Long-term readers of this blog will remember that IP Draughts had a phase of running text through the Bla Bla Meter. “Bla bla” is, he thinks, a polite version of the word that the creators of the tool had in their minds. A low score is good, a high score bad.
Court judgments deal with complex issues, and it is not always easy to reduce those issues to very simple language. But it can be done. In 2016, Peter Jackson LJ famously wrote a judgment in a family law case so that the children concerned in the case could understand it. His bla-bla score? A phenomenally good 0.07.
This was an exceptional case, and IP Draughts wouldn’t expect this approach in most cases. He found that some of the senior judges that he most admired had scores in the 0.16 to 0.19 range, while some others strayed into the .20s.
IP Draughts thought it would be interesting to see how the latest batch of judges to be appointed to the UK Supreme Court score, based on a random selection of passages from their judgments in the Court of Appeal. Lord Sales, Lady Arden and Lord Kitchin are good lawyers, but are they good writers? The results are in:
In the R v General Medical Council case (2018), Lord Justice Sales scored 0.18 and Lady Justice Arden scored 0.17. In the Nestle v Cadbury (2017) case, Lord Justice Kitchin scored 0.16 in one passage and 0.19 in another.
All very good scores. For comparison purposes, IP Draughts tested parts of the speeches of Lords Sumption, Briggs and Hodge in the Warner Lambert v Generics (2018) case in the Supreme Court. They all had similar scores in the 0.17 to 0.19 range.
IP Draughts stresses that this is not a measure of the quality of the decision, just the way it is expressed. But he is biased in favour of people who make an effort to express themselves simply and clearly.
By way of contrast, the judgment of Marcus Smith J in the CMA v Concordia International case (2019) scored 0.25. A reasonable score by an intelligent judge, but not fantastic.
According to this BBC news report, some judges in India have had their judgments sent back to them for rewriting, because the Indian Supreme Court found them unintelligible. The report quotes as an example the text below:
However, the learned counsel…cannot derive the fullest succour from the aforesaid acquiescence… given its sinew suffering partial dissipation from an imminent display occurring in the impugned pronouncement hereat wherewithin unravelments are held qua the rendition recorded by the learned Rent Controller…
The summum bonum of the aforesaid discussion is that all the aforesaid material which existed before the learned Executing Court standing slighted besides their impact standing untenably undermined by him whereupon the ensuing sequel therefrom is of the learned Executing Court while pronouncing its impugned rendition overlooking the relevant and germane evidence besides its not appreciating its worth. Consequently, the order impugned suffers from a gross absurdity and perversity of misappreciation of material on record.
IP Draughts put this text through the Bla Bla Meter and was disappointed to discover that it only scored 0.25. It should have been off the scale! IP Draughts has been in touch with the people who run the Bla Bla Meter. They acknowledged that there are some areas where the algorithm doesn’t work well, and that the above example should have received a much higher score.
Just for fun – though he may regret it – he put a couple of US Supreme Court judgments through the test. He randomly alighted on SAS Institute v Iancu, Director USPTO (2018) and tested extracts from the judgments of Gorsuch J and Breyer J. One scored 0.28 and the other 0.29. The consistency between their scores suggests that a different drafting style is favoured in the US Supreme Court.
Finally, he decided to put the famous Florence speech (2018, about Brexit) of Theresa May, the British Prime Minister, through the test. You might think that a politician would speak very plainly and have a score of less than 0.10. Her score? 0.21.
Oh, and this blog article (excluding the quotation from the Indian judge)? 0.06 – phew!