This is my 4-part harmony version of the first two phrases of the traditional song "Angels we have Heard on High".

"Angels we have Heard on High" arrangement, mm. 1–4

When I play it back, it sounds very thick and muddy, which is not what I hear with the same sound font in the Bach chorales I studied (so it can't be the sound font causing the problem). What am I doing wrong?

  • 2
    It sounds fine to me. Can you be more specific about where it sounds thick and muddy?
    – Aaron
    Nov 20, 2022 at 8:11
  • @Aaron More in the second phrase than the first, but I can't seem to pinpoint an exact place where I feel that it sounds muddy.
    – OprenStein
    Nov 20, 2022 at 8:47
  • 3
    Just noticed I made an error in my transcription. In your soprano, m.3 beat 4, is the Bb correct? I've always sung it with a C there.
    – Aaron
    Nov 20, 2022 at 9:11
  • @Aaron I transcribed the melody by ear, so it might be my transcription that contains an error.
    – OprenStein
    Nov 20, 2022 at 9:18
  • 1
    Maybe not the cause of 'muddiness', but I seem to remember a 'rule' that says after an octave jump (bass line), the next notes should go back in between those octave notes.
    – Tim
    Nov 20, 2022 at 9:18

6 Answers 6


The first obvious issue is that the tenor and alto spend almost the whole first bar moving in parallel fourths. Using parallel fourths is one thing, but doing it in such an extreme way means the two voices basically blend to a single, overpowering one. You don't want the middle voices to team up in this way, they should generally be subordinate to the main voices. Especially with this being the first bar, it changes the whole perception/expectation/calibration of the listener, so even if the rest is fine then it probably won't come across.
Oh, and the melody itself joins the parallelism with that A-C in octaves. This should have been an obvious issue to you.

Then bar two starts with quaver-movement in all the accompaniment voices under the dotter crotchet in the main voice. To me, that dotted note has a pivotal role in the melody, and it's a point of rest, moreso than the minim on beat 3. With so much movement below it that definitely isn't the case anymore. Now, this could be fine – I'm not really sure if my interpretation of the melody is best. But you should have a good reason for doing that, and I don't see any here.

Aaron talked at length about bar 3. Honestly I find the mentioned doubled notes far less problematic than the other issues.

What does not work for my ears though is the subdominant in first inversion on beat 4, leading into... a sixth-only suspension of the dominant also in first inversion? IDK, is that even a thing? I'd say what definitely clashes is the sustained A in the melody with the G-A-B♭ in the tenor. And beat two seems to still try being a C chord, but for that of course the doubled B♭ puts the seventh way out of proportion, all the while there's still an A sounding in the soprano. No, bar 4 really isn't working for me, I guess you must have meant it completely different – but then you failed to make your intention clear in the score.

I wouldn't say any of those things are, per se, typical muddiness-causers. However, all together they are definitely prone to causing that effect, via at least two mechanisms:

  1. The voices don't have proper roles. That makes it confusing to the ear what should be tracked how.
  2. You have chords with either questionable inversion or unnecessary dissonance, that neither build up proper tension nor resolve to anywhere nor create any distinctive, interesting sounds.

In summary, this arrangement sounds like the voices occasionaly don't know what they're doing.


To my ear, there are two problems, both in measure 3:

  1. On beat 2, the doubled G (i.e., the fifth of the chord) in the bass and soprano.
  2. On beat 3.5, the doubled E (i.e., the leading tone of the tonic key) in the bass and alto.

Allowing those are "the problems", try the following revision to measure 3, and see if that fixes the muddiness.

  • Soprano: no change
  • Alto: all 1/4 notes, F4-G4-D4-F4
  • Tenor: beats 1 & 2 no change; beats 3 & 4, eighth notes, A4-Bb4-C5-D5
  • Bass: all 1/4 notes, F3-E3-F3-D3

m. 3 revision

  • 3
    Is it me or are all the voicings unnecessarily close? I.e., the basses are high and the sopranos are low and the middle voices are squeezed in between. Nov 20, 2022 at 10:40
  • 2
    I'm reading this without listening to it, so take this with a grain of salt, but the E you're calling attention to isn't functioning as a leading tone, so the rule about not doubling it doesn't apply. And the doubling of the G on beat 2 is probably less of a problem than the fact that the bass has G at all. Second inversion chords normally appear only immediately before a dominant chord (I6/4-V-I).
    – phoog
    Nov 20, 2022 at 10:47
  • 3
    @ToddWilcox that is the usual cause of "muddy" texture -- close voicings in the middle to lower register.
    – phoog
    Nov 20, 2022 at 10:48
  • @phoog I wasn't applying a rule; I was responding to what I heard. Leading tone function aside, the two E's just didn't sound good to me. Ditto the Gs.
    – Aaron
    Nov 20, 2022 at 13:36
  • Looking back at this, after a few months. I thought that you were supposed to unconditionally double the 5th of a 6/4 chord?
    – OprenStein
    Feb 21, 2023 at 3:04

I do not know if this is the cause of your problem (which is also hard to determine if we do not know the SF you use), but you should try to keep an eye on voices moving in the same direction. Try to remove the diminuition and have a look at a reduced version:

enter image description here

Look at this slight alteration for an example how this could be done:

enter image description here

Here’s the MuseScore file for you to experiment with:


  • Aren't the parallel fourths of the A/T in the first and third bars still not ideal? Nov 21, 2022 at 23:14
  • @AndyBonner Yes, sure (I did not claim to find all problems with the part). These parallel 4ths in the first measure actually include a wrong resolution of a tritone bb-e -> a-d.
    – Lazy
    Nov 21, 2022 at 23:39
  • @Lazy "wrong resolution" LOL
    – Fid Rewe
    Nov 23, 2022 at 22:29
  • @FidRewe Do you have any point to make?
    – Lazy
    Nov 24, 2022 at 7:51

One source of the muddiness is in the tenor and bass voice spacing. If you rewrite the tenor voice at pitch this will become immediately evident to you. In general, however, I would say that the reasons are primarily poor spacing choices and poor part writing. The separate voices merge into an indistinct mass of sounds. Recommend rewriting the lower 3 voices. Also, if this is a text setting, keep the text in the score. This will keep you "honest" in your musical choices in terms of desired effect, singability, how the indiv. tessituras of the parts relate to the actual sound of the voice, overall setting of the text to music etc.


The problem here is very conceptual. Bach composed his 4-voice hymn arrangements primarily as basso continuo (b.c., figured bass), so staring with the undelying harmony. Also, his b.c. has a lot of specific rules, in particular what we call the "rule of the octave" and also very big attention to final harmonies of phrases (usually sticking to some early baroque traditions) and to cadences, preference of counter-motion etc. Another thing is that his approach was very rhythmical in the sense that the b.c. is not only a sequence of harmonies, but it also pays attention to the position within the bar, phrase or the whole hymn.

This itself usually very vell gives the specific voices, because a lot of underlying rules apply (e.g.: no parallel pure fifths, special treatment of 64 chords, parallel fourths allowed only in specific cases -- usually when dealing with a 5 6 ascending or 6 65 descending b.c., avoiding all voices raising or lowering, going into octaves and fifths with opposite motion, proper resolution of , but all of this with a lot of exceptions, some rules treated more importantly than others). All these rules mean that a lot of snippets of the inner voices are quite fixed, and you can only possibly place them in either A or T.

It's impossible to grasp all the rules he applied (our current age probably doesn't even know all of them), and it's very difficult to write down the intricacies. But the problem is that if you start from the voices and not from the harmonic and melodic (in all voices) rules, you IMHO can't really mimick Bach's music's clarity.


Your Bass and Tenor are actually too close, lots of 3rds. Remember that the little 8 below the Tenor clef means it sounds an octave lower, not where it is written. Small intervals in the medium-low register cause muddiness. That is why the distance between Bass and Tenor is, in general practice, bigger than between the 3 upper voices. My teacher used to say that the 'rule of octave' (not more than an octave of separation between voices) is not strict between Bass and Tenor.

Think of a piano. If you play C-E-G chord in the left hand territory, sounds muddy. But with the right hand is ok. That is why the left hand usually plays 4ths, 5ths and octaves, and smaller intervals are up, in the territory of the right hand.

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.