Major English Equivalents of the Polish Impersonal -no/-to Construction in Translated Academic Prose: A Parallel Corpus-Driven Study of Research Article Abstracts

This study explores some of the factors underlying the choice from among the major English translation equivalents of the Polish impersonal construction with -no/-to verb forms in academic writing. Based on a parallel corpus of 487 Polish abstracts of research articles in linguistics and their English versions, Multiple Correspondence Analysis is deployed as an exploratory technique to identify potential associations among three types of constructions employed in the English target texts and four usage features related to the Polish source texts. Binary Logistic Regression is then used on a subset of the data with two most frequent constructions to determine the significance and strength of the correlations. The results indicate that while the agentless passive is the default structural equivalent of the Polish -no/-to construction, three factors may prompt the use of an active-voice structure with an inanimate subject, which was found to be the second major equivalent. These factors include: the presence of a locative adjunct referring to the study being summarized, the semantic type of the process denoted by the -no/-to verb, and the length of the NP functioning as the complement of the verb. The fourth usage feature under examination, namely the position of the complement in relation to the verb, seems to be less significantly correlated with the translation choice between the two major constructions in English.