Short answer, “no”…
… but research published in October by a team based at University College London on whether the automatic analysis of cases of the European Court of Human Rights could predict outcomes concludes that “Our models can predict the court’s decisions with a strong accuracy (79% on average)”.
The project was undertaken by Dr Nikolaos Aletras, currently at Amazon.com but previously at the Department of Computer Science, UCL, Dr Vasileios Lampos, also of the UCL Department of Computer Science, Dr Dimitrios Tsarapatsanis, of Sheffield Law School, and Dr Daniel Preoţiuc-Pietro, of the University of Pennsylvania’s Positive Psychology Center. They looked at 584 applications under Articles 3 (inhuman or degrading treatment), 6 (fair trial) and 8 (private and family life), analysing features for the sections headed Procedure, Circumstances of the Case, Facts, Relevant Domestic Law and Practice, The Law (ie, the relevant provisions of the ECHR), and the full case. They then applied an Artificial Intelligence algorithm to find patterns in the text. To prevent bias, they selected an equal number of violation and non-violation cases.
According to Dr Lampos, the most reliable factors for predicting the court’s decision were the language used and the topics and circumstances mentioned in the case text. The Circumstances section of the text includes information about the factual background to the case: they observe that
“The consistently more robust predictive accuracy of the ‘Circumstances’ subsection suggests a strong correlation between the facts of a case, as these are formulated by the Court in this subsection, and the decisions made by judges. The relatively lower predictive accuracy of the ‘Law’ subsection could also be an indicator of the fact that legal reasons and arguments of a case have a weaker correlation with decisions made by the Court.”
They conclude that their empirical analysis
“indicates that the formal facts of a case are the most important predictive factor. This is consistent with the theory of legal realism suggesting that judicial decision-making is significantly affected by the stimulus of the facts”.
Dr Aletras said that though he and his colleagues did not see Artificial Intelligence replacing judges or lawyers,
“we think they’d find it useful for rapidly identifying patterns in cases that lead to certain outcomes. It could also be a valuable tool for highlighting which cases are most likely to be violations of the European Convention on Human Rights.”
For someone who tends towards the legal realist view of how judges decide cases, the outcome is not remotely surprising. That said, however, it will be interesting to see whether the authors’ suggestion
“that published judgments can be used to test the possibility of a text-based analysis for ex ante predictions of outcomes on the assumption that there is enough similarity between (at least) certain chunks of the text of published judgments and applications lodged with the Court and/or briefs submitted by parties with respect to pending cases”
can be confirmed in practice. It would have been useful to test that hypothesis on cases for which the actual outcome differed from that anticipated by legal commentators or where chamber judgments were subsequently overturned by the Grand Chamber.
It would also be interesting to see the technique applied to Article 9 cases. My suspicion is that they are probably trickier to analyse than cases decided under Articles 3 and 6 (though not, it should be said, under Article 8) because the facts in Article 9 cases only get you so far, given the court’s tendency to grant a wide margin of appreciation in matters of thought, conscience and religion. If recent cases are any guide, a recital of the facts that included the words “hijab” or “burqa” and “France” would be a racing certainty for nul points.
- 2016) Predicting judicial decisions of the European Court of Human Rights: a Natural Language Processing perspective. PeerJ Computer Science 2:e93. (
- Rosalind English, UKHRB: Computer algorithm predicts most Strasbourg judgments. [To whom thanks for flagging this up: we missed when it first came out.]