Volume 6 - Issue 18 (2026-09-20)

Evaluating Accuracy of Machine Translation of Iraqi Satirical Political Discourse: A Case Study of Faiq Al-Sheikh Ali's Twitter Page

1- Researcher: Zainab AbdulKareem Dinar

2- Researcher: Afyaa Osama Abdullah

Department of Translation - College of Arts - Al-Iraqia University- Baghdad- Iraq

This study examines the accuracy of machine translation (MT) in translating Iraqi satirical political discourse into English through selected tweets from Faiq Al-Sheikh Ali’s official Twitter page. By comparing Twitter’s auto-translation feature (powered by neural machine translation engines) with expert human translations, this qualitative case study evaluates English outputs to determine meaning preservation, pragmatic accuracy, and cultural resonance. Evaluation focuses on semantic accuracy, retention of sarcasm, interpretation of localized political allegories, and dialectal nuances. The findings reveal a significant performance gap in translating figurative and dialectal language. While literal propositional meaning is often transferred, MT engines consistently fail to convey humor, irony, and hyper-local references inherent in Iraqi political discourse, often generating misleading pragmatics. Human intervention remains indispensable for translating culturally loaded and politically sensitive dialectal texts.