TRANSLATING POETRY CONNECTED TO VISUAL ART: A CORPUS-BASED MULTIMODAL ANALYSIS OF FRENCH–ENGLISH EKPHRASTIC TRANSLATION

Abstract

This study explores how translators address the distinctive challenge of translating poems that are closely connected to visual artworks. When poetry and painting intersect, translation becomes more than a purely linguistic activity, it requires navigating the complex interplay between written language, visual imagery, and cross-cultural interpretation (Zheng, 2022). The research fills a notable gap in translation studies. While much has been written about poetry translation and, separately, about the relationship between verbal and visual art, few studies have systematically examined what occurs when translators must engage both at once (S Udhayakumar, 2018). This issue is increasingly relevant in contemporary artistic practice, where multimedia works combining text and image are common, yet translation theory provides little guidance for working with these hybrid forms (Dastjerdi et al., 2008). The study investigates several interconnected challenges. Translators of poem-painting combinations must preserve not only meaning and poetic form but also the visual-verbal relationships that create aesthetic coherence (Jiang, 2020). They must decide whether to reflect spatial features of the artwork, how to translate culturally specific references across text and image, and whether the presence of accompanying art allows greater linguistic freedom or imposes stricter fidelity. Using a corpus-based comparative approach, the research analyses French poems alongside the paintings that accompany them and their English translations. The methodology adapts existing translation assessment models to multimodal contexts, examining both linguistic shifts (semantic, syntactic, stylistic) and intersemiotic shifts (changes in the alignment between text and image). The corpus reveals consistent patterns in how translators balance these dual demands. Preliminary results indicate that translators tend to adopt source-oriented strategies, striving to preserve formal features and maintain strong correspondence between text and image. Translation shifts are relatively limited, with few semantic or syntactic alterations. Instead of encouraging creative deviation, the visual context appears to anchor translator choices, leading to versions that privilege fidelity to both language and artwork. This research offers valuable theoretical insights into intersemiotic translation and practical guidance for translators, publishers, and cultural institutions engaged with multimedia works. It also proposes pedagogical approaches for training translators to navigate the increasingly multimodal nature of contemporary culture.