If you would like to participate in discussions, please sign in or register.

In this Discussion

Most Popular This Week

    Due to changing needs and technologies, the SMT Executive Board has decided to retire SMT Discuss (effective Nov. 9, 2021). Posts will be preserved for archival purposes, but new posts and replies are no longer permitted.

    Melodic Minor versus the Hybrid Scale

    Dear Colleagues,

    Genuine melodic minor has been in existence for at least 300 years now – with raised 6th and 7th scale degrees in both directions. It is enough to open any work in a minor key written by J. S. Bach and see a descending fragment which contains the raised degrees. A most conspicuous example is the contrapuntal passage to the fugue theme in the C minor fugue (WTC, I). Jazz people often claim that it is they who started using genuine melodic minor on a regular basis, but let them hear some Bach music first.

    In view of this empirical evidence, it makes no sense to call "melodic minor" the combination of ascending melodic and descending natural minor scale. It would be ridiculous to use the term "melodic minor" for such a hybrid scale and to "invent" a new nickname for the true descending melodic minor which is pretty abundant in the literature.

    Of course, the hybrid scale (ascending melodic + descending natural) occurs even more frequently than genuine melodic minor itself, but this is no reason to call it "melodic minor". If some colleagues want to prove that the hybrid scale must be called "melodic minor", let them derive harmony from its "descending portion" and prove that these chords belong to the melodic minor mode. :)

    Therefore, I suggest that we stop confusing students worldwide with such statements as, "melodic minor has a different ascending and a different descending version". Instead, we shall explain to them that a melodic minor is a minor scale which has raised 6th and 7th degrees in both directions. As for the combination of ascending melodic and descending natural scale, we can say that this scale is also very frequent; it is easier to use vocally and being fully compatible with the key signature, it frequently neutralizes the modal influence of the opposite major scale.

    Let us remind ourselves that the harmonic, melodic and doubly harmonic versions of major and minor are all conspicuous examples of modal mixture. Whether you will say "scale degrees 6 and 7 in minor are raised" or "the upper tetrachord of natural major is borrowed into minor, thus giving rise of melodic minor" – you are saying the same thing.

    It is another matter that some teachers still explain to their students that "degrees 6 and 7 in minor are flexible, while their counterparts in major are firm (!), or "minor mode has more chords than major mode because of the flexibility of degrees 6 and 7". I think that such statements should be embarrassing for a professor of music to say...

    Conclusion. There is only one scale that can be called "melodic minor", and this is the scale which has degrees 6 and 7 raised in both ascending and descending direction. There is also a hybrid scale which combines the features of melodic and natural minor. Yes, melodic minor versus a hybrid scale. No tradition is broken, no tradition is created. Only the false tradition of uniting two different scales under a single uniform name will go away.


    Thank you, and best regards.



    Dr. Dimitar Ninov

    Texas State University


    Sign In or Register to comment.


    • 19 Comments sorted by Votes Date Added
    • "Jazz people often claim that it is they who started using genuine melodic minor on a regular basis, but let them hear some Bach music first."

      That’s an intriguing statement—most jazz pianists I know and study are obsessed with Bach!  But you do find the ascending form of melodic minor in descending passages in the common-practice period.  In any case, Debussy and composers early in the 20th-century were certainly using "genuine melodic minor," as Tymoczko has amply demonstrated, but the point here is that they were using different modes of melodic minor, not just the first mode which would be used in common-practice music.  

      The significant point about jazz is that jazz musicians study the chord-scale relationships of the modes of both diatonic and acoustic collections thoroughly and early on in their training, and that the pedagogy for this has been systematically disseminated for many decades.  It is therefore visually and aurally obvious to a jazz-trained theorist when composers of any genre use an acoustic collection, and do not confuse it with the octatonic collection.  We know the difference between a 7(#9) and a 7(#9#5).  And, indeed, the reason why these relationships are studied so thoroughly by jazz musicians is because jazz, particularly (but not only) starting around 1960, employs the acoustic collection not just "regularly" but constantly, uses all of its modes, and sometimes uses these modes as the focal point of a composition or improvisation.

    • Isn't the whole point of calling something "melodic minor" that the scale is intended to represent something, well... melodic?  If we're talking about Bach: Absent any harmonic rationale for further alteration, the default melodic motion in most Common Practice style minor pieces is to used raised 6 and 7 when ascending toward the tonic note and lowered 6 and 7 when descending toward the dominant note.  The descending raised upper tetrachord is generally used by Bach for harmonic reasons (often a supporting dominant harmony), not simply for melodic variety.  I have no problem putting a name on a scale that uses the raised upper tetrachord in both directions, but this proposal seems to contradict the entire rationale for why theorists came up with the name "melodic" minor in the first place.

      All that said, I personally find the teaching of the "three forms of minor" ONLY useful in later music, after the theorists created such abstractions and taught them to enough composers that they become relevant to the function of minor mode pitch collections.  Bach (and most other 17th and 18th century music) clearly didn't conceive of three "different scales" for minor.  To me, teaching these to students regarding Common Practice music thus cements an anachronism that is more harmful than helpful.  It leads to absurd questions like "Which minor scale is Bach using in this piece?" or "How do I choose which scale to use in my model composition assignment?"  If you're writing in Bach's style, you're not worried about some sort of abstract collection of pitches known as a "scale": you write the correct notes that lead toward the local melodic goal, subject to constraints of the underlying harmony.

      But I'm probably not the right person to ask about such things, because to me the whole emphasis on diatonic scales is wrapped up in the overemphasis of diatonicism when teaching Common Practice style.  There's a reason it's nearly impossible to find "pure diatonic" examples for your intro theory class out of the Bach chorales or whatever.  Even Mozart at the age of 5 knew that the "dominant note" got its own "leading tone" frequently at an interior cadence.  And that's precisely what "melodic minor" is (and all it is, historically): leading tones.  #7 resolves to 8 and b6 resolves to 5.  The rest of the notes around these leading-tone maneuvers are altered where necessary to avoid awkward intervals like augmented 2nds.  All the talk of an abstract "scale" to represent such things for Bach is merely a mnemonic aid that attempts to represent common default choices.

      However, since theorists reified these "three different scales," we're stuck teaching (and confusing) intro students with them.  And they do have utility for understanding later music written by folks who actually thought of scales this way.

    • I think that no scale has two forms except a hybrid one, otherwise that is a misundertsanding. Secondly, a mode may be an equivalent of a scale or an equivalent of a tonality (tonal area) with limited parameters (natural, harmonic, melodic). Besides, harmonic and melodic major are also used in music, but teachers teach their students that there is one major scale and three minors scales, which is ridiculous. "Melodic" does not mean "no harmony is supposed to be derived from that", and if you claim that there is a descending portion of melodic minor that is different from the ascending one, try to prove this by deriving harmony from the latter; you will end up in the natural mode. Of course, melodic minor is also a source of harmony, with some very nice realtively or true altered chords along the way.

      If you wish to call "melodic minor" a hybrid scale, then you have to explain to your students how you would call the descending melodic minor whose scale degrees 6 and 7 are raised. Would it not be ridiculous to "invent" a new name for that?

    • Rick, I do not know what exactly you call "acoustic" collection, but the first 13 harmonic tones form a diatonic collection for sure. Even the famous maj13+11 chord is purely diatonic by nature, for it is found on the fourth scale degree of natural major, and on the sixth scale degree of natural minor. Maybe you want to use th term "chromatic" or "altered". Also, jazz poeple call "an altered dominant" only the fully altered dominant (V7-5+5, -9+9) which discloses misunderstanding. A true altered chord is any chord which contains a chromatic interval. In this sense even V7-9 is a true altered chord, not becaise the ninth is minor, but because there is a diminished seventh in the chord.

    • Dear all,

      Dimitar wrote "Genuine melodic minor has been in existence for at least 300 years now – with raised 6th and 7th scale degrees in both directions." 300 years, that leads us back to 1717 or, say, just before Rameau's Traité d'harmonie. I cannot believe that any scale named "melodic minor" was known by then, nor even that a minor scale with raised 6th and 7th degrees was described, under any name, 300 years ago. Rameau himself writes: L’on prend ordinairement l’Octave de la Notte pour modele de tous les Tons mineurs (Traité, 1722, p. 345), "One usually takes the scale of as the model of all minor keys". This is a "Dorian" minor, with raised 6th degree, and indeed it was then the generally accepted model of the minor. (Also in Bach's WTC I, as I think I already mentioned before.)

      I tried to find the origin of the expressions "melodic minor", "harmonic minor", and the like, without (yet) coming to an answer. If anyone had information about this, I'd be most interested. Browsing through some of the treatises that I have in searchable electronic form, I found the following:

      • Calcott (A Musical Grammar, 1817, p. 130) distinguishes the "essential" minor scale, with ♮6 #7 ascending and descending, from the "accidental" minor one, with #6 #7 ascending and ♮7 ♮6 descending.

      • Reber (Traité d'harmonie, 1862, p. 199) does speak of the "melodic minor scale", but he means the minor scale used as a melody. He shows it both with ♮6 #7 and in various other forms, including not only #6 #7 ascending and ♮7 ♮6 descending, but also ♮6 ♮7 ascending and #7 #6 descending, in the same example!

      • Durand (Traité complet d'harmonie, 1881, p. 172) distinguishes two forms of the minor, the mineur 1ère forme with ♮6 #7 ascending and descending, and the mineur 2e forme, with #6 #7 ascending and ♮7 ♮6 descending. He does not use the expression "melodic minor".

      • Gevaert (Traité d'harmonie, 1905, pp. 102-107) names "modern minor" the minor scale with ♮6 #7, both ascending and descending, and describes a "variant" of the "modern minor", with #6 and #7, again both ascending and descending (pp. 105-107), which he says constantly mixes with the modern minor and the diatonic minor (i.e. Aeolian). He gives several examples, mainly from Bach, of all these forms of the minor and of their mixing together. He does not use the expression "melodic minor".

      I'll search further, but I'd be particularly grateful for any additional information that SMT members might have.



    • Dear Nicolas,

      Thank you for the research - it is very useful, indeed! When I said that genuine melodic minor has been used for 300 years, I meant the practice of using this scale and mode (as source of melody and harmony), and not necessarily the verbal expression "melodic minor" as a recognized term at that time.

      Nowadays, when terminology is clarified and a tonality may incorporate all kinds of scales and modes (as long as they are subordinate to a single major or minor triad) it does not make sense to call a natural minor scale "melodic" when it descends, does it? Yet teachers massively do that, and when you ask them how this "descending melodic minor" is different from the descending natural scale, they do not know what to say...that is the problem.


      You may combine an ascending natural scale with a descending harmonic or melodic scale, but this does not make the whole scale natural, does it? Then, why call "melodic" a hybrid scale which is a combination of ascending melodic and descending natural? That this does not make sense is evident from the harmony derived from the "descending portion" of melodic minor - it is just pure diatonic harmony from the natural mode. In view of the existence of genuine melodic minor, it does not make sense to call any other scale melodic that is different than the genuine one.

      If, on the other hand, we consider the combination of a lower Dorian tetrachord and an upper chromatic pentachord (sol, la-bemol, la becare, si-bemol, si-becare) a general minor scale from which three basic versions spring, I do not mind. The same operation may be done with the major scale, using a lower Ionian tetrachord and the same upper chromatic pentachord.





    • In addition to my comments earlier today, I can add this.

      Choron and Fayolle, in their Dictionnaire historique des musiciens of 1810, write in the article devoted to Blainville (the advocate of the "third mode", the Phrygian, along the major and minor), that "in 1778 Vandermonde, great mathematician (géomètre) and nevertheless mistaken mind, proposed four types of minor modes, [...] by the alteration of the 6th or the 4th note".

      This is a reference to Vandermonde's Système d'harmonie, published in the Mémoires de l'Académie des Sciences for 1778. Vandermonde describes there (pp. 3-4) the "minor mode properly speaking" (maybe the re mode, i.e. with #6), the "minor mode in ascending" and the "minor mode in descending" (perhaps the ascending one with #6 #7; and the descending one with ♮7 ♮6, which is the la mode), to which he adds a fourth type, with #4!

      In 1781, Vandermonde published a Second mémoire sur un nouveau système d'harmonie, in which he adds a fifth type of minor mode. He writes that in the previous Mémoire, #4 was meant to suppose #6 [A B C D# E F# G or G# A, apparently], but that he must now consider also the type with both #4 and ♮6, the one that justifies the augmented sixth [♮6–#4, say F♮–D#].

      This echoes Rich Pellegrin's mention that jazz musicians "study the chord-scale relationships of the modes". Vandermonde appears unable to justify the augmented sixth without describing a scale that can include it. But common-practice harmony, as John MacKay implies, is ready to accept accidental notes outside the local scale – ficta notes, one might have said in an earlier time.





    • Dear Nicolas,


      Yes, you are talking about doubly harmonic minor perhaps, the one that has raised fourth and seventh scale degrees. This chromatic mode is one way of justifying the typical altered subdominant chords (or altered double dominants) which contain a diminished third or an augmented sixth: IV6+, IV6+/5, and II6+/4/3.

      As for the statement of Rick's that jazz people study the chord-scale relationships, that is true, but not enough. There are chords which cannot simply be explained by a certain scale arrangement, such as some altered S and D chords which make use of the lowered 4th scale degree in minor, something unheard of in the jazz realm. For instance, if you say that E7 may resolve directly into Cm as an altered dominant chord, some people will look startled, others will cleverly justify this as an upper extension of the big dominant G (13-9) with a root omitted, although this chord cannot be explained naturally in a minor mode because of the conflict between 13 as chord tone and 3 minor as a scale tone, while a third group which has studied the principle of derivation of true chromatic chords will recognize a VII4/3 with a lowered 5th in minor mode - a great linear dominant which is studied in some harmony books along with other chords containing a lowered scale degree. Another curiosity in the altered realm is V7 with a diminished seventh in minor, which weakens the dominant function and may be used along with a lowered fifth. All these chords do not necessarily concern a particular scale, but are derived logically from the principle of building chromatic chords in major and minor.

      Even the fully altered dominant (V7-5+5, -9+9) has not been derived in the way jazz people explain it (as derived from the melodic minor scale a half step above the dominant root) but is simply the result of double alteration in opposite direction of degrees 2 and 6 in natural major mode. Better yet, degree 6 is not necessarily raised but b7 is used instead, giving raise to the major/minor blues feel in V7+9, where natural seventh collides with flat seventh. In minor the fully altered dominant's 9th tones are actually diatonic scale degrees: 6 and 7 in natural minor. Thus the fully altered label is a compromise but it keeps uniformity between major and minor modes, and the chord is a true chromatic chord anyway.


    • I'm a little behind in this thread, so let me here just respond to Dimitar's first comment to me. 

      Waters and Tymoczko in various places refer to the asending form of melodic minor as the acoustic collection, due to the overtone series.  The first 13 notes create "G melodic minor," due to the presence of both Bb and F#.  I used the two terms interchangeably in the post in order to point out their "collectional" equivalence.  (The pitch centers are labeled differently though; "G melodic minor" is collectionally equivalent to "C acoustic.")

      Regarding the altered dominant, in my experience as a jazz musician we refer to "altered dominants" in general, in the way that you do; i.e., an altered dominant is a dominant that has any chromatic alteration.  However, some people use the symbol "ALT" to designate the fully altered (b9#9b5#5) dominant you mention.  This is an unfortunate symbol in my opinion, because not all jazz musicians use or understand it, and therefore may play any of the altered dominants rather than a fully altered dominant. 

    • No doubt there are some sonorities that cannot be accounted for by chord/scale theory.  Chord/scale theory is only one piece of the puzzle.  Many jazz musicians, theorists, and composers feel that chord/scale theory is pedagogically overemphasized, and many (most?) of them do not derive chords from scales.  Scales are studied more as a way to gain fluency in improvising or composing with particular chords, since once does not want to simply arpeggiate chords in composing or improvising.

      That being said, there are many more options than just diatonic and acoustic modes.  For example, there are also modes of harmonic minor, harmonic major, double harmonic, etc.  I agree with Dimitar that the #4 b6 scale from Vandermonde, researched by Nicolas, is double harmonic.  Although in jazz terms, this would be a mode of double harmonic, as mode 1 of double harmonic is taught in jazz as 1 b2 3 4 5 b6 7 1 (i.e., the two augmented seconds are at either end of the scale, not in the middle).  As for the Vandermonde raised-4 raised-6, that would be the 4th mode of harmonic major.

      Dimitar - I don't understand why you are trying to derive the fully altered dominant chord from a major scale.  An unaltered dominant chord is "derived" from Mixolydian, which is diatonic to the major key.

      Lastly, Dimitar, I would caution against assuming that anything is "unheard of in the jazz realm," as jazz covers an extraordinarily broad range of music and the corpus of recorded jazz is vast.  Consider also that Ornette Coleman's album Free Jazz came out 56 years ago (at least half as long as jazz has even existed).  There's a lot out there.

    • I will say that the "melodic minor" scale as represented by ascending melodic minor and descending natural minor form is a well established convention in the cello pedagogy and historical etude books (and I image the same is true for other string instrument literature).  The raised 6th and 7th degrees used in a minor context, or lowered 6th and 7th are widely used in early music (Chant), music from the Middle Ages as musical ficta, the Renaissance modal system, Common Practice period (although the ascending/descending form is preferred, composers do break the rules as the Bach examples demonstrate.  Perhaps the features of the "melodic minor" scale are more important than the term: half minor, half major tetras, the raised 6th works well when the goal is the tonic (but composers also like the harmonic minor flat 6th raised 7th) for a more exotic sound, the 7th (both major and minor) functions as a leading tone in either major or minor keys or dominant 7th inversion or passing tone, and the variability of these degrees offer the potential for expanded harmonic variation.  It is certainly popular among jazz musicians to use the ascending melodic form in both directions possibly because that form offers more dissonance and harmonic complexity when expanded to 9ths, 11ths, and 13th chords.   

    • Carson, you write that the "melodic minor" scale is "a well established convention in [...] historical [cello] etude books". This is precisely what I am researching. Can you be more specific? Can you quote titles of books, and dates? Of course, cello books do not predate the 18th century, but I'd believe that their mention of the "melodic minor" (with its difference between the ascending and the descending), or their systematic description of it, even without name, hardly predates the second half of the 19th century.

      You also write that "the raised 6th and 7th degrees used in a minor context, or lowered 6th and 7th are widely used in early music (Chant)". This must be a mistake or a misunderstanding: there is no reason to raise or lower a degree in modal chant, where in addition there cannot exist a "minor context". It is only in polyphony that the need may arise and even there, it may not have been as "widely" used as you seem to think.

      Accidental raisings or lowerings of course are not the same thing as systematic raisings or lowerings over a span of time. A raised 7th degree at the moment of a cadence does not form an ascending melodic or harmonic minor scale. But it would be difficult to discuss that in real compositions: I am therefore more interested in theory texts, which may at least give an indication of how the things were conceived in their own time. Among those that I quoted, Durand (1881) is for the time being the first...

      Once again, I'd be grateful for titles and dates.



    • Dear Rick, thanks for all of your valuable comments! I agree with most of them. I will add that there are three basic forms of each major and minor scale: natural, harmonic and melodic. The natural and melodic major and minor are dual, that is - you can use the same collection of tones to create either a major or a minor scale. Thus, if melodic minor is the host, its fifth mode is melodic major, or if melodic major is the host - its fourth mode is melodic minor. For instance, C melodic minor is also a G melodic major. What is strange is that most jazz theorists do not see this connection and they call the fifth mode of melodic minor "Mixolydian flat 6" But if you do that, you should call melodic minor "Dorian raised 7" which only creates more confusing terms. People must be able to understand that the upper tetrachord of melodic minor is borrowed from natural major, and the upper tetrachord of melodic major is borrowed from natural minor. No Mixolydian flat 6 - Melodic major is the term, and the duality and modal mixture strongly imply that!

      But that is not all - the harmonic and melodic versions of major and minor are chromatic modes, for they also yield chromatic chords and intervals...Some theorists call them "conditional diatonic system" to point out that these are chromatic modes with low level of chromaticism and most of their chords are diatonic. But for their shame, some musicians do not understand that. Most textbooks show "the diatonic chords in minor" among which the augmented triad and some other truly altered chords are listed...of course this is ridiculous, for the augmented triad is a true chromatic (really altered) chord...

      The doubly harmonic modes of major and minor are also dual: C doubly harmonic major is at the same time F doubly harmonic minor. These modes also yield the popular  subdominant and dominant chords with an augmented sixth.

      120 years ago, N. Rimsky-Korsakov based his books of harmony on the natural and harmonic major and minor modes. Today I base the first part of my harmony book on six modes – the above mentioned natural, harmonic and melodic major and minor.


    • Hi Nicolas, yes I will be happy to provide you with some examples and dates as soon as I get to my library.  

    • I just want to know one thing. Will this be on the test? If so, which one? For the theory history class or the contemporary music theory class?

    • Stephen,

      It will definitely be on your jazz improvisation, arranging, theory, or aural skills tests!

      And, in my opinion, the huge body of jazz theory treatises spanning 50+ years is a significant piece of the history of Western music theory, and should be studied accordingly.

    • Dimitar-

      I think we're talking "across" each other here. I said the "melodic minor" is representing melodic defaults for a particular era; you say we can derive harmonies from melodic minor scales. I agree with your statement, but it does not contradict the rationale for why "melodic minor" came to be a pedagogical aid in the first place (as Nicolas documented).

      Also, you seem to be operating from an implicit definition of "scale" as "collection of pitch classes."  I do not accept that that is the only definition of "scale."  Scales (or modes, or other terms, depending on circumstances) can also include information about common melodic gestures.  Knowing how Gregorian modes (or Indian raga) function in actual repertoires frequently requires knowing something not just about collections of pitches, but how they relate to each other, which can include stereotypical melodic motions.  I believe our standard Western scale pedagogy has suffered a bit because we tend to view scales as abstract pc collections, even though many traditional musics around the world have more sophisticated notions of the basic "material" of pitches as used in performance.

      In any case, I have absolutely no problem coming up with a name for whatever scale or collection of pitches.  I just think it's an "uphill battle" to attempt to change the definition of such an established concept that already has a clear definition and a rationale that's built into its very name.

    • John,

      In fact, I am not arguing with anyone. I declare it makes no sense to call "melodic minor" a portion of the natural scale in case genuine melodic minor has been practiced for 300 years. You are free to agree or disagree with my declaration, and I respect your opinion.

      You say: "Scales (or modes, or other terms, depending on circumstances) can also include information about common melodic gestures." Right. So, out of two scales: the hybrid one with a descending natural portion, and the genuine one which goes unchanged – which one will you call "melodic minor"?

      You say: "I just think it's an "uphill battle" to attempt to change the definition of such an established concept that already has a clear definition and a rationale that's built into its very name." Certainly, It is not a battle for me. I teach this way, and in my classes I play genuine melodic minor as one of 13-17 scales I use to educate my students in aural II and III. They instantly understand the discrepancy between the name "melodic" and the use of two different scales within it (ascenind melodic and descenidng natural), and they know that if melodic minor will be on the test - it will be genuine (unchanged in both directions).

      Battle? With whom? With theorists who insist on the continual application of a rotten tradition and pretend genuine melodic minor does not exists in the old music literature?

      I will repeat my question to you: how would you call the melodic minor scale (yes, the one which has raised 6th and 7th degree in both directions and has been used so much by Bach and other Baroque composers)?

    • Hi Dimitar, I think John's point is that the idea of the ascending descending melodic minor natural minor form is fairly well established in the literature and pedagogy as well.  For instance if you go to imslp.org and search for "cello methods" you will find that a majority of the methods and material includes such [hybrid] examples.  I'm sure it may be similar for other instruments?  Maybe the pianists and other instrumentalists in the group could weigh in?  I agree that Bach uses the descending form of the  melodic minor and there are certainly some examples of that in the Bach Cello Suites, but on average it is a much smaller occurrence than the "ascending/descending" hybrid. In Suite II Praeludum in D minor if Bach uses a C# in a D minor context it is an auxiliary tone or a dominant [A] leading tone.  In the final ten measures before the ending there is a descending [A] G F E D C# B A  = A dominant 7th with a flat 6th resolving to D minor.  Perhaps that suggests that Bach interchanged tetras in many different combinations and variants.