Lesson with Marsbars (General Composition, Literature, History)

Marsbars:

From what we've talked, it seems to me the most adequate thing to do would be to first work on a little simple composition exercises.

The first exercise I like to do in these sort of situations, so I can get to know you better, is that you write something based entirely on how it sounds to you.

That is to say, write for any instrument or group of instruments, but the most important thing is that you write entirely "by ear". Don't write a piece, just fragments, ideas, or sketches of what you think sounds cool , interesting, etc.

I already saw some of you work, so I'm aware that you have some basic knowledge. But for this particular exercise,forget all you know about music (so to speak) and just "wing it" based entirely on how you think it sounds better to you.

The reason for this exercise is, like I said, so I can get to know you better as a composer which is crucial or we won't get any work done. If I say anything else, I may end up influencing you so I'll explain more as we go along.

Remember, don't try to write pieces, just sketches, fragments, it doesn't matter how long or short. It's only important that you do it all by ear. Help yourself with the computer, piano, or whatever other instruments (or ways to produce sound at all) you have.

The result you can post as a MIDI (or any other such sound file, as mp3, etc) as well as PDF, specially if you did it on the computer. Don't worry about the MIDI (If you picked that) sounding bad, that's not a problem.

Try to have somethin' at least by next week, but there's no rush if you don't~

If there are any questions, feel free to post them here, naturally.

Is this adequate? I composed this entirely using the computer; trying not to worry about theory.

lesson1 - byear.MID

Finale 2008 - [lesson1 - byear].pdf

lesson1 - byear.MUS

Good work! And so fast too~

From this I have a clearer idea on what we can work on. I see that you're really entrenched in classical rhythms, though your sense of harmony/etc is got a modern spin to it.

You're using also imitation techniques, and such other things, which is all fine. However, I think we should loosen you up some, specially the rhythms.

It works like this, we'll now use a technique taken from serialism, which is apply a specific set of numbers to a sequence of musical parameters. In this case, you're again free to write whatever pitches/instruments you want, that's not defined. However, the note durations are going to be defined by a number row.

You also get to choose your own row, and what it means. For example, a row of 5:

1 = quarter

2 = Whole

3 = 16th

4 = dotted 8th

5 = half

And, you can organize the sequence in whatever way you want. First as 12345, but then can be 54321, or in any order, or repeating numbers even. The idea is that you play around with it, again, don't make a piece, but rather a sketch or treat it as an exercise with a limited number of variations for the row. Remember that you are free to pick, again, what instruments and what actual note pitches are played. The only set-down parameter are the note durations, and the row you created.

Keep in mind, that there's no "wrong" or "adequate" exercise outcome for these things. So, don't worry about it and go crazy.

Post the row you used, as well as files (midi/pdf/mus) like before.

1 = quarters

2 = eighths

3 = halves

4 = whole

5 = dotted 8ths

6 = 16ths

I wrote this so the note values vary every measure or so, then I realized that you probably want each note to change with every note. I'll do it over if that's the case.

lesson2 - note_duration_rows.MID

Finale 2008 - [lesson2 - note_duration_rows].pdf

lesson2 - note_duration_rows.MUS

1 = quarters

2 = eighths

3 = halves

4 = whole

5 = dotted 8ths

6 = 16ths

I wrote this so the note values vary every measure or so, then I realized that you probably want each note to change with every note. I'll do it over if that's the case.

Yeah, I should've specified more. The idea is that within the same measure, you have for 123456 and then 654321 in two separate voices, and things like that. But trying to keep the row within a measure, so that each note value comes one after the other.

So, in one measure you'd have in one voice if it's 123456 with your row, it starts with a quarter and then come each of the other values until the 16th note at the end.

I have an example I did a while ago:

serialism.jpg the score, and

serialism1.mp3 Mp3 audio

The thing is, in my example I not only serialized the note durations, but I also serialized the pitches, and write a little scheme of how it flowed. It's detailed on the top of the score, though in German. Either way, it should give you an idea of how it works. Note I also did away with measures altogether, since I find them pretty pointless in something like this, though it helps to keep things organized sometimes.

Same row. Whenever the lower voice comes in it goes through the row backwards (654321).

This came out kind of random. I'm wondering how Schoenberg and everyone was able to express themselves with serialism so fluently. I wasn't feelin' it.

lesson2 - note_duration_rows_fixed.MID

lesson2 - note_duration_rows_fixed.MUS

Finale 2008 - [lesson2 - note_duration_rows_fixed].pdf

Good work.

That's a good question. A problem with serialism has always been that, when you have rows and things defined by numbers, you're actually taking the composer's intuition away from the composition process beyond the definition of the rows at the beginning. There are a lot of mixing and matching, specially by Schoenberg. However, Schoenberg shouldn't be considered that much of a serialist as he belongs to the atonality movement instead (despite his technique being seminal to the development of serialism later on).

Stockhausen and Luigi Nono are examples of serialist composers, for example Nono's "Il Canto Sospeso" or "La fabrica Iluminata" are examples of serialist compositions. Stockhausen's electronic pieces are pretty famous for trying to even take out other musicians from the whole equation, as serialism demands absolute tone per tone-precision, which real musicians aren't so good at when things get as complicated as it can get with these techniques.

Today, it's not very usual to shoot for the same goals that Nono or Stockhausen did in the 50s and 60s. Think of serialist techniques such as the one we used as ways to come to rhythms, harmonies or otherwise textures that can be used as musical material regardless of what technique we use to use them in.

For example, you can come up with a rhythm like we did serially, but use classical harmonic principles. If the note pitches are serialized, you don't necessarily have to always play them in the same way. 12 tone technique is pretty much a series of inversions/retrograding of the same row, but all that work is dependent on the composer.

Development, that is, is still dependent on the composer's intuition if we're talking about 12 tone music strictly. The difference is that serialism is the serialization of absolutely every single parameter, even things such as form, etc etc so that the piece technically "writes itself" once you have your formulas figured out. 12 tone technique is simply serializing a 12 tone row, where rhythm, form, and other such things are still left to the composer's intuition to control.

Another thing that is interesting about these techniques is that by eliminating the composer's intuition or "say" except for the definition of the rows, the piece can sound in a direction that the composer would have not reached simply "by ear", or otherwise intellectually. It can give rise to material that can later be reused, or developed in different techniques altogether.

In our case, the idea was that you experimented a little with it, and like you said, it sounded sort of "random." However, the aesthetic principle of things like this can only be appreciated if you give it some time to actually let it get in your ears. Specially rhythms that aren't metric and non-standard harmony. It's not important if you later decide to write atonal, serialist, or otherwise use any of these techniques at all, but it's a good idea to have had first-hand experience with them.

So, on to your next exercise!

Take the first measure of the previous exercise, keep the notes and rhythm. Then, only using this material, make variations of it. They can be only measure wide, and don't need to be done in any serial technique. For example, you can double some notes, alter the rhythm, invert the intervals, and so on. Do some 15 to 20 small variations.

Remember that the musical material has to be taken from the original measure, such as the pitches and rhythm. Try progressively variating things, start with small changes and move on to bigger variations. For example, you can get rid of all rhythm and just have all the pitches play at once, creating a sort-of cluster chord. There are a lot of possibilities, so give it a shot.

PS: I recommend that if you can, you listen to some of Stockhausen's electronic pieces. Like, for example, Gesang der J

Here you go. I have to say.. I'm getting thoroughly "loosened up" with these exercises. :w00t:

I listened to that Stockhausen piece and am wondering if all those sounds he synthesized himself or are they samples. Does he have any orchestral work I should be aware of? Interesting.

Finale 2008 - [lesson3 - variations].pdf

lesson3 - variations.MUS

lesson3 - variations.MID

Here you go. I have to say.. I'm getting thoroughly "loosened up" with these exercises. :w00t:

I listened to that Stockhausen piece and am wondering if all those sounds he synthesized himself or are they samples. Does he have any orchestral work I should be aware of? Interesting.

Well, even if we just started basically, what you did here does show that you're getting around trying out more things than before. I really liked how you solved this exercise, specially the rhythms and the intervals.

It should be evident by this point that, when it comes down to it modern techniques like 12 tone, serialism, and so on rely a lot on principles found in counterpoint. However, like you've done here, the note relationships don't need to behave in a certain way unless we're shooting for a different style.*

The idea with "loosening" up is so that you actually feel comfortable working with a good degree of freedom. Interestingly, where one would perhaps think "Well total freedom is easy as pie, you can do whatever you want!" it couldn't be further from the truth. The less "rules" you have, the harder it is to have the job done for you.

For example, let's go back to Mozart. Mozart wrote a lot of music. A whole, huge, lot. However, as he got older he'd write less and less music. In his later years (though he died young) his output was much smaller than when he was younger. This wasn't because he was running out of juice, no. It was because he was trying to go do something else, something different.

It's no surprise that the weirdest compositions come from this time, where he was trying to reach back to past techniques (counterpoint, Bach's style, etc etc) while at the same time trying to experiment with new ideas (at the time).

It is also evident that in the 20th century, as stylistic rules became more and more of a choice rather than a norm, composer output became progressively smaller unless composers did stick to a system. Sure, there were very prolific composers (Martinu, for example) but Ligeti's work is not even a fraction of Mozart's, in sheer volume. However, Ligeti had a lot more to think about when he wrote, than Mozart did.

A composer today has a million options to pick and choose from, and because of that it's important to know what one wants. Sadly, there's no formula for this, but it does come with experience.

About Stockhausen, the Gesang der J

Heres the row:

1: C#

2: C

3. A

4. D#

5. D

6. E

7. B

8. A#

9. F#

10. G

11. G#

12. F

I listened to Penderecki's "Threnody" and found it frightening to say the least. I'm curious how he approached writing such a monstrous piece. I love Ligeti's Atmospheres, particularly the use sound mass composition to create "walls of sound".

Finale 2008 - [lesson4 - 12 tone].pdf

lesson4 - 12 tone.MID

lesson4 - 12 tone.MUS

Excellent. Well, now that you've listened to these pieces, let's leave serialism a little to the side and go a little over how they work.

Penderecki's Threnody was written almost as a reaction to the 50s serialism, in a sense that the way it's written allows for some freedoms of interpretation. The score is written in a graphical manner in sections with the goal of uncertainty.

Both Atmospheres and the Threnody use a technique which comes from Henry Cowell (and to a lesser extent Charles Ives), which is that of Clusters. There's a common misconception about what clusters are and how they work, so let's fix that. The effect you heard as "wall of sound" in Atmospheres is caused by the moving cluster textures across the orchestra, and in Penderecki's case, the clusters are used in a different way (which brings the frightening-like character to the piece since horror movies and such have used these techniques extensively. However, the sound is much more jarring than Atmospheres.)

Let's define a regular "chord" in a sense of intervals. You can tell a major from a minor, a diminished from an augmented chord because of the differences between the intervals between the notes that make up the chords. Chords are built in the norm by distances of 3rds, small or big 3rds, in a row. So, in essence, a cluster fills up the holes between each note and the next, so that you can't really hear which notes are being played individually.

Indeed, a cluster is an attempt to unite the sound of more than a single note, as a "single sound" rather than a combination of sounds. The way it does this is by eliminating intervals between the notes, though there are many other things that can be done. There are many types of clusters, and as you heard in these two pieces, they can have wildly different uses.

What defines a cluster is:

Size - How many notes you have going.

Dynamic - If the same dynamic is applied through the entire cluster, or not.

Sparsity - How far apart the notes are.

The last one is particularly important. For example, if you take a tiny 3 note cluster, C - Csharp - D, and play those notes right next to eachother, you get a very compact cluster. However, if you play these three notes across 3 octaves so they are far apart, the effect is entirely different, while retaining the texture from the first cluster.

Ligeti's piece, Atmospheres, has entire groups of clusters that move through each orchestra section, across sections. The clusters become narrower, louder, more sparse, softer, and so on as the piece goes.

Curious fact: the score for Atmospheres is written in regular notation, and in fact it's in 4/4! Ligeti probably went for the easiest way to write it, and used 4/4 only as an organization metric, since the piece's rhythm is very difficult to determine to say the least. This tendency to use time signatures only as an organization aid comes far back from examples such as Stravinsky's "Sacre du Printemps" (Rite of spring) where the orchestra is accentuated outside of metric-beats, so that the time signature isn't defining the rhythm of the piece, but rather just keeping it readable for the musicians.

You can also see how different the implementation of the orchestra in Ligeti's piece is compared to what we usually have as "orchestration" for Romantic, Vienna Classic, and so on periods. The parameters such as Rhythm, Melody, harmony, and so on are so blurred that you don't perceive them in the traditional sense, despite their existence regardless.

This tendency to uncertainty is largely part of a quasi-revolt to the super-strict and exacting character of Serialism. To put it a little into historical context, in Europe there major turmoil with the 2nd world war, and in Germany (which was before the war one of the leading nations in the avant-guarde, modern music, and so on) was basically all beaten down. Composers who did not write in certain "aesthetics" (Like Wagner) were either forced into leaving the country (Hindemith, etc), stop composing (Webern, etc) or worse. And overall, there was a general "backwards" tendency in the 40s with Richard Strauss (Who had success during Nazi germany due exactly to his Romantic/Wagnerian-aesthetics) and the general Neoclassical trend.

I've said before that the only reason composers have generally looked to the past, was to find new inspiration, to try new things, and to understand where they are and what they can do to move "forwards" in a sense. Hindemith and Stravinsky (early Bartok) were composers who took this approach, and their Neoclassical/Neoromantic compositions are not just reviving old techniques or styles, they did a lot of things which were very different, new, and explored far deeper than just recreating or imitating something.

With Strauss a similar tendency can be seen ever-so-slightly but due to political pressure it never got that far. The 4 last Lieder ("Vier Letzte Lieder") from him can be considered a compact portrait of everything that has happened in music history until then, harmonically, and so on. Then there were composers such as Hans Pfitzner who were a lot more interested in political reasons (in his case, it's argued that he tried to gain popularity following Strauss by writing music the Nazis found acceptable. The outcome wasn't so bright for him, though, and he ended up being just a footnote in what otherwise was a terrible period of music history.)

Then, came the end of the European war, and people began flocking back to Germany, establishing festivals for new art in an attempt to make up for all the lost time during the war. From this attempt, came such festivals as the Documenta (Kassel) and the festivals/summer courses for new music in Darmstadt (where people such as Xenakis, John Cage, Stravinsky, Stockhausen, would meet with young composers of the time, do seminars, concerts, etc.)

The actual start of Serialism as "The serialization of all musical parameters" (opposed to 12 tone, which is just pitches, and there are other aesthetic principles beyond just that, making both rather different in practice.) can be traced to a piece from Messiaen. Messiaen himself was a teacher during the Darmstadt summer courses/festivals in 1949 and 1950, and his piece from this time "Mode de valeurs et d'intensit

I'm a little hazy on a few things:

Should I create my tone row with the harmonic implications in mind?

Let's say I decide to make a string quartet, does 12-tone forbid doubling a note in two or more voices?

What dictates the use of inversion and retrogrades, does it just cycle through all of them consecutively or am I free to choose when and where this changes?

I'm a little hazy on a few things:

Should I create my tone row with the harmonic implications in mind?

Let's say I decide to make a string quartet, does 12-tone forbid doubling a note in two or more voices?

What dictates the use of inversion and retrogrades, does it just cycle through all of them consecutively or am I free to choose when and where this changes?

Well, try to get a sound out of it that you like while still following the rows and working with the rhythm. The inversion/retrograding, etc etc, is all personal choice, you can as well not use any inversions or retrogrades at all and do only rows with different rhythms, pauses, and so on. There's a lot of possibilities, see what works for you.

If you do a string quartet, each voice has it's own row/inversion/whatever, you CAN double (two instruments that start from 0 will hit have the same notes, but you can vary that too in rhythm, etc etc.) If you do double-notes on strings (if it's a string quartet) the second note must also be from the row.

I decided to try a string quartet. The row is the same as before.

soprano: unaltered

alto: inverted

tenor: retrograde

bass: inverted + retrograde

I found 12 tone very difficult to work with. I had some trouble keeping track of the rows and I think I wound up repeating a few notes here and there. However, I could see how this system can be manipulated to suit the composer's needs. In the hands of a master, it can produce some surprisingly tonal sounding and/or beautiful results.

Finale 2008 - [lesson4 - 12 tone string quartet].pdf

lesson4 - 12 tone string quartet.MUS

lesson4 - 12 tone string quartet.MID

I decided to try a string quartet. The row is the same as before.

soprano: unaltered

alto: inverted

tenor: retrograde

bass: inverted + retrograde

I found 12 tone very difficult to work with. I had some trouble keeping track of the rows and I think I wound up repeating a few notes here and there. However, I could see how this system can be manipulated to suit the composer's needs. In the hands of a master, it can produce some surprisingly tonal sounding and/or beautiful results.

Not a bad exercise. Notice how depending on the rhythm, register, etc etc you can end up making very different harmonies, and yet at the same time maintain a borderline uncertainty. You can be on the edge of atonality, but never quite there or go directly for it.

I think that thus far, you've done pretty good overall. You've got a pretty active imagination. I particularly like what you did with the start of the quartet here, and you played some with the rhythms and pedals.

So, now let's leave serial techniques aside for now and look at something else that's pretty important for any composer. Besides music history, musicological fields (theory, etc) and techniques/styles, a composer has to have a really good grip on instrumentation, and more than that actual techniques related to each instrument.

Each piece that you write for any instrument has to be not only a search for getting the sounds you want into reality, but a research into the instruments/techniques you're using. If you write something for Tuba, for example, you might as well want know everything there is to know about it as to make the best use of the instrument. Specially when you're dealing with solo instruments, or small ensembles.

So, from this, I recommend that you listen to a couple of Luciano Berio's Sequenza pieces: III for female voice, and VI for viola solo. This is a whole different type of thing, and where I want to go with it is that sometimes just one single instrument is more than enough to get the point across.

In a sense, less is more sometimes, and don't underestimate an instrument only because classical literature has not been kind to it (poor Viola...)

If you don't have any objections, I'd suggest we actually start working on a piece. Think about what you want to do, what instruments to use, techniques or just the overall feel, sound or character. It doesn't matter if you can't decide on everything right away, but give it some thought and post it here.

I'd like to write for a small ensemble. Possibly a woodwind quintet/trio? I want to get a good grasp of the various sections of an orchestra. (Not like my knowledge of strings is any better) I'd like the piece to have a tonal center but I don't mind exploring other regions as well. I want it to sound "playful" and be rhythmically driving.

What's a good ensemble for my level of skill?

How about a wind trio, Clarinet, Bassoon and Oboe?

Sounds good. Where should I start?

Sounds good. Where should I start?

Well, first thing, read up on the instruments, ranges and so on. Next, start writing sketches and trying out things. Work a little see what kind of stuff you can use, want to use, or avoid.

PS: Take your time, really. It's important that you come up with material that you can use, rather than actually get "things done." You can draw everything else from this first planning stage, and so on.

I have a question. Let's say we have a melody a measure long, and the melody repeats itself while getting transposed down a semi-tone each measure. What would be the best strategy of harmonizing something like this?

I have a question. Let's say we have a melody a measure long, and the melody repeats itself while getting transposed down a semi-tone each measure. What would be the best strategy of harmonizing something like this?

Well, if you have a sequence like that you can simply transpose the previous harmonization, OR you can harmonize it new as if it modulated, etc. Depends on what notes, what melody, and so on. It's hard to tell without examples, but you can do a lot of things. You can even just use the same harmonization and let it become dissonant on purpose, and so on.

Speaking of which, if you're writing for winds, there are a couple of composers I want you to check out, the first one is Bohuslav Martinu, specially the madrigals for wind trio (oboe bassoon clarinet.) And the other is Jean Francaix who has written also a bunch of work for wind ensemble. I have a hunch they may be to your liking.

Say, if you gots stuff you can also show what you're working on. If you don't mind. If you haven't had time, that's fine too. But, uh, it'd be good if you kept me posted otherwise I can't be of much help. Just a reminder~

Sorry about this. I only have a few measures finished. (my dog ate my measures) I started working on a couple of other things and this has sort of drifted into the background. I suck at managing my time. Working on 3 different things on a single day is a monumental task in itself for me. This is embarrassing.

Funny thing is, what I have written has been sitting next to me on paper for the last 2 weeks. I just haven't been able to pick up a pencil. My time and energy has just been going in a thousand different directions lately. Time to get focused.

Here's what I have so far. I'm not so sure what's going on harmonically in the beginning. It's something like: V7 - I, V7 - I, V7 - I, getting transposed down each time and winding up on a C major chord.

woodwind trio.MID

woodwind trio.MUS

Finale 2008 - [woodwind trio].pdf

Hm.

I like what you got, do you have anything planned for the structure? If there is any? What do you want to develop in terms of material? Rhythms? Motives? Dissonance or harmony?

Also, remember that these people need to breathe so large passages need to be broken up into phrases. You can use slurs here to indicate when to breathe, etc etc.

As for the harmony, well, don't worry too much about it unless you have a specific scheme you want to adhere to. You can do a chord-map for example, and have free-modulations through it as the structure of the piece. As an idea, anyways.

I get the vibe you're treating the oboe as a sort of "primary" instrument and the other two as accompaniment. There's nothing wrong with that, but remember that in these types of small ensembles you can hear clearly the other voices, so you may want to consider how you balance them differently.

But, it's not really bad how you have it. Maybe it can be interesting to mix it up a little later on in the piece.

You also clearly have a motive/theme at the start, with a pretty particular harmony. Are you going to use it again? After the C major chord at the end of the fragment you could as well work with that sort of "theme" some, with some variations rhythmically maybe.

Like I said, it's pretty nice what you have so far. But, try to think ahead of your writing. Where do you want to go with this? Do you have any sort of particular length in mind?