> If the truth isn't somewhere in the middle, then by definition it must be on one of the two extreme edges

In one dimension. If both sides are fundamentally wrong, the middle is probably also mischaracterised.

Pro-and anti-phlogiston theorists [1] weren’t validated by a little phlogiston. They were superseded by oxygen theory.

[1] https://en.wikipedia.org/wiki/Phlogiston_theory

I'm no electrical engineer. But I know it's crucial to not let the phlogiston out of the wires.

> "In one dimension."

Not even in one dimension. But that is obviously also a categorization/composition issue and therefore subject to other, potentially fallacious and accordingly named, pitfalls.

> If the truth isn't somewhere in the middle, then by definition it must be on one of the two extreme edges.

Just because accurate results aren't to be found "somewhere (unspecific!) in the middle" doesn't mean that one a) finds them precisely in the (extreme) edges, or fringes, and b) that the middle is completely excluded, especially in the analysis and comparison of dynamic systems (e. g. macroeconomic analysis).

Another day, another infinite-dimensional Hilbert space.