for those of you that are interested in analysis of the music we love to learn and play, how do you approach a new piece? i want to analyze pieces as i learn them so i can get better at understanding composition in hopes of writing my own, but could use some advice of how to go about this.

i typically have been looking at intervals between notes and the horizontal progression of chords, trying to identify if it effects tension or release. been trying to identify the types of cadences being used, whether perfect / imperfect, masculine / feminine etc. been trying to break down the pieces into different sections i.e. for sonata form indentifying the exposition, coda, development, recapitulation etc.., trying to identify key changes and the building up to them. are they prepared? are they not? etc.

i dunno, i'm kinda new to all of this so i'm sure there is lots i don't understand yet. what do you look for or try to pin point? i've been trying to read some books but i feel like i'm in a wierd limbo state, either it's too simplistic or far too advanced. any good reccomendations for intermediate theory books?

i know a lot of people aren't fans of the analytical side of music, but for those that are...a little input please.

thanks :p

Hmmm...

It really depends on the style of the music you're looking for so this is what you look for first.

1. You need to realise what the music you're looking at is "made off", or what era it says it is. Beethoven writes differently than Ligeti, or Bach. A twelve tone piece will be differently approached than a tonal one, or something else, etc.

So this is your first big step. This usually does take a bit of time, and also includes further steps as well (I mean, if you've seen that it's a tonal piece, you'll probably also have noticed the tonality, and the changes elsewhere).

2. In tonal music, you should go for a harmonic analysis. I IV I 6/4 V and I etc... If something goes to a weird place... well it means that a modulation has taken place, plainly. Where and how it is up to you to find out. Just keep in mind that for a modulation to actually "work" successfully, a cadenza is also needed, no matter how small. You can't establish tonality without saying V I in some way.

3. You will notice repeated elements here and there. You can very well spot some phrases which are repeating themselves here and there, or maybe changing themes (1st and 2nd in a sonata, for example). After that larger and larger, to which at a point you'll realise the full form of the piece.

The 3rd step could also be 2nd very well. It's up to you. In short it's what people call "FORM". A sonata? A fugue? A scherzo? What is this? It's most important, because you can expect things afterwards and actualyl skip large chunks of the piece! I mean recurring themes don't take too much time, but if you don't know why they are there, or if they are in fact recurring then...

In non tonal music, the 1st step is rather important. You will need to realise what's going on and if there is something that "sticks out" sort of saying. Coherence is most important in composition and so things that are repeated (could be a pattern, could be an interval, could be a series, could be a rhythmic part, could be an orchestration quirk, whatever really), are the ones that will define what you will do next..

I work in small forms, so for me, I usually start with a harmonic language and melodic or textural idea. Everything else (if I'm lucky) proceeds from there.

I guess I look at the harmony too.

This is like analysing how one analyzes!

But for example something easy like Bach's Prelude in C, this is a good piece, as many are, to analyze. Seeing the harmonic progressions, etc... If you are into composing, try taking Bach's progression and change it up with different rhythms and melodic structure or make it more harmonic and less melodic.

You are not alone. I for one love analysis.
A true geek I suppose.

to add to pianobuff saying, you can also play with rythms, here's a nice video of the canadian brass jazzing-up Bach, no piano but very inspiring band!

http://www.youtube.com/watch?v=miB8p0Kgv5c

I've always thought the purpose of analysis is to get under the covers of a piece of music to determine why it is or is not effective. Thus the analysis must be geared to dissecting a particular piece of music. A quick look at the C major prelude might lead us to believe that a harmonic analysis of the 5 voice chorale would be sufficient, but does knowing the harmonic progression really tell us anything? Don't the register shifts really indicate the dramatic outline of this piece as much as anything else?

My point is that harmonic analysis isn't particularly useful in telling us why a piece is effective unless it's correlated with dramatic outline and other features. For example for me the climax of the Hammerklavier fugue is the fugal entry just following the false entry. Without looking at the score I'm willing to bet this entry is probably the highest entry of the piece. The false entry helps refocus attention on the theme and prepares the listener for the climactic entry. The music that follows gradually moves into lower registers and becomes less busy eventually coming to a stop, at which point Beethoven ramps everything up for the ending. Do we really need to know the harmonic analysis to tell us more? It would provide a more complete picture, but it's the dramatic details that will be most useful for future composers in determining how to structure the climax and ending of a fugue for effectiveness.

No analysis is complete without a harmonic analysis even with nontonal music. So I guess I agree with Nikolas for the most part, but I also like to look at matters of rhythm, texture, instrumentation and especially whether the composer writes effectively for the chosen instrument(s). There are so many questions that one can analyze regarding any piece that THAT is why determining why a piece is effective or not, to be able to gear your analysis to the salient issues. For me that is what's important in Nikolas' 1st step.