Hello, dear readers! Last week was a bit packed, and I had a mildly annoying cold, so I could not find the time to write a decent issue of this newsletter. Hopefully, today’s edition makes up for the gap. It features a short rant on dealing with perfectionism, some reading recommendations, some details about applications to the EUI doctoral programme in Law, as well as a few other opportunities related to AI and Law. And, of course, an adorable otter.
Today I want to think aloud about the ages-old question: when is a text good enough to be made public? For those who love the writing process, stopping might be a challenge because they are having fun with the expressive process. I am not one of those people: writing does not come easy to me, so I try to find how much I need to write to convey my points.1 What entertains me, instead, is the editing process: I am at my best after an initial expression of ideas, when I can fine-tune sections of my arguments, cut unnecessary passages, and so on.2 This is also a form of writing, of course,3 but here the challenge is slightly different: when should we stop tinkering with the text? If it is to be read by other people—and some texts are meant to be written rather than read—then we must finalize it at some point.4
The usual response to these challenges is to work on a text until it reaches the best possible form. Such an impulse is positive, especially in academia and other domains in which rigour is essential. And, in academic contexts, it offers a defence mechanism: if we rewrite and rewrite our text, we are more likely to avoid or respond to the criticisms we can anticipate, giving our ideas a strong presentation and saving us from the embarrassment of having somebody point out an obvious flaw in our labour of months or years.
Yet, at least in an academic context, the individual comfort offered by perfectionism has deep negative implications. At an individual level, it may be a stumbling block in one’s career, especially for early-career scholars who still need to secure permanent employment. But, in my view, the biggest harms come at a systemic level:
Our obsession with perfect expression means that some arguments never see the light of the day or only do so belatedly. Whenever that happens, the conventional wisdom in a discipline may have settled on ideas and methods that are less fruitful than the proposed alternative, which now faces an uphill struggle in becoming accepted;5
The bar for perfection is not universal and fair: it reproduces patterns of discrimination and privilege in scholarship. The insiders of an academic community can often get away with sloppy scholarship, while individuals from minority backgrounds and other people in particularly vulnerable situations must make titanic efforts to even get heard (and are sometimes ignored even then);
As a result of these asymmetries, perfectionism is not really useful as a filter for publications, either. Excellent publications often get rejected (or stuck into revise-and-resubmit loops) on the grounds of minor details, whereas even the best venues can publish loads of formally perfect but uninteresting work.6 If we want to make the volume of published scholarship manageable, demanding perfection at the point of submission (or even publication) does not seem useful.
Despite these undesirable consequences, many scholars—myself included—can be petrified by the challenges of determining when a text should be made available to the world. My approach to this issue, nowadays, is to see academic research as a team sport. Whenever I write something, I am not trying to provide killer arguments in favour of my positions, to solve once and for all a particular question, or to think thoughts no one has conceived before. Instead, I am trying to do something closer to Thomas Kuhn’s image of normal science:7 solving puzzles that add a little to our knowledge about certain objects of study, in a way that allows other people to continue from where I stopped. Accordingly, my standard of “good enough” is not one that ends debate on a topic, but one that keeps the conversation going and adds new perspectives to it. As the American philosopher Robert Nozick once said, “There is room for words on subjects other than last words.”
I do not intend to present this framing as a universal solution for the issues that stem from perfectionism. The most pressing of these issues are social in nature, so they cannot be solved by changing individual outlooks. Even at the individual level, the fact that I feel comfortable stopping when I do is shaped by various privileges,8 and as such it is not universalizable. Still, I wanted to write about this subject to articulate my thoughts on it and remind myself that enough is enough for a particular section of my thesis. And, if that helps other people facing similar struggles, that is even better.
Anyway, I should probably get back to dissertating now.
Recommendations
First, I need to get some self-promotion out of the way. After the latest newsletter, two writings of mine are now available. The first is an updated version of my AI Act paper with Nicolas Petit, now released as a Robert Schuman Centre Working Paper. In this paper, we draw from law & technology scholarship to highlight three issues faced by EU policymakers: how to frame general-purpose AI, the mismatch between the views of risks in product safety and human rights law, and the institutional capability and legitimacy gaps that emerge from the Act’s design. None of these issues can be fully eliminated, and our update shows that, in some ways, the Parliament and Council proposals deepen them. Still, we propose some ameliorative measures.
The second writing I would like to pitch you is a solo-authored paper, titled Automated Uncertainty: A Research Agenda for Artificial Intelligence in Administrative Decisions. An abridged version of my arguments in the paper is available at DigiCon, but my overall point is that introducing AI into administrative decision-making does not eliminate uncertainty. Instead, it introduces new sources of uncertainty about facts, what legal rules are applied, and by whom. By introducing the term “automated uncertainty”, I show how these factual and legal uncertainties are connected to the scientific uncertainty surrounding AI systems, and suggest the need for investigating their impact on how administrators make their decisions at the street level.
But that’s more than enough about myself. Now I’d like to share a few links that might interest you:
Mathilde Cohen, ‘When Judges Have Reasons Not to Give Reasons: A Comparative Law Approach’ (2015) 72 Wash & Lee L Rev 483.
Rebecca Crootof and BJ Ard, ‘Structuring Techlaw’ (2021) 34 Harv J L & Tech 347.
European Data Protection Supervisor, Opinion 44/2023 on the Proposal for Artificial Intelligence Act in the light of legislative developments.
Jie Fan, ‘Axiomatizing Rumsfeld Ignorance’ [2023] J Philos Logic.
Stephen Knox and Norin Arshed, ‘Street Level Discretion, Personal Motives, and Social Embeddedness within Public Service Ecosystems’ [2023] Pub. Admin. Rev. early access.
Johann Laux, ‘Institutionalised Distrust and Human Oversight of Artificial Intelligence: Towards a Democratic Design of AI Governance under the European Union AI Act’ [2023] AI & SOCIETY.
David Lawder, ‘US Drops Digital Trade Demands at WTO to Allow Room for Stronger Tech Regulation’, Reuters (25 October 2023).
Raymond Sun, Global AI Regulation Tracker.
Helen Toner and Timothy Fist, ‘Regulating the AI Frontier: Design Choices and Constraints’ (Center for Security and Emerging Technology, 26 October 2023)
Lee Vinsel, ‘How to Be a Better Reactionary: Time and Knowledge in Technology Regulation’ (Medium, 17 October 2023).
See also the US Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence.
Opportunities
PhDs at the European University Institute
The EUI, my doctoral alma mater, is now accepting applications for its PhD programmes in Law, History and Civilization, Political and Social Sciences, and Economics, all based in the metropolitan region of Florence, Italy. All of the programmes last from 4 to 5 years, and they offer an international environment for you to develop your research. Not to mention the small perks, such as free access to museums and the gorgeous sights from each EUI building. Applications are open until 31 January 2024, and activities begin on 1 September 2024.
Each department has its quirks and priorities, so I cannot speak for the other departments except to say that they feature world-class groups of researchers. The same is true of the law department, which has solid research on, inter alia, European Union law, public international law, and various private law topics. And readers of this newsletter will probably be aware—or pleased to know—that the EUI has a strong interdisciplinary research cluster on technological change and society, including many professors and PhD researchers from the law department.
If you want to learn more about the EUI doctoral programme, we will hold a few information sessions in November. And when I say “we”, it is because I will be involved in three of them. On 30 November, Professor Gráinne de Búrca and I will present the EUI PhD in Law more generally. Two days before, Nicolas Petit and I will hold two information sessions about ASPIRE, a programme that funds scholarships for researchers from the Global South interested in topics of innovation, competition, and technology policies (the first session in an Asia-friendly timeslot, the other one aimed at the Americas). In addition, Professor Claire Kilpatrick, my colleague Wojciech Giemza, and alumna Professor Patrycja Dąbrowska-Kłosińska will hold a session on 23 November aimed at potential candidates for the grants offered by Poland.
Each of these sessions will also feature Lorenzo Ghezzi from the academic service. Hence, they are a perfect opportunity to ask any questions about research at the EUI Law department, life in Florence, or the application procedure. And please feel free to spread the word about the programme and the information sessions to anyone who might be interested in them.
Other opportunities
The Law & Technology Research Group at Sciences Po Law School is calling for applications from early-career researchers (PhD stage or <5 years post-PhD) for the Young Digital Law conference 2024, on the theme of Conflicts & Communities in Digital Law. They invite short papers for panel presentations and extended abstracts for writing workshops, which should be submitted by 19 January 2024, with the conference taking place on 6-7 June 2024.
Prof. Dr. Anna Beckers from Maastricht University is looking for an aspiring researcher in law with an interest in sociological jurisprudence and science and technology studies. The PhD researcher is expected to engage in the “data” angle of the CHAINLAW project and research from a socio-technical perspective on supply-chain software systems and how such technical affordances regulate global production, and how they relate to the law. Applications are open until 11 December 2023.
Boston University (BU) invites applications for eight tenure-track faculty positions as part of a 3-year cluster hiring initiative in Artificial Intelligence (AI). They particularly welcome applications from minority candidates and candidates working on the foundations of AI, methodologies, or AI research inspired by or tackling issues in areas beyond computing. Review of applications will start on 1 November 2023 and will continue on a rolling basis until 15 April 2024.
IE University invites applicants committed to an academic career in an international and innovative environment to apply for 3 full-time, tenure-track, positions at IE Law School. Applications should be sent by 15 December 2023, and the appointment will begin September 1, 2024. They accept applications from all areas of the law, but Legal Theory and Philosophy and Law and Technology are listed as priority areas.
The mustelid coda
I hope you enjoyed this issue. If you haven't done so, please subscribe to receive future updates in your email inbox. And feel free to hit “Reply” and contact me with any comments, complaints, or suggestions. I am also happy to spread the word about job opportunities, events, and new publications that might interest those who read this newsletter.
In addition to its communicative value, writing is an essential part of the thought process, especially when we are grappling with complex ideas. Consequently, the decision of when to stop writing is often a decision about whether and how our thoughts about a particular topic have grown stable enough to form a cohesive(-ish) whole.
However, I am a terrible proofreader, much to the chagrin of my co-authors.
Of course, once a text is finished one can always revisit it. Some people build their careers by revisiting again and again the same topic (or narrow set of topics) in increasingly sharper formulations. This is a valid way to approach scholarship, but it is not for me.
There are many, many examples in each discipline. As an early-career scholar, and a not particularly brave one, I am uninclined to supply examples in writing, unless I am directly arguing against them.
For more on this, see Bright and Heesen on peer review.
My use of the term is meant as an analogy, not as an affirmation of the epistemic status of the law. Contra many theorists from Civil Law backgrounds, I believe that the idea of a “science of law” is a disservice both to the law and the (social) sciences.
Indeed, it took me some time to lower to a manageable level the impostor syndrome I feel writing about EU law as somebody who had a single class—not course, class—on EU law before coming to Italy. Even here, however, my adaptation has benefited from the privileges associated with being at the EUI.