Definitions of knowledge try to determine the essential features of knowledge. Closely related terms are conception of knowledge, theory of knowledge, and analysis of knowledge. Some general features of knowledge are widely accepted among philosophers, for example, that it constitutes a cognitive success or an epistemic contact with reality and that propositional knowledge involves true belief. Most definitions of knowledge in analytic philosophy focus on propositional knowledge or knowledge-that, as in knowing that Dave is at home, in contrast to knowledge-how (know-how) expressing practical competence. However, despite the intense study of knowledge in epistemology, the disagreements about its precise nature are still both numerous and deep. Some of those disagreements arise from the fact that different theorists have different goals in mind: some try to provide a practically useful definition by delineating its most salient feature or features, while others aim at a theoretically precise definition of its necessary and sufficient conditions. Further disputes are caused by methodological differences: some theorists start from abstract and general intuitions or hypotheses, others from concrete and specific cases, and still others from linguistic usage. Additional disagreements arise concerning the standards of knowledge: whether knowledge is something rare that demands very high standards, like infallibility, or whether it is something common that requires only the possession of some evidence.
One definition that many philosophers consider to be standard, and that has been discussed since ancient Greek philosophy, is justified true belief (JTB). This implies that knowledge is a mental state and that it is not possible to know something false. There is widespread agreement among analytic philosophers that knowledge is a form of true belief. The idea that justification is an additionally required component is due to the intuition that true beliefs based on superstition, lucky guesses, or erroneous reasoning do not constitute knowledge. In this regard, knowledge is more than just being right about something. The source of most disagreements regarding the nature of knowledge concerns what more is needed. According to the standard philosophical definition, it is justification. The original account understands justification internalistically as another mental state of the person, like a perceptual experience, a memory, or a second belief. This additional mental state supports the known proposition and constitutes a reason or evidence for it. However, some modern versions of the standard philosophical definition use an externalistic conception of justification instead. Many such views affirm that a belief is justified if it was produced in the right way, for example, by a reliable cognitive process.
The justified-true-belief definition of knowledge came under severe criticism in the second half of the 20th century, mainly due to a series of counterexamples given by Edmund Gettier. Most of these examples aim to illustrate cases in which a justified true belief does not amount to knowledge because its justification is not relevant to its truth. This is often termed epistemic luck since it is just a fortuitous coincidence that the justified belief is also true. A few epistemologists have concluded from these counterexamples that the JTB definition of knowledge is deeply flawed and have sought a radical reconception of knowledge. However, many theorists still agree that the JTB definition is on the right track and have proposed more moderate responses to deal with the suggested counterexamples. Some hold that modifying one's conception of justification is sufficient to avoid them. Another approach is to include an additional requirement besides justification. On this view, being a justified true belief is a necessary but not a sufficient condition of knowledge. A great variety of such criteria has been suggested. They usually manage to avoid many of the known counterexamples but they often fall prey to newly proposed cases. It has been argued that, in order to circumvent all Gettier cases, the additional criterion needs to exclude epistemic luck altogether. However, this may require the stipulation of a very high standard of knowledge: that nothing less than infallibility is needed to exclude all forms of luck. The defeasibility theory of knowledge is one example of a definition based on a fourth criterion besides justified true belief. The additional requirement is that there is no truth that would constitute a defeating reason of the belief if the person knew about it. Other alternatives to the JTB definition are reliabilism, which holds that knowledge has to be produced by reliable processes, causal theories, which require that the known fact caused the knowledge, and virtue theories, which identify knowledge with the manifestation of intellectual virtues.
Not all forms of knowledge are propositional, and various definitions of different forms of non-propositional knowledge have also been proposed. But among analytic philosophers this field of inquiry is less active and characterized by less controversy. Someone has practical knowledge or know-how if they possess the corresponding competence or ability. Knowledge by acquaintance constitutes a relation not to a proposition but to an object. It is defined as familiarity with its object based on direct perceptual experience of it.