Developing an organic songwriting process and avoiding over-complication

[Nyrrth]

I'm OK with coming up with rudimentary ideas and riffs, but find it quite hard to build on that and make it into a song in an organic way. Sometimes I'll string together a bunch of riffs, but it never really leaves me satisfied as it often feels forced.

I also tend to end up getting lost in the maze of trying to over-complicate the idea. I'm into progressive rock and metal, so I'm naturally attracted to odd time signatures, polyrhythms, textures etc. I might come up with a part that's somewhat complex, but then I end up trying to outdo myself and make it even more complex and so on. End result - I get frustrated, lose interest in the idea altogether, and move on to the next one. Rinse, repeat.

Anyone else deal with this and have any advice on how to go about developing an idea into a song?

Here's my Soundcloud page:

xxPorcupinexTreexx's sounds on SoundCloud - Create, record and share your sounds for free

The first track there is an example of what I'm talking about. I'm not feeling inspired by how to proceed with it, and the somewhat jazzy part at the end is something I just threw in there, but I'm not sure if it works with the rest of the stuff.

For those interested, the second track there is an Opeth cover I did myself (including the manly female voice at the end - thanks to pitch-shifting and the formant plug-in in Logic!). I'd appreciate your comments on it though from a recording/mixing standpoint - since I'm very new to the whole DIY production thing (as another post I made today in the recording forum would suggest).

Cheers!

 

[Overtone]

I like the way the main section builds up. I think the break either comes in too early or it is too long, because the main section could have kept going a while longer. So maybe the clean would work as a quick break if it was shorter and then back into the main theme, or you could put it later and make that a transition into a new section. It is all about thinking about how the different parts can be used to make a flow that is interesting.

Ideas seem to come more easily when I
-take a break. if the chords are stuck in my head i come up with melodies naturally while I am making lunch or walking outside or whatever
-jump to different instruments.
-start recording or sequencing

 

[Solodini]

In uni, I had a reharmonisation exercise in which I had to take a song, come up with a new accompaniment and then come up with a new melody to replace the original one. If you have an idea which you've accompanied, come up with a new accompaniment for it, a new melody for that and see if they work as bases for consecutive sections.

 

[Mr. Big Noodles]

It sounds good, but it rambles a bit. You guys are identifying separate sections in the music, but I just don't hear it; there's not enough contrast to justify a new section anywhere. Rather, what you have recorded is sort of like a 'theme group'. I would advise composing something in a different key with a different feel, then smashing the two together and see how it sounds.



--------------



I'm going to talk for a while about form, since we're here now. Form is, perhaps, the most essential part of songwriting, so that's where you want to put your focus. One achieves interesting form by arranging a series of contrasting and related sections, so as to create flow and multi-dimensionality in the music. Some good forms to look into:

Binary: AB, or AABB
Ternary: ABA
Repeated binary: ABAB, or ABABABAB... however long you need it
Strophic: AAAAA... Kind of monotonous, but I'm bringing it up anyway.
Rondo: ABACADAE... Alternates an A section with new material in some arrangement.
Five-part rondo: ABACA
Seven-part rondo: ABACABA
Arch form: ABCBA, may be extended to something like ABCDCBA.
'Song' form, or 32 bar form: AABA (typical of American jazz standards), basically a ternary form.
The other song form: ABABCB, or verse-chorus-v-c-bridge-c

The idea with most Western musical forms is that there is an idea stated at the beginning (A), then a departure (B), and finally a return (A). Repetition is obviously a fundamental part of this equation. Sometimes, this material is extended so that the expository material is AB, then the departure is C, and the return may be AB, A, or B. None of this stuff is set in stone, though, so do whatever you want. However, I encourage you to examine form with an open mind. When I say it with letters, it might seem really dry and that there is no real difference between AB and ABA, but you'll find that each form has a particular character. I'll demonstrate how these things work real quick.

The most basic form is just one section that repeats, the strophic form. We see this sort of thing a lot in folk melodies that have eighty-thousand verses.

So, if you want your music to sound like a children's lullaby, repeat the same eight measures over and over again. This form doesn't go anywhere.

Binary forms similarly are pretty static, but the addition of a contrasting section breathes some life into the repetition.

Art Bears - The Winter Wheel


0:18 - A
0:45 - A
1:15 - B
1:39 - A
2:05 - A
2:34 - B
2:52 - Coda based on B material (not part of the form)

You don't see a lot of binary form in pop music, as it's quaint. Being that this form consists of limited materials, it still doesn't go very far and has that folky feel to it.

Inserting a third section tends to provide a bit of contrast and welcome relief from the old material, particularly if there are repeated sections before C.

Opeth - Harvest


0:00 - Introduction (not a part of the form)
1:06 - A
1:36 - B
2:14 - A
2:44 - B
3:17 - C
4:18 - A
4:48 - B
5:26 - Coda

There are forms that have a defining characteristic, such as the rondo form which is known for its repeating A theme. And here's a rondo from everybody's favorite dickbag:

Yngwie Malmsteen - Fugue (despite the fact that it has nothing to do with a fugue)


0:00 - A
0:33 - B
0:49 - A
1:14 - B
1:30 - C
1:58 - D
2:25 - A
2:57 - B
3:14 - A
3:29 - D
3:51 - a (smaller because it is truncated; sometimes a short A is used just to wrap things up, such as in rounded binary form [ABa])



----------------



All in all, this might not be the answer you're looking for, but having command and understanding of form will help you to conceptualize and write songs easier and better.

 

[Nyrrth]

Thanks, folks! Some very good ideas there. The sample I had wasn't intended to necessarily flow that way. It was just a bunch of ideas I put together, which may or may not be part of the same song when it's all finally done.

All in all, this might not be the answer you're looking for, but having command and understanding of form will help you to conceptualize and write songs easier and better.

While it was probably not the answer I expected, I think it was a huge eye opener in terms of analyzing music! I listened to all those examples and looked at your analysis, and they make total sense. Thank you!

I would advise composing something in a different key with a different feel, then smashing the two together and see how it sounds.

As far as key changes are concerned, do you have any tricks (for lack of a better word) on how sometimes these are accomplished? I'm vaguely familiar with concepts like modulation to the dominant, or even for that matter something like Coltrane changes. Don't ask me to describe it though! I just know of those ideas, so if needed I could look up the concept to understand how it's actually done. But in general, are there some ground rules or methods to accomplish key changes that don't sound completely random?

Lastly, perhaps another question for SchecterWhore, but others are of course welcome to comment. In the original sample on my Soundcloud, how would you break down those sections into A, B, C etc? Going by your analysis I would probably think it is something like:

Intro: Not really part of the form
First heavy Riff: A
Quiet part with bass: Not really part of the form, some kind of bridge
Second Heavy Riff: B
Jazzy end: C

Would you agree? You also mentioned "part of a theme group", so could you elaborate on what you meant?

 

[Mr. Big Noodles]

As far as key changes are concerned, do you have any tricks (for lack of a better word) on how sometimes these are accomplished? I'm vaguely familiar with concepts like modulation to the dominant, or even for that matter something like Coltrane changes. Don't ask me to describe it though! I just know of those ideas, so if needed I could look up the concept to understand how it's actually done. But in general, are there some ground rules or methods to accomplish key changes that don't sound completely random?

Let's break modulation down real fast. We change key in order to make music more dynamic and tonally interesting. Certain key relationships have a special place in tonality, such as moving to the dominant, the relative major/minor, and the parallel major/minor. You might class the subdominant in those special relationships, as well, but it's a little rarer. Then, there are modulations that are more colorful, such as moving by mediant relationships (thirds, ex: A major to C major, E major to G# minor), by minor second (cheesy), or by tritone (also cheesy, but happens occasionally). Some modulations, such as movement to the supertonic, do happen often enough, but I don't think it's a big deal.

Depending on where you start (in a strict academic model; there are no absolutes, but we need to make a few assumptions), there are expectations for how a modulation will unfold. For instance, if a piece begins in a major key, it is generally assumed to move to the dominant key area, and if not that then the relative minor. If you look at the ternary form, ABA, then apply that key scheme to those sections, you get the simplest and most fundamental chord progression in Western music - I V I. If you go to the relative minor, then it's expected that vi is just a stepping stone to V, since that V-I progression is so important.

When you start with a minor key, it's expected that you go to the relative major, although modulating to the major dominant happens on occasion, and I think that modulating to the subdominant is somewhat more common with minor keys. The minor mode is more unstable, so it has some options, but if you want to sound safe and balanced, then going to the relative major gives two easily delineated characters.

I'm going to talk for a while about a piece I composed for school a few years back. It was a five-part rondo (ABACA) in C minor. I think my key scheme was such:

A - Cm
B - Gm
A - Cm
C - No real key, modulated around a lot.
A - C

If you look at what's going on there, we have that repeating A that is a feature of rondo form. The program for the piece was one of journey and return, so I wanted it a bit unstable. Instead of taking it to the relative major (Eb) for the B section, I opted for the more colorful minor dominant, giving the sense of further departure. Then, back to the A section, we're still wandering around. In the C section, there's a lot of instability, and the final return of the A is in the parallel major, some sort of jovial arrival. In this way, there is a sense of transformation from the beginning to the end of the piece. So, I'm matching my pitch centers to my concept.

I remember also looking at a Robert Schumann piece that took a chord progression (Something like I iii vi ii V I, I forget.), then expanded each of those chords to become its own key. If the key was D major, for example, it would move like this: D major, F# minor, B minor, E minor, A major, D major. For that, he was taking a smaller idea and expanding it to a larger form. Kind of process-oriented, but effective.

Hopefully, this will give you an idea of how to pick your keys.

-------------

As far as actually getting to those keys, these are the types of modulation:

Direct modulation, or phrase modulation - End your phrase in the first key, begin next phrase in the second key. No preparation needed. I've heard this referred to as "Hollywood modulation", implying that commercial film music is lazy shit. It's all good, though. This is usually pretty effective.

Common chord modulation - A chord that is present in two different keys is used as a pivot point to get to the new key. For instance, if we're in D major and want to get to C major, you might have a progression like D G A7 Bm Em F G7 C. Yellow is D major, red is C major, and blue is the chord that is present in both keys.

Common tone modulation - Same as above but only with one tone that is present in both keys.

Chromatic modulation - Any modulation that uses a secondary chord in order to modulate. Let's go from D major to C major again: D G A7 D7 G7 C. The green chord is a secondary dominant chord, V7/V in the key of C major.

Enharmonic modulation - Takes a dominant seventh chord or a diminished seventh chord and changes its function. For instance, G7 is either V7 of C, or an augmented sixth chord leading to F#. Here's a rather distant modulation: D G A7 G#m. The A7 is acting both as V7 in the old key, as well as +6/i in the new key. This is similar: D G A7 G#7 C#m. Same deal, A7 is +6/V in the new key. With diminished seventh chords, you rely on the fact that it's a symmetrical chord, so B°7, D°7, F°7, and G#°7 are essentially the same thing. A song that I wrote uses a progression that exploits this: F#m9 G#°7 Am9 B°7. The diminished chords are acting as vii°7 for four different keys at any given time.

Change of mode - Not really a modulation, just going from one mode to the next while staying on the same pitch center. For example, E minor to E major, or F phrygian to F lydian.

And I'll get to the other questions in a bit...

 

[Mr. Big Noodles]

Lastly, perhaps another question for SchecterWhore, but others are of course welcome to comment. In the original sample on my Soundcloud, how would you break down those sections into A, B, C etc? Going by your analysis I would probably think it is something like:

Intro: Not really part of the form
First heavy Riff: A
Quiet part with bass: Not really part of the form, some kind of bridge
Second Heavy Riff: B
Jazzy end: C

Would you agree?

You got it, dude. Spot on. My personal feeling is that it's not finished, though, and sounds like it's all part of the same section. Which brings me to the next point...

You also mentioned "part of a theme group", so could you elaborate on what you meant?

Imagine you have a box with some apples in it. Some are red, some are green, some are large, some are small, but they're still all apples. In your excerpt, you have three different apples (Fuji, Granny Smith, and Jazz Apples ). Since that's all you've recorded, we have to call those different sections. But imagine somebody gives you a crate full of citrus fruits. When you place the citrus next to the apples, there is a much clearer division, even though the apples are different in and of themselves. This is what is meant by "theme groups".

Sonata allegro form does this. It's a dramatic form, so the idea is that there are two themes that have opposing characters with the intent that a conflict will arise between them and be resolved after the climax of said conflict. Those themes will consist of multiple melodies that share similar characteristics, hence theme groups. The form is like a ternary form, consisting of the following sections:

Exposition - Contains the two opposing theme groups, and usually closing theme. Sets up the material for the rest of the piece. Theme I and Theme II are in different keys. The exposition is frequently repeated in order to fix the themes in the ear, but this practice begins to die out around 1860 or so.

Development - The themes are pulled apart and manipulated. The music explores the tonal space, modulating all over the place, getting destroyed and being built back up. These consist of a number of phases in which the music changes character.

Recapitulation - The first theme comes back in its original key, then the second theme returns but in the same key as the first theme (resolution of conflict).

And there's usually a coda.

Ludwig van Beethoven - Symphony 5, mvt. 4


Unfortunately, this is a shit recording of this movement. I have a recording of Otto Klemperer conducting the Philharmonia Orchestra that is at a better tempo and doesn't have musicians screwing up everywhere. Additionally, this recording omits the repeated exposition, which I think is a mistake.

Exposition

Theme Group I -
0:00 - Subtheme 1
0:33 - Subtheme 2 - Modulates to the new key.

Theme Group II - 0:58

Closing Theme - 1:25

Development

2:00 - Phase 1 - Morphs Theme I, lets listener know we're going somewhere else now.
2:32 - Phase 2 - Sequential modulations, distorts sense of tonal center.
2:58 - Phase 3 - Works up to dominant pedal.
3:34 - Phase 4 - Messing around with Theme II.

Recapitulation

Theme Group 1 -
4:00 - Subtheme I
4:32 - Subtheme 2 - Doesn't modulate.

Theme Group 2 -
5:02 - In tonic key area

Closing theme - 5:37

We've been saying that introductions and codas and such aren't formally significant, but Beethoven likes to turn his codas into a second development section.

Coda

5:53 - Theme 1 modulating.
6:16 - Authentic cadences give the sense of things finishing up...
6:23 - … but it keeps going.
7:04 - Theme 2 material, I think. I'm not going to double check.
7:15 - Based on closing theme
7:37 - Big fucking cadence.


This next one is perhaps a clearer and more modern approach to this whole "theme group" thing:

Opeth - White Cluster


The themes that enter at 0:00 and 0:22, respectively, have a similar character, no? They both are heavy as shit! Dun-dun-dun-dukka-dun! We'll say that those are both in Theme Group I. At 1:46 and 2:33, we have the themes that are a part of Theme Group II. Then, 3:42 is like the closing theme from the Beethoven thing. 4:29 begins some sort of departure (as a development section does in sonata allegro form, except it doesn't really develop anything here). Then, 7:17 is a recapitulation to the second theme group.

 

[Inverted11]

For those interested, the second track there is an Opeth cover I did myself (including the manly female voice at the end - thanks to pitch-shifting and the formant plug-in in Logic!). I'd appreciate your comments on it though from a recording/mixing standpoint - since I'm very new to the whole DIY production thing (as another post I made today in the recording forum would suggest).

Excellent cover, and the female voice modulation sounded great! I think the bass could use a little more eq in the mixing department.

 

[Nyrrth]

Wow, that's a lot to digest! While I read and re-read your post, I have some questions that come to mind right away.

(thirds, ex: A major to C major, E major to G# minor)

I'm a bit confused here. A to C would be a minor third, while E to G# would be a major third, correct? Did you mean thirds in general (major or minor doesn't matter)? Also, if I'm right in the previous sentence, in your example looks like the minor third chord (C) is being played as a major chord, and the major third chord (G#) is being played as a minor chord. Is there a way to represent this?

Chromatic modulation - Any modulation that uses a secondary chord in order to modulate. Let's go from D major to C major again: D G A7 D7 G7 C. The green chord is a secondary dominant chord, V7/V in the key of C major.

A bit confused here too. The green chord in the above example was D7, but wouldn't G7 be the V7 chord in the key of C major? Just wondering if I'm paying attention and pointing it out correctly, or missing something completely!

That's it for now....there will probably be more when I'm able to comprehend the rest!

 

[Nyrrth]

You got it, dude. Spot on. My personal feeling is that it's not finished, though, and sounds like it's all part of the same section.

Yes, it is in no way finished! Just a smattering of ideas strung together. As far as being in the same section goes, I assume you meant part of the same theme? After going through your post, I'd probably think that the jazzy bit at the end would be its own theme group though (at least when the song is properly constructed). The two heavy parts could be considered as two parts of the same theme since they have the same sense of aggression, but the jazzy part sort of digresses from everything else. Any thoughts on why it comes across as being in the same theme?

 

[Nyrrth]

Excellent cover, and the female voice modulation sounded great! I think the bass could use a little more eq in the mixing department.

Thank you! As far as the "female" voice goes, though, all I'll say is that I wouldn't hit on her based on that masculine voice (well, that, and that my wife wouldn't be too happy if I did!)

 

[Inverted11]

Thank you! As far as the "female" voice goes, though, all I'll say is that I wouldn't hit on her based on that masculine voice (well, that, and that my wife wouldn't be too happy if I did!)

Don't worry, my wife wouldn't be too happy either.

 

[Mr. Big Noodles]

I'm a bit confused here. A to C would be a minor third, while E to G# would be a major third, correct? Did you mean thirds in general (major or minor doesn't matter)? Also, if I'm right in the previous sentence, in your example looks like the minor third chord (C) is being played as a major chord, and the major third chord (G#) is being played as a minor chord. Is there a way to represent this?

What I'm talking about here is mediant relationships. Mediant refers to the space of a third between tonics or roots. A major to C major is a mediant relationship, as is A major to C minor, A to C# major or minor, Ab major or minor to C major or minor, etc. Mediant relationships can occur either on the level of chord progression (Major chords ascending by minor thirds at 0:12 here.), or on the larger level of key relationships.

The A section is in A minor, the B section is in F major. It's just modulating a third away. You know, that kind of thing.

A bit confused here too. The green chord in the above example was D7, but wouldn't G7 be the V7 chord in the key of C major? Just wondering if I'm paying attention and pointing it out correctly, or missing something completely!

Look again. D7 is V7/V (pronounced "Five-seven of five") in C major. It's a secondary dominant chord, meaning that it's supposed to lead to another function within the key. D7 normally resolves to a tonic of G, but we're only using G as a dominant instead. Alright, big chord progression. I'll make the chromatic chords red.

C C7 F B7 Em Am A7 Dm D7 C/G G7 C

I V7/IV IV V7/iii iii vi V7/ii ii V7/V I6-4 V7 I

Notice the pattern - V7/IV goes to IV, V7/iii goes to iii, V7/ii goes to ii... V7 of anything goes to whatever that anything is. You dig? And I6-4 is figured bass for a second inversion tonic chord, in case you didn't know.

Yes, it is in no way finished! Just a smattering of ideas strung together. As far as being in the same section goes, I assume you meant part of the same theme? After going through your post, I'd probably think that the jazzy bit at the end would be its own theme group though (at least when the song is properly constructed). The two heavy parts could be considered as two parts of the same theme since they have the same sense of aggression, but the jazzy part sort of digresses from everything else. Any thoughts on why it comes across as being in the same theme?

"A" and "B", as you call them, are the exact same texture. The notes are different, but the mood is the same between them. "C", indeed, is markedly different, but A and B are kinda insubstantial as far as thematic material goes, and C is in the same area tempo-wise. So, rather than hearing two complete sections, I hear something that's textural and hazy, and a melody. It's a process of unfolding, not a bunch of neat little sections that line up for easy analysis.

That's it for now....there will probably be more when I'm able to comprehend the rest!

Come at me, bro.

 

[LetsMosey]

@SchecterWhore - great insight!

 

[Guitarman700]

Schecterwhore is the best thing about this forum.

 

[revlover]

Schecterwhore is the best thing about this forum.

 

[Ryan-ZenGtr-]

A different approach to the problem of writing, compared to SchechterWhore's in depth suggestions:

A friend came over to jam and write one day. He was a great guitarist, left handed and a fine artist and tattooist. He was also studying to become a luthier as left handed guitars were harder to come by in the styles he liked.

He gets setup to play, the red light goes on and ... nothing. He says he needs to warm up a bit, proceeds to plays something amazing. I said to him, why not record that?

He said; "It's not good enough."

I struggled a lot with this phenomena. I call it "The Cool Filter"; ideas which do not surpass expectations are rejected. The trouble arises when so many ideas are rejected there is nothing to work on.

When I'm working on a group of ideas for development into material I always seek to get an arrangement so I know which ideas will be developed. I end up throwing away lot's of ideas as they are hard to include in an arrangement. My understanding now is that arrangement is king.

Record everything!

 

[Nyrrth]

SchecterWhore - yes, I think I understand the whole mediant relationship concept, thanks! I also get what you meant by V7/V, V7/iv etc. I neglected to look at what was after the "/", but now I understand what you were talking about. The V7 is basically the dominant chord in the key of what the original chord is modulating to, if I'm still with you.

More later...

 

[Mr. Big Noodles]

Bingo. vii°7/V (or whatever) fulfills the same function. One last thing: you'll notice that in my progression that I have V7/V I6-4 V7 I at the end. I6-4 is delaying the dominant chord - the "I6-4 V" figure is called a "cadential 6-4".