“I don’t always know, but must choose what works best.”
Such as, how reason extracted from an understanding of the real world can assist in choices that are more accurate in their answers or conclusions. Reason is more reliable than, faith extracted from beliefs that do not connected to the understanding of the real world, like faith in prayer for healing over medical treatment or faith in some fable about the origin of everything, that doesn’t match the existing knowledge about the origin of everything.
This analysis of which to choose can be navigated with choosing that which is the most reliable.
Reliabilism: a general theory of knowledge.
Reliabilism strives to ascertain whether the general belief-forming process by which method the belief was formed and how reliable it is to produce a high ratio of true beliefs. Reliable standards for true beliefs commonly rely heavily on asking good questions and bing skeptical or reserved before establishing an answer.
How can know whether a process of true belief is reliable in general or not? Justificational status of a belief must somehow depend on the way the belief is caused.
Wishful thinking, gut feelings confused reasoning, hasty generalization or guessing are not proper ways of justification if they nether hold or require evidence. Their common feature is unreliability: they tend to produce false beliefs in a large proportion of the cases.
Critical thinking, logic, or good inductive/deductive reasoning are proper ways of justification if they relate to evidence. Their common feature is reliability: they tend to produce true beliefs in a large proportion of the cases. What do these processes have in common? They all reliable: most of the beliefs that each process produces are true.
Definition of justification: the definition of an (ultimate) epistemic term should rely upon only non-epistemic terms, otherwise the definition would be empty of circular. Epistemic terms: justified, knowledge, having good reasons, having evidences, demonstrated. Non-epistemic terms: belief, truth, cause. A reliable inference process provides justification to an output belief, if its input beliefs were themselves justified.
How could the justified ness of the inputs have arisen? By having been caused by earlier reliable processes. This chain must end in reliable processes having only non-doxastic inputs, such as perceptual inputs. Thus, justifiedness is a matter of a history of personal cognitive processes. This historical nature of justifiedness implied by reliabilism, and contrasts sharply with (internalist) foundationalism and co herentism. 1