enter image description here

In this concerto there are a couple of places where parallel octaves are used. I I am a harmony student but This is the first score that I have studied and see what just about every harmony book and counterpoint text teaches against. So what gives? Did Mozart not care about parallel perfect octaves and if they are so forbidden why do they appear in the very first score I have looked at? And most importantly, why are we taught not to use them?

  • 1
    Those octaves are doubled voices which are entirely parallel and don't count as "different" voices in the counterpoint sense. The entire craft of orchestration rests on such judicious doubling. For instance, cellos and basses basically play parallel octaves for the entire classical period! Apr 16, 2023 at 13:00
  • Doubled voices just in 2 chords?
    – user35708
    Apr 16, 2023 at 19:09
  • 3
    Note that no one should be teaching you not to use them. What is good to teach is that there was a period of practice in Western Europe where they were essentially never used. Music theory is descriptive, no proscriptive. As in, it's about what has been done, not what can or should be done. Apr 16, 2023 at 19:39
  • @ToddWilcox there's plenty of proscriptive music theory out there, and the most notable such proscription is that against parallel fifths, followed closely by the rule against parallel octaves; both were observed very strictly by practicing composers for several centuries before Mozart and at least a century after. It wasn't simply that they were "never used"; students were explicitly taught that they were not to use them.
    – phoog
    Apr 17, 2023 at 12:35
  • People are writing answers about why that rule doesn't really apply here, which is interesting and valuable (I too have had this rule cited to me in a place where I now understand it didn't apply), but what I'm most curious to learn is why the rule exists at all. In what way does it (a) sound bad and/or (b) impair contrapuntal voice movement? Apr 17, 2023 at 15:52

6 Answers 6


Counterpoint theory generally handles the case of writing for a certain number of independent voices. Here it is a general rule to avoid certain parallels as these impair the independence of the voices.

An real world score such as this violin concerto is not made up of only independent voices. Take for example the violins in an orchestra: In a Mozart time orchestra you’d have about 6-8 first violins, all playing the same part. These can be seen as parallel unisons, but you would not consider them independent voices in the first place. And similarly not all parts of the score need to be independent voices in a contrapunctal sense. You will find though that on a large basis if you distill the core subjects and counterpoints from such a score there will not be such parallels, as these can quickly make the whole thing sound quirky.

  • The point is that if you play this part just with the violins, The parallels are not only audible but make the violin section sound thin at that point. those other parallels with the viola are awful too.
    – user35708
    Apr 16, 2023 at 19:03
  • @armani Are they? youtube.com/watch?v=MOZ2CsKWoBY&t=427s
    – Lazy
    Apr 17, 2023 at 6:16
  • 4
    @armani but the piece wasn't written to be played only with the violins. Also, you're allowed not to like Mozart's music (if this even is by Mozart) or his style of orchestration. Even in pieces of undisputed authorship these techniques of orchestration are present; if you find them tolerable in those cases and awful here then perhaps you can identify the reason behind the difference and potentially contribute to the scholarly debate about the piece's authenticity.
    – phoog
    Apr 17, 2023 at 12:24
  • @armani the viola and the cello are playing the same voice at that little section before the fermata, the eightnotes G-G-Fsharp-Fsharp-B-B-A in both parts. The viola is playing every second eight note an octave up, which gives some extra life to the line. Sep 24, 2023 at 19:57

The prohibition is when writing counterpoint, which Mozart is not doing. When one instrumental part doubles another, there is no attempt to maintain the independence of voices — in fact, specifically the opposite.

  • @armani It’s not math or engineering. Context matters. There aren’t any fixed, consistent rules. When Bach was writing counterpoint, voice independence was key. When Mozart was writing homophony, doubling sounded appropriate. The entire history of Western music is one of constant change in practice. What was idiomatic for one composer was completely ignored or contradicted by another composer. Just because you lose points for doing something in a music theory class doesn’t mean it wasn’t a popular thing to do at some point in the last 400 years. Apr 16, 2023 at 18:57
  • Under that premise one could say that all instances of perfect parallel intervals were "doublings". No. For it to be "doubling" there has to be some consistency... This is not what I have indicated in the score notes I added. Those are parallel perfect intervals and they do exactly what parallel perfect intervals do, which is thin out the texture.. Maybe with the whole orchestra it is not so obvious but it is with the 3 part violins
    – user35708
    Apr 16, 2023 at 19:07
  • 1
    @armani You’ve marked two passages in the score. Viole and Celli can very well be seen as doubling. You will see that with second half of 496 Viole and Celli play the same notes, the Viole just alternating between octaves. The first Violin is of course not doubling, it is triple stops, i.e. playing multiple strings at the same time. This mostly serves to add more notes to the tutti, which is generally a situation where polyphony not the first thing on your mind, and is written in a way that is hopefully comfortable to play.
    – Lazy
    Apr 17, 2023 at 12:47

Counterpoint is one thing, this concerto is another. Voices here can (and do) get 'doubled', and as such, with so many different voices, those 'rules' have to be disobeyed.


As other answers indicate, it is because scores are often not engaged in "part writing" per se. The classical period is characterized by instrumental figurations that are not directly subject to voice leading rules. Such figurations were also common in the baroque period.

The middle voice of the violin triple stops might better be a D instead of an A at the downbeat of 497. (I'm no violinist, but I assume it's playable because the next chord has D in that position.) It's also possible that Mozart didn't write this piece (the attribution is uncertain) and that if he had he would have done it differently.

But in general, this passage is characterized by relatively free doubling. It's basically a three-voice texture from the voice leading perspective (as is often the case with a series of 6 chords). The violas are more or less doubling the basses, though not strictly so, just as the first violins are more or less doubling the first oboe. The part writing is basically this:

%%score {RH|LH}
V:RH clef=treble
[b d]|[a/ d][g/ d]|[f/ d]
V:LH clef=bass

The differing figures and the manner in which various parts jump from one voice to another are concerns of a different discipline from part writing: welcome to the world of orchestration.

  • Free doubling? Now thats a new one. Can you explain what that is? My textbooks show doubling as a consistent duplication of one of the voices. Like a bass and cello might double at the octave. They dont double just sometimes. Doubling is clearly doubling and I dont see how my notes I highlighted can be conceived as doubling
    – user35708
    Apr 16, 2023 at 18:50
  • If orchestration is a different discipline, what is the point of part writing then?
    – user35708
    Apr 16, 2023 at 18:53
  • @armani look at the viola: D-B-G-F♯-B-A. Now look at the basses: G-F#-B-A. Look at the oboe: B-A-G-F♯. Compare the first violin: G-B-A-G-F♯. They are playing more notes in common than not. It's common in Mozart for the violas to double the basses without necessarily playing exactly the same thing. The same is true of melody instruments. Look at some other Mozart concertos or symphonies and you'll find other examples.
    – phoog
    Apr 16, 2023 at 20:01
  • "What's the point of part writing": to establish the norms for harmony and voice leading in a homophonic texture, or indeed in a stile antico polyphonic texture. In a newer style concerted texture, the rules are a bit relaxed. Consider your own recent question: "I wrote this 5 part texture this morning only to discover that all I had done was double the tenor line an octave higher. ... I was going to elaborate the top line for a solo violin part." Look at the violins and winds in other orchestral pieces by Mozart. It's common to have different melodic material following the same shape.
    – phoog
    Apr 16, 2023 at 20:05
  • 1
    @armani not completely, no, but seriously, I looked at one other violin concerto of Mozart and one symphony and I found examples of parallel octaves and unisons from this sort of quasi doubling, whatever you want to call it, both violas with basses and violins with woodwinds. If it's that easy to find it must be idiomatic. There's also a similar phenomenon in baroque music, especially choral/orchestral, where the texture stays basically in 4 parts but the voice leading in the orchestra and chorus are different. Parallel octaves occur but there are still (at least) 4 independent parts.
    – phoog
    Apr 17, 2023 at 11:09

To answer the question briefly and succinctly: because orchestral writing does not have the same principles nor is bound by the same rules as theoretical counterpoint. Or to quote the probably overused saying, this is apples and oranges.

When writing for an orchestra, there are various concerns (balance [between voices and between instrumental groups], texture, clarity) which do not exist when writing a mere counterpoint exercise. Doubling an important voice (in this instance: the bass, which is doubled [with octave jumps] by the violas; and the top note of the "melody", which is played by both the violins and the oboes) to give it more prominence is a natural solution to this.

In short, the issue here isn't one of counterpoint: these simply aren't independent voices (in the same way that the strings doubling the lower voices in this Bach cantata [BWV 182, mvt 7, see score image below] are not; or the flute and violin doubling the chorale melody at the octave later on are not; an independent voices), and thinking of it through the lens of counterpoint misses the forest for the trees. What you're looking at here is more properly called orchestration. BWV 187, mvt. 7, mes. 1-3

To answer your penultimate question (and with that put to rest any doubts about the final one), you will note that, if you disregard the doubling (in the Mozart as well as in the Bach example), the voice-leading still very much does follow the usual rules of voice-leading.

  • Re. before anybody says something about the belated answer: one important element missing from previous answers was the fact that the topic being discussed here has a name, i.e. "orchestration", and the main reasons for that.
    – AlexJ
    Sep 24, 2023 at 17:43

If one views music "theory" as making predictions about how various musical constructs will be perceived, what music theory says about parallel perfect intervals is that a group of instruments playing parallel octaves and fifths is apt to sound somewhat like a single instrument with a different tonal character.

Having some melody lines played by trumpets, some played by violins, and some by a tuba, is apt to make a piece more interesting than if every melody line were played by the same instrument, and having a melody line played mostly by violins except for a few carefully chosen accents which are played by trumpet can space things up, but only if transitions between instruments "make sense".

Likewise, using parallel fifths and parallel octaves is fine if the transitions between the portions of the music where lines move in that fashion and the portions where they move independently are sensibly placed in the context of the piece. Music theory suggests that such transitions may be perceived as being analogous to transitions between different instruments, but whether that's a good or a bad thing would depend upon what one is trying to accomplish.

Your Answer

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