Possibilities of Language Technology in Lexical Analyses of Canonical Hungarian Bible Translations

The paper tries to attempt an unorthodox analysis of contemporary Hungarian Bible translations by using the software Sketch Engine. An important condition for computer-assisted linguistic analysis is a sufficient amount of textual data. In the case of Bible translations, “sufficient quantity” is easily attainable, since in this case a single translation represents the whole corpus. The corpus in this case is a properly annotated text: six Hungarian Bible translations. The paper highlights the fact that the computer-based text analysis can still reveal new features of Bible translations that are not found in the linguistic and hermeneutical analyses. In addition to the various layers of analysis of language use, it focuses on data like type-token characteristics, unigrams, bigrams, possible collocations, terms, orthographic error types, and orthographic variations of the chosen translations. Analyses show new correlations between translations or prove the well-known connections between translations, denominations, and sacred language use. Data and statistics can be used for several purposes – here, to gain more knowledge about Hungarian Bible translations. Statistics do not influence the work of the translators, and no far-reaching conclusions can be drawn from them. But they do provide a basis for a more detailed interpretation of a kind that is not possible during human reading.