Parallel octaves and fifths are not allowed in four-part writing because it undermines the independence of the two voices. But it is perfectly acceptable to go from 4 to 3 voices when necessary. Aren't the two things pretty much the same since you lose a voice in both cases?

Here is an example of 4 voices going to 3 voices

enter image description here

  • 1
    Can you post an example of 4 to 3 voices, especially if you're getting than from a particular textbook. I'm not sure what you really mean. Obvious piano texture can change number of voices all the time, but even in choral music - the usual type for part writing/harmonization exercises - some voices can drop out in passages. Commented Jun 15, 2022 at 14:05
  • 2
    I added this for you, please have a look
    – user35708
    Commented Jun 15, 2022 at 17:43

4 Answers 4


It is not acceptable to go from 4 to 3 voices in strict 4 part writing. Something like this

enter image description here

would be totally not be acceptable. What is acceptable is multiple voices falling onto the same note at some point while progressing differently like this:

enter image description here

By four part writing we usually mean "writing four independent voices", not "writing for four voices". And the reason why here parallel fifths are not allowed is that parallel fifths tend not to work as independent voices. Meanwhile it is not a problem for two voices to share a note (although played on a piano this does have the effect of the note being played only once).

  • 1
    What do you mean by "strict?" A textbook assignment? In real music, ex. a 4 part fugue, 1 voice dropping out to leave 3 voices is very common. Commented Jun 21, 2022 at 12:48
  • You are still going from 4 voices to 3 in your 'acceptable' example, and haven't really explained why 3 voices is any better. (And indeed, two voices moving to one unison note is done very frequently, specifically to avoid consecutive fifths.) So this doesn't answer the question.
    – benwiggy
    Commented Nov 14, 2023 at 10:34
  • @benwiggy No, I’m not? Also why would I need to explain why 3 voices would be better? I do make no point about 3 voices being better? And yes, this does not strictly answer the question, as the question is essentially asking for justification for a not exactly correct statement, which this answer addresses.
    – Lazy
    Commented Nov 14, 2023 at 17:14
  • @MichaelCurtis You will find that music theory is abstracted from real music. Music theory does not tell you how to write good music, it gives you a framework for understanding how certain things work that can then be applied to actual music. In actual music the only rule is that everything is allowed as long as it sounds good, and even if it doesn’t you simply should not expect people to like it. This does include paralles unisons.
    – Lazy
    Commented Nov 14, 2023 at 17:21

The example you give is not contrapuntal 4-voice writing, it's keyboard style 'bass line plus block chords in the RH'. But even then, it's good to keep the density constant, the slips from 4-density to 3-density aren't HUGH sins in this style, but I'd count them as minor faults.

Of course, this sort of thing is fine, and can be very effective. We ARE in SATB world here, and the very deliberate unison passage is highly effective.

enter image description here


I get the basic argument: if 1 of 4 voices was in parallel fifths, and the only objection to parallel fifths is their lack of part writing independence, then such a texture should be simply regarded as 3 independent voices, and if 3 independent parts is OK, the 3 independent voices with a 4 voice in parallel fifths should be OK too.

Here is the problem: the fairly common point...

Parallel octaves and fifths are not allowed in four-part writing because it undermines the independence of the two voices.

...is more of an apology for the strict part writing rules of the common practice era.

The fact is parallel fifths are prohibited, because the sound was undesired during the common practice era.

It isn't really an issue of part independence. If it were, then parallel thirds/sixths would not be so readily accepted.

The difference between parallel fifths and parallel thirds is relatively small. In a passage of parallel thirds, like this...

enter image description here

...it would be silly to say the two violin parts are truly independent lines. It's a conspicuous instrumental doubling more akin to an orchestration effect.

Aside for the purely aural difference between parallel fifths and thirds - some people like organum, some like Fauxbourdon - there is one technical difference between parallel fifths and parallel thirds. When playing parallel thirds the intervals qualities change between major and minor, are always consonant, and don't depart from the key signature. But when playing parallel fifth at some point you will either hit a dissonant diminished fifth or in attempting to make it a consonant perfect fifth you will deviate from the key signature.

So, those are two reason, which are not specifically about part independence, why parallel fifths are avoided: harmonic taste and maintaining consonances within a key signature.

  • But it is precisely the temporary loss of independence which makes the sound of parallel 5ths undesirable in that style. Psychoacoustically, perfect 5ths blend into each other such that several in a row can be perceived as one line, whereas the same cannot be said for 3rds. Conversely, in the case of parallel 3rds, you can still hear that there are 2 voices even when their contours are the same. This cannot always be said of parallel 5ths though, especially if the timbre is the same, because they blend into 1 sound.
    – ibonyun
    Commented Jun 16, 2022 at 19:59
  • But I like your point that it is style-dependent. This is, after all, why power chords are a thing. In rock and metal, the blending is often desirable because it also makes it sound thicker than a single note line but doesn't muddy the harmony the way 3rds would (especially in a lower register).
    – ibonyun
    Commented Jun 16, 2022 at 20:02
  • I disagree. I don't think there is a big distinction for listeners that parallel fifths sound like one voice but parallel thirds will be recognized as two. Especially for untrained listeners it seems to me most people can identify very little about the technicalities of texture and instrumentation. Commented Jun 16, 2022 at 21:29
  • What an untrained listener can or cannot articulate is beside the point. What matters is that they will hear the difference between parallel 3rds and 5ths.
    – ibonyun
    Commented Jun 16, 2022 at 23:39
  • Hearing difference is not the same as hear two independent parts. Why is the listener beside the point? Listeners are the point of music. I think you overlook the matter is about independent parts and two parts playing parallel anything don't sound independent, they sound as one unit, in plain English they are doing the same thing. The degree on interval separation is a mere technicality. Commented Jun 17, 2022 at 12:44

The reason parallel octaves and fifths are frowned upon in part writing hearkens back to antiquity when just intonation and pure intervals were in vogue. An octave is the fundamental and 1st harmonic, and a fifth is the 1st and 2nd harmonic of a single note. These intervals blend easily into one note when movement is parallel, even in the 12-tone equal tempered system. These days, no one is interested in perfect harmonic interleaving of intervals, unless as historical curiosity, and all kinds of artifacts are found, the least of which is phase shifting. The restriction, therefore, doesn't have much relevancy in modern music.


Your Answer

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