The Translation of Pashto Poetry into English: A Comparative Study of Human and Machine Translation

Abstract

The translation of Pashto poetry is challenging due to its rich sound pattern, cultural metaphors, and imagery. This study aims to compare the translations of selected Pashto poems by machine and human translators to analyze the effectiveness of translation and justice to the meaning of the source text. This study uses Dastjerdi et.’al model for qualitative comparative analysis of the poetry sample. It carries the analysis at two different levels: textual (rhyme, rhythm, lexis, alliterations) and extra-textual (cultural references, ideology, spiritual tone). The translations of ChatGPT were generated through specific prompts similar for all poems. The results of this study show that human translators have a tendency towards domestication, simplifying the cultural relevant terms for the target audience, which sometimes leads to the loss of poetic tone and ideological depth. On the other hand, ChatGPT's translations tend to preserve the cultural depth and essence of Pashto poetry as instructed. ChatGPT uses a foreignization strategy for translation of Pashto poetry to preserve cultural norms and depth. The study concludes that artificial intelligence chatbots, especially models like ChatGPT, show great potential in literary translation when guided accurately.