Forensic Language Analysis Lab
Summary
- We provide expert linguistic analysis of language used as evidence in criminal or civil forensic cases to address questions of disputed and anonymous authorship, authorship profiling, and text authenticity.
- We research how each individual is linguistically unique and how this knowledge can be applied to solve forensic problems.
- We also study Artificial Intelligence (AI) language, including the detection of AI-generated texts, linguistic impersonation and text‑based deep‑fakes.
About the research and consultancy
We research how linguistic methods can be used to analyse language as evidence, particularly in cases where the authorship of a text is disputed.
Our main research focus is linguistic individuality: how unique each individual’s language use is, how stable it remains across time and context, and how this knowledge can be used in forensic casework.
This includes research on text deep‑fakes, AI‑assisted impersonation, and the linguistic differences between human authors and AI systems.
A core part of our work examines the impact of Artificial Intelligence (AI) on authorship analysis: both how AI tools can support forensic analysis and how AI changes the nature of writing itself.
We also apply our methods to historical and literary cases, such as the Jack the Ripper letters and newly discovered texts linked to Charles Dickens.
We regularly run casework groups for undergraduate and postgraduate students, using publicly available data to investigate real‑world authorship questions as part of research‑led teaching.
Previous groups have worked on: the Zodiac killer, the mystery of John Titor, and the identity of Erdnase.
Research impact
Our research directly supports our consultancy work in the United Kingdom and internationally. We work with law enforcement agencies, legal professionals and other organisations on cases where language plays a critical evidential role.
Our expertise is applied in investigations involving disputed or anonymous texts, including serious criminal offences, fraud, threats and cases of author impersonation. In each case, we provide clear, evidence‑based analysis suitable for investigative and legal decision‑making.
While most consultancy work involves written language, our research also supports the analysis of spoken language data where appropriate. We do not provide consultancy in Forensic Speech Science (voice analysis).
Beyond individual cases, our work has wider impact. It informs policy and best practice around the use of AI in education and the justice system, contributes to public debate on the reliability of linguistic evidence, and supports fair, evidence‑based outcomes in high‑stakes environments.
What consultancy services do we offer?
We offer the following consultancy services:
- Authorship analysis: who authored this questioned document? We work on all kinds of types of texts, such as emails, sets of text messages, and longer legal documents.
- Authorship profiling: what clues can be inferred about the author of this questioned document from its language? For example, we can study a questioned document to determine whether the author is a native speaker of English or not (if not, which language, if yes, which dialect), and other socio-demographic characteristic, such as gender/age/social class/education/ethnicity or any other information that can be extracted from the language. NB: we do not perform psychological assessments.
- AI usage and text deep-fakes: was this questioned text produced by AI or by a certain person? Is this text genuinely from a specific person or is it a deep-fake produced by AI? For example, we offered analyses to determine whether a student accused of using AI in higher education genuinely produced their text or not.
We also offer a short course to introduce Forensic Linguistics. The course can be taken as a one-day course or over eight weeks, depending on the intended learning goals.
Meet the team
- Michael Cameron
Title: “Each to their own language: A study of linguistic individuality”,
Supervisors: Dr Andrea Nini and Dr Colin Bannard. - Sadie Barlow
Title: “Lexical chunks and the nature of Idiolects”,
Supervisors: Dr Andrea Nini and Dr Edoardo Manino. - Matilde Basilicati
Title: “Individual language change across a lifespan. Using Authorship Analysis to investigate the degree of stability in a person’s idiolect”,
Supervisors: Dr Andrea Nini and Dr Dmitry Nikolaev. - Baoyi Zeng
Title: “Can a robot impersonate a human? Studying machines’ ability to mimic linguistic identity”,
Supervisors: Dr Andrea Nini and Dr Dmitry Nikolaev. - Ben Cross
Title: “Machine Learning and Natural Language Processing for Authorship Identification”,
Supervisors: Dr Riza Batista-Navarro and Dr Andrea Nini. - Thomas Hales
Title: “Large Language Models for Authorship Analysis”,
Supervisors: Dr Andrea Nini and Dr Dmitry Nikolaev.
- Nicholas Quincy Hopper
- Emma Bell
- Shehza Muzafar
- Asha Joseph
Publications and further information
- Cameron, M., Nini, A., Bannard, C. Linguistic individuality in lexicogrammatical alternations. https://doi.org/10.31234/osf.io/uvtrb
- Nini, A. (2023). A Theory of Linguistic Individuality for Authorship Analysis. Elements in Forensic Linguistics. Cambridge, UK: Cambridge University Press. https://doi.org/10.1017/9781108974851
- Dunn, J. & Nini, A. (2021). Production vs perception: The role of individuality in usage-based grammar induction. In Proceedings of the Workshop on Cognitive Modeling and Computational Linguistics, 149-159, Online: Association for Computational Linguistics. https://doi.org/10.18653/v1/2021.cmcl-1.19
- Fonteyn, L. & Nini, A. (2020). Individuality in syntactic variation: An investigation of the 17th-century gerund alternation. Cognitive Linguistics, 31(2), 279-308. https://doi.org/10.1515/cog-2019-0040
- Nini, A., Halvani, O., Graner, L., Titze, S., Gherardi, V., Ishihara, S. (2016). Grammar as a behavioral biometric: Using cognitively motivated grammar models for authorship verification. Humanities and Social Sciences Communications, 13, 455. https://doi.org/10.1057/s41599-025-06340-3
- Nini, A. (2026). idiolect: An R package for forensic authorship analysis. Journal of Open Source Software, 11(119), 7575. https://doi.org/10.21105/joss.07575
- Ishihara, S., Kulkarni, S., Carne, M, Ehrhardt, S., Nini, A. (2024). Validation in Forensic Text Comparison: Issues and opportunities. Languages, 9(2), 47. https://doi.org/10.3390/languages9020047
- Nini, A. (2019). Developing forensic authorship profiling. Language and Law/Linguagem e Direito, 5(2), 38-58. https://ojs.letras.up.pt/index.php/LLLD/article/view/6116
- Nini, A. (2019). Corpus analysis in forensic linguistics. In Chapelle, C. A. (ed), The Concise Encyclopedia of Applied Linguistics, 313-320, Hoboken: Wiley-Blackwell.
- Nini, A., Wood, C., Bowles, H. (2025). The Dickens signal: Investigating the authorship of “The Two Brothers”, a ghost story transcribed from shorthand. Textus, English Studies in Italy, 3, 25-52. https://www.rivisteweb.it/doi/10.7370/119419
- Nikolaev, D., & Shumilin, M. (2021). Some Considerations on the Attribution of the ‘New Apuleius.’ The Classical Quarterly, 71(2), 819–848. https://doi.org/10.1017/S0009838821000987
- Donlan, L. & Nini, A. (2022). Forensic authorship analysis of the Ayia Napa rape statement. In Picornell, I., Perkins, R., and Coulthard, M. (eds), Methodologies and Challenges in Forensic Linguistics Casework, 29-43, Hoboken, NJ: Wiley-Blackwell.
- Nini, A. (2018). An authorship analysis of the Jack the Ripper letters. Digital Scholarship in the Humanities, 33(3), 621-636. https://doi.org/10.1093/llc/fqx065
- Zeng, B. & Nini, A. Authorship impersonation via LLM prompting does not evade Authorship Verification methods. https://arxiv.org/abs/2603.29454
- Gan, Q., Dunn, J., Nini, A., Adams B. (forthcoming). A multi-dialectal, longitudinal corpus of human-AI hybrid language production. In Proceedings of the International Conference on Language Resources and Evaluation, European Language Resources Association (ELRA).
- Barlow, S., Manino, E. & Nini, A. ‘A simpler method for calculating likelihood ratios in authorship verification’. 6th European Conference of the International Association for Forensic and Legal Linguistics, University of Montpellier Paul Valéry, Montpellier, France. 07/2026.
- Basilicati, M. & Nini, A. ‘Authorship verification with LambdaG on Ancient Greek: A case study of Plato’s seventh letter’. 6th European Conference of the International Association for Forensic and Legal Linguistics, University of Montpellier Paul Valéry, Montpellier, France. 07/2026.
- Cameron, M., Nini, A. & Bannard, C. ‘Same story, different chunks: Idiolectal patterns in language production’. 6th European Conference of the International Association for Forensic and Legal Linguistics, University of Montpellier Paul Valéry, Montpellier, France. 07/2026.
- Hopper, N. & Nini, A. ‘The effect of formality on the length of idiolectal bundles’. 6th European Conference of the International Association for Forensic and Legal Linguistics, University of Montpellier Paul Valéry, Montpellier, France. 07/2026.
- Zeng, B. & Nini, A. ‘Can AI fool a forensic linguist? Detecting LLM impersonation of an individual’s language’. 6th European Conference of the International Association for Forensic and Legal Linguistics, University of Montpellier Paul Valéry, Montpellier, France. 07/2026.
- Tompkinson, J. & Nini, A. ‘Assessing the suitability of forensic authorship analysis methodologies for speech data’. The 33rd Annual Conference of the International Association for Forensic Phonetics and Acoustics (IAFPA), Leiden University, The Hague, Netherlands. 21/07/2025.
- Nini, A. ‘Examining an author’s individual grammar’. Comparative Literature Goes Digital Workshop, Digital Humanities 2025. Universidade Nova de Lisboa, Lisbon, Portugal. 14/07/2025.(see Github repository)
- Barlow, S. & Nini, A. ‘A corpus analysis of idiolectal n-grams’. 13th International Corpus Linguistics Conference (CL2025), Aston University/Birmingham City University/University of Birmingham, Birmingham, UK. 30/06/2025.
- Nini, A. ‘Language models as cognitively realistic mathematical models of grammar’. Plenary talk at the 13th International Quantitative Linguistics Conference (QUALICO 2025), Masaryk University, Brno, Czech Republic. 28/06/2025.
- Zeng, B. & Nini, A. ‘Can AI fool a forensic linguist? Detecting AI impersonation of an individual’s language’. Investigating Human and AI Creativity and Imitation Symposium, University of Leeds, Leeds, UK. 06/06/2025.
- Tompkinson, J. & Nini, A. ‘Evaluating the usefulness of embedding phonetic representations into an authorship analysis-based framework for the comparison of spoken data’. 5th European Conference of the IAFLL – International Association for Forensic and Legal Linguists, Aston University, Birmingham, UK. 26/06/2024.
- Cameron, M., Nini, A. & Bannard, C. ‘Linguistic Individuality in Lexicogrammatical Alternations’. 5th European Conference of the IAFLL – International Association for Forensic and Legal Linguists, Aston University, Birmingham, UK. 26/06/2024.
- Brown, G., Nini, A. & Kirchhübel, C. ‘Likelihood ratio based authorship verification methods applied to forensic voice comparison tasks’. 5th European Conference of the IAFLL – International Association for Forensic and Legal Linguists, Aston University, Birmingham, UK. 26/06/2024.
- Barlow, S. & Nini, A. ‘The idiolectal information in long character n-grams: An experiment using paraphrasing’. 5th European Conference of the IAFLL – International Association for Forensic and Legal Linguists, Aston University, Birmingham, UK. 26/06/2024.
- Nini, A. ‘Cognitive Linguistic forensic authorship analysis using the likelihood ratio framework’. 5th European Conference of the IAFLL – International Association for Forensic and Legal Linguists, Aston University, Birmingham, UK. 26/06/2024.
- Nini, A. ‘How Linguistics can help to eliminate unnecessary complexity in modern Natural Language Processing’. Keynote at the Interdisciplinary Perspectives: Bridging Sociological Studies in the Digital Age conference. King’s College London, London, UK. 11/06/2024.
- Nini, A. ‘A more interpretable method for the analysis of an author’s language’. Keynote at the Corpus Linguistics Symposium: Style and Authorship. University of Leeds, Leeds, UK. 07/06/2024.
- Nini, A. ‘A formal model of lexicogrammatical individuality’. 16th International Cognitive Linguistics Conference, Heinrich Heine University Düsseldorf, Düsseldorf, Germany. 10/08/2023.
- Nini, A. ‘The Statistical Approximation Hypothesis: A Cognitive Linguistic explanation for the effectiveness of function word frequency’. Digital Literary Stylistics Special Interest Group Workshop, Digital Humanities 2023, University of Graz, Graz, Austria. 10/07/2023.
- Nini, A. ‘Why do function word frequencies vary across individuals? Evidence in favour of The Statistical Approximation Hypothesis’. 12th International Corpus Linguistics Conference (CL2023), Lancaster University, Lancaster, UK. 04/07/2023.
- Nini, A. ‘Using the likelihood ratio framework in real authorship identification casework: The General Impostors with Writeprints method’. 4th European Conference of the IAFLL – International Association for Forensic and Legal Linguists, University of Porto, Porto, Portugal. 20/07/2022.
- Nini, A. & Ishihara, S. ‘The likelihood of lexicogrammatical overlap’. IAFL15: The 15th Conference of the International Association of Forensic Linguists. Aston Institute for Forensic Linguistics, Aston University, Birmingham, UK (online). 14/09/2021.
- Nini, A., Cameron, M. & Murphy, C. ‘Experimental evidence on the individuality of lexicogrammar’. International Construction Grammar Conference 11. University of Antwerp, Antwerp, Belgium (online). 20/08/2021.
- The 2nd Roundtable on Practices and Standards in Forensic Authorship Analysis. University of Manchester, Manchester, UK (online). 07/07/2021.
- How Language Nerds Solve Crimes
Otherwords, PBS, 26 January 2024 - It is too easy to falsely accuse a student of using AI: a cautionary tale
Times Higher Education, 10 July 2023 - Rally in support of woman in Cyprus ‘rape’ case
Victoria Derbyshire, BBC Two, 6 January 2021 - Cyprus rape case: Experts cast doubt on teenager’s confession
The Times, 3 January 2020 - Woman Who Accused 12 Men of Rape Is Guilty of ‘Public Mischief’ in Cyprus
The New York Times, 30 December 2019 - UK teenager’s rape retraction ‘not written by native English speaker’
Sky News, 31 October 2019 - Jack the Ripper letters ‘faked to sell newspapers’
The Times, 1 February 2018 - Jack the Ripper letters suggest newspaper hoax
BBC News, 1 February 2018