Quantum Distributions and Text Studies: Temperature and Literature

Abstract

Rank-frequency distributions of words in texts have a number of similarities to particle distributions in statistical physics. This makes it possible to propose a new set of parameters by which texts that are examples of complex systems can be attributed. In particular, it is possible to show the relation of these parameters to the typological classification of languages by their level of analyticity and to illustrate the evolution within several language “lineages”. The proposed parameters are calculated from the frequency data of words rarely occurring in texts. It turns out that this part of а rank-frequency distribution is characterized by more stable behaviour, in contrast to the high- frequency lexicon certain other authors used in similar studies. One of the parameters used in classification is an analogue of temperature in physics. Its smaller values correspond to languages with a higher level of analyticity (less developed word inflection, replaced by a greater number of auxiliary words and a fixed word order in sentences). The proportion of rarely used vocabulary in such languages is different from languages in which inflection is well developed. Our approach is demonstrated using the example of translations of Antoine de Saint-Exupéry’s novella The Little Prince into nearly forty languages and various translations of the Gospel of John. The latter are used to study the historical development of language given that this religious text was translated in several different centuries. The results indicate new dimensions of previously known concepts. They are considered in the broader context of the linguo- statistical and linguo-philosophical achievements of Wilhelm von Humboldt, Morris Swadesh, Joseph Greenberg, Gabriel Altmann, Reinhard Köhler, and other linguists.