International Journal of Philology

Translation of modern German realities using DeepL

Svitlana Amelina, Rostyslav Tarasenko, Liying Shen
Download article
Abstract

The relevance of this research lies in the increasing role of machine and automatic translation systems and their potential use as auxiliary tools for translators. This study aimed to explore the capabilities requiring the translator to rely on background knowledge or conduct additional reference searches for accurate interpretation. Some accurate translations of realities were found in the DeepL system as transliterated matches. In certain cases, generalised versions of matches proved successful. However, realities from the past three years, despite being widely distributed across the Federal Republic of Germany, were translated incorrectly. For these, a descriptive translation approach is recommended. The results of this study can be utilised by translators to enhance the accuracy of translating texts containing cultural and linguistic realities of the DeepL translation system in rendering the realities of contemporary German language into Ukrainian. A variety of research methods were employed, including analysis and synthesis, comparative and descriptive approaches, and generalisation. The findings revealed that the accuracy of translating realities using the DeepL system varies across thematic groups. The distribution of thematic groups, along with the number and proportion of accurately translated realities, is summarised in the accompanying tables. The highest accuracy was observed in the thematic group “Geographical Names”, while the lowest (zero) was found in the group “Ironic Everyday Realities”. The study determined that the most successful translations occurred with complex nouns (composites) when their meanings directly corresponded to the combined meanings of their components. Conversely, translation results were unsatisfactory when the meanings deviated from this direct correspondence. Abbreviations were left untranslated, 

Keywords

machine translation; thematic groups; machine translation systems; German-language realities; adequate translation

Suggested citation
Amelina, S., Tarasenko, R., & Shen, L. (2024). Translation of modern German realities using DeepL. International Journal of Philology, 15(3), 69-82. https://doi.org/10.31548/philolog/3.2024.69
References

[1] Banitz, B. (2020). Machine translation: A critical look at the performance of rule-based and statistical machine translation. Translation Notes, 40(1), 54-71. doi: 10.5007/2175-7968.2020v40n1p54.

[2] Berger, D. (2018). DeepL Pro: New subscription service for professional translators, companies and developers. Retrieved from http://surl.li/alzyep.

[3] Doherty, S. (2016). The impact of translation technologies on the process and product of translation. International Journal of Communication, 10, 947-969.

[4] Kabachenko, І., & Malashevska, І. (2024). The difficulties of translating German realities. Transcarpathian Philological Studies, 2(33), 108-112. doi: 10.32782/tps2663-4880/2024.33.2.19.

[5] Kirsche, H.-J., & Meinung, L. (1984). The German reichsbahn from A to Z. Berlin: Transpress.

[6] Kostikova, I., Shevchenko, A., Holubnycha, L., Popova, N., & Budianska, V. (2019). Use of machine translation technology for understanding scientific and technical texts. Journal of Theoretical and Applied Information Technology (JATIT), 97(4), 1355-1366.

[7] Lee, D.T.S., Zhou, Zh.Q., & Tse, T.H. (2020). Metamorphic robustness testing of Google Translate (Technical report HKUCS-TR20-03). In ICSEW’20: Proceedings of the IEEE/ACM 42nd international conference on software engineering workshops (pp. 388-395). Hong Kong: Hong Kong University of Science and Technology. doi: 10.1145/3387940.3391484.

[8] Li, Z., Xiao, L., Lin, R., & Xiao, Zh. (2023). Metamorphic robustness testing for DeepL translation. Journal of Physics (Conference Series), 2456, article number 012018. doi: 10.1088/1742-6596/2456/1/012018.

[9] Maslianko, P.P., & Sielskyi, E.P. (2021). Method of system engineering of neural machine translation systems. KPI Science News, 2, 46-55. doi: 10.20535/kpisn.2021.2.236939.

[10] Mediavilla, S. (2024). An analysis of Google Translate and DeepL translation of source text typographical errors in the economic and legal fields. Revista de Llengua i Dret, 81, 88-105. doi: 10.58992/rld.i81.2024.4188.

[11] Mukataeva, Ya. (2022). Realia as a type of non-equivalent vocabulary and their classification. Scientific Bulletin of the International Humanitarian University. Series: Philology, 55, 174-177. doi: 10.32841/2409-1154.2022.55.40.

[12] Olkhovska, A., & Lavrenova, M. (2021). An experimental study of the impact of using machine translation systems on the quality of translation of texts in the field of law. Bulletin of V.N. Karazin Kharkiv National University. Series: Foreign Philology, 94, 74-78. doi: 10.26565/2227-8877-2021-94-11.

[13] Pym, A., & Torres-Simón, E. (2021). Is automation changing the translation profession?. International Journal of the Sociology of Language, 270, 39-57. doi: 10.1515/ijsl-2020-0015.

[14] Rodríguez de Céspedes, B. (2019). Translator education at a crossroads: The impact of automation. Living Languages, 64(1), 103-121. doi: 10.1515/les-2019-0005.

[15] Ryabova, K. (2024). Comparative analysis of machine translation techniques performed by Systran, O.translator, and M-translate. Scientific Notes. Series: Philological Sciences, 209, 315-321. doi: 10.32782/2522-4077-2024-209-47.

[16] Shilinska, I., & Kuziv, M. (2022). The use of machine translation systems in teaching translation of scientific and technical texts. Scientific Bulletin of the International Humanitarian University. Series: Philology, 53(2), 149-152. doi: 10.32841/2409-1154.2022.53-2.34.

[17] Sopylyuk, N., & Tsarenko, I. (2019). Comparative analysis of machine translation systems for economic discourse (using the example of French-Ukrainian language pairs). Bulletin of the Alfred Nobel University. Series: Philological Sciences, 2(18), 289-296. doi: 10.32342/2523-4463-2019-2-18-25.

[18] Sun, Z., Zhang, J.M., Harman, M., Papadakis, M., & Zhang, L. (2020). Automatic testing and improvement of machine translation. In G. Rothermel & D.-H. Bae (Eds.), ICSE ‘20: Proceedings of the ACM/IEEE 42nd international conference on software engineering (pp. 974-985). doi: 10.1145/3377811.3380420.

[19] Veselovska, G., & Radetska, S. (2021). Machine translation: Its typology, advantages and disadvantages. Current Issues in the Humanities, 35(7), 23-27. doi: 10.24919/2308-4863/35-7-4.

[20] Wang, H., Wu, H., He, Zh., Huang, L., & Church, K.W. (2022). Progress in machine translation. Engineering, 18, 143-153. doi: 10.1016/j.eng.2021.03.023.

[21] Wyndham, A. (2021). Inside DeepL: The world’s fastest-growing, most secretive machine translation company. Retrieved from https://surl.li/xzulzo.

[22] Yvette, G., Barry, H., & Koehn, Ph. (2019). Translationese in machine translation evaluation. Retrieved from https://arxiv.org/pdf/1906.09833.