Musical Idea 35: More Shadowing Techniques

06.07.2014 § 3 Comments

See previous shadowing ideas here:


Shadowing a rhythm makes it more even, that is, makes its intervals more similar. Because you take half from neighbors which are different and combine them, all the new intervals share something more with each other than they did before. Repeating the process over and over will approach the state where all intervals are the same (but never reach it). Look at our earlier example. Shadow 5, 6, 3 … It comes out to 5.5, 4.5, 4 … Shadow it again and it’s 5, 4.25, 4.75 …  Again 4.625, 4.5, 4.875 … Again 4.5625, 4.6875, 4.75 … Again 4.625, 4.71875, 4.65625 …you can see that we’re asymptotically approaching 14/3 = 4.667 because the sum of 2, 8, and 4 is 14 and that’s 3 total intervals.

As you approach maximal evenness, an alternating pattern emerges in the shadows — every other shadow is nearly identical. The shadows in-between are nearly identical, too, just offset halfway.

Repeatedly grinding away what is distinct about a rhythm until it’s indistinguishable from a regular pulse is one thing, but it may be far more interesting to begin with the ground away thing and then work your way back out from it. The differences in intervals of an extremely shadowed rhythm are imperceptibly small, but they are still proportional to each other and serve as a key, or seed, such that when you unshadow them they will gradually reveal what they had hidden inside them. A cyclic difference set may take some time to shadow down into a regular pulse, but the rewards of unshadowing into one from that nubbin could be grand.

One experiment I would try is dealing with multiple iterations of shadows together. If you only layer two together — one rhythm and its shadow — the resulting intervals will come in groups of two identical ones at a time. Selectively switching out strikes between one rhythm and its second shadow could have a cool effect (more so if you haven’t yet shadowed to the point of near-identifcalness). Layering two or three or four layers of shadows together, distinguished from each other (or not) by pitch or timbre (in this case, rather than intra-rhythm distinctness of the strikes, there would be inter-rhythm distinctness but each rhythm would have homogenous strikes) could be interesting. You could have a variable window of shadow visibility — say, varying from 2.5 layers of shadows to 3.333 layers of them, and that window sliding up and down from the original rhythm into a few shadows deep or maybe a few un-shadows shallow.



So far we’ve only been considering a shadow strike to come exactly halfway in-between strikes. But what if our hands don’t move quite the same speed out of a clap as they do into the next? In that case, they’d reach the maximum point furthest away from the clap, not at the centerpoint of the interval between claps like usual, but off from that centerpoint. Supposing we could move outward more quickly than we could move inward (because of resistance of the compressed air or something?) then the shadow rhythm would shift backwards, approaching the original rhythm.

The more quickly we can pull out of the clap, the more similar the shadow rhythm gets to the original. Note that this is not merely shifting the shadow around, since the shadow (usually) cannot be brought to match the rhythm through a translation; we’re moving the shadow within each interval, proportional to that interval. In other words, rather than the shadow occurring 0.5 of the way through each interval as usual, it may occur 0.14 of the way through, or 0.99 of the way through. Theoretically, if either the outward or the inward motion of our clapping became instantaneous, then our shadow rhythm would be identical to the original.

What these proportions create is a whole continuum of degrees of shadowness. I might call any shadow that’s not precisely halfway a lopshadow. Certainly you could get interesting variations on traditional shadowing by repeatedly shadowing by the same lopsided proportion, or maybe by alternating proportions that sum to 1 (like .35 then .65).

This also introduces another method of funkily gradually transitioning from a rhythm into its shadow (or back). And it’s a stealthier one. The diverging di-strikes can produce awesome effects, but aren’t exactly subtle about how they’re accomplishing it. This way you can smoothly evolve through shadow layers without temporarily doubling the number of strikes in your beat, too.


Lopshadow. (top) A traditional shadowing of the clave son. (bottom) A .75 lopshadowing, showing that the hand comes down much faster than it goes up, creating a rhythm that is halfway between the clave son and its shadow.

Pitch Shadowing

Shadowing ideas can also be applied to pitch intervals, too. A shadow tuning, that is, would be the set of pitches exactly halfway in-between its parent tuning’s pitches. Funky shadowing for pitch can bring out a whole other side of the effect really, as you can hear each element undergoing fluid evolution individually through time, rather than the entire thing fluidly but each element of it only by comparing across iterations.

What might be particularly interesting is to do both pitch and rhythm funky unshadowing at the same time. You’d begin with a boringly repetitive banging of chords with all the same intervals with no variation in duration between them. In bursts of chaos the chords and their durations would keep reorganizing themselves until finally a varied key and correlated rhythm was uncovered.


§ 3 Responses to Musical Idea 35: More Shadowing Techniques

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

What’s this?

You are currently reading Musical Idea 35: More Shadowing Techniques at cmloegcmluin.


%d bloggers like this: