I have been learning the acoustic guitar for about 3 years, but just recently started writing my own songs. I am currently learning music theory.

Given that starting point what advice would you give to a novice composer? Books to read? Anything to avoid?

Just to clarify I am not asking how to be a great or the greatest composer, but rather any advice that would be helpful to new composers.


AbstractDissonance raises some good points; you don't 'need' theory to write good music, in the same way that you don't need good grammatical knowledge emote an elegantly constructed sentence when speaking. In both cases however, knowledge helps.

Understanding how something works pretty much always improves your ability to utilise it; this is no less true for music than it is for anything else.

If you have been playing for three years; you'll probably find that you already know much of what you need to write good music. But i'll make some suggestions anyway.

Chord progresssions

Understanding chord progressions is essential to writing good music since they form the basis of many/most styles of music. In order to fully understand them you need to understand the major scale and how the chords in basic form relate to it. This will instantly have you writing recognisable tunes and eventually even hearing progressions in the music that you listen to. Once you have the diatonic triads (check out this awesome tool for a jump start) down and you are making music with them you might want to look at 7th/extended chords.

Circle of Fourths/Fifths

The circle of fourths/fifths is a tool extensively used by song writers the world over; these types of progression can be heard in almost all forms of popular music in one way or another. Here is a chart I found on the web:

Circle of fourths/fifths

Going counter clockwise the chords are a perfect fourth apart, going clockwise, a perfect fith. You can start in any key of the circle and the pattern remains the same. Here is another perhaps clearer version of the circle.

So in C a: i - iv - v - vi progression would be C F G Amin (obviously you could extend these chords however you wanted).

This is by no means a complete description; I just think these things are worth researching and learning for anyone who is writing songs.


This is a question with no answer. No one knows how to compose great music and anyone can compose crap. There is no formula or equation. You can learn all the theory, techniques, etc but some guy that knows nothing can come along and blow you out of the water. Why? Feeling, talent, etc...

It also depends on what you want to do. If you just wanna create simple child songs then it will be easier than composing songs that will effect the direction of humanity.

No one can tell you now to be a great composer because it depends on the person(everyone is different) and no one knows. The best I can do is simply tell you that you do not need theory to write good music. What is most important is that you are able to express yourself through your instrument. This means you have an intimate connection to your instrument. That it is an extension of you.

Also, copying off others is a good way to learn. Take a song and change it up. Pick your poison... change the chords, the rhythm, the lyrics, etc... We started out by copying other humans to learn to talk, walk, read, etc... music is not going to be much different.

Music "theory" is about communication. It's so you can tell me how you did something but you can't tell me WHY you did it. The "why" is what makes the difference between someone good and someone bad. It's an emotional thing. But to be able to express this at all you have to be able to get it out of your head and that involves an instrument(whether a guitar, pencil, piano, mouth, etc...). Be good at your instrument and it will be easier for you to be a good composer.

  • 6
    Overall I agree, but the question was, basically, do you have any advice for a new composer with some limited music theory. And trust me, I have certainly composed some crap. :) I am trying to minimize that, but... – Anonymous Jan 29 '11 at 23:39
  • 1
    Sure, most people compose crap. I've learned 10 years worth of music theory and it's done little good. I can compose a lot of baroque and classical stuff using it but it doesn't sound great. You can't compose music by the numbers. It's fine to learn theory in depth BUT only after you have developed your ear. Else you'll most likely use theory as a crutch. I know a lot of great musicians that don't know any theory but they use some "advanced" harmonies, correct voice leading, counterpoint, etc... All because they use their ears. – Anonymous Jan 30 '11 at 0:52
  • I come from a mathematics background and the theory was easy for me to learn and implement(after doing real analysis for a year picking up a music theory book was like reading Dr. Seuse). It did nothing for me musically because it doesn't tell you when to use something. It does give you common practices and you can use that to fool some people into thinking you know what your doing but it will never get you to the level of greatness(only your desire to express yourself and devotion to it will). – Anonymous Jan 30 '11 at 0:55
  • 1
    The idea of learning theory is so you can create music that is acceptably close to the classical composer masters, like Bach and Beethoven. If you lack inspiration or good ideas then theory can't help you but at least you'll have a fighting chance creating something passable. A side-effect is you can intelligently discuss and understand music that is non-trivial and be able to understand the harmonic structure of it giving you another fighting chance of playing through the changes without embarrassing yourself. None of it guarantees genius though. – Anonymous Jan 30 '11 at 3:58
  • 3
    You are wrong: It is not a question with no answer - it is a question with potetially many good (and different) answers! What you mean is probably "It's a question that does not have ONE correct answer." – awe Feb 2 '11 at 12:45

There's a chord progression map based on harmonized scale. here's a link


and here's a sample in a key of C

chord map in key of C

This one really helped me alot. All you need is to go from C or Am jump to any chord but next you should go where arrows leads or vice versa... green charts is for substitution for example you can change this progression: C F Em Dm to this C F B Em or even so C F B-Em Dm.. Try it it's lot's of fun =)

  • Awesome! Haven't seen anything like this before. – Bella Feb 16 '11 at 20:00

I improved my songwriting skills by attending songwriting circles. The general idea is that a group of songwriters meet once a week or once a month and everybody brings a new or newly-revised song.

You play your song, and people tell you what's good or bad about it, then you do the same for the others' songs. When these circles work they are a powerful tool. They're popular in folk muslc communities.


I doubt that knowing music theory has alot to do with being a good songwriter, but it probably cannot HURT.

Singable songs are ussually pretty simple. There is no reason to think you are cheating by using only 3 chords. Once you have the basics down, maybe then, use theory to add a bit of complexity, and this is where some theory might come in handy.

As an example, I, who don't know hardly anything about theory, had written this progression, actually, I found a list a chords and rearranged them and they sounded cool, anyways, I wrote out the progression for someone who DID know theory, and they were able to tell me just by plugging in the chords to their understanding of theory, what chord to play if I wanted to send the whole thing up a fret, or what you call, modulation.

That is one example of how theory can help, but I don't think it is a necessity. Sometimes, it takes more guts to write a song with just 3 chords, its so simple, you feel like a dufus, but it works.


I would study what works for you. Pick some pieces you like, and pick them apart. Think about exactly what in the piece makes you react favorably; is it the instrumentation or the arrangement? The sentiment of the lyric (if any)? How about the melody? Or is the chord progression out of the ordinary?

Also try this with pieces that don't work for you. What specifically turns you off? And in both cases, think about why the composer/performer decided to do it one way and not another. Would another way fix what's wrong, or break what's right?

And keep in mind something saxophonist Charlie Parker is reputed to have said (seriously paraphrased): Learn all your theory, then forget all that and play.


As DRL pointed out, a bit (or a lot) of theory cannot hurt. However, when you study theory, you are often stylistically thrown into jazz (if you're interested in 7ths), or basic classical music, and it does not immediately relate in a useful way.

Still, the notion of chord as having functions, movements that you can easily identify (your ear knows them, you just have to learn the names) and of understanding the basic notion of harmonic rhythm will be extremely useful at some point; not to mention rhythm if you need some grounding there.

From a rock/metal/grunge/whatever guitar player's point of view, though (that would be mine), you might be more interested in riffs and the way they interact. Then I suggest studying modes, but applying that directly to riffs you love or wrote. If you know metal, you know phrygian, for instance.

Also, my personal advice is : play from the heart first, either alone or as a band. Then study the music that comes out, which might point to new directions, which may or may not stay in the piece. Repeat as often as necessary. Bon appétit.

For any french speaker, I'd heartily recommend "La Partition Intérieure". It's primarily aimed for jazz improvisation, but there' so much knowledge in there, and it's written to help general understanding and musical applications, so anyone will find something in there to keep these creatives juices going.


It is useful to know some basic theory. See DRL's answer.

When composing for guitar, it can also be exciting to experiment with chords. Keep the basic keys in the back of your head, but try out some combinations that might look extremely advanced on paper, but are easy to play on a guitar. This means you don't sit with pen and paper when you compose, you just play around with your guitar, and take notes when you hear something that sounds interesting (it might not even be "valid" chords, or if you find it out, the chord notation looks like a mathematic equation).

Last I want to say that writing music is hard work. You need to practice to be good at it. You also need some talent, but you don't know that for sure until you have tried for a while and see the results. If your friends like the music you create, you might have potential to create some decent stuff. If you (truly) like the music yourself (be honest!), you have potential to be great - we often tend to be our hardest critics ourself...