Improving part-writing for my melody ?

[Palestr1na]

Hi y'all,

I've been preoccupied with harmony and piano playing for a few years and also composed some decent songs. However, I feel I'm not making much progress in part-writing and I'm looking for pointers. Here's a melody I composed:

Bzepic2.png

Listen:
https://easyupload.io/0ulc4y

I've gone through some youtube videos on part-writing and generally they just talk about what to double or not double, and to avoid paralel motion. (Yes, I'm aware that in the beginning I broke the rules by doubling the third.) In composing this, I've mainly been concerned with having as much contrary motion as possible. However, there are plenty examples in popular songs with lots of paralel motion. I have the vague notion that paralel motion is only appropriate when there are attractive tones (e. g. leading tones). But otherwise, I have no idea how to really improve this melody.

[jancivil]

"plenty examples in popular songs" of parallel fifths etc vs 4-part harmony writing that's restricted as such to get a certain style affect amounts to two rather different goalposts. All considerations are contextual, musical. There is no such notion as you've surmised I've ever heard of. The stricture is about parallel aka consecutive perfect fifths or octaves within your 4-part basis. There is no particular influence on that coming from a leading tone being present. In your last cadence, the leading tone A# moves down to F# in the I harmony. Why? Is this a pseudo-full cadence? If it's supposed to be the final cadence it's incomplete. There can be reasons for it along the way, but at this stage, you want to be a bit more square.

There are enough problems in this to amount a rather large expenditure of time sorting it by specifics, if the goal is classic good part-writing. Youtube should very much not be your idea of part-writing school, you need a coherent contiguous course. I can however point to some concepts:

There is a good reason for the restriction to one M3 in a major harmony. It's not weighted right, and particulary beginning the exercise on I6 with two 3rds is dodgy (and there is probably no good reason for it). All these inversions, what is your reasoning for these? It isn't forming a very good bass line, is it?
More typically your first exercises will involve a given top tune or bass line with Roman Numerals and figured bass. Melody writing is a somewhat different lesson, albeit you should be apprised the idea of part-writing is 'write good parts'; eg., the independence provided by contrary motion is the point, not contrary motion as an abstraction to guarantee anything through itself.

[Palestr1na]

I didn't mean paralel fifths. (A practice proscribed in older classical music, ie music which is more strictly diatonic.) I merely meant paralel movement. I'm aware that the voices can have independence despite paralel movement, for example, when the bass prefigures a note which (right afterwards) appears in the soprano melody. I sometimes do this with good success.

Someone told me, in the simplest terms, that there were only 2 voices in this melody of mine. So I introduced more contrary motion and made this version which should have 4 voices. (But that doesn't mean it's good.) And the big idea here was that there should be a contrast of light and deep between the beginning and the end of the melody. So while I probably should just simply dedouble the third, I think it's absolutely necessary for the song to start off, in this way, with a "I" chord in first inversion.

Sometimes I come up with a really great song, and the bass seem as if it's its own melody. But in this example I posted, it just sounds more like it's cobbled together from disparate things. (I'm pleased with the soprano, though.)

[jancivil]

"A practice proscribed in older classical music"
this kind of remark is why one will say 'all decisions are musical/context bound'. as well as having said already, as regards contrary motion, "no guarantee" per se of anything other than itself. I don't recall any such thing in CPP part-writing (which is largely oriented in JS Bach, who worked before there was any 'classical' period. Parallels such a thirds abound in the choral music, which is the model. Parallel fourths as well.).
So what is that, from Fux? You seem to be picking up factoids off the internet. You need a real course, over canvassing people on the 'net or browsing Youtube. You're going to wind up with things to unlearn.

So you gave one reason for the one choice. Ok, no problem but what's the reason for the D# doubling? You can do less. In any case, a larger goal of this kind of exercise is to learn to write good parts.

You're also mixing two very different tasks. You want to place learning good part-writing foremost in your compositional aspirations if they include this particular area of technique. Hence my remark on 'more typically' the teacher writes up the basis for the exercise; giving one line of 4 and indications of the harmony and what is wanted in the bass (which is figured exactly), giving musical puzzles (with specific goals) to work thru in your learning to write parts, rather than worry about the melody's quality (conflicting goals at the same time, another would be worrying about the quality of the chord progression) yourself.

[TribeOfHǫfuð]

What to double and what not? In Fux style? Just don't double the third. Yeah, he has a few more sayings on doubling here and there in Gradus, but nothing no one has ever given a fck about, least of all Bach and Mozart. It is the consecutive fifths and octaves that are his demons, and he invented them himself, so let him suffer.

And FWIW: I agree with @Jan on the inversions and the bassline. What is that about?

Anyway. Stay of the thirds as much as possible and thats that if you want to follow the old hag.

[jancivil]

For example, one type of goal is to write with a strong root movement; another is what will you do differently for weak root movements (where a progression has a lot of stepwise or third root movement); what kind of parts for a classic cadence (incl. what harmonies are classic. Certain figures are given in the bass, and the movement follows type (eg., I6/4 to V[5/3]); V4/2 to I6).

[TribeOfHǫfuð]

Sometimes composer wannabees don’t read all they should and instead they exchange cheap tricks that will make them pass an exam at a mediocre level. Here is one from my music school in the prime 90s: In whatever part in 4/4, it always seems to work if you keep the complete triads to everything on-beat while using any mediating chord, substitute or inversion off-beat. It will sound trivial but make you pass.

[Palestr1na]

So what is that, from Fux? You seem to be picking up factoids off the internet. You need a real course, over canvassing people on the 'net or browsing Youtube. You're going to wind up with things to unlearn.

I've been watching a french professor with a Youtube channel who takes most of his stuff from Fux and Palestrina (whose music I particularly enjoy). I should probably at least order some decent books on part-writing, since I don't have too much time to concentrate on composition courses. (Most of my time goes into philosophy and lit, and learning to play instruments, and anyways most the pop songs I compose have far simpler tasks than this.)

So you gave one reason for the one choice. Ok, no problem but what's the reason for the D# doubling? You can do less. In any case, a larger goal of this kind of exercise is to learn to write good parts.

I guess I just wanted the first chord to sound fuller, but seeing I can't/won't place the bass (the bottom D#) lower, there's not much else to do than to keep it or to remove it. (Or to arrange the melody for two instruments.)

[amulitz]

If you want the sense of two or more melodic lines functioning at the same time, or if you have four voices going simultaneously and want them to each have melodic integrity (i.e., be heard as a melody individually), then you have to avoid parallel fifths, fourths and octaves because the overtones are so similar, the listener loses the sense of multiple melodies. Of course, parallel motion is used often, and if you look hard, you can find parallel fifths on occasion in Bach, but not more than one time in a piece.

[TribeOfHǫfuð]

If you want the sense of two or more melodic lines functioning at the same time, or if you have four voices going simultaneously and want them to each have melodic integrity (i.e., be heard as a melody individually), then you have to avoid parallel fifths, fourths and octaves because the overtones are so similar, the listener loses the sense of multiple melodies. Of course, parallel motion is used often, and if you look hard, you can find parallel fifths on occasion in Bach, but not more than one time in a piece.

Nope, that is historical after-rationalization. Fux did not explain himself in debth but just made the assumption as part of his definition of counterpoint and slightly suggests later that a choir will have trouble singing parallel fifhts and octaves. However, as Knud Jeppesen pointed out, Fux's references to Palestrina in the intro did not justify a focus on parallel octaves and fifths in particular, because Palestrina had no special restrictions on these that do not go for other intervals. Your explanation about overtones is neither Fux' own nor in accordance with Palestrina's polyphony, which he referred to. It is scholastic after-rationalization understood in terms of harmony and tonality that did not dominate medieval polyphony.

[Palestr1na]

If you want the sense of two or more melodic lines functioning at the same time, or if you have four voices going simultaneously and want them to each have melodic integrity (i.e., be heard as a melody individually), then you have to avoid parallel fifths, fourths and octaves because the overtones are so similar, the listener loses the sense of multiple melodies. Of course, parallel motion is used often, and if you look hard, you can find parallel fifths on occasion in Bach, but not more than one time in a piece.

Avoidance of parallel fifths and fourths is almost taken for granted, when you're composing in a diatonic (not modal) idiom. That wouldn't necessarily be the case if I was attempting to compose in a more chromatic idiom, or an idiom where some of the contrapuntal rules were inverted, like Debussy, or in a so-called polytonal idiom (some theorists argue that there is no such thing as "polytonal") like Darius Milhaud whose music I don't find in any way disorienting. (For example Milhaud's String Quartet No. 5)

[TribeOfHǫfuð]

Forget all the "naturalistic" explanations as if the brain is wired to certain harmonies by birth. It is nothing but conventions, and conventions change by the tides of time.

[jancivil]

I'd like to see something cited as a source for conflating parallel P4ths with parallel P5ths, for the Common Practice Period.

[TribeOfHǫfuð]

Well, now you mention it, I like to see that too...but it all smells somewhat of the childhood diseases we discussed in the other thread; mistaking musical conventions for universals given by evolution or the almighty himself.

[jancivil]

I tried to verify it. There's nothing. I found a fair bit of rules that aren't really (one was the 6/4 as purportedly strictly forbidden except the cadential one on the 5 bass with the 5/3 resolution, yet there is are two definite things known as the passing and the neighboring 6/4 in the baroque), but that one isn't in any list. I think it's more rules gleaned from conservatory of youtube than anything.

It doesn't appear to happen very often but I never heard it argued in my life, and I felt like I was 4-part writing for a living for a couple years.