Why do we need key signatures such as E♯, B♯, C♭, and F♭? Take a look at the scales for E♯ and B♯:

enter image description here enter image description here

E♯ has 4 sharps and 4 double sharps. The key of F is exactly the same, and it only has one accidental in the key signature (B♭). I understand why it's sometimes necessary to use the note E♯, but using the E♯ key signature seems like it makes things more confusing.

B♯ has 3 sharps and 5 double sharps, while C major has no accidentals in the key signature at all! Which do you think is easier to understand?


9 Answers 9


They're just extreme versions of enharmonic scales-that is, scales that exist in an identical sounding key but are spelled differently. It simply has to do with the fact that we have to have as many keys as possible to allow correct spellings of chords and whatnot. For example, A# minor is the relative minor of C# major (they share all the same notes). Now, the key of E# major is ridiculous on its own, but E# is the dominant of A# minor, and should we be writing a sonata in an old style, then we would need E# available to us as it is a necessity to writing in A# minor for any length of time. Like I said, these are kind of extreme keys: they exist primarily for theoretical reasons (and because theory teachers and professors say they have to exist) as opposed to everyday use and practical ones.

  • 1
    A nice "practical" illustration is the E flat minor prelude and D sharp minor fugue in book one of Bach''s "48" (BWV 853). The prelude has a climax about 3/4 of the way through, on the "Neapolitan 2nd" chord of F flat major (i.e. 9 flats). The fugue has a set of entries in the dominant-of-the domimant key of E sharp minor (i.e. 9 sharps)..
    – user19146
    Mar 31, 2015 at 20:57

Regarding key signatures, you'd be hard pressed to find a musical work which has a key signature in either of those keys you've mentioned. The screenshot of your examples are even without any key signature; it's difficult for a staff to accommodate that that many accidentals and they were likely never intended to be able to encompass every possible key signature. Your example instead illustrates major scales based upon them. This is likely done as an exercise for a student of music theory in order for them to engage with pitches not commonly found in music, and a major scale is a convenient means of summarizing several of them. As you said yourself, notes like E# are sometimes necessary. The same goes for each of the other pitches found in your example.

C-major and F-major scales are undoubtedly more readable than the same sort beginning on B# or E#, but musicians should nevertheless become comfortable with reading pitches of double sharps, flats, et al, despite their infrequency. Modal mixture, altered chords, and composers that avoid enharmonic spellings spring to mind as a few reasons to be exposed to these sorts of pitches.

  • 3
    It's possible (and not uncommon) to have a major scale passage within a section of music that has a different key signature. If a piece of music in C# major (with sharps F#, C#, G#,D#, A#, E#, B#) has a scale passage starting on a note a four half-steps above the key note, notating that as (E# F## G## A# B# C## D## E#) would require accidentals on four notes (since E#, A# and B# are in they key signature). Notating it as (F G A Bb C D E F) would require accidentals on all seven notes. Even notating it as E# G A A# B# D E E# to exploit the sharps in the key signature would still...
    – supercat
    Dec 17, 2017 at 18:34
  • ...likely require putting accidentals on more notes than would be needed using double sharps, since the Anat and Enat would cancel ut the A# and E# in the key signature, making it necessary to restate them on the next note.
    – supercat
    Dec 17, 2017 at 18:36
  • @supercat but no piece of music should ever be written in C sharp major.
    – phoog
    Feb 2, 2021 at 10:48
  • @phoog: If a piece is going to modulate between parallel major and minor keys, and the key note pitch s supposed to be between C and D, I would think that having it modulate between C# major (seven sharps) and C# minor (four sharps) would be cleaner than having it modulate between Db major (five flats) and Db minor (eight flats), or between Db major and C# minor (causing the staff position of the key note to change when modulating to a parallel key).
    – supercat
    Feb 2, 2021 at 17:42

Another issue not yet mentioned is that, especially when using a computer to edit music, one may want to perform a sequence of transposing operations that create weird key signatures, but only normalize key signatures after performing all the steps. If one transposes a piece of music up, and then transposes a portion of it down by the same amount, it will often be desirable that the downward transposition should precisely cancel the effect of the upward transposition. If key signatures are normalized between the two operations, however, the operations might not cancel.

For example, if a piece of music had section which switched a few times between B major (five sharps) and B minor (two sharps) and was transposed up a major third, that would yield key signatures of D# major (nine sharps) and D# minor (six sharps). If the D# major portion were normalized to Eb (three sharps) and then the section was transposed back down, the result would be a mixture of Cb major (seven flats) and B minor (two sharps), with a consequence that matching notes in the major and minor key would appear at different staff positions.

One would generally want to normalize key signatures before printing them out for purposes of performance, but being able to have unusual key signatures during the editing process can allow computers to maintain distinctions during editing (e.g. the difference between Cb major and B major).

  • Normalizing the key signatures at every step does not break the round trip in this example because a normalized key signature should not have more than six sharps or flats. A better example would be G flat major and F sharp major, which both have six.
    – phoog
    Feb 2, 2021 at 10:45
  • @phoog: C# major (seven sharps) is rare, and I'm not sure I've ever seen it in print, but Cb major (seven flats) is common, and I've seen it in print much more often than B major or F# major.
    – supercat
    Feb 2, 2021 at 15:47

Adding to the existing answers, they are useful in tuning systems that use more than 12 notes.

For instance, 19-tone equal temperament will need key signatures with up to nine sharps or flats. That is, it needs key signatures for:

  • F♭ major (8 flats, now one chromatic semitone above E major)
  • B𝄫 major (9 flats, now one chromatic semitone above A major)
  • G♯ major (8 sharps, now one chromatic semitone below A♭ major)
  • D♯ major (9 sharps, now one chromatic semitone below E♭ major)

These scales logically exist but, you're right, it's hard to imagine a circumstance where we'd need them! Occasionally it's appropriate to use a scale outside the scope of key-signatures (they only go as far as 7 sharps or 7 flats, we don't use double sharps or flats in key-signatures). G# major is not ridiculous.


Once you start talking about keys and scales you're basically ticking the box that says 'I agree never to use the same letter name consecutively'. Practical considerations, such as ease of reading, do not trump this. For better or worse, this proscription of consecutive letter names lies so far back in the musical theory genome that to upset it would lead to a period of chaos; so we persist.


Why are key signatures like E♯ and B♯ necessary?

They're not necessary, for which we should be thankful, because for all practical purposes they do not exist.

Theoretically, they do exist, because we can extend the circle of fifths infinitely. You could have a key signature of F♭♭♭ if you wanted, but only in theory. In practice, there is no good reason for this, much less a need for it.

There is sometimes reason to spell an individual pitch as B♯, or some double sharp, or what have you, and there might be occasion to spell a chord with such a root, but there is never reason to spell an entire key that way. In fact, key signatures should normally be limited to six flats or sharps. This is why the parallel major of C♯ minor is D♭ major (for example, in Beethoven's "Moonlight" sonata).

  • 1
    The parallel major of C# minor is C# major, not Db major. Just because Beethoven wrote the second movement of his piano sonata Op.27 No.2 in Db major, doesn't make those two keys the same.
    – Divide1918
    May 27, 2021 at 10:23
  • 1
    @Divide1918 but they are the same. How would one play the second movement differently if it had been written in C sharp major? You press the same keys; you get the same pitches. Sure, in some theoretical sense there's a difference, but Beethoven didn't choose the keys to invoke that difference. He chose them to make the music easier to read and because C sharp is the same as D flat.
    – phoog
    May 27, 2021 at 12:50
  • 1
    They sound the same because the piece was written for piano, an equal temperament instrument. But just because they sound the same, doesn't mean that they are the same. Parallel keys have the same tonic by definition, so you're basically saying that C# and Db are the same note, which is absurd.
    – Divide1918
    May 27, 2021 at 13:23
  • @Divide1918 it is not because it is an equal temperament instrument but because it is a 12-tone instrument. C sharp and D flat are the same note, on every 12-tone keyboard and on most keyboards with more than 12 tones. Do you imagine that Beethoven used D flat because he had some subtly different tonic in mind? Of course not. He used it because it's the same note and it has a simpler key signature. Even with strings or voices or trombones, if you were to play a piece with movements in these keys using just intonation, you'd take your root C sharp and D flat to be the same frequency.
    – phoog
    May 27, 2021 at 19:00
  • @Divide1918 if C sharp and D flat are different, how are they different? The frequency is identical. There is no difference in their tendency of chromatic motion (up or down), because they are in both cases the tonic pitch. What benefit do we have, in this context, by maintaining a distinction? There is none. If we were talking about a piece in G, or in any other key, there would be, but we're not.
    – phoog
    May 27, 2021 at 19:05

There are no key signatures of E# or B#! But there are B# and E# in REAL key signatures (and Cb and Fb too). For example, C# Maj has all notes sharped, so that is the easiest example. NO real key signatures have double sharps and these keys are NOT needed. Yes some music had double sharps, just NOT in the key signature. There are 7 sharps and 7 flat and one all natural (C Maj) key signature. That is all! It is important to note that there used to be a note between B and C and between E and F to facilitate different tunings to make the music more harmonious, on Harpsichords and I believe Clavichords, where the player would choose one or the other to fit with the remaining notes being played. Different tunings (temperaments such as even and equal etc) have compromised the notes enough so as to prevent this need. I think the old kind was "Just temperament" and "Pythagorean" or other tunings. Hope this makes sense...

  • +1 for pointing out that the OP's notion (key sigs for E# major and B# major) was wrong, and discussing what actually exists (keys which include an E#, possibly with a B#, too).
    – Rosie F
    Dec 16, 2019 at 10:27

off drpylon's A#m example:

Without this "strange" enharmonic notation, the chord built on the 4th degree would be notated as: natural F, G#, natural C. Looking at this on a score would not look right; I read somewhere that the one of the purposes of enharmonic notation is to allow the writer (and reader) the ability to visually see what appears to be a normal chord at first glance, such as while sight reading.

In other words, it would be easier to sight read e#, g#, b#, rather than natural f, g#, natural c, hence an actual advantage of this strange enharmonic notation.

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.