History of a Dog’s Dinner
Keith Ewing and Conor Gearty on the Police Bill
The rule of law means different things to different people, but at its core it means that government must be conducted in accordance with the law, and must have legal authority for its actions. The principle was established as long ago as 1765, in Entick v. Carrington, a seminal case in which the King’s Secretary issued a warrant authorising two messengers to enter the home of John Entick and search for seditious papers. There was neither common law nor statutory authority for this action, which Entick successfully challenged in the courts, recovering damages from the hapless messengers (said to be ‘as much responsible as their superiors’) for the trespass to his property.
In giving one of the great judgments in the history of English jurisprudence, Lord Camden expressed alarm at the idea that ‘the secret cabinets and bureaus of every subject in this kingdom’ could be ‘thrown open to the search and inspection of a messenger, whenever the Secretary of State shall think fit to charge, or even to suspect, a person to be the author, printer or publisher of a seditious libel’. Under the authority of the warrant issued in this case, Lord Camden reflected, Entick’s house could be searched, and ‘the lock and doors of every room, box or trunk ... broken open’.
In upholding Entick’s claim for damages, the court not only made it clear that the State must have legal authority before it violates the legal rights of citizens, but pointed to the need for such authority before anyone’s person or property could be violated or subjected to search and seizure. ‘By the laws of England,’ Lord Camden said, ‘every invasion of private property, be it ever so minute, is a trespass. No man can set his foot upon my ground without my licence, but he is liable to an action, though the damage be nothing.’ Anyone who so much as ‘bruises the grass’ or ‘treads upon the soil’ of another ‘is bound to shew by way of justification, that some positive law has empowered or excused him’. That the power was ‘essential to government, and the only means of quieting clamours and sedition’ was no justification for conduct which could be sustained only by ‘the text of the statute law, or by the principles of the common law’.
Why should the ‘rifling’ of someone’s house and the removal of his ‘most valuable secrets’ be a matter of such concern? The reasons given by Lord Camden are as relevant now as they were in 1765. First, the power in Entick v. Carrington was exercised merely on suspicion, before anything had been established against the citizen who was subjected to the treatment. The intrusion into his property was ‘executed against the party, before he [was] heard or even summoned; and the information as well as the informers [were] unknown’. Second, Lord Camden drew attention to the vagueness of the procedures accompanying the execution of the power, with the search taking place ‘in the presence or the absence of the party, as the messengers shall think fit, and without a witness to testify what passes at the time of the transaction’. Third, Camden was concerned that the innocent person was ‘as destitute of remedy as the guilty: and the whole transaction ... so guarded against discovery that if the officer should be disposed to carry off a bank-bill, he may do it with impunity, since there is no man cap able of proving either the taker or the thing taken.’
None of these is an argument against surveillance by the police or any other State agency, and they were not intended as such in Entick v. Carrington. They are, however, compelling arguments in favour of limited powers, restrained in the manner of their exercise and capable of being deployed only after the most testing form of authorisation. State powers of surveillance are exercisable against the innocent as well as the guilty and are inevitably employed in great secrecy and without the knowledge of the target. They also constitute the most serious violation, not only of property rights recognised by the common law, but also of fundamental freedoms recognised by the European Convention on Human Rights. Even if judged to be necessary, such powers should be subjected to the most searching, safeguards, limiting their exercise and protecting the citizen from their over-enthusiastic use.
The principle of English law that reached its highest expression in Entick v. Carrington demands that a judicial warrant be required before a person’s liberty is violated by the State. This applies both when an individual is arrested and detained and when it is proposed to enter and search his or her property. Both these situations are governed by the Police and Criminal Evidence Act of 1984. In recent years, however, legislation has been passed that enables the Home Secretary to issue warrants both to the security services and to the police to interfere with the personal privacy of individuals. Although the dilution of the role of the judiciary at this stage in the process is to be deprecated, and would have been savaged by such a judge as Lord Camden, it is crucially important that the power to interfere with someone’s privacy which these laws give to state agencies be authorised by someone other than the security services or the police themselves.
As a general rule, an arrest must be authorised by a magistrate’s warrant, though in many cases this formality may be dispensed with. But, except in cases where the so-called ‘general arrest conditions’ are met, the power to arrest without a warrant applies only to ‘arrestable offences’ – defined by the 1984 Act as offences for which someone could expect to be sentenced to five years for a first conviction. After arrest, a person may be detained by the police for only 24 hours, except in the case of a ‘serious arrestable offence’, when a detention for a further 12 hours may be authorised by a senior police officer. Any detention beyond 36 hours requires a magistrate’s warrant, and may be granted only in the case of a serious arrestable offence, for further periods of 36 hours up to a maximum of 96 hours. Thereafter, the person must be charged or released.
So far as entry and search are concerned, a judicial warrant is again required, and, again under the 1984 Act, a warrant may be granted only where there is reason to believe that a ‘serious arrestable offence’ has been committed and there is material on the premises which is likely to be of substantial help in the investigation of an offence. The warrant will normally be granted by a magistrate, but in some cases the consent of a circuit court judge is needed, and in others the material in question may not be recovered by the police at all. The former category applies to ‘excluded’ and ‘special procedure material’ including personal records, journalists’ files and business records. The latter applies particularly to ‘items subject to legal privilege’ – to communications between lawyer and client.
But this principle of judicial authorisation has been seriously undermined in recent years. The first step was the Interception of Communications Act of 1985, which made it an offence to tap telephones or open mail without a warrant, while providing that the police and the security services need only apply to the Home Secretary, the Foreign Secretary, the Northern Ireland Secretary or the Secretary of State for Scotland, as appropriate, rather than to a judge. The 1985 Act formed the basis of the Security Service Act of 1989 and the Intelligence Services Act of 1994, which together confer statutory powers on MI5, MI6 and GCHQ. Executive rather than judicial authorisation was again to the fore. Particularly controversial was Section 3 of the 1989 Act, which empowered the Home Secretary to issue a warrant to the security services authorising ‘entry on or interference with property’.
This power was extended in 1994, and again in 1996, when the role of the security services was broadened to include supporting the police and other law enforcement agencies in the prevention and detection of serious crime. This meant an increase in the circumstances under which surreptitious entry could be made on the authority, not of the courts, but of a Secretary of State. The only difference between this and the situation in Entick v. Carrington is that the modern executive has secured a Parliamentary sanction for its actions. The point was well made by Lord Browne-Wilkinson during the debate on the second reading of the 1996 Act, which he described as a ‘major constitutional shift’. As he understood the Act, ‘an officer of the security services’ can obtain ‘a warrant that is not signed by a judge or a magistrate but by the Home Secretary or someone to whom that power is delegated’. That officer, he continued, ‘can enter, search and remove things from a private house. The policeman with whom he is co-operating can do no such thing. He requires a warrant.’
There is one important exception to the requirement (albeit increasingly diluted) for prior authorisation. In a process patently without formal legal authority, the Home Office has allowed the police to place bugging devices on people’s property without recourse to any procedures set out in any Act of Parliament or found in any decision of the courts. The relevant Home Office guidelines stipulate only that the use of such devices be authorised by the relevant chief constable if he is satisfied that the police investigation which is underway concerns ‘serious’ crime; that normal methods of investigation have been tried and failed (or must in the nature of things be unlikely to succeed if tried); that there is good reason to think that the use of the equipment would lead to an arrest and conviction, or would prevent an act of terrorism; and that the use of the equipment is operationally feasible.
The guidelines exhort chief police officers to ensure that the invasion of privacy is ‘commensurate with the seriousness of the offence’. In other words, they embrace a doctrine of proportionality which we might expect to see in any legislation on this subject. They also specifically provide that where the targets of surveillance ‘might reasonably assume a high degree of privacy, for instance in their homes, listening devices should be used only for the investigation of major organised conspiracies and of other particularly serious offences, especially crimes of violence’. This, too, is a welcome qualification, and one that might be expected to make its way into any statute on the subject.
But the problem identified by Lord Camden remains: there is no authorising statute, so it is not at all apparent what action can be taken by the citizen if any of the criteria are breached. How does the citizen enforce restrictions on the perpetration by the police of unlawful acts, acts no less unlawful for the unpalatable fact of Home Office collusion and the apparent endorsement by the courts, whose own position is all the more extraordinary in that some senior judges have been in the vanguard of the opposition to the Government’s attempt in the ill-fated Police Bill to place on a statutory footing a version of Home Office procedures giving more leeway to the police? Despite this political engagement on the part of some-senior judges, the juridical response to the use of these procedures has been far from satisfactory.
It is enough to refer to the 1996 decision of the House of Lords in R. v. Khan, which was concerned with the use in legal proceedings of evidence obtained by means of bugging devices unlawfully placed by the police. The appellant in the case was not the most attractive of characters: he had been jailed for three years in connection with the importation of heroin. The case against him, however, was based entirely on a tape-recording of a conversation between him and someone else obtained by means of a listening device placed on the outside of the home of his interlocutor by the South Yorkshire Police. This evidence was ruled admissible by the trial judge despite the fact that ‘the attachment of the listening device had involved a civil trespass, and had occasioned some damage to the property.’
The question for the House of Lords in the appeal was simply whether the evidence was admissible. The Lords held that the trial judge was correct to admit it despite the discretion vested in criminal courts under Section 78 of the Police and Criminal Evidence Act of 1984 to exclude evidence obtained in circumstances the admission of which ‘would have ... an adverse effect on the fairness of the proceedings’. ‘As a matter of English law,’ the Lords reminded us, ‘evidence which is obtained improperly or even unlawfully remains admissible.’ It is one of the more striking defects of the legal system that, while John Entick could sue to recover damages for trespass, as the law currently stands he could still be convicted on the basis of the evidence unlawfully obtained. Paradoxically, the end (state necessity) may not justify the means (violation of civil liberties) in civil litigation, but it most certainly does in criminal proceedings.
The powers given by Parliament to the various security services at least have the virtue of requiring authority in advance from someone other than the concerned state agency itself. The police practice in respect of the bugging of private property is unlawful because it is unauthorised by legislation or the common law. Yet in proposing now to legislate to give the police extensive powers in this area, the Government has turned to this unlawful precedent. Before it was butchered by the House of Lords, Part III of the Police Bill proposed to permit the police to authorise themselves to enter a person’s premises, to interfere with his or her property and to place surreptitious listening devices. The only comparable provision on the statute book is the notorious Section 9 of the Official Secrets Act of 1911; and even this section, around which controversy has raged for decades, normally requires a judicial warrant for entry and search for an offence under the Act. Only in cases of ‘great emergency’ does the Act permit a police officer to authorise an entry and search and even then only when ‘in the interests of the state’ the officer believes ‘immediate action [to be] necessary’.
The key provision of the Government’s proposal was that ‘no entry on or interference with property or with wireless telegraphy shall be unlawful if it is authorised under this section’. The proposed power so to authorise was extraordinarily wide – without limit both as to the nature of the property and the nature of the interference. It allowed for the bugging of premises, business as well as domestic, with listening devices and anything else the police may have deemed appropriate. There would be no restriction: lawyers’ offices, medical practices, newspaper offices or even a priest’s confessional could all be entered, interfered with or bugged on the say-so of a police officer. All of this is a far cry from the existing legal powers of the police on entry and search to be found in the 1984 Act.
Under that Act, it is impossible to obtain by court order or any other means items subject to legal privilege, and it is possible to obtain personal records, journalistic and other confidential material only after a careful procedure has been gone through, involving an application to a circuit judge. Even the most trivial of entries unconnected with an arrest or other special emergency requires an application to a magistrate for a warrant, to establish that the police have reasonable grounds for believing that ‘a serious arrestable offence has been committed,’ that ‘there is material on premises specified in the application which is likely to be of substantial value ... to the investigation of the offence’ and that ‘the material is likely to be relevant evidence.’ If the Government’s proposals were passed into law, the police would be able effectively to bypass these stipulations completely.
The only limitation on the proposed new power applied to the purposes for which it could be used. Entry into, or interference with, property could be authorised only where the authorising officer thought it necessary ‘on the ground that it is likely to be of substantial value in the prevention or detection of serious crime’; and was satisfied that ‘what the action seeks to achieve cannot reasonably be achieved by other means.’ There was no mention of any need to respect privacy or to apply any principle of proportionality. ‘Serious crime’ was broadly defined, covering conduct which involved the use of violence or resulted in substantial financial gain as well as conduct by ‘a large number of persons in pursuit of a common purpose’. It would have applied also to offences for which someone could reasonably expect to be sentenced to three years or more on a first conviction.
The application of the new procedures to the detection and prevention of ‘serious crime’ was one of the most remarkable features of Part III of the Bill, when the power to obtain and execute a search warrant applies only to ‘serious arrestable’ offences, a much more narrowly defined category. This could have a number of bizarre consequences. Suppose, for example, that the police were investigating an offence which was not thought sufficiently serious to authorise magistrates to issue a search warrant. In such a case they could nevertheless authorise themselves to enter and search private premises, and in the process open drawers and cupboards, rifle through papers, seize relevant material and place bugging devices. In other words, the Bill proposed paradoxically that more direct violations of civil liberties should be permitted in those cases where the matter under investigation is less serious. At the very least this required some explanation.
The decision in Entick v. Carrington provided strong reasons why secret powers of this kind, if they are to be granted at all, should be subject to tight safeguards. Two forms of control come readily to mind, both of which have been discussed in the controversy over the Police Bill. First, effective scrutiny before the power to interfere with property is granted, so that the case for using this power has to be made in the strongest terms – the sort of control exemplified by the traditional system of judicial warrants. Second (and less reassuring), the possibility of scrutiny and supervision after the event, so that any exercise of the powers would have to be justified to an independent third party.
Perhaps predictably, the Government (supported by Labour) proposed the latter and less effective option. The power to authorise the use of bugging devices would have rested initially with a chief constable or, in cases of emergency, with his deputy. A judicial commissioner would then have been appointed to investigate complaints about the improper use of the power, and to conduct an annual review of the way the chief constables had used their powers. These arrangements were to an extent modelled on those currently operating under legislation dealing with telephone tapping and the powers of the security and intelligence services, whereby judicial commissioners review their operation and report annually to the Prime Minister.
Fortuitously, however, the flaws in this form of supervision were effectively exposed by the recent revelations that telephone tapping files have gone missing, notwithstanding provisions in the Interception of Communications Act of 1985 designed to promote their security. Quite apart from failures of this kind, ex post facto scrutiny is no substitute for independent judicial authorisation in advance. After the event, the scope for action by even the most vigilant commissioner would be narrowly circumscribed by the terms of the Bill and further diminished by the fact that he or she could act only long after the alleged abuse had taken place. Moreover, all that the most effective scrutiny after the event can do is to reveal the existence of authorisations which should never have been made in the first place. As such it offers no remedy to the individual whose privacy ought never to have been violated.
Quite apart from the inevitable failings of ex post facto review, there are a number of simple questions to be asked and contradictions to be explored. If a judicial warrant is necessary to enter and search the property of someone suspected of having committed an offence and against whom there is evidence, on what rational ground should one not also be required to enter and place a device on the property of someone who may or may not be suspected of having committed an offence, and against whom there may be no credible or reliable evidence? If a warrant from a circuit judge is necessary before access is permitted to ‘special procedure’ or ‘excluded’ material, on what rational ground should one not be needed before a device is placed which would reveal information that in printed form would be regarded as ‘special procedure’ or ‘excluded’ material? If a warrant is necessary (albeit from the Home Secretary) before the police may intercept telephones to prevent or detect serious crime, on what rational ground should one not also be necessary before placing other surveillance devices, particularly where the latter (unlike telephone interceptions) may be compounded by a requirement to trespass onto someone’s property? Indeed, if Part III of the Bill were passed, what would stop elements within the police pushing for further changes, in telephone-tapping, judicial warrants and so on, so as to make the relevant powers also contingent on a police rather than on a judicial or political evaluation of facts? It would be more difficult to resist such pressure, the point of principle having already been compromised. And if all these safeguards were to fail, what would then stop the police arguing for the power to authorise themselves to detain suspects for up to four days, rather than for the 36 hours that at present applies?
The government has rightly been excoriated for attempting to introduce these far-reaching provisions into law. Not only do they go beyond the current (unlawful) practice, they are also self-evidently contrary to constitutional principle. It is tempting to think, however, that the changes proposed in Part III of the Police Bill do not matter much because, even if enacted, they will be struck down, either by British courts or eventually in Strasbourg, as a breach of the terms of the European Convention. Regrettably, the Convention is not the civil libertarian panacea it is widely believed to be, for two reasons. First, our existing legislation on wire-tapping and bugging has already been found to be in compliance with its terms; and, second, in such cases as R. v. Khan, the Convention does little to influence British procedures.
The procedures established under the Interception of Communications Act of 1985 and the Security Service Act of 1989 have both been held not to be in breach of the Convention guarantee of privacy, despite the fact that in both Acts executive action is authorised by an executive rather than by a judicial warrant. (Perhaps we should be grateful that at least this measure of prior authorisation is required, given the procedures to be found in Part III.) The 1985 Act was held, on a complaint by Campbell Christie of the Scottish TUC, to be compatible with the Convention. Similarly, three different complaints alleging that the procedures in the 1989 Act fell short of the demands of the Convention were rejected by the European Commission of Human Rights. One of these complaints was brought by Patricia Hewitt and Harriet Harman, and on that occasion the Government ‘without admitting as much, accepted ... a reasonable likelihood that the Security Service had compiled and retained information concerning their private lives’.
In the absence of any evidence or indication that the system was not functioning as required by domestic law, the European Commission found that the framework of safeguards achieved a satisfactory compromise between the requirements of the State and the rights of the individual – this is despite the fact that neither the tribunal set up under the 1985 Act nor its 1989 counterpart ‘had ever made a determination in favour of a complainant’. It is possible that a similar conclusion would be reached by the Commission in respect of Part III of the Police Bill even in its original form.
The value of the European Convention was tested again, but from a different angle, in the 1996 Khan case. The question here was whether evidence should be excluded because the manner in which it was obtained violated the protection of privacy as guaranteed by Article 8 of the Convention. In what was the first case before an English court to raise the question whether a trial judge is bound to follow the Convention in determining how to exercise his or her discretion under Section 78 of the 1984 Act, the House of Lords held that ‘if the behaviour of the police in the particular case amounts to an apparent or probable breach of some relevant law or convention, common sense dictates that this is a consideration which may be taken into account for what it is worth.’ The last few words indicate that this was hardly likely to advance matters very far, and so it proved.
The case law of the European Court of Human Rights offers few obstacles to the admission of illegally obtained evidence in criminal proceedings. The leading case is Schenk v. Switzerland (1988), in which the applicant was convicted of attempting to incite the murder of his estranged wife. The most telling evidence against him was contained in an unlawfully recorded telephone conversation. Schenk argued that this meant his right to a fair trial under Article 6 of the Convention had been compromised. The Court of Human Rights disagreed, finding, by a majority of 13 to four, no violation of this or any other provision of the Convention. The Convention, the majority declared, did ‘not lay down any rules on the admissibility of evidence as such, which [was] therefore primarily a matter for regulation under national law’. It is thus open to question whether Khan would have been decided any differently if determined by reference to the European Convention rather than domestic legislation.
The simple lesson from Strasbourg is that this is a problem which is not going to be resolved by the European Convention on Human Rights to the liking of the civil liberties lobby. But if it is a political problem to be resolved at Westminster, where do we go from here? In responding to opposition from all quarters, the House of Lords in January’s debate has, in the memorable words of Michael Howard, presented the Government with a dog’s dinner, by passing two mutually inconsistent amendments, at the instance of Labour and the Liberal Democrats respectively. There are now three options for him to consider: he can seek to push through his original proposals (which initially enjoyed the support of the Opposition’s Home Affairs team); or he can adopt the Labour amendment; or the Liberal Democrat one.
It is around the Liberal amendment that defenders of civil liberties should unite. The Labour proposal that bugging operations should be authorised by a judicial Commissioner is cumbersome, and sustainable only in the absence of a simple and more effective alternative. It is likely that the proposed Commissioners would be based mainly in London and this would diminish the opportunity for applications to be granted following hearings in person, and would also increase the likelihood of the police authorising themselves on the grounds of urgency. The Liberal Democrats in contrast have provided a principled, workable and sensible proposal that bugging operations should take place only after a warrant has first been obtained from a circuit court judge. It is, moreover, a proposal which could usefully be extended to other forms of police and security service surveillance, such as telephone tapping.
As the parties meet to consider the various options, there is above all a need for a principled solution to the constitutional problems presented by the current practices. The least satisfactory outcome of the current confusion would be for the Bill to implode in mutual recrimination, leaving the police to carry on as before without legal authority for a practice which has the evident support of the courts, while we wait for the problem to be addressed after the election by a possible Labour government – even though the Party has shown itself to be confused and uncertain on constitutional principle, despite its commitment to reform. A solution to this problem requires more, however, than simple adherence to what has become the slogan of prior judicial authorisation.
A number of principles ought to inform future legislation.
1. It is essential that legal authority is established to control bugging by the police. The current situation whereby police forces can act unlawfully with the evident approval of the Home Office is wholly unacceptable.
2. The power to enter or interfere with private property should apply only in respect of serious arrestable offences as defined by the Police and Criminal Evidence Act of 1984; other police powers of surveillance (including the interception of communications) should be similarly constrained.
3. An authorisation to enter or interfere with private property should be granted only by a judicial officer, preferably only by a circuit court judge. In deciding whether to issue authorisation, the judge should be required to be satisfied that the degree of intrusion into the privacy of those affected is commensurate with the seriousness of the offence.
4. An authorisation should be granted by a circuit judge only where there is good reason to think that use of the equipment will be likely to lead to an arrest and a conviction, or where it is judged appropriate to the prevention of acts of terrorism.
5. The power to enter or interfere with domestic premises should be subject to specific regulation. In the case of domestic premises, the use of bugging devices should be permitted only for the investigation of major organised conspiracies and other particularly serious offences, especially crimes of violence.
6. The special status of premises such as those of medical practitioners, solicitors and barristers, newspapers and churches should be recognised in the legislation. It is questionable whether bugging devices should be permitted at all in such locations, but if they are, it should only be in clearly specified cases.
7. Evidence obtained as a result of the unlawful use of surveillance devices should not be admissible in legal proceedings.