finnhiggins
03-30-2007, 01:22 AM
BIG POST AHEAD WARNING!
This started out as a reply to the "Matrix" thread about Thomas Lang's concept on his new DVD. I think it's a badly-designed exercise, and I explained why over in that thread a little. But rather than just bashing, I think I should make this a bit more general and talk about exercise design: Why you should be doing it, and some things that I think objectively make practice exercises good or bad.
First up, what do I mean by exercise design: I mean literally sitting down and writing exercises out for yourself that you're going to practice. Why do it? There are lots of drum books on the market, after all.
My suggestion is that every great drummer does their own exercise design. That's why they're able to put their own books and teaching method DVDs out on the market: They just take the stuff they design for themselves and write it up tidily to sell. Now, that doesn't mean you shouldn't buy DVDs or books: it is great to see how other people have designed their exercises and take inspiration from that to do your own. But I don't think it's healthy to just practice religiously out of other people's books.
For my purposes I separate drum books into three types:
1) Personal books. This would be something like Bill Bruford's book, David Garibaldi's book, things like "West African Rhythms for Drumset", many of the marching/snare solo books etc etc. These are books that document one drummer's personal journey as a player and write up a lot of the stuff they've developed for themselves musically. Some are more about transcriptions - like Bruford or Garibaldi - some are about stylistic material like the Royal Hartigan african rhythms book.
2) Technical books. These would be things like Stick Control, Accents & Rebounds, Master Studies, Gary Chaffee's "Patterns" series or Gavin Harrison's two books on rhythm. These aren't so much about music as they are about developing facility, and for the most part they're pretty much definitive works - once a particular book does a certain aspect of technique/facility stuff right then it is pretty rare for another book to come along that does the same thing better. We're still using Stick Control, after all.
3) Material books. I'd tend to see these as being things like The New Breed or Syncopation which give you reading material that can be used to design your own exercises, although there are plenty of existing interpretations that can be used with either.
Now, of those books I tend to think that everybody should have the key books from #2 and #3. You need to go through Stick Control to get some control of the sticks. You need to have Syncopation handy so that you can use it as a tool in designing your own exercises. But the #1 books are, for me, totally optional and entirely up to your taste. If you get a book of that description and don't really think the results are hip then ignore it. My suggestion is that we should ALL be writing our own personal book, and that other people's books are just there for you to take inspiration from.
So what makes a good exercise? I reckon a good exercise is something that meets the following goals:
1) It works on something that you are currently unable to do. If you're just playing stuff that you can play easily then it's playing, not practice. Get gigs if you want to do this, they'll make you focus harder because there's an audience watching.
2) When played correctly it sounds musical to your ears. What you practice will come out in your playing, and if your exercises sound unmusical to you then you'll develop playing habits that you dislike and which make you unhappy.
3) It can be directly applied to a style of music that you're enthusiastic about listening to and playing. Otherwise... why bother?
If there are exercises written by other people that fit those criteria then by all means practice them. But if you want to sound like you, odds are you're going to need to start writing your own.
Now, I just want to get back to the Lang stuff for a minute. Lang's "Matrix" system is a particular class of exercise that's very common in music education: it's a combinations exercise. You take a certain number of things that can be combined, and you learn all the mathematical permutations. It's a very useful way of studying, because it gives you a defined workload and feeling of progress, and once you get good at studying permutations exercises they're ultimately quite quick to work through. Much quicker than learning thousands of unrelated transcriptions would be.
But Lang's stuff is also a sub-class of these exercises: it's a technically-derived "combine everything" exercise with no musical selection of combinations involved. You just take all the stickings and combine them with all the other ones, in lots of different note groupings.
I've seen a few of these kinds of systems in the past, and I do understand why people find these systems attractive. The idea of "learning to do everything" rather than just learning some stuff out of a book that somebody else thought was hip is rather appealing. But comparing the two approaches they're both rather lacking in important ways:
Option 1: Learn from books of pre-existing musical material, like "West African Rhythms for Drumset" or John Riley's jazz books.
Advantages: Material you're studying is actually musically structured and has an existing musical purpose. You know the material is all going to be of reasonably good value given the time put in.
Disadvantages: It's not your own material. If you go right through the book then you're only ever going to be a second-rate copy of whoever wrote the book for you, because it requires no original thought.
Option 2: Learn everything, create permutation-based systems for combining every pattern you can think of.
Advantages: You'll learn a lot of co-ordination and cover a lot of ground in detail, and you're not studying something that is some other guy's idea of musical playing so you're unlikely to come out a clone of another player in musical terms.
Disadvantages: You're not studying something that is anybody's idea of musical playing, instead cutting off your nose to spite your own face by practicing thousands of hours of material that not even YOU think sounds any good - you just think you have to do it because it's "part of the system".
So I'd suggest that the best approach is one that combines the advantages of both systems: You need to be practicing material that you believe is musically valid, yet which is not entirely something that somebody else has written for you. You can also use the powerful training benefits of combinations practice by creating exercises that combine patterns in a way that you find musical.
The difficulty with this third approach is that it actually requires thinking, opinions and a lot of consideration of the material you play. Effectively you have to write your own drum book and then be the first person to learn it all, which is a lot more work than just practicing. Practicing is just learning - duplication. Exercise design is where you get to be original and creative, or create facility to be spontaneously creative later when you're playing freely with the material you're studying.
There's no reason that this kind of practice can't lead to Lang-like technical results, too - it just requires Lang-like amounts of practice time to go with it...
I think a good template of all this is the system that Horatio Hernandez put together for himself and outlines in "Conversations in Clave" - which is a damnably hard book, if you ask me. That book blends a bit of both - it uses certain stylistic requirements (clave patterns) and then trains co-ordination by running rhythmic permutations against that. Sure, it's not going to cover every combination of hands and feet - but it will teach you enough permutations to play freely against a clave, while simultaneously helping you understand the rhythmic feel of resolving to different positions in the bar against the clave pattern.
Another good example would be the systems that Alan Dawson used to teach from the Ted Reed "Syncopation" book. Each system has a specific musical concept or sound, but it teaches you to play freely inside changing rhythms using that concept. It has musical thought put into constructing the systems, yet can be applied to lots of technical permutations.
I'm not suggesting that everybody go away and learn Conversations In Clave or read up on Dawson's methods, but just that they're a good template for a co-ordination book that we could all be writing for ourselves. I'm working on some stuff right now (attached) which is actually all based on technical/movement patterns and ostinato permutations - but I've tried to keep it all fairly anchored to a specific musical purpose that is interesting to me at the moment.
To explain, I was trying to work on a few things:
1) I wanted to develop more freedom in the particular feel dictated by the 1-bar-clave bass drum pattern shown on the ostinato sheet. It's a really common feel that shows up in latin styles and also heavily in Balkan marching music. I've been listening to a fair bit of balkan brass music lately, particularly stuff like Boban Markovic (get the album "Hani Rumba" if you can, it's really cool) and was really digging the feels they manage to build with just snare and bass drum. Really loose but really sick and intricate - it's worth checking out.
2) I've got some technical problems that need a bit of dedicated study of certain stroke combinations - sometimes when I have up strokes and down strokes with one hand falling in between full strokes in the other it causes some issues with the full strokes. I settled on a F-D-T-U-D-T-U-F (full-down-tap-up-down-tap-up-full) sequence as the best one to isolate the particular technical problem I was having, particularly when it is played with both hands in different displacements.
3) I want to keep developing my dynamic control in the bass drum, particularly while playing changing accents over it with the hands so I'm accenting in three limbs in a fairly independent kind of way.
So I decided I was going to build all of that into some material that I could use to work on all that stuff. Initially since it's too challenging I could split it up into individual exercises for each of the above points, but as I improve at each they unify into single exercises that can be practiced all at once.
What I ended up doing was writing a simple computer program (yeah, I'm a geek..) that took all the possible displacements of the FDTUDTUF stroke sequence in each hand and combined them every possible way in the left and right hand, played as a single-stroke roll. This gave me a whole pile of rhythms (outlining the accents), and since I'd made the program output to Lilypond .ly file format I could generate a MIDI file of those rhythms being played, as well as notation of them.
In a Lang-like system I'd just learn all of these over my ostinatos. But if you ask me that's a waste of time, because most of them sound crap.
So to bring some musical judgment into the exercise, I took the MIDI file and loaded it into Cakewalk Sonar. I put the MIDI file of rhythmic combinations on one track, then made another track which just contained a loop of my bass drum ostinato. After that I spent about an hour listening through all the one-bar groove phrases that came out of that and noting down which ones I thought sounded pretty hip. After going through the remainder a few times I settled on sixteen bars worth of stuff that I quite liked, and structured it roughly into the attached melody sheet.
So what I now have is a 16-bar exercise, any one bar of which requires the exact same series of strokes in each hand to play correctly. So I can use any bar of that as a dynamics and stroke technique exercise, helping me co-ordinate up and down strokes with one hand against full strokes in the other in a cleaner manner. That lets me sit down with a mirror and this material and be a technique geek.
Then once I've got comfortable playing any one bar of this exercise I can sit down and learn to play either of the ostinatos against it. They're just two of the eight possible accenting permutations (2^3), but they're easy ones and they each have a particular feel.
So by working on this stuff I develop dynamic technique in the hands and with the bass drum, and I improve my vocabulary in a particular style of playing in a way that I find musically sound. Obviously this kind of technique-derived rhythmic material only has a fairly limited set of possibilities musically, so I combine it with typical traditional snare drum patterns transcribed from recordings and use both to improvise over a particular ostinato.
I've been doing this for a week or two now, and I can feel my co-ordination and groove in this particular type of playing improving dramatically, plus it's helping sort out some nagging issues in my hand technique.
The way I've put the exercises together is very technology-centric, but basically it touches all the points I think that home-grown exercises should:
1) Is it challenging to you, and working on specific problems in your playing?
2) Is it musically valid, have you listened to the results objectively (without considering how difficult different patterns are) and decided whether they're something you like?
3) Can it be directly applied to playing music that you're enthusiastic about?
Permutations are a great way of dealing with #1, but to deal with #2 and #3 you also need to include some consideration of actual musical material and gather understanding of how different styles work. That comes from listening and transcription.
So my suggestions for exercise design to help you be more creative and interesting in the music you want to play:
1) Make them about a specific concept you find exciting. This could be a new idea of your own, or a concept you hear played on a recording that you want to expand on. Learn to read if you can't, and transcribe music that you like. You'll start seeing patterns, and these patterns will lead to conceptual ideas.
2) Read up, and use existing exercise systems as your templates for the exercises you design. Learn about Ted Reed/Syncopation based exercises. Learn about Gary Chaffee and his fatback combination exercises. Check out "The New Breed", check out Conversations In Clave etc. These systems actually work, and lots of drummers have used them - so if you can modify this approach to use in your own practice then it'll probably work efficiently too.
3) Edit. If you use a combination-based system make sure that all the combinations you're working on are actually musically interesting to you, and can be used in a way that you enjoy. Don't just blindly play through a billion combinations because they're there.
4) Apply. Try to make sure that the exercises you design are sufficiently grooving that you can actually play them over records and try to make them feel right, because it's the stuff beyond the notes that make something feel right stylistically as much as it is the patterns themselves.
5) Record the results. Check back in six months against the stuff you wrote up, and make notes on what worked and what didn't. What made it into your playing that you think is hip, what bad habits you think the exercise caused. Then keep all this in mind when you design future exercises.
This started out as a reply to the "Matrix" thread about Thomas Lang's concept on his new DVD. I think it's a badly-designed exercise, and I explained why over in that thread a little. But rather than just bashing, I think I should make this a bit more general and talk about exercise design: Why you should be doing it, and some things that I think objectively make practice exercises good or bad.
First up, what do I mean by exercise design: I mean literally sitting down and writing exercises out for yourself that you're going to practice. Why do it? There are lots of drum books on the market, after all.
My suggestion is that every great drummer does their own exercise design. That's why they're able to put their own books and teaching method DVDs out on the market: They just take the stuff they design for themselves and write it up tidily to sell. Now, that doesn't mean you shouldn't buy DVDs or books: it is great to see how other people have designed their exercises and take inspiration from that to do your own. But I don't think it's healthy to just practice religiously out of other people's books.
For my purposes I separate drum books into three types:
1) Personal books. This would be something like Bill Bruford's book, David Garibaldi's book, things like "West African Rhythms for Drumset", many of the marching/snare solo books etc etc. These are books that document one drummer's personal journey as a player and write up a lot of the stuff they've developed for themselves musically. Some are more about transcriptions - like Bruford or Garibaldi - some are about stylistic material like the Royal Hartigan african rhythms book.
2) Technical books. These would be things like Stick Control, Accents & Rebounds, Master Studies, Gary Chaffee's "Patterns" series or Gavin Harrison's two books on rhythm. These aren't so much about music as they are about developing facility, and for the most part they're pretty much definitive works - once a particular book does a certain aspect of technique/facility stuff right then it is pretty rare for another book to come along that does the same thing better. We're still using Stick Control, after all.
3) Material books. I'd tend to see these as being things like The New Breed or Syncopation which give you reading material that can be used to design your own exercises, although there are plenty of existing interpretations that can be used with either.
Now, of those books I tend to think that everybody should have the key books from #2 and #3. You need to go through Stick Control to get some control of the sticks. You need to have Syncopation handy so that you can use it as a tool in designing your own exercises. But the #1 books are, for me, totally optional and entirely up to your taste. If you get a book of that description and don't really think the results are hip then ignore it. My suggestion is that we should ALL be writing our own personal book, and that other people's books are just there for you to take inspiration from.
So what makes a good exercise? I reckon a good exercise is something that meets the following goals:
1) It works on something that you are currently unable to do. If you're just playing stuff that you can play easily then it's playing, not practice. Get gigs if you want to do this, they'll make you focus harder because there's an audience watching.
2) When played correctly it sounds musical to your ears. What you practice will come out in your playing, and if your exercises sound unmusical to you then you'll develop playing habits that you dislike and which make you unhappy.
3) It can be directly applied to a style of music that you're enthusiastic about listening to and playing. Otherwise... why bother?
If there are exercises written by other people that fit those criteria then by all means practice them. But if you want to sound like you, odds are you're going to need to start writing your own.
Now, I just want to get back to the Lang stuff for a minute. Lang's "Matrix" system is a particular class of exercise that's very common in music education: it's a combinations exercise. You take a certain number of things that can be combined, and you learn all the mathematical permutations. It's a very useful way of studying, because it gives you a defined workload and feeling of progress, and once you get good at studying permutations exercises they're ultimately quite quick to work through. Much quicker than learning thousands of unrelated transcriptions would be.
But Lang's stuff is also a sub-class of these exercises: it's a technically-derived "combine everything" exercise with no musical selection of combinations involved. You just take all the stickings and combine them with all the other ones, in lots of different note groupings.
I've seen a few of these kinds of systems in the past, and I do understand why people find these systems attractive. The idea of "learning to do everything" rather than just learning some stuff out of a book that somebody else thought was hip is rather appealing. But comparing the two approaches they're both rather lacking in important ways:
Option 1: Learn from books of pre-existing musical material, like "West African Rhythms for Drumset" or John Riley's jazz books.
Advantages: Material you're studying is actually musically structured and has an existing musical purpose. You know the material is all going to be of reasonably good value given the time put in.
Disadvantages: It's not your own material. If you go right through the book then you're only ever going to be a second-rate copy of whoever wrote the book for you, because it requires no original thought.
Option 2: Learn everything, create permutation-based systems for combining every pattern you can think of.
Advantages: You'll learn a lot of co-ordination and cover a lot of ground in detail, and you're not studying something that is some other guy's idea of musical playing so you're unlikely to come out a clone of another player in musical terms.
Disadvantages: You're not studying something that is anybody's idea of musical playing, instead cutting off your nose to spite your own face by practicing thousands of hours of material that not even YOU think sounds any good - you just think you have to do it because it's "part of the system".
So I'd suggest that the best approach is one that combines the advantages of both systems: You need to be practicing material that you believe is musically valid, yet which is not entirely something that somebody else has written for you. You can also use the powerful training benefits of combinations practice by creating exercises that combine patterns in a way that you find musical.
The difficulty with this third approach is that it actually requires thinking, opinions and a lot of consideration of the material you play. Effectively you have to write your own drum book and then be the first person to learn it all, which is a lot more work than just practicing. Practicing is just learning - duplication. Exercise design is where you get to be original and creative, or create facility to be spontaneously creative later when you're playing freely with the material you're studying.
There's no reason that this kind of practice can't lead to Lang-like technical results, too - it just requires Lang-like amounts of practice time to go with it...
I think a good template of all this is the system that Horatio Hernandez put together for himself and outlines in "Conversations in Clave" - which is a damnably hard book, if you ask me. That book blends a bit of both - it uses certain stylistic requirements (clave patterns) and then trains co-ordination by running rhythmic permutations against that. Sure, it's not going to cover every combination of hands and feet - but it will teach you enough permutations to play freely against a clave, while simultaneously helping you understand the rhythmic feel of resolving to different positions in the bar against the clave pattern.
Another good example would be the systems that Alan Dawson used to teach from the Ted Reed "Syncopation" book. Each system has a specific musical concept or sound, but it teaches you to play freely inside changing rhythms using that concept. It has musical thought put into constructing the systems, yet can be applied to lots of technical permutations.
I'm not suggesting that everybody go away and learn Conversations In Clave or read up on Dawson's methods, but just that they're a good template for a co-ordination book that we could all be writing for ourselves. I'm working on some stuff right now (attached) which is actually all based on technical/movement patterns and ostinato permutations - but I've tried to keep it all fairly anchored to a specific musical purpose that is interesting to me at the moment.
To explain, I was trying to work on a few things:
1) I wanted to develop more freedom in the particular feel dictated by the 1-bar-clave bass drum pattern shown on the ostinato sheet. It's a really common feel that shows up in latin styles and also heavily in Balkan marching music. I've been listening to a fair bit of balkan brass music lately, particularly stuff like Boban Markovic (get the album "Hani Rumba" if you can, it's really cool) and was really digging the feels they manage to build with just snare and bass drum. Really loose but really sick and intricate - it's worth checking out.
2) I've got some technical problems that need a bit of dedicated study of certain stroke combinations - sometimes when I have up strokes and down strokes with one hand falling in between full strokes in the other it causes some issues with the full strokes. I settled on a F-D-T-U-D-T-U-F (full-down-tap-up-down-tap-up-full) sequence as the best one to isolate the particular technical problem I was having, particularly when it is played with both hands in different displacements.
3) I want to keep developing my dynamic control in the bass drum, particularly while playing changing accents over it with the hands so I'm accenting in three limbs in a fairly independent kind of way.
So I decided I was going to build all of that into some material that I could use to work on all that stuff. Initially since it's too challenging I could split it up into individual exercises for each of the above points, but as I improve at each they unify into single exercises that can be practiced all at once.
What I ended up doing was writing a simple computer program (yeah, I'm a geek..) that took all the possible displacements of the FDTUDTUF stroke sequence in each hand and combined them every possible way in the left and right hand, played as a single-stroke roll. This gave me a whole pile of rhythms (outlining the accents), and since I'd made the program output to Lilypond .ly file format I could generate a MIDI file of those rhythms being played, as well as notation of them.
In a Lang-like system I'd just learn all of these over my ostinatos. But if you ask me that's a waste of time, because most of them sound crap.
So to bring some musical judgment into the exercise, I took the MIDI file and loaded it into Cakewalk Sonar. I put the MIDI file of rhythmic combinations on one track, then made another track which just contained a loop of my bass drum ostinato. After that I spent about an hour listening through all the one-bar groove phrases that came out of that and noting down which ones I thought sounded pretty hip. After going through the remainder a few times I settled on sixteen bars worth of stuff that I quite liked, and structured it roughly into the attached melody sheet.
So what I now have is a 16-bar exercise, any one bar of which requires the exact same series of strokes in each hand to play correctly. So I can use any bar of that as a dynamics and stroke technique exercise, helping me co-ordinate up and down strokes with one hand against full strokes in the other in a cleaner manner. That lets me sit down with a mirror and this material and be a technique geek.
Then once I've got comfortable playing any one bar of this exercise I can sit down and learn to play either of the ostinatos against it. They're just two of the eight possible accenting permutations (2^3), but they're easy ones and they each have a particular feel.
So by working on this stuff I develop dynamic technique in the hands and with the bass drum, and I improve my vocabulary in a particular style of playing in a way that I find musically sound. Obviously this kind of technique-derived rhythmic material only has a fairly limited set of possibilities musically, so I combine it with typical traditional snare drum patterns transcribed from recordings and use both to improvise over a particular ostinato.
I've been doing this for a week or two now, and I can feel my co-ordination and groove in this particular type of playing improving dramatically, plus it's helping sort out some nagging issues in my hand technique.
The way I've put the exercises together is very technology-centric, but basically it touches all the points I think that home-grown exercises should:
1) Is it challenging to you, and working on specific problems in your playing?
2) Is it musically valid, have you listened to the results objectively (without considering how difficult different patterns are) and decided whether they're something you like?
3) Can it be directly applied to playing music that you're enthusiastic about?
Permutations are a great way of dealing with #1, but to deal with #2 and #3 you also need to include some consideration of actual musical material and gather understanding of how different styles work. That comes from listening and transcription.
So my suggestions for exercise design to help you be more creative and interesting in the music you want to play:
1) Make them about a specific concept you find exciting. This could be a new idea of your own, or a concept you hear played on a recording that you want to expand on. Learn to read if you can't, and transcribe music that you like. You'll start seeing patterns, and these patterns will lead to conceptual ideas.
2) Read up, and use existing exercise systems as your templates for the exercises you design. Learn about Ted Reed/Syncopation based exercises. Learn about Gary Chaffee and his fatback combination exercises. Check out "The New Breed", check out Conversations In Clave etc. These systems actually work, and lots of drummers have used them - so if you can modify this approach to use in your own practice then it'll probably work efficiently too.
3) Edit. If you use a combination-based system make sure that all the combinations you're working on are actually musically interesting to you, and can be used in a way that you enjoy. Don't just blindly play through a billion combinations because they're there.
4) Apply. Try to make sure that the exercises you design are sufficiently grooving that you can actually play them over records and try to make them feel right, because it's the stuff beyond the notes that make something feel right stylistically as much as it is the patterns themselves.
5) Record the results. Check back in six months against the stuff you wrote up, and make notes on what worked and what didn't. What made it into your playing that you think is hip, what bad habits you think the exercise caused. Then keep all this in mind when you design future exercises.