Article published In:
Languages in Contrast: Online-First Articles

Corpus-based contrastive studies and AI-generated translations

 | University of Oslo, Norway
This article addresses the issue of using AI-generated translations to perform contrastive analysis. The aim is to establish whether bidirectional translation corpora (and by extension human translators) have become superfluous, given the success of AI-generated translations. To investigate this, results from a previous study of English bring and Norwegian bringe based on bidirectional data are compared with results from a study based on AI-generated data. The AI-generated translations show a markedly higher Mutual Correspondence between the verbs than the human translations. This AI-translation effect may give an inaccurate picture of how equivalent the verbs really are. In cases where AI and human translations deviate, the latter are characterised by semantically more specific verbs. The study highlights some features only available in bidirectional corpora, including the possibility of taking individual variation into account. The findings suggest that bidirectional corpora (and human translators) still have a role to play in contrastive studies.
Keywords: AI-generated translations, GPTese, cognates, Mutual Correspondence, English/Norwegian
Article outline
  • 1.Introduction
  • 2.Bring vs. bringe in the English-Norwegian Parallel Corpus
  • 3.AI-generated translation
  • 4.Procedure of generating translations with GPT-4
  • 5.Contrastive analysis of bring and bringe based on AI-generated vs. human translations
    • 5.1Quantitative overview
    • 5.2Quantitative and qualitative analysis of patterns of use
      • 5.2.1The ditransitive pattern
      • 5.2.2The monotransitive pattern
      • 5.2.3The complex transitive pattern
      • 5.2.4Fixed phrases
  • 6.Limitations of the study
  • 7.Discussion
  • 8.Concluding remarks
  • Acknowledgements
  • Notes
  • References
  • Corpus
Published online: 20 September 2024
DOI logo
https://doi.org/10.1075/lic.00051.ebe
References (32)
References
Ahrenberg, L. 2017. Comparing Machine Translation and Human Translation: A Case Study. Proceedings of the Workshop Human-Informed Translation and Interpreting Technology (HiT-IT). Varna, Bulgaria, 7 September 2017. Association for Computational Linguistics. 21–28. DOI logoGoogle Scholar
Altenberg, B. 1999. Adverbial connectors in English and Swedish: Semantic and lexical correspondences. In Out of Corpora: Studies in Honour of Stig Johansson, H. Hasselgård and S. Oksefjell (eds), 249–268. Amsterdam/Atlanta: Rodopi. DOI logoGoogle Scholar
Baker, M. 1993. Corpus linguistics and translation studies. Implications and applications. In Text and Technology: In Honour of John Sinclair, M. Baker, G. Francis and E. Tognini-Bonelli (eds), 233–250. Amsterdam/Philadelphia: John Benjamins Publishing Company. DOI logoGoogle Scholar
2004. A corpus-based view of similarity and difference in translation. International Journal of Corpus Linguistics 9(2):167–194. DOI logoGoogle Scholar
Bjorvand, H. and Lindeman, F. O. 2007. Våre arveord. Etymologisk ordbok. Oslo: Novus.Google Scholar
Bokmålsordboka. 2024. The Language Council of Norway and the University of Bergen. Available at [URL] [last accessed 15 January 2024].
Brahmbhatt, A. 2023. GPT-3.5 vs GPT-4: an in-depth analysis of OpenAI’s language models. Available at [URL] [last accessed 5 April 2024].
Brown, P. F., Della Pietra, S. A., Della Pietra, V. J. and Mercer, R. L. 1990. The mathematics of Statistical Machine Translation: Parameter Estimation. Computational Linguistics 19(2), 263–311.Google Scholar
Cambridge Dictionary. 2024. English Dictionary Online. Cambridge University Press. Available at [URL] [last accessed 15 January 2024].
Ebeling, J. 1998. The Translation Corpus Explorer: A browser for parallel texts. In Corpora and Cross-linguistic Research: Theory, Method and Case Studies, S. Johansson and S. Oksefjell (eds), 101–112. Amsterdam: Rodopi. DOI logoGoogle Scholar
Ebeling, J. and Ebeling, S. O. 2015. An English-Norwegian contrastive analysis of downtoners, more or less. Nordic Journal of English Studies (NJES) 14 (1): 62–89. DOI logoGoogle Scholar
Ebeling, S. O. 2017. Bringing home the bacon! A contrastive study of the cognates bring/bringe in English and Norwegian. Kalbotyra 7001: 104–126. DOI logoGoogle Scholar
Gellerstam, M. 1986. Translationese in Swedish novels translated from English. In Translation Studies in Scandinavia, L. Wollin and H. Linquist (eds), 88–95. Lund: CWK Gleerup.Google Scholar
Hasselgård, H. 2010. Contrastive analysis / contrastive linguistics. In The Routledge Linguistics Encyclopedia, Third Edition, K. Malmkjær (ed.), 98–101. London: Routledge.Google Scholar
Hendy, A., Abdelrehim, M., Sharaf, A., Raunak, V., Gabr, M., Matsushita, H., Kim, Y. J., Afify, M. and Awadalla, H. 2023. How good are GPT models at machine translation? A comprehensive evaluation. DOI logoAvailable at [URL] [last accessed 8 April 2024].
House, J. 2023. Translation: The Basics, 2nd ed. London: Routledge. DOI logoGoogle Scholar
Hutchins, W. J. 2001. Machine translation over fifty years. Histoire Épistémologie Langage 23(1), 7–31. DOI logoGoogle Scholar
Johansson, S. 2007. Seeing Through Multilingual Corpora: On the Use of Corpora in Contrastive Studies. Amsterdam/Philadelphia: John Benjamins. DOI logoGoogle Scholar
Kanade, V. 2023. What is ChatGPT? Characteristics, uses, and alternatives. Available at [URL] [last accessed 4 April 2024].
Klaudy, K. and Károly, K. 2005. Implicitation in translation: Empirical evidence for operational asymmetry in translation. Across Languages and Cultures 6(1), 13–28. DOI logoGoogle Scholar
Koehn, P. 2020. Neural Machine Translation. Cambridge: Cambridge University Press. DOI logoGoogle Scholar
Mandal, S. A. 2019. Evolution of Machine Translation. Towards Data Science. Available at [URL] [last accessed 13 July 2024].
Marr, B. 2023. The Top 10 Limitations of ChatGPT. Available at [URL] [last accessed 5 April 2024].
Muftah, M. 2022. Machine vs. human translation: A new reality or a threat to professional Arabic-English translators. PSU Research Review, Emerald Publishing Limited. Available at [URL] [last accessed 5 April 2024].
NAOB. 2024. Det Norske Akademis Ordbok. Det Norske Akademi for Språk og Litteratur. Available at [URL] [last accessed 15 January 2024].
OED. 2017 / 2024. Oxford English Dictionary. Oxford University Press. Available at [URL] [accessed 25 July 2017 / 15 January 2024].
Quirk, R., Greenbaum, S., Leech, G. and Svartvik, J. 1985. A Comprehensive Grammar of the English Language. London: Longman.Google Scholar
Sinclair, J. 1999. A way with common words. In Out of Corpora: Studies in Honour of Stig Johansson, H. Hasselgård and S. Oksefjell (eds), 157–179. Amsterdam: Rodopi. DOI logoGoogle Scholar
Summa Linguæ. 2021. Rule-Based vs. Statistical vs. Neural Machine Translation. Available at [URL] [last accessed 4 April 2024].
Vanmassenhove, E., Shterionov, D. and Gwilliam, M. 2021. Machine translationese: Effects of algorithmic bias on linguistic complexity in machine translation. Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, 2203–2213. Available at [URL] [last accessed 4 April 2024]. DOI logo
Wikipedia: The Free Encyclopedia. 2024. GPT-4. Available at [URL] [last accessed 13 July 2024].
Corpus
English-Norwegian Parallel Corpus (1994–1997), Dept. of British and American Studies, University of Oslo. Compiled by Stig Johansson (project leader), Knut Hofland (project leader), Jarle Ebeling (research assistant), Signe Oksefjell (research assistant). Available at [URL] [last accessed 15 January 2024].