You’re missing the point.
I am NOT questioning the general theory.
I’m speculating in what breaks DS spokes, which usually have considerably higher tension than the NDS spokes.
Either they break b/c of that particular spoke being undertensioned compared to the other DS spokes.
Or there is something else going on that keeps the NDS spokes from breaking.

And I think you’re wrong about ”not enough tension to fatigue”

Consider a spoke alternating between rattling around and barely load-carrying. Would that too fatigue?
Sure, a big proportional change in load is a big fatigue driver.
But what if that ”big proportional change” is between near zero and very little load?
As far as I remember, there is a lower limit to when fatigue can occur as well. If the cyclic load change is small enough, steel doesn’t fatigue.
That’s one of the frequently stated advantages to steel frames & forks. Designed and dimensioned ”right”, they’re expected to last ”forever”.