It is as yet impossible to rigorously define terms like rationality (rational, rationalism) and reason (reasonable, reasonableness, reasoning) without ending up in a circular definition, as we do not yet have a good understanding of how intelligence – i.e. that quality which allows entities to reason – functions. (Obtaining this understanding is one of the major goals of the study of artificial intelligence, since an important part of recreating something artificially is to understand how it works in nature.)
It is much easier to identify specific examples that are not rational (and to explain why they are not) than it is to give a set of rules by which rationality can be clearly identified.
It is somewhat less easy, though still possible, to work out a set of guidelines by which rationality may be detected with a reasonably low error rate.
- The words "rational" and "rationality" are often misused to advocate solutions which serve some specific, obvious goal at the expense of less obvious goals
- Example: "Rational economic theory says that the minimum wage [or pollution laws, or safety inspections, etc.] harms businesses and reduces profits." If your goal is to improve conditions for a community, it is not rational to accept a solution which maximizes corporate profits at the expense of that community's well-being.
- The concept of rationalization is often conflated with "rationality".
- Rationality does not require dispassion, nor does rational dialogue require participants to suppress their feelings. While emotion can often interfere with rational judgement (and strongly-expressed sentiments can interfere with rational dialogue), a judgement that was made on a rational basis may lead to strong emotion as an entirely rational reaction. It may even be irrational not to experience emotion, depending on the nature of the question.
- rational (adjective form)
- rationality detection
- rationalization (falsely conflated with rationality)
- Aumann's agreement theorem
- The rule of succession is a rational, mathematical way of estimating the odds of something which has never been observed to occur.