Writing Accidentals

Hi,

Just a a generall question that has always niggled at the back of my mind.

When writing accidentals, are there any 'rules' for deciding when to use the enharmonic equivalent or not? For example why would a composer choose to write Dd instead of C#?

Thanks alot.

Typically it goes based on function. For example, if you were in the key of C major and wrote a ii#5 chord, you would use an A#. If you were writting a I7 chord, you would use a Bb. As you can tell, the function of the note(s) in these examples is very different.

Also, when there's no function system or tonality altogether, there's an open choice. Sometimes composers choose to write sharps instead of flats because they can sound different in certain instruments, but also because some people think things such as sharps being more "aggressive" than flats. I don't know.

Then there's the ease of reading, which I guess varies. It's good when you change the actual notes when going up or down, but not entirely necessary. Like, C C# isn't so great, you can probably read it better as C Db. But it's pretty much free, though it's good to avoid using things like E# and Fb, or Cb without a good reason. I guess overall it depends what systems you're using. If you're using 12 tone, your row is going to contain all the accidentals as you will always write them (C# as opposed to Db if you wrote it like that in the row.)

If it's something using functional harmony, the keys and chords give what accidentals they use. Hence a C sharp major chord isn't written with flats, obviously.

My position:

In the score: Derive from the harmony, as Nik suggests....

BUT,

In parts: make it easy to read. Most second tenor saxophonists don't care that they have the b9 of the chord, if it's a Db in a sea of sharps, write a C# for the sake of readability.

:whistling:

The score must mimic the parts though. Never have parts that are different than the score or else most conductors will get very very annoyed. Beware of flying batons!

Both Robin and Justin are correct... first, the score must match the parts, and second, make it easy to read. It's okay to "misspell" a chord if the ultimate effect is easy to read.

To go further with this topic: your accidentals should be logical. When going up a chromatic scale, it looks better to read C C# D D# E F F# G G# A A# B C (sharps going up). Going down the scale, it looks best if it reads C B Bb A Ab G Gb F E Eb D Db C (flats going down). There are exceptions, of course, and you'll need to tailor lines so that they look "right" in whatever key you're in. The reason this looks better is that you don't have to throw in a lot of natural signs, so the music doesn't look as "cluttered" with accidentals.

The most important aspects have been mentioned, however it should be pointed out that in harmonically complex passages of classically tonal music you often have to decide between two principles of writing accidentals, while being forced to ignore the other: Either you focus on vertical harmony or on the melodic line/voice leading. In the first case, as has been mentioned, you write the accidentals according to the harmonies you write. In the second case you look at where each tone moves/resolves to: Accidentals "bend" the notes in a certain direction, so you try to resolve the notes in this direction, so that flats are resolved downwards and sharps upwards. So you'd rather write the melodic line d#-e-f#-g than eb-e-gb-g. This of course only applies when a note actually moves to a neighboring half tone. You'd still rather write g-f#-e than g-gb-e for two reasons: first, it is seen as preferable to actually move to different diatonic base tones if possible and not to change tone alterations on one single tone; second, certain intervals are commonly seen as simpler than others: g-f#-e only consists of minor and major seconds, both "simple" intervals. G-gb-e however consists of an augmented unison and a diminished third, which are considered more complex and less "natural", since they aren't intervals one would commonly hear as such if played alone (i.e. if you played gb-e on a piano most people would hear it as the major second f#-e and not as a diminished third). There are many more such common "guidelines", many of which aren't formalized but just customary. Note however that these examples were made entirely disregarding the key of the piece, which would in reality matter quite a bit too of course.

However, this was to show that some of these traditional habits of voice leading are quite independant of tonality and have more to do with our notation system and readability.

But most times you don't have either harmonically oriented accidentals or melodically oriented ones, but a compromise. In a lot of traditionally tonal music it doesn't even matter, since the the two solutions are congruent, as harmonies are generally built to resolve all voices correctly. But there are still many cases where a choice has to be made and one thing sacrificed for the other. This is where you will see great individual differences between composers, some which lay their focus more on individual lines (say, Schumann/Brahms/Sch

It depends on the resolution. If the piece is in C major, you can borrow Ab from C minor or G# from A minor (which is the relative minor of C major). The Ab and G# are enharmonic equivalents of each other, but Ab generally goes down to G (dominant of C), while G# goes up to A (tonic of relative minor).

Although I've seen many scores where a sharpened note goes downwards (or flatted upwards), so you are free to renotate accidentals if you have a good reason.

The score must mimic the parts though. Never have parts that are different than the score...

...the score must match the parts

Touche.

I always keep it harmonically accurate in the score, and don't bother editing the score enharmonically until the parts are extracted...then, while formatting the parts, I'll notice something that should be changed. I suppose the new 'linked parts/score' capabilities of Finale will help with this... :hmmm:

For the time being, I'm the only one who might see the scores, so it doesn't really matter to me....

ALSO, my approach will differ depending on WHAT I'm writing - if it's background pads, then the harmony takes precedent; if it's contrapuntal melodic writing, the line will be more important... I'm a little goofy I guess.

Thanks alot guys. That's helped an awful lot.

I am a firm believer that ease of reading is an important qualification when determining which accidentals to write. I stacked a chord B Eb F# once in a horn quartet because the Eb was easier to read for the performer, rather than the D#. A conductor or analyst would freak about it, but I still stand confident in my decision.

Things like that... I just think that when it comes to hearing your music, you don't need them to jump through extra hoops so your C# major chord is diatonically correct.

On the other hand, if you go nuts with enharmonics and replace every D# with Eb, you might end up with a line going, say, E Eb C# in a B major tonality, and I would feel in that case that you're needlessly screwing with my brain. If the individual part has any internally obvious diatonicity or linearity, I'd prefer "hard" but sensible accidentals. That's me, though...

It's much easier for a performer to read notes that are written all as sharps or all as flats - So, like EldKatt says, if you're writing Eb's, Bb's etc., don't put in F#, write Gb.

...well, unless the next note is a G...

It's pointless to make hard-and-fast rules on this, it can be a very subjective determination on how to notate this.

Well yeah, there will always be exceptions...

my simple advice?

write it in a way that the harmony makes sense.

(of course, if its atonal then i guess its your choice....)

The accidental depends on the resolution. These usually correspond to the chromatic scale. If it resolves upward then its raised.

Of course it isn't always done that way since it can be hard to read(such as when double accidentals come into play) or in special cases when form comes into play(notable one is Beethoven's 7th symphony where its written as C# but he writes it as a Db intentionally because of what happens on the recapitulation IIRC).

For example, if you have a passing tone like C C#/Db D then you don't write C Db D since its obvious that the Db is moving upward and hence it is a C#.

Now its true that for fixed pitch instruments Db and C# are the same it is not true for non-fixed pitched instruments. In this case Db and C# are truly different and C C# D will sound different than C Db D(although most people cannot hear the difference). So in reality these are two different notes and hence have different functions. Hence we should retain this in fixed pitch instruments because they might be played on non-fixed pitch instruments.

Although there are theoretical reasons to notate them properly too. For example, If you have the progression Cmaj G7 Cmaj you obviously don't write the G7 as F##7 because its functioning as a G and not an F##7(assuming we are in the key of C).

If, say, you have a modulation from C to G then its the same thing... but in some cases, as has been pointed out, you might make a enharmonic modulation because it is easier to read. e.g., maybe modulating from E to C#m could be reinterpreted as E to Dbm and notated that way because, say, we might make end up wanting to modulate to C# later.

E.g., E-> Db#m -> Db or E -> C#m -> Db instead of E -> C#m -> C# both which hide whats really going on.

It's pointless to make hard-and-fast rules on this, it can be a very subjective determination on how to notate this.

That is basically it. Pretty much everyone here has been right. :-P

So you believe that there is no difference between an augmented 4th and diminished 5th? or that notating them the same accomplishes what?

Do you think a so called II7(dominant 7th) chord is the same as a V7/V as some people like to call it?

One thing you are not taking into account is context and thats what music is all about. F# != Gb even if out of context they have the same sound. Just cause you write it down which ever way you want doesn't change the fact of what you hear and only obscures what you are trying to express.

Sure in most cases it doesn't really matter but ultimately you are obscuring what you are trying to express. If you do it intentionally its ok but if you have no clue then its wrong.