Hi,

I've tried searching the forum as there seems to plenty of Gr theory activity -but no luck on this particular question!

Most source of harmonisation theory state one has to avoid \\ 5ths (and octaves) under many guises.

However, one thing that I have never got the bottom of is the use of "simple", consecutive, root position chords - i.e I-II, or I-IV, or I-V, or any combination really.

To my mind thay all contain \\ 5ths by definition - so when I look at a lot of examples in the ABRSM Th books most seem to at least contain IV-V and V-I root chords (both at cadence points and not).

I don't have problems with any progression using inversions etc as they get rid of the \\5ths easily but
I think I have yet to grasp somethng fairly fundamental here as to when consecutive root chords are allowed and when not .

One theory I am working on is that this does not apply to any "root position chords" that just have the bass as the root, with other combinations of 3rds and 5ths in the upper 3 voices that change (somehow) across the consecutive chords as we are interested only in \\5ths between the same voices.

It only really applies if they are simple root, 3rd,5th chords - like a simple piano left-hand three note chord.

It could just be my interpretation of what I see in the examples where you just get the roman naming of the chord and not the full written out score parts....

Any advice would be greatly appreciated - am I going in the right direction ?

Yep you're definitely on the right lines.

Working out harmonies and voicings is a bit like sudoku - there is at least one acceptable answer to each problem - even with a series of root position chords you juggle round the voicing of the chords to get rid of all the no-nos, such as consecutive 5ths & octaves and so on. Sometimes you might invert a chord if you can't see any other way to avoid consecutives. Sometimes you might solve one problem and then see you have created another, but there is normally a solution to get the 'right' answer. Occasionaly you might have to accept a less-than-ideal solution (such as a leading note in chord V falling to the 5th of chord I), but always aim to follow all the rules - that way, you will build up discipline in your harmony.

If you have keyboard skills, I would thoroughly recommend playing all your harmonisations - if you can learn what the no-nos and the 'correct' answers sound like, it will all start to make sense. And, of course, if you can learn to play Bach originals, and understand what he is doing, you can have no better model.

Good advice all round!

Context is important - in this case I am looking at Q1 option 1 where you have to only provide the extended roman chords to harmonise a given melody and not fill in all the parts - so to me I was looking at say "IV-V" with my "simple pianist's hat on.

In this case I can now (I think) quite happily provide consecutive roots position chords as the actual make up of the inner "voices" aren't written down, just the bass note (root).

Phew..

Of course when it comes to the figured bass and part writing questions, that is when I will probably hit what I was trying to describe, but I'll be able to spot them and work out different ways etc

Thanks for your help here - a bit of sanity at last!

Yes, in that case, you don't have to worry about consecutives, as you could write something like vi - ii7 - V7 - I, knowing that there's going to be a way to harmonise that progression without creating consecutives. So all you really have to worry about is if the melody notes fit the chords you've chosen.


Have fun - I have to say that Bach choral harmonisation is one of the theory things that gives me the greatest pleasure and satisfaction ... it has the problem-solving aspect of sudoku or crosswords, but with the joy of a playable, musical result!

EDIT - have amended roman numerals to take account of minor chords as that's the way it's done these days. I'm sure we didn't do the lower case thingy in my day (30 years ago), but maybe my memory is going.

Yup - I come form the same sort of viewpoint, I love a good logical puzzle, with, as you say, maybe more pleasure in working out how it actually works and the listening experience is increased !

Cheers