Research Article

Detecting the Human in the Machine: A Forensic Linguistic Approach to Distinguishing Human, AI, and Hybrid Authorship in Turkish

Volume: 20 Number: 2 October 19, 2026

Detecting the Human in the Machine: A Forensic Linguistic Approach to Distinguishing Human, AI, and Hybrid Authorship in Turkish

Abstract

The proliferation of large language models capable of producing fluent Turkish prose confronts linguistic scholarship with a problem inverse to its longstanding concern: rather than analysing texts of known human provenance, scholars now face the task of determining whether a given text was authored by a human writer, generated by a machine, or composed through some combination of the two. Existing detection systems based on perplexity and burstiness suffer from opacity, English-language bias, and an inability to handle hybrid composition. The present study advances a forensic linguistic alternative grounded in the presence/absence coding framework and Jaccard similarity coefficient developed by Grant (2013) and adapted to Turkish in Kocagül (2022). Drawing on a corpus of 336 argumentative texts produced by 56 participants across six task conditions, the analysis applies 61 active linguistic features distributed across lexical, syntactic, structural, AI-specific, and data-driven categories. Three findings emerge. AI-generated Turkish texts exhibit a coherent feature profile, most distinctively elevated semicolon and comma density, sentence-initial discourse markers, recurrent ‑Arak converb chaining, and formulaic closing vocabulary. The within-author Jaccard baseline (0.41) substantially exceeds all human-to-machine comparisons (0.22-0.30), while the AI-to-hybrid pairing reaches 0.52, confirming the persistence of the AI lexical profile across editorial intervention. In edited hybrid texts, 41 of 61 features (67%) remain shifted from the writers’ baseline toward the AI baseline, indicating that lexical and discourse-organisational features pass below the editor’s threshold of attention. The findings characterise an AI-typical linguistic profile in argumentative Turkish and contribute a Turkish-language case to a corpus linguistic and forensic literature whose principal benchmarks have until recently excluded the language.

Keywords

References

  1. Amasyali, M. F., & Diri, B. (2006). Automatic Turkish text categorization in terms of author, genre and gender. In C. Kop, G. Fliedl, H. C. Mayr, & E. Métais (Eds.), Natural Language Processing and Information Systems (NLDB 2006) (Lecture Notes in Computer Science 3999, pp. 221-226). Springer. https://doi.org/10.1007/11765448_22
  2. Argamon, S. (2018). Computational forensic authorship analysis: Promises and pitfalls. Language and Law / Linguagem e Direito, 5(2), 7-37.
  3. Berber Sardinha, T. (2024). AI-generated vs human-authored texts: A multidimensional comparison. Applied Corpus Linguistics, 4(1), 100083. https://doi.org/10.1016/j.acorp.2023.100083
  4. Bevendorff, J., Bonet Casals, X., Chulvi, B., Dementieva, D., Elnagar, A., Freitag, D., Fröbe, M., Korenčić, D., Mayerl, M., Mukherjee, A., Panchenko, A., Potthast, M., Rangel, F., Rosso, P., Smirnova, A., Stamatatos, E., Stein, B., Taulé, M., Ustalov, D., Wiegmann, M., & Zangerle, E. (2024). Overview of PAN 2024: Multi-author writing style analysis, multilingual text detoxification, oppositional thinking analysis, and generative AI authorship verification. In N. Goharian, N. Tonellotto, Y. He, A. Lipani, G. McDonald, C. Macdonald, & I. Ounis (Eds.), Advances in Information Retrieval (ECIR 2024) (Lecture Notes in Computer Science 14613, pp. 3-10). Springer. https://doi.org/10.1007/978-3-031-56072-9_1
  5. Canbay, P., Sezer, E. A., & Sever, H. (2020). Detection of stylometric writeprint from the Turkish texts. In 28th Signal Processing and Communications Applications Conference (SIU) (pp. 1-4). IEEE.
  6. Chun, J., & Elkins, K. (2023). The crisis of artificial intelligence: A new digital humanities curriculum for human-centred AI. International Journal of Humanities and Arts Computing, 17(2), 147-167. https://doi.org/10.3366/ijhac.2023.0310
  7. Çöltekin, Ç. (2010). A freely available morphological analyzer for Turkish. In Proceedings of the Seventh International Conference on Language Resources and Evaluation (LREC’10) (pp. 820-827). European Language Resources Association.
  8. COPE Council. (2023). COPE position: Authorship and AI (English). https://doi.org/10.24318/cCVRZBms

Details

Primary Language

English

Subjects

Linguistics (Other)

Journal Section

Research Article

Publication Date

October 19, 2026

Submission Date

May 22, 2026

Acceptance Date

October 18, 2026

Published in Issue

Year 2026 Volume: 20 Number: 2

APA
Kocagül, H. (2026). Detecting the Human in the Machine: A Forensic Linguistic Approach to Distinguishing Human, AI, and Hybrid Authorship in Turkish. Cankaya University Journal of Humanities and Social Sciences, 20(2), 579-594. https://doi.org/10.47777/cankujhss.1957142
AMA
1.Kocagül H. Detecting the Human in the Machine: A Forensic Linguistic Approach to Distinguishing Human, AI, and Hybrid Authorship in Turkish. CUJHSS. 2026;20(2):579-594. doi:10.47777/cankujhss.1957142
Chicago
Kocagül, Hülya. 2026. “Detecting the Human in the Machine: A Forensic Linguistic Approach to Distinguishing Human, AI, and Hybrid Authorship in Turkish”. Cankaya University Journal of Humanities and Social Sciences 20 (2): 579-94. https://doi.org/10.47777/cankujhss.1957142.
EndNote
Kocagül H (October 1, 2026) Detecting the Human in the Machine: A Forensic Linguistic Approach to Distinguishing Human, AI, and Hybrid Authorship in Turkish. Cankaya University Journal of Humanities and Social Sciences 20 2 579–594.
IEEE
[1]H. Kocagül, “Detecting the Human in the Machine: A Forensic Linguistic Approach to Distinguishing Human, AI, and Hybrid Authorship in Turkish”, CUJHSS, vol. 20, no. 2, pp. 579–594, Oct. 2026, doi: 10.47777/cankujhss.1957142.
ISNAD
Kocagül, Hülya. “Detecting the Human in the Machine: A Forensic Linguistic Approach to Distinguishing Human, AI, and Hybrid Authorship in Turkish”. Cankaya University Journal of Humanities and Social Sciences 20/2 (October 1, 2026): 579-594. https://doi.org/10.47777/cankujhss.1957142.
JAMA
1.Kocagül H. Detecting the Human in the Machine: A Forensic Linguistic Approach to Distinguishing Human, AI, and Hybrid Authorship in Turkish. CUJHSS. 2026;20:579–594.
MLA
Kocagül, Hülya. “Detecting the Human in the Machine: A Forensic Linguistic Approach to Distinguishing Human, AI, and Hybrid Authorship in Turkish”. Cankaya University Journal of Humanities and Social Sciences, vol. 20, no. 2, Oct. 2026, pp. 579-94, doi:10.47777/cankujhss.1957142.
Vancouver
1.Hülya Kocagül. Detecting the Human in the Machine: A Forensic Linguistic Approach to Distinguishing Human, AI, and Hybrid Authorship in Turkish. CUJHSS. 2026 Oct. 1;20(2):579-94. doi:10.47777/cankujhss.1957142

Çankaya University Journal of Humanities and Social Sciences
CUJHSS, e-ISSN 3062-0112