Thinking about bad doctrine (AI, Law, and Otter Things #22)
Welcome to the latest issue of AI, Law, and Otter Things! This time, I ended up writing a monographic rant about legal doctrine. More specifically, about what is bad doctrine and what we should do about it. After these words, I share some interesting stuff I've read lately, before concluding with some otter images.
In a previous issue, I manifested my sympathies for doctrinal legal research. Even if one should resist the ultra-formalist claims that the law is a system that must be read entirely on its own terms, this system still has an internal logic—as warped as it may be at times—that warrants some study. Such study can be an important part of the life of the law, integrating various sorts of concerns into the legal game and providing practitioners with a systematic view of their everyday practice. Yet, one might easily respond that, in practice, the doctrine is not as helpful as it can be.
A few months ago, a judge from one of the Brazilian state courts posted that one should "Ignore the doctrine. 99% of it is rubbish.". Since Brazil is a civil law jurisdiction, the statement has led to angry reactions from legal scholars, who have dedicated quite a few lines of text to protect the honour of the Rechtslehre from the brutal attacks of an Instagram post by a judge who, let's face it, is not really a towering figure among Brazilian legal practitioners.
This critique is a blunt cousin of other arguments against "merely" doctrinal thinking, such as claims that legal academia (in general or in a specific jurisdiction) produces too much doctrine about a given topic. As these arguments express judgment upon actually existing doctrinal thinking—rather than systematic claims such as the evergreen thesis of a crisis of doctrinal thinking—, we can only assess them if we have a standard of what counts as rubbish doctrine in the first place.
I am sure there are deep studies about this kind of thing, but I believe personal experience can provide us with a good starting point for the debate. After all, I don't think it is too bold to say that most of my legal readers have, at some point, read a piece of legal scholarship and thought "Well, this is crap." Furthermore, many of us have produced that crap at some point—I surely have, and please don't let me know if you haven't. Sometimes these reactions are purely emotional—"How dare this human being disagree with my clear argument from the paper I just wrote?". Yet, it seems premature to think we cannot formulate this judgment of rubbishness in rational terms.
A first candidate reason why a piece of doctrine is rubbish might be that it is wrong: it might rely on inadequate statements of fact or applicable law or draw inferences from these starting points that are either a poor fit with the legal system or clash with accepted propositions in the doctrine. Such wrongness is always tentative, since changes to the law or doctrinal consensus on other topics may render a previously unacceptable idea more palatable to legal practitioners. Nevertheless, I don't feel particularly uncomfortable with the idea that someone may be wrong about the law, even if a statement of wrongness is always conditional.
Being wrong, however, is neither a sufficient nor a necessary condition for rubbish doctrine. A piece of writing might be wrong and yet signal directions that might be useful for practitioners and further scholarship. This seems to be the case of my first paper on automated decision-making: I am no longer convinced by the specific doctrinal positions I held there, but it suggests an approach for connecting Articles 22 and 25 GDPR that has been useful to a handful of scholars elsewhere. Conversely, one may get things right and still fail to contribute meaningfully to scholarship, for example by defining your thesis in terms so narrow that it feels like a fill-in-the-blanks exercise. We must therefore seek our distinction elsewhere.
Another easy target for criticising doctrine is language, as doctrinal scholarship can be arid, use arcane vocabulary, and all the usual complaints. And, let's face it, most of us aren't exactly literary wordsmiths, either. But these complaints are a lot more acceptable when they come from outside the legal profession. Legal writing face similar issues and some legal cultures even place a premium on baroque prose. This is certainly the case in Brazil, where law school often pushes students to "write fancy", but I would invite any reader that sees this habit as exclusive to Civil Law cultures to open the terms of service of the nearest online platform. So, it seems difficult to argue that la doctrine is especially bad when it comes to crimes against the vernacular.
I believe it is more useful to define rubbish doctrine in methodological terms. What makes me think "This is bad", rather than "This is wrong" or "It physically hurts me to read this text", is a certain sloppiness in argumentation. Sloppy doctrine can appear, for example, through the inadequate examination of our starting points. This is particularly challenging when dealing with interdisciplinary matters, as we often lack the means to probe the validity of some theoretical claims made outside the law. If we rely on outdated scientific results, take at face value technological claims about AI made for marketing purposes, or rely on inaccurate estimates of the environmental impact of AI systems, we might waste good thought on flawed premises.
We might also produce rubbish doctrine if we are sloppy with our modes of reasoning. Legal arguments are composed of specific elements—what sources can be accepted, how to draw inferences from these sources, and so on—, and many trees have been felled to advance our understanding of their interplay. If a piece of doctrine fails to observe the rules of play regarding these elements, it provides an inadequate description of the legal system as it stood when being analysed.
At its worst, sloppily constructed doctrine might be "not even wrong" because one cannot meaningfully evaluate the argument it presents. This can happen, for example, if a doctrinal argument is grounded on excessively indeterminate concepts—of the kind that can be stretched to fit anything— or if the inference processes are internally inconsistent. (Thanks to Victor Nascimento for calling my attention to this point) Whenever that is the case, doctrine no longer is a useful guide to help us arrive at a reasonable decision but merely an instrument to give our decision a veneer of legal rationality.
Sloppy doctrine might thus warrant the label "rubbish" if it provides too distorted a description of the legal system to be of any use for thinking about legal issues. It would be self-deluding to say that no such doctrine is published. Even though we all exercise self-criticism while writing—and pre-publication review mechanisms such as peer-review provide some form of systemic control—these are at best insufficient to ensure that only good doctrine is published. Yes, even in your favourite journal. But would any sensible criteria of sloppiness force us to agree with the judge's claim that 99% of doctrinal thinking is rubbish? Here, we have a clearly subjective line-drawing exercise, which will be directed by many factors beyond purely academic considerations.
But even if one is willing to grant that much doctrinal scholarship is rubbish, doctrinal scholars should not lose much sleep over the situation. After all, legal scholars frequently draw insights from ideas that had been discarded, abandoned as unpractical, or otherwise rejected. For a lawyer trained in a Civil Law tradition, the most obvious example would be the various ingenious uses generation after generation has made of Roman law instruments. If we are taking the metaphor of doctrinal rubbish somewhere, we might as well go all the way and remember rubbish can often be recycled or reused.
Stretching the refuse metaphor beyond any semblance of reasonableness, one might also think about reducing the production of rubbish. This reduction seems particularly appealing in the case of doctrine that is "not even wrong", as its lack of conceptual sharpness might contaminate any arguments that try to salvage it. By containing rubbish doctrine before it causes too much damage, we could ensure it does not contaminate scholarly debates and legal practice, thus reducing the probability that new rubbish comes into place.
These rubbish-containing approaches could be, at least to a certain extent, adopted through the individual and systemic filters described above. Yet, their desirability must be weighed against two other factors. The first one is that doctrinal production is a form of intellectual experimentation, and some experiments eventually go wrong, often in ways that leave behind little but rubbish. Rubbish-suppressing norms in doctrinal scholarship thus push scholars (at least those wanting a career) towards risk aversion in their thinking, and an overly cautious scholarly culture might end up being ossified and unable to face challenges posed by social change.
Allowing the publication of some rubbish doctrine might also be useful from a pedagogical perspective. Given the complexity of legal reasoning, there are many ways to produce rubbish, some more subtle than others. But doctrine operates at a different tempo than legal practice, so legal scholars actually have the time to understand whether and why something is rubbish. If properly identified, rubbish doctrine provides us with a sandbox where we can analyse errors that would be very costly if committed by a lawyer or judge. This does not mean that anything goes when it comes to doctrine, but it suggests that we might miss a bit if we adopt excessively strict criteria to avoid the eventual production of rubbish.
Doctrine, unlike legal practice, can afford to get things wrong on the first try. After all, most of the harms of doctrinal rubbish only appear when said rubbish steers other forms of legal thinking, such as judicial decisions. Therefore, one can afford to have lots of rubbish in doctrinal thinking, so long as it is possible to distinguish good doctrine from bad doctrine and make sure that the former is accessible. If that is the case, doctrine as a whole can be a valuable social practice even if—and in some aspects because—a lot of it turns out to be rubbish.
A few interesting reading
Speaking Through Time (Legal-Phi, 4 January 2022)
In this blogpost, Lucas Miotto (Maastricht) interviews Martin David Kelly, who presents his theories on how the meaning of statutes changes over time. To deal with “always speaking” utterances, which are not meant to produce effects at a specific moment in time but rather through an entire period, Kelly distinguishes between the linguistic meaning of the utterance and its “content”, that is, its conditions of application. This interview provides an interesting introduction to the ideas Kelly develops in his thesis, and it includes some very illustrative examples (such as the content of “room temperature” when it comes to serving wine).
Barbara Cosens and others, ‘The Role of Law in Adaptive Governance’ (2017) 22 Ecology and Society.
The law is often conceptualised as a top-down, rigid approach to regulation. However, if adequately deployed, it can be an enabler and not a constraint to adaptive governance by creating conditions for the emergence of adaptive regimes, eliminating barriers to their development, and legitimising them.
Yanqiu Tao and others, ‘Trend towards Virtual and Hybrid Conferences May Be an Effective Climate Change Mitigation Strategy’ (2021) 12 Nature Communications 7324.
Conferences have a substantial environmental impact, as shown by a life cycle analysis of the various transportation and organising factors involved in their realisation. Even though virtual and hybrid conferences also have considerable footprints, the authors propose various models of conference organisation that show these approaches result in a considerably lower environmental footprint (close to 90% for purely online events, but even hybrid events with local gatherings can achieve substantial reductions).
Tony Freeth, ‘Wonder of the Ancient World’ (2022) 326 Scientific American 24.
A fascinating dive into the mechanisms of the Antikythera machine, an analog computer found in an Ancient Greek shipwreck.
Finally, some otters
Thank you for your attention, and please feel free to contact me if you want to follow up on any of these topics (or chat about other stuff)!
[tweet https://twitter.com/hokutotomerry/status/1482673870574264320]