Abstract
Background. The Bengali language is replete with figurative expressions derived from animals, the meanings of which are often culturally opaque when rendered into English. This exploratory study investigates the manner in which Bengali-speaking university students evaluate three English rendering strategies for such expressions: direct metaphor, single simile, and detailed simile.
Methods. Eleven culturally salient animal items were administered as 33 questionnaire prompts to 78 Bengali-speaking, non-native English-speaking students at the University of Rajshahi, Bangladesh. Each participant was asked to rate every item on a four-point scale, indicating how acceptable they found each item, and to provide open-ended commentary. This was guided by Abdul-Raof's (2006) typology of figurative comparison, Pierini's (2007) structural model of the simile, and Lakoff and Johnson's (2008) Conceptual Metaphor Theory. Descriptive acceptability rates were computed for each strategy and animal. A chi-square test with Bonferroni-corrected pairwise comparisons was used to compare strategies at the aggregate level.
Results. The highest level of acceptability was demonstrated by detailed similes (66.3%; 95% CI: 63.1-69.4%), followed by direct metaphors (58.4%; 95% CI: 55.1-61.6%) and single similes (54.2%; 95% CI: 50.9-57.5%; χ² = 27.01, df = 2, p < 0.001, Cramer's V = 0.10). The utilization of detailed similes was found to be significantly more acceptable than the employment of direct metaphors or single similes. However, no significant differences were observed between the former and the latter. The effect sizes were found to be negligible, and analysis of individual cases indicated that acceptability was closely associated with the apparent degree of semantic overlap between the Bengali and English connotations of each animal, rather than being solely dependent on the form of the translation.
Conclusions. Within this Bengali-speaking university student sample, detailed similes were adjudged the most effective rendering strategy. However, translation form explained only a minor proportion of the observed variation in judgments. Apparent cross-cultural conceptual overlap and the specific wording of each stimulus appeared to be more significant factors. It is important to note that the stimuli varied in terms of semantic content and form. In addition, no target-culture English comparison group was included. Therefore, it would be prudent to interpret these findings as descriptive patterns in this specific sample, rather than as an established account of Bengali-English cross-cultural translation.