As is widely recognised, and often repeated, legal professional privilege serves an important public interest. It allows parties to take professional legal advice, and to prepare for and conduct litigation, without the fear that what they say as part of that process will come back to haunt them later.
Privilege was described by Lord Hoffmann in Morgan Grenfell as a “fundamental human right” and a “necessary corollary of the right of any person to obtain skilled advice about the law”. The importance of the principle is underlined by the fact that, where it applies, its protection is absolute. The court cannot weigh the public interest in maintaining privilege against the competing public interest in having a case decided on all relevant evidence. If material is privileged, it is not admissible, no matter how much evidential value it may have.
Against that background, it is all the more surprising that the question of whether privilege applies under English law often depends on fine distinctions and formalities, rather than matters of substance. This blog post will look at some of the anomalies.
The test for privilege
The requirements that must be satisfied in order to qualify for protection under either of the two main strands of privilege are well known:
- Legal advice privilege applies to confidential lawyer/client communications for the purpose of giving or obtaining legal advice, regardless of whether litigation is in prospect.
- Litigation privilege, in contrast, applies to material prepared for the dominant purpose of litigation that is reasonably in prospect, and includes third party as well as lawyer/client communications.
That may all sound straightforward. But those familiar formulae hide a multitude of difficulties.
Information from third parties
If a lawyer emails a third party (the client’s accountant, say) to ask for information needed in order to advise the client, and the accountant responds, that exchange of emails isn’t privileged (assuming there is no litigation in prospect). It is not a lawyer/client communication and so is not covered by legal advice privilege, even though it is (in our example) for the purpose of giving legal advice.
Equally, if the lawyer telephones the accountant to discuss the information, that communication is not privileged and either the lawyer or the accountant could be required to answer questions about it if giving evidence in subsequent proceedings. And if the lawyer takes a note of the conversation, that note may not be privileged either, unless it can be said to betray the trend of the lawyer’s advice. (That was the test adopted in The RBS Rights Issue Litigation, where the court applied the Lyell v Kennedy (No 3) (1884) principle, relating to a lawyer’s selection of third party copy documents, to the rather different context of a lawyer’s notes of witness interviews conducted as part of an internal investigation. As to whether it’s the correct test, the current state of the law is rather unclear. The point was left open by the Court of Appeal in SFO v ENRC.)
In contrast, if the lawyer telephones the accountant and, rather than taking a note of the call, summarises the key points in an email to the client following the call, that email will quite clearly be privileged. It is a lawyer/client communication for the purpose of giving legal advice. Even if the email does not contain actual legal advice, it will be privileged so long as it’s part of the continuum of communications relating to the matter on which the lawyer is advising. That much is clear from the case law (see, for example, Property Alliance Group v RBS).
So the question of privilege depends on the form in which the information received from the third party is recorded. If it is (only) in a lawyer/client communication to give or obtain advice, privilege will attach. But if it is in a written communication from the third party, or a note of an oral communication? Here there is no privilege, it seems, even though the purpose is identical.
Copies of non-privileged documents
If a lawyer emails the client summarising an article the lawyer has found online, which is relevant to some aspect of the matter on which the lawyer is advising, that email is privileged. The same will be true if the lawyer quotes from the article, or includes a link to the article in the body of the email. In each case, the email will be a lawyer/client communication for the purpose of giving legal advice.
But if, instead, the lawyer downloads the article and attaches a copy of it to the email, privilege will attach to the email itself (as a lawyer/client communication) but not, it seems, to the attached copy of the article, unless the lawyer’s selection of that article betrays the trend of the legal advice (applying the principle in Lyell v Kennedy) or, perhaps, if it can be said to fall within the (rather random-sounding) exception for copy documents created for a privileged purpose where the client has never had the original (deriving from The Palermo (1883), which remains good law, for the moment, though it has been the subject of much criticism by both courts and commentators and the boundaries of the principle are not entirely clear).
So, again, the question of privilege may depend on a matter of form: whether the lawyer summarises or quotes from the relevant source, which is privileged, or alternatively attaches a copy, which is unlikely to be, at least subject to the outcome of the appeal in FRC v Sports Direct, due to be heard in October. Much of the attention in relation to that case has, understandably, been focused on the High Court’s rather controversial decision that there was no infringement of Sports Direct’s privilege in requiring it to produce its privileged material to its auditor’s regulator, for the purposes of an investigation into the auditor. But the Court of Appeal has also given permission to appeal the High Court’s finding that attachments to lawyer/client communications do not thereby become privileged, even where they are communicated for the purpose of giving/obtaining legal advice. That decision will be one to watch on both counts.
Information provided to third parties
The principle of limited waiver is now well established under English law (as applied, for example, in another decision in the Property Alliance Group v RBS litigation where the court accepted that disclosure of privileged documents to regulators on confidential terms did not, in itself, result in a wider waiver of privilege). It means that privileged material can be shared with a third party, on confidential terms, without losing privilege against the rest of the world. Equally, it is clear that a confidential communication will be privileged to the extent that it evidences privileged material, even if the communication itself does not meet the tests for legal advice or litigation privilege, for example, because it is not a lawyer/client communication and there is no litigation in prospect (see for example Re Edwardian Group Ltd).
So, if a lawyer writes to the client’s financial adviser summarising the lawyer’s advice to the client, the communication will be privileged to that extent. (This assumes, as will almost certainly be the case, that the adviser owes an express or implied duty of confidentiality to the client.) Or if the lawyer sends the financial adviser a copy of a letter of advice to the client, that letter will remain privileged.
But let’s say the lawyer’s letter to the client did not contain any legal advice; it merely summarised factual information relevant to the matter on which the lawyer is advising and is therefore privileged as part of the continuum of communication, referred to above. Here the position is different. If the lawyer copies the letter to the financial adviser, it remains privileged. But if the lawyer merely summarises the same facts in a letter to the financial adviser, that will not be privileged, at least outside the litigation context (where it will likely be privileged if, for example, sent to a potential witness). But again that seems a mere accident of form.
And a similar problem can arise even in the litigation context, because of the Court of Appeal’s very strict interpretation of the dominant purpose test for litigation privilege in WH Holding v E20. In that case, the court held that the test will not be satisfied unless a communication or document was prepared for the dominant purpose of obtaining advice or evidence in relation to litigation, rather than the conduct of the litigation more broadly. On that test, for example, it seems that communications with litigation funders will rarely be privileged, save to the extent that a privileged document is copied to a funder, or the communication summarises legal advice or evidence obtained for the litigation.
Again, however, that leads to the oddity that a summary of information (as opposed to advice or evidence) provided to a litigation funder is unlikely to be privileged, even though it may be integral to the conduct of the litigation, unless that summary is contained in a lawyer/client communication that is shared with the funder. Once again, form prevails over substance.
These are just a few examples of the anomalies that abound when contemplating the application of the rules of privilege in specific factual scenarios. That leads to the unsatisfactory situation that privilege may be gained, or lost, by structuring communications in a particular manner. And where documents are created by those who aren’t steeped in the rules on privilege, the question of whether or not the protection can be claimed may seem a matter of accident.
What is the solution? In my view, many of the difficulties would be solved by developing a unified test for privilege based on the dominant purpose of the document or communication, rather than depending (as is currently the case for legal advice privilege) on who is communicating with whom. So, if I had my way, privilege would apply to any communication or document that was brought into being for the dominant purpose of either obtaining legal advice or preparing for or conducting litigation (in a reasonably broad sense, not just advice or evidence as is currently the case for litigation privilege).
That may not solve every problem relating to the application of privilege, and there might still be a need for some refinement around the edges. But it would give a clear and workable starting point, which would be much more likely to achieve the underlying purpose of the protection, without the current anomalies.