Obiter remarks now binding: Court of Appeal (radically) alters doctrine of precedent

04 May 2020

Public Law and Judicial Review

Last week a five-judge Court of Appeal handed down R v Barton and Booth [2020] EWCA Crim 575. The decision establishes that R v Ghosh [1982] QB 1053, on dishonesty in criminal law, is no longer binding. Barton and Booth is a bold decision with profound implications for the doctrine of precedent. It makes for interesting reading to lawyers operating in all areas of law.

The basic question in Barton and Booth was whether the Court of Appeal should follow its own decision of Ghosh, or the more recent Supreme Court dicta of Lord Hughes in Ivey v Genting Casinos (UK) Ltd (t/a Crockfords Club) [2017] UKSC 67. Lord Hughes opined that the test for dishonesty in Ghosh was wrong and that an alternative approach should replace it.

Had the Supreme Court actually found that Ghosh was wrong, the matter would have been straightforward. But it did not. As their Lordships recognised in Barton and Booth (at para. 93), Ivey was strictly obiter on the dishonesty point. Lord Hughes explained that dishonesty was not germane to the particular gambling offences in play on the appeal (at paras. 43-45, 49 and 75). And as the appellants in Barton and Booth observed (at para. 93), the Supreme Court did not even appear to hear argument on the point.

Remarkably, the Court of Appeal considered Ivey to be binding. Giving the judgment of the Court, Lord Burnett CJ stated the principle as follows (at para. 104):

Where the Supreme Court itself directs that an otherwise binding decision of the Court of Appeal should no longer be followed and proposes an alternative test that it says must be adopted, the Court of Appeal is bound to follow what amounts to a direction from the Supreme Court even though it is strictly obiter. To that limited extent the ordinary rules of precedent… have been modified.”

And, at para. 106:

We would not wish it to be thought that we are following Ivey reluctantly. The concerns about Ghosh have resonated through academic debates for decades. Lord Hughes’s reasoning is compelling.”

In reaching this decision their Lordships drew from R v James [2006] QB 588, which concerned the partial defence of provocation (as it then was). Here, another five-judge Court of Appeal followed the ruling of the Privy Council in Attorney-General for Jersey v Holley [2005] 2 AC 580 rather than the earlier bare-majority decision of the House of Lords in R v Smith (Morgan) [2001] 1 AC 416. In Holley, a majority of the Board was critical of the reasoning in Smith and endorsed the minority position instead.

James was justified by “exceptional features” that were doubtful ever to arise again, and “should not be taken as a licence to decline to follow a decision of the House of Lords in any other circumstances” (James, paras. 43-44, per Lord Phillips CJ). The thrust of the decision was that Holley saw an enlarged panel of nine Law Lords effectively overrule Smith having set out to “clarify definitively” the state of English law; and the Court of Appeal should follow it notwithstanding that their Lordships were formally sitting as the Privy Council, not the House of Lords. In the circumstances the result of any appeal to the House of Lords on the substantive law would have been a foregone conclusion (James, para. 39).

Lord Burnett CJ described the position in Barton and Booth as strongly analogous, indeed “stronger because the ordinary rules of precedent require us to follow decisions of the Supreme Court” as the successor to the House of Lords (at para. 102):

The undoubted reality is that in Ivey the Supreme Court altered the established common law approach to precedent in the criminal courts by stating that the test for dishonesty they identified, albeit strictly contained in obiter dicta, should be followed in preference to an otherwise binding authority of the Court of Appeal. As in James, we do not consider that it is for this court to conclude that it was beyond their powers to act in this way.”

His Lordship was careful to stress that the principle only applies where all the judges in the Supreme Court have agreed to direct that a Court of Appeal authority should no longer be followed (para. 104).


Barton and Booth will of course interest criminal lawyers for its determination on the meaning of dishonesty. But the wider significance lies in its radical change to the doctrine of precedent. Albeit in limited circumstances, Supreme Court obiter dicta can now displace otherwise binding Court of Appeal authority.

The decision exposes two interesting issues.

First, it establishes that the Supreme Court can change the substantive law other than through resolving the appeal before it. This is constitutionally seismic. The courts are meant to be dispute-resolvers, there to determine the legal rights and obligations of the parties in the case at hand. When senior courts make law, they do so in the process of settling the disputes before them. Hence why the principle to emerge from the decision – the ratio decidendi – is limited to those legal findings that were necessary to determine the dispute. Hence why there is a distinction between ratio and obiter remarks at all.

Practice directions aside, the courts have never made it their business to engage in abstract, blue-skies law reform. This is the role of Parliament, as required by fundamental principles like the separation of powers. For good reason there is a crucial distinction between how courts and legislatures make law.

Obiter remarks are not of course uncommon. They often contain useful analysis for future judges to consider and weigh up. But elevating obiter remarks to the status of binding authority is a significant step indeed. It is far more than the limited modification to the existing setup that Lord Burnett CJ believed it to be. Allowing Supreme Court judges to change the law by speculating on abstract matters beyond the facts of the appeal, and potentially without having heard proper argument on the point, tempts them into Parliament’s law-making territory. The implications will require careful management in future.

Second, and with respect, the analogy with James is questionable. Yes, it involved the Court of Appeal preferring a non-binding tribunal to the ordinarily binding House of Lords, and in that sense modified the doctrine of precedent. But the Court of Appeal was following the Privy Council’s ruling on the provocation issue, and it was essential for the Privy Council to rule on that issue in order to resolve the appeal. In the circumstances, and given the Privy Council personnel, there is logic in the James position that Holley should be treated as a ruling by the House of Lords.

Barton and Booth involves a different situation. The Court of Appeal was not following any ruling of the Supreme Court at all. It was following the academic reflections of a senior judge as to what the law should be. However senior the judge, and however compelling the reflections may have been, they were not a judicial determination on the issue. Once again, dishonesty was beyond the scope of the Ivey appeal and the Supreme Court seemed to hear no argument on whether Ghosh should be abandoned. It was certainly not the “undoubted” reality, as Lord Burnett CJ put it in Barton and Booth (at para. 102), that the Supreme Court was overruling Ghosh and thereby seeking to alter the doctrine of precedent.

The distinction between deciding an appeal (as in Holley) and speculating beyond the appeal (as in Ivey) is core to the fundamental constitutional distinction between the respective roles of Parliament and the courts. Barton and Booth is a significant ruling whose effects may well be felt for many years to come.

Dr Alex Williams is a member of the Cornerstone Barristers Public Law team.