if a model was right, you wouldn't give up and call it merely useful.
This is only true if a correct model is useful, which is certainly not always true. Even further: the most correct model isn’t necessarily the most useful model even when it is useful.
If I remember correctly evolutionary simulations consistently show a fully accurate perception of the world is generally actively harmful even when there’s no associated resource cost. Autistic analyzers often have more accurate models of social dynamics but do worse at socializing. Blind optimism, undeserved confidence unfounded worries etc are all extremely useful, and moving to a more accurate view is less useful.
It’s possible to have wrong and useless models of course, but that’s the point of the adage.
p
This is only true if a correct model is useful, which is certainly not always true. Even further: the most correct model isn’t necessarily the most useful model even when it is useful.
If I remember correctly evolutionary simulations consistently show a fully accurate perception of the world is generally actively harmful even when there’s no associated resource cost. Autistic analyzers often have more accurate models of social dynamics but do worse at socializing. Blind optimism, undeserved confidence unfounded worries etc are all extremely useful, and moving to a more accurate view is less useful.
It’s possible to have wrong and useless models of course, but that’s the point of the adage.
More options
Context Copy link