• Home
• About
• Login
• Register
• Search
• Current
• Archives
• Announcements
• Pre-2013 Index

Much of the doctrinal development in this area of Article 8 ECHR case law that is concerned with 'criminality information sharing' has stemmed from a body of litigation around the issuing by the police (in partnership with the Disclosure and Barring Service) of 'Enhanced Criminal Record Certificates' (ECRCs), which are required by employers where there are public protection dimensions to employment . This litigation has effectively 'peaked' in doctrinal terms in the form of the Supreme Court decision in R (T and another) v Home Secretary and another [2014] UKSC 35. Appropriate 'filtering' rules for the contents of ECRCs, used in large swathes of employment sectors, including education, health, social care, criminal justice, and so on, were deemed to be essentially non-existent in the same case in the Court of Appeal in 2013. This Court of Appeal judgment then led to reforms in the ECRC system through a series of Orders laid before Parliament in the summer of 2013. The declaration by the Court of Appeal that the ECRC filtering rules then in place were incompatible with Article 8 of the European Convention on Human Rights was upheld in the Supreme Court judgment itself. The reforms to ECRC filtering rules themselves, however, have been (unsuccessfully) challenged as being 'light touch'; merely precluding a single, first conviction or caution for some less violent offences from being shared, and as long as a sufficient period of time has passed [62]. As was noted by Simon J (at para. 65) in this recent unsuccessful challenge in R (W) v Secretary of State for Justice [2015] EWHC 1952 (Admin), it was necessary to recognise "that almost any system which could be devised may lead to harsh results at the margins". Of course, in systematising any decision making process with rules, rather than more elastic criteria, there is always a risk of overly reducing discretion (in ECRCs, of the police to take into account the circumstances surrounding the relevant conviction).

It is fair to say that there is more guidance regarding the sharing of criminality information, however complex, than there is for intelligence retention. It was hoped that the Supreme Court decision in Catt would have provided more clarity and so we now turn to the doctrinal issues in this case.

6. The recent Supreme Court decision in Catt

The recent Supreme Court decision in Catt[63] has shown a move toward greater recognition for the realities of policing based on 'intelligence', whether of a value from relatively low to very high. But it is possible to have a range of views as to the future impact of this decision to deem lawful the retention of scraps of intelligence about one peaceful or non-violent protestor as part of the 'jigsaw' of intelligence on the more violent group with which they associate.

Some basic factual points to note about the case, now that we are offering up a more detailed critique of the judgment in this Part of our piece, are that Catt has no criminal convictions, is 92 years old, and has never been violent personally at anti-arms trade rallies or other political demonstrations (although groups with which he associated had engaged in violence). Catt has been arrested, but not charged, on two separate occasions at protests, for the offence of obstructing a highway. Records on police databases recorded his presence at protest events, his appearance, demeanour and peaceful protests activities (including sketching police and protestors on a notepad etc.). Catt had already achieved, through the use of subject access requests under the Data Protection Act 1998, an idea of the scope of the records that were held about him and his protest activities and associations, and had already convinced the Metropolitan Police authorities that the disparate pieces of information concerning him should no longer be connected with the use of a 'nominal profile' - an index entry, in effect, which marks an individual of note to investigators of criminality, here in the public protest context. What remains are mentions of John Catt in the reports and intelligence describing other activists and demonstrators at particular events and protests which he attended. The Supreme Court noted in its judgment that extensive weeding of 'nominals' had gone on over a period of time in order that this indexing of individuals was proportionate. It accepted the police's argument that to undertake the wholesale weeding-out of the residual component records of those whose nominal profiles were removed would result in an enormous, potentially unworkable, administrative burden - but that argument as to costs and resources was not significantly entered into, since the Supreme Court determined that the retention of those more fragmentary intelligence records concerning John Catt could be justifiably retained, given his links as a member of various groups to the more violent and radical activities of others. Importantly, the Supreme Court decision in Catt means that these fragmented intelligence records need not be automatically deleted as they form part of a 'jigsaw' of intelligence on the protests.

Giving up part of that intelligence 'jigsaw' in the context of policing public disorder, domestic extremism and other public protection risks is an anxious exercise for the state. Tellingly, given the wider context of the impact the outcome of Catt will or would have, it has been written that:

" For anyone whose job it is to protect others from harm, one of the most difficult tasks is to decide when to delete personal information. Personal data collection by the public sector, and the sharing of those data, is often subject to criticism in the media, and to the risk of close scrutiny in the courts. Yet taking a decision to delete can never be an exact science, and there is often an underlying fear that a deleted piece of information might turn out to be the missing link that would have identified a risk to another person." [64]

There are a number of particular competing policy issues in the domain of retaining information on police databases, which are highlighted by Catt. The first such policy concern is the notion that individuals should be as free from state intrusion and monitoring as the necessary requirements of public protection will permit - the classic civil libertarian perspective - and that the state should be able to forget misdemeanours, as well as forgive them, in the aim (particularly in this case) of avoiding the chilling of public protest in a democratic system with a tradition of the same. In Catt in the Court of Appeal, where John Catt was successful, Moore-Bick LJ noted, for example, that "Even information of a public nature, such as a conviction, may become private over the course of time as memories fade, thereby enabling people to put their past behind them …" [65]

However, there do exist in UK society a range of groups who present a public protection risk - and there is the argument, reflected in the eventual outcome of John Catt's case, that there are those individuals, who, even as peaceable 'hangers-on' in more radical, even violent, circles, must forgo a better protection of their personal 'informational privacy' or 'right to informational self-determination'. In Catt in the Supreme Court, Lord Sumption (at para. 31) [emphases added] argues that:

"The composition, organisation and leadership of protest groups who are persistently associated with violence and criminality at public demonstrations is a matter of proper interest to the police even if some of the individuals in question are not themselves involved in any criminality. The longer-term consequences of restricting the availability of this resource to the police would potentially be very serious. It would adversely affect police operations directed against far less benign spirits than Mr Catt. Organised crime, terrorism, drug distribution and football hooliganism are all obvious examples. One cannot look at an issue of this kind simply in relation to Mr Catt."

There was a particular kind of sub-narrative which developed in the Catt judgment in the Supreme Court which had not surfaced to the same extent in earlier iterations of the case. There was a sense, as a kind of classic 'floodgates' concern, that if John Catt were to succeed in the face of public policy arguments for the retention of his data within police intelligence databases, and have those records then deleted, then this might lead to a situation where intelligence retention in very sensitive areas of UK policing might be fatally compromised. Chief amongst these issues as raised in the Supreme Court judgment in Catt was the idea of intelligence gathering and retention in relation to the histories and 'criminality information' of domestic violence perpetrators.

As Lord Sumption has in obiter dicta in Catt (at para. 43):

"[Particular] kinds of offence are often characterised by the development of abusive behaviour over a long period of time. This is especially true of domestic violence, a difficult and sensitive area in which the protection of persons at risk may require sensitive monitoring over a considerable period."

Similarly, and perhaps more passionately, Lady Hale (at para. 54) commented that:

"It is well known that, for a variety of reasons, complaints of domestic violence are often not followed through to prosecution and conviction. But it is vital for the police, when responding to any new complaint, to know whether there have been similar complaints in the past. Domestic violence often escalates in seriousness with each new incident, and the police have to be aware of this when considering how to respond. It is not too dramatic to say that lives have been saved as a result."

In the same decision again, once more in obiter, Lord Toulson (at para. 76) also concluded that, in relation to the retention of the types of criminality information reaching the lower thresholds of certainty only (such as complaints or allegations, as opposed to convictions, with the latter having reached the standard of proof of guilt 'beyond reasonable doubt', for example):

"The response of the police to complaints about abusive conduct may well be affected by knowing whether similar earlier complaints have been made against the same person, either by the same or by other complainants. In those circumstances I do not consider it to be unlawful for the police to adopt a standard practice of retaining a record of such complaints for several years, but with a readiness to be flexible in the application of the practice."

So the judgment by the Supreme Court in Catt, in many ways, was not just about, and does not touch upon the regulation of a single police database, in relation to the intelligence records describing just one individual of little risk to society: instead, the decision in Catt could be seen in the minds of the Supreme Court justices in the relevant panel to have distinct ramifications for the policing of harmful behaviours across society as a whole. In this regard, the Supreme Court could be seen to be reflecting current policing concerns regarding the importance of effective data retention and sharing. In its Building the Picture report, HMIC commented on the importance of consistent data sharing in observing crucially, from a policy perspective, that:

"In the light of case law and high-profile cases such as Jimmy Savile's long period of sex offending, we are materially concerned about the extent to which the police service is responding fully to the responsibilities inherent in a changing environment, where speedy access to up-to-date and relevant information is essential. For example, we found cases where forces had not revisited their position since the whole of the police service completed local information management implementation plans in 2010. In this regard, the absence of appropriate audit and assurance regimes (to check that information is being appropriately assessed, retained or disposed of) is especially worrying, and needs to be addressed swiftly." [66]

However, while the relevant lines of jurisprudence and regulatory policy might be converging post-Catt to at least some extent, it is not so clear that our own recent case law, and particularly the outcome in Catt, is consistent with the European view of similar police operational intelligence issues. If we consider the manner in which the European Court of Human Rights, with its quasi-supervisory jurisdiction over the UK Supreme Court, has come to determine some of its own views on police intelligence and criminality information retention, in relation to one particular case at least, we can see a marked difference emerging in comparison with contemporaneous UK jurisprudence. The Registrar of the European Court of Human Rights has observed (at least in a press release), that in Brunet v France (2014) (application no. 21010/10), a case concerned with the potential deletion of (and refusal to delete) an allegation of domestic violence from a police intelligence database, it was the case:

"that the information contained in the database was quite intrusive in nature. While that information did not contain the individuals' fingerprints or DNA profile, it consisted of details on identity and personality, in a database that was supposed to be used for researching crimes. In addition, the retention time of the personal record, 20 years, was particularly lengthy in view of the fact that Mr Brunet had not been found guilty by a court and that the proceedings had been discontinued… the Court took the view that the State had overstepped its margin of appreciation in such matters, and that the rules for the conservation of records in the STIC database, as applied to Mr Brunet, did not strike a fair balance between the competing public and private interests at stake. Accordingly, the impugned retention could be regarded as a disproportionate interference with Mr Brunet's right to respect for his private life and was not necessary in a democratic society." [67]

The difference with the outcome in Catt is marked. The UK jurisprudence, in the form of Catt at least, treats domestic violence perpetrators and groups of protestors, violent or otherwise, exactly alike for the sake of the 'jigsaw' argument of operational necessity; while the European Court of Human Rights, perhaps, might prefer to distinguish the risk posed individually by a non-violent protestor such as John Catt from the societal risk posed by an alleged domestic violence perpetrator. We shall see if John Catt is granted permission to proceed to a hearing by the Strasbourg Court. To date, however, little has been written about the decision in Brunet v France, since:

"Sadly, the judgment in Brunet, which is in French, does not give a detailed explanation of the French retention scheme. It is therefore difficult to make precise comparisons between the French and domestic schemes. However, it would seem that [UK regulation of police intelligence] may be open to challenge on the basis of this judgment." [68]

The doctrine of the margin of appreciation might support the UK government if Catt case were to be re-examined at the level of the Strasbourg court. It is possible to make comparisons with the facts of Brunet and the situation in Catt but the common law basis of much public order policing in the UK, including powers to gather and to use police intelligence might mean that in the UK, police intelligence retention in the public order policing context of Catt might be readily deemed justifiable by the European Court of Human Rights through applying the margin of appreciation. It is interesting, though, to note that the Brunet decision was given only minimal consideration in the Supreme Court judgment in Catt.

UK case law is already moving on in a way, post-Catt, that might yet see considerable shifts in the regulation of police intelligence, if not precisely on the same point as in Catt (that is, on the retention of intelligence about relatively peaceful protestors). We commented upon the recent case of Davis[69] in section 3 above. Although the Court ultimately disapplied section 1 of the 2014 Act, for Bean LJ the European Court decision in Digital Rights Ireland must be interpreted to mean that (para. 95) 'a wholly innocent person's data might be accessed in order to assist in the detection of serious crime by others. The need for access to data is not limited to data directly attributable to particular individuals suspected of having committed serious crimes'. This would seem to reflect strongly the views of the Supreme Court in Catt on the nature of intelligence and data-led policing. The Court however went on to determine that (at para. 98), as per the European Court of Justice decision, access to communication data retained in bulk by communications service providers requires 'independent approval.' This kind of independent judicial oversight is not required for decisions to collect and retain police intelligence at the moment.

7. A broader commentary and critique of the Supreme Court decision in Catt

A commentary, or a critique, of the Supreme Court decision in Catt is ultimately a commentary or critique of the judgment of Lord Sumption in the case - it is the first of the judgments, the lengthiest, and the most uncompromising in tone. Lord Sumption addresses directly the issue of Article 8 ECHR engagement in the case, and the notion of Article 8 as a qualified right, by observing (at para. 6) that: "it is clear that the state's systematic collection and storage in retrievable form even of public information about an individual is an interference with private life… It follows that the present appeals turn on article 8(2) of the Convention, and in particular on whether the retention of the data is (i) "in accordance with law", and (ii) proportionate to its objective of securing public safety or preventing disorder or crime."

However, for Lord Sumption in Catt (at para. 12), in his judgment, the quality of a lawful interference with qualified rights being 'in accordance with the law' "does not mean that the law has to codify the answers to every possible issue which may arise. It is enough that it lays down principles which are capable of being predictably applied to any situation." Lord Sumption seems to have in mind the notion that the indefinite retention of police intelligence, or nearly all police intelligence, is what is being challenged, and that the case was one to use to determine principles applicable to the retention of data about any group or individual by the police - where that person or group is a public protection risk - rather than set thresholds for the deletion of data or intelligence where a certain (low) profile of risk is presented on the facts.

Interestingly, Lord Sumption deployed a four-part, or 'quadripartite', test as to the proportionality of an interference with rights in his judgment in the Supreme Court in the case of R (Bank Mellat) No.2 v HM Treasury [2013] UKSC 39, stating that a measure of the proportionality of an interference with ECHR rights required:

" exacting analysis of the factual case advanced in defence of the measure, in order to determine (i) whether its objective is sufficiently important to justify the limitation of a fundamental right; (ii) whether it is rationally connected to the objective; (iii) whether a less intrusive measure could have been used; and (iv) whether, having regard to these matters and to the severity of the consequences, a fair balance has been struck between the rights of the individual and the interests of the community."

While, as David Hart QC has observed, Lord Reed in Bank Mellat constructed a four-part proportionality test that is less vague than the proportionality standard applied by the Strasbourg court, the Strasbourg court cannot apply as nuanced a proportionality analysis as a national court can do, since the Strasbourg court must recognise more explicitly the nature of the domestic, national 'margin of appreciation'[70]. Lord Sumption in Bank Mellat, in accordance with this more granular, domestic approach that can be favoured by the Supreme Court, deployed the above 'quadripartite' test in his judgment, saying that the proportionality test could be 'sufficiently summarised' in that way. However, this quadripartite proportionality test is absent from the text of Lord Sumption's judgment in Catt. In Catt (at para. 17), Lord Sumption noted that:

"In my opinion, the retention of data in police information systems in the United Kingdom is in accordance with law. The real question on these appeals is whether the interference with the respondents' article 8 rights was proportionate to the objective of maintaining public order and preventing or detecting crime. For this purpose, it is necessary to look separately at the two cases [that of Catt, and that of T] before us, for the relevant considerations are very different."

However, in Catt, when considering the proportionality of the interference with the Article 8 ECHR rights of both Catt and T, Lord Sumption did not expressly apply the four-part test he deployed in Bank Mellat. The approach Lord Sumption (and other Supreme Court justices in the panel for the case) seems to have taken is that the challenge in Catt was to the police intelligence system of retaining pertinent information of various kinds, about individuals, and therefore about 'risky' groups, in the 'jigsaw' concept outlined above. We do not see whether Lord Sumption has considered that a less serious, or more minimal interference in the rights of an individual like John Catt (as per the third part of the quadripartite proportionality test) would include the systematic review and deletion of his records over time, however costly, where a person has demanded it. In essence, there is no substantive consideration of the idea of a 'more minimal interference' principle in the case to the detriment of the rights of those in the same situation as John Catt in the foreseeable future. The principle of the overarching necessity of the 'jigsaw' is established, to the benefit of operational, intelligence-led policing, particularly in relation to public order circumstances and risks.

Tellingly, there is also no specific consideration in the entire Catt judgment as to whether the retention of the data in Catt's case, or T's case, as individual matters, was something that addressed a 'pressing social need' - the test regularly and routinely used by the European Court of Human Rights to determine if the interference with the qualified right in Article 8 was 'necessary in a democratic society' (and thus part-way to being deemed lawful, in alliance with tests for 'legitimacy' and 'proportionality'). Lady Hale does admittedly highlight that in her view, the crucial factor is as much whether the retention of information was 'necessary in a democratic society' as whether it was proportionate retention [49]. Lord Toulson also considers the issue of the necessity of retaining information about John Catt in the relevant database; and, however, finds the required necessity lacking, and would have seen Catt's case (though not T's) succeed [65] [71]. Perhaps if Catt were to appear at Strasbourg, the discussion would turn more on whether the retention of police intelligence concerning an individual, such as Catt, who presented such little risk of violence as a demonstrator, really did constitute, as an interference with the Article 8 ECHR right to respect for private life, a response to a 'pressing social need' to monitor and place under ongoing surveillance those who presented a risk of public disorder or serious harm.

However, as noted above, Lord Sumption deploys the test for 'legitimacy' ('in accordance with the law') but not a test for proportionality of his own devising - or anyone else's. Since the relevant test for 'necessity', at least in the form of a 'pressing social need' is also not deployed in his judgment, it is difficult not to turn, questioningly, to the traces of policy and those ideological stances relied upon by Lord Sumption in justifying his dismissal, ultimately, of both of the cases brought by Catt and T alike.

Perhaps the outcome in Catt would have been different if the case was about a nominal record, rather than a passing reference to an individual in a database; since Lord Sumption implies that an unlinked record is not as particularly intrusive as an 'indexing' or 'linking' record that provides greater 'searchability'. But this seeming nuance of database operation and design does not take into account the true, current functionalities or potential future functionalities of database technologies; whereby (assuming steps are taken to consolidate the variety of records management systems used by police forces) all and any scraps of police intelligence will be summoned as easily as those currently linked by PNC/PND 'nominal records', or even more so.

Lord Sumption's policy stance seems to be that the mere inclusion (even the indefinite inclusion) on a police intelligence database, is, ultimately, nothing to be concerned about, or nothing to be complained about. To quote him at length, he writes (in Catt at para. 27) that:

"The retention in a nominal record about a particular person or in an Information Report about a demonstration of information about other persons such as Mr Catt who were participating in the same event does not carry any stigma of suspicion or guilt. Mr Catt takes exception to what he regards as the inference that all those mentioned as participating in events such as Smash EDO protests are "extremists". But that is not a fair inference. The relevant police units are concerned with "extremism", in the sense of the pursuit of a political cause by criminal means, but it does not follow that all those who are recorded as attending these events are being characterised as extremists in that or any other sense. Unlike the records of criminal convictions or cautions, the information would not be regarded as discreditable to those who were merely recorded as attending an event at which they were not alleged to have committed offences. But in fact, the material is not usable or disclosable for any purpose other than police purposes, except as a result of an access request by the subject under the Data Protection Act. It is not used for political purposes or for any kind of victimisation of dissidents. It is not available to potential employers or other outside interests. There are robust procedures for ensuring that these restrictions are observed. Finally, the material is periodically reviewed for retention or deletion according to rational and proportionate criteria based on an assessment of danger to the public and value for policing purposes."

There are two key points to make about this long passage quoted here. Firstly, when he writes that 'unlike the records of criminal convictions or cautions, the information would not be regarded as discreditable to those who were merely recorded as attending an event at which they were not alleged to have committed offences ', Lord Sumption does appear to be treating 'intelligence' differently and regarding it as less intrusive, at least in Mr Catt's case, than criminality information as we would define it, taxonomically. We might assume that, in Lord Sumption's view, Mr Catt was hardly on the radar, let alone under the microscope. Secondly, there is an assumption here about the subjective feeling experienced by those placed under state surveillance in public places, whether overtly, or as some people might assume when protesting in public, covertly, and the idea that there is no reason to fear potential political victimisation from police surveillance. An appreciation of the extent to which police surveillance might have been a part of the long-term employment 'blacklisting' practices in the construction sector [72], or the way that undercover officers have been alleged (and have even admitted) to infiltrating protest groups contesting laudable causes such as police discrimination and victimisation itself[73], gives a different part of the jigsaw on police intelligence practices that Lord Sumption has overlooked.

There is also no sense of any recognition here of the argument that, for those who do feel retention of surveillance and intelligence records describing them is an intrusion by the state that warrants justification, the effect might be a 'chilling' of the desire to protest, and corollary reduction in the ability to utilise one's own freedom of expression. For Lord Sumption (at para. 31), criticisms or indeed fears and scepticism toward the purpose for retaining information on police databases:

"need to be considered in the light of some basic, and perhaps obvious, facts about the nature of intelligence-gathering. Most intelligence is necessarily acquired in the first instance indiscriminately. Its value can only be judged in hindsight, as subsequent analysis for particular purposes discloses a relevant pattern. The picture which is thus formed is in the nature of things a developing one, and there is not always a particular point of time at which one can say that any one piece in the jigsaw is irrelevant. The most that can be done is to assess whether the value of the material is proportionate to the gravity of the threat to the public."

There is instead a prominent consideration given to the issue of resources, costs and pragmatisms by Lord Sumption, who writes in Catt (at para. 32) that:

"The current weeding process in relation to nominal records involves an assessment of the threat posed by the subject of each such record. Mr Catt is not the subject of a nominal record, but merely appears as part of the cast in incidents with which the subjects of nominal records are associated. To fillet all the nominal records not simply in order to review the retention of information relating to the subject of the record but to examine the individual position of every other person mentioned in it would be a major administrative exercise. The alternative of not retaining information in a nominal record about any other members of the cast would significantly undermine the value of the record."

The current required scope of the 'administrative exercise' of reviewing then deleting or retaining police intelligence such as 'event histories' is set out in the guidance published by more than one relevant information regulator; and it is this guidance to which the next Part of this piece now turns.

8. Recent changes to guidance on the retention and deletion of police records made following the decision by the Supreme Court in Catt

To a certain, more practical extent, the oversight and regulation of police intelligence databases is achieved through the self-regulation by the police of intelligence databases and the role of the relevant College of Policing guidance on 'Authorised Professional Practice' (or 'APP') [74], the relevant statutory Code of Practice in the form of the Management of Police Information guidelines [75], and the newly created guidance published by the National Police Chiefs' Council (the 'NPCC') on the deletion of records from police systems [76]. Importantly, sitting on top of this regulatory framework is the important constitutional notion that, in the UK at least, "[r]equests for the deletion of information which are refused are challengeable by way of judicial review." [77]

Recent guidance on the use, scope and retention of Police National Database and other local/specialised police records is to be found in the form of the latter document, produced by the NPCC (the inheritor of the regulatory role previously undertaken by ACPO, or the national Association of Chief Police Officers), which notes that:

"Locally held records, whether stored on other electronic document management systems or in manuscript, are managed by Chief Officers in accordance with Accredited Professional Practice (APP) [on] Information Management issued by the College of Policing. In this regard, it should be understood that this Guidance does not currently extend to the deletion of custody photographs." [78]

As for the aims of the NPCC, as the new quasi-regulator of police intelligence or criminality information management practices, their recent guidance observes, with a heavy nod towards the importance of compliance with both the law and central government policy, that:

"The Government wants to protect the civil liberties of innocent citizens, whilst giving police the powers they need to identify suspects and solve crime using DNA and fingerprints (hereafter referred to as biometric information). The Government also recognises that there is a requirement for the police to hold certain information about an individual's criminal antecedents for their policing purposes and to satisfy the requirements of criminal justice partners e.g. the courts. This information includes convictions, out of court disposals and other 'Event Histories'." [79]

With the Catt decision in the Supreme Court validating the status quo of criminality information or police intelligence designed as largely permanent, the NPCC have outlined the relative inflexibility and comprehensiveness they envisage for the single largest repository of data used by the police:

"PNC records are retained until a person is deemed to have reached 100 years of age. However, Chief Officers can exercise their discretion, in exceptional circumstances, to delete conviction records, specifically those relating to non-court disposals e.g. adult simple cautions and conditional cautions as well as any 'Event History' owned by them on the PNC but only where the grounds for so doing have been examined and agreed… A person may have an 'Event History' recorded on the PNC even though they have only come to the attention of the police, LEA or NPPA on one occasion and regardless of whether that one occasion resulted in the person being convicted of an offence." [80]

There are some criticisms we might make of the wording of the recent NPCC guidance, in an Appendix outlining factors affecting decisions to delete records from the PNC, however. This Appendix to the guidance asserts that:

"There is no set criterion for the deletion of records e.g. "beyond reasonable doubt" or "balance of probabilities"; it is for Chief Officers to exercise professional judgment based on the information available." [81]

But this assertion (which is about the standard of proof required to convince the chief officer concerned to order a deletion) is flawed. Diktats about a threshold for a standard of proof, which should rightly be the 'balance of probabilities' in a matter of determining civil rights, such as this, should also only be an assertion about the necessary certainty with which a chief officer must appreciate the component elements of the pertinent factual circumstances of an applicant seeking the deletion of the removal, and when weighing them in the balance. The decision as to whether to delete or retain data on the PNC following an application by an individual to have it removed must be a decision based on the key requirements of necessity, legitimacy and proportionality, and chiefly the latter, when appreciating those factual circumstances. This key paragraph in an appendix to the relevant guidance somewhat disastrously undermines the rights of potential applicants seeking the deletion of their PNC records, or 'Event Histories'.

Another element of the wording of the recent NPCC guidance is somewhat at odds with the rationale deployed by the majority of the Supreme Court in deeming the retention of police intelligence records about John Catt to be justifiable and therefore lawful. While in the Catt judgment Lord Sumption, and to a lesser extent other Supreme Court justices, actually acknowledged the risks posed by certain groups of (potential) offenders in society as a justification for retaining information about their associates more broadly, the NPCC guidance actually states that:

" Chief Officers will consider applications on an individual basis and will not set retention periods for groups of individuals, however defined. "[82]

And rightly so, since to do otherwise would arguably be an unlawful 'fettering of discretion' by the chief police officer who did engage in this kind of group profiling. But where does this leave our understanding of the supposed need to retain information about individuals on the fringes of organised crime, terrorism, drug distribution and football hooliganism, as per Lord Sumption in Catt, however? In many ways it could be said that Catt has done little but create a concern over the potential chilling effect of the judgment on the enjoyment of the right to protest in public.

9. Conclusion: Looking ahead

In reality, legalities of police retention of personal information, whether as 'criminality information', or 'police intelligence' data, are never going to be completely settled. Technological advances in the field of intelligence gathering, recording and data analysis alone would see to that, we suspect.

The resources issue, in terms of the higher costs that would come with the more nuanced 'weeding' of records, combines powerfully with arguments for the use of fully comprehensive intelligence databases to help the police protect the public, in the face of any privacy rights or civil liberties arguments to the contrary. And yet those latter arguments, in favour of stronger and continuous police accountability over the regulation of 'criminality information' and the retention of records identifying individuals on intelligence databases, must nonetheless continue to be made.

We would tentatively suggest a number of conclusions and recommendations arrived at through our discussion in this piece of both the law relating to police intelligence and evolving police regulatory guidance on the retention of that police intelligence:

There is a lack of definition around the notion of police intelligence itself, both in case law and in guidance and policy (as 'soft law'). It can mean information collected in the course of investigations, or it can mean the product of analysis (intelligence products) used to draw inferences and conclusions, and make decisions. The retention of unassessed intelligence could be viewed as less intrusive than interpreted or assessed intelligence (for instance, retaining the fact that Mr Catt attended a demonstration is less intrusive than recording him as an associate of violent demonstrators with the implicit label that this suggests). We conclude that criteria in case law and regulatory guidance for intelligence retention lack clarity and coherence. We believe that not enough attention has yet been paid to existing and potential electronic data analysis techniques. Further cases are inevitable, and future applicants may well use Davis to support an argument for independent review of retention decisions. Ultimately case law is not the appropriate vehicle to which police decision-makers should refer. The Government needs to address, through regulation or code, the extent to which information about all of us is 'on the radar' while ensuring that safeguards exist to prevent misuse of the microscope. There are worse starting points than those criteria for the governance of criminality information sharing, as identified by Lord Neuberger in L and recently endorsed by Mr Justice Behrens in SD, of reliability; gravity; relevance; currency; impact; and the overall proportionality of the sharing (and here the retention) of information.

Furthermore, there appears to be a lack of corporacy in the management of information across police forces, suggesting the need for primary legislation covering both the definitions, and the ongoing retention and use of categories of information in order to improve consistency of approach. It would be advisable for such statutory framework to go further than asking Chief Constables to have regard to the rules. Chief Constables may resist any suggestion of compulsion, but as the technological aspects of policing become ever more important, so does the importance of ensuring that information accessed is accurate and relevant. Such statutory framework would inevitably need to address the difficult arguments as to when information, and in particular intelligence, should be deleted, forcing an acceptance that forces must be prepared to delete in some circumstances, despite the risk to future investigations. Just as importantly, any new framework must facilitate the police's ability to make valuable links across records in different forces and at the same time, restrict an individual force's ability to make unilateral, blanket or automated decisions about deletion of certain records.

One uncertainty on a practical level that the Catt decision in the Supreme Court has created is with regard to the regulatory and oversight role that the Information Commissioner's Office [83] will have in relation to police intelligence databases in the future, after obiter dicta in Catt (at para. 45), where Lord Sumption observed that: "the parties [in Catt] have gone through three levels of judicial decision, at a cost out of all proportion to the questions at stake."

Indeed, the recent guidance from the NPCC on the process for potential deletion of police records states that

"A person affected by the processing of personal data can request an assessment from the Information Commissioner on whether the retention and processing of their personal data complies with DPA requirements… If the Information Commissioner is asked to make an assessment the Chief Officer should provide all the information that the Information Commissioner reasonably requires to make the assessment… If the Information Commissioner determines that the processing is likely to contravene the DPA then the Chief Officer must review the original decision to retain the relevant records having due regard to the Information Commissioner's assessment." [84]

We wonder though if the ICO is best equipped (or resourced) to deal with challenging and specialist necessity and proportionality human rights arguments. There is no single independent oversight body for police databases, since their operation is regulated by a combination, in practice, of the College of Policing, the NPCC, HMIC, the Home Office, and more than 40 police forces all across the UK - not to mention legal oversight from the courts and the ICO. There has been a proposal by the Independent Surveillance Review for a single National Intelligence and Surveillance Office [85]; and the oversight, governance and accountability functions of this body could be extended to cover the retention of police intelligence on databases as well as interception of communications/bulk communications data. In addition or in the alternative, there may well be room for an Information Management Commissioner, independent from the Inspectorate function, with a role to motivate and assist police forces with information management requirements. Such bodies are likely to uncover inconsistencies in approach due to the sheer number of police authorities, all independently making decisions about retention. This adds another dimension to the finely balanced debate over the desirability of a national police force in England and Wales. [86]