Journals / Uluslararası Dil ve Çeviri Çalışmaları Dergisi / 2026 / Cilt: 5 Sayı: 1

Linguistic and Cultural Fidelity in AI-Driven Translation: A Comparative Analysis of DeepL and Human Translations in English-Turkish Literary Short Stories

Yapay Zekâ Destekli Çeviride Dilsel ve Kültürel Sadakat: İngilizce-Türkçe Edebi Kısa Öykülerde DeepL ve İnsan Çevirilerinin Karşılaştırmalı Analizi

Mustafa Dolmacı(SELÇUK ÜNİVERSİTESİ)0000-0002-2503-6072
Pages
92–109
Publication Date
2025-06-24
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

This study evaluates AI-driven translation tools in literary contexts and recommends hybrid models that combine AI supported machine translation efficiency with human cultural and creative experience.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:This study investigates the differences between DeepL’s machine translations and human translations of English-to-Turkish fictional short stories, utilizing Katharina Reiss’ translation criticism model to analyze linguistic components.
📊
Key Findings & Data:The study compares 381 sentences translated by DeepL with professional human translators from five literary works, identifying semantic, lexical, grammatical, stylistic, and nonsensical differences.
💡
Conclusion & Impact:This study evaluates AI-driven translation tools in literary contexts and recommends hybrid models that combine AI supported machine translation efficiency with human cultural and creative experience.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:makine çevirisi · insan çevirisi · kısa hikâye

Abstract

This study investigates the differences between DeepL’s machine translations and human translations of English-to-Turkish fictional short stories, utilizing Katharina Reiss’ translation criticism model to analyze linguistic components. The study compares 381 sentences translated by DeepL with professional human translators from five literary works, identifying semantic, lexical, grammatical, stylistic, and nonsensical differences. Lexical differences are most common (33.86%), followed by semantic (22.05%), near-synonymous (20.73%), grammatical (9.16%), nonsensical (8.40%), and stylistic (5.77%). DeepL translated simple, straightforward words well but struggled with culturally nuanced expressions, figurative language, and contextual adjustments. While human translators utilized target-culture-focused procedures to retain literary aesthetics, DeepL used literal, source-text-focused methods, resulting in complex narrative incoherencies. The investigation found 32 incorrect DeepL translations, highlighting the need for literary post-editing. The results underscore DeepL's limitations in conveying stylistic nuances and cultural allusions, reinforcing the indispensable function of human translators in preserving the artistic and emotional richness of literary texts. This study evaluates AI-driven translation tools in literary contexts and recommends hybrid models that combine AI supported machine translation efficiency with human cultural and creative experience.

Özet

Bu çalışmada, İngilizceden Türkçeye çevrilen kurgusal kısa öykülerde DeepL’ın makine çevirileri ile insan çevirileri arasındaki farkları, Katharina Reiss’in çeviri eleştirisi modeli kullanılarak dilbilimsel bileşenler açısından incelenmektedir. Beş edebi eserden alınan ve DeepL ile profesyonel insan çevirmenler tarafından çevrilen 381 cümle karşılaştırılmış ve farklılıklar belirlenmiştir. En yaygın farklılık türü sözcüksel farklılık olup (%33,86), ardından anlamsal farklılık (%22,05), yakın anlamlılık (%20,73), dilbilgisel farklılık (%9,16), mantıksız (yanlış çeviri) olma durumu (%8,40) ve üslup açısından farklılık (%5,77) olarak sıralanmıştır. DeepL, basit ve doğrudan kelimeleri başarıyla çevirirken kültürel nüanslı ifadeler, mecazi dil ve bağlamsal uyarlamalarda zorlanmıştır. İnsan çevirmenler hedef kültür odaklı yöntemlerle edebi estetiği korurken, DeepL kaynak metne bağlı kelimesi kelimesine çeviri stratejileri kullanmış, bu da karmaşık anlatısal tutarsızlıklara yol açmıştır. Çalışmada 32 hatalı DeepL çevirisi tespit edilerek edebi metinler için son düzenleme gerekliliği vurgulanmıştır. Sonuçlar, DeepL’in üslup inceliklerini ve kültürel göndermeleri aktarmadaki sınırlarını ortaya koyarken, edebi metinlerin sanatsal ve duygusal zenginliğini korumada insan çevirmenlerin vazgeçilmez rolünü pekiştirmektedir. Bu çalışma, yapay zekâ destekli çeviri araçlarını edebi bağlamda değerlendirmekte ve yapay zekânın verimliliği ile insanın kültürel-yaratıcı deneyimini birleştiren hibrit modeller önermektedir.

🔖

Cite This Article

Dolmacı, M. (2025). Linguistic and Cultural Fidelity in AI-Driven Translation: A Comparative Analysis of DeepL and Human Translations in English-Turkish Literary Short Stories. Uluslararası Dil ve Çeviri Çalışmaları Dergisi, Cilt: 5 Sayı: 1, 92–109. https://doi.org/10.63673/Lotus.1662868

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (16)

  1. Chopin, K., The story of an hour, Full Text - The Story of an Hour - Owl Eyes. Available at: https://www.owleyes.org/text/the-story-of-an-hour/read/chopins-short-story
  2. Chopin, K. (1894/2018). The story of an hour | Bir saatin öyküsü (A. Tunalı, Trans.). Mütercim. https://mutercumanblog.wordpress.com/2018/12/06/the-story-of-an-hour-bir-saatin-oykusu-1894/
  3. Deacon, A. and Wilde, O. (2015) The selfish giant. London: Red Fox. Gentzler, Edwin. Contemporary Translation Theories. London and New York:
  4. Guerberos-Arenas, A., & Toral, A. (2022). Creativity in Translation: Machine Translation As a Constraint for Literary Texts. Translation Spaces, 11(2), 184–212.
  5. Ibáñez Moreno, A., & Esther Domínguez Mora, M. (2025). Google Translate versus DeepL in Spanish to English translation of Don Quixote. Translation and Translanguaging in Multilingual Contexts, 11(1), 65-87.
  6. Jackson, S. (1991) Charles. Mankato, MN: Creative Education.
  7. Ke, Z. (2024) ‘Comparison between human translation and machine translation take Lolita as an example’, Communications in Humanities Research, 32(1), pp. 58–64.
  8. Karabayeva, Irina & Kalizhanova, Anna. (2024). Evaluating machine translation of literature through rhetorical analysis. Journal of Translation and Language Studies. 5. 1-9. 10.48185/jtls.v5i1.962.
  9. Kolb, W., Dressler, W. U., & Mattiello, E. (2023). Human and machine translation of occasionalisms in literary texts: Johann Nestroy’s Der Talisman and its English translations. Target, 35(4), 540-572.
  10. Noriega-Santiáñez, Laura & Corpas Pastor, Gloria. (2023). Machine vs Human Translation of Formal Neologisms in Literature: Exploring E-tools and Creativity in Students. Tradumàtica tecnologies de la traducció. 233-264. 10.5565/rev/tradumatica.338.
  11. Reiss K. (2000) Translation Criticism - The Potential & Limitations. Categoriesand Criteria for Translation Quality Assessment. Transl. by E.F.Rhodes, Manchester/New York, St. Jerome/American Bible Society.
  12. Rivera-Trigueros, I. (2022). Machine translation systems and quality assessment: A systematic review. Language Resources and Evaluation, 56, 593–619. https://doi.org/10.1007/s10579-021-09537-5
  13. Sarhsian, Evelina & Zinchenko, Olha. (2024). Utilizing Machine Translation Technology for Reproducing Short Prose Literary Texts. Studia Linguistica. 101-110. 10.17721/StudLing2024.24.101-110.
  14. Short stories: The open window by Saki (no date) East of the Web. Available at: https://www.eastoftheweb.com/short-stories/UBooks/OpeWin.shtml
  15. Thriven, C. 2002. Cultural Elements in Translation. The Indian Perspective. Translation Journal. Volume 6, No. 1 January 2002
  16. Toral, A., & Way, A. (2015). Machine-assisted Translation of Literary Text. Translation Spaces, 4(2), 240–267. Wilde, O. (1986) Bencil Dev. Istanbul: Esin Yayinevi.