19

Is the meaning very context dependent, or can it be defined in general terms? Does the meaning change by author or period?

Sometimes it's used to refer to anything that stays within a specific tonal key or context, but sometimes when people talk about diatonic scales they refer to seven note scales specifically, even though other scales (like pentatonic) can be diatonic too. Sometimes it's just used as opposite of "chromatic".

According to Encyclopedia Britannica:

Diatonic, in music, any stepwise arrangement of the seven “natural” pitches (scale degrees) forming an octave without altering the established pattern of a key or mode—in particular, the major and natural minor scales. Some scales, including pentatonic and whole-tone scales, are not diatonic because they do not include the seven degrees.

But in Sound on Sound we can read that church modes are diatonic too, so it's not really about major and natural minor scales:

The definition of a diatonic scale is that there are five whole-tone and two semitone intervals in the series and that the semitones must always be separated by at least two whole-tones. Using '2' to symbolize the whole-tone steps and '1' for the semitone steps, the major diatonic scale corresponds to the interval series 2212221. No matter what note you start on, following this prescription yields a major diatonic scale — the white keys starting on C is one example. It turns out that all possible diatonic scales are constructed by starting somewhere in the major diatonic scale and continuing until you reach the same note you started on. Those are generally referred to as the church modes: Dorian for 2122212, Phrygian for 1222122, Lydian for 2221221, and so on.

Are harmonic and melodic minor not diatonic then? The i V i progression isn't diatonic?

According to Wikipedia, "diatonic" can apply to:

Musical instruments, intervals, chords, notes, musical styles, and kinds of harmony

And also puts a time context to the concept

They are very often used as a pair, especially when applied to contrasting features of the common practice music of the period 1600–1900.

Even when staying within similar contexts, the concept can take various very similar forms, but that still differ in something important. There's maybe a place where the definitions converge? Or is it just one of those concepts that vary among authors?

Exactly what does "diatonic" mean?

5
  • 2
    My answer is on the question - 'Can a diatonic scale have sharps and flats?' Sadly, there is more than one 'definition'.
    – Tim
    Nov 27, 2019 at 10:41
  • 1
    I find this an interesting q and await answers from cognoscenti. For me diatonic is close/synoymous to scale and tonality.
    – Rusi
    Nov 27, 2019 at 11:58
  • 1
    I like this question as it goes on the ground of our completely unsure knowledge and ignorance of fundamental terms in music theory. ;) Nov 27, 2019 at 15:09
  • 1
    Very related: this answer.
    – Matt L.
    Nov 28, 2019 at 9:22
  • 1
    There are very few words (if any) that have an “exact” meaning. Almost all words have shading of meaning depending on context.
    – Dave
    Nov 28, 2019 at 15:02

9 Answers 9

15
+150

As others have mentioned, the word diatonic comes from ancient Greek music theory and literally means "through [whole] tones." Ancient Greek music tuned its scales using intervals of perfect fourths called tetrachords. A diatonic tetrachord was one that was tuned with two whole tones on the top, and the remainder left on the bottom (roughly a semitone), like descending E-D-C-B in our modern scale.

These tunings in ancient Greece were contrasted with chromatic methods of tuning the tetrachord, which generally involved intervals smaller than whole tones and therefore often resulted in some consecutive semitone-sized intervals, like we find in our modern chromatic scale.

The reason this background is important is because it gave birth to two somewhat different ways of using the term diatonic today:

(1) The first comes directly from Greek scale construction. By adding an additional diatonic tetrachord (A-G-F-E) to the E-D-C-B one I mentioned above, we can get a complete descending scale for an octave: E-D-C-B-A-G-F-E, with a distinctive pattern of whole steps and half steps. (I give the scale in descending order, as that was typically how Greeks would think of tuning it.)

Those notes were also the "white notes" on our modern piano. This pattern derived from the Greek diatonic scale thus gave birth to all the diatonic medieval modes, from our modern C major and A natural minor scales, to the Dorian, Phrygian, Lydian, and Mixolydian modes, all employing that same diatonic scale from ancient Greece. In that original sense of diatonic, all the modes using that specific set of notes with its pattern of whole steps and half steps (like the white keys on a piano) are derived from a "diatonic scale."

(2) The second usage comes later historically, not emerging until around the 18th century when our modern major/minor key system came into place. During the late 16th century, there was a strong revival of interest in ancient Greek music theory that began in Italy and spread more widely. With it, the concept of things like chromatic tunings became of more interest. Chromatic notes became associated specifically with notes outside of the standard diatonic scale (consisting, again, mostly of what we think of as the "white notes" on the piano).

Thus, when major and minor keys became standard in the 18th century, the word diatonic retained its association with the "primary notes of the scale," while chromatic referred to notes "outside the primary scale." For major keys, the seven standard notes were thus diatonic, and any accidentals could be considered chromatic notes. For minor keys in classical style, it was a bit more complicated, as the 6th and 7th scale degrees often were employed in both raised and lowered forms. Traditionally, the word diatonic related most directly to the so-called natural minor, because of the rationale from point (1) above.

However, modern music theory sources are inconsistent with the way they use diatonic in relationship to minor keys. Generally speaking, standard uses of both the raised and lowered forms of scale degrees 6 and 7 in minor are often discussed as "diatonic" in many modern music theory books. That is, the leading tone is generally raised in dominant function chords. The sixth scale degree can be raised to progress melodically in smooth motion to the leading tone. The sixth scale degree is generally flattened in other contexts (especially when leading to scale degree five) and the seventh sale degree may also be flattened to progress melodically to the lowered sixth. All of these are often thought of as standard "diatonic" patterns in minor, while more exotic uses of chromaticism with scale degrees 6 or 7, or uses of other accidentals on other scale degrees, would be considered "chromatic."


This sidesteps the question of whether melodic and harmonic minor scales are diatonic, which is really a matter of opinion and how exactly you create the formal definitions.

From my perspective, the usefulness of the term diatonic in the second sense is not about scales specifically, but rather about whether particular notes, intervals, and chords are considered "chromatic" or not. To my mind, if the goal of the word diatonic now is to determine the standard uses of notes of the scale, then a leading tone in minor is necessary in classical style, so it's obviously "diatonic" by the second meaning. Intervals and chords built using that leading tone are also, by extension, "diatonic." Hence, the part in the OP's original question that asks whether i-V-i is "diatonic" is that such a progression obviously is, according to the second meaning.

But is a harmonic minor scale "diatonic" though? I'd personally say no. My rationale is that generally when people speak of "diatonic" in relation to scales, they are using that historical definition (1), with its specific pattern of whole steps and half steps. Some would allow other scales to be labeled "diatonic" too, but the criteria get increasingly loose if we were to admit harmonic minor (with its augmented second) into the mix.

On the other hand, chords and harmonic progressions that use the standard minor key accidentals are thought of as "diatonic" according to definition (2). By that definition, we might consider a use of a raised 6th scale degree fine (and "diatonic") if necessitated by melodic motion, but a purely coloristic use of the raised 6th degree (as in a simple i-IV-i) progression might be thought of as a "chromatic" variant of the iv chord.

I think the usage I outlined in the last couple paragraphs accords with how standard music theory texts often use the terms today, but there is still some variation in official definitions and use.

0
8

The definition that I've seen most often (composition, harmony, analysis, counterpoint, and form books for the most part) relates to the Common Practice Period key structure. Some extension are made to earlier forms. (I don't remember seeing the term used much for post-romantic music.)

In major keys, "diatonic" refers to melodies and harmonies using notes from that key. For minor keys, "diatonic" refers to scale steps 1,2,b3,4,5 and both b6 and b7 and 7. Both form of the mutable steps (6 and 7) are termed "diatonic." If a modulation or even a tonicization occurs, one may use diatonic to refer to either key depending on what's needed. This leads to the next paragraph.

One cannot define "diatonic" without some reference to the term "chromatic." Music that stays in a single key for the most part with a few chromatically altered notes may be said to use "inessential chromaticism." "Inessential" means that one can analyze the piece without worrying about the chromatic notes. Perhaps Schenkeristas would say that the chromaticism disappears in deeper structures. These would things like a secondary dominant, a Neapolitan Sixth, an Augmented Sixth, perhaps an augmented or diminished chord now and then. (I would like to extend "diatonic" to these harmonic structures and to using accidentals to create a half-step in things like trills or neighbor tones, but I haven't found anyone else doing this. I just use the idea for myself in composition.)

Some music (Wagner, Chopin, Mozart, Bach, Gesualdo, Beethoven, etc.) write music where the chromatic parts are fundamental to the structure. (Schenker's deep structures contain these chromaticisms.) They are not small-scale deviations from the "diatonic" structure but essential. (Called in some of the books, "essential chromaticism.")

In pre-Baroque music, I've seen the term diatonic used to refer to pieces that do not need accidentals to switch modes. Of course, the Bb vs B (or B vs H) contrast seems to be called diatonic by some authors and chromatic by other.

The original usage for Greek tetrachords classifies them by the terms diatonic, chromatic, and enharmonic. The meanings of these terms has changed but he words linger on.

3
  • 1
    ‘One cannot define "diatonic" without some reference to the term "chromatic."’ Mmm, that's my understanding — that diatonicism is about scales and note sets that mostly involve intervals of two (or more) semitones, as opposed to chromatic scales and note sets which are only single semitones apart.
    – gidds
    Nov 27, 2019 at 20:46
  • So, in this case, i V i progression would be diatonic? And for that matter, all melodic and harmonic minor based harmonies would be diatonic too?
    – NPN328
    Nov 28, 2019 at 7:30
  • 1
    In CPP harmonic, the minor mode is treated as a single object. One can derive the three scales, but all three are often used in the same piece. Only raised 6 with lowered 7 together seems to be avoided (probably because this sounds a bit like a major scale formed a second below the tonic.) Also the augmented chord on III isn't common though uses in the late 19th century. Likewise (and this is the only thing I have found universally avoided) is using the raised 6 as an upper neighbor to 5 (probably because half-step neighbors are really popular.)
    – ttw
    Nov 28, 2019 at 15:42
7

According to my trusty 'bible' - aka 'Oxford Companion to Music' -

Diatonic scales are the major and minor, made up of tones and semitones, (inc. +2 in the harmonic minor) as distinct from the chromatic scales, made entirely of semitones. Thus the modes are also diatonic.

Diatonic passages, intervals, chords and harmonies, all constructed from the notes of the prevailing key are, whilst chromatic passages, intervals, chords and harmonies are not.

Some authorities won't include harmonic and melodic minor notes, and define the intervals of diatonic scales to be the well known TTSTTTS.

I have a feeling there's not one correct answer...

2
  • 1
    Indeed. Can the altered scale be diatonic? It's a mode of melodic minor... But if not, is E7 diatonic to A minor? I'm not sure what I believe anymore :)
    – user45266
    Nov 27, 2019 at 22:31
  • @user45266 - if all minor scales contain diatonic notes, therefore are diatonc, thn yes, of course, any mode of melodic (or harmonic) minor will be diatonic. Just like Dorian, Phrygian et al. IF...
    – Tim
    Apr 5, 2020 at 10:13
5

Diatonic is each scale you can play equal to the scales with the white keys. So any other mode that can be fitted in a same pattern of 5 whole tone steps and 2 half tone steps arranged in the same way as the white keys of a keyboard is diatonic.

This means: all scales like wwhwwwh, whwwwhw, hwwwhww, wwwhwwh, wwhwwhw, whwwhww, hwwhwww, will always have a pattern ww and www if they have a range more than an octave.

As we see the pitch doesn't play a role and neither the root tone of the scale.

Examples:

All major scales and all natural minor scales,

all church modes without accidentals and

all Greek modes build on 5 w and 2 half steps as described above.

All intervals derived from this scales are diatonic intervals.

So far what I know without looking consulting other sources.

N.B.

The Greeks had diatonic scales built of 2 tetrachords downstep like e,d,c,b,a,g,f,e and as we can read in wiki they also had "chromatic scales" with other half steps and "enharmonic scales" with quarter steps.

https://en.wikipedia.org/wiki/Musical_system_of_ancient_Greece

Mind that there are 2 definitions of "diatonic scale":

a) the traditional classical definition

b) the modern extended definition

I'm posting here a google translation of the German wiki site, which is more detailed than her English sister:

The root tone series c-d-e-f-g-a-h corresponds to the white keys of the keyboard. These are the notes of the C major scale.

Diatonic scales are usually seven-degrees (heptatonic) scales, dividing the octave space into five whole and two semitones. They differ from non-ionic ladders by the following necessary characteristics:

All scale steps are derived from different root tones, which is externally reflected in that their names all begin with different letters. Between adjacent stages, there are no excessive or reduced intervals. For example:

The "classical" diatonic scales (major, minor, and church modes) additionally fulfill the condition that they can be composed (by adding a further whole-tone step) of two diatonic tetrachords. Also, the tones of these scales can be obtained by fifth layering.

The seven pitches of any diatonic scale can also be obtained by using a chain of six perfect fifths.

Any sequence of seven successive natural notes, such as C–D–E–F–G–A–B, and any transposition thereof, is a diatonic scale.

In extension of this original strict definition, sometimes even those scales are called diatonic which merely fulfill the condition of subdividing the octave into five whole and two semitones. Examples are the acoustic and the altered scale.

In addition, according to the present understanding, also scales can be regarded as diatonic, containing less than seven tones, such as. As the anhemitonic-pentatonic ladders, which divide the octave space in three whole steps and two thirds.

The diatonic scales in the narrower sense ("classical" diatonic scales) also include the church tones and the modal scales that today draw on them.

In the broader sense these scales are also diatonic:

the melodic minor scale upwards

the altered scale used mainly in jazz

the acoustic scale

the anhemitonic-pentatonic ladders

Not diatonic scales

These scales are not diatonic or not completely diatonic:

the harmonic minor scale, as it contains an excessive second (hiatus)

the gypsy ladders, because they contain excessive seconds (hiatus)

the whole-tone scale, since the last whole-tone step needed to reach the octave is in reality a diminished third (in the notation ais-c from c)

the chromatic scale

the modes with limited transposition possibilities of Olivier Messiaen

My conclusion:

Now deriving chords and intervals of these scales in a broader sense (extended definition) will lead to a confusion that we better talk about diatonic chords and intervals derived from the scales defined in the classical meaning.

The following list e.g. counts all 3 kind of minor scales to the diatonic scales:

Diatonic scales (German)

4
  • Agree with most of this - but don't think an aug2 from harmonic minor is enough of an indiscretion to exclude it!
    – Tim
    Apr 5, 2020 at 10:17
  • I’m not sure whether I got what you mean: You say the harmonic minor scale is diatonic too. O.k. , this will then be the extended meaning, yes? Apr 5, 2020 at 12:50
  • Some believe it is, some don't. I do.
    – Tim
    Apr 5, 2020 at 13:08
  • Smile ... as some questions here (especially of mine) are still disqualified as opinion based. ;) Apr 5, 2020 at 15:29
4

I agree with the Encyclopedia Britannica in restricting the definition of diatonic and chromatic to within the Western heptatonic scales.

The way the Wikipedia article opposes diatonic and chromatic in different applications also makes sense to me. This distinction evolved at the interface of when melody gave rise to harmony in the Western context, as key-consciousness became a factor, marking the start of the Common Practice period in the early Baroque era.

I would say that it is instructive to conceptualise this as a continuum, from the purely diatonic like 1st species counterpoint to heavy late Romantic chromaticism. Beyond these borders, pentatonic and modal tonality (with their absence of key-defining leading tones) past the diatonic border, and whole-tone harmony and atonality at the chromatic end.

Hence, within the boundaries of "Common Practice" in the Western canon, there are varying definitions of "diatonic", which the Wikipedia article gives a good rundown of. All agree that:

  • major scales
  • natural minor scales

... form the bedrock of what it means for an interval, a melody and a harmony to be "diatonic". Where they differ is how "diatonic" the other forms are:

  • harmonic and melodic minor scales;
  • augmented fourth chords and half- and fully diminished 7th chords, which arise from these extended minor scales;
  • other chords with accidentals outside the key (Neapolitan 6th, secondary dominants, chromatic mediants).

So different authors do have different definitions.

4

The general way that this is used is "using only notes which can be found in one of the major scales".

I think most people would take issue with Brittanica about the pentatonic - the pentatonic scale is a subset of a major scale, and therefore is diatonic, at least in my reasoning. (Whole tone is not though.)

A diatonic instrument is only capable of making major scale notes (leaving aside things like note bending). For example, diatonic harmonicas (versus chromatic ones).

Diatonic harmony contains chords (or sequences) which again only use major scale notes. Therefor, a min 7 chord is diatonic, as it uses a subset of the major scale notes (even though it does not use all of them).

No, harmonic and melodic minor are not diatonic. Neither is a minor chord followed by its dominant seventh. For instance no major scale contains these notes:

C / D / Eb / F /G / B

which are the notes in Cmin and G7.

-1

This is something that gets needlessly convoluted:

Diatonic simply means "belonging to a specific scale". My only references are the ones you've already posted, it's just that the definition gets skewed.

So B is diatonic in the key of C major, but G# is not. Eb is diatonic in the key of C Dorian, but is not diatonic in the key of C Lydian.

Whether a note is diatonic, therefore, depends on the note and the scale in question. You can't answer the question without both pieces of information.

1
  • 1
    It doesn't as there are lots of scales that aren't diatonic. Mar 3, 2020 at 1:07
-3

If you are willing to dare to challenge the dictionary definition you will find great value in accepting that diatonic is the same as 'di-tonic' – two tonics.

The two tonics are the first and fifth notes (I and V) as the V is harmonically closest to the I.

This is exactly why modern music theory is built on fifths – the fifth is where the scales increment – first C, then G is the next key, then D ... a fifth up, etc.

Now look at the composition of each fifth interval, such as CDEFG.

CDEFG's intervals are WWHW.

WWHW is also the interval pattern for GABCD.

Point being that major scales are build on overlapping patters of WWHW ... each V becomes the I for the major scale of the next incremental key (incrementing by the number of accidentals).

Would you like minor you say?

Ok, why not go backwards ... WHWW ... going down by fourths.

Each of these two patters is 'diatonic' because the beginning and end are the same ... a "W" ... like a puzzle piece so they fit end to end in each direction... like a puzzle piece where the left side is the inverse of the right side so they perfectly fit.

A pattern like HWWW isn't diatonic. It's the pattern of a mode.

A pattern like HHWW can't work because the two H's are together. In diatonic patterns:

1) Matching ends so they can fit like puzzle pieces.

2) Two or three W's between the H's ... as they overlap at the V note ... the I note of the next key is the V note of the previous key.

11
  • 1
    Please stop trying to push this definition. The route word is dia which is "through, across" rather than di which is two.
    – Dom
    Mar 3, 2020 at 1:32
  • Well as it's correct ... and I've explained why ... and no one has yet explained why it's wrong, why are you opposed to this insight? Two ... across ... point to point ... two tonic system... I explained why ... the explanation is why the conventional definition falls short. If the explanation is correct then why is the conclusion unreasonable? The only reason it's become dia-tonic is to convey that it's reversible. What's unreasonable about this? Mar 3, 2020 at 4:37
  • There's a lot of problems: 1. you are double counting. The full pattern WWHWWWH so the WWHW double pattern you mention makes no sense in terms of the construction and the fact it overlaps with the next scale pattern built on the dominant is irrelvant. 2. There's not two tonics. There's one which is the first scale degree by definition. The fifth is the dominant. 3. The circle of 5ths has nothing to do with scale construction and even if you base scale construction on it, you would end up with the lydian scale not the major scale. There's more, those are what will fit in this comment.
    – Dom
    Mar 3, 2020 at 5:30
  • 1. Oops. you're right. They don't overlap. Wrong word. They concatenate to bring you to the tonic of the next key. WWHW + WWHW = WWHWWWHW ... brings you to D, two keys upward. 2. There's no two tonics. The V becomes the I for the next key. 3. Yes, the circle of fifths isn't used for scale construction. Mar 3, 2020 at 12:31
  • All of your explanations are based on the circle of 5ths which has nothing to do with the definition of diatonic. So if you agree with 3, why does most of this post talk about the circle of 5ths? You are stretching the definitions you are using so much it muddies the water which is the exact opposite point of this question.
    – Dom
    Mar 3, 2020 at 14:09
-3

Thank you all for nice thread of discussion. There is some light on that. I shall explain here in most simple and most convincing way.

Actually, it is quite easy to understand why it is called as di-a-tonic scale. It basically means a scale with TWO (di) tonic centers.

The one tonic (root) note is, (say) 'C' in C-Maj. scale. The second tonic (root) note is (say) 'A' as root note in A-min. scale.

Thus same set of notes creates TWO (di) centers forming major and minor scale hence called diatonic scale.

1
  • It's a nice way to think of it, but by this definition, each scale actually has seven "tonics", one for each mode. @Athanasius wrote correctly when he explained that Western music's "diatonic" scales were named for the ancient Greek "diatonic" scales. A brief explanation and etymology is given here: en.wikipedia.org/wiki/Diatonic_scale#History
    – Aaron
    Jul 28, 2020 at 5:06

Not the answer you're looking for? Browse other questions tagged or ask your own question.