Probabilistic intepretation of not purely referential text

The article describes the tools of creation of the not purely referential text and its interpretation algorithm. The algorithm is defined as a number of subsequent actions leading to interpretation results. The algorithm is analysed on the data of Stanislaw Lem’s “Cyberiada” that presents a not purely referential situation [Quain 2000] in the text as a stylistic device and a product of linguistic play, which includes non-referent use of names; intertext references which are not identified by the interpreter; and creation of the text by means of genres’ mash-up (a fairy-tale and philosophical narration). In the latter the not purely referential situation in the text depends on the mash-up of two genres’ balancers. The reader is forced to use many ways of text interpretation that increase the not purely reference degree of the text. The interpreter prefers their own ways of interpretation that form the semantic field of the text. Each reader can chose one or few meanings, triggering the probabilistic interpretation. At the same time a non-referent sign is a new word with a referent in a possible text world. The interpreter does not have any tools and rules for interpretation of the non-referent sign, thus only the probabilistic interpretation here can be applied. The algorithm of the analysis of the not purely referential text is a number of subsequent actions, including identification of the system of interpretative filters, and creation of a referent of a non-referent sign. The interpretative filters like phonetic associations, grammar markers, and sign context use are regarded in the paper. These tools are applied for the interpretation of non-referent names. The intertext references may be identified only if the interpreter reads the reference text. If he does not, there will be no intertext in the text. If the reader models the referent of the non-referent sign, if he can identify the intertext references, they will decrease the degree of not purely referential text.