Reading Law's Great Unread: Qualitative Computational Methods, Artificial Intelligence and the New Empirical Legal Research
Bibliographic record
Abstract
How will new computational technologies change legal research and our visions of what law is? Inspired by the work of digital humanists, Bourdieu, and sociologists of literature, this dissertation explores how the methods of “distant reading” can be used to develop new classes of critical insights about law. After situating the project theoretically, this dissertation reports on a series of new computational studies about Canadian law. Chapter 1 measures Canadian statutory and regulatory law, showing that law has grown unevenly over the past decade and a half. Chapter 2 uses new artificial intelligence to transcribe and analyze Supreme Court of Canada hearings, revealing gendered and linguistic speaking patterns among justices. Chapter 3 shows how computational methods can be deployed to detect inconsistency and discord in a jurisprudence, in this case Canada’s law of terrorism. Chapter 4 uses machine learning to study refugee law jurisprudence, particularly showing how it has developed over the past decade. Chapter 5 leverages new computational techniques to analyze Social Security Tribunal of Canada decisions regarding employment insurance appeals and suggests that new computational analyses might usefully change legal education. It concludes by considering how some visions of computational legal analysis—despite the sweep and scope of their projects—are part of old and traditional visions of what law is.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.031 | 0.052 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.006 | 0.007 |
| Science and technology studies | 0.008 | 0.079 |
| Scholarly communication | 0.015 | 0.018 |
| Open science | 0.003 | 0.005 |
| Research integrity | 0.002 | 0.004 |
| Insufficient payload (model declined to judge) | 0.006 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".