Stop and Search, Terrorism and the Human Rights Deficit

This paper discusses the judgment of the House of Lords in Gillan [2006] UKHL 12, where the law lords examined the compatibility of the power in ss. 44 and 45 of the Terrorism Act 2000 to stop and search individuals with the Human Rights Act 1998. It is argued that the decision of the House of Lords that ss. 44 and 45 are compatible with Convention rights was wrong for a number of reasons. Not only did the House of Lords fail to interpret and apply the applicable Convention rights correctly, it also missed an important opportunity to develop the principle of legality in the context of the Human Rights Act. Moreover, the House failed to deal adequately with the troubling question of whether the exercise and use of the stop and search power under the Terrorism Act was racially discriminatory. These issues take on a particular importance not only because of the greater use of the power since the so-called ‘7/7 attacks’, but also because of the disparate impact that the use of the power has on racial minorities.


I. Introduction
On 9 September 2003 Kevin Gillan, a PhD student at the University of Sheffield, was en route to a peaceful demonstration against the arms trade when he was stopped and searched by two Metropolitan police officers. Gillan was delayed for 20 minutes while the officers exercised their powers under ss. 44 and 45 of the Terrorism Act 2000 (TA) 1 to search for 'articles connected with terrorism'. 2 The search yielded nothing incriminating and no arrest was made. Gillan subsequently brought an action by way of judicial review against the Metropolitan * Department of Law, UWE Bristol. My thanks to Ed Cape, Howard Davis and the anonymous referee for their very helpful comments on this paper. The usual disclaimer applies. 1 Hereafter the Terrorism Act 2000 (c.11) is referred to as the 'TA' and ss. 44 and 45 of the Terrorism Act as 'ss. 44 and 45'. 2 44.
(2) An authorization under this subsection authorizes any constable in uniform to stop a pedestrian in an area or at a place specified in the authorization and to search-(a) the pedestrian; (b) anything carried by him.
Police Service contending, amongst other things, that the stop and search power contained in Part V of the TA was incompatible with his Convention rights as scheduled to the Human Rights Act 1998 (HRA). 3

II. An Exceptional Police Power: the Background
Sections 44 and 45 were enacted in 2000 by Parliament as part of the counterterrorist powers in Part V of the TA. The TA is primarily a consolidating measure intended to provide the UK with a permanent legal regime for dealing with terrorism. These permanent powers combine many of the previous statutory provisions passed as temporary, emergency provisions. But once again powers that were once viewed as exceptional and temporary have become permanent and unexceptional. 4 A legal regime designed to be a temporary derogation (3) An authorization under subsection (1) or (2) may be given only if the person giving it considers it expedient for the prevention of acts of terrorism. (5) If an authorization is given orally, the person giving it shall confirm it in writing as soon as is reasonably practicable. 45. (1) The power conferred by an authorization under section 44(1) or (2)-(a) may be exercised only for the purpose of searching for articles of a kind which could be used in connection with terrorism, and (b) may be exercised whether or not the constable has grounds for suspecting the presence of articles of that kind.
(2) A constable may seize and retain an article which he discovers in the course of a search by virtue of section 44(1) or (2) and which he reasonably suspects is intended to be used in connection with terrorism.
(3) A constable exercising the power conferred by an authorization may not require a person to remove any clothing in public except for headgear, footwear, an outer coat, a jacket or gloves. (4) Where a constable proposes to search a person or vehicle by virtue of section 44(1) or (2) he may detain the person or vehicle for such time as is reasonably required to permit the search to be carried out at or near the place where the person or vehicle is stopped. Exceptional violence requires an exceptional response. But with the passage of time Parliament has in most cases come to the view that these once temporary COMMON LAW WORLD REVIEW and developed to deal with a specific threat has been replaced by a permanent regime and intended to counter general threats. Indeed, the radiating effect of emergency laws has in modern times contaminated other parts of the criminal justice process. The powers contained in Part V of the TA have their origins in the Criminal Justice and Public Order Act 1994 (CJPOA). Before the passage of the 1994 Act the police would have to rely on s. 1 of the Police and Criminal Evidence Act 1984 (PACE) to stop and search individuals suspected of being involved in terrorist activities. From the police perspective s. 1 of PACE suffered from the handicap of requiring a reasonable suspicion. 5 The CJPOA changed this situation by amending the Prevention of Terrorism (Temporary Provisions) Act 1989 (PTA) to include a standardless power to stop both vehicles and people. In other words, the power is not constrained by any conditions precedent that need to be satisfied before it can be lawfully exercised. 6 By way of contrast, an ordinary stop and search under PACE requires that a police constable has 'reasonable grounds for suspecting that he will find stolen or prohibited articles'. 7 Under the PTA a police officer could stop an individual and search him for 'articles of a kind which could be used for a purpose connected with the commission, preparation or instigation of acts of terrorism' 8 although not the actual person. The power was qualified through the requirements that it must have been both authorized by an appropriate officer and its exercise must be connected with terrorism relating to either Northern Ireland or the affairs of the UK. Unless renewed the power lapsed after 28 days. These powers were judged to be necessary by the then Home Secretary in order to deal with an anticipated upsurge in Irish Nationalist violence aimed at targets in London, which subsequently failed to materialize.
In 1996 Parliament extended the power to include the search of persons. 9 The authorization process differed in two respects from the previous regime. First, the authorization could be given orally by the appropriate officer, although it would need to be confirmed later in measures are essential bulwarks in the 'war against drugs/crime/terror'. The exception has once more become the norm. A good example of this tendency in action is of course the changes to the right to silence enacted first in the context of Northern Ireland, and subsequently in the rest of the UK. 5 In the alternative an officer might have decided to make an arrest under s. 14 of the Prevention of Terrorism (Additional Powers) Act 1996, c. 7. 6 The CJPOA itself contained a similar new power in s. 60 where an appropriate officer can authorize standardless searches and seizures if he reasonably believes that incidents involving serious violence may take place in any locality in his area. 7 PACE, s. 1(3). 8 Prevention of Terrorism (Temporary Provisions) Act 1989 (c. 4), s. 13A, inserted by CJPOA, s. 81. 9 Prevention of Terrorism (Additional Powers) Act 1996 (c. 7), s. 5 inserting s. 13B into the PTA 1989. Although a person could now be searched as opposed to anything they were carrying, s. 13B limited that power: 'Nothing in this section authorizes a constable to require a person to remove any of his clothing in public other than any headgear, footwear, outer coat, jacket or gloves' (s. 13B(4)).
writing. And second, the Home Secretary was granted the power to superintend the authorizations. 10 As with much emergency legislation before, 11 and since, 12 the 1996 amendments and additions were rushed through Parliament with little effective debate, in this case in just one day. 13 The new provisions were justified on the grounds that the government feared a terrorist attack during the 80th anniversary of the Dublin Rising. Of particular concern was the possibility of attacks by nationalists armed with pocket incendiary devices. It subsequently became apparent that the need for such provisions had been considered some time before. 14 It was these temporary provisions that were re-enacted in the permanent TA, albeit with some amendments. 15 As it now stands s. 44 empowers a senior police officer to authorize searches under s. 45 where he 'considers it expedient for the prevention of acts of terrorism'. 16 An authorization may be made orally although it must be confirmed as soon as is reasonably practicable in writing. 17 Moreover, every authorization made under s. 44 must be confirmed by the Home Secretary within 48 hours, or it will lapse. In any case, an authorization lapses after 28 days unless renewed. 18 While authorized, a police officer may search vehicles and people for 'articles of a kind which could be used in connection with terrorism . . . whether or not the constable has grounds for suspecting the presence of articles of that kind'. 19 The power is thus derived from the authorization. And because it represents a departure from the normal standard applicable to police searches the power has, albeit to a limited extent, been surrounded by safeguards. 20 First, the power needs to be authorized by the appropriate officer with knowledge of the area where it will be applied. That activation is subject to confirmation by the Home Secretary. And second, given the extraordinary nature of the power, the authorizations should last for a limited time. However, so that the police can rapidly respond to a threat should it suddenly arise, the relevant officer has the power to authorize searches orally. Indeed, it might be argued that the scheme in Part V seeks to balance the need for a special power against the broader interests of a free society by limiting the departure from the legal norm to situations where it is necessary. However, whatever the theory, since the Terrorism Act came into force there has been a substantial period where a series of 'rolling authorizations' has been in force in the Metropolitan and City of London police areas. Moreover, not only were these authorizations not a response to specific threats, but stops and searches under the authority of ss. 44 and 45 were authorized even before 9/11. In practice an exceptional power intended to be employed on a limited basis has, through administrative action, become quasi-permanent. By contrast, the powers under the old PTA had been used more sparingly. 21

III. Judgments
In the High Court Mr Gillan challenged the exercise of the powers on three grounds. 22 First, it was contended that Parliament clearly intended that the authorization of the power in s. 44 was to be limited to circumstances where there was judged to be an imminent threat to a specific location and where ordinary police powers were likely to be insufficient. Brooke LJ, who gave the judgment of the court, disagreed. Parliament had clearly intended that an authorization could cover an entire police area and not simply a location as s. 60 of the CJPOA did. 23 Parliament intended to grant a broad discretion to the appropriate officer because 'a need might crop up at any time for the targeted anti-terrorism patrols, the mobile road checks or the other elements of the anti-terrorism strategy, and it is understandable why, if there is a long-standing general threat, the police wished to be able to exercise these powers anywhere in the police district'. 24 In deciding to issue an authorization, matters of public safety and national security would need to be considered, and these were matters requiring expertise possessed by the police and Home Secretary. Some degree of deference was therefore required. 25 The second strand to the application was the contention that the police had acted unlawfully when exercising their ss. 44  Where a police officer of or above the rank of superintendent reasonably believes that-(a) incidents involving serious violence may take place in any locality in his area, and (b) it is expedient to do so to prevent their occurrence, he may give an authorization that the powers to stop and search persons and vehicles conferred by this section shall be exercisable at any place within that locality for a period not exceeding twenty four hours. they had insufficient guidance on how to exercise that power. This was an issue which required anxious scrutiny: 'The wider the power, and the more it impinges on personal liberty, the more anxious the court will be to ensure that it is used to achieve the purpose for which it was granted and not for any ulterior or extraneous purpose'. 26 However, in the current case there was just enough evidence to show that the powers had not been misused. 27 Nevertheless, the police needed to review their training and procedures in this area to ensure that the powers were not misused. 28 The third head of challenge related to the HRA. Notwithstanding the important human rights implications bound up in this case, the High Court summarily dismissed this argument. In Brooke LJ's opinion, if any Convention rights were infringed there was adequate textual justification in the European Convention on Human Rights (ECHR) to uphold their compatibility. 29 Only if the police were using these powers as part of day-to-day policing on the streets of London would there be force in this submission. 30 Gillan appealed. The Court of Appeal dismissed the appeal. In essence the Court of Appeal agreed with the High Court but it differed from the decision in the court below in two important aspects. First, the Convention rights points were more fully argued and considered by the Court of Appeal. These were ultimately unsuccessful. However, Lord Woolf CJ made a number of important points in relation to the application of Convention rights. With respect to the applicability of Article 5, the coercive detention on the street for the purposes of a search fell into a grey area under Article 5. 31 His lordship concluded that in the final analysis 'a short detainment pursuant to a stop and search will normally fall outside Article 5'. 32 Moreover, the scheme contained in the TA could not be said to be arbitrary. 33 The second point of difference with the High Court was the Court of Appeal's evident unhappiness with the absence of evidence submitted by the respondents by way of justification. 34 Lord Woolf CJ was rightly concerned with the paucity of evidence. Similarly, Lord Woolf CJ also expressed his concern at the absence of clear directions being given to officers as to how to employ their special powers. The instruction offered was quite simply 'lamentable'. 35  In his unsuccessful appeal to the House of Lords Gillan raised essentially the same points. In dismissing the appeal the House of Lords handed down a judgment which has important implications for Convention rights and the rule of law. In common with the judgments below, a unanimous House of Lords agreed that the powers in question were indeed sweeping but nevertheless justified given the terrorist threat which, they argued, the United Kingdom currently faces. 36 There were no grounds for concluding that Parliament intended the powers to be used only where necessary, given the express condition that their use be 'expedient'. The scheme was intended to give the police some degree of flexibility, especially in the post-9/11 environment. The powers were, however, subject to a 'wide variety of restrictions and safeguards' 37 and the possibility of civil action in the case of any alleged misuse. Similarly, the rolling authorizations could not be faulted, complying as they did with the letter of the law. 38 The applicant's arguments with respect to Convention rights were no more successful. The House of Lords unanimously decided that Article 5 had not been infringed. Lord Bingham concluded that as the procedure was so brief there was no deprivation of liberty within the terms of Article 5(1). 39 But even assuming that there was an interference with the right to liberty then in his lordship's opinion it would be justified within the terms of Article 5(1)(b). 40 The arguments with respect to Article 8 were similarly rejected. The search did not fall within the terms of Article 8 according to a majority. 41 Nevertheless, Lord Brown did conclude that privacy rights could be infringed although any interference was justified. 42 No issue arose under either Article 10 or Article 11.
Before the House two further points were pressed, neither with any success. Firstly, the applicants contended that the regime in ss. 44 and 45 of the TA was inconsistent with the principle of legality contained in the ECHR. 43 The authorizations were inaccessible, not having been published, and the power to stop and search was unfettered, broad 36  Everyone has the right to liberty and security of person. No one shall be deprived of his liberty save in the following cases and in accordance with a procedure prescribed by law: . . . the lawful arrest or detention of a person for non-compliance with the lawful order of a court or in order to secure the fulfilment of any obligation prescribed by law; 41 Above n. 36 at para. 28, per Lord Scott who concluded that an infringement would be 'theoretical'. 42 Ibid. at para. 75. 43 Ibid. at para. 32. and ill-defined. 44 Thus the power could be exercised arbitrarily, thereby negating the principle of legality. Lord Bingham was far from persuaded: 'The principle of legality has no application in this context'. 45 The legislation was detailed, specific and unambiguous. The powers in ss. 44 and 45 were defined with considerable precision. The authorization to bring the power into effect was not law and did not need to be publicized. Indeed, the effectiveness of the provisions may be undermined if there was a requirement to publish. The regime contained sufficient safeguards. 46 Moreover, in Lord Bingham's opinion, a police officer is not free to act arbitrarily and would be open to civil suit if he did. 47 In the 'real world a police officer will always have some reason for selecting a particular individual as a person to be stopped and searched'. 48 The best safeguard according to Lord Hope was to ensure that the powers were properly exercised through guidance and training. 49 The second additional point on appeal concerned the potentially discriminatory use of the powers. Although ultimately the law lords agreed that no racial discrimination had in fact occurred, they arrived at that conclusion by different reasoning. For instance, Lord Scott concluded that the use of the powers disclosed no discrimination on racial grounds and even if it had existed any such use could nevertheless be upheld as lawful within the terms of the Race Relations Act. 50 However, Lords Hope and Brown were more exercised by this point. As Lord Hope put it, 'the risk that it will be employed in a discriminatory fashion cannot be discounted entirely'. 51 The decision in the Roma Rights case 52 stood as a reminder that discrimination on the basis of racial origin was always suspect. Thus a person cannot be selected for a stop and search simply on the grounds of their racial origins even if the search is necessary to prevent terrorism. 53 The fact that a person is Asian is not a legitimate reason for the exercise of the powers, although ethnicity could be relied upon in drawing up a profile of possible terrorist suspects. 54 55 Indeed, while al-Qaeda remained one of the principal terrorist threats to the UK it was inevitable that a disproportionate number of Asian people would be stopped and searched under the powers. 56 While Kevin Gillan was ultimately unsuccessful in his challenge to ss. 44 and 45 of the TA, his case is nevertheless an important one for a number of reasons. First, Gillan's case throws into relief the inevitable tension between the liberty of the individual on the one hand and the counterterrorist requirements of the state on the other. The power in ss. 44 and 45 of the TA places 'the liberty of every man in the hands of every petty officer' 57 through the adoption of, in effect, a standardless discretion that arguably facilitates arbitrary action. 58 In such a context it is especially important that the judiciary develops an effective notion of the principle of legality to constrain the scope and exercise of what are, by any measure, exceptional powers. The principle of legality is an integral part of the Convention forming the keystone in the arch supporting the guaranteed rights and freedoms. Without a robust notion of the rule of law underpinning Convention rights and freedoms, the guarantee of rights and freedoms would be essentially theoretical. It is a significant weakness of the decisions that the issue of legality was inadequately dealt with.

IV. The Interpretation of Convention Rights
In Gillan's case the courts, particularly the House of Lords, interpreted Convention rights in a conservative manner that was overly deferential to the executive. This approach to judicial interpretation is nothing new under the HRA. But Gillan provides yet another clear example of the poverty of thought and simple lack of imagination of the senior British judiciary when deciding difficult cases under the HRA. For example, in reaching its decision the House of Lords was not assisted by Strasbourg authority. The European Court of Human Rights has yet to consider the interaction between Convention rights and police powers to stop and search individuals randomly. Thus their lordships were unable, as they have been in other HRA cases such as Bellinger, 59 to apply European jurisprudence. 60 Such circumstances create difficulties for the courts. On the one hand is the admonition for a court under the authority of s. 2 of the HRA 'not to cast around in the 55 61 And on the other hand is the view exposed by Lord Bingham in Ullah where his lordship concluded that a court should not 'without strong reason dilute or weaken the effect of the Strasbourg case law . . . it is of course open to member states to provide for rights more generous than those guaranteed by the Convention, but such provision should not be the product of interpretation of the Convention by national courts, since the meaning of the Convention should be uniform throughout the states party to it . . . The duty of national courts is to keep pace with the Strasbourg jurisprudence as it evolves over time: no more, but certainly no less.' 62 The problem with such a doctrine is that there is a temptation to treat the case law of Strasbourg as offering 'off the peg' solutions to cases. Where there is case law in point this, it might be argued, is not overly problematic. However, reliance on Strasbourg case law brings with it an inherent difficulty. The jurisprudential technique of the European Court of Human Rights has never been to identify explicitly the principles and purposes of the rights and freedoms in the Convention. 63 The European Court of Human Rights focuses heavily on the facts of the individual applications rather than engaging with the conceptual and philosophical issues underlying them. Consequently the Court has tended not to expound and elaborate its decisions into a coherent body of jurisprudence and as a consequence its case law is rather formulaic. For example, the proportionality test is often applied without a weighing of the limbs of the proportionality test. 64 The doctrine in Ullah has simply translated these flaws into British human rights jurisprudence, and has been responsible for the impoverished state of domestic human rights jurisprudence of which Gillan is an example. 65 This approach is explicitly predicated on the justification that the HRA was intended by Parliament to provide a domestic forum for the redress of Convention related claims without the need for an application to Strasbourg. 66 However, while this is true, for some reason the other intention of the HRA has been ignored. 'Enabling courts in the United Kingdom to rule on the application of the Convention', the White Paper, Rights Brought Home argued, 'will also help to influence the development of case law on the Convention by the European Court of Human Rights on the basis of familiarity with our laws and customs and of sensitivity to practices and procedures in the United Kingdom . . . Our courts' decisions will provide the European Court with a useful source of information and reasoning for its own decisions'. 67 In practice the doctrine in Ullah not only frustrates that intention, it has left the courts without the intellectual tools to reason adequately in human rights cases. 68

i. The Right to Liberty and Freedom of Movement
Mr Justice Holmes once remarked that 'it is one of the misfortunes of the law that ideas become encysted in phrases and thereafter for a long time cease to provoke further analysis'. 69 The jurisprudence of the European Court of Human Rights with respect to Article 5 is a good example of such a phenomenon. The leading case, Guzzardi v Italy, 70 is frequently applied in an almost mechanical fashion. And this is problematic since the test that the European Court of Human Rights enunciated under Article 5 for determining whether or not there has been a deprivation of liberty provides 'neither bright-line rule nor anything that begins to approximate to a clear definition'. 71 The European Court of Human Rights simply unhelpfully asserted, and has continued to do so ever since, that 'the difference between deprivation of and restriction upon liberty is nonetheless merely one of degree or intensity, and not one of nature or substance'. 72 Such an elastic definition provides little concrete guidance as to how it might apply in different contexts. In practice the Guzzardi test provides an almost open-ended judicial discretion for the application of Article 5. This characteristic of the jurisprudence is, no doubt, what exercised Sir Gerald Fitzmaurice to dissent in Guzzardi. Applying this nebulous test in Guzzardi the European Court of Human Rights extended Article 5 to cover a situation that was more properly dealt with by Article 2 of Protocol 4.
With such limited guidance it is perhaps not surprising that British courts have struggled when applying Article 5. Indeed, given the flaws of European case law in this context it is arguable that British courts could make a useful contribution to the development of Convention jurisprudence by injecting some analytical rigour. However, in Gillan instead of approaching the interpretation of Article 5 afresh and reasoning from principle, the House of Lords simply applied by analogy the 'encysted phrases' of Guzzardi v Italy. 73 No doubt because our courts believe they are duty bound to follow the jurisprudence of the European Court of Human Rights, no attempt was made to look for the underlying purpose of Article 5 or indeed to consider whether the security interests of the individuals had been adversely affected. Although we are told that the HRA is a constitutional instrument, being in effect a British bill of rights, 74 little attempt has been made to interpret it as such by the higher judiciary. In almost every comparable jurisdiction courts, when faced with such issues, begin with a purposive interpretation of the extent to which any constitutional guarantee applies. The purpose of a constitutional guarantee is to be understood in light of the interests that it is designed to protect. And this analysis 'is to be undertaken, and the purpose of the right or freedom in question is to be sought by reference to the character and larger objects of the Charter itself, to the language chosen to articulate the specific right or freedom, to the historical origins of the concept enshrined, and where applicable, to the meaning and purpose of the other specific rights and freedoms with which it is associated within the text of the Charter'. 75 Moreover, it is axiomatic that as it is a constitution that is being interpreted, the guarantees must be read and interpreted in a generous manner. 76 However, British courts have treated these principles as mere shibboleths that can be occasionally recited, and then safely ignored. In Gillan there was no attempt to interpret Article 5, or indeed the other applicable Convention rights, purposively. Indeed, it is hard to say that any of the applicable Convention rights were by any standard interpreted generously. There is a clear danger here that unless the HRA is interpreted as a constitutional instrument it will in time represent not a 'living tree', 77 rather a judicial bonsai.
The adherence to the rule in Ullah's case meant that a wealth of useful authority was overlooked. Although the European Court of Human Rights has not considered the issue of whether a police stop and search interferes with Convention rights, the issue has been 73 Subsequently, the absence of full and proper analysis in Guzzardi has generated more forceful criticism from members of the House who did not sit in Gillan considered in other jurisdictions. 78 In Gillan the House of Lords concluded that the powers in the TA were more an interference with freedom of movement than the right to liberty. The interference with the right to freedom of movement by random stops was considered by the Zimbabwean Supreme Court in Elliot v Commissioner of Police. 79 In Elliot the Supreme Court was asked to consider whether the absolute and uncontrolled discretion to stop members of the public randomly and demand the production of their identity cards was consistent with the Constitution. Under s. 10 of the Zimbabwean National Registration Act any person who failed to produce their identity card was guilty of an offence. 80 Gubbay CJ held that the impugned provisions infringed the constitutional guarantee to freedom of movement. The power in s. 10 permitted a random stop. The section did not specify any grounds or cause for stopping the person.
Consequently it could be employed in an entirely arbitrary manner. An officer could act on his own whim and out of malice or spite. As Gubbay CJ observed: individual police officers will have different reasons. Some may tend to stop the young, others the elderly; some only men, others women; some the poorly attired, others those with the outward appearance of affluence; some those who are jacketless, others those without handbags; and so on. Racial considerations may even be a factor. If it is unnecessary that the stoppage be based upon reasonable suspicion that the person is not carrying an identity document and that all the police officer need say is 'I stopped him because I was seeking persons without identity documents', how many innocent persons will be so accosted to apprehend one without an identity document? 81 His lordship went on to conclude that simply because an officer was acting within the law did not mean he is not doing so arbitrarily. There was 'no essential link between the two concepts'. 82 A random stop under the power no matter how brief and for whatever purpose was a detention, and thus an infringement of the right of freedom of movement. 83 And as this infringement could not be justified on the grounds of public order the section was unconstitutional. However,

The European Commission of Human Rights (hereafter the Commission) in B v
France 10179/82 13/05/87 DR 52 at 120 did consider the compatibility of the French law that required a person to produce proof of identity in limited circumstances and thereafter to detain a person at a police station for up to six hours while their identity was confirmed. The Commission held (at 126) that this was compatible with Art. 5(1)(b) representing a 'fair balance' between the right to liberty and an obligation prescribed by law. The Commission declined to decide whether there had been a deprivation of liberty within the terms of Art. 5 on the basis that the law covered the situation that the applicant complained of. while freedom of movement is not a Convention right, it is nonetheless a right at common law. 84 Yet there was no consideration of this by their lordships.
Be that as it may, the difference between liberty and freedom of movement is one of degree. It is arguable that the House of Lords was wrong to conclude that the power in ss. 44 and 45 does not affect the right to liberty. 85 For instance, Lord Bingham concluded that an individual subject to the powers in ss. 44 and 45 would be 'detained in the sense of kept from proceeding or kept waiting . . . [but] there [would be] no deprivation of liberty'. 86 This is quite simply wrong. Article 5 is a right with a considerable pedigree. The possibility of arbitrary interference with the liberty of the subject has long been a central concern of English common law. Blackstone, for instance, observed that: the law of England regards, asserts, and preserves the personal liberty of individuals. This personal liberty consists in the power of locomotion, of changing situation, or removing one's person to whatsoever place one's own inclination may direct; without imprisonment or restraint, unless by due course of law . . . [The] confinement of the person, by secretly hurrying him to gaol, where his sufferings are unknown or forgotten is a dangerous engine of arbitrary government . . . [Moreover] the confinement of the person, in any way, is an imprisonment. So that the keeping a man against his will in a private house, putting him in the stocks, arresting or forcibly detaining him in the street, is an imprisonment. 87 For Blackstone the right to liberty and the right to personal security were absolute or primary rights necessary for freedom. 88 Similarly Dicey concluded that 'the right to personal liberty as understood in England means in substance a person's right not to be subjected to imprisonment, arrest, or other physical coercion in any manner that does not admit of legal justification'. 89 This English heritage has clearly been transposed into Article 5 for, as the European Court of Human Rights has made consistently clear, that Article is intended to protect individuals from arbitrary deprivations of their physical liberty. 90 Article 5 of the ECHR is a guarantee of 'individual liberty in its classic sense, that is to say the physical liberty of the person . . . its aim is to ensure that no one should be dispossessed of this liberty in an arbitrary fashion'. 91 The 'protection against arbitrary arrest and detention', contained in Article 5, concluded that the European Commission of Human Rights 'was one of the principal considerations of the drafters of this treaty'. 92 That concern created a strong right that is one of the 'first rank' rights that protect the individual in a democratic society. 93 Consequently, Article 5 is subject only to a set of very narrow and specific limits. In essence Article 5 is the classic articulation of the protection of physical liberty through a bodyguard of legality. Furthermore, as the text makes clear, Article 5 extends to the protection of both the liberty and security of the individual. While the security interests of the individual have received little attention from the European Court of Human Rights over the years, nevertheless what case law exists does allow two important conclusions to be drawn. 94 First, the type of 'security' interests contemplated by Article 5(1) are those which are related to the physical liberty interests of the individual. Second, the guarantee of 'security' is in essence a protection against the arbitrary interference with the physical liberty guaranteed by Article 5. 95 Moreover, as the text makes clear, the framers of the ECHR clearly intended that Article 5 apply to situations that were not arrests because, although an arrest includes detention, detention does not necessarily include arrest. 96 There are a range of situations where agents of the state conduct coercive procedures, which are not arrest, but which involve the restraint or control of an individual, 97 preventing people going where they please. Thus not every deprivation of liberty is an arrest. 98 Detention includes an element of state compulsion, which might include but is not limited to arrest, but which nevertheless interferes with the physical liberty and security of the individual. At one end of the spectrum detention might include forcible admission to a hospital under the Mental Health Act, whereas at the other end it could include short-lived interferences such as compulsory blood tests 99 or stops and searches. However, on the basis of the decision in Gillan, one might conclude that Article 5 applied only to arrests and not simply detentions in the context of criminal justice. 100 Yet in the context of a stop under the TA a refusal to submit to a police officer's order to submit to a search constitutes a criminal offence, as does wilfully obstructing such a search. 101 Both are punishable by imprisonment. 102 As Lord Bingham observed in Gillan, a member of the public is under a duty not to obstruct a police officer who is conducting a search within the terms of the Act. 103 There can be little doubt that the right to liberty 'is implicated by laws which impose imprisonment as a penalty for their non-observance'. 104  with liberty'. 108 Thus when faced with a request, a demand or direction from an agent of the state, does a person reasonably regard himself as free to refuse? 109 Criminal liability for refusing to comply is in effect compulsion and thus engages the right to liberty: 110 'any criminal liability for failure to comply with a demand or direction of a police officer must be sufficient to make compliance involuntary'. 111 As was noted above, Lord Bingham concluded that a person stopped under the authority of ss. 44 and 45 was not detained but merely kept waiting. On reflection this view cannot stand. To begin with, the fact that the stop was likely to be short-lived is a separate issue from whether or not the detention is arbitrary. Whether or not the stop was arbitrary should have warranted separate consideration under Article 5. Furthermore, a person who is required to stop his car by a red traffic light is kept waiting. After a short delay he is allowed on his way. No doubt some might argue that this too is an interference with the right to liberty. After all, an individual who ignores a red light and drives through risks not only an accident but also a conviction under the road traffic laws. However, such regulation differs in two important aspects from the power in the TA. First, under the TA there is a direct assumption of the control of a specific individual rather than the universal regulation of a class of people. Second, the TA powers are a direct gateway to the criminal justice process. A power to stop and search an individual is intended, in part, to discover evidence of criminality. As with Mr Funke, albeit in a slightly different context, individuals stopped under the TA power enjoy a Hobsonian choice. 112 Either they submit to the search and its attendant risks or they resist and face the possibility of a criminal conviction. Of course it is most unlikely that the average citizen is aware of their exact legal position and it is likely that, when faced with such a demand from a police officer, he or she will comply rather than risk assault or arrest for obstruction. Psychological compulsion in such circumstances can make the restraint of liberty involuntary and thus amount to a deprivation. 113 Police stops are inherently coercive, both legally and psychologically. 114 It is, of course, important that in seeking to interpret the right purposively we do not overshoot that purpose through an overgenerous interpretation. Clearly, an overgenerous interpretation of Article 5 to cover a large number of temporary and transitory interferences with the right to liberty might hamper the effectiveness of law enforcement. Therefore in the absence of direct coercion to the 108  individual, Article 5 will clearly not apply. Thus, for example, when you stop at the traffic lights Article 5 will not be engaged. But at the same time we must not lose sight of the fact that a narrow interpretation would render the protection afforded by Article 5 'theoretical and illusory'. 115 Moreover, as one of the principal purposes behind Article 5 is to prevent arbitrary deprivations of liberty, setting a high threshold for the engagement of Article 5 effectively removes from scrutiny areas of low-level discretionary powers that have historically been abused.
ii. Privacy Leaving Article 5 to one side, with respect to the Article 8 right to privacy a more expansive interpretation was also possible for, as Feldman correctly observes, 'a personal search in public is a particularly embarrassing and upsetting experience . . . it is also a major interference with people's right to privacy'. 116 The European Court of Human Rights has repeatedly held that the right to privacy is a broad concept 'not susceptible to exhaustive definition', depending in large part on a determination of what reasonable expectation of privacy a person enjoys in a particular context. The privacy interests protected do not necessarily have to be the core ones envisaged by Article 8 for that guarantee to be engaged. Thus in certain circumstances a person will enjoy a right to privacy in public. 117 Here American law offers an interesting perspective. Textually the Fourth Amendment is, of course, very different from the provisions of the ECHR: 'The right of the people to be secure in their persons . . . against unreasonable searches and seizures, shall not be violated'. Thus American jurisprudence might seem to be of limited utility. However, the interests that it protects are broadly the same as those protected by the ECHR in this context. Just as Articles 5 and 8 protect the liberty/security and privacy interests of the individual so does the Fourth Amendment. 118 The Fourth Amendment not only protects individuals from unreasonable searches which interfere with their legitimate privacy interests but also their liberty/security interests. 119 For instance the US Supreme Court has long held that 'no right is held more sacred or is more carefully guarded by the common law than the right of every individual to the possession and control of his own person free from all restraint or interference of others'. 120 Thus in Terry v Ohio the US Supreme Court held that the right to security guaranteed by the 115  Fourth Amendment 'belongs as much to the citizen on the streets of our cities as to the home owner closeted in his study to dispose of his secret affairs'. 121 When a person is accosted by a police officer and prevented from walking away then for the purposes of the Fourth Amendment he has been seized. 122 The US Supreme Court has emphatically rejected a de minimis approach whereby such stops were not to be considered arrests for the purposes of the Constitution. As the Court observed, such stops and searches when conducted in the street are not simply a petty indignity, being 'a serious intrusion on the sanctity of the person which may inflict great indignity and arouse strong resentment, and it is not to be undertaken lightly'. 123 Furthermore, the US Supreme Court has held that bags and luggage enjoy constitutional protection under the Fourth Amendment. Indeed, bags enjoy a greater expectation of privacy than cars for, as the Supreme Court observed, 'luggage is intended as a repository of personal effects'. 124 It therefore seems reasonable that a person enjoys a reasonable expectation of privacy for materials carried in a bag, which might range from the mundane to the highly intimate.
Privacy rights under Article 8 are also implicated by the recording and storage of data relating to the exercise of the powers under the TA. Every such stop and search has to be recorded by law. In conducting the stop and search an officer will fill out Form 5090. On that form he is required to note the ethnic group of the detainee, and he must also ask for that person's name, address and date of birth. 125 Copies of Form 5090 are subsequently fed into the 'Stops Database'. 126 No doubt the 'Stops Database' is invaluable in assisting senior officers to comply with their duty to produce reports on the use of stop and search powers. 127 However, this form of systematic recording and storage of personal information relating to identity 128 has been held by the European Court of Human Rights to infringe the right contained in Article 8(1), 129  Article 8 is, of course, a qualified right, and a breach of Article 8 would only follow if the TA could not be saved under Article 8(2). For a measure to be upheld under Article 8(2) the state must show that the impugned measure has a legitimate aim, is necessary in a democratic society and is prescribed by law. 131 In the shadow of recent terrorist attacks no one could seriously argue that the TA does not pursue a legitimate aim. The provisions of the Act are clearly justified as necessary for the prevention of crime or to protect the national security interests of the United Kingdom. That is not in doubt. However, Article 8(2) does not grant a state carte blanche to do whatever it considers necessary. In other words, the powers must be proportionate. However, where the powers wielded by the state are exceptional in nature then a higher standard of scrutiny is required. In Klass v FRG, 132 for example, the applicants had their communications intercepted and monitored under German law. They complained that this constituted an interference with the right to privacy guaranteed by Article 8. The European Court of Human Rights did not agree, holding that the interference was justified. Nevertheless, in dismissing the application the European Court of Human Rights made a number of important observations. To begin with, the Court noted that in the age of increasingly sophisticated terrorism a state must be able to take measures to counter such threats effectively. 133 Thus exceptional powers may, at times, be justified. Nevertheless, the state does not enjoy an unlimited discretion in such matters. Critically states must not rely on pleas of necessity in the context of counterterrorism to adopt measures that undermine democracy rather than protect it. Such security measures are only tolerable in a democracy so far as is strictly necessary. 134 Exceptional measures must therefore be accompanied by effective safeguards to ensure that they are not abused. 135 Simply because the context is challenging this does not license the state to adopt permissive discretionary powers in the absence of effective safeguards.

iii. Freedom of Expression
Moreover, in practice the powers in the TA can act as a form of indirect prior restraint. 136 The immediate purpose of the powers is the detection and prevention of terrorism. However, the powers can also be used for a secondary purpose: that is, restraining speech, as the Chapter of Magna Carta, which is pointed against arbitrary power, was violated'. These cases are interesting not simply because they warn against unbridled discretionary police power, but also because at their core they concern free speech. Although primarily concerned with the legality of searches, the government would have been unable to bring prosecutions for seditious libel without such documentary evidence. Thus, as Stuntz contends, these cases were 'classic First Amendment cases in a system with no First Amendment and no vehicle for direct substantive judicial review'. 143 In this sense the TA power is thus a good example of the unintended effect of emergency legislation. As Waldron notes, such emergency powers are often used for other forms of state regulation unrelated to their original counterterrorist purpose. 144 Yet in Gillan the possibility that the powers in the TA might be used to restrict free speech were dismissed as remote. 145

V. The Principle of Legality: the Problem of Permissive Powers
The rule of law is an inherent part of the Convention scheme for the protection of rights and freedoms. For legislative acts to comply with the requirements imposed by the rule of law, a law must provide a measure of protection against arbitrary interferences by public authorities with Convention rights and freedoms. 146 Nowhere is this requirement more important than Article 5. The principle of legality is an essential part of the architecture of Article 5. Article 5 states that 'no one shall be deprived of his liberty save in the following cases and in accordance with a procedure prescribed by law . . .
[for] the lawful arrest or detention of a person . . . in order to secure the fulfilment of any obligation prescribed by law'. The House of Lords concluded that because the obligation to stop was imposed by virtue of the TA this complied with the requirement that the procedure be 'prescribed by law'. In a procedural sense their lordships were quite correct. However, the concept of legality contained in the Convention requires more than simply legal authority to interfere with the liberty of the individual. 147 Such a narrow conception of legality would do little to prevent the arbitrariness that Article 5 is intended to prevent. Thus the European Court of Human Rights insists that legality has both procedural and substantive aspects. The Convention requires that the law be sufficiently accessible and precise in order that arbitrary applications can be avoided. 148  requirement. In the context of stop and search powers the usual safeguard has traditionally been the requirement of a 'reasonable suspicion' on the part of the police officer. 150 Such a criterion, in theory at least, not only limits and controls the employment of intrusive powers to situations where society's interests in the detection of crime outweigh those of the individual, but also prevents abuse by providing an objective standard against which an officer's conduct can ex post facto be measured. The TA lacks this criterion. 151 The power in the TA is limited only by Parliament's command that it be used 'only for the purpose of searching for articles of a kind which could be used in connection with terrorism'. 152 The potential for abuse is thus largely hidden because the law delegates to the officer a power with no adequate basis of control. In the absence of an effective controlling criterion the law makes Convention rights dependent on the exemplary behaviour of police officers.
Before the Court of Appeal and the House of Lords counsel for the applicants in Gillan had sought to argue that the power in the TA was not 'prescribed by law' in that it was imprecise and vague. This argument was bound to fail for, as Lord Bingham observed, the TA 'defines and limits the powers with considerable precision'. 153 However, both counsel and the Bench missed the essential point with respect to legality. The power is not vague; rather it is permissive. In this context the key question is, 'are there sufficient safeguards in the TA to protect Convention rights?'. The permissiveness of a provision in these circumstances means that rather than preventing abuse, it actually facilitates it. The power may be used for purposes beyond those licensed by Parliament without any effective legal oversight or redress. The power authorizes arbitrary interferences with Convention rights that are in effect not 'prescribed by law'. In Vasileva v Denmark the European Court of Human Rights held that a power to authorize detentions should not be used in a punitive manner outside its intended purpose. 154 Similarly, in the South African case S v Z 155 the applicants were a number of juvenile offenders who had been sentenced to reform school but because of systemic failures had been kept in penal custody instead. The Eastern Cape High Court concluded that deprivation of their liberty was arbitrary and thus, because it fell outside the purpose for which it was imposed, inconsistent with the constitutional guarantee to liberty. 156 Indeed, with permissive powers the danger that they will be abused in this manner is even greater, as the case of Walter Wolfgang illustrates. 157 The detention of an individual for heckling at a private function where there was no evidence of a terrorist connection was clearly punitive and in no way connected with the detection or prevention of terrorist crime. When a power can be used for an inappropriate purpose it can be used arbitrarily.
Indeed, rather than vindicating the requirement inherent in the Convention that restrictions on rights be in harmony with the principle of legality, a permissive power subverts that principle. Such powers allow for pretextual stops. In other words, an individual is stopped on the pretext of counterterrorist reasons but in reality the stop is for other purposes. Such a power, in the absence of effective safeguards, carries with it an ever-attendant danger that it will be exercised arbitrarily. Although the term 'arbitrary' has been frequently cited it has not received much judicial clarification either in the European Court of Human Rights or courts applying the HRA. However, what constitutes arbitrariness has received greater consideration under the International Covenant on Civil and Political Rights (ICCPR) and domestic bills of rights based on its provisions. 158 This jurisprudence is most instructive. At the outset the term 'arbitrary' was included in order to cover acts that may be lawful in the sense that they were consistent with the law but which were nevertheless unjust. 159 Indeed, the term was deliberately included to ensure that unjust and not simply unlawful acts were covered. 160 Subsequently, the Human Rights Committee (HRC) concluded that the term 'arbitrary' 'must be interpreted more broadly to include elements of inappropriateness, injustice and a lack of predictability'. 161 To that list, as Nowak suggests, might be added capriciousness and disproportionality. 162 The much cited decision of the HRC in Hugo van Alphen v The Netherlands provides a good illustration of the principle in action. In Alphen the author was arrested on suspicion of forgery. He was remanded in custody for nine weeks. The author alleged that his 157  arrest and his detention were intended solely to place pressure upon him to waive his absolute right to legal professional privilege and to make incriminating statements. The HRC held that the ICCPR required more than the author's treatment being authorized by law. His treatment needed to be reasonable in the circumstances. Any remand to custody had to be necessary, which in the author's case it was not. In fact it was arbitrary. Similarly, in A v Australia the author of the complaint had been detained for a period of four years in a detention camp for unlawful migrants. In determining that the detention was arbitrary an examination of both the necessity and proportionality of the detention was necessary.
Before national courts the approach of the HRC was approved by the Privy Council in Fok Lai Ying v Governor in Council, 163 although without much discussion, and subsequently in more depth by the New Zealand Court of Appeal in R v Goodwin. 164 In Goodwin the appellant had been convicted of the manslaughter of his infant daughter. During questioning, but while not under arrest, the appellant asked if he could leave the police station. He was told that he was not free to go. In a subsequent interview the appellant confessed. The appeal turned on the admissibility of the confession and thus whether the appellant had in fact been arrested or detained. The statement that the appellant was not free to go induced a belief that he had to remain. He was thus detained for the purposes of s. 22 of the New Zealand Bill of Rights Act 1990. With respect to the question of arbitrary detention Richardson J concluded that 'whether an arrest or detention is arbitrary does not turn on its lawfulness but on the nature and extent of any departure from the substantive and procedural standards involved . . . an arrest or detention is arbitrary if it is capricious, unreasoned, without reasonable cause, if it is made without reference to an adequate determining principle or without following proper procedures'. 165 The principle of legality requires there to be a criterion or criteria against which executive action must be exercised and can thereafter be independently and impartially reviewed. If a discretionary power is unfettered then there is a clear potential for abuse. In these circumstances it is not enough simply to say that such a permissive power is justified because it assists in the detection and prevention of crime. Those goals can be pursued in a way that does not require a power that represents a dangerous exception to the supremacy of law. 166 Similarly, the compatibility of a permissive power with Article 8(2) was considered by the European Court of Human Rights in Herczegfalvy v Austria. 167 The application concerned, in part, the interception and control of Herczegfalvy's correspondence by his legal guardian 163  and the psychiatric hospital where he was detained. Before the European Court of Human Rights the applicant argued that the interception was not authorized by law and could not therefore be justified under Article 8(2). In reply the Austrian state argued that the authority for such interceptions could be found in the Austrian Civil Code and the Hospitals Act. Section 51 of the Hospitals Act stated that a detained patient 'may be subject to restrictions . . . as to their contacts with the outside world'. 168 However, the law contained no indication of any criteria that should be employed when exercising the power, and any mechanism by which the decision could be reviewed ex post facto. 169 The European Court of Human Rights held that this lack of precision was incompatible with the requirements of Article 8(2). The Court concluded that 'if a law confers a discretion on a public authority it must indicate the scope of the discretion'. 170 Moreover, in the absence 'of any details at all as to the kind of restrictions permitted or their purpose, duration and extent or the arrangements of their review . . . the provisions do not offer the minimum degree of protection against arbitrariness required by the rule of law in a democratic society'. 171 The principle of legality thus has two important aspects, one preventative and one remedial. First, a discretion that has the potential to interfere with Convention rights should not be so permissive that it allows arbitrary interferences with rights to shelter behind it. Generally, a law that 'confers a discretion must indicate the scope of that discretion'. 172 This is not the 'impractical and unworkable prescription' that Lord Steyn thought it was. 173 Where the legislature 'has given a plenary discretion to do whatever seems best in a wide set of circumstances [then] there is no "limit prescribed by law"', 174 for where there are 'no standards governing the exercise of the discretion granted by the [law] the scheme permits and encourages an arbitrary and discriminatory enforcement of the law'. 175 Permissive powers invite manipulation. They can be used for purposes for which they were not granted, as ss. 44 and 45 of the TA demonstrate. While the legislature might mandate their use for a specific purpose, operationally because the power is expressed in permissive terms it can be used for a range of other purposes. Such permissive powers facilitate the sotto voce infringements of Convention rights in a manner that defeats the very purpose of the Convention. For example, ss. 44  misused for public order purposes. 176 Moreover, such powers deny equality before the law. As they have no standard they can in effect give free rein to the personal stereotypes or prejudices of the individual officer in their application. 177 Thus, a discretion must be controlled by some objective criterion or criteria. As the US Supreme Court noted in Brown v Texas, when 'a stop is not based on objective criteria the risk of arbitrary and abusive police practices exceeds tolerable limits'. 178 There needs to be a controlling criterion that guarantees 'that an individual's reasonable expectation to privacy is not subject to arbitrary invasions solely at the unfettered discretion of the officers in the field'. 179 As White J observed, the requirement of reasonableness as a control on law enforcement searches was required by the Constitution in order that the privacy and security interests of the individual were not arbitrarily invaded. 180 The requirement of reasonableness represents a means of reconciling the competing interests of the individual's privacy and security from arbitrary interferences on the one hand and the government's interests in the detection and prevention of crime on the other. 181 In cases where the legislature has departed from the reasonableness standard, then there must be some mechanism containing an objective standard that ensures that the individual's privacy and security interests are not left at the whim of the official exercising a standardless power. As the French Conseil constitutionnel observed, 'the performance of checks entrusted by the law to the criminal investigation police must be based on purely objective criteria, to the exclusion of all forms of discrimination prohibited by the Constitution'. 182 Similarly, in Re Ontario Film and Video Appreciation Society the Ontario Board of Censors enjoyed a broad statutory power to censor films. There were, however, no standards constraining the exercise of the censorship discretion under the Theatres Act or any of the regulations made under the authority of the Act. The Ontario Supreme Court held that: The Charter requires reasonable limits that are prescribed by law; it is not enough to authorize a board to censor or prohibit the exhibition of any film of which it disapproves. That kind of authority is not legal, for it depends on the discretion of an administrative tribunal. However dedicated, competent and well-meaning the board may be, that kind of regulation cannot be considered as 'law'. It is accepted that law cannot be vague, undefined, and totally discretionary; it must be ascertainable and understandable. Any limits placed on the freedom of expression cannot be left to the whim of an official; such limits must be articulated with some precision or they cannot be considered to be law. 183 This issue also arose before the South African Constitutional Court in Dawood v Minister of Immigration. 184 At issue in Dawood was the validity of s. 29(9) of the South African Aliens Control Act. The Act granted immigration officials a discretionary power to exclude foreign spouses from the country while their applications for immigration permits were considered, by refusing to renew or grant a temporary permit in the meantime. The exercise of the discretionary power had the potential to interfere with the constitutional right to dignity that underpinned marriage. The issue for the Constitutional Court was whether an unconstrained discretion could meet the requirement that limits on constitutional rights be 'prescribed by law'. The Court held that a broad discretionary power could not constitute a 'law' for the purposes of s. 36 of the Constitution, the limitation clause. 'It is an important principle of the rule of law that rules be stated in a clear and accessible manner', O'Regan J concluded, for 'if broad discretionary powers contain no express constraints, those who are affected by the exercise of the broad discretionary powers will not know what is relevant to the exercise of those powers or in what circumstances they are entitled to relief'. 185 It is not enough simply to say that the power can be exercised in a manner compatible with the Constitution or read down into conformity. It is for the legislature to provide guidance on when it is justifiable to limit rights under a discretionary power either in the Act or via delegated legislation. 186 Without criteria the exercise of the power will introduce an element of arbitrariness that is inconsistent with the constitutional protection of rights. 187 Second, the presence of an objective standard enables effective ex post facto review and scrutiny that will ensure the lawfulness of any interference with Convention rights and provide a remedy if necessary. Here a court must not be hampered in testing the necessity of limits placed on Convention rights by rules such as the level of scrutiny or the absence of an objective legal standard that provides sufficient means to measure the legality of executive action. A standardless power is a good example of the latter. Decisions made under its authority do not need to be justified under objective criteria. Instead they rest on subjective assessment made by the exercising officer which ex post facto a court is unable to scrutinize effectively in terms of the Convention simply because no standard is required by 183 (1983)  the statute. There will inevitably be cases where an individual's Convention rights are infringed and yet he will be unable to obtain an effective remedy. For instance in Smirnov v Russia 188 the applicant complained about a search of his apartment. The applicant was representing a number of individuals in proceedings although he was not under suspicion himself. The search warrant issued by the Public Prosecutor was broadly drafted, granting the police an unrestricted discretion in its execution. 189 The police were free to seize what they liked, and indeed did so. The European Court of Human Rights held that not only was the search not carried out 'with relevant and sufficient grounds' but there was also an absence of effective safeguards to constrain the exercise of the discretion and to prevent abuse. 190

VI. Legality and Discrimination
But perhaps the strongest argument that the safeguards are insufficient is the possibility that the standardless power might easily be employed arbitrarily on the basis of racial prejudice. Any unbridled discretion that has the potential to be employed in an arbitrary manner has implications for the equal application of the law. 191 Here the Home Office's statistics do not make for very comfortable reading. In a typical year before the 7/7 attacks, 192 an average of 8,120 stops and searches were conducted by police forces in England and Wales. 193 In the Metropolitan area, the most ethnically diverse part of the United Kingdom, there were 5,245 stops and searches under ss. 44 and 45, leading to 57 arrests of which 2 were related to terrorism. 194 Both individuals arrested for terrorism were white. However, over a quarter of those stopped and searched were not white. 195  This pattern is not surprising. Code A of PACE states that 'there may be circumstances . . . where it is appropriate for officers to take account of a person's ethnic origin in selecting persons to be stopped in response to a specific 200 terrorist threat (for example some international terrorist groups are associated with particular ethnic identities)'. 201 In other words, ss. 44 and 45 of the TA when read with the PACE Code allow an officer to engage in racial profiling. 202 There are, as Swiney suggests, 203 two definitions of racial profiling. Under the narrow one racial profiling occurs when race is used as the sole criterion for determining whom to stop and search. This is the definition favoured by the House of Lords in Gillan. The second, broader definition occurs when race or ethnicity is one factor that is taken into account amongst others. Indeed, race can influence decisions subconsciously. This broader definition is favoured by the US Department of Justice. 204 The over-use of police powers against minority communities is, of course, nothing new and since the Lawrence Inquiry purportedly is a matter of great concern. 205 Such communities have been subject to arbitrary powers of stop and search before. 206 In many ways ss. 44 and 45 are similar to the 'sus' laws that were repealed when empirical evidence demonstrated their over-use against certain minority communities. 207 Where a stop is based solely on race it will be discriminatory. 208 As McIntyre J observed, 'distinctions based on personal characteristics attributed to an individual solely on the basis of association with a group will rarely escape the charge of discrimination'. 209 In United States v Brignoni-Ponce the US Supreme Court held that a stop made by a border patrol solely on the basis of racial appearance was unconstitutional. While Mexican appearance could be a relevant factor it did not on its own justify stopping individuals. 210 Racial discrimination, as the European Court of Human Rights noted in Timishev v Russia, is 'a particularly invidious kind of discrimination and in light of its perilous consequences requires from the authorities special vigilance and vigorous reaction'. 211 The Council of Europe urges contracting states to ensure that the principle of nondiscrimination is applied in the context of policing and, when dealing with emergencies, the Council urges respect for the rule of law, and an absence of discrimination coupled with adequate safeguards. 212 'There is no more effective practical guaranty against arbitrary and unreasonable government', concluded Jackson J, 'than to require that the principles of law which officials would impose upon a minority must be imposed generally . . . conversely, nothing opens the door to arbitrary action so effectively as to allow those officials to pick and choose only a few to whom they will apply legislation, and thus to escape the political retribution that might be visited upon them if larger numbers were affected'. 213 We might therefore have expected a robust judicial response in Gillan to what appears to be a pattern of discriminatory stops and searches. 214 Article 14 is after all an important weapon in preventing the development of patterns of discrimination within society. Article 14 is intended to shield individuals from objectification based on stereotypes, particularly those who have suffered such stereotyping in the past. 215 Profiling has a pernicious effect in practice. It encourages people to deal with persons from different races as 'black', 'white', 'Asian' and so on, rather than individuals worthy of equal concern and respect. 216 However, because of the standardless sweep of the power in ss. 44 and 45 there is a danger that it might act as a mechanism for giving effect to crude stereotyping. Those who possess the crude proxy characteristic, namely certain racial origins, are taken to possess the material characteristic. The proxy is based on a crude sociological generalization for which there is little evidence, and creates an atmosphere where crude pejorative stereotypes will be employed either consciously or unconsciously. 217 Of course, discrimination on the basis of race cannot be justified on utilitarian grounds. 218 Everyone is entitled to be treated as an individual and not as a statistical unit. 219 On the facts the Roma Rights case would be good authority for arguing that the current practice of stop and search under ss. 44 and 45 is directly discriminatory, being 'inherently and systematically discriminatory on racial grounds'. 220 However, in Gillan Lord Brown was able to distinguish the Roma Rights case on the grounds that an individual's ethnic profile was not the sole factor in determining whether to stop and search under ss. 44 and 45. Racial origins were simply part of an overall profile. 221 The police would need to have regard to other factors when deciding to stop and search under the authority of ss. 44 and 45. In other words, there was no intentional discriminatory use of the power in a racially motivated manner. Thus the House of Lords saw little difficulty in employing race as in effect a key element of a terrorist profile, although some members of the Appellate Committee were clearly troubled by the manifest discrimination. Lord Brown concluded that it was 'inevitable that so long as the principal threat [ s. 44 TA will be of Asian appearance'. 222 To many, no doubt, Lord Brown's statement appears to be an incontrovertible fact. For instance, all but one of the 7/7 bombers were British Asians. However, simply because one set of terrorists fits the profile it does not mean that future ones will. The key question here, surely, is are we simply succumbing to the temptation to fall into pejorative stereotyping? Indeed, how strong is the correlation between the proxy trait, namely race, and the material one, that is a propensity to commit terrorist crime? It might be argued that as profiling potential offenders is a well-established law enforcement practice it should be used in this context; after all, profiling is seen by many as a reliable and effective investigatory tool. 223 But the key question in this context is how rational is profiling? By focusing on race we are ignoring an important fact, namely that Islamic terrorism is driven by a fundamentalist interpretation of a religion. This form of terror is not linked exclusively to a particular ethnicity in the same way, for example, as Irish terrorism was. 224 The traditional paradigm of viewing terrorism as connected to territorial disputes is not appropriate post 9/11. 225 Moreover, the al-Qaeda network is simply not a terrorist group like the IRA. For instance, while some of Bin Laden's aims might have a territorial element to them, the Islamist groups that form the al-Qaeda network are not committed to the liberation of an ethnic homeland from an occupier. The motivation is of a different order, springing from a divine injunction to defend one's religion from its enemies. Islamic terrorists will inevitably be a heterogeneous group of nationalities and ethnicities, as the patchy records of those trained in al-Qaeda's Afghan camps reveal. In contrast to Irish republican terrorism, for instance, where the majority of operatives were Irish or of Irish descent, Islamist terrorists could literally be anyone. Ethnic targeting will therefore only ever be limited in effect. It is not uncommon for white, mixed race and black individuals to convert to Islam. Some will inevitably become Salafists or Islamists. 226  There may of course be a coincidence between the two, as in Kashmir. But Islamic terrorists do not subscribe to notions of states and sovereignty for the simple reason that they believe that the world belongs not to men but God. 226 Broadly speaking Salafists are Muslims who believe that society should be ordered as it was in the time of Mohammed, in accordance with the Koran and hadith (the record of the words and deeds of Mohammed). Islamists are a similar group in that they too hope to establish a purely Islamic society. However, Arab-Afghans but recruits from the West including converts. 227 Such a convert was of course Richard Reid, the so-called 'shoe bomber' who attempted to commit mass murder by blowing up a trans-Atlantic flight with an explosive device concealed in his shoes. Reid is of mixed race, being of British and Jamaican descent. 228 Similarly in November 2005 a white, former Christian, middle-class, Belgian national, Muriel Degauque, 229 took part in a suicide attack against US forces in Baghdad, killing six others in the process. In fact there is growing evidence that the suicide bombers who attack cities such as London in the future will not be Asian in appearance but white. The presence in the alleged August 2006 conspirators of white converts should not be a surprise to anyone. Faced with an organized offensive against it al-Qaeda has simply adapted its tactics. In 2005 The Global Media Front, a recognized mouthpiece of al-Qaeda, published a document on the Internet indicating that al-Qaeda intended to use white converts to undertake attacks. Al-Qaeda refers to such people as 'Rakan Bin Williams'. The site went on to place the use of such converts in the context of al-Qaeda's strategy of surprise and adaptation: 'They restricted the Arab and were attacked by those from Pakistan . . . you will not be able to find any solution because our next soldier cannot be put under surveillance or restricted in his movements or arrested'. 230 235 But then stop and search in general has rarely been an effective means of apprehending offenders. 236 Moreover, the irrationality of using race as a proxy trait is a good example of probability neglect. Not surprisingly we are gripped by a degree of fear when we think about terrorism. In such circumstances we naturally tend to focus our thoughts on the adverse outcomes while ignoring the likelihood of their occurrence. Such neglect of probability 'can produce significant problems for law and regulation'. 237 The majority of us have no idea how probable it is that we will be injured or killed in a terrorist incident. In the absence of this information we rely on our own perceptions. How easy is it for us to recall or visualize a salient example of the event in question? Images of terrorism are readily available. One can quite easily recall the images of the aircraft hitting the World Trade Centre in New York or the exploded bus in Tavistock Square in London. Terrorists of course realize this for, as Sunstein observes, such people (whether they know it or not) 'show a good understanding of the availability heuristic simply because a highly publicised terrorist attack 238 is likely to create more fear than base statistics warrant'. 239 Although we are far more likely to die in a car crash or from chronic illness we tend to perceive terrorism as a more pressing threat to ourselves and society, and in light of this it is easier to justify extraordinary measures to deal with it. We allow our strong emotions, principally fear, to drive out qualitative judgments, 240 and spend a disproportionate amount of time planning for worst-case scenarios. Thus we tend to focus on high risk information, believing it to be more compelling and accurate, and plan our response accordingly. For instance in the wake of 9/11 in the United Kingdom the government conducted exercises based on scenarios involving the use of biological and nuclear weapons. 241 In September 2003 the government undertook Exercise Osiris II at London's Bank Underground station simulating a chemical attack. However, when the much anticipated attack came on 7 July 2005 it was conducted by a small cadre of bedsit fanatics wielding devices manufactured from household chemicals. 242 In such a climate it is inevitable that representative institutions will neglect probability too. In fact 'a good deal of legislation and regulation can be explained partly by reference to probability neglect when emotions are running high'. 243 The legislative response to terrorism carries with it a severe risk of probability neglect. 244 Sensationalist saturation media coverage of terrorism hardly helps, triggering the availability heuristic. 245 Overreaction is inevitable, with broad sweeping and largely ineffective emergency powers being the result. Sections 44 and 45 of the TA are a good example.
Furthermore, the discriminatory effect of the operation of ss. 44 and 45 of the TA is also counterproductive. 246 If society as a whole is to reduce the risk of Islamist terrorism then moderate Muslims need to be supported. However, the over-use of emergency powers creates a danger that Muslims will be stereotyped and alienated. In the end, faced with what is perceived to be a partial and aggressive use of police powers, there is a danger that the Muslim community will feel stigmatized, and become increasingly angry and alienated. In such circumstances it is probable that we will lose the support of the very community we need to tackle this problem effectively. 247 Individuals who see the police and criminal justice system as unfair are unlikely to assist the police by informing on suspicious activity or indeed testifying. 248 Arguably, in the long term such crude powers can hinder law enforcement by undermining the legitimacy of the police. Disillusioned young men are of course the very people that terrorists seek to recruit, especially at a time when the concept of Islamic Umma has attained greater potency in general, and for some the bonds of Tawheed are strengthening. 249 Moreover, in the post-Lawrence world the use of such powers in a discriminatory manner does nothing to improve relations between the police and other minority communities who are not seen directly as the source of the threat. For ordinary people police legitimacy is driven more by their perceptions of procedural fairness than by instrumentalities such as their effectiveness in fighting crime. 250 The legitimacy of the police is, in the public's mind, linked to judgments on the fairness with which the police make decisions and exercise their authority. Indeed, unequal treatment is a major source of dissatisfaction. 251 Regardless of their ethnicity 'people cooperate with the police when they view the police as legitimate'. 252 Unfairness or perceived unfairness will lead to alienation, defiance and non-cooperation. 253 Somewhat depressingly we have been here before. In 1983 a Home Office research paper concluded that 'the repeated use of stop and search powers can have an adverse effect on police/community relations to the point where any contact with the police comes to be seen as . . . a sign of official hostility'. 254 The report went on to conclude that 'without a secure base of community support the use of powers of this kind, however extensive or circumscribed, rapidly becomes burdensome and ineffective'. 255 Indeed, this has been a persistent concern of both inquiries and researchers in the intervening 25 years. 256 Moreover, the perception of a fundamentally unfair criminal justice system contributes to broader racial divisions in society. 257 When politicians talk of 'redressing the balance' the costs of such adjustments do not fall evenly across society. As Waldron has pointed out, it is often minority communities that carry the heaviest burden. 258 Of course, beyond the specific context of terrorism there is the wider question of the role of government in preventing discrimination. Arguably in a free and democratic society government has a special duty not only to ensure that discrimination on prohibited grounds is prevented but also to act as a teacher. Where the government is seen to act in a discriminatory manner it acts in a way that not only undermines its own efforts to prevent such action in private parties, it also acts inconsistently with its own laws and policies intended to eradicate discrimination. Not only are the costs of such action higher and more hidden than in the private sector, 259 discriminatory action by the government is likely to be observed by the wider population and this will inevitably sow the seeds of distrust if not fear of members of the principally targeted group. In such an atmosphere discrimination will flourish. 260 In these circumstances we need to be especially sure that emergency powers are both necessary and effective. However, proving direct racial discrimination would be difficult. It would involve an admission by an officer that he had relied on racial stereotypes in the exercise of his discretionary powers. 261 But of course, under vague discretionary powers people can be stopped at whim, and where there is no controlling criterion then no reason will need to be given. 262 Be that as it may, while the exercise of ss. 44 and 45 might not constitute direct discrimination it might nevertheless constitute indirect discrimination. In Jordan v United Kingdom the European Court of Human Rights concluded that 'where a general policy or measure has disproportionately prejudicial effects on a particular group it is not excluded that this may be considered as discrimination notwithstanding that it is not specifically aimed or directed at that group'. 263 The power in ss. 44 and 45 is facially neutral. It does not single out any particular group for differential treatment. 264 It is silent as to whom it should be applied. However, indirect discrimination is intended to deal with discrimination that hides behind neutral criteria.
Although the law in its form is non-discriminatory it is nevertheless discriminatory in its operation as the burden of its operation falls more heavily on members of discrete and insular minorities. In other words, under ss. 44 and 45 there is a differential impact on members of protected groups. Stops and searches are not intentionally discriminatory. The police intend to intercept terrorists preparing to commit crimes or to gather intelligence. However, in effect the operation of the power is to single out specific groups for special treatment. As the statistics make clear, the operation of ss. 44 and 45 is a paradigm case of indirect discrimination, for the powers when applied interfere with the Convention rights of a higher number of members of racial minorities than whites. In effect it is discriminatory. The European Court of Human Rights has cautioned against simply relying upon statistical analysis as sufficient to prove a prima facie case of discrimination. 265 However, where 'undisputed official statistics' reveal indirect discrimination then this is sufficient proof. 266 Once a prima facie case of indirect discrimination is made out by a claimant, then the burden of proof shifts to the defendant to demonstrate that there has been no discrimination, and if there has, that it can be objectively justified. 267 For ss. 44 and 45 to be objectively justified within the terms of the Convention they will need to pursue a legitimate aim in a proportionate manner. 268 However, there is an immediate problem here. There is no express legal authority for such profiling. A similar situation arose in the South African case City Council of Pretoria v Walker. 269 Here the Council differentiated between categories on the basis of race by charging, in a different manner, residents of different suburbs. The effect was to discriminate indirectly against the residents of predominantly white suburbs. The Constitutional Court held that the decision of the Council to establish such a charging regime was not authorized by law, either expressly or by necessary implication. Therefore, the Council's policy was not authorized by law and it could not seek to justify its policy. 270 Arguably, it is the case that profiling under the TA is not authorized by law and is thus unlawful.
Leaving the issue of legality to one side, few would seriously doubt that the prevention and detection of terrorism is a legitimate aim. Although the ss. 44 and 45 power was originally enacted in 1996 as an expeditious response to what is now a historic threat, it has now become an important weapon in counterterrorist policing post 9/11. There can be no doubt of the threat posed by contemporary terrorism. Given the brutal and bloody nature of terrorism the adoption of an effective approach to counterterror policing that is preventative is both highly desirable and wholly understandable.
If we accept that some form of special powers are thus essential to deal with this terrorist threat, the question then becomes, which ones are likely to be effective? In other words, is there a rational connection between the legitimate aim and the legislative means chosen to meet it? Sections 44 and 45, we are told, are an important part of the counterterrorist arsenal. For example, the Association of Chief Police Officers claims that ss. 44 and 45 are an invaluable tool because they deter and disrupt terrorists, particularly their reconnaissance of potential targets. 271 Indeed, the House of Lords was itself impressed by similar witness statements to that end. The ss. 44 and 45 powers remained necessary to disrupt terrorist activity and to gather intelligence. 272 Lord Bingham thus felt able to conclude that, in light of the statement, there was 'no evidence of any kind to contradict or undermine this testimony'. 273 In doing so Lord Bingham followed a long line of authority in English law whereby the judiciary adopt a highly deferential attitude in cases involving national security and effectively accept the ipse dixit of the executive. However, the Home Office's own statistics tell a different story. In fact a thorough examination of the statistics brings into question the issue of whether the power is effective, and thus necessary. For example, between 2001 and 2005 there were 24,781 stops and searches under the power. These stops resulted in 364 arrests for offences other than those connected with terrorism, and just 36 for those related to terrorism. Unfortunately, we do not know how many of the 36 arrests for terrorism resulted in a conviction for such offences. As has already been noted, since the 7/7 attacks the TA power has been used more frequently. Consequently the number of arrests has risen. In 2005/2006, 19,064 exercises of the TA power resulted in 59 arrests for terrorist related offences. This 0.3 per cent success rate represented an improvement on the previous combined average of 0.14 per cent. 274 Nevertheless, this success rate can hardly be said to represent a power that is so effective that it can be described as necessary. Such a low success rate is not surprising given the difficulty in combating contemporary terrorism, where the protagonists are indistinguishable from their law-abiding compatriots. The inescapable conclusion must be that the power in the TA is largely a symbolic one given its limited impact on improving security. Be that as it may, it is tempting to conclude that had there been an instance where ss. 44 and 45 have succeeded in foiling an attack it would have been referred to. In the absence of specific evidence the suspicion must be that, as with ordinary powers of stop and search under PACE, the impact of stop and search in disrupting and deterring terrorist crime is limited. 275 Instead, as with PACE, the TA powers are in large part directed at controlling disorderly and dangerous groups rather than actually detecting crime. 276 In this context race is used as a trigger for surveillance and control. Moreover, the statistics seem to indicate that these provisions are, par excellence, what Thomas calls 'comfort legislation' 277 whose benefits are primarily psychological. 278 The presence of police officers stopping and searching people in public places is no doubt to some people reassuring. But the price for providing such reassurance is the significant expansion of policing and police powers. 279 Equally, no specific evidence was presented showing how ss. 44 and 45 have yielded usable intelligence. Moreover, there is a further 'strong argument to be made that statesanctioned discriminatory treatment of vulnerable and disadvantaged groups should never be permitted even if it could be defended as rational'. 280 Not only are there more efficient ways of collecting intelligence, the power is most unlikely to disrupt or prevent terrorist attacks. The correlation between the goal of collecting intelligence and the means chosen to achieve it seems implausible in light of the statistics. 281 Finally for ss. 44 and 45 to be objectively justified, the means chosen to limit the right should do so no more than is necessary to achieve the objective. In Gillan the limitation was imposed not by the statute but by its application. As we have already seen, a vague, over-broad power has the potential to limit rights and freedoms more than is necessary because the law does not clearly demark or control the limits of the power. Under such a power limits on Convention rights will vary with every application. In such a context the key question is what, if any, safeguards have been put in place to prevent abuse? For instance, what guidance has been issued on the lawful exercise of the power? This is not merely some academic exercise. The potential for abuse of security legislation has a 'long if somewhat tawdry pedigree'. 282 Indeed, the need for safeguards is probably greater in times of crisis when the temptation is to place the exigencies of security before all else. In the context of the TA clearly Parliament intended that the power be used only in connection with activity related to terrorism. Yet the power has resulted predominantly in arrests for crimes not connected with terrorism. Thus Parliament or the courts should put in place incentives and structures to ensure that discretionary powers are not abused. In fact Parliament did not put in place any effective safeguards to prevent the abuse of this power. Rather than addressing the issue of whether and indeed when racial profiling should be used, thereby surrounding the power with proper safeguards, the PACE codes were amended. According to paragraph 2.25 of PACE Code A such action is justified because certain international terrorist groups are associated with particular ethnic groups. This presumably is what the Code means by an 'objective threat'. The Code is silent on how the threat is to be determined. Instead it is left to the various police forces as to how the assessment is to be made. The Metropolitan Police Force has issued Police Standard Operating Procedures to determine how that assessment is to be made. 283 This is surely significant, for nothing in the TA sanctions racial profiling. Yet the PACE Code contemplates treating different classes of people differently. Arguably, this issue should have been addressed directly by Parliament.
It is nonetheless important to recognize that safeguards are not limited to ensuring that the power is exercised in a manner compatible with Convention rights. Ex post facto safeguards have a role to play too. The House of Lords concluded that an action in the county court would provide an effective remedy for an abuse of the power. There are, however, a number of problems with this. First, the power is unlikely to be used against the economically favoured or powerful. It is most likely to be used against the disadvantaged. 284 Second, the abuse of ss. 44 and 45 is likely to remain largely hidden. The innocent will rarely pursue a claim. Litigation is expensive and timeconsuming. Moreover, many of those stopped simply lack faith in the legal system to vindicate their rights. 285  for concern in granting such a vague discretion. It is unlikely to be used as much against the economically favoured or powerful as against the disadvantaged. It is interesting that the home in Davis v Russell, supra, was described as lodgings, and those in Eccles v Bourque and the present case were apartments. As Professor Paul Weiler has stated, "abuses of police power will rarely affect respectable members of the middle classes", but will instead "focus on the poor and on marginal, minority groups"'.

VII. Conclusion
In the end the House of Lords found that ss. 44 and 45 were compatible with the Convention. But assuming, for the sake of argument, that the courts had identified a possible incompatibility between the Convention and ss. 44 and 45 of the TA, what might have been an appropriate remedial response? Here, for example, a court could scrutinize the power as exercised. Sections 44 and 45 of the TA authorize the lawful stop and search of an individual for the purposes of detecting terrorist related crime. Under ss. 44 and 45 a stop and search for other purposes is unlawful. Although in Gillan itself there was no question of either an unlawful search or bad faith on the part of the police, it is undoubtedly true that the power has been instrumental in far more arrests for non-terrorist related offences than for those contemplated by the TA. The power has also been used in questionable circumstances, as the Wolfgang episode rather publicly demonstrated. But given the structure of ss. 44 and 45 a court simply has no way of knowing what the motivations of an individual officer were. As we have seen, there is no objective standard which purports to control the exercise of the power in ss. 44 and 45. Be that as it may, if a stop and search under ss. 44 and 45 produces evidence of criminality other than that related to terrorism, it is still admissible notwithstanding the fact that power may have been employed unlawfully. 287 In this situation a court would no doubt follow R v Khan 288 where evidence procured through infringements of both English law and Convention rights was nevertheless admissible at trial. In England and Wales, unless there is an adverse impact on the fairness of proceedings a trial court will not normally 289 exclude unlawfully obtained evidence. 290 The rule is a strict one. In fact it might be argued that in practice the strictness of the exclusionary rule fails to provide adequate protection of Convention rights. Not only is the infrequently exercised judicial discretion to exclude evidence on the basis of unfairness a narrow one, 291 it has been rightly described as somewhat weak. 292 In these circumstances it is not unreasonable to argue that in the context of ss. 44 and 45 the exclusionary discretion is unlikely to provide an effective remedy for the infringement of Convention rights. Moreover, given that Convention rights are implicated, a new prophylactic rule is necessary to ensure that such rights are not infringed if possible. 293 After all, rights enshrined in a constitutional document deserve protection consistent with that status. 294 In this context, at least, one possible alternative to the exclusionary rule would be an inclusionary rule, similar to the one that Scots law takes with respect to unlawfully obtained evidence. Where unlawfully obtained evidence is led in Scotland there is a presumption in favour of exclusion unless the Crown can convince the court that the irregularity should be excused. 295 The rule in Lawrie v Muir allows a court to balance 'the interests of the citizen to be protected from illegal or irregular invasions of his liberties by the authorities [with the] interests of the state to secure evidence . . . necessary to enable justice to be done'. 296 In fact a new inclusionary rule would have a number of benefits. First and foremost, such a rule would uphold Convention rights by providing an effective remedy. Second, the rule would provide an incentive against the routine use of the special power, limiting its use to cases where terrorism might be involved. Such a prophylactic rule may help to prevent the ss. 44 and 45 power being unlawfully used on the pretext of a stop and search for terrorist offences. This would undoubtedly help to give effect to the intention of Parliament in expressly limiting the use of the power to terrorism cases. Third, given the difficulty of detecting unlawful exercises of the power, requiring the exercise to be justified before the Crown may lead the evidence ensures that the evidence has in fact been gathered lawfully. This may go some way in helping to maintain public confidence in the criminal justice process. Conduct that undermines the rule of law is the sort of conduct that a court should not tolerate, much less endorse. 297 And finally, an inclusionary discretion would allow a court adequately to weigh the public interest in prosecuting those suspected of grave crimes. To a large extent a balancing test answers Cardozo's conundrum. 298 Alternatively, it may be possible for the courts to read ss. 44 and 45 down by implying into the provision a requirement that an officer should have a reasonable suspicion before he exercises the power. 299 In the South African case Lawyers for Human Rights v Minister of Home Affairs 300 the constitutionality of an immigration officer's statutory discretion to detain a person he believed to be an illegal immigrant was challenged. The power in s. 34(8) of the South African Immigration Act 2002 to detain 'illegal foreigners' was not constrained by any criteria. There was for example no requirement that the officer possess a reasonable suspicion that an individual was an illegal foreigner. Thus under s. 34(8) a person who was not an illegal foreigner, but who had been declared as such by an immigration officer, could be lawfully detained and deported. The High Court declared s. 34(8) to be unconstitutional. However, on appeal the Constitutional Court reversed this decision. Yacoob J concluded that if the interpretation taken by the High Court was correct, then s. 34(8) was indeed unconstitutional. However, the section could be interpreted in a constitutional manner by requiring that a reasonable suspicion be present for the discretion's lawful exercise. The adoption of such an interpretation would not unduly strain the language of the statute. However, a court might take the view that the legislation is so permissive that it is not possible to read the sections down. The protection of basic rights 'should not be dependent upon a reliance on the continuous exemplary conduct of the executive, something that is impossible to monitor or control'. 301 As Sopinka J observed in Little Sisters, 'if the legislation itself lends itself to the repeated violation of Charter rights . . . the legislative scheme itself is partially responsible and must be remedied'. 302 A declaration of incompatibility under s. 4 of the HRA is thus the only appropriate remedy. 297 R v Grant [2005] 3 WLR 437 at para. 56, per Laws LJ. 298 The People v Defoe (1926) 242 NY 13 at 23-4, per Cardozo J: 'The pettiest peace officer would have it in his power through overzeal or indiscretion to confer immunity upon an offender for crimes the most flagitious. A room is searched against the law, and the body of a murdered man is found. If the place of discovery may not be proved, the other circumstances may be insufficient to connect the defendant with the crime. The privacy of the home has been infringed, and the murderer goes free. Another search, once more against the law, discloses counterfeit money or the implements of forgery. 'A mistake by the judiciary in times of terrorism', concluded Justice Barak, 'is worse than a mistake of the legislature and the executive in times of terrorism . . . the reason is that the judiciary's mistakes will remain with the democracy when the threat of terrorism passes and will be entrenched in the case law of the court as a magnet for the development of new and problematic laws'. 303 The decision of the House of Lords in Gillan is one such mistake. The deferential judgment failed to give full and proper effect to a number of important Convention rights, in effect leaving the executive free to introduce broad sweeping powers of stop and search more widely. Experience shows that the 'relaxation of basic constraints on official power is rarely confined to situations of the emergency that gave rise to them'. 304 At the same time the judgment in Gillan weakens the normative value of legality by approving a power that may be exercised arbitrarily in a manner that denies equality of treatment.