Learning Structure

I cannot write a structured piece of music. I can write a few good parts here and there, but it inevitably ends up as a jumbled mess of parts that don't really make sense together. Now, if I'm going to ask about structure, I can't just ask, "Hey, guys, how do I write structured pieces?" So I'm going to break my question down into a number of more pointed ones.

1. First of all, there's the question of logic. How can I tell if one section of a piece follows "logically" from another? Is this just something I have to do by feel, or is there more to it than that? Intuitively, I think it's a matter of similarities. But even when I make sure that each part of my piece is similar to the last part in some way, it still doesn't sound structured! So what am I doing wrong?
   
2. Is it better to try to learn large-scale "overall" structure first, or focus on structuring each part adequately, and then on joining them up?
   
3. Are transitions fundamentally important to structure? What makes a good transition? Should a transition be different from its adjoining parts, or similar to each one in some way, or what?
   
4. What are some good books on musical structure that can help me actually string together a coherent piece?
   
5. Finally: how did you learn about structure? I know it takes "hard work," but that doesn't tell me anything. What kind of work does it take? Do you consciously try to improve your structure, or are there other elements of composition that one should work on first?
   

Edit: On second thought, this might go better in the Advice and Technique subforum. Can a moderator move this?

1. Treading on muddy waters here, but I think I get the drift of your question. There are three main principles to organize a musical form: Repetition, contrast and variation. Quite self-explanatory.

2. You could have an idea for the overall form first, and start out with the big picture and then fragment it into smaller subsections which have their own form (say you want to do an overall ternary ABA form, and decide you want a second level of organization where A and B have their own form) The reverse approach is possible (generating form by addition of smaller units into a bigger form that becomes the building block for a bigger structure). Perhaps going back and forth from global to local is the best way. Try the different strategies. As long as you stick to formal protopypes, regular themes of say 4+4 bars, or 8+8 you should be on your way to coherent organization.

3. Transitions are not always required. Start out with plain sectional forms and worry about transitions when you are ready to tackle more complex and less rigid forms. Sectional forms have clear-cut sections which are fairly autonomous (except when they are harmonically open) and follow a very limited number of archetypes (strophic, binary, ternary, rondo, theme with variations...)

4. The Schoenberg text recommended by Elliott Carter is a good one.

5. I had a dedicated course about musical form in the Conservatory, but the fundamentals were shown in other subjects since first year, so when I took musical form, it was just a matter of consolidating, organizing, systematizing and going deeper into already very internalized concepts. The core ideas are actually very simple and easy to grasp. You will need to analyze scores, listen to models and play them (playing is the best way to internalize harmony and form, by the way). Form and harmony are linked, especially in tonal music. You need some prior knowledge of harmony in order to understand some organizational principles of the music, mostly dealing with cadential points (the endings of the different sections).

I have to agree with saratro here: start out writing simple chorales first. Then move to basic secitional forms like: Binary, Tenary, and Rounded Binary. Here, you can about sticking to create basic themes and forms and then you can progress onward to more complex forms that exist and then twist them around....

You're an idiot.

If you can not act like a mature adult here, then this is not the place for you. :)

. How preposterous would it be for example if a novelist were to resort to dictionaries and scholastic journals on the English language for inspiration?

A writer must learn grammar before he can write a novel, however. :closedeyes:

Not true, just a proper guide on grammar and style.

That may work, I guess. The best model is to score study the structures and know how they work.

Would you rather be a stickler to some stupid demeaning acronym like CROCK? (C for contrast, R for repetition...) I think it's an insult to the artistic mind and a misrepresentation of what music really is. A child who is taught to listen well should be able to grasp analysis better than a child who is taught the other way round, anyway. More importantly, the child who listens is having a much more musical experience. Play the music and let the beauty do its trick. Read them a poem they might enjoy. Not bog them down with an artificial roadmap and a laundry list of solecisms.

I disagree. An "artificial roadmap" is absolutely vital to understanding. With the aid of a standard model, we are able to assess and gain an understanding of the benefits and drawbacks of sticking to or deviating from the model. What you are advocating is just shooting in the dark. You can look at as many scores of classical sonatas as you like and you'll probably eventually notice a pattern that resembles sonata form, or you can read about sonata form which will allow you to look at scores with an improved understanding. More fool you if you won't take advantage of 200 years worth of academic leg work that has already been done for you but the rest of us are just going to read the book. Also, music isn't language; the analogies don't work.

Also, music isn't language; the analogies don't work.

I'm interested in this position. What makes you say that music isn't language? What are the differences?

[/size][/font][/color]

I'm interested in this position. What makes you say that music isn't language? What are the differences?

Translate this sentence into music.

Translate this sentence into music.

Okay, I don't think that all language is necessarily transposable. You can't necessarily translate English into a formal language like predicate logic, for example, but they're both still languages. Under a broad definition of language - one that includes, for example, formal languages like mathematics - music qualifies as a language. It has syntax and semantics, and that means it qualifies as a language.

For the record, some English sentences can arguably be translated into music. If you wanted, could you not translate the sentence, "I'm sad," into music? Or any other emotional expression, for that matter?

Right. You make studying music sound like it's a bad thing. Would you criticise a student painter for visually studying the masterpiece itself as opposed to reading solely the texts formed by opinions of critics and historians? "More fool" me, indeed. Come off that ad hominem bandwagon.

Where did I say you should "solely" read texts? I said you should read the theory so you can then look at the scores with an improved understanding instead of wasting time trying to discover everything for yourself. Looking at a Mozart sonata with a working knowledge of sonata form is going to be a lot more productive than looking at it with no prior knowledge.

Also, you need to understand what an ad hominem is. I essentially said that your position is a foolish one to hold, not that you are a fool, thus attacking your argument and not attacking you personally. Maybe you should come off the "using Latin phrases incorrectly in an attempt to sound intelligent" bandwagon.

Okay, I don't think that all language is necessarily transposable. You can't necessarily translate English into a formal language like predicate logic, for example, but they're both still languages. Under a broad definition of language - one that includes, for example, formal languages like mathematics - music qualifies as a language. It has syntax and semantics, and that means it qualifies as a language.

For the record, some English sentences can arguably be translated into music. If you wanted, could you not translate the sentence, "I'm sad," into music? Or any other emotional expression, for that matter?

The goal of language is to communicate information, ideas and concepts as clearly and succinctly as possible. Any attempt to do this with music leads to ambiguity. While you could write a passage of music that communicates "I'm sad" to you, it wouldn't necessarily communicate that to everybody else. Only when you told them that the passage meant "I'm sad" could they definitively understand the passage to mean "I'm sad". Music alone can't communicate anything definitively. It may have latent meanings of its own but these are always ambiguous until put into some kind of decisive context.

So, I suppose you could justifiably call music a language, but if you are going to do that you need to hold it to the criteria of what makes a good language. If you do that, you'll find that it comes out poorly in every aspect because you're trying to put a square peg in a round hole. Let music just be music; it's pretty good at it.

Right. How about reading a sonata as it is while studying it at the same time with a good teacher?

That would be ideal but not everyone here has access to a teacher so a book such as Schoenberg's can act as the guide for the student to then go and study the scores.

Nonetheless here are t reccomended texts you might use:

1. Structure and Style, Leon stien

2. Fundamentals of musical composition, Schoenberg

3. Musical composition, Brindle

4. Wallace Berry - Form in Music; Structural Functions in Music

I do know quite a few book buffs who can barely write a proper melody or harmonise a simple tune if they have to. In their head, they wrote an authentic 2 part invention, when in fact they really didn't. Conversely, I know some who constantly seek to perfect their understanding through music itself who are comparatively at ease with whatever they set out to do. Most composers I admire, dead or alive, belong to the latter category. Some expressed it fervently, eg. Balakirev, Delius and Richard Strauss (Paraphrase: "If great masters of the past (referring ironically to Beethoven and Brahms, the so called dry and uninspired architects) never bothered themselves with theorising, why should we?"). To composers who hear more than they draw graphs and pie charts, the bookish composer paradox may not necessarily apply. I think the book that best substitutes a teacher is one that forces you to look at the pieces as whole works of art. Not a book of dry and uninspired prescriptions that could otherwise ruin the first impression of music for a kid or beginner.

The book isn't supposed to inspire; it is supposed to be useful. If you want inspiration, go and stand on a mountaintop or something; if you want to understand the technical aspects of musical structure, read the book (while also studying scores, so you can see what you are reading about in its proper context). You're building up some ridiculous straw man who has his head stuck in books and is oblivious to everything else but theory and arguing against it when nobody here is holding that position.

For the record, Mozart, Haydn and Beethoven all used Gradus ad Parnassum to study counterpoint so if you're going to suggest that the so-called masters had an aversion to theoretical study, you would be wrong.

Aren't you a joy? Guess what is both inspiring and useful? If I want inspiration, I listen, play and analyse some scores. Stand atop a mountain? Is that something you would recommend doing in place of some hard listening, perhaps right after the compulsive flash card and regurgitation rituals? I think not. The straw man, or whatever you want to call it, is real. I know quite a few of their kind. Don't even have to look hard.

How do you analyse the scores? Are you going to tell me that you discovered it all for yourself, completely isolated from traditional theoretical models? I think not. Listening, playing and looking through scores is inspiring. I found it so inspiring that I went and read some rather "dry and uninspiring" books in order to better understand the workings of the inspiring music I was hearing, which led to me being even more inspired when I was able to listen to the works again with a full understanding.

I'm sure the straw man is real, but they aren't here. You're arguing against a position that I'm not holding! You seem to have turned my position which is: read books to provide context for score study, into some ridiculous academic parody along the lines of: don't look at scores at all and rely solely on theoretical books. Go and find the people who hold this position and argue it with them. You're wasting your time with me.

But compare the quality of music of average amateur composers today to those of the forgotten salon music dilettantes of yesterday. Let's be objective here and not be biased against any specific genre. You'd be hard pressed to find even a handful of improvisers like Richard Grayson in today's culture. You'd probably get a crowd who proclaim themselves masterly practioners of tonality, who reckon their musicianship proportional to the height of the stack of books they read, but lack the sensitivity to be able to even write a basic classical pastiche, something which the basest sensationalist opportunist charlatan of the 19th century could do effortlessly.

Your comparison is flawed because you don't know of any amateur composers from the 19th century. If their music was published, they weren't amateurs. Just like any era, there are some amazing musicians alive today; there are also some amazingly bad musicians alive today. You seem to be making some strange generalisations based on nostalgia for an era that you weren't even a part of. How the hell would you have any idea whether the basest sensationalist opportunist charlatan of the 19th century could effortlessly write basic classical pastiche or not? Perhaps you could give an example of such a piece of work, though I expect that you can't because it is a completely fallacious statement.

A lot of the answer to you question will depend on what kind of surface style you intend to write in. I don't think you can really seperate deep-level structure from what happens 'on top' of the music. Some composers deliberately choose a style in which the sections, be that a melody, a chord, a texture, are very clear (Poulenc, mid-baroque, some minimalists, film scores) whereas at the opposite extreme some write their music as what seems like an endless transition from start to finish (Wagner, Sibelius, contemporary spectralist and other schools of minimalism). So the terms of what constitutes 'logical' structure and development will depend on what your long-range goal of the piece is. As you seem to be just starting out learning to compose, I would guess that you're probably experimenting more with the former style, which is fine. You might want to plan out different sections for a piece based on forms you've already encountered (things like minuet and trio movements or variations) and then come up with related ideas for the surface material, which you can then work out what goes where.

You could alternatively take a single idea or process and follow it through. Some good examples of this appear in contemporary music, such as Arvo Pärt's Cantus in Memorium Benjamin Britten, which uses just a descending scale played at different speeds in the string orchestra. Or the minimalist works by Steve Reich and others where two versions of the same thing are played with one slightly faster than the other and the piece ends when they coincide again. A simple canon is another example of this.