truthlikeness a term introduced by Karl Popper in 1960 to explicate the idea that one theory may have a better correspondence with reality, or be closer to the truth, or have more verisimilitude, than another theory. Truthlikeness, which combines truth with information content, has to be distinguished from probability, which increases with lack of content. Let T and F be the classes of all true and false sentences, respectively, and A and B deductively closed sets of sentences. According to Popper’s qualitative definition, A is more truthlike than B if and only if B 3 T 0 A 3 T and A 3 F 0 B 3 F, where one of these setinclusions is strict. In particular, when A and B are non-equivalent and both true, A is more truthlike than B if and only if A logically entails B. David Miller and Pavel Tichý proved in 1974 that Popper’s definition is not applicable to the comparison of false theories: if A is more truthlike than B, then A must be true. Since the mid-1970s, a new approach to truthlikeness has been based upon the concept of similarity: the degree of truthlikeness of a statement A depends on the distances from the states of affairs allowed by A to the true state. In Graham Oddie’s Likeness to Truth (1986), this dependence is expressed by the average function; in Ilkka Niiniluoto’s Truthlikeness (1987), by the weighted average of the minimum distance and the sum of all distances. The concept of verisimilitude is also used in the epistemic sense to express a rational evaluation of how close to the truth a theory appears to be on available evidence. See also CONFIRMATION, INFORMATION THEORY , INSTRUMENTALISM , PROBABILIT. I.N.