27

I understand the construction of the scales; natural minor, harmonic minor (raises the 7th) and melodic minor (raises both 6&7 when ascending).

What I don't understand is why, when I play melodic minor descending, it flattens the 6&7 ,giving me the natural minor. So my questions is: Why is a descending melodic minor scale the same as a natural minor scale? Thanks.

John
  • 1,934
  • 3
  • 19
  • 26
Slick
  • 271
  • 3
  • 3

8 Answers8

39

The main “purpose” of the raised degrees in ascending melodic and harmonic minor is to create a strong leading note:

X:1
L:1/4
M:none
K:Am
V:1 clef=treble
e ^f "V₇ ↗"^g "i"a
%

Clearly, this leading tone only makes sense if you actually do resolve it upwards. If you go down instead, this can rather leave a feeling of unresolved tension.

X:1
L:1/4
M:none
K:Am
V:1 clef=treble
a "↗"^g "?"^f "??"e
%

In harmonic minor, this unresolvedness is not as apparent because the attention will be more drawn to the exotic (if not somewhat jarring) 1½ tone step:

X:1
L:1/4
M:none
K:Am
V:1 clef=treble
a (^g "!↘"f) e
%

Melodic minor avoids these 1½ tone steps to create smoother melody lines. When going up, you still want the leading tone, hence you raise the ⅶ and also the ⅵ. But when going down, it's more sensible to just leave both of them natural.

Of course that doesn't mean you can't use raised ⅵ and ⅶ degrees when going downwards. In particular if one voice already has one of these raised tones, it would be problematic to un-raise them in another voice. For example, the fugue that RRR brought up has this passage:

X:1
L:1/16
M:C
K:Am
%%score (T1 T2) (B1)
V:T1           clef=treble
V:T2           clef=treble
V:B1           clef=bass
% 1
[V:T1] (d8            d2 )
[V:T2] A,2dc   BA^G^F E2
[V:B1] G,2^F,E,F,2^G,2 A,2

Here, the bass is already going up in melodic minor to the A root, and especially if the bass does this you really want the leading note effect, so it is F♯-G♯-A. At the same time, the alto needs to cross the ⅵ / ⅶ territory to get down to the Ⅴ (e). Using a g-natural there would cause a pretty bad clash between the voices:

X:1
L:1/16
M:C
K:Am
%%score (T1 T2) (B1)
V:T1           clef=treble
V:T2           clef=treble
V:B1           clef=bass
% 1
[V:T1] (d8            d2 )
[V:T2] A,2dc   BA"⚡"GF E2
[V:B1] G,2^F,E,F,2^G,2 A,2

On alternative would be to use harmonic-descending there

X:1
L:1/16
M:C
K:Am
%%score (T1 T2) (B1)
V:T1           clef=treble
V:T2           clef=treble
V:B1           clef=bass
% 1
[V:T1] (d8            d2 )
[V:T2] A,2dc   BA^G"¿¡"F E2
[V:B1] G,2^F,E,F,2^G,2 A,2

but again, that sounds a bit Spanish. Especially in a middle voice it's more innocuous to just adapt to the context of melodic-minor from the bass voice there, even if it means traversing the scale in the “wrong” direction.

Aaron
  • 70,616
  • 10
  • 97
  • 243
leftaroundabout
  • 34,776
  • 66
  • 147
  • The problem with this standard answer is that every time you alter any note in the scale it changes the key. When the key changes the chord set changes. In Western music the scale always matches the chords. – RRR Dec 27 '16 at 07:54
  • 4
    @RRR I don't think "every time you alter any note in the scale it changes the key" is a standard music theory viewpoint; if anything, it's the opposite: ideas like the "melodic minor" are an (arguably clumsy) attempt to allow variations in tonality *without* seeing every variation as a key change. – Нет войне Dec 27 '16 at 08:24
  • @RRR that's depending on what you mean by "alter any note in the scale", of course. If it's a 'temporary' deviation, especially in a melody, that could just be a bit of melodic 'colour'. If you permanently alter a note, then you may be in a different key, but it would only be definitively seen as a key change if you'd changed the tonal centre of the piece too. – Нет войне Dec 27 '16 at 08:29
  • @topo morto the whole idea is that every time you alter any note in the scale it changes the key! So, every time you change a note you change the key. That's what key means. There's no variation in the note set without a change in the chord set. Otherwise scales have no meaning. You can't have it both ways. The Harmonic minor and the major and the melodic minor are distinct entities. THey can be strung together horizontally but not vertically. Otherwise the result is nonsense. – RRR Dec 27 '16 at 10:53
  • 3
    @RRR what you say may very well hold true for Jazz, but in classical music there's not really any such thing as “the chord set”. Harmony arises from counterpoint between different voices, and what chords can be created this way is just a consequence of the notes used by each voice. And it's not necessarily possible to say what scale each single note belongs to; quite often an accidental is just used to cause a secondary leading effect without really shifting the tonal center. Especially for longer chromatic runs, it would be impractical to interpret every accidental as a key change. – leftaroundabout Dec 27 '16 at 11:36
  • 1
    @RRR ...whilst in pop/rock, you might have a fairly simple cyclic chord progression made up of a set of chords, the constituent notes of which wouldn't fit in the diatonic scale - but in many cases it wouldn't be common to analyse such a chord progression as involving key changes. – Нет войне Dec 28 '16 at 10:01
  • @RRR are you saying that every accidental in a piece of music is a key in change? – SaggingRufus Dec 29 '16 at 12:46
5

Because it gives a smoother melodic line when running up and down the scale. hence "melodic" minor.

Is this maybe arising from the common misconception that there's "a scale for a chord" and you're meant to stick to it? Not so, and you've just discovered one example!

Laurence
  • 84,790
  • 5
  • 59
  • 178
4

As already mentioned, a strong leading tone, one semitone lower than the root note, sounds better with a smaller change. This then made a tone and a half gap between notes 6 and 7, so 6 was also put up a semitone (from where it was in the natural minor). The natural minor contains all of the notes found in the relative major. Hence the same key sig. When, in early music, the melody went down, it sounded fine to use the notes from the natural minor, but ascending melodies were thought to be better sounding using the sharpened notes.

As an aside, all three minors use the same first 5 notes in their scales: natural then continues with notes from the relative major; harmonic continues with natural minor notes, except the raised leading tone; melodic continues with the parallel major notes ascending. There is also the jazz melodic, which favours the old ascending melodic notes both ways.

Modal minors are for another day...

Tim
  • 183,051
  • 16
  • 181
  • 444
  • For a long time we were quite content with "modal" flattened sevenths. Then, for a time, we became obsessed with dominants and tonics, which required the sharpened leading note. Don't knock it, a WHOLE lot of excellent music came out of that system (and still does). – Laurence Aug 26 '16 at 15:02
  • I'm not knocking anything! Of course it works (worked), and gave us music that sounded as it did. But so did tuning until ET came along. Would we revert? Probably not, generally speaking. Or did I miss the point? – Tim Aug 26 '16 at 17:23
  • 1
    ET? The "call home" guy? – Laurence Aug 26 '16 at 20:11
  • @Laurence I believe Tim means "Equal Temperament," as distinguished from "Well Temperament" e.g., in "The Well-Tempered Clavier" (q.v.). – David Conrad Aug 27 '16 at 23:58
  • 1
    @Tim I just realised from your answer why some diplomat called it the Jazz Melodic, cos he didn't want to be bothered by loony ideologs. "Well ishnot YOR Melodic, ishda the Jazz Melodic, shee? na leeme alone!". – RRR Dec 27 '16 at 08:33
  • @RRR possibly the basis of most advances in theory! :) – Нет войне Dec 28 '16 at 10:31
2

Another way to look at things (popular in the Common Practice Period) is to consider that each key has two modes, major, and minor. The most important (or commonly used) notes in a major key are the usual diatonic notes (like C,D,E,F,G,A,B,...) when written with the key note as 1. A minor key has a minor third (C,D,Eb,F,G...) and two mutable tones on steps 6 and 7. Both the "natural" versions (Ab, Bb) and the "raised" versions (A, B) are for the most part considered diatonic to the key. Historically, the major mode seems to have come from the Lydian and Mixolydian modes. The minor mode comes from the Dorian and Phrygian modes. Fashions in cadential and melodic procedures lead the major and minor modes. The two most obvious are the raised seventh degree at a cadence and the avoidance of an augmented second between the usual sixth and raised seventh scale degrees, especially in vocal music.

The use of raised steps in ascending and lowed steps in descending passages is a reasonable rule, but it really only holds when the passage has tonic-oriented underlying harmony (like c-minor chords in the key of c-minor.) With subtonic-oriented harmony, the tendency is to used both scale steps in their lowered form (Ab and Bb against an F minor chord). Similarly, the raised scales steps tend to be preferred against dominant harmony (A and B vs a G or G7 chord.)

There are several other exceptional procedures; the most common is to ignore any of the above procedures for color reasons. It sounds a bit more exotic to use the augmented second. This is no problem for instrumental music. Also, the Ab-B scale step is often used against a dominant harmony with a dominant ninth chord (G9 in the key of C). Arpeggiating a dominant ninth chord uses F,G,Ab,B, four succsive scale steps. The raised 6 step is rarely (if ever) used as an upper neighbor the fifth step (Ab is the neighbor to G). (I'm not sure about lower neighbors to the seventh step, even in a major key.)

There are several books that describe these developments. They do not all agree with each other.

Joel Lester: "Between Modes and Keys" Joel Lester: "Compositional Theory in the Eighteenth Century" Carl Dahlhaus: "Studies on the Origin of Harmonic Tonality"

These and a search through the Internet (and Google Scholar) can be helpful.

ttw
  • 24,207
  • 1
  • 31
  • 76
2

It doesn't.

bach example

(from The Chord Scale Theory & Jazz Harmony by B. Nettles and R. Graf / Berklee College of Music)

As you can see from the third bar of this wonderful example by @Matt L. it descends the same way as it ascends, or it ain't the Melodic Minor! (It's usualy the Aeolian i.e. the Relative Major.)

If Am "Melodic" "descends" as A,G,F,E,D,C,B,A then thats the Aeolian. (i'm going to call it A Aeolian, but you can call it C Aeolian if you like.)

A Aeolian is the same as the key of C Major. The key of C major is not the same as the key of A Melodic Minor. Thats a key change. You aren't in A Melodic Minor at that point. End of story.

For the purposes of any kind of basic improvisation at all the Melodic Minor must be played descending the same way as ascending, (A,G#,F#,E,D,C,B,A) as J.S. Bach shows us.

Very obviously, from any point of view, the basic baby A Melodic Minor chord set:

---Am, Bm, C+, D, E(7), F# dim, G# dim

has absolutely nothing to do with the set of notes:

A,B,C,D,E,F,G

And you know what? Playing them in the other direction isn't gonna make them work either!

And if you're not intending to improvise, how can practicing two completely different sounds and somehow "pretending" that they have anything to do with each other help a child, let alone an adult?

Where on earth did this misleading musical pedagogy come from? I'm not being rhetorical here, I'd really like to know. I've asked dozens of top musicians and they all shrug and admit that it's daft, but oy vey!

So gentle seeker, breath a sigh of relief. You have been lied to.

It's not a question. The question is moot. It's two different things, and music teachers be crazy!!

For a hilarous example of the above, which descends, slowly but inevitably, (as this thread well might) into an appauling cat fight, check out [http://www.abrsm.org/forum/index.php?showtopic=54590]

Enjoy!

RRR
  • 448
  • 4
  • 6
  • I got hauled over the coals for calling melodic minor a key. However, I'm with you all the way (except mode naming!). It's the old adage - if it sounds good... But someone had to make up a rule for it, didn't they? – Tim Dec 27 '16 at 07:56
  • @Tim WHO hauled you over the coals? And whats your beef with mode nameing ? Do you mean that I should call It C Aeolian in this example? I just added a link to the post. it's fun. – RRR Dec 27 '16 at 08:17
  • @Tim What does the number by my name mean? it keeps going up!!?? – RRR Dec 27 '16 at 08:21
  • @RRR http://meta.stackexchange.com/questions/7237/how-does-reputation-work, assuming you're not just showing false modesty :) – Нет войне Dec 27 '16 at 08:27
  • @topo morto thanks and er ...blimey! how Trumpesque! Positively Byzantine! (the scheme not your comment). Obviously designed by and for males only. Don't worry, I have no modesty whatsoever, false or otherwise. – RRR Dec 27 '16 at 08:40
  • @RRR the scheme is just designed to give a little bit of emotional validation to the dreadful types of people you will meet on these stack exchange sites. Pay as much or little heed to it as you will. – Нет войне Dec 27 '16 at 09:17
  • @topo morto lol "dreadful types of people"! DEPLORABLES. Napoleon invented medals for when there was no booty. GIVE US CASH! – RRR Dec 27 '16 at 10:58
  • 1
    I disagree with your conclusion – it's generally _not_ wrong to naturalise the ⅵ and ⅶ when going down, in fact it usually makes sense to do this – but +1 for bringing up an example where the circumstances indeed make it necessary to keep these degrees raised while going down. I added a discussion of that to my answer. – leftaroundabout Dec 27 '16 at 12:23
  • @leftaroundabout It's not WRONG to lower the 7 and 6. It's merely not Melodic! It's Aeolian! (And If you just lower the 7 it's Dorian). I invite you to show me one single example of what you refer to from any where at all that's NOT patently Aeolian. Proove it. You proove it, I'll thank you for showing me the way. – RRR Dec 27 '16 at 13:36
  • @RRR yes, it is Aeolian. Melodic-minor is just a variation of Aeolian after all, at least in baroque / classical music. In impressionism, Jazz etc., melodic-minor is also used in a different sense that has more to do with modes / chords like you say, but that's quite difficult from the classical notion of melodic-minor (in fact I'd rather call this different, as it has little to do with melody). – leftaroundabout Dec 27 '16 at 14:19
2

The main purpose of the raised sixth scale degree was because, for the longest time in melody writing, writers wanted to avoid the augmented second interval between the sixth and seventh scale degrees.

So the theorist was left with a dilemma, you want a strong raised leading tone so that the Dominant to Tonic resolution in the cadences but still this augmented second led to unsatisfactory melodies.

So they came up with the idea of raising the sixth note of the minor scale as well to solve these problems they were having. The melodic minor has its roots very much in melody writing.

When your line is descending then you can have a minor seventh and then also you would not want a raised sixth when your scale is descending, so that is how we came up with the natural minor form (ie a minor scale that has all the notes of the relative Major.)

Neil Meyer
  • 14,365
  • 7
  • 40
  • 88
0

Since major scale and natural minor scale were the two modes in the modal system, I postulate that the melodic minor scale was accidentally found by lower the 3rd a half step. Take C major scale for example, if you were just only flatting one-note, the E a half step to Eb, you're automatically in the C melodic minor scale ascending (let's just ignore about the 3 flats key signature). There was no need to raise the 6th and the 7th for they were already there, to begin with. We just follow the interval distances to form other melodic minor scales.

This comes to the 2nd postulation of mine about why do we have the natural minor scale descending in the melodic minor scale. If someone sings a full C melodic minor scale, with the descending staying as exactly as the ascending intervals, the natural chance of singing the C major scale will be like 98% when descending because we get used to the Fa-Mi Ti-Do of a major scale, it is quite difficult (Not impossible, but hard) to sing the note in Eb! Try it

Sean
  • 9
  • 2
0

This "why" may be a question with different edges

  • The convention of the name "melodic minor scale" is that, a convention that surely appeared at some point in the history of musical theory; probably it appeared first in a book and we would have to do an historical study on the author, how music was conceived at that time, and so on. On this, there is a tendency to now call the "melodic minor scale" the scale that rises and falls in the same way, and that could be the convention of this time, but we always need to consider that traditionally the minor melodic scale went up as a melodic minor scale and went down as a natural minor scale.
  • Another thing may be the "why", as aesthetic reason, of using "melodic minor scale" with that decrease other than its rise. Well, each style has its followers and detractors, I prefer the aesthetics of the harmonic minor scale, but I understand that the second augmented, is very striking, very different from the rest of seconds and probably at that time they did not want a second That emphasizes so much, in both cases conserving the vii as leading tone (not sub tonic), of course.
Pillqu
  • 21
  • 5