As a music theory teacher, I think a lot about why musicians need theory. Certainly some of my students will forget everything I’ve taught them and yet somehow manage to do what they need to do anyway. Others will retain and put to use the techniques I have imparted in class in a very direct way. Probably the largest group will fall somewhere in between.
These thoughts have led to a few provisional decisions about the relationship between music theory and music practice, the latter being, as I see it now, the spectrum of activities from composition through performance to listening. As one moves along this spectrum, the need for theory becomes less urgent.
Consider composition: While an uncontrolled analytical impulse can bottleneck rather than free ideas, music theory can open the road for composers facing, in the midst of work, the inevitable question, “what next?” by asking, “what has been done?” The more analytical tools one brings to music already written or conceived, the greater one’s knowledge of this music will be, and the more likely ways of moving forward will come to light.
Performance: Here I’m thinking about those who play music composed by someone else from memory or a score. Analyzing a piece of music is a wonderful aid to memorization, certainly more reliable than just memorizing where your fingers go. But sometimes musicians develop a memory gifted enough not to need the assistance intellectual knowledge of a piece can provide. When memorization is not an issue, theoretical knowledge helps musicians contextualize the moment in which they find themselves. Sometimes, though, a gifted ear will discern musical context innately.
And listening: Certainly one can enjoy a symphony without knowing what sonata form is. But the more background one brings to any experience, the richer the experience is likely to be. Knowing more about the sorts of things that happen in compositions helps make listening a more active, engaging process. And yet, the consequences for flagging attention and ignorance on the part of listeners are small compared with the same coming from performers and composers.
I tell my students that music theory is ear training, and ear training is music theory. Even if students don’t find much direct application for the knowledge imparted in theory classes, students should take from these classes a better ear and, with it, the increased ability to appreciate any music they encounter.
Entries (RSS)
There are aspects of music theory that do indeed form the building blocks needed for any composer. Things like notation, counterpoint, solfege, etc. However, it’s been decades since I wrote anything with a key signature, and in all honesty, I don’t even remember them anymore. So that makes me wonder how critical some things are for composers to learn.
Still, it’s important to have a foundation, even if one ends up not really using it. What is perhaps most important, though, is exposure to a lot of different music. Not just classical, and not just Western. If you look at many musical traditions, what drives them isn’t so much harmony, but rhythm. With the possible exception of some of Scelsi’s music and some drone works, rhythm is at the heart of much of the music I tend to listen to, and I suspect this is true for most people as well.
Teaching music theory to composers is a blast, but I really love teaching performers best. When we analyze a piece, we always do it with an ear to how we might play it. So, I ask them, now that you have discovered x about this passage, will you play it like this, or like this? I’m not trying to tell them how to play a particular passage, but how to think through their interpretation.
Broad generalization: almost all instrumental teaching is focused on refining ones sound – intonation, color, phrasing. Here’s the analogy I ask my students to think about: if you go to a play, would you rather hear an actor with a beautiful voice who has no idea what the play is about, or an actor with a beautiful voice who understands every nuance of the character and plot? Music theory gives performers the tools to understand what they are saying to the audience – when applied properly, it makes the difference between an attractive performance and a compelling performance.
Interesting thread. I composed for several years prior to taking music theory. Last year I took Music Theory and Sight-singing I + II. This year I’ll be taking III+IV and then moving on to Counterpoint, Orchestration, etc. I’m an older student and go to school at night.
The most difficult thing for me is seeing/hearing how I actually apply what I’m learning. Am I composing better? Listening better? Not really sure. I probably am but it may not be due to theory. I do feel solfeg has helped develop my ear a bit. Made me listen more carefully. We haven’t discussed form much yet so can’t say how that would help. Also, I’d like to learn more theory about modern and non-Western music. Nevertheless, I recognize composing is a trade as well as an art. Theory may not help me come up with inspiring ideas–but when I do come up with inspiring ideas I hope I will have the craftmanship to make the most of those ideas.
Another question is: What theory do you need? I suppose most of us have been raised on a solid diet of harmony and counterpoint and are well-equiped to handle a broad range of classical music, and this is knowledge I’d not want to live without. But it does not necessarily make one well-equiped to play Feldman, Lucier or Antoine Beuger. (Or even Kurtag!). There is something curious about the fact that traditional harmony is still the core of the curriculum and that say, psychoacoustics is generally thought of as a somewhat marginal subject.
To me the key issue of teaching music theory to performers isn’t memorization but interpretation. I’m not a good enough performer to _completely_ appreciate how a theoretical understanding of a piece can aid interpretation, but I’ve seen it done. John Heiss, at NEC, teaches an amazing class called “Interpretive Analysis” which is all about learning to interpret music theory elements of a piece and using that knowledge to aid your performance interpretation — he also does a lot of coaching of NEC ensembles. The examples I can think of are pretty rudimentary, but if, for instance, you can identify a deceptive cadence as such you migh play it differently, leading up to it as if you’re coming to a real cadence and then, say, pulling back a little to emphasize the fact that you’ve just defied expectations. Again, that’s a pretty simple and obvious example, but that principle applies to much more subtle elements as well. If you’re good at identifying structure, you can play in ways that emphasize and clarify that structure, if you’re good at analyzing counterpoint you can emphasize the ways in which the notes are supposed to relate to each other, if you’re good at analyzing harmony you can emphasize the meanings of how the music moves through chord progressions, etc.
For a composer, I think the key is the ability to take the general principles of music theory, which you can only really learn and understand by deeply studying specific cases of musical practice, and applying those principles to the kind of music you’re writing. I don’t follow the rules of species counterpoint, but my study of the subject has made me better at writing my own counterpoint and better at voice-leading. Schoenberg makes one of my favorite examples of this phenomenon — I don’t recall what piece it was, but a few years ago a group of NEC grad students and I were listening to some Schoenberg piece and discussing it. Some people thought the chords were beautiful, but I actually found them pretty ugly; and yet even though I found them ugly I also thought that they _worked_. The reason was simple — Schoenberg was using approximations of the rules of species counterpoint to control his voice-leading, so even though I didn’t like the chords themselves they all followed logically from one to the other.
Lawrence — great analogy with acting. A not entirely related anecdote: Bernard Rands once insisted to us that no string players ever practiced pizzicato — which is why pizz always sounds bad. I haven’t exactly had bad pizz experiences, but I recall Rands’s story often anyway.
As for species counterpoint and so forth: I remember David Lewin justifying its study by claiming students would get to know a particular musical language (16th-century sacred music). Learning the components of one language gives you an idea of how you might build your own.
Beyond all practical reasons (which are important), is the fact that through the study of harmony, analysis, counterpoint, etc., the student knows and comes closer to music better, it gives him or her a far better understanding of it, which is indispensable for any serious musician.
Theory isn’t ear-training per se.
If we consider ear-training as teaching literacy, relating the heard language of music to the notation on the page, then theory is grammar. It certainly helps with the comprehension of what’s on the page, and is as much or more descriptive as prescriptive. But it’s not “the sounds”.
Contemporary (as distinguished from popular) music seemingly works to distinguish itself by the breaking and bending of the rules of classical theory: tonality, acceptable harmony, phrasing, cadences etc. So one reason to learn theory is to appreciate the evolution of music to this point, and what the logical way forward should be.
When I first began writing I was very grateful for the classical forms: reinforcement of tonality with harmonic progressions, the 8 and 16-bar phrase ending with one of several possible cadences, the changing of keys in an orderly manner. Form gives the beginning composer hope in that there is a framework or structure upon which one can build musical expression and the results will be acceptable, at least to historically sanctioned sensibilities. For this reason alone I believe music theory has great value – those who feel initially inadequate to the task of writing music will have a method to cling to and work with.
The great problem – and the the great freedom – of contemporary music is that it requires no such constraints. A good musician once told me that the music isn’t on the printed page – it is “out there”, just in front of the players and beyond the influence of theory. The great problem for the composer is how to make the symbols on the page that will make great music in the ears of the listeners.
Consider composition: While an uncontrolled analytical impulse can bottleneck rather than free ideas, music theory can open the road for composers facing, in the midst of work, the inevitable question, “what next?†by asking, “what has been done?â€
No, no, no! That’s music history you’re talking about. Music theory is the study of the abstract processes in terms of which music — in general — is understood. As such, it does not specifically concern what any particular composer actually did at any particular time. What it does concern is how a listener thinks of a composer’s notes; it should be very easy to see why this kind of study is an absolutely fundamental part of the training of every musician — composer, performer, or listener.
Indeed, I would go so far as to say that, by its very nature, music theory is precisely what “musical training” consists of — with things like music history or instrumental skill being optional (if nonetheless very useful) add-ons.
Consider composition: While an uncontrolled analytical impulse can bottleneck rather than free ideas, music theory can open the road for composers facing, in the midst of work, the inevitable question, “what next?†by asking, “what has been done?â€
No, no, no! That’s music history you’re talking about. Music theory is the study of the abstract processes in terms of which music — in general — is understood. As such, it does not specifically concern what any particular composer actually did at any particular time. What it does concern is how a listener thinks of a composer’s notes; it should be very easy to see why this kind of study is an absolutely fundamental part of the training of every musician — composer, performer, or listener.
Indeed, I would go so far as to say that, by its very nature, music theory is precisely what “musical training” consists of — with things like music history or instrumental skill being optional (if nonetheless very useful) add-ons.
(I’m not sure why my previous comment posted twice. Hopefully that won’t happen with this one.)
To forestall any misunderstanding, let me add that when I describe music theory as studying the processes in terms of which music is understood, I am taking it for granted that this necessitates actual experiential familiarity with the sounds of music (i.e., ear training, which I would consider the most rudimentary–and indispensable– form of music theory).
How can traditional music theory even answer such basic interpreter-centric questions as (I riff):
1. Is this part a climax or is it an introduction of a new material chain? How do I tell the difference? How do I play the difference?
2. These chords seem to point to a section I played with manner X. Should I play them again with manner X or should I play it differently to enhance the contrast? How do I know what would be better?
3. How do I know when to play up contrasts of materials that are alike and how do I know when to downplay contrasts in this piece?
4. How do I analyze this piece to find the techniques and materials that will allow me to perform this piece in a manner that will demonstrate the interesting connections between textural or melodic materials?
5. How do I know when it’s OK to stray from the music directions in the score? Can the music tell me this? Is the way the music is notated giving me subtle clues as to how to perform it beyond the obvious?
So much bad contemporary music performances are because the musicians just don’t GET the piece. Or they play up one possible interpretation without taking into account the other possible interpretations which could better show the real complexity of the intersections between materials and textures.
I think we really need an entirely new musical theory that would be for interpreters and not composers. This theory would also be very cool for composers. I would love to hear a great performer talk about how she decided to slow down at this bit in the piece, etc and change her playing style to enhance the development of blah blah blah… I would suggest there’d be a lot of collective listening and discussion during listening.
Truly essential music gestalts defining types of musical material/textural similarities, types of musical formal directionalities and mis-directions are still not well expressed in the music theory literature. Also, we are very bad at talking about and analyzing musical textures and textural transitions. But I digress… ;)
James is right – we are, to an extent, conflating theory and analysis, and they should be two different things. Theory gives us tools for analysis, in the way that studying music notation gives a composer tools for writing music. The confusion comes because, as with notation, it is impossible to understand theory without analyzing specific pieces.
Good analysis, though – at least the kind I was trying to describe above – uses theory, history, acoustics, and anything else that might be relevant in order to understand how a composition works. It doesn’t provide the one answer to how a given piece should be played, at least not very often. More often it provides a strategy for obtaining a more coherent performance.
James — what I wrote about drew on my own experiences as a composer. First I establish a broad outline. Then I start writing music. But even with the outline, I still sometimes don’t know what should come next. In several pieces, I have found the way forward by going backwards into the music I’ve written and looking at what I had done. Inevitably we don’t know everything about what we have written. Asking questions like “what defines this music rhythmically/timbrally/dynamically, etc.” (questions one grows accustomed to through studying music theory) has helped me discover ways by which I may, say, establish contrast or just grow the music. The more one is used the hunting down the answers to such questions, the more likely one can find many different ways forward through the creative process.
How about this: if you were to dance (about) architecture, the relationship between the dance and architecture would be analogous to that between music theory and music. Nothing wrong with music theory, and certainly nothing wrong with dancing (about) architecture, but don’t expect your buildings to stay up better ;-)
tig
Jeff H.:
The questions you raise are exactly the sorts of questions that Heinrich Schenker was interested in, and which he attempted to address in his writings. Whatever one thinks about the particulars of his own analyses, it should be clear that he was headed in the right direction. For the life of me, I can’t understand why there have been so few serious efforts to discuss 20th-century music in the same way.
David S.:
It seems I misunderstood what you wrote. I thought you were talking about “what has been done” in the sense of other music by other composers, which, in the context, would have meant you were espousing a Pistonian view of the nature of music theory (one which would hold that studying music theory consists of acquiring knowledge of the “practice of composers”) . Your most recent comment, however, makes it clear that your notion of “music theory” may actually be much closer to mine than I had feared. So there may not be any disagreement after all.
James, I agree that Schenker’s conversation with music was a very good thing. My graduate courses in Schenkerian analysis were a profound experience for me but in post-tonal music (whatever we call it) I can’t see how an analysis which looks for skeletal structures inherited from common practice technique can inform except coincidentally. When I was studying Schenkerian analysis though, I had a great professor and he would often cite the inadequacies of ALL music theories to allow for a real analysis of music the way it is heard. He cited Tovey (general approach in discussion) and Knuth? as theorists that allowed for a discussion about way music can be broken down and talked about in a humanistic way, not in a notational way. A way that would help discussions with interpreters and their needs (as I discussed previously). For one thing, how can we even look at music without discussing its performance? It doesn’t exist! Music is not notes on a page. ;)
Modern theory with its counting, chord-naming and simplistic pattern recognition has nothing to do with how music is composed (by good composers) or played. These useless approaches today have festered into the various ways music is now perceived and conversed about. And I believe it has festered into how music is written – bad music… and this is tragic… but don’t get me started… ;)
Jeff H.:
but in post-tonal music (whatever we call it) I can’t see how an analysis which looks for skeletal structures inherited from common practice technique
Aha! Precisely my point. Theoretical structures are not “inherited from common practice technique”; they are inherent in the psychology of a listener. This is exactly the mistake I (wrongly) thought David was making: confusing theory (or analysis) with history. Compare and contrast:
Theory/Analysis: “This B is [i.e. sounds like] a leading tone to a C”
History: “In 18th-century music, a leading-tone B normally resolves immediately to C”.
See the difference? Notice that there is only a one-way dependency relation: the historical proposition depends on a theoretical concept (“leading tone”) in order to make sense; but a theoretical concept such as “leading tone” refers to the workings of the human mind, and as such is conceptually independent of the facts of musical history.
When I talk about applying Schenker’s ideas to 20th-century music (or any music, for that matter), what I basically mean is taking a work (or a passage) and diagramming (or “graphing”) it — much like one does with sentences in linguistics. The point is to explain, as specifically as possible, the hierarchical position and conceptual function of every note in the piece — identifying which notes are leading tones to which other notes, etc. (One does not necessarily have to literally draw a graph in order to accomplish this, but it can often be helpful.) The extent to which the notes in a piece actually behave like those in, say, an 18th-century work is a historical question that one may ask after performing the task of analysis.
James wrote: Theoretical structures are not “inherited from common practice techniqueâ€; they are inherent in the psychology of a listener. This is exactly the mistake I (wrongly) thought David was making: confusing theory (or analysis) with history. Compare and contrast:
Theory/Analysis: “This B is [i.e. sounds like] a leading tone to a Câ€
History: “In 18th-century music, a leading-tone B normally resolves immediately to Câ€.
The distinction you’re making between the two is clear and part of a sound & needed approach. But the “T/A” example is still freighted with historical assumption; i.e., “leading tone”, solfege letter-names, and “is” (“sounds like” doesn’t really fix that, either). These are still up in the tree, not the roots of the matter. Real theory should be much more universal than this.
The distinction you’re making between the two is clear and part of a sound & needed approach. But the “T/A†example is still freighted with historical assumption; i.e., “leading toneâ€, solfege letter-names, and “is†(â€sounds like†doesn’t really fix that, either). These are still up in the tree, not the roots of the matter. Real theory should be much more universal than this.
Being “up in the tree” is different from being “freighted with historical assumption”. The concept of “leading tone” may not be “ultimate”, in that it may be reducible to other, more primitive theoretical concepts. But, at least as I’m using it here, it is still a theoretical concept itself.
It may be helpful to consider an analogy. In the early universe, there was no water around, since water requires the bonding (and therefore the existence) of hydrogen and oxygen. But that doesn’t mean that the idea of “water” is “freighted with historical assumption”. It’s still a concept of physics/chemistry, one that is, indeed, reducible to the more primitive concepts of “hydrogen”, “oxygen”, and “chemical bonding”; and there was a time in cosmic history when those concepts were relevant to the description of physical processes, while “water” was not.
Similarly, there may be types of music where concepts like “leading tone” (or, indeed, “B”) are not relevant to analysis (what music do you have in mind?). But that doesn’t mean that such concepts aren’t legitimately theoretical in nature.
But the leading tone doesn’t seem to arise from any fundamental physical principle. It doesn’t take much listening outside the Western European Cultural Tradition to find all kinds of music where the concept doesn’t exist. And even within the “WECT”, by the time you hit Schoenberg and Varese the concept gets pretty useless, and by now, with Cage, Xenakis, Feldman, early Reich, Coates, Dhomont, Frey, etc., we have a plethora of valid approaches that don’t consider leading tones.
The leading tone concept is a useful part of a theory, but a theory of a fairly particular cultural practice. What gets most of us grumblers, I think, is that most places teach Practice, but call it Theory. It’s like teaching clothing design with the fundamental forms given as shirt, pants, and skirt.
But the leading tone doesn’t seem to arise from any fundamental physical principle
Why should it? Is there any reason to think music theory should be directly reducible to physics?
And even within the “WECTâ€, by the time you hit Schoenberg and Varese the concept gets pretty useless
I don’t happen to agree. On what do you base this conclusion?
What gets most of us grumblers, I think, is that most places teach Practice, but call it Theory.
Even worse: under the name of theory, they teach practice — using a bad theory to do so!
But the leading tone doesn’t seem to arise from any fundamental physical principle
Why should it? Is there any reason to think music theory should be directly reducible to physics?
Not at all, though there are some aspects of both acoustics and psychoacoustics that should be taught right from the begining. But at best, the rest is akin to anthropology; primarily descriptive analysis of relative human systems. Given that, \\\”WECT\\\” should show up quite a bit later, with much less — or more integrated — focus.
And even within the “WECTâ€, by the time you hit Schoenberg and Varese the concept gets pretty useless
I don’t happen to agree. On what do you base this conclusion?
That it\\\’s not there for me (though I can understand a Schoenberg possibly thinking that they were somehow continuing the practice in an extended form). Traditional theory won\\\’t cut it, but if Schenker can convincingly chart the course through Arcana (not to mention Ionisation!), I\\\’m all ears. And if that works, it\\\’s on to Coptic Light!
What gets most of us grumblers, I think, is that most places teach Practice, but call it Theory.
Even worse: under the name of theory, they teach practice — using a bad theory to do so!
Now there\\\’s where we agree, James! (Nice blog, by the way. Keep it up.)