It is not difficult to think of constitutional rules that are criticized, defended, or often both, on normative grounds that are more or less fact-free—not for what they actually are, but for what their critics or defenders believe they are or ought to be. In the United States, the Citizens United decision comes to mind. In the United Kingdom, Lewis Graham argues, a similar fate has befallen section 3 of the Human Rights Act 1998 (“HRA”), which provides that “[s]o far as it is possible to do so, primary legislation and subordinate legislation must be read and given effect in a way which is compatible with the … rights” protected by the European Convention on Human Rights.
Graham notes that section 3, “perhaps more than any other provision in the HRA, has been subject to serious criticism in the literature.” He does not mention the Bill of Rights Bill, which the last Conservative government introduced in an ultimately failed attempt to replace the HRA; if enacted, it would have eliminated section 3. For the record, although very critical of the Bill as a whole, I was sympathetic to that aspect of it at the time.
The reason for the hostility section 3 has aroused is that rights-compatible interpretation might, in the hands of a sufficiently adventurous court, amount to a rewriting of the legislation being “interpreted,” and so risk subverting democracy, the separation of powers, and the rule of law. A leading judgment has indeed suggested that, in applying section 3, courts can assign to statutory language an “unnatural or unreasonable” meaning, provided it “is intellectually defensible”, and “do considerable violence to the language and stretch it almost (but not quite) to breaking point”.
Graham’s argument, however, is that—despite these and other provocative statements in some early section 3 cases, on which scholarship (and, I would add, teaching) tends to focus—UK courts have not been so adventurous as to justify this criticism. He develops this argument by identifying and reading all the cases where judges have deployed section 3, and assessing the ways in which they have done so. Beyond the importance of this study for scholars and others interested in the evolution of the UK’s constitution, it offers valuable lessons on the usefulness, and also the limits, of good empirical scholarship in constitutional law.
Graham observes that “after an initial (perhaps understandable) enthusiasm for section 3,” its use has levelled off at a handful of cases a year, from all the levels of UK courts combined. Hardly a torrent of judicial rewriting of legislation, then, but a steady trickle. About one case in three concerns the right to a fair trial, whether in criminal law (which altogether supplies about one in five section 3 cases), or in a civil context. Graham suggests that this “reduces the risk of judges improperly trespassing upon the terrain of other constitutional actors,” which seems fair.
Reinforcing the theme of judicial non-radicalism are Graham’s findings as to the manner in which legislation is “read and given effect” so as to be compatible with Convention rights. Although the cases say that this can be done in fairly disruptive ways, including by effectively replacing statutory language with different, rights-compatible terms, this is rarely done. The most popular technique is to adopt a novel interpretation that “obviates the need for any explicit addition or deletion of the legislative text,” while “additions” are largely preferred to “deletions.” Just how different these methods really are surely depends on the individual case. This is one of the inevitable limitations of empirical scholarship that concerns itself with large numbers of cases rather than focusing on individual examples. But Graham is probably right that the courts’ choices reflect a desire to give effect to legislative intent, adjusting legislation for details that may have been overlooked rather than re-writing it in a manner that would disrupt its scheme.
The longest section of Graham’s article, and perhaps the most compelling in making the case for judicial moderation, if also the most complicated, concerns the self-imposed limits courts have observed in applying section 3. For one thing, “the case law is replete with examples of judges seeking to identify and preserve the intention of the legislature,” though it also contains the odd indication of a willingness to disregard this—normally—central factor in statutory interpretation. For another, although the application of section 3 often means that courts will frequently give statutory words unexpected meanings, the terms of a statute impose limits on this approach, and sometimes prevent it altogether. Analogous cases and on-point precedents matter, as do the views of the parties—most notably the government, which would rather have the court rely on section 3 “than having to respond to a declaration of incompatibility,” which is likely to follow if doing so is not possible. Last but not least, courts may refrain from relying on section 3 when the choice of a rights-compatible interpretation involves considerations of fundamental policy or morality, or consequences judges may not be in a position to anticipate.
Graham concludes that “[s]ection 3 is powerful, but is perhaps not as radical as it is sometimes portrayed as being.” Indeed, “on the rare occasions where the courts have adopted a particularly radical interpretation, this has been at the insistence of all parties to the dispute,” including the government. They are aware of the constitutional limitations on their role, and demonstrate this awareness in how they decide cases—if not always in how they talk about their decisions. Critics should not focus on rhetoric which overtakes the reality, but on what courts actually do.
I have more to say about what the takeaway from Graham’s article for section 3 critics such as me. First, though, let me note the way in which it enriches the conversation and can serve as a model for constitutional scholarship. Writing about a few salient cases, let alone about a few punchy statements made in a few salient cases, is easy. Insofar as this remark is critical, the criticism is directed at myself as much as at any reader who may feel targeted by it. But my point is not that narrowly-focused scholarship has no value. It is, rather, that a survey of an important area of the law, such as that which Graham’s article provides, can, as in this case, supply essential context within which the salient cases and punchy statements can be assessed, and perhaps discounted. If the striking examples set out the direction in which the case law develops, then a focus on them is justified. But if they turn out to be outliers, then the question of why that is so becomes more interesting than criticism.
A survey is better positioned to supply this context when, as here and indeed elsewhere in Graham’s scholarship, it is backed by numbers and not just a vibe-check. But the presence of figures alone is not enough; the trick is to know what to count. The significance of Graham’s study lies in his ability to count not merely outcomes, let alone some personal characteristics of the judges who reached these outcomes, but features of the reasoning of the cases he studies. That is to say, although his work presents itself as empirical, and has the advantage of big-picture rigor that is the attraction of empirical scholarship, it is infused with a doctrinal scholar’s interest in legal reasoning, instead of abstracting it away as empirical legal scholarship too often does. This is where the true value of Graham’s work lies.
That said, the limits of such work, like its strengths, have to do with the questions it asks. This is, of course, not a criticism of Graham’s article, but simply an acknowledgment of the fact that no piece of scholarship, and no method, is by itself sufficient to ask and answer all important questions about a given topic. By design, empirical work can no more than hint at the causes of the trends it brings to light. And while it can and should inform normative views, it cannot and should not define them.
With respect to article 3 of the HRA, Graham makes a strong case that the courts have been more thoughtful than their critics often give them credit for, and more moderate than they themselves sometimes let on. His article is not meant to explain this discrepancy, although his recent post on the UK Constitutional Law Blog, which explores the role of ideology in the ostensibly purely merit-based appointments to the UK Supreme Court might point at one part of the explanation. Another might lie in the existence of a tendency by the various actors in the UK’s “unwritten” constitution to maintain an uncertain equilibrium by staking out ambitious claims without prosecuting them to their full extent, but that is a far more complicated topic than can be explored here.
Graham recognizes that “critics may allege that so long as a more permissive, activist approach is possible we should remain worried,” but argues that we should, at least, discount the possibility, in light of the experience he describes. That is fair, so far as it goes. But that does not answer the question whether Parliament was entitled to delegate a legislation-shaping power to the courts even if they expected it to be used with moderation, or whether this delegation should be retained even though we now have good reason to think that it is and will be used thoughtfully and without radicalism. Even within these constraints, it remains a delegation of legislative power, and is vulnerable to criticism on that basis (purely normative criticism, of course, since there is no question of a non-delegation doctrine applying in the United Kingdom).
However, even if one is inclined to remain critical of section 3 on this basis, as I am, Graham’s article has the virtue of clarifying what lines of criticism can be productive, and which ones should be abandoned. Our views, like those of section 3’s supporters, will be better informed and more focused. Graham’s blend of empirical work and doctrinal reasoning helps us all become better thinkers.
