Judicial overreach, the concept that courts have exceeded their constitutional power to interpret and develop common law and strayed into political decision making or policy development, has long haunted both academic and political writing on constitutional affairs. Commentators increasingly frame judicial overreach as systemic, arguing that there is a trend towards judicial activism, as opposed to, say, a handful of erroneous decisions.
But, how would one know whether there was systemic judicial overreach? Much has been written on the limitations of anecdotal reasoning, as well as, on the other hand, on what a solely empirical approach omits. As such, which diagnostic toolkit to be reached for is not immediately obvious. Nonetheless, getting the right one is clearly essential in assessing whether systemic judicial overreach really exists. This short blog post seeks to introduce the two major methodologies which are currently used in the debate over judicial overreach, explaining the opportunities and risks involved in each.
A good deal of the scholarship on the phenomenon is written with the assumption that this overreach exists in the first place and, consequently, advises on which actions which should be taken to curb it. Typically, such scholarship has used an ‘anecdotal’ approach to make the case for the existence of judicial overreach. This approach, borne out through, for example, the Judicial Power Project’s list of ‘50 Problematic Cases’, looks to eminent decisions, particularly within the realm of public law, to examine what they have to tell us about judicial overreach. Cases such as R (Evans) v Attorney-General [2015] UKSC 21 are often at the forefront of work in this vein, and are considered as qualitative evidence, emblematic of the wider problem. Although as Paul Craig has remarked, “I do not mind how many times Evans is cited, it still only counts as one case”.
This anecdotal approach can serve as an impactful way of illustrating judicial attitudes, though, and permits a level of detailed qualitative analysis which is lacking from other methodologies. Decisions of the highest appellate courts, particularly ones in which there is an extensive discussion of the relationship between the judiciary and the legislature (such as Jackson v Attorney General [2005] UKHL 56) can provide an important window into how the courts approach constitutional questions.
However, one may just as easily find cases which demonstrate judicial restraint. The recent decision of RA, AA and MZZ v Foreign, Defence and Home Secretary [2026] EWCA Civ 3 (summarised in this case note) saw the Court of Appeal set aside a mandatory order made by the Divisional Court on the basis that it transgressed the constitutional boundary between the executive and the courts’ role. In some senses, then, this anecdotal form of evidence gathering has an obvious problem: that, given the vast number of judicial decisions made each year, one can usually find what they are looking for, and consequently, it can prove very difficult to weigh the findings of each side to diagnose the overall approach of the courts.
A further limitation is that anecdotal reasoning is particularly ill-suited to capturing the day-to-day work of courts hearing judicial review claims. It is striking that many of the decisions which are heralded as examples of judicial overreach are decisions of the highest appellate court. Such cases, though, more often turn on distinctly constitutional facts, because they necessarily involve an arguable point of law of general public importance. Indeed, both Miller 1 and Miller 2 begin by the Supreme Court explicitly noting what it is that they are and are not deciding, given the extraordinary facts of the cases.[1] Cases such as these are far removed from the everyday work of the lower courts, yet arguably, it is the decisions of these lower courts which set the tone of approaches to judicial review. Where anecdotal reasoning is used to justify widespread reform to the law of judicial review, it is particularly important that this reasoning accurately represents what is happening in the lower courts.
Even though anecdotal approaches are ill-suited to shedding light on these more ‘ordinary’ instances of judicial review, thus far many consultations into how the law of judicial review should be reformed have included much anecdotal evidence. For instance, the Independent Review of Administrative Law Report (2021) identified concerns of judicial overreach as a topic of interest. In a short summary of the evidence of judicial overreach which had been presented to the Review. The report described how:
“Concern has been expressed to us that examples of such judicial overreach have begun to creep into the caselaw. Professor Jason Varuhas argued in his submission of evidence to us that there are “increasing instances of doctrines or approaches which are very difficult to reconcile with the idea of judicial review as a supervisory jurisdiction”.
As an example (of several), he pointed to the Divisional Court’s decision in R (Litvinenko) v Secretary of State for the Home Department.[2] In that case, he argued that the Court set aside the Secretary of State’s decision not to hold a public inquiry into the death of Alexander Litvinenko simply because it “disagreed with reasons given by the Minister””.[3]
It may well be the case that the Divisional Court’s decision in R (Litvinenko) v Secretary of State for the Home Department is an example of judicial activism. It may also be the case that a number of cases can be found which are, like Litvinenko, indicative of such activism. This activism may even be systemic, but an anecdotal form of reasoning is an unconvincing way of demonstrating a widespread, much less systemic, attitudinal issue on the part of the courts. Indeed, the Independent Review approached this evidence cautiously, advising ‘judicial restraint’ and highlighting a number of cases in which judicial dicta explicitly recognised the constitutional limits of their role.[4]
There are alternatives to anecdotal evidence. One is a more empirical approach which seeks to evaluate the behaviour of the judiciary by examining judgment metrics such as how often the government succeeds, or which grounds of review are successful. Empirical or ‘systematic’ studies have focussed on (amongst many other issues) particular duties within judicial review, the approach that the courts take in reviewing ombud decisions, and the use of the Human Rights Act 1998 to challenge statutory instruments.
There are real benefits to this research methodology, which are set out in greater detail in this excellent piece by Lewis Graham. The empirical approach better captures what Paul Craig terms the “normal incidence of judicial review” by providing a comprehensive overview of what is happening in judicial review claims. This also counterbalances one great difficulty with the anecdotal approach, because under the empirical approach, equal weight is given to each case within the dataset, whatever decision the court came to. This removes the tendency to disregard contrary evidence and assists in creating a clear and accurate picture of what is happening actually happening across judicial reviews.
It is also true, though, that the empirical approach has an outcome-driven focus on decisions, in which the overall result of the case determines how we characterise the contribution it makes to the judicial overreach debate. For example, if a case if decided in favour of the government, that is typically all that is taken into account by this method. But, many cases which have a result which is ‘friendly’ to an orthodox understanding of the judiciary’s role and Parliamentary sovereignty include judicial dicta, sometimes very powerful and authoritative judicial dicta, which cut in the opposite direction. As Brian Christopher Jones has written, “these wider aspects of the judicial voice are not easy for empirical studies to capture, but they can produce significant ripple effects.” An empirical approach alone cannot account for the nuances which can be contained within one case and the subsequent impact which judicial dicta may have on constitutional boundaries.
It is difficult to design a perfect diagnostic toolkit, but it is important to be aware of the limitations of the toolkit one is using when seeking to diagnose the attitude of the judiciary. Each approach has merits and demerits and combining the two may facilitate a more representative and balanced approach moving forward, because the accuracy of the diagnosis of judicial overreach is important. Where reform is to be made to the law of judicial review, this change must be evidence-backed. This is a time in which some have argued forcefully that the courts need to be ‘reined in’, particularly in respect to judicial decisions concerning the European Convention on Human Rights.
Such arguments are crafting policy: at the end of 2025, the Conservative Party announced that the next Conservative Government will leave the European Convention on Human Rights, following a review into “the use of legal challenges to block government’s doing what they were elected to do”. Reform UK have taken the same position. Where a diagnosis of judicial overreach is, in part, being used to mark the path that policy will take, particularly in the realm of fundamental rights, it is essential that we make the correct diagnosis.
Cassandra Somers-Joce.
Cassandra Somers-Joce studied law at Oxford University, where she is currently a non-stipendiary lecturer. She has previously taught public law at King’s College London, in addition to holding research roles at the University of York and Durham University.
The Constitution Society is committed to the promotion of informed debate and is politically impartial. Any views expressed in this article are the personal views of the author and not those of The Constitution Society.
[1] Miller 1 at paragraph [3]; Miller 2 at paragraph [1].
[2] [2014] EWHC 194 (Admin).
[3] See the Independent Review of Administrative Law (2021) Report, Page 58, Paragraph [3.7].
[4] See the Independent Review of Administrative Law (2021) Report, Pages 61- 63, Paragraphs [3.18]- [3.24].