Two fun websites
Jim Kroger
http://members.w-link.net/~jsss/
--
Watch out for spam block
List of 30 online jazz guitar instruction websites:
http://www-personal.umich.edu/~jimkk/jazzfast.html
Steve Carter
while. I disagree with his scale suggestions, but that's a horse of a
different color.
I wrote a program that suggests scales, also. I call it THoTH. I'll be
writing it up as part of my Player's Journal on my web site, but
here's a little about it.
THoTH stores information about tunes in a database: title, composer,
style, etc. That's pretty mundane. But it also reads MusicPrinter Plus
(MPP) files, determines (from the "sheet music") what the chords are,
analyzes the chord progression, determines the chord-scales, and
stores that information in the database. So I can, for example, say
"show me a list of tunes that use Lydian b7". Or "show me a list of
tunes that use both a min7b5 chord and a dom7b9 chord. Then I can
click on one of the tune names and go directly to MPP, where I can see
the notation, print it, play it as MIDI, etc. This came in pretty
handy when I was teaching at Berklee.
I did a presentation on THoTH at a conference on music and cognition
at McGill University in 1997. Some day I'd like to do more with THoTH.
Steve
Steve Carter
Solo Jazz Guitarist
www.frogstoryrecords.com
Dan Adler
> I'm spending way too much time on the web lately....
>
>
> http://members.w-link.net/~jsss/
>
> http://www.changes98.com/
Jim,
What do you get from these sites? Maybe it makes sense for horn
players who can't play piano, but this all seems like stuff you are
better off figuring out for yourself on guitar.
Mike Archbold
> I'd seen Jazz Scale Suggester years ago, so it's been around for a
> while. I disagree with his scale suggestions, but that's a horse of a
> different color.
>
Hi, I wrote the program... a little background about it...
When I started I was just trying to learn the CLIPS language, and
decided to use jazz improv as an domain for developing an expert
system. (Expert systems have been around since the 70s.)
Originally I had no plans to develop anything "serious." I just had a
ii-V-I, ii-V, that sort of thing, and no user interface. I thought I
could develop a working rule base in two weeks. Eighteen months later
I was satisfied that I had taken the idea as far as I could. I
literally just kept tinkering with it until I thought it was "done".
But the thing was accidentally developed for the most part.
One interesting part came in trying to devise a method of sorting out
tonal areas, especially when you have chords in close proximity that
may or may not sound like a single key. Basically, the way I coded
it, if a major chord is encountered (and similarly for minor), the
rule base generates candidates for neighboring chords. So if the
program sees a C chord, it generates a D minor, E minor, F major, and
so on which may or may not attach to neighboring chords. Thus it
utilizes a generate-and-test paradigm.
But what happens if you see an F and a D minor near the C chord? Dm
and F are both in the keys of C and in F. Exactly how close does the
F and Dm chords have to be to the C chord for the cluster to be in the
key of C, and not hinting an F major cluster? Typically we read
charts going forward, but we can also go backwards too. The program
doesn't start at the beginning and go forward either. It looks for
the most promising areas, like a major chord ~anywhere~, and generates
all the candidates concurrently. It might have to throw out (RETRACT)
candidates. In the example above the program would generate
candidates in the keys of F and C for the Dm and F chords
simultaneously. It has to decide which candidates to throw out
depending on what relations exist.
I like the ideas you have in your program. Also some ideas I had
were: enhancements to make the suggestions more difficult as the
player advanced, a facility to listen to the solo and suggests
alternates, catering the suggestions to the level of theory the person
has. (Most of the rock guitar players I've known have very little
interest in sheet music or theory. I can't tell you how many players
I've met whose eyes gloss over when you start talking about theory.)
Also, I think a fuzzy logic approach might be easier to implement.
The program has about 200 or so rules, and it would be dangerous
modifying that thing now, as I've forgotten what I was doing for the
most part. There was a period about two years ago when I was working
on it day and night for three weeks, my guitar in one hand and typing
into the program with the other.
To make a long story short, I spent wwwwaaaayyy more time than I
should have on the program, but I think in the end it works fairly
well for a person who is possibly a rock guitar player who wants to
get into jazz. It was kind of accidentally developed, and in any AI
program the problem is usually the "combinatorial explosion", but for
the majority of the tunes I tried it on, the program seems to produce
decent sounding scale suggestions over a problem domain that is widely
variable. I read once that the number of possible outcomes to a chess
game was about the same number as the total atoms in the universe.
The number of possible jazz charts is enormous too, and also the
scales used are a matter of taste; therefore, many programs are
possible.
Steve Carter
Nice to hear from you. A pleasant surprise, actually. I don't think
there are a lot of people in the world who have attempted what you and
I have attempted: an AI program for jazz chord-scales.
The premise of my presentation on THoTH at the 4th International
Conference on Music and Cognition, was that, to a large extent,
chord-scales *could* be determined by simply moving forward through
the progression. My system was just a pretty simple set of if-then
statements, but what made it work -- at least in my mind :-) ) was
that it was based on Bill Leavitt's chord-scale approach. Bill used to
say, "The ear has an eye, but no memory." No one understood him. :-)
But his premise was that the scale for a chord was determined by the
effect of that chord on "the preceding tonal area." My app is pretty
good at coming up with what Bill woulld have considered "the right
scales", no matter what chord progression you throw at it. Within
limits, of course. I've mostly used it on standards.
I used to teach harmony at Berklee, and I know that THoTH could get an
A on a Harmony 2 final.
I had a lot of fun writing it.
From what you say, your app is really more of a "real" AI application,
whereas mine is more of a brute-force app, based on some rigig rules.
To be honest, I forget how I determined which chords are in the "key
of the moment" -- but it works pretty well. I did kind of kludge
modulations -- by looking for a ii-V, but if you do put in the ii-Vs
in the sheet music (MPP file), THoTH catches the modulations pretty
well. It does OK with "Body and Soul" -- which is more than I can do
some nights :-).
At the time, I also spent "way too much time" writing it, but these
days I work days as a software engineer, and gig -- when I can find a
gig -- so maybe it was good practice.
Steve
On 1 Sep 2002 16:34:36 -0700, jaz...@w-link.net (Mike Archbold)
wrote:
Steve Carter
Solo Jazz Guitarist
www.frogstoryrecords.com
Jim Kroger
available? What did you write it in? I'd like to get more details on
its constrction, being an old expert system writer. Maybe we can do
that offline as it's OT. Ever thought of making it into a commercial
product? Another idea, ever thought about making it into a web site?
I'd pay for it. I already paid for JSSS :-) .
Hey, here's an idea.....make it into a palm pilot application.
Jim
"Ars longa, vita brevis"
Jim Kroger
wrote:
>Hi, I wrote the program... a little background about it...
>
>When I started I was just trying to learn the CLIPS language, and
>decided to use jazz improv as an domain for developing an expert
>system. (Expert systems have been around since the 70s.)
>
<SNIP>
Hi Mike, nice to make your aquaintance. I discovered your program last
night and immediatly bought it (with a price like that how coud I
not?). I am tickled pink with it.
I suspected you were using CLIPS. It's really about the only thing
around now. Interesting you only needed 200 rules. Also your generate
and weed approach is interesting. My reaction to the program, aside
from being completely delighted, is a wish for more thorough
explanation. Others may use it to get ideas to put to practical use,
but I'm interested in it as a learning tool, in that I'd like to
understand how to choose scales over chords better. It would be great
if it generated a few possibilities and explanations for each. It
would be great if it included deeper explanations, for example,
instead of (I paraphrase from memory) "a minor harmonic creates
tension over a...," an explanation of why this is the case. But I
guess we could take this down to the very basics "an interval is the
distance between...." so where do you stop. At least with the
explanations you provide I can research the answers and learn and
that's why I'm so pleased. Anyway, I know what it's like to try and
modify a program you haven't touched for a while, no thank you. But if
you get brave, here's a thought: would a simple printout of the rules
fired and the problem space at that time (or whatever it's called in
CLIPS, the current set of facts, variable assignments, etc) provide
some of the more extensive explanation of its thinking I'm looking
for?
Anyway, thanks much for the great program.
Jim
--
Watch out for spam block
List of 30 online jazz guitar instruction websites:
http://www-personal.umich.edu/~jimkk/jazzfast.html
"Ars longa, vita brevis"
Jim Kroger
just posted may explain why.
I mean what else am I gonna do, you won't answer my emails so I have
to figure it out somewhere ;-). I'm not smart enough to figure it
out on my own.
I really do like your playing though, and I'm still working on your
"Green Dolphin Street." The good thing about updating my web site is
it induced me to visit some of the players' sites I hadn't been to and
started listening to clips. Really some great musicians in this
newsgroup.
Jim
--
Watch out for spam block
List of 30 online jazz guitar instruction websites:
http://www-personal.umich.edu/~jimkk/jazzfast.html
"Ars longa, vita brevis"
Mike Archbold
Jim,
Thanks for the compliment. These are all good ideas. For my part I
feel like I took the idea as far as I could go with it (or wanted to
go with it). One of the reasons it only costs $10 is that, as you
said, the explanations may not be comprehensive. Some kind of
question and answer facility would be nice. There are tons of things
that can be done with this idea, and I suspect in the future we'll see
more of this kind of application with a lot fancier things going on
that will probably make my program look like child's play. Jazz
improvisation is a perfect domain for expert systems since the art
exhibits all kinds of constraints and regularities, and I certainly
don't own any patent on the idea.
Mike
Steve Carter
Nah, students would use it to cheat on Harmony exams and my friends
who teach harmony would beat me up. ;-)
Yes, I've thought a lot about making THoTH commercial. It's a long
story, so let's discuss it offline. I will be starting a THoTH section
on my site soon. At first, just discussing the history and theory,
with maybe some screen shots. Later, I hope, actually making the app
available, somehow, on line.
Steve
On Mon, 02 Sep 2002 04:50:37 GMT, jimkkre...@umich.edu (Jim
Jim Kroger
wrote:
>Jim,
>
>Thanks for the compliment. These are all good ideas. For my part I
>feel like I took the idea as far as I could go with it (or wanted to
>go with it). One of the reasons it only costs $10 is that, as you
>said, the explanations may not be comprehensive. Some kind of
>question and answer facility would be nice. There are tons of things
>that can be done with this idea, and I suspect in the future we'll see
>more of this kind of application with a lot fancier things going on
>that will probably make my program look like child's play. Jazz
>improvisation is a perfect domain for expert systems since the art
>exhibits all kinds of constraints and regularities, and I certainly
>don't own any patent on the idea.
>
>
>Mike
Mike... well, it's a great little program and I'll get many times my
money's worth out of it, so thanks again. Very interesting to see two
of my interests combined like this. It's already spurred me to check
out new things. Like you probably, I don't have time to pursue it,
unfortunately. I guess we'll have to leave it up to Steve...
Keith Freeman
>should have on the program, but I think in the end it works fairly
>well for a person who is possibly a rock guitar player who wants to
>get into jazz.
I speak as someone who wishes that the wealth of excellent educational material
now available was there thirty years ago, when I first decided to try to play
jazz guitar. It's really only since I discovered and started using Aebersold
playalongs that I made any progress, and then only in the last couple of years
that I've got to the stage where my practising produces a noticeable
improvement. So at the age of fifty, a big concern for me is not wasting time.
And that brings me to my point: I have the idea that there is far too much
emphasis on the scalar approach in a lot of the educational material. Of course
it's important to know when learning a chart that D- G7 C are in the key of C,
but I question the usefulness of suggesting (or at least reinforcing the
notion) that "you can play the C major scale over that". In my own case I found
that produced aimless improvisations. It wasn't until I started outlining the
chords (with arpeggios) that my impros started to sound like music. Indeed, as
Emily Remler pointed out in her video, in a fast bop tune there isn't usually
time to do more than outline the chords.
But most of the books and videos seem to start with scales. I found trying to
work with Aebersold's written-out scale suggestions merely resulted in
information overload. Apart from not producing very interesting music, running
the scales doesn't let the chord function (dominant, tonic, subdominant etc.)
be heard.
My guess is that someone starting out would do better to concentrate on the
triads and seventh chords (and the triads in the chords). It may even be that
the notes "in between" will take care of themselves, that a student could find
them by ear. (As I don't teach, I have no way of testing this, but perhaps some
of you who do teach already apply it, or would be willing to try it on a
student.)
-Keith
(just my EUR 0.02)
Rick DelSavio
chord tones.
There's a bunch, (he said as he plugged his site,) of examples of this technique
over II-V's and "Giant Steps" at the site below. I often wonder how guys like
Tal, or Billy Bauer etc. were able to "do it" without all the material that we have
now. Anyway...Good luck to you. Rick
Keith Freeman wrote:
--
Jazz Guitarist/Educator
Check out lessons and original music @
http://www.rickdelsavio.com
Steve Carter
I agree with you and Keith that arpeggios are really the place to
start. Sometimes I think we teach scales because they are easy to
teach. ;-)
As for material, I know I wrote a lot in the process of trying to
figure this stuff out for myself. For example, (plugging my own site)
that arpeggio study on "God Bless the Child" is the kind of thing I
did a lot while learning to play over changes.
As Howard Roberts used to say: there are only three things that go in
to any melody: scales, arpeggios, and intervals. So I've tried to
write exercises in all three. Of course, there's more to it than that
(Howard was being a bit facetious). There's motivic development, "the
long line", deep structure...
Steve
Mike Archbold
These are all good ideas. Future programs I suspect will do a lot more than mine.
Actually my opinion is that it would be best to think along two dimensions, arpeggio and scale simultaneously.
Personally I like being aware of the underlying scale, so there are quite a few notes to choose from, though as
you well know some sound better than others. I don't think there is anything wrong with starting out with a
scale approach. I suspect we all eventually gravitate toward the color tones anyway. The program points out
some notes to accent here and there.
In general the program is just intended to get somebody started. Part of the idea came from talking to my next
door neighbor in the late 90s. He is a rock guitar player and he wasn't really aware of all the relationships
that go on between chord/scale. He had the minor pentatonic down, barre chords, and so forth, but I couldn't
come up with a way of getting him to see the big picture. It seems like part of the problem is the magazines
that list tab. It's sort of a paint-by-the-numbers affair with that stuff. So if you take a guy grounded in
the tab-paint-by-numbers who is interesting in jazz but turned off by the complexity, what do you do? He was
aware of a lot of the note names on the fret board, but wasn't keen on making the connection to sheet music.
I'd call out "A" and he'd know where to go for the most part. Basically that problem is one factor that got me
started on the program. It's not a panacea of course, rather one possible method of getting started. If you
have a guitar player who has the pentatonic and rock chords down, how can you transition into jazz? Eventually
what happened to me was that I got tired of ringing ears and three chords. I love rock, but there was something
pleasant about a major seventh chord... well anyway I'm rambling.
Rick DelSavio
hung-up on scales initially.
Which may in large part be because "Rock" progressions are so modal and
one "can" riff/shred (God I hate that term!) all afternoon with E
Aeolian. Anyway... That was interesting to hear what Howard Roberts said
about 3 things...I recall hearing Zappa say on a radio interview that
either a note is a chord tone (consonant) or it isn't. (Dissonant) And
depending on which chord-tone it is not, that will yield various degrees
of tension/dissonance. I think that we should ask a Jazz sax instructor,
for instance, would they start a student off with emphasis on one or the
other. Or both. Hmm..Thanks, Rick
Steve Carter wrote:
--
Jim Kroger
beginning of one of his chapters, in which he teaches improvising
over chord tones first.
Jim
>Keith wrote:
> It wasn't until I started outlining the
>chords (with arpeggios) that my impros started to sound like music. Indeed, as
>Emily Remler pointed out in her video, in a fast bop tune there isn't usually
>time to do more than outline the chords.
--
Watch out for spam block
List of 30 online jazz guitar instruction websites:
http://www-personal.umich.edu/~jimkk/jazzfast.html
"Ars longa, vita brevis"
Joey Goldstein
Jim Kroger wrote:
>
> I think Joey G. would agree with you. You virtually paraphrased the
> beginning of one of his chapters, in which he teaches improvising
> over chord tones first.
Sure I agree.
But I also think that as far as learning the fretboard is concerned and
starting to get your ears together is concerned that scales are still
the place to *start*.
The whole major/minor tonal key system is predicated on the major and
minor scales. The most important chords in tonal music (maj, min, dim
and aug triads) are subsets of the major and minor scales. On the
guitar, the easier arpeggio fingerings for these chord types are best
discovered as subsets of the easier scale fingerings for the major and
minor scales.
But someone approaching the jazz repertoire trying to iterate an entire
scale over each chord (or trying to use one scale over a series of
chords) will quickly be frustrated.
The main trick to being able to play over changes is to be able to *play
the changes*. Soloing over changes is in many ways a lot like comping
over changes. You fill in X amount of time on the sound of one *chord*
then the next *chord*. If you play the notes of the chord at the right
time your listener will hear that clarity and you yourself will develop
the clarity to know exactly where you are in the progression. If however
your concentration is on merely playing the notes of a scale at the
right time you may not even know what chord you are playing on or where
you are in the form of the tune.
Of course all music making is a balancing act and too much "vertical"
chord tone melody will need to be contrasted with some "horizontal" key
based scalar stuff to achieve a good balance. In the end, whether your
approach is primarily vertical or horizontal to be able to play
effectively over this repertoire you have to be aware of what chord you
are on at all times and how every note you play relates to that chord
(unless of course you're just winging it which can be another nice contrast).
So yes there is an over emphasis on scales in much of the jazz and pop
instructional material that is readily available. But some of the blame
has to be put on the student too because young students invariably
overblow the importance of their own early accomplishments in their own minds.
Learning your scales is just the tip of the iceberg, a very very big
iceberg. Congratulations are in order once you accomplish this because
it was real hard work but now you have to get on with learning how to play.
Learning your arpeggios is just another tip of the iceberg sticking up
out of the water at some other spot. Get on with it.
> Jim
>
> >Keith wrote:
>
> > It wasn't until I started outlining the
> >chords (with arpeggios) that my impros started to sound like music. Indeed, as
> >Emily Remler pointed out in her video, in a fast bop tune there isn't usually
> >time to do more than outline the chords.
>
> --
> Watch out for spam block
>
> List of 30 online jazz guitar instruction websites:
> http://www-personal.umich.edu/~jimkk/jazzfast.html
>
> "Ars longa, vita brevis"
--
Joey Goldstein
Guitarist/Jazz Recording Artist/Teacher
Home Page: http://www.joeygoldstein.com
Email: <joegold AT sympatico DOT ca>
Mike Archbold
learning how to develop an expert system and also that it forced me to do some serious thinking and a lot of
research. I got a lot out of it. The problem was that I got started.... once you get started on something like
this it has a tendency to snowball, unless you are smart and delete it while you still have the chance :)
Nazodesu
<freeke...@compuserve.com> wrote:
> And that brings me to my point: I have the idea that there is far too
> much emphasis on the scalar approach in a lot of the educational
> material. Of course it's important to know when learning a chart that
> D- G7 C are in the key of C, but I question the usefulness of
> suggesting (or at least reinforcing the notion) that "you can play
> the C major scale over that". In my own case I found that produced
> aimless improvisations.
You're right. Aimless improvisations for beginners should be done
using different references. Just kidding.
I think that the major scales should be learned in all positions, with
or without aimless improv, with which I have no problem. It's not
detrimental. It's harmful. It's useful. It's not music, but this is
a critical building block.
Arps should be learned second they are a subset of scales. One should
then learn the basic chord types in arpeggios in all positions. The
problem is this is much more conceptually difficult to learn and to
use. So many players continue to saw away on one-size-fits all scales
over a series of chords.
Somewhere in there one should also learn myrid triads and seventh
chrods in all positions.
> It wasn't until I started outlining the chords (with arpeggios) that
> my impros started to sound like music. Indeed, as Emily Remler
> pointed out in her video, in a fast bop tune there isn't usually time
> to do more than outline the chords.
Actually that's the point that I usually abandon my generally
chord-dominant thinking. I don't have time to outline chords
repetively, and when I do it sounds dull and rigid. I usually work in
a more scalar-dominant approach in these circumstances.
> But most of the books and videos seem to start with scales.
I think that's good. I think most people don't go beyond the
"starting" parts of videos and books.
> I found trying to work with Aebersold's written-out scale suggestions
> merely resulted in information overload. Apart from not producing
> very interesting music, running the scales doesn't let the chord
> function (dominant, tonic, subdominant etc.) be heard.
>
> My guess is that someone starting out would do better to concentrate
> on the triads and seventh chords (and the triads in the chords).
I think for the novice it's difficult to understand how triads and
larger chords fit in to the bigger picture. It's more difficult
thereafter to merge in scales on top of it. Just my view.
> It may even be that the notes "in between" will take care of
> themselves, that a student could find them by ear. (As I don't teach,
> I have no way of testing this, but perhaps some of you who do teach
> already apply it, or would be willing to try it on a student.)
Hardly any point of using a teaching approach that assumes the student
will play by ear, though. Certainly they can do that with out any
teaching approach at all.
Having said that I've never thought a specific methodic approach to
teaching "playing by ear" is given enough thought. The logic and
discipline of it can certainly be taught and practised and drilled by
teachers with their students.
Keith Freeman
>did a lot while learning to play over changes.
them out yet, but they certainly look like the way to go.
For guitarists I think there's another problem with the scale-based approach,
which is that it encourages us to stay in one position. Following my Holy Grail
;-} experience with Garrison Fewell at the summer course I'm now working a lot
more with the triads from the chords and sliding/hopping around the neck. While
the "find all the triads/7th chords in one position" approach may look more
logical, I suspect that it's actually more difficult for a beginner (which I
was for many many years!) than the moveable triad approach (taking the triads
on strings 432 and 321 as the starting point). The latter also has the
advantage of forcing you to spell out the subdom, dom and tonic functions. I've
found it's opened out my playing enormously.
I've noticed that an arpeggio played over two or more positions sounds a lot
more dynamic than the same arpeggio played in one position, partly because the
sound is less homogeneous and partly because you can use stronger fingerings.
-Keith
Keith Freeman
>On the
>guitar, the easier arpeggio fingerings for these chord types are best
>discovered as subsets of the easier scale fingerings for the major and
>minor scales.
If by "easier scale fingering" you mean one that stays in position and avoids
stretches, I wonder if this is really true. I know I had a lot of difficulty
learning (and fingering) all the seventh chord arps in the major scale that way
- after all, each one has a different fingering! I suspect that if I'd started
out with the approach I learned from Garrison (see my reply to Steve) it would
probably have been easier - it certainly wouldn't have been any more difficult.
By shifting around the fretboard you're able to recycle your fingerings,
plus you learn the fretboard much more thoroughly. Plus the phrasing is more
dynamic, more jazzy. For one thing it lets you get into sweep fingerings, so
you can introduce a bit of velocity right from the start.
Garrison hinted that once we had got this technique down and learned the
phrasing we could always go back to our old fingering styles (the "logical"
ones we'd all learned from the books). I wonder if I shall, though...
For those who don't know Garrison I should perhaps point out that he
deliberately harks back to the greats of the 50s, Wes, Grant, Kenny etc. While
some of them may not have had the chops of today's players, I don't think
anyone can deny that they had groove, feel, soul, and that came from their
phrasing, which in turn came from their fingering. And I guess that's my bag
too..
-Keith
Keith Freeman
>
>I think that's good. I think most people don't go beyond the
>"starting" parts of videos and books.
-Keith
Dan Adler
> I think Joey G. would agree with you. You virtually paraphrased the
> beginning of one of his chapters, in which he teaches improvising
> over chord tones first.
Jim,
When I had time to teach, I used to get very good results with people
starting with arpeggios and expanding them to "chord scales" (using
Joe Pass terminology rather than the mode terminology) rather than the
other way around. This approach is explained in detail here:
http://danadler.com/misc/ii-v-i.txt
Enjoy,
Mike Archbold
thing that drove Nietzsche nuts, namely the mistaken belief in opposites. We read some persons who are in favor
of a scale oriented approach; others in the arpeggio. The fact is that they are not contrary the one to the
other. Both arpeggios and scales are necessary. The benefit of arpeggios is that they serve to highlight the
chord, but the negative is that if you want to play a fast run, you may well wind up doing all kinds of crazy
gymnastics because there are only so many notes in an arpeggio. Also, playing nothing but arpeggios would soon
drive your audience to distraction, as they would no exactly what to expect. You wouldn't for instance play an F
over an Em7 chord in the key of C even in passing if you were using strictly arps. Scales have the advantage of
lots of notes to choose from, making it easier to play fast runs. Though we know not all of the notes in a
scale will always sound so hot, like a D over an A7 chord.
The problem with starting a player on nothing but arpeggios is that it becomes necessary to shift over every
chord. That's tons of arpeggios. Later on this is necessary, but just to start off, it's akin to driving a
stick shift instead of an automatic. You have to watch the road and also worry about shifting. Given that the
player will eventually learn how to solo with scales, there is no reason not to start with scales and move to
arpeggios once the player gets bored with the diatonic sound. At least the player can start out making some
music. Taking All the Things You Are the player would have to come up with tons of arpeggios to get though it,
but about a third as many scales. At least the player can get off to some start without being overwhelmed.
I'm not arguing against arps, but both need to be learned.
Adam Bravo
news:1103_10...@news.w-link.net...
> gymnastics because there are only so many notes in an arpeggio.
? There can be as 7 notes in an arpeggio.
Also, playing nothing but arpeggios would soon
> drive your audience to distraction, as they would no exactly what to
expect. You wouldn't for instance play an F
> over an Em7 chord in the key of C even in passing if you were using
strictly arps. Scales have the advantage of
> lots of notes to choose from, making it easier to play fast runs. Though
we know not all of the notes in a
> scale will always sound so hot, like a D over an A7 chord.
What is your definition of arpeggio? I mean, an arpeggio doesn't have to
only have the roote, 3rd, 5th, and 7th - it can go on to the 13th just like
a scale.
> The problem with starting a player on nothing but arpeggios is that it
becomes necessary to shift over every
> chord. That's tons of arpeggios. Later on this is necessary, but just to
start off, it's akin to driving a
> stick shift instead of an automatic.
My mother and aunt were both forced by their parents to learn stick shift
before they could get their license, in case they had to drive stick shift
for whatever reason.
You have to watch the road and also worry about shifting. Given that the
> player will eventually learn how to solo with scales, there is no reason
not to start with scales and move to
> arpeggios once the player gets bored with the diatonic sound. At least
the player can start out making some
> music. Taking All the Things You Are the player would have to come up
with tons of arpeggios to get though it,
> but about a third as many scales. At least the player can get off to some
start without being overwhelmed.
It's more difficult to make something musical from scales.
Playing scales starting from the root in eighth notes always ascending is
really not very musical sounding at all - very rare in most jazz. But
playing arpeggios in eighth notes ascending from the root is quite common.
It gets old quickly, but it's at least usable for a little bit.
Joey Goldstein
Keith Freeman wrote:
>
> I agree. The only point I'm not so sure about is:
>
> >On the
> >guitar, the easier arpeggio fingerings for these chord types are best
> >discovered as subsets of the easier scale fingerings for the major and
> >minor scales.
> If by "easier scale fingering" you mean one that stays in position and avoids
> stretches, I wonder if this is really true. I know I had a lot of difficulty
> learning (and fingering) all the seventh chord arps in the major scale that way
> - after all, each one has a different fingering!
Well I am no position to contradict your own experience but with the
large number of students I've had over the years and with myself it has
been a good way to start.
You really have/had trouble playing the diatonic 7th chord arps from the
C major scale in Pos VII?
For me it was easy to start playing over changes by moving major scale
fingerings around the fretboard key by key. I was not making real strong
lines but i was following the form of the tune. Of course I had to start
with the major scale fingerings that I was the most comfortable with
which, at that time, just happened to be the 5 fingerings in the 1st
Leavitt book, 2 of which were much easier than the others because they
do not involve any finger strteches.
For lots of kids these days the 3 notes per string fingerings seem to be
a good place to get started as far as scales are concerned. But for the
types of chord tone soloing I am advocating below they are not as easy
to work with conceptually as the strict position-style fingerings.
If I was playing Recordame, for example, I could use the the G major
fingering at Pos VII for the A dorian section. Then at the switch to C
dorian I just moved the same finger pattern to Pos X where I could stay
right up until the Bbmaj7 chord. For the stuff in Ab (Bbm7 Eb7 Ab7) I
could use the same pattern in Pos VIII, etc. etc.
So I WAS learning to follow the progression and learning to follow the
progression while you're soloing is the first stumbling block for most
kids coming out of rock because they are used to using one scale, the
blues scale, for everything all the time. That tune changes scales in
some places that are pretty tricky to feel "naturally" so you have to
learn to time yourself bar by bar, chord by chord, even when you're just
using the major scale. The same approach works for Joy Spring, Scapple,
Ornithology or whatever.
The next step to getting stronger lines was to isolate the chord tones,
i.e. the arpeggios, for the individual chords from within each of those
scale fingerings. After that, and this is the hard part, you have to
start looking for ways to use voice leading in your lines, i.e. you have
to study the ways that one chord tone on the first chord moves into the
nearest chord tone on the next chord. And this is not so easy to see
while using a series of moving position type scale fingerings. For this
you need to look at other things like playing all those scales and
arpeggios in one position or on one string or on pairs of strings or on
groups of strings or in block chords.
But definitely, in my experience, scales, especially the easier scale
fingerings, are the place to *start*.
> I suspect that if I'd started
> out with the approach I learned from Garrison (see my reply to Steve) it would
> probably have been easier - it certainly wouldn't have been any more difficult.
> By shifting around the fretboard you're able to recycle your fingerings,
> plus you learn the fretboard much more thoroughly. Plus the phrasing is more
> dynamic, more jazzy. For one thing it lets you get into sweep fingerings, so
> you can introduce a bit of velocity right from the start.
>
> Garrison hinted that once we had got this technique down and learned the
> phrasing we could always go back to our old fingering styles (the "logical"
> ones we'd all learned from the books). I wonder if I shall, though...
>
> For those who don't know Garrison I should perhaps point out that he
> deliberately harks back to the greats of the 50s, Wes, Grant, Kenny etc. While
> some of them may not have had the chops of today's players, I don't think
> anyone can deny that they had groove, feel, soul, and that came from their
> phrasing, which in turn came from their fingering. And I guess that's my bag
> too..
>
> -Keith
--
Mike Archbold
start off with scales, since you at least don't have to shift completely every chord change to something else if
you have several chords in sequence that fall under the same key. It's not going to sound terrific but at least
the player can get a somewhat easier start. Particularly for a rock guitar player who is used to a scale
approach over a set of chords -- this idea is familiar to him/her. I completely understand your stance on
arpeggios but in my view it is not an either/or situation. Mike
Jim Kroger
Thanks very much Dan, I'll look again. By the way, your CDs arrived
today (2!!). I've just popped in Vibetones, it sounds wonderful. I've
been playing Green Dophin Street on my computer in continuous cycle
while I've been writing the last few days. I think I'm getting close
to being able to replay all the solos in my head. And funny, I don't
get tired of it....though how much attention I focus on it varies with
the difficulty of the writing. This is going to be a good model to
study because I really like the solos. I really appreciate you
sending the CDs, the sound from my computer is not good but I'm
setting up my good home hifi system finally and the CD is going to
sound very nice on there. Also looking forward to hearing the second
CD. Keep up the great work and let me (us) know when the next one is
availalbe...
Thanks again,
Jim
Keith Freeman
>stick shift instead of an automatic.
Over here in Europe (I'm in Holland, and learned to drive in England) the
vast majority of people start learning to drive on cars with manual gearboxes!
-Keith
Keith Freeman
>starting with arpeggios and expanding them to "chord scales" (using
>Joe Pass terminology rather than the mode terminology) rather than the
>other way around.
Glad to hear someone has tried this and found it useful.
Another point that occurs to me while experimenting with my new-found shifting
arpeggios is that not having a fixed scale pattern in your head makes it easier
to experiment with other ways of filling in the spaces between the arpeggio
notes (chromatic notes, neighbour notes). Whereas I have to 'force' myself to
play e.g. a major 2 in a half-dim arpeggio (D E F Ab rather than D Eb F Ab) if
I'm playing in position, if I'm thinking the arpeggio shape and using slides
with the first and third finger it comes much more naturally.
-Keith
Keith Freeman
>only have the roote, 3rd, 5th, and 7th - it can go on to the 13th just like
>a scale.
Yes, the arpeggios I'm working with often go up to the 13th.
-Keith
Dan Adler
> Thanks very much Dan, I'll look again. By the way, your CDs arrived
> today (2!!). I've just popped in Vibetones, it sounds wonderful. I've
> been playing Green Dophin Street on my computer in continuous cycle
> while I've been writing the last few days. I think I'm getting close
> to being able to replay all the solos in my head. And funny, I don't
> get tired of it....though how much attention I focus on it varies with
> the difficulty of the writing. This is going to be a good model to
> study because I really like the solos. I really appreciate you
> sending the CDs, the sound from my computer is not good but I'm
> setting up my good home hifi system finally and the CD is going to
> sound very nice on there. Also looking forward to hearing the second
> CD. Keep up the great work and let me (us) know when the next one is
> availalbe...
Jim,
Thanks! It's very flattering to know that someone has more than a
passing interest in something I played. Usually, I don't think of my
playing as being on a memorable level, but, hey, I'm not going to
argue with you :)
Mike Archbold
Look fellas, if you really want to start out with arpeggios, have at it! I don't care really. Sorry my stick-
shift analogy wasn't any good. If you want to make a kid learn 31 arpeggios in each position on the neck for
All the Things You Are to get started, go right ahead. Never mind he's used to one pentatonic scale for the
whole solo if he comes from a rock background. Oh, hell no, here man, learn 31 arpeggios to get started. Who
am I to stop anybody. Sheesh, I hate newsgroups. It's like the Monty Python argument sketch of contradictions.
Keith Freeman
experiences and insights with the folks here. If something useful comes out of
it, great, if not, never mind.
>If you want to make a kid learn 31 arpeggios in
>each position on the neck for
>All the Things You Are to get started
Well, I don't think I would start with All The Things, I'd choose something
easier for a beginner. By the time we tackled that song I'd want the student to
be able to see the triads within the arpeggios: there are a lot of common
triads between Fm7 Bbm7 Eb7 (that's just Bbm7 with an optional resolution from
Ab to G) and Abmaj7. Bear in mind that the way I learned this from Garrison
involves seeing the triads on the fingerboard as I go up the arpeggio (for
instance in Bbm7: Db, Fm, Ab).
> Never mind he's used to
>one pentatonic scale for the
>whole solo if he comes from a rock background.
Is it really relevant where someone is coming from if they want to play jazz
guitar? Aren't rock guitar and jazz guitar two different instruments? If a
golfer came to you to learn tennis, would you get him to adapt his golfing
swing to the new game? Seems to me that whatever kind of guitar someone has
played - rock, jazz, folk - jazz is going to mean starting from scratch. That's
certainly the way Garrison approached it at the summer course, and he had a lot
of rock guitarists to contend with. It looked to me like his approach was going
to bear fruit more quickly than the way taught in all of the books I've seen
(and I've a shelf full of them!).
-Keith
Lumpy
> ...Sheesh, I hate newsgroups. It's like the
> Monty Python argument sketch of contradictions.
This is opinions. Contradictions is down the hall.
lumpy
--
www.digitalcartography.com