Who our judges are matters. The reasons for this are well rehearsed.Footnote ¹ It matters because everyone should have an equal opportunity to become a judge regardless of who they are. It matters because a diverse judiciary, which is representative of society, has greater democratic legitimacy. Finally, it matters, we are told, because a more diverse judiciary produces better quality decision-making as a variety of perspectives are considered.Footnote ² So much is true. But who our judges are also matters because once we accept this, we have to let go of Lord Reid’s fairytale judge: the judge hidden in Aladdin’s cave and keeper of the ‘magic words’.Footnote ³ This is nothing new. For some time now, it has been generally accepted that, in Lord Reid’s words, ‘for better or for worse judges do make law’.Footnote ⁴ And if this is true – and I think it is – if judges are, at least on some occasions, making law, then who these individual judges are matters. Which brings us to Lewis Graham’s book: Judicial Individuality on the UK Supreme Court. Footnote ⁵

Graham starts from the position that judicial neutrality is impossible.Footnote ⁶ Judges, he suggests, are human with their own ‘values, preferences, ideologies, ideas and biases’.Footnote ⁷ And, just like the rest of us who often bring our own life experiences into account when making decisions, it is ‘impossible to imagine that judges’ do not do the same.Footnote ⁸ In other words, judges bring their individual preferences into their decision-making and so into their judgments.Footnote ⁹ This is not controversial. However, Graham seeks to push this further, revealing not only that judges do this but, more interestingly, characterising and determining the extent and impact of their individuality on their decision-making. The result is a significant contribution to the growing literature on judicial studies.Footnote ¹⁰

Focusing on the UK Supreme Court during its first decade, Graham seeks to identify not only which preferences have more influence on the Justices and how this influence might be measured, but also how the Justices differ from each other as a result of their preferences.Footnote ¹¹ This is no mean feat. He reviews a total of 682 cases, somewhat modestly referring to his methodology as ‘an exercise in sophisticated counting’.Footnote ¹² He uses a number of different measurements in order to discover each Justice’s individual preferences and the effects of these on their judgments,Footnote ¹³ identifying categories of information about each Justice sitting in the case: the type of judgment (lead, dissenting or concurring judgment) and length, whether they sided with a particular type of party, the outcome and whether this indicated an approach to certain areas of law, and the type of legal reasoning and tools used.Footnote ¹⁴

Of course, there are, as Graham notes, limitations with his methodology, as there are indeed with any research. Graham recognises and addresses them, noting, for example, the subjectivity of manual coding, neatly stating that ‘if judges are human, so are coders’.Footnote ¹⁵ Nevertheless, the use of a coding table and set guidelines for each category, and cross-checking by outside coders ensures that his coding is robust.Footnote ¹⁶ He also appreciates that his methodology is only able to identify one preference at a time and, as such, is not able to demonstrate conclusively that that preference alone caused the judicial behaviour in question.Footnote ¹⁷ Nevertheless, he argues that ‘correlation can be very significant’,Footnote ¹⁸ particularly when comparing different Justices’ behaviour and preferences.

Chapters 3 to 8 set out the results from Graham’s ‘sophisticated counting’.Footnote ¹⁹ Unsurprisingly given the number of cases considered, there is a tremendous amount of data within these chapters and there is the risk that the reader could become overwhelmed. Graham manages this well. Each chapter deals with a particular category of information gathered by Graham during his review, setting out the data clearly, using tables and graphs effectively to bring the data to life. He also analyses the data in several different ways to give a more reliable picture of the emerging patterns and, once his analysis is complete, he often ties this to views that the Justices have expressed judicially and extra-judicially.Footnote ²⁰

Many of the emerging themes about the individual Justices are fascinating. In Chapter 3, for example, on ‘the form or nature of the judges’ output’Footnote ²¹ we discover that Lord Sales was most likely to give the lead judgment,Footnote ²² Lord Hope was most likely to concur,Footnote ²³ and Lord Rodger was most likely to dissent.Footnote ²⁴ Chapter 4 goes on to look at ‘how likely an appeal is to be allowed, and whether appeals from certain courts are treated more favourably than others in this respect’.Footnote ²⁵ Whilst Graham finds that there is ‘less evidence of a substantive difference between the judges in the appeals context’,Footnote ²⁶ there are some. Lord Phillips and Lady Hale’s higher rate of allowing appealsFootnote ²⁷ is likely, he suggests, because both judges sat as Presidents, possibly making them less deferential to lower courts – although Lord Neuberger bucks this trend.Footnote ²⁸

In subsequent chapters, Graham shifts his focus to consider substantive areas of law and legal reasoning. Chapter 6 looks at tort, contract, crime, EU law, devolution, immigration and asylum, and tax cases. Despite involving small datasets,Footnote ²⁹ it is still possible to identify individual judicial preferences or tendencies. Lord Sumption, for example, was less likely to find a breach of EU lawFootnote ³⁰ or to adopt a pro-tax position,Footnote ³¹ while Lord Kerr was more likely to find a party liable in tortFootnote ³² and to uphold a devolution challenge.Footnote ³³ Chapter 7 focuses solely on human rights law because, as Graham argues, these cases (of which the Supreme Court hears a large number, sometimes with larger panels) often involve legal and political controversy where the Justices might use their discretion.Footnote ³⁴ And again, Graham’s attention to detail allows you not only to see the Justices as individuals, but to confirm suspicions as to the stance they might adopt. It is unsurprising, for example, that the much missed and respected Lord Kerr was ‘the most rights-friendly judge’,Footnote ³⁵ often by a significant margin.Footnote ³⁶ Then in Chapter 8, Graham moves on to consider how the Justices approach statutory interpretation, the use of precedent, jurisdiction, and remedies. Once again the data is instructive. Lady Hale, for example, was more likely to accept that the Supreme Court had jurisdiction to hear a case, matching views that she had expressed judicially.Footnote ³⁷

Without doubt, there is a risk that the amount and richness of the data might overwhelm the reader. Graham seems to appreciate this, using Chapter 9 to gather together key themes in individual profiles of the Justices in which he attempts to establish their judicial and political ideology. These profiles are a fascinating read and together they ably demonstrate his key thesis: that the Justices are individuals, with their own preferences, and that these on occasion inform their judgments.

There, of course, have to be limits to any research: a line must be drawn somewhere. Coding the first ten years’ worth of the Supreme Court’s cases seems a clear and reasonable line. However, Graham goes further in Chapter 10, providing an assessment of the Supreme Court from 2020 to 2022. He notes the changes to the composition of the Court in 2020 and, following a review of the cases, concludes that ‘the Court has more recently adopted a position which is generally more conservative in nature but is not characterised strongly by judicial restraint across the board’.Footnote ³⁸ This tracking of the changing nature of the Supreme Court is valuable not only for those of us interested in the way in which the Supreme Court works but to anyone interested in the courts and judging. I hope that Graham continues to provide regular updates.

You will have noticed that this review has not yet discussed Chapter 5, in which Graham discusses ‘the success rates of different types of litigant’, which he argues ‘may allow some inference of [the Justices’] underlying sympathies’.Footnote ³⁹ He does so by forming a view as to whether a particular Justice in a given case is more or less sympathetic to the state, ‘socio-economic underdogs’ and women. This is the chapter that I was most looking forward to reading. After all, if Graham’s research shows that the Justices prefer certain types of parties, and this preference may be linked to who they are, then this might – indeed should – have important consequences for who our judges are. Unfortunately, at least when it comes to social class, this does not appear to be case.Footnote ⁴⁰

Graham begins by noting that the Supreme Court Justices are predominantly from higher social class backgrounds, with many having ‘attended elite schools such as Eton’ and later either Oxford or Cambridge Universities.Footnote ⁴¹ He also suggests that they mostly remain within this social class, with most of them having careers at the commercial bar before joining the bench.Footnote ⁴² He says that it is ‘easy to see why some have therefore suggested that judges tend to side with certain parties out of self-interest, behaving in a manner which protects their own social interests as well as those of a similar status’.Footnote ⁴³ Graham then attempts to address this issue by investigating how likely it is that the Justices decide in favour of a group of claimants he categorises as ‘socio-economic underdogs’.Footnote ⁴⁴

Of course, this simply prompts a prior question as to who is a socio-economic underdog? Graham includes here individuals who might be most obviously disadvantaged: ‘those who tend to hold a particularly disadvantaged position in society more generally, such as homeless people, prisoners, immigrants or those facing deportation and those relying on social welfare’.Footnote ⁴⁵ However, his definition also extends beyond such individuals: where ‘in a given case, there exist[s] a recognised social or financial disparity between the parties’ and so the ‘weaker party’ is disadvantaged, then they should be classified as an underdog.Footnote ⁴⁶ So, for example, in a dispute between a tenant and a landlord, the tenant is the socio-economic underdog; and in a dispute between a worker and an employer, it is the worker and so on.Footnote ⁴⁷

Graham finds that on an ‘individual level, judges are – if only just – more likely to side against the underdog than they are to side with them’.Footnote ⁴⁸ If so, then it might also be suggested that the Justices, who are drawn predominantly from the higher social classes, are more likely to side against a party who is from the lower social classes. Put another way, the Justices are more likely to decide in favour of a party that is from the higher social classes, like themselves. If this is the case, this is incredibly significant: it suggests that the Justices are deciding cases in line with their own class interests.

To be clear, I do not think that this is what Graham intends to suggest. There are two reasons for this. The first is methodological. Graham’s ‘socio-economic underdog’ does not necessarily accord with a party’s social class. Social class is a difficult concept to define, and Graham does not include any definitions and proxies of social class in his research.Footnote ⁴⁹ Furthermore, his concept of a socio-economic underdog is relative between the parties before the Court: it is not necessarily an indication of either party’s position outside this relationship or, indeed, the courtroom. Rather his coding suggests that ‘an “ordinary” citizen’ is automatically ‘in a weaker position than public authorities or the government’Footnote ⁵⁰ notwithstanding their economic, cultural, and social capital. This is problematic. Take, for example, Mueen-Uddin v Secretary of State for the Home Department – a case that falls outside Graham’s time period.Footnote ⁵¹ Mr Mueen-Uddin brought a claim against the Home Secretary for libel and breach of statutory duty pursuant to the General Data Protection Regulation. According to Graham’s methodology, Mr Mueen-Uddin is a socio-economic underdog. In fact, he is a founding member of the Muslim Council and has held ‘a number of prominent public and charitable positions in British society’,Footnote ⁵² including at Musim Aid whose total gross income in 2023 was over £21 million.Footnote ⁵³ Of course it is not possible to identify definitively Mr Mueen-Uddin’s social class from these facts, but the point is that when it comes to social class, things – whether individually or relationally – are not as clear cut as Graham’s research suggests.

The second reason lies in the scope of Graham’s field of study. From the outset he is clear that he ‘is not directly concerned with the admittedly fascinating question of why those preferences might be held’ preferring to leave that to ‘others’.Footnote ⁵⁴ Given the ‘ample evidence’ set out by Graham ‘that the personal preferences and legal and political ideologies of judges affect decision-making on the UK Supreme Court’,Footnote ⁵⁵ it is hard to resist the conclusion that this is a missed opportunity. After all, insofar as a person’s preferences and ideologies are influenced by their identity characteristics and their life experiences (which are themselves informed by these characteristics), then it follows that this must be the case – at least on some occasions – for our judges. Therefore, who the judge is can, on occasions, make a difference to how a case is decided and/or the outcome reached. This is not a novel argument. Sherrilyn A Ifill makes the point in relation to race,Footnote ⁵⁶ as does Erika Rackley in relation to gender.Footnote ⁵⁷ Greater diversity would not only introduce a wider variety of preferences and ideologies to the bench, but given that decisions are better for having been informed by a variety of perspectives, it might also improve the quality of decision-making.Footnote ⁵⁸ This is important for the judiciary as a whole but particularly so for the Supreme Court given its especial lack of diversity.Footnote ⁵⁹ Graham recognises this diversity argument at the end of his book:

But why stop here. After all, if we can show that the judges may have certain preferences as a result of their identity characteristics and consequential life experiences, the argument for diversity is strengthened even further. Of course, the responsibility is not solely Graham’s. Either way his research is a valuable part of this process. It is now for the un-named others to use, and build on, his research to demonstrate why the judges have these preferences, to continue to make the argument that who our judges are matters – not least because it offers the best chance of a diverse judiciary at all levels.

In the meantime, law students, lawyers, parliamentarians, and judges alike would benefit from reading Graham’s book. After all, it can only be a good thing to know more about the judges who hold ‘the most power, and [deal] with the most interesting and difficult cases’,Footnote ⁶¹ and the individuals beneath the robes.