AI ALIGNMENT FORUM
AF

Wikitags

Rationalization

Edited by Eliezer Yudkowsky, Ruby, Grognor, et al. last updated 13th Nov 2021

Rationalization is the act of finding reasons to believe what one has already decided they want to believe. It is a decidedly terrible way to arrive at true beliefs.

“Rationalization.” What a curious term. I would call it a wrong word. You cannot “rationalize” what is not already rational. It is as if “lying” were called “truthization.” – Rationalization

Rationality starts from evidence, and then crunches forward through belief updates, in order to output a probable conclusion. "Rationalization" starts from a conclusion, and then works backward to arrive at arguments apparently favoring that conclusion. Rationalization argues for a side already selected; rationality tries to choose between sides.

Rationalization can be conscious or unconscious. It can take on a blatant, conscious form, in which you are aware that you want a particular side to be correct and you deliberately compose arguments for only that side, looking over the evidence and consciously filtering which facts will be presented. Or it can occur at perceptual speeds, without conscious intent or conscious awareness.

Defeating rationalization - or even discovering rationalizations - is a lifelong battle for the aspiring rationalist.

See Also

  • Motivated skepticism, motivated cognition
  • Filtered evidence
  • Fake simplicity
  • Self-deception
  • Litany of Gendlin
  • Occam's Imaginary Razor
  • Hope, oops
Subscribe
2
Subscribe
2
Discussion0
Discussion0
Posts tagged Rationalization
23Understanding and avoiding value drift
TurnTrout
3y
7
Add Posts