Solution Space

also Restricted Solution Space


The set of answers a problem permits; the tighter its constraints, the more candidate answers can be ruled out as errors trivially.

A problem’s solution space is the set of answers it allows. CF treats restrictions on that space not as mere bookkeeping but as error-detection mechanisms: every constraint that excludes a region of the space converts the answers in that region into immediately detectable errors. Saying “the answer can’t lie between 5 and 50” means any value there is wrong by inspection; saying “the answer must be a whole number” makes any fraction a spotted error, and even enables cheap correction by rounding at each step so small errors never compound.

CF’s distinctive move is to treat shrinking the solution space as a deliberate strategy for reliable thinking. The smaller the space, the more of it is ruled out for free, leaving only a handful of candidates to scrutinize. The limiting case is a binary question — yes or no — where almost everything is excluded and detection of wrong answers becomes trivial. This is why yes-or-no questions are generally easier to reason about and get right. They are also less ambitious and individually less valuable, but CF endorses the small-steps tradeoff: ten easy, low-error steps beat one large risky leap. People stall by attacking a whole undivided problem at once instead of breaking it down.

This connects solution-space restriction to CF’s broader account of error correction and mechanistic thinking: useful constraints, like error bars, are valuable precisely because they make most mistakes cheap to catch.


See also

Referenced by


Sources

  1. Error Correction Math and Types Primary criticalfallibilism.com
  2. Most Factors Aren't Borderline Supporting criticalfallibilism.com
/term/solution-space/