Ok, you finally forced me to go pull out my engineering handbook and study a bit. Here's it is in short:

For a flat plate, which a kayak rudder usually is, the lift generated by a change in the angle of attack falls off faster with a high aspect ratio then it does with a lower aspect ratio. Ie, the longer and skinnier the rudder, the worse it works for steering. The "burble point" [where lift abruptly drops] being about 38 degrees angle of attack for an aspect ratio of 1, while only 25 degrees for an aspect ratio of 9. This means a higher aspect ratio rudder is far more likely to not turn a boat, especially if the rudder is thrown hard over. Efficiency is great, but not if it means the loss of steering control as a result.

I also extrapolated the data out for a G-438 airfoil (symetrical). Roughly the same.

Yes, a high aspect ratio is more efficient then a lower aspect ratio, but (and it's a big important but) only when the angle of attack is zero. This high efficiency is significant for the keel, daggar board, skeg, leeboard or whatever straightline low attack angle device is being used. Not so as much for the rudder which works by not having an attack angle of zero.

This is bolstered by my observation of rudder designs and most of the design recommendations for them that I have seen from boat designers, who usually suggest an aspect ratio of between 1 and 3 for boat rudders.

> A high aspect ratio rudder will provide more efficient turning
> moment than a low aspect rudder. The main reason you don't see them
> on sailboats is structural. America's Cup boats tend to have high
> aspect keels and rudders, but they can afford the neccessary materials.
> It is generally a good idea to keep a rudder from sticking below the
> keel because, you don't want the rudder to break off. This problem
> is solved in kayaks (and some sailboats) by having the rudder kick
> up.