Peters & Peters

The Disputes Brief

Weekly insights on the latest commercial judgments

Picken J: zombie hunter?

April 19, 2026

In their 2024 article, Jonathan Barnard KC and Robert Dacre likened the decision in Three Rivers DC v Bank of England [2003] EWCA Civ 474 (“Three Rivers (No.5)”) to a zombie, staggering on despite various courts taking successive pot shots at it. In his judgment in Aabar Holdings v Glencore [2026] EWHC 877 (Comm) (“Glencore”), Picken J, took a further swipe at Three Rivers (No.5), holding that legal advice privilege can be claimed over intra-client documents sent between or created by members of the “client group” for the dominant purpose of seeking legal advice. Picken J recently hastened the demise of the so-called “Shareholder Rule”, which impinged on when privilege could be asserted. Will his intervention also cause the downfall of the zombie?

Three Rivers (No.5) – the rundown

In 1991, the Bank of Credit and Commerce International (BCCI) collapsed, amidst allegations of widespread fraud and financial manipulation. The British Government set up an independent inquiry (“the Bingham Inquiry”) into the supervision of BCCI in the following year, in which the Bank of England (“the Bank”) made submissions. BCCI’s liquidators and creditors later sued the Bank of England for misfeasance in public office. During the litigation, the Bank claimed that various documents that had come into existence during the Bingham Inquiry were protected from disclosure because they were subject to legal advice privilege. Given the non-adversarial nature of the Bingham Inquiry, the Bank accepted it could not claim that the documents were protected by litigation privilege.

The dispute over the application of legal advice privilege came down to 4 categories of documents: (1) those prepared by Bank employees, which were intended to be sent to and were in fact sent to its solicitors; (2) those prepared by Bank employees with the dominant purpose of the Bank’s obtaining legal advice, but not sent to its solicitors; (3) those prepared by Bank employees, without the dominant purpose of obtaining legal advice, but sent to the Bank’s solicitors; and (4) those falling into categories (1) to (3), but where the employee of the Bank had become an ex-employee.

The Court of Appeal concluded that the Bank was not entitled to privilege in any of the 4 disputed categories of documents. It held that legal advice privilege could not be claimed for documents other than those passing between the client and their legal advisers and evidence of the contents of such communications. For the purpose of identifying whether legal advice privilege applied to the documents concerned, the “client” was a small group of Bank officials, referred to as the Bingham Inquiry Unit (“BIU”), who were authorised to and did communicate with the Bank’s solicitors. Employees of the Bank who were not in the BIU could not be classified as the “client” and therefore documents created by them or forwarded by them to the BIU, even though for the purpose of obtaining legal advice, were not privileged.

Criticism

The narrow client approach in Three Rivers (No.5) has been widely criticised in this jurisdiction; other common law jurisdictions have declined to follow it (detailed commentary on this can be found in the article linked at the top of this piece). It is not hard to see why. The doctrine of legal professional privilege seeks to enable parties to consult their lawyers in confidence and for their lawyers to be “free to give honest and candid advice on a sound factual basis, without fear that these communications may be relied on by an opposing party if the dispute comes before the court for decision” (Ventouris v Mountain [1991] 1 WLR 607 at 611C-D). Large corporates act through many individuals. For a corporate to understand its legal position and for legal representatives properly to advise on it, the corporate will need to obtain information from various sources. Having regard to the rationale for the privilege, as Tomlinson J (whose decision was overruled in Three Rivers (No.5)) put it, “it must be axiomatic that it is the confidentiality of the whole process of communication which requires protection, not just those documents which can be recognised as comprising the actual or final communication” ([2002] EWHC 2730 (Comm) at para.30).

Glencore

In Glencore, Picken J was asked to decide whether internal communications between members of the “client” and/or documents created by a member of the “client” could be covered by legal advice privilege. Relating this to Three Rivers (No.5), the question would have been whether communications internal to the BIU, or documents created by the BIU, which were not sent to the Bank’s solicitors, could be subject to legal advice privilege.

Picken J reviewed Three Rivers (No.5) and the authorities cited within it. Having done so, he considered that the judgment did not concern intra-client communications of this nature, despite acknowledging that the order made by Tomlinson J at first instance (which was upheld by the Court of Appeal) appeared to include this category of documents. As a result, he felt there was no restriction upon him finding that the documents concerned could be subject to legal advice privilege, provided that they were created for the dominant purpose of seeking legal advice.

In arriving at this finding, Picken J considered that there was no justification for a more circumscribed approach. In particular, he identified that, if such an approach were to be adopted, an engagement or instruction letter, identifying the issue on which legal advice would be sought, would be privileged. However, another document created by the “client”, which identified the issue on which advice would be sought, would not be. In practical terms, Picken J noted there was no difference between the two types of document and it was, therefore, “illogical” to permit legal advice privilege to apply in the one case but not also the other.

Conclusion

Unfortunately, it appears that Glencore, which sought to assert privilege, backed away from an argument it had initially adopted during disclosure that Three Rivers (No.5) had been wrongly decided. However, one can hope that Picken J’s judgment will, if appealed, give an opportunity for Three Rivers (No.5) to be reviewed. Otherwise, the zombie will keep on staggering.