Piano World Home Page Forums Our Most Popular Forums Adult Beginners Forum Chorales for CATS

Cus, take your time as say. A phrase at a time no matter what the schedule is is fine. I'll post something shortly about the D major chorale

Cus! One way to get at the voice leading is to write phrases. Another way is to take them apart! All here is to that 2nd goal! So here are some comments about what I see. Keep in mind that I'm seeing the phrase through MY understanding of voice leading. Other opinions likely abound.

1. The first phrase. It's all about establishing D maj as the key. If you listen to the phrase you can hear that pretty easily. "What's the key" is a question best answered by listening and looking or looking and listening.

2. The pickup chord (D maj) that connects to the chord on the first beat of m1. IF we look at that connnection strictly according to Schoenberg's law of the shortest way you can see that Bach doesn't follow the law of the shortest way!

The only way to deal with that one is to ask "Why not?" !! Well, what Schoenberg would say about something like this is musical necessity takes precendence over rules. Which means, what are the necessities of that chord connection?

a. A law of the shortest way connection would be to go from the pickup chord to a G maj chord that would be (from bottom to top) G B D G. But the problem, the reasons that connection isn't made according to the law of the shortest way is the melody skips from F# in the pickup chord to B in the 1st chord of the 1st measure. So what Bach is doing is writing lines that harmonise the melody he has rather than lines that follow the most restrictive rules. If he followed restrictive rules he wouldn't allow that skip from F# to B.

b. The solution is: The melody skips up to B. Bach brings the bass to G. The inner voices at that point connect the pickup chord to the 1st beat chord through law of the shortest way style.

c. You could then conclude from that that in any style of music, jazz included, if a melody skips somewhere past a law of the shortest way connection, well, that's absolutely fine. And you'll hear/see this in Bill Evans all over the place

Which means the voices will just continue on their way with LSW (except for the melody note).Since the connection discussing is made of 2 root position chords then doublings here happen according to basic guidelines. Double the root! No parallel 5ths or octaves to be seen. Lesson being if 3 out of 4 voices stick to LSW and the voice that doesn't is the melody, well, that's fine. Do the best that can be done with the rule. And that's that.

3. The chord connection from beat 1 to beat 2 (first measure). Beat 1 is a root position IV chord. Beat 2 is a I chord in first inversion. That's usually notated as a Roman numeral (I) with a small subscript 6 following the Roman numeral. The 6 means the top voice is a 6th higher than the bass note.

The doubling for this chord. The FIFTH is doubled. Not the root? If you play/sing all the voices you can hear why. The doubled fifth simply makes for more efficient, economical, ergonomic voice leading. In other words, that doubled fifth makes the voice leading sound smooth. Could say more about this connection but ... the next connection (beat 2 to beat 3) is another rule breaker!

4. Beat 2 to beat 3. The standard practice is DO NOT DOUBLE THE THIRD IN A FIRST INVERSION CHORD. If that is an important rule, well, Bach just BROKE it! You gotta love this guy who breaks the rules in the theory books. But all is well even though he's a subversive rule breaker!

a. The thing is some theory books will say doubling the 3rd is ok IF the thirds are approached through contrary motion. Which is exactly how Bach approaches them. And it's how he leaves them. The big lesson is contrary motion often takes precedence over the other rules. Which means if we (or Bach) needs to break a rule or do something that's not commonly done, the mitigating circumstance, if one is needed, will often be contrary motion.

5. The pickup chord through to beat 4: Bach's gone from a I chord to a I chord! Go to the next measure and the big deal there you'll see is he's going from V to I. So that first phrase is basically about establish the heck out of D maj and that's it!

6. IF we put in "picky" Roman numerals underneath EVERY chord we'll get a bunch of chord progressions that show note to note what's going on. But those note-to-note chord progressions won't necessarily show us what's going on over the entire phrase. And, again, the entire phrase is simply about establishing D maj as the home key.

7. A great thing to do with something like this first phrase is rewrite it! Meaning keep the melody intact. Everything else is open to change as you wish. ... I'll try to rewrite that phrase letter today and post the solution. Pretty much no matter what solution any of us choose, Bach's will be better! That's why he's Bach! BUT ... if we take our solutions and compare them to Bach's solutions that's like getting a lesson. As in "I did it this way" but "Bach did it that way." Then the question is WHY did Bach do it that way?

8. There are millions of other ways to look at all I've just described. So all above is one path into the forest so to speak.

Will clarify anything that's not clear if needed ...!!

Again, you made a great recording of the chorale!

Many thanks for your wonderful analysis Mark.

1. I love the big picture of establishing the key as D maj.

2. I checked out the Bill Evans Beautiful Love Take 2 transcription. I agree that Bill uses a lot of arps, and his lower voices continue to voice-lead when he leaps around.

Here you are treating the G of m2 beat 1 as a bass.

4. Here you are treating the G as a tenor.
Can both bass and tenor sing the same note ?

7. Voice-leading is so beautifully addictive. Here is my go at Phrase 1,
https://app.box.com/s/shmlyx4rei9aie8w97mc

the sheet
https://app.box.com/s/l6gttw30jz8bw61p38g4

Of course it's nowhere near the exquisite sound Bach achieves.
One main reason is that contrary motion which Bach uses in m2 is so appealing to the ear.

Cus,

Glad you saw that that voice leading stuff from Bach is in Bill Evans. THAT'S the PROOF!

About bass and tenor singing the same note. YES. They can both do that. In the old days music was only for voice. It's from that tradition that you've probably seen in the Inventions that sometimes two voices end up on one note.

A unison for voice leading is like an octave. Meaning NO parallel unisons! For root position chords where the root has to be doubled then having tenor and bass land on the same note is perfectly fine for that.

I just listened and looked at the reharm of Bach's melody. You've taken some liberties there are that are outside the scope of Bach-style voice leading. But at the same time some of what you've done are liberties you will find in jazz.

When you do find stuff in jazz that's not in Bach it doesn't mean that jazz "broke the rules" or doesn't care about the rules. It just means jazz extended common practice to be more inclusive.

There's a lot of detail following about the first chord connection (the pickup to the downbeat of measure 1). For anyone following along here's the Bach Chorale we're talking about.

https://dl.dropboxusercontent.com/u/19895150/O%20Haupt%20voll%20Blut%20und%20Wunden.png

1) Your first chord connection. Of course the D maj triad is perfect. Then it goes to vii half-dim 7. When a perfect fifth goes directly to a diminished fifth, which is what your reharm does, that's a parallel fifth. Which we avoid in the style of Bach.

But, in jazz, well, yes, you'll find that and it's not wrong. So that's why the first chord connection doesn't actually sound, well wrong. Because it's not stylistically correct for Bach but it is for jazz.

2) The vii half-dim 7 that's built on C# (the chord on the downbeat of m1). Take a look at the soprano. You've LEAPT up to that note which is a B and it's heard with C# as the root. So it's a seventh - a dissonance. That 7th, if you want to use it, should be part of a PREPARE SUSPEND RESOLVE (down) pattern.

But, there are other ways to get to to dissonances. Another possible pattern would be CONSONANCE - leap up into a) DISSONANCE - RESOLVE (down by step). Which is what you've done. F# in the soprano in the pickup chord is a consonance. B in the soprano (over the C#) in the bass of that chord is a dissonance. So you've leapt up to it from the F# (to the B) and then resolved down by step.

In a traditional theory class FIRST you'd learn the PREPARE SUSPEND RESOLVE PATTERN. Then perhaps next would be the CONSONANCE leap to DISSONANCE and RESOLVE down by step pattern that you've used.

3) There another stylistic thing to point out. In jazz a doubled root is a doubled root. Pretty much it's fine wherever/whenever it happens. In Bach-style voice leading of course on a root position chord we know up to now - just double the root.

The problem with when a chord is built on the ascending 7th scale degree a doubled root is a doubled leading tone. (which is what the chord you've put in on the downbeat has). Leading tones like to ascend (resolve to the 1st scale step). So a doubled leading tone is a good way to introduce parallel octaves because the leading tone always looks to ascend.

But there's more! The D in the bass and the D in the alto (in the pickup chord) move in parallel motion to the 2 C#s on the downbeat of m1. Parallel octaves!

The par. octave here is easy to fix. Move the alto UP to E instead of DOWN to C#. And then you also have contrary motion which is good (D and D resolving to C# and E). The E also would be the third of the chord on the downbeat so you'd have a fuller chord and that would be nice too.

I'm going to stop here. Because in those two chords we've got a ton of detail. And what happens with voice leading is one wrong turn and all the voices in all the other chords start to turn to funny places. Like taking a wrong turn on the road. If you sat nav doesn't reset it'll think every single turn that follows is the wrong turn!

* * * * *

The way to find alternatives to what you've written on the downbeat of m1 is just to ask "What are all the chords in the key of D major that include a B natural? The answer is E minor, G maj, B min, and, sure C# half-dim 7.

A question could be "Why did Bach choose the G major triad that he did for the chord on the downbeat? Well, the first answer is probably he heard it that way! But "why" did he hear it that way? The answer is in the chorale.

If you want, see if you can re-do the pickup to the downbeat to see if there's a way to improve the solution you wrote. I'll hold off now on giving a solution. But on the other hand. This is detailed stuff. So just say so if you'd rather that I put up a solution.

Hope this helps! If you have questions or comments (or other solutions) let's discuss.

Hi Mark
Thanks so much for reviewing my files and for your detailed feedback.
You're right, at this stage, I need to keep on resetting the sat nav.

I might have taken a wrong turn here again but I have another go at the reharm of Phrase 1.
https://app.box.com/s/6r4i2cy9nwa18zvjp6ib

the sheet
https://app.box.com/s/gksg3rr1jn0nht7kr50f

Cus,

First, here are 3 solutions to the first phrase of that chorale.

https://dl.dropboxusercontent.com/u/19895150/3solutions.png

And, NONE OF THE THEM ARE ANY GOOD!

There are 3 things I take from this (and also what the the attempts you've made with it)

1. To see "how good" Bach was try to re-do what he did.

2. Your solutions and my solutions tried to change too much. If we had looked to make the smallest possible change or changes perhaps something else would have turned up that worked better. Or maybe not ... !

3. Compare what Bach did at any point to what we did. Of course his version is better. But the lesson is look and listen to see "why" the different parts of his version work better. Look and listen to see where stuff went wrong in the solutions we attempted.

If you look at the downbeat of the 1st measure in my third solution, you'll see I dropped the bass line an octave. So that's a starting point for this other way of going about it, which is, looking to re-write but doing it as little as possible. But as you can see that by the 4th beat in the measure the rewrite has gone astray.

* * * * *

In these beginning stages what you probably want to do with each and every chord connection you make is (1) scan through the voices and look specifically for parallel octaves and fifths and (2) check that all doublings are ok.

Do you see the parallel octaves in your phrase from the pickup to beat 1 of the first measure?

There's nothing easy about re-writing what Bach did. But sometimes the path has to go through the thicket if you know what I mean ... Luckily Bach's thicket sounds good ....

Hi Mark
Thanks for posting your 3 reharms. I can hear that each reharm's changes become closer to Bach's originals.
And I can see the parallel octaves in Example 3: m2 beat 4.

Also I can see my error of parallel octaves linking the pickup to m2 beat 1.
I'm kinda glad I had two painful goes because it really opened my eyes & ears to Bach's genius.
One of the main things I learnt through my reharm is how Bach drops the octave to create enough room for the contrary motion when he wants the bass to ascend.

For doublings, is it OK to double the root for 1st & 2nd inversions ?

Hi Cus,

Here I will disagree with you. I'm not sure I got any closer at all to the original phrase!

The thing w/voice leading as you're finding is a lot of people find it to be really painful tedious going at first! Because there's a lot of repetition and a lot of detail that has to be observed. And pretty much only after those details are absorbed and second nature does the payoff begin to come.

So even though your rewrites didn't go to where you wanted them to go, it is true that your 2nd rewrite had fewer parallels than the first one (in connecting the first chord!) That may not seem like much progress. But it's huge. Over time you'll see the parallels on the page more clearly and you'll avoid them. And you'll become sensitive to how they sound. So in terms of jazz you'll decide whether you want to admit that sound. Or not.

That drop the octave trick that you mentioned. That's exactly what that's there for me. Dropping down or jumping up an octave is a get-out-of-jail-for-free card!

We haven't actually talked about what makes a smooth line other than to say use the law of the shortest way whenever possible. But there's some other stuff lurking about in Bach's chorales (and in the one we want to write) that goes to making smooth lines. So that other stuff is

1. Except for octave (leaps), avoid leaps larger than a major 6th in any voice. But the M6 is ok.

2. Avoid tritone leaps that DON'T resolve down by step. This is mostly in the bass.

3. Avoid lines, for the most part that outline a tritone or a 7th. Meaning you might have a line that fills in the notes (walking by step through them all) between a tritone or a 7th. But because the tritone or the 7th in this case "frames" the line it'll be prominent. So it's avoided because the smooth flow of line is preferred over whatever "special" effect might come from making a tritone or a 7th prominent in the flow of a line.

5. When lines "outline" something, like a tritone or a 7th, you might also call that a "boundary interval" - as in the 24 shapes! The basic idea is the low points and the high points in the line will stand out. When they do stand out it's a special effect that draws attention away from the overall smoothness of all the lines.

6. Most of what I've described above as "don't do it because it's not in the style" found it's way into jazz. Actually it found it's way into classical music at least 50 years and really more years before jazz.

So jazz didn't break rules so much as over time what was once strict in voice leading (do this, this, this, but not that) became permissive (do this, this, this, and ok, that thing over there is fine too). That happened because each set of ears in every succeeding generation evolve to accept new possibilities.

Although there are some who will say each succeeding generation cares less and less about the rules and tradition.

There's always push/pull!

Ok! I didn't mean to write quite so much but there is that much going on in the background when connecting just two chords.

To look to the future the sound gets into your ears. All the calculating and detailing we're doing now simply disappears. Or to couch this in mysticism of philosophy - Ludwig Wittegenstein, brother of Paul Wittgenstein, said "He must so to speak throw away the ladder, after he has climbed up on it." (Learn the rules and then forget them!)

Hope this helps!!

Hi Chorale CATS

I've been following your thread on and off as best I could. Don't really have the time, nor am I at the level, to really participate. But I thought I might pick up a book of the Chorales and maybe try some as sight reading or just as a supplement to everything else I'm working on.

Is there a particular edition you all favor? I was thinking of This One

Hi Jim,

The edition you linked to is generally considered the standard. But Dover has edition as well that's a little less expensive. More importantly, it's got larger print.

http://www.amazon.com/Harmonized-Chorales-Keyboard-Dover-Music/dp/0486445496

You can also download all of the chorales for free from imslp

http://imslp.org/wiki/Chorale_Harmonisations,_BWV_1-438_(Bach,_Johann_Sebastian)#Selections_.28Nos.1.E2.80.93371.29

but then you have to print everything.

Everything and everyone starts somewhere so if you have question about anything at all in the chorales we're happy to help you. No worries about levels!

The great thing about the chorales is there's stuff in them to be learned at every level - from total complete beginner through to accomplished professional musician/artist.

Hope this helps -

Hi Jim
Welcome to CATS and thanks for popping in to say hello.
I'm a beginner too; when I started on the chorales in May, I had negligible piano technique, and no classical theory background.

With sight-reading, a breakthrough for me was Fred Hersch suggesting you can pass your thumbs around, so it's OK for your RH to play the tenor, and it's OK for your LH to play the alto.

Also feel free to start where-ever. Chorale 1 in G maj is not the easiest, and the C maj ones can be difficult.


Hi Mark

Thanks for your encouragement.
One of the key learnings for me when you corrected my reharms was to check all voice combos (incl bass-alto), not just the obvious combos which are soprano-alto, bass-tenor.

The jump-the-octave trick is going to be useful for me when I reharm ATTYA A2 sec where the soprano drops really low.

Today I worked on BMV 322 Gott sei gelobet Phrase 4 and found a tenor example which illustrates your points below. Except for point 2. where his tritone appears to resolve UP.
https://app.box.com/s/iuqf8k4oro3ylzj1od2r

Cus, just calling it the "jump-the-octave" trick as you've done is great! Because now you know EXACTLY what the possibility of that jump allows.

Well, my point #2 was too ambiguously stated. In fact, to be honest as I look at that I'm not sure what I was saying (mea culpa!)

What I SHOULD have said was

Avoid tritone leaps in general. Because they're not so easy to sing (relative to other stuff in a chorale). AND because the tritone is so distinctive it's going to stand out. IF it stands out, then it's talking away from the smoothness of the line. Because all things considered, smoothness of line is the PRIME OBJECTIVE.

But none of that explains the fact that you found a tritone leap!

The C leaps down a tritone to F# and then it resolves up by step. And notice how after that Bach just keeps coming up by step. So he fills in the tritone. That "fill in the spaces in the leap is a very common device. Everyone uses it - classical composers, jazz musicians etc. The basic idea is leap in one direction and then step back (as many times as needed) in the other direction. Sometimes smaller leaps go in the bigger leap. Sometimes it's not quite as clean as I'm describing. But you sure you see the principle ...

So Bach LEAPS down a tritone and then walks back up for step. In so doing he fills in the tritone. If you follow the tenor line from that C over to the end of the phrase the tenor line ends on B. Which is to say EVERYTHING in between that C and B is just basically filler. Sing the line and see if feels like that to you. In other words does the larger sense of the line go from that C (the high point of the phrase) to B (the last note of the phrase).

Assuming it does that's the law of the shortest way at work. In a STRUCTURAL sense. ... structural meaning some notes are more important and essential than other notes. The STRUCTURAL notes are the framework on which the phrase hangs.

Another observation about that tritone leap. Notice how the first note of the leap and the second note of the leap - they're both consonances as they're heard against the other voices. Now play or sing the tenor line by itself. And you'll probably hear the tenor line there sounds like Bach leapt down from the 4th scale degree to the leading tone. And the leading resolved to .......... (fill in the blank of where it resolved.

And the last observation is notice that Bach more or less buried this stuff that we're talking about in the tenor line. Had the leap been in an outer voice it would be much more noticeable. That's not to say Bach wouldn't do it. It's just that when it happens in tenor or alto voices it's not as noticeable as it might be if it was in a bass or soprano voice.

If you have questions about this just ask. Or if you hear it differently than the way I've explained it we can discuss more too ...

Hi Mark
I tried to sing the tritone leap today and had NO idea, I agree the interval is very difficult to sing indeed !

Yes I could hear the motif as C,B which is established in the preceding pitches, and I could hear that the tritone filler/embellishment as an example of motivic development.



And I heard the tenor line as a C7alt resolving to the consonant B which is a chord tone of G maj.

Cus, I'm not quite following you with the C7alt. But hearing C resolve to B is a good step! And hearing the tritone as not easy to sing is an excellent step as well!

Hi Mark
I do like your interpretation of F# as a leading tone. Because that is in the spirit of voice-leading, which is all about the choice of note.

I guess I was thinking of the progression IV7 leading to I. The F# suggested C7alt to me.

Hi Cus,

Ah, I see what you're saying!

Something else to point out about the Chorales is at any given time there is a chord going on. But most of the time Bach's going to going to and from I chords or V chords. Most of the time. There are tons of exceptions. So a good way to look at any of the phrases is figure where does it start. Look at the pickup chord if there is one and the following chord that comes on the downbeat of the next measure.

Next step is where does it end? Look at the chord under the fermata. Then kind of look to everything else in the middle to get a sense of how the opening of the phrase is getting to the end.

One of the things about the chorales is the chord progressions will break down into simple Roman numeral analysis - which would be "functional" harmony. "Functional" meaning the chords have functions. V has a function. I has a function. Etc. The functioin of I is to establish key.

But. Having explained that. Looking at functional harmony in the chorales is really best when looking very limited segments. Like first chord vs last chord. Or a chord or two before the last chord going into the last chord.

The reason is is that if the phrase is say 3 measures long in 4/4, There are 12 chords,probably. But those 12 chords are there because of the voice leading. Those 12 chords aren't there because Bach was thinking about chord progressions. So looking at those 12 chords in a row doesn't actually show a whole lot. Except that Bach was using chords! Or maybe a different way of saying it is harmony in classical music makes structure. Composers new that I was resolution and V was the opposite. So they composed to exploit that.

That's very different than jazz where the chords are there because they're chords and it's specific chord progressions that define a tune. So in jazz having the chords lined up one after another is exactly what you want.

... maybe a next step with the Chorales is to look at a phrase and identify ONLY the large outline. Where does the phrase start. Where does the phrase end. Once we have that we can see how everything else in the phrase is working to and from those start and end points.

Actually, IF for some reason we HAVE to break a chorale down into a chord progression. For whatever reason, one way to do it would be to take the first chord of each phrase. So this would be looking at the structure of the chorale and not the details.

I'll find a good phrase or two and post them with some observations about phrase start and end points. Feel free to do the same too! (But only if you want to, of course!)

Mark and Cus,

Thanks for the welcome and the advice.

Looks like I bought myself an early Christmas present. Due to the holidays I have an extra week between lessons, so I should be able to find some time to start fiddling with the Chorales a bit. I'm looking forward to it.

(I could have sworn I posted a message similar to this a few days ago, but dang if I can find it anywhere. Old age must be catching up with me )

Hey Jim
That's great your book has arrived.

Hey Mark
Thanks so much for the very helpful note of big picture chords.
Yes you are right, it's so different to the functional harmony in jazz tunes.

I'd love to analyze my current Chorale which is BMV 322 as I'm not even sure what key it's in: C maj, G7 or G maj ?

Here I take Phrase 1
https://app.box.com/s/1fprrlyjehdqadcak5cw
and identify the pick-up as G7, m1 beat 1 as G7 and the target chord as E min.

Jim,

I with you. I thought I posted this message a while ago!

Glad you have the Chorales. We're here if you have questions or comments. All questions are good and encouraged!

Cus, good work on that excellent chorale you've chosen. Could you as next step play and sing the bass voice of the 1st phrase. And then if you could tell Jim & I what key it sounds like?

mmmm, very interesting exercise.
I primarily hear it as G7.
When I kept on playing it over and over, I also hear it as C maj.