watch?v=jdvsRJL4Mtw

[2]: https://www.youtube.com/watch?v=iBVkYGLEUpg

There are things that are useful to the extent that they're not inaccurate, but most of it is very incomplete with regards to how most of the music most people enjoy today works. Recorded music is not mentioned, blues tonality is not mentioned, there is no theory of groove or syncopation, and there's nothing about lyrics. Without these things, "music theory" is not useful for studying modern popular music.

Your book recommendation looks quite interesting too, thank you. I'm not sure if Ethan Hein's blog is a great starting place for a beginner, but it's some of the best "theorizing" around recorded music that I know of.

The blog you linked to is full of conventional music theoretical discussion such as:

"The blues is based around a very different set of harmonic expectations than the ones underlying classical music. In the blues, major and minor tonality are freely intermingled. Dominant seventh chords can function as tonics. Tritones need not resolve."

Although the blues uses devices not commonly found in classical music, the language being used to discuss it is conventional, main stream music theory: "major", "minor", "tonality", "tritone", "dominant", ... No surprises!

Someone looking for basic music theory will not get into this at all.

Secondly, parallel major and minor does occur in classical music. So do dominant sevenths functioning as tonics: at least secondary tonics.

Moreover the people who came up with the blues tradition (and jazz) were not completely isolated from hearing classical music.

Lastly, there is orchestral music that doesn't consist of "powdered wig" scales conforming to cycle of fifths progressions. Let's see what comes to mind: Moussorgsky's "Night on Bald Mountain", or Ravel's Boléro, ...

"not completely isolated from hearing classical music"

Many of the most important jazz composers and musicians were classically trained and were intimate with the type of music theory we're talking about here. Thelonius Monk could reel off Chopin and Rachmaninoff etudes from memory. "Take the A Train" and many other tunes we associate with Ellington were written by Billy Strayhorn, who's main interest was classical music, but found that the classical music establishment at the time would not accept a black composer or musician. Jazz was the only genre in which he was allowed to work. (Ellington/Strayhorn and Monk are considered the most important composers in Jazz, and their initial training was in European classical music.)

Jazz is a development of Western classical music, and knows it. In fact, it's a rather miraculous revival of the improvisatory tradition that was central to European art music until it died out with the rise of symphonic music in the 19th century. If you listen to some contemporary jazz pianists, for example, Marcus Roberts, there's no mistaking the strong influence of European classical music.

> Jazz is a development of Western classical music

only of Western classical music?

Wait, what? You're claiming that basic intervals and chord are useless for understanding music that wasn't written by people in powdered wigs? I'm very confused. You need that to understand playing Brown Eyed Girl on an acoustic guitar. In fact, Hein's blog covers much of the same material.

a somewhat more structured introduction with exercises and examples to boot: http://www.musictheory.net/

+1 for this actual intro. The OP is more like Cliff's Notes.

The guide I usually see people point to is RavenSpiral's https://www.scribd.com/doc/5220863/Ravenspiral-Guide-to-Musi...


I found the PDF using Google: https://www.google.com/#q=Ravenspiral+Guide+to+Music+Theory+... (Scripd.com is annoying, like a PDF download scam)

That's very good indeed, and much closer to my personal idea of a useful theory guide.

Thanks for posting the link to Hein's blog which I was previously unaware of.


Especially of interest.

Understanding real music theory (as in the OP and musictheory.net exercises) will allow you to understand folk and popular music, as well.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact