Now, this would be an easy enough question if:

  • The piece in question had a title containing the key, or
  • The piece would have a home key, modulate to another key and then come back to finish in the home key.

My question is concerned with all other instances regardless of genre.

Instead of saying "This piece is in key X", is saying "this piece has multiple keys, X and Y and Z" the more correct version?

If these pieces are purposefully reduced to a single key, what may be the motivations behind it?

  • 1
    Maybe compounded more so by 'Symphony in X'?
    – Tim
    Mar 21, 2021 at 11:37

7 Answers 7


This answer covers key labels and descriptions with a heavier focus on popular music styles, as opposed to classical music:

I think the handiest way to lump all of that information together is usually "starts in X" rather than "is in X", when pieces of music have multiple keys throughout. Obviously if there is one main key (jazz standards which modulate in certain sections but return back, for example "Beyond the Sea", would fit this bill), then "this piece is in X" makes sense. But this question is asking specifically about songs and music where that is not the case.

If there are multiple keys and no clear primary key with which to label the piece, it is probably best practice to use a more descriptive statement. Here are some situational examples of what I mean by that, since every song will be different:

The Beatles' song "Let It Be" is in the key of C major. (This is not up for debate.)

"Enter Sandman" by Metallica is in the key of E minor. (The parts in F# minor lead strongly back down to E minor.)

Beyonce's "Love On Top" is in C major. (Many modulations up a half-step at the end, but still primarily C major - the ending is clearly a repeated variation of the main part of the song, so C has a convincing claim to being the key of the song.)

"Layla" by Eric Clapton is in D minor, but drops everything down a half-step to C# minor in the verses. (It makes sense to say that D minor is the primary key, since it is only the verses that temporarily shift.)

Frank Sinatra's recording of "Fly Me to the Moon" can be viewed in C major or in its relative key A minor. (There is a continuous blur between the two relative keys.)

"God Only Knows" by the Beach Boys has E major competing with A major to make a weak and fluid tonal center. (Though E is likely the most viable candidate, the song weakens the key by constantly flirting with other keys such as A major and avoiding strong resolutions to the E chords.)

"Bohemian Rhapsody" by Queen begins by tonicizing the key of Bb major and modulates to various other key centers. (It is not feasible to pick a single key to label the entire song with, but since it starts in one clear key, this can be used to communicate the necessary information.)

"Giant Steps" by John Coltrane cycles between the keys of B major, Eb major, and G major. (It is useless to single out any one of the three key centers.)

Schoenberg Piano Concerto Op. 42 is atonal. (Defies any tonal logic, and no key could make a claim to represent this composition.)

The concept of a key should not be applied to "4'33"" by John Cage. (Assigning a key signature to silence is pointless.)

Some might disagree with some of the descriptions I gave, but this is fairly subjective, and my point is that each case will be different and require different methods to communicate the necessary key-related information. There are many different ways that multiple keys could be present in a piece of music, and they must be described situationally as a result. It is not always going to be possible to state unequivocally that "X key is the key of this song".

Generally, the most important purpose for giving one single key for a song is to get every musician on the same page as to what to play, and in cases with multiple keys present, as long as one key is given for one agreed-upon moment in the music, every other part of the music can be deduced with some transposition. The trouble with that often lies in two places: either it is not obvious what the key at a particular moment is, or there is no obvious moment to select as representing the key of the entire piece, or potentially both. Thus, occasionally music may require a more detailed description of its varying tonalities in order to communicate unambiguously what to play.

  • 1
    Nice list. I'd add 'Unforgettable', seems firmly in G to start, ends up in C. Written with a C key sig., though. And I can't figure out where it modulates! 4'33" - just checked, and it's written in 7#, modulating to 6b, but there's a version in key C for poor sight-readers.
    – Tim
    Mar 22, 2021 at 8:27
  • And strangely, we tend to consider the last chord as representing the key, rather than the first.
    – Tim
    Mar 22, 2021 at 8:34
  • 1
    I originally asked this question from a "music automation" standpoint - for automatic key detection, the standard procedure is to come up with a single key, even if the piece contains many modulations, without a clear single key label. I was mainly questioning that standard, and it seems to me that it is indeed a problem yet to be solved... Mar 22, 2021 at 11:17
  • 1
    @MorganKendal Unambigious automatic key detection: Reinterpret the key as "the key signature that minimizes the number of accidentals". Not quite the same as the traditional meaning, but it has an objective metric. Less so if the music has many modulations, since you can reduce the number of accidentals by switching keys more frequently. But that's akin to compression, there is probably a way to describe an optimal number of key signature changes. It needs a 'cost' of key changes so it doesn't produce a key change for every single accidental. Mar 22, 2021 at 12:47
  • 3
    @MorganKendal if you're doing any sort of "music automation" I'd recommend not using "songs" as the smallest "collection of sound". Songs can be far too complicated. I'd use sections of music that combined into larger ones. It makes sense for most music to talk about the key signature of one section of music, even in extreme cases like "Giant Steps". I'd say its very rare not to be able to break music up into discrete sections with describable characteristics, but common not to be able to apply those characteristics to the larger work.
    – bendl
    Mar 22, 2021 at 14:33

It's a historical convention. Generally, multi-movement pieces are named using the main key of the first movement. The convention dates back (at least) to the suites for various instruments written as the idea of "keys" being important was gaining interest. Old Gregorian Chants were often named by appending the mode to the first few words of the chant. This practice has continued.

The question is also interesting in that some chants transposed modes (and there was a difference in B and Bb) but the naming convention has stuck for at least a millennium.

  • There also used to be a rule that a piece had to end in the same key in which it started. This included multi-movement pieces, so the last movement had to be in the same key as the first. The only common exception to this rule is that pieces in a minor key would often end in the (relative or absolute) major. This key change might not happen until the very last note in some cases (J.S. Bach did that a lot). Other exceptions existed, but were very rare until around the 20th century or so. Mar 22, 2021 at 13:33

Is "this piece is in key X” a valid statement?

Sometimes, at least.

  1. Baroque-era minor-key pieces frequently end with a major chord. Obviously this stretches the limits of the question, since it can't really be said there are multiple sections in different keys, but it's worth a mention insofar as the piece does technically end in a different key that it began and is still consider to be "in minor".

  2. Chopin Ballades #2 (in F major, Op. 38) and #4 (in F minor, Op. 52) start in one key but end in another. In Ballade #2, the first 46 measures are in F major, but the remainder of the piece (158 measures) is in A minor. Ballade #4 begins in C major (7 measures), but is predominantly in F minor (232 measures).

  3. George Gershwin's "Impromptu in Two Keys" again stretches the question, but is worth noting. In this case, the right-hand part is in D major while the left-hand part is in Eb major. Although the piece ends with a cadence in Eb major, it would be misleading to say it's "in the key of...." This might be a case where it would be appropriate to say "This piece is in the keys of D and Eb major."

  • 1
    I'm certain the F major initial material returns in the middle of Chopin's Ballade No. 2, and I've always interpreted the start of Chopin's Ballade No. 4 as a giant dominant prep. What's more interesting, IMO, is that the start of Chopin's Ballade No. 1 in G Minor can easily be interpreted as being in A flat major.
    – Dekkadeci
    Mar 21, 2021 at 15:07
  • 1
    @Dekkadeci Yes on both counts: In Ballade #2, the F Major material has a substantial return in the middle of the piece as well as a return, transformed to A minor, at the end. And in Ballade #4 it is an extended dominant, though that isn't clear until the entrance of the main theme. Nice call on the Ab major opening of Ballade #1!
    – Aaron
    Mar 21, 2021 at 16:03

The main key is normally the opening movement and the final section.

The other movements are usually in the relative keys or mediant keys.

E.g. Beethoven’s Piano concerto Cm (2nd movement in E major!)

Also the different sections in the sonata form are modulating to the dominant or relative key.

In music there is much more convention than theory based. Bach‘s prelude for lute (bwv 999) transcription for harpsichord begins in Cm ending in G major.

The purpose of this labeling may be to avoid endless titles in the announcement for concerts and sheet music. (Speculation!)


As a short addition to the answers: In former times, it was not uncommon for composers themselves to designate a piece as, for example, "ex C", Latin for "out of C" meaning "starting out from C". So a piece might have been called "Prelude and Fugue ex g". Thereby the composer accounts for all possible modulations ensuing.


You've already considered the case of a classical multi-movement work with a well-defined tonal 'home' key, and a piece mostly based on a minor key but with a final section like a prolonged 'Tierce de Picardie' (Tchaikovsky 5th Symphony springs to mind). These often have a key as part of the title. Obviously this is valid.

Otherwise, well, it's valid if it's valid. If there's a demonstrable 'home key' it's valid to say so. If there isn't, well, don't!


It's a convention of structure and naming, not a description of how much material is in any particular key.

Tonal music conventionally starts and ends in the same key. That key is often used in the title, like "Sonata in C Major."

In terms of how much material is in various keys through out a piece of music, even a typical classical sonata might have, very roughly speaking, only half the material in the starting/ending key. A title like "Sonata in C Major" is not at all indicating how much material is C major, or other keys.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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