Secretary of State for the Home Department & Anor v IAB & Ors, R (on the application of)

On 7 February 2024 the Administrative Court is due to hear a challenge to the lawfulness of regulations removing the requirement for houses in multiple occupation to be licensed if asylum seekers are to be placed there. We are concerned on this appeal not with the merits of that claim but with an important issue concerning the evidence put forward by the Defendant Secretaries of State. They have served documents in which most of the names of civil servants in grades below the Senior Civil Service (“SCS”) are redacted. The Claimants say that the Defendants are not entitled to make these redactions. Swift J (“the judge”), in a judgment of 17 November 2023, upheld the Claimants’ objections and ruled that they are entitled to disclosure of the documents without such redactions. The judge gave permission to appeal to this court, noting that the case raised an important issue of practice and procedure.

In total, prior to the hearing below, the Defendants had disclosed four tranches of documents running to more than 500 pages. Each of the tranches of disclosure included redacted documents. Disclosure was given without explanation (either generally, or document by document) of why the passages had been redacted. In a skeleton argument filed for the hearing of the renewed application for permission to apply for judicial review on 19 October 2023, the Secretaries of State referred to redaction of the names of "junior civil servants" (by which they meant any civil servant outside the grades that comprise the SCS, regardless of age or experience). The Civil Service currently employs about half a million people, of whom approximately 2% are in the SCS. The Government is asserting the right for the other 98% to remain anonymous, save in exceptional cases – in practice, unless their identity bears directly on the decision whether to grant judicial review -- in any documents put forward in evidence in judicial review cases.

The evidence from the Defendants resisting an order for unredacted disclosure included witness statements of Phillip Smith, Head of Specialist Appeals and Litigation at the Home Office and Joanna Key, Director General at the Department for Levelling Up, Housing and Communities. Mr Smith writes:

“5) The usual approach to disclosure of documents across government is to identify members of the SCS and to include their contact details, consistent with the publication of those details in the Civil Service Yearbook, but not to do so for junior officials. On occasion this principle may not have been observed in respect of junior officials, either through error or conscious decision (perhaps because the name was considered relevant), but the expectation remains that junior officials are entitled to a greater degree of protection from personal exposure than their SCS colleagues. This may be described as an expectation of confidentiality.

6) Conversely there are rare occasions on which even the details of SCS officers are redacted, where it is considered that not doing so would expose those officers to particularly high risks. This is subject to legal advice that disclosure is required.

7) In many instances it is a trivial task for a person to infer the email address of a junior official once their name is known. Those with a relatively common name have a limited measure of protection from the inclusion of a disambiguating factor in their email address, but those with less common names have no such protection. For example, there are five men named Philip Smith with active Home Office email addresses but only one Phillip Smith.

8) There is also a risk in all cases, even if small, that the disclosure of names will undermine the welfare of civil servants, for example through harassment by communications sent directly to them. Within my own command I have a Senior Presenting Officer (“SPO”) who has regularly received abusive communications from a person whose appeal he presented before the Upper Tribunal (“UT”), before the underlying appeal progressed to the senior courts. SPOs are Senior Executive Officers, three grades below the SCS.

9) Knowing the name of the SPO, and wrongly holding him responsible for all the consequences of the adverse immigration decision, the appellant was able to indulge in a campaign of harassment. The abuse progressed to the point that earlier this year I had to instruct the Government Legal Department to make clear to the appellant that any further instances would be met with both civil action and referral to the police for investigation of possible offences of harassment and/or malicious communications.”

Ms Key states:

“11.

While junior civil servants - namely those at grades up to and including Grade 6 - do perform important advisory and management functions, they do not fulfil decision making roles, and accountability for advice or recommendations they may help provide always rests with a senior civil servant. It is on that basis that the names of senior civil servants are routinely disclosed, and those of junior civil servants are not. The exception to this approach would be if the identity of one or more junior officials was directly relevant to the claim before the Court, which applies in this case only to the Home Office policy lead, Tahira Shah. This is dealt with separately by the Home Office witness statement.

12.

Redacting the names of junior civil servants helps protect their privacy and safety, as part of our duty of care, especially relating to cases that are contentious and may attract public attention. This can prevent them from becoming targets of unwarranted personal blame, and through that, harassment, threats or retaliation, which can adversely affect their welfare. SSLUHC is concerned that junior officials have an expectation of confidentiality, and therefore routine disclosure of this sort retroactively and without specific cause will undermine this reasonable expectation without junior officials having had any opportunity to adjust their behaviour accordingly. A change to this long-held position could lead to this becoming more routine and have an adverse effect on Government policy delivery.

13.

I understand from Mr Andrews of the GLD that in recent years there have been examples of names and contact details of civil servants entering the public domain in association with contentious decision-making, resulting in their identification with the decision concerned on social and even mainstream media. In one specific case, this involved the publishing of Mr Andrews’ correspondence on behalf of GLD. This exposed Mr Andrews to offensive messages from members of the public. The civil servants involved in these examples, and Mr Andrews, a relatively junior GLD lawyer, were simply carrying out their public duties pursuant to Government policy and were in no position publicly to defend themselves on their own account, which makes this particularly concerning. Whether or not the relevant individuals had the same expectation of privacy as junior civil servants, the incident is illustrative of the general risks.

14.

In my view, the specifics of this case and the policy being challenged has the potential to be contentious, given it relates to the provision and quality of asylum accommodation. This specific policy has already been subject to media scrutiny and active engagement on social media. There is a heightened risk of harassment, or unwanted attention in the event that the names of junior civil servants are disclosed.

15.

More generally, redaction of the names of junior civil servants encourages open communication within Government. Officials in the department conduct their work on the understanding and expectation that their names will not enter the public domain where this is not necessary. If junior civil servants fear their names may be disclosed in legal cases, it could have a chilling effect on government as they might be hesitant to express concerns or provide candid advice, which could hinder effective decision making, or more widely discourage participation in public service.

16.

Different considerations apply in relation to senior civil servants whose names and positions are routinely published by their departments. As a result, they are publicly identifiable and associated with the work of a particular department, and accordingly their expectations of privacy are different to junior officials.

17.

I am aware of the recent judicial criticism of redactions in cases such as FMA & Others v SSHD [ 2023] which indicate that the names and related information of officials should not generally be redacted from official documents when they are disclosed pursuant to the duty of candour. On the basis of advice and precedent, the names of junior civil servants were not considered relevant to this claim or disclosable. I recognise that it is important in applying redactions that there is no material effect on the intelligibility of the disclosed material. To that end, care has been taken that all email chains contained within the disclosure bundles shared are easy to follow. For example, job titles (where included within the email at all) and domain names (i.e. ‘@levellingup.gov.uk or @homeoffice.gov.uk’) are retained in all disclosure.”

The Defendants also placed before the judge a witness statement of Jonathan Marron of the Department of Health and Social Care made in the case of R (Good Law Project Ltd) v Secretary of State for Health and Social Care [2021] EWHC 1223 (TCC). That case was a challenge to the “fast lane” policy of procurement of personal protective equipment during the COVID pandemic. O’Farrell J had to rule at an interlocutory stage on a similar, but not identical, application for redaction to that in the present case. Mr Marron stated:

“4.

DHSC, and HMG more widely, has a general approach of applying redactions to the names of individuals below Senior Civil Servant level in public-facing documents, in order to protect junior staff who are often not key decision makers. For example, for any and all responses to Freedom of Information requests, we consistently apply a section 40(2) exemption (personal information) to avoid the release of this information.

5.

Junior civil servants therefore have a fair expectation that in the course of performing their roles, especially under instruction from more senior civil servants, they will not be vulnerable to their names or other personal details being released in relation to the work that they do, particularly if this is on sensitive areas. It is incredibly important for the Department’s ability to empower these junior civil servants to effectively perform their roles, to be able to protect them from exposure in public documents that might invite criticism. By contrast, Senior Civil Servants operate under the expectation and understanding that as senior decision-makers, they invite a higher level of scrutiny and accountability, including the release of their names in public documentation.

6.

In the instance of this specific judicial review, the Court will be aware of the high volume of individuals named throughout an even higher volume of disclosed documents. The vast majority of junior civil servants whose names appear in these documents are not relevant to the facts of the case (at times they are even simply in copy on an email chain), let alone decision makers in the matters that the Court is concerned with.

7.

If the Court were to allow this application, it would be disproportionately distressing to the vast majority of junior civil servants whose names would be released, due to the high likelihood of media scrutiny and harassment connected to this claim.

8.

Considering that the majority of these individuals were volunteers who offered to support Government and the general public in the emergency COVID-response, releasing their names and exposing them to this level of scrutiny could negatively impact the Government’s ability to draft effective junior civil servants into such emergency/high profile roles in future.

9.

Releasing these names within the confines of a confidentiality ring helps to maintain protection from public exposure and scrutiny for these civil servants, whilst still allowing the Claimants to follow the relevant trails of evidence.”

The principal issue which Swift J had to decide was, in his words, “is it permissible for the Secretaries of State, as matter of routine, to redact the names of civil servants outside the Senior Civil Service from documents disclosed in proceedings?” He held that it was not:

“12.

Two points of context are material. The first is that it is well-established that the duty of candour is an obligation of explanation rather than simply an obligation of disclosure. The substance of the obligation is well put by Sir Clive Lewis in his “Judicial Remedies in Public Law” 6^th edition 2021, at paragraph 9-098. The obligation exists to ensure that a defendant explains, whether by witness statements, or the provision of documents, or a combination of both, the reasoning process underlying the decision under challenge. In the present case the Secretaries of State have, to date, chosen to discharge their candour obligation by disclosure of the documents in the four disclosure bundles. No witness statements have been provided. The second point of context is the criterion for disclosure of documents in judicial review proceedings. The standard applied by the court when asked to decide whether disclosure of a document is required is whether disclosure is necessary for the fair and just determination of an issue in the case: see Tweed v ParadesCommission for Northern Ireland [2007] 1 AC 650 per Lord Bingham at paragraphs 3 and 4, Lord Carswell at paragraph 38, and Lord Brown at paragraph 52.

13.

It follows that the correct premise is that by making the disclosure they have already made, the Secretaries of State accept that disclosure of those documents is necessary for the fair and just disposal of the issues in this case or, at the least, per Lord Bingham at paragraph 4 of his speech in Tweed, that the disclosed documents are “significant to its decision”. In this case the documents disclosed, which evidence the decision-making process were, no doubt, disclosed in support of the Secretaries of State’s response to the challenges on the Tameside ground: the Secretaries of State will rely on these documents to support their case that the decisions rested on proper enquiry into and consideration of relevant matters. This is the context within which the Secretaries of State’s general submission on relevance must be considered.

14.

The practice of redacting, of blanking-out parts of documents disclosed in litigation on the ground that the part redacted is irrelevant, is long-established. One obvious situation is where a part of a disclosable document does not concern the subject matter of the litigation. The position in claims under CPR Part 7 goes significantly further. In GE Capital Corporate Finance v The Bankers Trust [1995] 1 WLR 172, Hoffmann LJ stated (at pages 174B and 175G and H):

“It has long been the practice that a party is entitled to seal up or cover up parts of a document which he claims to be irrelevant …

… In my view, the test for whether on discovery part of a document can be withheld on grounds of irrelevance is simply whether that part is irrelevant. … There is no additional requirement that the part must deal with an entirely different subject matter than the rest.

The Peruvian Guano test must be applied to the information contained in the covered-up part of the document, regardless of its physical or grammatical relationship to the rest. Relevant and irrelevant information may, as in this case, be contained in the same sentence. Provided that the irrelevant part can be covered without destroying the sense of the rest or making it misleading, a party is permitted to do so.”

15.

The Secretaries of State’s submission on relevance relies on the logic explained by Hoffmann LJ. The submission is to the effect that notwithstanding that each document under consideration was properly disclosable, it is then possible to remove by redaction any part of the document that does not directly bear upon one or the other of the Claimants’ grounds of challenge. This includes the names of the civil servants, though could also include much else.

16.

The logic that drives the Secretaries of State’s submission extends well beyond the mere redaction of the names of civil servants outside the Senior Civil Service. It would permit redaction of the name of any and every civil servant, save where the identity of the person went to the legality of the decision, and would permit the removal of any part or word in the text of a document that did not in some way directly concern a ground of challenge. Moreover, the same reasoning would apply for all public authorities before the courts in all judicial review claims; the submission made does not identify any logical distinction between civil servants in government departments and persons employed by local authorities or by any other decision maker whose powers are derived from public law.

17.

I accept that the outcome of the grounds of challenge in this case will not depend either on the identity of the decision-maker or of any other person involved in the decision-making process. The Claimants do not contend otherwise. However, I do not consider the correct approach to redaction of disclosed documents in judicial review pleadings can be driven only by the purity of Hoffmann LJ’s logic. What is required to discharge the obligation of candour when a public authority chooses to meet that obligation by disclosure of documents must, at the least, be fully informed by the purpose of the candour obligation. Redaction, sentence by sentence or line by line, as a matter of course, runs against the grain of an obligation aimed at ensuring public authorities responding to judicial review claims should explain the reasoning underlying the decision under challenge. ...”

The judge then cited the judgment of Sir John Donaldson MR in R v Lancashire County Council ex p. Huddleston [1986] 2 All ER 941, which said that once an applicant has obtained permission for judicial review “it becomes the duty of the respondent to make full and fair disclosure” and that this is “a process which falls to be conducted with all the cards placed upwards on the table and the vast majority of the cards will start in the authority’s hands”. The judge continued:

“17...This explains the premise and extent of the duty of candour. A document that has been disclosed in judicial review proceedings ought not, absent good reason, be redacted on grounds of relevance in any way that impairs either the actuality or the appearance of a "cards face upwards" approach. So far as concerns the relationship between the courts and public authorities described by Sir John Donaldson (no longer a "new" relationship), the "cards face upwards" reference also makes the point that appearance has a part to play, not the least because the premise for disclosing the document at all is that disclosure is necessary for the fair and just determination of the case.

18.

Redaction leads to significant practical difficulties. The present case is an example of a common situation where email exchanges and other contemporaneous documents are disclosed to explain a decision-making process. Most decisions made within central government now involve significantly sized groups of civil servants. On any occasion one civil servant within the group might be the sender of the message, might be the recipient of the message, or might (usually, will probably) be copied in. Sometimes (as in this case), the civil servants within the group are spread across different government departments. At the least, redacting names makes the decision-making process and the significance of each document disclosed more difficult to understand. In some instances, it may obscure the significance of a document almost completely. When correspondence and other documents are disclosed for the purpose of evidencing a decision-making process it will rarely be the case that it will not assist the court's understanding of that process and the decision itself to know by whom or to whom documents were sent, forwarded, or copied. In most cases, when this information is redacted, any outsider's understanding of the documents (and for this purpose the court is an outsider) is significantly hampered. Misunderstanding and misinterpretation become commonplace. When documents are disclosed, and parties then rely on them by including them in the hearing bundle, the court is under a practical obligation to consider those documents with a view to making sense of how the information in the documents bears upon the legality of the decision under challenge. All this is made much more difficult and much more time-consuming when (for example) successive strings of email correspondence, each pages long, are entirely anonymised. The same point applies to names redacted in the body of correspondence or other documents. All such redactions only detract from the intelligibility of the document and impair achievement of the purpose for which the document was disclosed in the litigation.

19.

The Secretaries of State's response, that any concerns are about no more than "making reading documents a little bit easier", is glib. First, ensuring that documents disclosed in litigation to explain a decision-making process are readily intelligible is an objective worth achieving for its own sake. It is notable that the Secretaries of State's proposal to deal with problems of intelligibility (both in this case, and generally) was to replace redacted names with a list of ciphers; an approach that would be laborious, prone to error, and even when error-free would only add a new layer of complexity to the task of understanding the narrative of the decision-making process from the documents disclosed.

20.

Second, an approach to compliance with the obligation of candour that, as a matter of routine, hides detail that aids the court's understanding of the public authority defendant's explanation of the decision under challenge, is antithetical to the purpose of the candour obligation. Third, the appearance created by the Secretaries of State's approach is a matter of genuine concern. Reasonable and well-informed members of the public will readily understand that there are occasions (few in number) when documents disclosed in aid of the fair and just determination of legal disputes must be redacted as some information in the documents is sensitive. Considerations of national security and instances where public interest immunity can be asserted are obvious examples, and there will be others. However, a practice by which information, not sensitive per se, is routinely removed from documents risks undermining confidence that appropriate legal scrutiny is taking place under fair conditions, because it will be apparent that the routine redaction builds in a possibility that the sense or significance of a document may be overlooked.

…

22.

Drawing these points together, the principle that ought to guide the approach in judicial review proceedings is that absent good reason to the contrary (which might, for example, include that the information in question was subject to a legal obligation of confidentiality), redaction on grounds of relevance alone ought to be confined to clear situations where the information redacted does not concern the decision under challenge. The names the Secretaries of State seek to protect are not in this class. Names of civil servants should not routinely be redacted from disclosable documents; redaction should take place only where it is necessary for good and sufficient reason. This conclusion is consistent with the obligation of candour and with the general principle of cooperation between public authorities and the court that is one foundation for judicial scrutiny. This approach will also guard against the practical difficulties caused by excessive redaction…

23.

The question that remains is whether, set against this general position, there is sufficient reason to support the Secretaries of State's submission that the names of civil servants outside the Senior Civil Service should, as a matter of routine, be redacted from disclosable documents.

24.

The Secretaries of State advance several points relying on the contents of the witness statements referred to at paragraph 10 above. The first is that the names of civil servants outside the Senior Civil Service should be removed because they have a "reasonable expectation of confidentiality" i.e., that civil servants have a general expectation that the fact they have been involved in a particular decision-making process will remain confidential even when the decision is subject to legal challenge. This expectation does not arise from any matter connected to the subject matter of any decision; it rests simply on the fact they are civil servants.

25.

I do not consider any such general expectation (even assuming it exists in practice) could be reasonable. No such expectation would attach to any person as a matter of general employment law. Moreover, when at work civil servants are not involved in anything that can be described as a private activity, they are exercising public functions as part of the public service of the country. It is also material that while the Secretaries of State's submission refers to the class of "junior civil servants" this label was applied only to distinguish them from the civil servants working in grades comprising what the government refers to as "the Senior Civil Service". Therefore, the distinction between "junior" and "senior" civil servants is akin to the distinction between junior and leading counsel and is not necessarily any indication of age or experience. The class of "junior civil servants" includes civil servants with significant responsibilities.”

Swift J said that he did not find the examples of harassment given by Mr Smith and Ms Key compelling as an argument for redaction. He said that the two examples in the evidence did not suggest a widespread problem and that generalised concerns should not provide for an approach to disclosure in judicial review claims. He concluded that no sufficient reason had been shown either from general considerations or the circumstances of the current case to warrant the redaction from disclosable documents of the names of civil servants outside the SCS.

Sir James Eadie KC submits that it is permissible in principle to redact parts of disclosed documents on the basis of irrelevance to the issues in the case. He relies on three decisions in particular. The first was the passage from the judgment of Hoffmann LJ in GE Capital Corporate Finance v The Bankers Trust [1995] 1 WLR 172 at [50] cited by the judge. The second was Shah v HSBC Private Bank (UK) Ltd [2011] EWCA Civ 1154, a case concerning whether disclosure by a bank of the writers and recipients of internal documents needed to be made. Lewison LJ referred to the observations of Hoffmann LJ in GE Capital and held that the same approach to the sealing or concealing of parts of documents applied in the “changed landscape” of the Civil Procedure Rules.

The third decision on which Sir James placed great reliance was that of O’Farrell J in the Good Law Project case cited above. She declined to order that the defendant Secretary of State disclose unredacted evidence to the claimant (where the redactions covered the names of junior civil servants), and instead ordered that the unredacted documents be disclosed in a confidentiality ring (as the defendant had proposed).

The core of the Appellants’ submissions was set out at paragraph 17 of Sir James’ skeleton argument:-

“The names and identities of JCS [junior civil servants] will generally be irrelevant. The mere fact that a civil servant was involved at some point in considering or discussing the issues relating to the decision does not render their identity relevant. Nor does being named (e.g., as a recipient of an email) in a document which has been disclosed. Relevance depends on the fact in question (here, e.g., the name/identity of the recipient of an email) bearing in a material way on the issues in dispute.”

The Appellants accept that there might be an exception if there were a tenable allegation of bias or some other allegation dependent on the actual identity of the particular junior civil servant involved in the challenged decision-making. But even in the case of the actual decision-maker “as the identity itself would not usually affect the issues in the case, partial redaction and/or ciphering may then remain permissible”.

The Appellants submit that there is no duty to disclose “irrelevant” information only in order to improve intelligibility. It is argued that “the redaction of individual names within a sentence does not affect the sense of that sentence”. Even if redaction of names did affect the document’s intelligibility, that issue should be resolved by alternative means which could include ciphering (including by provision of separate reference schedules) or providing descriptions of the individual’s department, unit or job title to distinguish between different persons.

The Appellants referred us to the decision of the Upper Tribunal (Administrative Appeals Chamber) in Cox v Information Commissioner and Home Office [2018] UKUT 119 (AAC) in which the tribunal said, in the context of requests under the Freedom of Information Act 1998, that “it is likely to be easier to demonstrate a need to release personal information about more senior decision-makers than about more junior staff”.

Finally, it was argued that, for the reasons given by Ms Key, the routine release of names of civil servants would have a “chilling effect on public administration”.

Ms Dubinsky KC submits that it is antithetical to the duty of candour to make redactions which obscure the context, significance or intelligibility of the documents disclosed or which apply a policy of redaction by default. Such an approach is not on proper analysis supported by any authority. She points out that, although the Appellants disavow any absolute rule of redacting names, it is clear from their arguments that they envisage only the narrowest exceptions; and that there is no practice or intention of analysing on a document-by-document basis whether redaction is justified or whether it impairs intelligibility in a given document.

As to the asserted expectation of confidentiality of the names of civil servants, the skeleton argument for the Respondents states pithily:

“The Appellants’ argument is circular: the Appellants generally redact, consequently junior civil servants have a reasonable expectation of confidentiality, consequently the Appellants can continue to redact.”

The claim that a class of information can presumptively be redacted is inconsistent with “the important, calibrated and exacting nature” of the duty of candour. Permitting redaction of a class of information on the basis of an asserted public interest circumvents the need for a public interest immunity claim: yet, as Singh LJ said in R (Hoareau) v Secretary of State for Foreign and Commonwealth Affairs [2018] EWHC 1508 (Admin) at [19], “there is no such thing as a class claim to PII any longer. The balancing exercise is undertaken by reference to the contents of the particular document in question.”

Mr Vassall-Adams KC, appearing for JUSTICE as intervener, reminded us that ever since Scott v Scott [1913] AC 417 it has been held that new departures from the principle of open justice must be authorised by statute. He asked us to note that CPR 39.2(4) provides that at a hearing “the court must order that the identity of any person shall not be disclosed if, and only if, it considers non-disclosure necessary in order to secure the proper administration of justice and in order to protect the interests of that person.”

I agree.

I also agree.