This blog aims to present the research project Lexhnology to a variety of different audiences: researchers, political and economic representatives, learners of legal English and the general public. The blog contains a general presentation of the project and its team, current news about events and publications, and job advertisements.

In a nutshell, the ANR project Lexhnology aims to create an online learning resource for learners of legal English, specifically for reading comprehension of American judicial decisions. The project is financed by the French National Research Agency (Agence nationale de recherche, ANR). It is led by Mary C. Lavissière (CRINI, Nantes Université). Other partners include LS2N (Nantes Université), ATILF (CNRS & Université de Lorraine), and LAIRDIL (Université de Toulouse). We provide further details in the paragraphs below.

Lexhnology’s objectives and research hypothesis

Given the growing extraterritoriality of American law, this domestic law is increasingly impacting other countries’ jurisdictions, as the recent data privacy issues for transferring personal data from Europe to the USA through the GAFAM show. The capacity to understand the reasoning supporting judicial decisions is a universal necessity in the fields of law. Within this context, it is of prime importance that foreign or second-language (L2) users of legal English be able to analyze case law.
Thus, identifying rhetorical divisions of these texts is a precursor for users’ understanding and exploiting of the reasoning behind judicial decisions. Indeed, there is a growing premise that communication is greatly facilitated by making text divisions explicit. This notably helps L2 learners to identify the information structure and communication objectives necessary to understanding documents in languages for specific purposes (LSP) [1]. Within linguistics, these discourse divisions have been called moves 2 (e.g. claim, support, refutation).

Move analysis was originally developed by [2] to teach L2 learners of academic English how to structure their articles in a way that adheres to the prototypical research article. While prescriptive recommendations about writing judicial opinions exist [3], very little linguistic research approaches moves in legal genres with an descriptive or analytic objective, see [4] [5], [6], [7]. Within natural language processing (NLP), the task of move analysis is called argumentative structure analysis and requires performing several processes: text segmentation, communicative role classification, segment linking and relation labeling. Like in linguistics, the attention has so far been focused on analyzing academic and scientific writing [8]. Despite the recent massive increase of access to online legal texts and the development of NLP tools for legal language, no technology is currently able to automatically identify moves in case law. However, automatic moves analysis in case law is challenging for the predominant deep learning approach in the domain, because, as mentioned by [9], legal documents are long, linguistically complex, and difficult to understand “even for humans because everyday terms can take on new technical meanings in a legal context”.

Finally, teaching moves to L2 learners is a widely accepted method in LSP L2 teaching/learning and may help learners understand the legally-binding rationale behind judicial decisions. In this context, a tool for automatically detecting moves with NLP methods would be of great aid to L2 students of legal English. However, the effectiveness of such a teaching method for increasing the speed and accuracy of comprehension of legal English has not yet been measured.

These gaps raise several research questions:

  1. What typology of moves and prototypes for judicial opinions can be established using move analysis combined with corpus linguistics augmented by NLP
    methods ?
  2. Does interaction with a corpus with annotated moves improve reading comprehension of case law?
  3. What needs and use cases exist for L2 learners of legal English and LSP teachers concerning the task of reading case law?
  4. How does access to a corpus with annotated moves foster learner and teacher autonomy in reading case law in L2 students and jurists?
  5. How do current neural language models represent the discourse structure underlying the moves? How can self-supervised learning take discourse information into “attention”? How can linguistic knowledge be injected into computational models, making them more self-explanatory? How to design predictive
    neural architectures to make them discourse aware ?

To answer these questions, Lexhnology will take an innovative interdisciplinary approach – linguistic, NLP, LSP teaching/learning. It will define moves, automatizing their detection and identification, while measuring the effectiveness of using moves in LSP teaching/learning. By creating an annotated corpus and tagging tool this project fills the gaps in the research about moves. This corpus and tool will be integrated into an open access platform for students, teachers, and jurists Lexhnology will contribute to the strategic priorities for research identified by the French government and the National Research Agency (ANR): interdisciplinary research, digital
transformation, artificial intelligence, education and technology, and digital humanities. The project adheres to Unesco’s open science recommendations (SDG43).

Search OpenEdition Search

You will be redirected to OpenEdition Search