Theory of tonality

In this article you are about to discover a different approach to the tonal music theory which will make learning musical principles about 50 times easier. I claim that what would take you weeks or months using the traditional approach will reduce to mere hours using my system. By using my method, you will be able to play any chord progression in any key with only very little need of actual learning. This may sound impertinent, but you can very easily verify that yourself, since this wouldn’t take you much time. I don’t suppose any preliminary knowledge, except that you are familiar with the names of intervals (such as perfect fifth or major third), most common scales (such as major and minor) and that there is something called “chord progressions”.

If you are used to call notes A, B, C, …, G, you can easily fall into the illusion that classical (and most of the modern) western music is based on whole steps and half steps, i.e. that you build scales (such as a major scale) by successively adding whole steps or half steps. It is not quite so; in fact, western music is based on perfect fifths. Let’s start with the note F; what note is the perfect fifth above it? It’s C. And above it? G. And so on… By stacking fifths, you get the following sequence:

F C G D A E B

Did you notice anything interesting? By starting at F and adding perfect fifths, we got all the notes of the C major (or A minor) scale. This is not a coincidence. Any major scale can be built so that you take one fifth below the tonal center (F below C in this case) and five fifths above the tonal center (G D A E B in this case).

So, to form a G scale, we would simply start with G and include one fifth below it and five above. However, until now, we haven’t given any name to the perfect fifth above B, so lets extend our “vocabulary” first. The traditional approach is to start anew and call it F again, but with a sharp sign. The next is called again C, but again with a sharp sign, and so on:

F C G D A E B F♯ C♯ G♯ D♯ A♯ E♯ B♯

What should we do with the left side? Let’s call the perfect fifth below F B♭, then E♭ and so on (we now repeat the same sequence from the right):

F♭ C♭ G♭ D♭ A♭ E♭ B♭ F C G D A E B F♯ C♯ G♯ D♯ A♯ E♯ B♯

The names we’ve chosen (using sharps and flats) may seem a little ad hoc, but the logic behind this naming will be explained later. If you don’t believe me that these all are indeed fifths, you can convince yourself by playing the notes on your instrument. What should come after B♯ and before F♭? Another copy of F C G D A E B with something called a double sharp sign (x) and a double flat sign (♭♭):

… E♭♭ B♭♭ F♭ C♭ G♭ D♭ A♭ E♭ B♭ F C G D A E B F♯ C♯ G♯ D♯ A♯ E♯ B♯ Fx Cx …

So, we know how to call the notes, but what is this all good for? As I have said, a major scale can be formed by taking one fifth below to tonal center (by “tonal center” I mean the “base” note of the scale, i.e. C in C major) and five fifths above it. So what happens if we start with G as our tonal center? By looking at the line above, you see that one fifth below it is C, five fifths above it are the notes D A E B F♯. Put together, the G major scale consists of

C G D A E B F♯

Similarly, if we start with, say, A, we obtain

D A E B F♯ C♯ G♯

Now, if I tell you to tell me the notes of A♭ major, you shouldn’t hesitate; just look at the line and read seven notes starting with the one on the left of A♭:

D♭ A♭ E♭ B♭ F C G

This itself may be quite useful, but we still did not get to the real purpose, namely that chord progressions are determined by their relative position to the tonal center in the line of fifths. For example, C – F – G is what would traditionally be called the I – IV – V progression (or tonic, subdominant, dominant, and this is the most common progression in western classical music). Notice that F is on the left of C, G on the right. The same pattern applies to any other key; for example, in the key of G the same progression would be G – C – D (by “the same” I mean it would sound completely the same, just a little higher or lower). In the key of A♭? The same! One left, one right, or A♭ – D♭ – E♭ (just look at the line; the segment above suffices).

What about the I – vi – ii – V – I progression (also one of the most common chord progressions)? In C major, this is

C – Am – Dm – G – C

(m denotes that the chord is minor, since the chords at the vi and ii degrees of a major scale are minor). By looking at the line you can see that at first you actually jump three letters to the right and then return “stepwise” back to C. This has to be the same in all keys, so in G, this leads to

G – Em – Am – D – G

(Remember, you start with G in the line and jump three times to the right and then return “letterwise” back). In A♭:

A♭ – Fm – B♭m – E♭ – A♭

Cool, we know how to find these chord progressions in any key when we look at the line of fifths. The problem is… Usually when we play there’s no line of fifths to look at. How to solve this problem? Rename the notes.

The ultimate solution

This may sound harsh, but what I actually want to do now is to rename the notes (that is, to write different symbols instead of A, B, …, G). In fact, it isn’t such a big deal; it took me about half an hour to learn to name all the piano keys using the new names with certainty. So the rule is:

Instead of D I shall write 0, the perfect fifth above D (i.e. A) shall be denoted 1, the perfect fifth below (i.e. G) shall be denoted –1 and so on, so that our line of fifths becomes

… –7 –6 –5 –4 –3 –2 –1 0 1 2 3 4 5 6 7 …

or aligned with the previous line:

D♭ A♭ E♭ B♭ F C G D A E B F♯ C♯ G♯ D♯
–7 –6 –5 –4 –3 –2 –1 0 1 2 3 4 5 6 7

C = –2 in the new notation. What are the notes of C major?

–3 –2 –1 0 1 2 3

And what about, say, B = 3 major?

2 3 4 5 6 7 8

Naming notes of any key becomes trivial. But in fact, it is not the notes in a key what matters, the important thing is what is their function. Every note is determined its relative position to the tonal center. I shall denote this position by +n or –n, for example, 1 is +1 of 0, 5 is +2 of 3, –3 is –1 of –2 and so on. The tonal center itself will be denoted by +0. What does it mean in terms of chord progressions? For example, the progression I – IV – V can be written as +0 –1 +1. In 0-major this means 0 –1 1, in 3 major this means 3 2 4, in 7 major this means 7 6 8 etc. Minor chords and other types of chords can be indicated using m or another traditional symbols. For example, we can write

+0 +3m +2m +17 +0

for the usual I – vi – ii – V7 – I progression.

The point is that once you learn how to play chords using this naming system on your instrument, you know how to play any chord progression in any key. For example if you know how to play chords –10 major, –9 major, and so on, –10 minor, –9 minor and so on, on the piano (its actually quite simple; it took me only one evening to learn how to play chords from –10 to 10 (both major and minor) on piano instantly (although I am a guitarist and I hadn’t known the fingerings until that evening; if you are a pianist, you just have to remember where are the notes in this naming system)). For example, just after one evening of training with no previous knowledge, I was able to play the I – vi – ii – V progression (i.e. +0 +3m +2m +1) in all keys, because it’s trivial. Want to start in 2 major (E major)? Just play the chords 2 5 4 3 2 (or more precisely 2 5m 4m 3 2). In 4 major (F♯ major)? Play 4 7 6 5 4. In –6 major (A♭ major)? Play –6 –3 –4 –5 –6. How long would it take take to you to learn these progressions using the traditional letters? Just compare:

2 5 4 3 2 4 7 6 5 4 –6 –3 –4 –5 –6
E C♯ F♯ B E F♯ D♯ G♯ C♯ F♯ A♭ F B♭ E♭ A♭

If you don’t already remember them, the letters appear completely random. This is exactly the reason why many musicians are able to play well in one key and are completely lost in another key. Using numbers… there’s nothing to remember. Everyone can count to ten.

Melody and chromaticism (where are the sharps and flats?)

You may now say “Well, that’s all nice, harmonies are simple, but what about melodies? They must become utterly complicated to think about.” This, in fact, is not really the case. Lets start by explanation, what the flats and sharps really are: ♯ simply means +7, ♭ means –7. For example, F = –3, so F♯ = –3 + 7 = 4. Similarly, B = 3, so B♭ = 3 – 7 = –4. This explains the traditional notation. We can just write any value using A, B, …, G (that is, the values from –3 to 3) with enough sevens added or subtracted (notice that x (double sharp) is just +7+7, i.e. +14).

This explains also the staff notation. We chose the lines and spaces to denote notes from –3 to 3 in melodically ascending order. If we want to denote a note outside this range, we simply do that by adding or subtracting seven from a note (we write a sharp or flat sign in front of it). Luckily, this note is melodically always in the place where it should be, that is, F♯ lies between E and G. So, when you know the numbers of the notes on the staff, reading sharps and flats shouldn’t be a problem, –3♯ is just a way to write 4, –2♯ = 5 etc.

Have you ever been taught that flats tend to “move down” and sharps tend to “move up”? This has a deeper reason. There’s a rule that applies very often:

A tone outside the current scale tends to return to the scale chromatically (one half step up or down) to the closest scale tone. However, this “closest” is in the sense of the line of fifths, so, for example, in C major, F♯ = 4 wants to move to G = –1, not to F = –3, because G is harmonically closer. Other way to state this rule is that a tone moves either by +5 or by –5. For example, G♯ = 6 tends to 6 – 5 = 1 = A in C major, whereas G♭ = –8 tends to –8+5 = –3 = F.

Finally, return to the problem of melodies. How does, for example, A major look like written melodically? It’s

1 3 5 0 2 4 6

This is not really that hard if you know the whole/half-tone structure of major scales, namely that it is WWHWWWH. Whole tone up is +2, half tone up is (inside a scale) is –5. You can therefore read the scale as +2 +2 –5 +2 +2 +2 –5. This applies to any scale. If we choose C major, we get

–2 0 2 –3 –1 1 3 –2

This isn’t really harder than to learn A B C♯ D E F♯ G♯ A or C D E F G A B C.

Other types of scales

When you denote non-diatonic scales such as the whole tone scale, double-diminished scale and other exotic scales, the notation gets messy because it is based on the diatonic scale. However, it is only the notation that’s problematic, the line of fifths has no problem with such scales. For example, the whole tone scale is created by stacking whole steps, that is, +2. So C = –2 whole tone scale would be –2 0 2 4 6 8 (that is, C D E F♯ G♯ A♯). Similarly, double diminished scale is created by alternating whole steps and half steps. Half steps in the context of a scale are the closest possible, i.e. –5. For example D = 0 double diminished scale would be 0 2 –3 –1 –6 –4 –9 –7 (D E F G A♭ B♭ C♭ D♭). Notice that the letter D appears twice, but no such irregularity occurs in numbers.

We also haven’t discussed the minor scale until now, but it is also quite simply. By looking at the A minor scale, we can see that a minor scale is created by taking four fifths below the tonal center and two above, so, for example, G = –1 minor scale is –5 –4 –3 –2 –1 0 1 (i.e. E♭ B♭ F C G D A). Harmonic and melodic minor are exactly what is denoted in the traditional notation. For example, A melodic minor is A natural minor (i.e. –3 –2 –1 0 1 2 3) with F = –3 and G = –1 raised to F♯ = 4 and G♯ = 6, that is, the whole scale is –2 0 1 2 3 4 6. You can see that this scale has “gaps”; it doesn’t contain 1 and 5. The effect of these gaps is that this scale is not stable, it tends to return to the A natural minor scale.

Leave a Reply

6 Responses to “Theory of tonality”

  1. ole kirkeby says:

    The fifth below the tonal center is the fourth in major (F to C, for example), and that is usually called an ‘avoid note’ because it sounds like it doesn’t really belong to the scale when you play it on a strong beat. If you leave it out, you get a more logical sequence — root plus stack of five fifths — and that is indeed the safe scale I call hexatonic on m3guitar.com.

    Do you feel strongly about using D as the reference? If you use Ab instead, then the mapping is very close to my clock notation. Even hours are the same, odd hours are mirrored across the center, and we can then claim the two systems are variations on a common theme.

    Finally, here is a pedantic comment: I find the minus sign as a symbol confusing because I associate it with a movement down in pitch. In your system one semitone up is -5 (I noticed you suggest using +7 instead of -5 for a sharp) and three semitones up is -3. I would probably prefer a symbol that somehow indicates ‘anti-clockwise’ steps instead (underscore? a subscript?).

    • Kubaz says:

      I have also thought about six tone scales, although in a slightly different manner. However, the problem is that this would be incompatible with our staff notation, because flats and sharps would then have to mean +6 and -6, and therefore the linear order of the notes would be distorted. I have experimented with pentatonic staff notation (which still works well, because +5 and -5 are semitones) and which has an interesting property – major chords and minor chords have the same shape (represented by noteheads) as the patterns pressed by fingers on the fingerboard in the M3 tuning.

      Using D as the reference tone is probably the only logical choice, because the notation is symmetric about it. However, if you want to forget the traditional notation and think only about the twelve tones as written on a circle, then yes, Ab is a better choice, because then, for scales close to C major, you don’t have to go over the “gap” (from 11 to 0, resp. 12). Nevertheless, the difference from your clock notation is quite important. Here, every scale forms a continuous segment (for example from 3 to 9). In the clock notation, the tones are spread everywhere.

      Actually, for the twelve tone system, I have probably a better naming convention. Use three consonants (I use d, m, s) and four vowels (I use a, e, i, o). Call every note as a combination of these (da, de, di, do, ma, me, …). The tones can be arranged in such a way that cycling d – m – s – d means adding a major third, cycling a – e – i – o – a means adding a minor third. For example, “damame” is the “da major triad”, masase is the “ma major triad”, sidido is the “si major triad” etc. Similarly, dademe is the “da minor triad”, mamese is the “ma minor triad” and so on. After a little practice, naming fast all tones in any chord becomes quite trivial.

      As for your pedantic comment — I have developed a whole system of more convenient notation since I published this article here. I would probably like to assemble all of my ideas in a paper I am writing (including the “dadedido” naming convention, diagrams presented in the section Melodic Scales, cycle of fifths with Ab as the base and written using numbers, the pentatonic staff etc.), and then publish it somewhere. However, the paper is still not complete. When I finish it, I’ll let you know.

      • ole kirkeby says:

        Okay, if you want to hang on to the sharps and flats from traditional music notation I agree it makes more sense to use D as the reference. Compared to my clock notation it is then the even hours that are mirrored across the center. For some reason I find that hard to visualize.

        My preferred notation is a chromatic staff without any sharps and flats. I put in a post, “Modified Schoenberg notation for m3guitar”, on the Music Notation Project Forum in March 2010 (http://groups.google.com/group/musicnotation/browse_thread/thread/66bf04ae9d078fff, you need to log in to Google groups to view it). You might want to float your idea over there but be warned that the feedback you get is rather unpredictable. A few members are constantly promoting their own systems which is very tiresome. I didn’t find the dialog terribly stimulating and it is a while ago I have followed it. However, I recently found out that another mathematician, Andreas Griewank who is professor at a university in Berlin (http://www.math.hu-berlin.de/~griewank/neu2010/index.html), has written a paper on some aspects of notation and the major thirds tuning (http://www.matheon.de/preprints/7046_mathtune.pdf). I have exchanged emails with him, and he is keen to continue that type of work and possibly collaborate on it. Like you, he is also interested in abstract notation and how to make sightreading easier, as opposed to me who is mainly trying to come up with a notation system that works well for the one instrument I am playing. If you are considering to publish your system you might want to contact him.

        • Kubaz says:

          Thanks for the provided links, I’ll look at them. Unfortunately, the images in your post don’t exist, but I think I can imagine how it looks. I developed almost the same notation (just without the dashed lines) independently and actually transcribed the Moonlight sonata into it when I was learning it some while ago (probably every guitarist playing in the M3 tuning would develop this kind of staff after some experimentation, as it is so easy to read). However, the problem is that one has to be able to read the traditional notation as well. It’s like learning Chinese using Pinyin; this might be a great tool for you if you want just to speak, but if you want to be able to read and write Chinese, there’s no way around learning the characters, so its good to find a way to make this easier.

          As for the “hard to visualize” thing — the point is that this notation is not to be visualized, but to be computed. For example, the I – vi – ii – V – I progression is just 0 +3 +2 +1 0, which is trivial to translate into any scale (e.g. 3 6 5 4 3 in 3-major). There’s no need for visualization. What I somewhat don’t like are the negative numbers. I tend to think only in scales with sharps to decrease the amount of negative numbers I have to work with, and this is probably not a good practice; however in a cyclic notation (in any variation of it), I don’t like the jump between 11 and 0 (resp. 12 and 1). Negative numbers provide a great tool for understanding the traditional musical notation and non-equally-tempered tunings, but they are not that useful for fast thinking on an equally-tempered instrument.

          I am wondering whether some combination of the “da de di” notation and the clock notation (either based on fifths or semitones) wouldn’t be the best approach. It would be possible to use the numbers 0 to 11 and call them not “one, two, three, …”, but “da, mo, si, …” (or whatever would be the convention for the clock notation). Then, chord structures would be obvious from the ‘linguistic logic’ and either overall harmonic structure (if the numbers were based on fifths) or the chromatic structure (in case of your clock notation) would be readily available from the number representation.

          • ole kirkeby says:

            Aargh, I have removed the links to the graphics of the modified Schoenberg notation. I had forgot about that, sorry. The drawings were really ugly and, as you say, it is not difficult to figure out what they should look like (and then you look at my drawings and it is a turn-off — not good).

            Whenever you come up with an idea that is related to my clock notation, please drop me a line. There are only a handful of us working on this, and yet I think it has great potential, so let’s try to coordinate our efforts.

        • Viaro says:

          Hey there, Ole Kirkeby. I have browsed your M3 website, and am really interested in the development of your clock notation. Do you envision it as a tool to help the use of the letter system with sharps and flats, or do you envision the letters eventually being unnecessary?

          I would like to attempt a synthesis of your clock notation and Kubaz’ line of fifths – the goal being to remove as many intermediate steps between the raw intervals and the music. Should we exchange email addresses to discuss this further?