About the Journal
The Journal of Cross-disciplinary Research in Computational Law (CRCL) invites excellence in law, computer science and other relevant disciplines with a focus on two types of ‘legal technologies’: (1) data-driven (e.g. predictive analytics, ‘intelligent’ search) and (2) code-driven (e.g. smart contracts, algorithmic decision-making (ADM), legal expert systems), and (3) their hybrids (e.g. code-driven decision-making based on data-driven research).
Legal practice is where computational law will be resisted, used or even fostered. CRCL wishes to raise questions as to (1) when the introduction of legal technologies should be resisted and on what grounds, (2) how and under what conditions they can be integrated into the practice of law and legal research and (3) how their integration may inform, erode or enhance legal protection and the rule of law.
Please subscribe for updates on upcoming articles and issues.
Please note that at this moment we are preparing the first issue, which will appear in 2022. We will open the journal for submissions in the course of 2022, do not hesitate to contact the editors if you have a proposal for a special issue or an article.
Launching of online first articles
The journal has kicked off mid-November 2020 with an invited article by Wolfgang Hoffmann-Riem, former Justice of the German Constitutional Court in Karlsruhe, who was part of the Court when it decided the seminal case that established the fundamental right to the guarantee of the confidentiality and integrity of information technology systems. By inviting him to contribute to the opening issue of CRCL we emphasise our close attention to legal practice.
The second online-first article (late November 2020) is authored by Mireille Hildebrandt, founding member of journal's editoral team. Her article concerns the normative alterity that is inherent in the different technologies that articulate legal norms, arguing that the normative affordances of text-based technologies enabled the rise of the rule of law. She contends that the transition to data- and code-driven 'legal' technologies will transform legal practice and thereby the mode of existence of modern positive law.
In the third online-first article (beginning December 2020) Reuben Binns, with a background in both computer science and philosophy, investigates the difference(s) that make(s) a difference in text-driven law as compared to data- and code-driven decision-making systems. This involves issues of procedure (getting to a right answer in the right way), discretion (dealing with the indeterminary inherent in text-driven systems), anticipation (the role of new case law) and a number of other incompatibilities that are all highly relevant when considering the integration of 'legal technologies' into legal practice or legal scholarship.
The fourth online-first article (half December 2020) argues the crucial importance of delay in law and the rule of law. Whereas efficiency is often associated with speed, Laurence Diver, founding member of this journal’s editorial team, argues that the slowness of a text-driven legal practice may be a feature rather than a bug when it comes to legal protection. To make this point, Diver coined the concept of a ‘hermeneutic gap’ and in this article he explores to what extent ‘slow computing’ might contribute to sustaining this gap before bridging it in the context of legal practice.
In our fifth online-first article (March 2021), Emilie van den Hoven argues that the epistemic opacity that characterises the systems that comprise computational 'law' threatens us with ever-greater 'hermeneutical injustices': constraints on our ability as citizens to make sense of the law, and to account for its place in our lives, which in turn threaten the dignitarian underpinnings of the rule of law.
Our sixth online first article (November 2021) is the first of a series of articles on the issue of interpretability in machine learning – in the context of data-driven ‘law’. In this article Elena Esposito leaves the well-trodden path of trying to figure out how legal technologies informed by machine learning actually reach their conclusions (output). Instead, she argues that we must learn to interact with these technologies, to use and control them rather than submit to them. Whereas she speaks of ‘communicating with’ and her replier, Federico Cabitzo speaks of ‘relating to’, both emphasise the need to reject objectivist assumptions, while nevertheless actively engaging with them in order to regain the agency that could otherwise be lost in translation.
Our seventh online first article (January 2022) squarely addresses the static temporality that defines machine learning (ML) systems, and its relevance for automated decision support systems that inform legally relevant decisions. ML systems can only be trained on historical data, and the relevance of such data may deteriorate over time, especially in the case of the law, which is a moving target as it adapts to changing circumstances. The author, Sylvie Delacroix, explains how this affects the interpretability and contestability of these systems over time. She proposes to introduce ‘ensemble contestability’ features capable of achieving long-term contestability. Her replier, Zachary Lipton, agrees that the temporal dynamics of ML must be taken into account when considering the impact of ML on the agency of those subject to their decisions. He then moves to succinctly develop three points of critique, arguing, in particular, that Delacroix’s post hoc ‘ensemble contestability’ solution is not viable and cannot do the work that is needed. In the course of his reply, he manages to trace the history of many relevant debates, thus providing a rich resource beyond the usual focus of explainable AI. Delacroix counters by distinguishing between the ‘interpretable object’ and the ‘subject capable of interpretation’, arguing that Lipton’s objections concern the ‘interpretable object’, whereas her concern sides with the subject capable of interpretation. Here we see the need for and the salient results of a genuine, respectful cross-disciplinary conversation.