Monday evening was the launch of the latest ‘thematic’ issue of the UNSW Law Journal. This issue’s theme is ‘The Individual Judge’, which is also the title of Kiefel CJ’s 2014 speech and paper, where she first said ‘collegiality is not compromise’. The paper is one of three where she has defended the High Court’s practice of attributing judgments largely written by one judge to all the judges who agreed with it. While there is much of interest in the new issue of the UNSWLJ, only one of the articles responds directly to Kiefel CJ’s stance. The paper by Andisheh Partovi et al sets out five arguments for correctly attributing authorship of judgments: ensuring individual accountability (an argument I also put here), discouraging free riding, and serving the interests of judges, academics and lawyers. More importantly, the authors acted on their views by outing the likely authors of the joint judgments of the High Court from 1987 to 1995, when Sir Anthony Mason was Chief Justice. Needless to say, their list is absolutely fascinating.
So, who, out of the eight judges who sat in that period, likely wrote the most important joint judgments of the Mason Court? Mason. In particular, the authors think Sir Anthony was very probably the author of the following five constitutional joint judgments
- Union Steamship Company of Australia v King (1988) 166 CLR 1 (on extraterritorial state laws)
- Cole v Whitfield (1988) 165 CLR 360 (the key modern case on s 92)
- Barley Marketing Board (NSW) v Norman (1990) 171 CLR 182 (on s 92)
- Smith Kline & French Laboratories (Aust) Ltd v Cth (1991) 173 CLR 194 (on High Court appeals)
- Botany Municipal Council v Federal Airports Corporation (1992) 175 CLR 453 (on s 109)
out of 11 that they could attribute with a high probability, and a further 13 non-constitutional joint judgments (including the Newspaper Rule and Spycatcher cases, as well as the likes of Myer Emporium and M v M) out of 76 such judgments. More generally, they found that Mason CJ, Brennan and Deane JJ together wrote nearly half of the joint judgments of the Mason Court. The method used to make these determinations is described in the paper at length. In short, it requires processing edited text files of the judgments for the presence and frequency of various words and word combinations, and parts of speech and combinations. The paper includes table of the various judges’ linguistic tells, for example identifying Mason’s use of phrases such as ‘the respective’ and ‘might be’, and his frequent use of ‘Wh-‘ pronouns, compared to (say) Gaudron J ,who likes to use ‘context’, ‘and thus’ and adverbs.
If the paper’s list of the very probable authors of some 87 Mason Court joint judgments isn’t interesting enough, there is also the potential future uses of their methodology. As the authors say:
Future research could examine attribution of authorship in joint judgments in other eras of the High Court’s history, such as the French Court, or for other Australian courts.
Such research could yield insights into shifts in High Court practice. For instance, I’d be curious to see whether the paper’s finding that all judges in the Mason Court wrote at least one criminal law joint judgment would be replicated in more recent courts. In his foreword to the thematic issue, former Chief Justice Robert French said that Partovi et al’s paper:
makes for fascinating, if rather unsettling, reading. It might encourage movement towards express attribution of authorship in judgments, particularly if the technique of computational linguistics creates a non-trivial probability of identifying the wrong person.
The paper in fact canvasses the risk of error at length, noting various reasons why the chance of mis-attribution of the high probability judgment writers, while non-trivial, is not high, while urging a ‘cautious approach’ in the case of a further thirty or so other judgments where the probability of the attributions is less high. I fully expect that all future academic works on any joint judgment of the Mason court will include a footnote to Partovi et al’s paper, noting any high probability attribution they make. And I would be thrilled if they or someone else does the hard work to allow similar footnotes for all High Court joint judgments, including future ones.