Lego Blocks - how I do improvisation

[Mark Kleinhaut]

So rather than get stuck into that, I’d say if you mean that improvising music is primarily an intuitive act I’d tend to agree. There’s a high level guiding decision making going on, but you can’t be thinking about on the note by note level


Sent from my iPhone using Tapatalk

we’re on the same page said otherwise, what goes in slowly comes out very fast!

[Christian Miller]

we’re on the same page said otherwise, what goes in slowly comes out very fast!

It goes in very slowly for me haha

[Tal_175]

What doesn’t help with answering that question is people seem to have different definitions of ‘theory’ anyway.

I’d go as far to say that nothing I have discussed qualifies as theory. But that’s quite a minority position. I tend to define theory as something which aims at an explanation of how music works, not simply categorise objects or schemata.

I'd go so far as to say that even the same people have different definitions of theory in different posts. Ha ha

Music “theory” is generally a way of categorising ahem… schemata … generally because you want to talk about them with others.

I remember in the past you said you have a different definition of theory. But I have to say, sometimes I have difficulty keeping track of which definition you have in mind when I read your posts. The distinction you are making I think is valid, but I wish we had different terms for these notions.

[Tal_175]

The endless debates between theory vs ear players isn’t really on topic with what I was saying about Lego improvisation except I wanted to emphasize that the Lego analogies were in no way intended to take sides in such debate. How we acquire knowledge, organize and access such knowledge, and whether analysis is part of that process is a different subject entirely. I realize it’s nearly impossible for this to be avoided, and maybe it shouldn’t be, but I’d like to think there’s general consensus that theory has little or no role when engaged in the actual act of improvisation. Is that not so?

That's true. I usually rely on the use of theory and other "conscious" processes during practice but I agree with what you said earlier that even that's a questionable use of theory. Even during woodshedding one should put exploration and ear based work to the forefront.

Ear based and concept based practices encourage very different mental attitudes. Even though a concept based practice is thought to also train ears by exposure, I think there is something inefficient about it. A lot of the effort goes to getting good at visualization and muscle memory, rather than ears.

One area that's hard not to rely on a conscious process is playing chords. I don't know what would be a purely "ear based" method of developing multi part harmony skills on guitar. Is your process for accessing chords also purely ear based (even grips).

[Mark Kleinhaut]

That's true. I usually rely on the use of theory and other "conscious" processes during practice but I agree with what you said earlier that even that's a questionable use of theory. Even during woodshedding one should put exploration and ear based work to the forefront. Ear based and concept based practices encourage very different mental attitudes. Even though concept based practice is thought to also train ears by exposure, I think there is something inefficient about it. A lot of the effort goes to getting good a visualization and muscle memory, rather than ears.

One area that's hard not to rely on a conscious process is playing chords. I don't know what would be a purely "ear based" method of developing multi part harmony skills on guitar. Is your process for accessing chords also purely ear based (even grips).

For me, the polyphonic stuff that goes between the grips is more interesting and creates a texture of moving voices that’s far more orchestral than just playing chords. I do that without any thought for naming the notes although I absolutely know the sounds from a practical sense. My only thoughts about grips I recognize as such are just to locate myself on the fretboard, markers to keep track of the form (for tunes) functioning as waypoints, and only exact name I’ll think about is the root.

[Mick-7]

I’d like to think there’s general consensus that theory has little or no role when engaged in the actual act of improvisation. Is that not so?

No, if I'm given the chart of a tune I've never played or even heard before and will be improvising on it, theoretical analysis can make the difference between playing it well or poorly, especially if it has complicated or unusual chord changes.

[Mark Kleinhaut]

No, if I'm given the chart of a tune I've never played or even heard before and will be improvising on it, theoretical analysis can make the difference between playing it well or poorly.

I see your point. I guess it would depend on difficulty level for me as well as if there is another chordal instrument and the harmonic clarity of the bassist. I can navigate most charts by ear once I’ve been through the form once. I’ll look at the changes while soloing, play a few myself if needed, but would rarely need to think about scales or even notes.

[Rob MacKillop]

No one has so far mentioned Harmolodics … I’ll go into hiding now…

[John A.]

What doesn’t help with answering that question is people seem to have different definitions of ‘theory’ anyway.

I’d go as far to say that nothing I have discussed qualifies as theory. But that’s quite a minority position. I tend to define theory as something which aims at an explanation of how music works, not simply categorise objects or schemata.

So, by my understanding, recognising a twelve bar blues progression is not theory, but discussing why and how the twelve bar blues works the way it does is theory (and overlaps with the field of academic musicology.)

This rules out a lot of things people tend to classify as “music theory” in everyday usage, including a lot of the ABRSM Music Theory syllabus we teach to kids in the UK lol. So it’s not mainstream use, I’ll admit.

At the other end people seem to argue that any music at all is theory because it contains theoretically identifiable musical objects like chords and scales and so on.

This seems an unhelpful definition because it is so wide as to overlap with “music”.

So rather than get stuck into that, I’d say if you mean that improvising music is primarily an intuitive act I’d tend to agree. There’s a high level guiding decision making going on, but you can’t be thinking about on the note by note level.

You have to internalise whatever you are learning to the level of embodied or tacit knowledge. Which is obvious from experience, no?

Sent from my iPhone using Tapatalk

I agree with you on use of the word "theory" in musical instruction and conversation. The philosophical idea of praxis (the union of theory and action) strikes me as a better term. When we wear our musician/improvisor/composer hats, most of what call "theory" is a mishmosh of "why did that work?" and "how can I do it again?" OTOH, for those seeking doctoral degrees in music, "theory" mostly means "they did not realize this is why they did that or that it's the same as what those other people did."

I also agree that the "lego" concept probably does underline how most people improvise. The human mind is a module-and-pattern-seeking-and-repeating engine. It's at such a fundamental level of cognition, communication, and behavior that we mostly aren't aware of it (unless we're consciously training ourselves to do something pattern-and-module-based, such as music or sports or computer programming).

In jazz (and I would say blues as well, and probably any other form that involves improvisation and stresses the individuality of the performer), a person's style is the blocks of note choices, phrasings, and articulations one tends to choose and how one strings them together (along with the the timbre one's physiology brings out of the instrument).

[Christian Miller]

I'd go so far as to say that even the same people have different definitions of theory in different posts. Ha ha


I remember in the past you said you have a different definition of theory. But I have to say, sometimes I have difficulty keeping track of which definition you have in mind when I read your posts. The distinction you are making I think is valid, but I wish we had different terms for these notions.

Note the use of the scare quotes in the post you quoted. I’m not sure if I’ve used those consistently or not, but I try to. I’m implying “I don’t think this the categorisation of schemata is actually music theory, and I think it’s problematic to do so, but I am acknowledging others do use the term to mean that, so let’s call it that for the sake of this argument.”

Which is quite high context, tbf. But people do use scare quotes in this way, so it’s good to look out for it. John A uses them similarly in his post.

I don’t think the definition I’ve cited above has changed much in the past few years. The issue comes from the fact that most here (or elsewhere) don’t subscribe to my narrow definition of theory. I’m ok with it. So we can discuss “theory”.

Sent from my iPhone using Tapatalk

[rpjazzguitar]

In the theory vs ear argument, there is a blind alley where the ear-ists insist they're playing by ear and the theorists insist that they're applying theory, even if they don't think of it that way. The language we use isn't up to the task. "Schemata" seems to be going down a similar blind alley.

One advantage of "schemata" is that it seems to include the rhythmic content of the line in a way "theory" usually does not, although, since I can't define "theory" precisely enough, I have no way to be certain.

I don't subscribe to the argument that players tend to do this a certain way. I find myself doing different things at different parts of the same tune, different things if I know the tune, different things depending on the complexity of the harmony, etc etc etc. If somebody applies one word to describe all of that, it will be insufficiently specific.

[starjasmine]

No, if I'm given the chart of a tune I've never played or even heard before and will be improvising on it, theoretical analysis can make the difference between playing it well or poorly, especially if it has complicated or unusual chord changes.

Taking that a step further, if I'm NOT given a chart of a tune I don't know, I'm using relative pitch and theory to understand what I'm hearing so that I can learn the tune on the fly quickly. This skill also underpins the musical conversation that happens on the bandstand: the soloist might hear a sonority in the comping that he reflects in his line or the rhythm section might hear something in the solo that they adapt to in their accompaniment.

[rpjazzguitar]

Taking that a step further, if I'm NOT given a chart of a tune I don't know, I'm using relative pitch and theory to understand what I'm hearing so that I can learn the tune on the fly quickly. This skill also underpins the musical conversation that happens on the bandstand: the soloist might hear a sonority in the comping that he reflects in his line or the rhythm section might hear something in the solo that they adapt to in their accompaniment.

Some players have the ability to hear somebody play a wrong chord. And, they know both what it was and what it should have been.

What theory helps you identify a mistake?

Most of us can tell when a chord is wrong if we know the tune. Some of us have bigger ears than that. I don't think theory has much to do with it. Of course, some may claim that, somehow, they're using theory, but I don't think the word "theory" stretches that far.

[starjasmine]

>>What theory helps you identify a mistake?

I would define "a mistake" as "something that doesn't fit" - as opposed to a substitution that works but is not what is customarily played in a particular spot. You don't need to know the tune to tell that someone has hit a clam.

The theory (or harmony study ... not going down that semantic rathole...) that helps you identify a mistake is the mental-connected-to-aural model that allows you to

1. identify the difference between the expected note and the one you actually heard
2. map that note to the progression that is going on (because you can hear that - again, you don't need to know the tune to know that you are expecting I in a ii V I and someone hits a b3)
3. use your theoretical knowledge to map that to standard notation: "I in this key is C and I heard b3 which is Eb instead of hearing M3, which is E."

All of this can happen in real time if your ears are connected to your mental model of harmony. You hear that b3 and instantly know that b3 is Eb in the key of C that was already established.

>>Some players have the ability to hear somebody play a wrong chord. And, they know both what it was and what it should have been.
That skill is not uncommon. At uni, most of the staff that would teach any sort of performance-related class could and did do that. I've also had the piano player call me out on chord subs he didn't like in a big band. Hit a clam in a pit band gig? Yes, the director will catch it and ask you about it.

[Christian Miller]

Re: mistakes. That’s usually what we call an unintended element, no? “I didn’t mean to play that note”?

Playing a composition it’s more or less a cut and dried thing. That note was wrong.

Improvising however, different thing.

Record yourself on a gig, listen to it right after and then listen back a year later. Totally different experience.

What you will have forgotten is your intention, and you will hear the music as it IS, not in comparison with the idea of it in your mind, and your memory of the process of playing it. Intention is all consuming for the musician in the act of creation ….

… and utterly irrelevant to the audience because they are not mind readers.

Sent from my iPhone using Tapatalk

[Christian Miller]

Some players have the ability to hear somebody play a wrong chord. And, they know both what it was and what it should have been.

What theory helps you identify a mistake?

Most of us can tell when a chord is wrong if we know the tune. Some of us have bigger ears than that. I don't think theory has much to do with it. Of course, some may claim that, somehow, they're using theory, but I don't think the word "theory" stretches that far.

There's no point knowing what a #11 is unless you can sing and hear it cold.

[starjasmine]

… and utterly irrelevant to the audience because they are not mind readers.

Sent from my iPhone using Tapatalk

But they know whether it "sounds right" or not. That can be subjective: a "right" note in the style of Monk might not be "right" at your country gig :-)

Rule # 1 of the bandstand: don't TELL them its wrong by making a face or giving the stink-eye to another player :-)

I do believe that we are now officially off-topic...:-) Sorry. Back to your regularly scheduled JGO discussion, which is already in progress.

[rpjazzguitar]

You're reading a chart, along with the rest of the band. The pianist plays the chord in the chart, the guitarist plays the wrong chord. (This is not a note in the middle of an improv which is subject to taste -- it's a wrong chord).

Everybody hears that there's a clash. It's not an artistic choice - unless your art is playing clams.

Some musicians will know what chord the guitarist played in error. Some will know the chord name. Some will know the individual notes.

Are they using theory? Not the way I understand the word "theory". But, as I wrote earlier, that word doesn't have a specific enough definition for some discussions.

Maybe it's a little like looking at a drawing done with a box of crayons and knowing the names of each color. For the 8 crayon box, most of us can readily do that. For the 256 crayon box, I suspect that most of us could learn it.

With music, this kind of identification comes much more easily to some than others -- and many people, including players, will never be able to do it, no matter how much they try, or, so I believe. It's something about the brain's wiring, not how much theory you have pumped in.

Now, maybe you could construct a test where you let the examinee take as much time as they need to try to figure out the identity of the incorrect chord. I can imagine that some theory might prove useful in that situation. But, the naturals seem to know instantly.

The very best musicians I know personally can do this.

[starjasmine]

Maybe it's a little like looking at a drawing done with a box of crayons and knowing the names of each color. For the 8 crayon box, most of us can readily do that. For the 256 crayon box, I suspect that most of us could learn it.

• There are 12 crayons in the box: the notes of the chromatic scale.
• Four combinations are prevalent: the triads that make maj, mi, aug, and dim chords.
• Beyond that, you can add extensions (diatonic 7, 9, 11, 13) or alterations (such as flatting the fifth or any of the extensions)
• That's an inclusive "or" you can extend AND alter; for example, G7b5b13

Not nearly as hard as 256 crayons... which I think might be akin to microtonal music :-)

You learn to identify those basic colors and what extensions (7th, 9th, 11th, 13th) sound like. As an experienced player, you can probably already do this intuitively (e.g., you know a 13th chord when you hear one) but just don't have the formal harmony framework to fit it into. (At least, I gather from your posts that you can read and play but haven't studied harmony.)

Now, maybe you could construct a test where you let the examinee take as much time as they need to try to figure out the identity of the incorrect chord. I can imagine that some theory might prove useful in that situation. But, the naturals seem to know instantly.

The very best musicians I know personally can do this.

This skill can be learned. That's what formal ear training in conjunction with study of harmony does. Yes, people with "perfect pitch" do this a whole different way, and (in most cases) do it better, but relative pitch combined with harmonic analysis on the fly is absolutely learnable. It takes years of serious effort - and a structured pedagogy with coaching helps a LOT, which is what the typical university music program provides - but it absolutely can be done.

The big band example I cited was when I was subbing for the usual guy and reading a chart. I probably played a maj7#11 intentionally in place of a triad, and the pianist did not play the #11. My on-the-fly substitution clashed with his (I think he played a straight ma7 there). There was no way either of us could have known in advance what the other was going to do. And a lot of people wouldn't have even called it a clash - but he noticed it and brought it up when the director stopped us for other reasons. I took his note and played a straight ma7 there from then on. I was pleased that he actually was paying attention to what I was doing.

The test you described is very similar to the transcription tests given in formal harmony classes, except that the number of chances you get to identify the chord is not unlimited. On a typical graded transcription exam, you get three passes to transcribe the piece in standard notation. There are various strategies for accomplishing this, but a common one is to try to grab the melody and bass on the first pass, fill in the harmony on the second pass, and fix up anything missing or wrong on the third pass.

[Aiq]

As for the mistake (unintentional) in real time there may exist examples when listening to the recording later one might say “Say, that was OK.”

[Mark Kleinhaut]

Aren’t ear training and theory completely different subjects? If you play me a chord and I say that’s a G7b5 it’s a simple statement of fact. True or false, nothing’s theoretical about it. When I studied with Ted Dunbar at Rutgers many many moons ago, he would turn his back and play all kinds of voicings and we would name them. Aural flash cards. We did this all the time. Layer he would break out his books, Theory of Tonal Convergence - a whole different boogie. It occurs to me these are still all a bunch of Legos. Just different shapes and colors.

[rpjazzguitar]

• There are 12 crayons in the box: the notes of the chromatic scale.
• Four combinations are prevalent: the triads that make maj, mi, aug, and dim chords.
• Beyond that, you can add extensions (diatonic 7, 9, 11, 13) or alterations (such as flatting the fifth or any of the extensions)
• That's an inclusive "or" you can extend AND alter; for example, G7b5b13

Not nearly as hard as 256 crayons... which I think might be akin to microtonal music :-)

You learn to identify those basic colors and what extensions (7th, 9th, 11th, 13th) sound like. As an experienced player, you can probably already do this intuitively (e.g., you know a 13th chord when you hear one) but just don't have the formal harmony framework to fit it into. (At least, I gather from your posts that you can read and play but haven't studied harmony.)

That can be challenging enough. But, on the bandstand, the bass isn't always playing the root and the key might suddenly change, and some minor and major type chords have the same notes, some dominants and different minor chords have most of the same notes, among other things that happen and make this more difficult.

To the poster who wondered if I've studied harmony. The answer is, I don't know. I've read Jazz Theory and Nettles and Graf, among others. But, a ground-up study of harmony integrated with ear training, no. I don't know what I don't know.

I do recall that in High School no guitarist I knew had any theory at all and some were much better than others at taking stuff off records. There is such a thing as a gift. Hard work can certainly produce improvement, but some people just have it.

[starjasmine]

Aren’t ear training and theory completely different subjects? If you play me a chord and I say that’s a G7b5 it’s a simple statement of fact. True or false, nothing’s theoretical about it. When I studied with Ted Dunbar at Rutgers many many moons ago, he would turn his back and play all kinds of voicings and we would name them. Aural flash cards. We did this all the time. Layer he would break out his books, Theory of Tonal Convergence - a whole different boogie. It occurs to me these are still all a bunch of Legos. Just different shapes and colors.

They are separate skills that, used together, contribute to overall musicianship. WRT your example, how do you know it's G7b5 and not, say, Ab7b5? Unless you have perfect pitch, someone had to give you a reference pitch. That reference pitch could be G or it could be some other note. You hear the interval between your reference pitch and the root of the chord you want to identify, and then (in my case, at least) use an understanding of harmony to construct the chord and name it as shorthand (G7b5) or as individual notes (G B Db F) and say "Hey guitarist, you played a Db in bar 27; could you please make that a D natural?"

[starjasmine]

That can be challenging enough. But, on the bandstand, the bass isn't always playing the root and the key might suddenly change, and some minor and major type chords have the same notes, some dominants and different minor chords have most of the same notes, among other things that happen and make this more difficult.

All of which you can hear if you have studied those things. Yes, some things pose challenges, but one can hear a LOT.

To the poster who wondered if I've studied harmony. The answer is, I don't know. I've read Jazz Theory and Nettles and Graf, among others. But, a ground-up study of harmony integrated with ear training, no. I don't know what I don't know.

That was me, and I didn't mean to offend. Sorry if I did. You have hit the nail on the head, though: integrating the study of harmony with ear training is the key. That study produces the following skills:
- If you hear it, you know what it is and you can play it or write it.
- If you see it written in standard notation, you can audiate it (hear it in your "mind's ear" or sight-sing it) and play it on your instrument.
- If you play it, you know what will come out of your instrument BEFORE you play it and you can write it down in standard notation.

As I've mentioned, it takes work to acquire this skill set. But many people do acquire it. It is worth noting that acquiring this skill is much easier if you are taught instead of trying to do it on your own.

I do recall that in High School no guitarist I knew had any theory at all and some were much better than others at taking stuff off records. There is such a thing as a gift. Hard work can certainly produce improvement, but some people just have it.

I won't disagree that we all have varying levels of talent at different things. But hard work and practice can produce strong results. There are some at both ends of the bell curve. But the majority of players are in the middle: they can learn these skills to varying degrees.

[rpjazzguitar]

All of which you can hear if you have studied those things. Yes, some things pose challenges, but one can hear a LOT.



That was me, and I didn't mean to offend. Sorry if I did. You have hit the nail on the head, though: integrating the study of harmony with ear training is the key. That study produces the following skills:
- If you hear it, you know what it is and you can play it or write it.
- If you see it written in standard notation, you can audiate it (hear it in your "mind's ear" or sight-sing it) and play it on your instrument.
- If you play it, you know what will come out of your instrument BEFORE you play it and you can write it down in standard notation.

As I've mentioned, it takes work to acquire this skill set. But many people do acquire it.


I won't disagree that we all have varying levels of talent at different things. But hard work and practice can produce strong results. There are some at both ends of the bell curve. But the majority of players can learn these skills to some degree.

No offense taken at all. I'm a fan of your posts.

If you have ears, you can pick up the language. If you have to develop your ears, it's going to be some work. My guess is you don't catch up to the person who always had big ears. I don't know how to quantify the potential amount of improvement.