Online Compositon Lesson

[stevel]

This is something we were working on for the music web here, and it's kind of stalled at the moment but I thought because Boneman and others had mentioned something like this, I would post this as an experiment.

Please read through and see if you'd like to participate. If so, post a reply saying so, and then do Lesson 1. We'll check them and then see what everyone thinks of the format and if we should continue in this fashion, or take an alternate approach.

File attached.

[ttw]

Great idea (especially for those of us with no formal training.)

Here is my first attempt.

[stevel]

Originally Posted by ttw

Great idea (especially for those of us with no formal training.)

Here is my first attempt.

Tony, could you post your excerpts as PDF files?

Steve

[stevel]

Originally Posted by ttw

Great idea (especially for those of us with no formal training.)

Here is my first attempt.

Here are Tony's comments: I'll go ahead and comment while they're fresh in my mind:

less1_00: Little variety at all.
Melodies are similar in shape.
Measure 6 has a different apex. About the only variety.
Implied harmony is tonic and dominant.

Agreed. It's a little "sing-songy" and it's a little boring. Maybe too much unity. It ends different than it begins, but still very "same" throughout (but then again, so is Ode to Joy!)

less1_01: A little more rhythmic variety.
Melodies are similar in shape but more note lengths in piece.
Measures 6 and 7 now contrast rhythmically with the others.
Implied harmony is tonic and dominant.

What's interesting to me is that you chose to make your long note the highest note of each little unit as well - the bigger skip in the last unit seemed out of place in the former example, but know, with it being held, along with the earlier two gives it, in my mind, a 2nd level of interest: Now you have the overall shape of the melody, but you have a 2nd point of interest in these three held notes that all "jump out" of the texture a bit.

In doing so however, I think the 3rd unit (measures 6-7), even though as you point out contrast with the others, seems out of place - almost a little too simple to belong with its relatives.

less1_02: A little more melodic variety; some skips of thirds.
Some contrast between conjunct and disjunct parts.
Measures 6 and 7 now contrast a bit less.
Implied harmony is tonic and dominant.
Faster harmony changes implied by melody.

Yes, I think this went the other way - while there's more variety in terms of melody and intervallic content, the directionality and "sequential-ness" (for example C-E B-D or A-C B-D) tends to make it less varied again. Do you have some particular love of measure 6-7? You seem to be changing units 1, 2 and 4 constantly, but leaving unit 3 (mm. 6-7) identical or highly similar in each example. If you ask me, 6-7 is the weakest unit of the bunch, especially in the former example. In this example, the constant motion of the other units do make 6-7 less contrasting as you note, and is more like the first lesson, though not as interesting to me.

l

ess1_03: A little more melodic variety; some skips of thirds.
Some contrast between conjunct and disjunct parts.
Some melodic inversions.
The new measure 1 and 9 suggested by the new measure 8
Implied harmony is tonic and dominant.
Faster harmony changes implied by melody.
Half cadence in measure 8 balanced by authentic cadence in 16

You have sort of a compound melody here. For the benefit of those who don't know, a compound melody is one where you basically use one instrument or line to imply two or more lines. Patterns like C-G-D-G-E-G can imply two parts - one part playing C-D-E and another part playing G-G-G. In this example Tony, I feel that the 2nd note of each unit is creating an "accompaniment": G D* E F E - D | F D* G E D - C | - so your melody *really* is:

G EF | E D | F GE | D C (rhythm like | H QQ | H H | per unit).

So you could actually go through and take out the 2nd note of each unit, and maybe for some of the other measures that are a little different, take out the highest note or lowest note (the one skipped to) which will likely create QQ H pattern instead of H QQ in most cases. This would not significantly change the melody (the "main" melody) and would actually provide more interest than the QQQQ H H pattern which gets redundant after a while. It would be interesting (maybe for you, and for our other participants) if you provide some variations of this last example with "missing note 2" in a couple, a few, many, or all of the measures (all may need to put the missing note on a different beat for variety though).

So overall, I think your second example is the most interesting (or "strongest") here. In the other cases, I think your attempts at introducing melodic and intervallic (and potentially harmonic) variety backfire because they tend to end up with less rhythmic variety and shape (directional) variety, such that overall the excerpt weighs too heavily on the unity side. However, the final example could certainly be "fixed up" by omitting some of those "unnecessary" notes and providing some more rhythmic variety.

I'd like to hear your comments Tony - if you agree or disagree with anything I've said, and anyone else who wants to chime in in agreement or disagreement with my comments will hopefully shed light on how we all approach things and help us discover if there certain things we favor or don't favor about music.

Best,
Steve

[reith]

In the exercise you restrict composition to just 5 pitches. Is one to write 8 - 10 bars using only those 5 absolute pitches or can they be transposed?

[stevel]

Originally Posted by reith

In the exercise you restrict composition to just 5 pitches. Is one to write 8 - 10 bars using only those 5 absolute pitches or can they be transposed?

Transpositions would be OK, but, for example, I gave a range of an octave or possibly a 10th, so I think: C [D F G A c] d for example would be permissible.

Steve

[ttw]

Here's the pdf.

I don't disagree with any of the comments. I did read the instructions as suggesting a 5 note range (fifth) rather just any 5 notes. That's ok.
I have a couple of others (also with a fifth range) coming. Of course, one could use sixteenth and eighth notes to get many more notes in 8 measures. (Normally, I seem to just three main note values except at cadences or for some ornaments.)

One thing that's hard is not to expand 8 to 16 measures without considering how that would expand to 32. Thus in a future expansion, I probably would have changed measures 6-7 more. Anyway, that's the way I usually do things.

[ttw]

Another version. Basically, it suffers from too much sameness.

[stevel]

Originally Posted by ttw

Another version. Basically, it suffers from too much sameness.

I disagree:

Example 1, obviously too same (not bad though)

Ex 2 (9-16) is very conjunct, but nice and folksy and chant-like. Promising.

Ex 3 (17-24) is, in my opinion, wonderful - that is, it has these folk and chant-like elements, but breaks up the conjunct-ness with some nice skips. There's a constant reiteration of the A, which makes it sort of "Dominant-like" (not unlike the reciting tone of early music) and you hold the D until the last note which is very nice (you did this in the previous example too but I think the impact is lessened by the more step-wise motion). The only thing I'd say is, my ear wants to hear a repeated E in m. 20 to keep the flow going - however, if it were accompanied, another part could provide that motion.

Ex. 4 (25-end) - This really does seem like you've combined the best elements from the previous ones, and added something new. I really think the D in 28 should be a half, without the E moving up to the F - The E sort of "takes away" from both the D and the F, lessening the impact of each (plus it's like you're implying D minor/dorian and then F Major/Ionian in sections). In 34-35 you do something nice - the repeated A notes - now, you've repeated the F notes before, but they tend to get "absorbed" because of their placement and "tonicity". The As seem to stand out as a "fresh" element here - I think you should go with that. I think mm. 38-39 should have the repeated A notes as a feature, in one or both measures. In fact, you might even be able to duplicate 34-45 exactly and have it be ok (since it's more D than F centered now - though I'd probably introduce some variety). I also like that this is 17 measures - it adds to the folksy/chanty-ness of it.

One thing I'll mention here Tony - it looks a little like you're intentionally trying to take motives and retrograde or invert them, or "explode" them - of course part of that could be because of the range and note value restrictions - an example is 38-39 which are retrograde inversions of each other (and there's other relationships because of the 4 notes). This may or may not have been intentional but sometimes these have the same kind of effect as C E D F E G F A - where it sounds very "pattern-y" - so that might be part of what you're hearing as "sameness".

Very Good (IMHO).

Steve

[stevel]

I did read the instructions as suggesting a 5 note range (fifth) rather just any 5 notes. That's ok.

There's a couple of relatively well-known Stravinsky piano miniatures for "Five Fingers" and in some cases, they do stick to a 5th, so that's part of the inspiration (though it doesn't have to be a range of 5th specifically).

I have a couple of others (also with a fifth range) coming. Of course, one could use sixteenth and eighth notes to get many more notes in 8 measures. (Normally, I seem to just three main note values except at cadences or for some ornaments.)

Well, most music uses only a couple of duration values! But here's something to think about Tony - you have a tendency to do 1:1 or 2:1 ratios - Q Q Q Q, or H H, or Q Q H, or H Q Q, etc. Don't forget Q H Q - SYNCOPATION! I would also allow Quarters tied over the bar to create half note! (tying a half to a quarter would be cheating though :-).

You might also try two values not in a 2:1 ratio - what about Half notes and 8th notes? I don't think I disallowed rests either!

One thing that's hard is not to expand 8 to 16 measures without considering how that would expand to 32. Thus in a future expansion, I probably would have changed measures 6-7 more. Anyway, that's the way I usually do things

Yes - and I think it is a good idea to be considering what would be coming next. But these are just experiments :-)

Steve