Banjo revelation: training the hand for arpeggios

An unexpected revelation (maybe) about the banjo's picking hand. The first basic "forward-roll" exercise that you find in books is pretty straightforward, focusing on the first three strings, thus:

A little weird to someone trained in flatpicking, but I do find it "physiologically logical" and just need to train myself.

Now...at some point you get to an exercise involving the fifth string, and that's where things started to get interesting for me. Note this exercise:

(all open strings...
T means to play with thumb,
I with index finger,
and M with middle finger)

This plays havoc with something I have taken for granted: that strings constantly ascend as they get farther away from you. My thumb wants to be "responsible" for the bass note in the arpeggio, and while the thumb does play the first note in the arpeggio (which is the bass note), it does not play that note again in the same measure. Amazing how hardwired that concept can be! I suspect I must now be feeling like all those sweep-picking guitarists who have such a hard time learning alternate picking in odd time signatures. (Since I learned alternate picking from the beginning--thank you Guitar Craft--that never bothered me.)

Of interest here, though, is how that fifth-string exercise relates to this one:

Aside from trying to decouple my brain from what it already "knows", I found myself wondering why the second note in the second exercise is picked by the index finger, but the same note in the third exercise is picked by the middle finger. There's got to be a reason for that!

That reason occurred to me when I tried alternating between the two patterns. Ding! I think it's a way of signalling to your hand how it's going to spread for the arpeggio, at least initially. In the first fifth string exercise, your thumb and index finger maintain a constant distance from one another; the whole hand simply moves from "thumb-over-third-string" to "thumb-over-fifth-string" and the middle finger picks up on the first. The three fingers are equidistant from one another, and the whole hand could stay locked that way if desired. In the second fifth-string exercise, the hand spreads after the first note, with the middle finger "staying" on the first string and the index finger picking up the second while the thumb travels all the way out to the fifth. Here the fingers are not equidistant from one another. Does the first upstroke note signal to the hand what to do next?

I'm not claiming to "know" here--among other things it's quite possible that different permutations of the exercise might nullify this concept--but I do know that when I started thinking of it in this way, I immediately got both faster and cleaner. Call it a working theory, and something to pay attention to!

Today's work seemed to confirm as well that the "logic" of the right hand includes assigning the thumb the task of the "1" count. Most banjo music I suppose uses enough notes that this is not ever really a problem, but I am at least theoretically interested in high-speed odd time signatures, and it will be interesting to see how the logic applies. (With flatpicking, the subject of the 1-count always being a downstroke can get contentious, and in Guitar Craft we learn--conspicuously--not to assume that the 1 will automatically be a downstroke. Apparently this was instilled so well that I am scanning the horizon early to see it coming with the banjo!)

Okay, let's try that again later and see if it holds...