# Minor Key (and other key?) Roman Numeral Analysis

What is the general standard in RNA for the natural minor key, as well as keys that are neither major nor nat. minor?

So in studying RNA, I found the major key pretty straightforward: I ii iii IV V vi viio.

But the minor key made things a little more confusing:

situation = use RNA to describe chords built from the notes of the natural minor key, (e.g. C nat. minor key),

people might generally do one of the following:

method 1: ("Traditional notation" in above table)

use the same scale degrees from the (same root) major key (C major key), altering them to describe the natural minor key (C nat. minor key).

If C major key's last(?) diatonic chord, the leading tone diminished triad (B D F) = viio, then C nat. minor key's last diatonic chord, the subtonic major triad (B♭ D F) = ♭VII.

Chords with non-CDEFGAB roots need accidentals. e.g. "7" = last diatonic note of its major key / leading tone (B), while needing accidental to describe last diatonic note of nat. minor key / subtonic, so "♭7" (B♭).

or

method 2: (not in the above table)

assign new scale degrees to the nat. minor key, so diatonic triads: i iio III iv v VI VII.

C nat. minor key's last diatonic chord, the subtonic major triad (B♭ D F) = VII.

or

method 3: ("Alternative notation" in above table???)

know what the triads are, and chord quality is not listed. Unless otherwise specified, iii will always mean a diatonic 1 3 5 chord built off of the mediant (root). Upper/lower-case ignored?

C-minor key: vii = subtonic major triad.

Methods 1 and 2 are more or less clear, but method 2 seems to me more intuitive when building diatonic chords in its own key. If RNA is dependent on a key, and the key does not change throughout a given sequence, then isn't method 2 the least ambiguous (shows chord quality) and most immediate/practical/relevant (it uses its own key)? I feel like method 1 sort of deifies(?) the major key, like an absolute point of reference.

The table above... I'm guessing the "Chord symbol" row is always upper-case/non-case-sensitive, and needs accidentals for notes non-diatonic to its (same root) major / taking the scale degrees from its major key. I'm not quite sure as for the "Alternative notation" section (shouldn't "I" be "i"? how to show non-diatonic chords' qualities?)

My main question is, apart from if hopefully I'm understanding things correctly,

What is the general standard in RNA for the natural minor key, as well as keys that are neither major nor nat. minor? (harmonic minor key(?), ascending melodic minor, non-western pentatonic(?)...)

I assume I can use method 2 for any scale, make it a key, forget about the accidentals, assign tonic = I/i, ... subtonic/leading tone/whatever last diatonic note = VII/vii (if heptatonic), and express chord qualities using upper/lower-case and symbols. This seems most intuitive (to me).

But resources I found use method 1 to describe chords built from the harmonic minor scale, but using the natural minor key as the point of reference, instead of the major key. Is this because there isn't a harmonic minor key? Or is there but just not used? Is the idea to use the closest key (i.e. most diatonic overlap), but that the key must be reasonably familiar to most (i.e. major/natural minor)?

e.g. if I wrote a chord progression in a song that uses only the diatonic notes of the A harmonic minor scale, how would I be expected to describe it using Roman Numeral Analysis?

A) method 1, A major key as reference (as in borrowing its scale degrees)

B) same method 1, but A natural minor key as reference

C) method 3, in A harmonic minor key. Can I do this? Does anyone? Would anyone care?

I would love to ask more, but this is perhaps already too long. Thank you immensely, I'm beyond thankful that this site exists!

• +1 - good question! And where does the melodic minor (ascending/jazz) figure? Your last example - key sig. of G# is weird.
– Tim
Commented Feb 5, 2020 at 7:51
• Hi, thank you so much! I'm not quite sure what you're asking, though! Do you mean if I made the asc. melodic minor into a key and then based the RNA notation on that? I used the A harmonic minor scale (arbitrary example) as a key (G#), as one would do with a natural minor or major scale, and then applied method 3, as in chord qualities aren't notated but are implied through diatonic 1 3 5. (like how some might label C-major leading tone dim. triad (B D F) as just VII, or vii, even though vii<sup>o</sup> is less ambiguous!) It is a little weird! I can feel my brain getting warm. Commented Feb 5, 2020 at 8:30
• RNA basically uses i, I, io, I+ for the main triad forms, based obviously on the 3rd and /or 5th intervals, but as you ask, which scale notes are applicable...
– Tim
Commented Feb 5, 2020 at 9:02
• C) method 3, in A harmonic minor key. Can I do this? Does anyone? Would anyone care? I think Bartok did it. Nobody can forbid you to notate just one sharp in the beginning of the staff near the clef. I wouldn't care. Commented Feb 5, 2020 at 17:04

There is not one absolute standard. But Kostka/Payne's system in Tonal Harmony will allow you to write a unambiguous symbol for any of the four triad types (major, minor, diminished, augmented) on all twelve possible roots within any major or minor key signature. It will also handle diatonic seventh chords and at least a large variety of non-diatonic seventh chords.

Upper case means major, lower case minor, `o` means diminished, `+` means augmented, and roots chromatically altered from the key signature are prefixed with an accidental. There are a few additional symbols to deal with seventh chords, but let's skip that.

Kostka/Payne uses this for minor keys...

...the circled ones are most commonly used.

(I take issue with one point. The minor chord rooted on the dominant. In minor keys, that chord in first inversion is common with a descending bass line, like `i v6 iv6 V`. But that's only a detail about what is most common.)

I think part of your confusion is about minor keys versus the three types of minor scales. At the risk of starting a music theory argument, chords are not built from fixed scales, they are intervals above a root. That is especially apparent in minor where the sixth and seventh degrees are flexible. Kostka/Payne's RNA tells you: key signature, root, chord quality, and inversion. So, `Cm: #vi°` written out means: in `C` minor, a diminished triad, rooted on the raised sixth scale degree `A` natural. That gives us a chord. It has nothing to do with a scale. Some will say it's from the ascending melodic minor scale, but I think that is not a helpful way to understand minor key harmony.

The only odd-ball case with their system in minor is the leading tone triad. That chord's root does require an accidental to alter it from the key signature, but they don't prefix the `vii°` with an accidental.

In your three examples, the first two match up with Kostka/Payne, with the exception of not prefixing the leading tone triad with a sharp.

The third example seems confusing, because it uses all lower case for three different chord qualities.

Do you have to use Kostak/Payne? No. But it is a widely known textbook. Most of the links you listed included symbols that match Kostka/Payne. I think that shows they are presenting a pretty well accepted standard. If you don't use their textbook, I would at least get some other well known resource. Something authoritative and preferably published.

There isn't a 'general standard'. Hence the confusion.

I prefer the system ('traditional' according to Wikipedia) where scale degrees are related to the major scale and minor/major chords are lower/upper-case. Minor or major dominant chords are v or V respectively. (Does a 'natural minor' v deserve the functional label 'dominant' anyway?)

But the other system does get used, and is doubtless considered 'standard' in some circles.

• I remember seeing Method 2 get used in my Royal Conservatory of Music Harmony textbooks and exercise books. Nowadays, I tend to hedge my bets by putting left-side accidentals in parentheses for minor keys. Commented Feb 5, 2020 at 17:11

Excellent question! Unfortunately, there does not appear to be a universal standard within the musical community at large. Even looking around this site will give varying systems of notation. However, allow me to provide an argument for the system I prefer:

I tend to favor the system labelled above as "traditional" (defining every symbol relative to the tonic major scale), because it needs no external reference to a specific tonality. If you know what the tonic note is, all chords can be converted to RNA, even those with little relation to the tonic. [B D♯ F♯] in the key of C minor to me is ♮VII, or even just VII if I know I will be understood. Defining everything in relation to the major scale eliminates the possibility that one chord label can mean two different things; under other systems, is III in some key of C an E major triad or an E♭ major triad?

One disadvantage, though, is that one must write out accidentals on certain scale degrees of the relatively common minor scales - even something as simple as E♭ in C minor must have an accidental in its symbol (♭III). To me, this is an acceptable trade-off for the security of perfect clarity; to others, this is an inefficient system bogged down by the need for explicit labels for simple concepts. I myself also like that the system performs very well in complex harmonic landscapes that stray quite far from diatonic major tonalities.

Of course, my preferred system is not an accepted standard, so it's generally a good idea to be able to at least understand what the many systems are trying to communicate. In certain circles, traditions such as RNA nomenclature can vary wildly, some even with regional modifications (check out the Nashville Number System for a well-known specific example)! Fortunately, all RNA systems were created in order to make sense, and often times context will make it very obvious what the symbol represents. It all comes down to how explicit one wishes to be and also what one believes their intended audience will understand.