Information Content of Words in Texts
In a previous study, Andersen (2002a) postulated that information of words in texts has to be examined from two aspects yielding two distinct measures called “speaker’s information content” (SIC) and “hearer’s information content” (HIC) which may differ in amount, i.e. SIC ≠ HIC, and cannot always be mechanically evaluated from the frequencies of words in the text. The idea is derived from the Fitts—Garner controversy in mathematical psychology (cf. Fitts et al. 1956; Garner 1962, 1970; Garner, Hake 1951; Coombs, Dawes, Tversky 1970; Evans 1967; Attneave 1959, 1968). Obviously, the problem is quite told but has not penetrated into linguistics as yet.
KeywordsInformation Content Word Length Geometric Distribution Latent Length Latent Entity
Unable to display preview. Download preview PDF.