Redacting for anonymisation: Article 8 v Article 10 in child protection context

Panopticon has reported recently on the ICO’s new Code of Practice on Anonymisation: see Rachel Kamm’s post here. That Code offers guidance for ensuring data protection-compliant disclosure in difficult cases such as those involving apparently anonymous statistics, and situations where someone with inside knowledge (or a ‘motivated intruder’) could identify someone referred to anonymously in a disclosed document. The Upper Tribunal in Information Commissioner v Magherafelt District Council [2012] UKUT 263 AAC grappled with those issues earlier this year in the context of disclosing a summarised schedule of disciplinary action.

Redaction is often crucial in achieving anonymisation. Getting redaction right can be difficult: too much redaction undermines transparency, too much undermines privacy. The Court of Appeal’s recent judgment In the matter of X and Y (Children) [2012] EWCA Civ 1500 is a case in point. It involved the publication of a summary report from a serious case review by a Welsh local authority’s Safeguarding Children Board. The case involved very strong competing interests in terms of Article 8 and Article 10 ECHR. For obvious reasons (anonymity being the key concern here) little could be said of the underlying facts, but the key points are these.

A parent was convicted in the Crown Court of a serious offence relating to one of the children of the family (X). The trial received extensive coverage in the local media. The parent was named. The parent’s address was given. The fact that there were other siblings was reported, as also their number. All of this coverage was lawful.

The local authority’s Safeguarding Children Board conducted a Serious Case Review in accordance with the provisions of the Children Act 2004 and The Local Safeguarding Children Boards (Wales) Regulations 2006. Those Regulations require the Board to produce an “overview report” and also an anonymised summary of the overview report. The relevant Guidance provides that the Board should also “arrange for an anonymised executive summary to be prepared, to be made publicly available at the principal offices of the Board”.

Here two features of the draft Executive Summary were pivotal.

First, reference was made to the proceedings in the Crown Court in such a way as would enable many readers to recognise immediately which family was being referred to and would enable anyone else so inclined to obtain that information by only a few minutes searching of the internet.

Second, it referred, and in some detail, to the fact, which had not emerged during the proceedings in the Crown Court and which is not in the public domain, that another child in the family (Y), had also been the victim of parental abuse.

The local authority wanted to publish the Executive Summary, seeking to be transparent about its efforts to put right what went wrong and that it has learned lessons from X’s death. It recognised the impact on Y, but argued for a relaxtion of a restricted reporting order to allow it to publish the Executive Summary with some redactions. It was supported by media organisations who were legally represented.

The judge (Peter Jackson J) undertook a balance of interests under Articles 8 and 10. He allowed publication, with redactions which were (in the Court of Appeal’s words) “in substance confined to three matters: the number, the gender and the ages of the children.”

In assessing the adequacy of these redaction, the Court of Appeal considered this point from the judgment of Baroness Hale in ZH (Tanzania) v Secretary of State for the Home Department [2011] UKSC 4, [2011] 2 AC 166, at paragraph 33:

“In making the proportionality assessment under article 8, the best interests of the child must be a primary consideration. This means that they must be considered first. They can, of course, be outweighed by the cumulative effect of other considerations.”

Munby LJ thus concluded (paragraph 47 of this judgment) that “it will be a rare case where the identity of a living child is not anonymised”.

He recognised, on the other hand, that Article 10 factors always retained their importance: “there could be circumstances where the Article 8 claims are so dominant as to preclude publication altogether, though I suspect that such occasions will be very rare.”

On the approach to anonymisation through redaction, Munby LJ had this to say (paragraph 48):

“In some cases the requisite degree of anonymisation may be achieved simply by removing names and substituting initials. In other cases, merely removing a name or even many names will be quite inadequate. Where a person is well known or the circumstances are notorious, the removal of other identifying particulars will be necessary – how many depending of course on the particular circumstances of the case.”

In the present case, the redactions had been inadequate. They did not “address the difficulty presented by the two key features of the draft, namely, the reference to the proceedings in the Crown Court and the reference to the fact that Y had also been the victim of parental abuse” (paragraph 53).

Far more drastic redaction was required in these circumstances: to that extent, privacy trumped transparency, notwithstanding the legislation and the Guidance’s emphasis on disclosure. In cases such as this (involving serious incidents with respect to children), those taking disclosure decisions should err on the side of heavy redaction.

Robin Hopkins