Independent report

The Terrorism Acts in 2020 report of the Independent Reviewer of Terrorism Legislation (accessible version)

Updated 6 July 2022

The Terrorism Acts in 2020

REPORT OF THE INDEPENDENT REVIEWER OF TERRORISM LEGISLATION ON THE OPERATION OF THE TERRORISM ACTS 2000 AND 2006, AND THE TERRORISM PREVENTION AND INVESTIGATION MEASURES ACT 2011

By JONATHAN HALL Q.C.

Independent Reviewer of Terrorism Legislation

Presented to Parliament pursuant to Section 36(5) of the Terrorism Act 2006

© Crown copyright 2022

This publication is licensed under the terms of the Open Government Licence v3.0 except where otherwise stated. To view this licence, visit nationalarchives.gov.uk/doc/open-government-licence/version/3.

Where we have identified any third party copyright information you will need to obtain permission from the copyright holders concerned.

This publication is available at www.gov.uk/official-documents.

Any enquiries regarding this publication should be sent to us at:

Direct Communications Unit
Home Office
2 Marsham Street
London
SW1P 4DF

ISBN 978-1-5286-3308-6 E02743840 04/22

Printed on paper containing 40% recycled fibre content minimum

Printed in the UK by HH Associates Ltd. on behalf of the Controller of Her Majesty’s

Stationery Office

EXECUTIVE SUMMARY

  • Statistical monitoring of terrorism legislation is undermined by the increasing yet understandable reluctance of members of the public to self-define their ethnicity (Chapter 1).

  • All 3 completed terrorist attacks in Great Britain were Islamist and all involved serving or released prisoners. Islamist Terrorism remains the principal threat in Great Britain. Weaponised Right Wing Terrorism featured prominently in disrupted late stage attack plots (Chapter 2).

  • Yet more internet-based terrorist organisations were banned. Terrorism legislation is operating at its limits in the face of online dissemination of terrorist publications and encouragement of terrorism in closed groups (Chapter 3).

  • During 2020 the only category to see an increase in terrorism-related arrests was the under 18s (Chapter 5).

  • Amendments to the Terrorism Act and associated Codes of Practice are necessary to deal with accessing remote data at ports, and with the retention of mobile phone downloads (Chapter 6).

  • There needs to be greater attention to measuring risk reduction and neurological diversity in the use of special civil orders against suspected terrorists, and the continuing denial of legal aid is unexplained and wrong in principle (Chapter 8).

  • A major joint agency operation (Operation ARBACIA) was mounted against Dissident Republican terrorists in Northern Ireland. Public perception should not affect the use of terrorism powers. Steps must be taken to ensure greater consistency in terrorism sentencing (Chapter 9).

  • Scotland saw the first prosecution of an incel terrorist (Chapter 10).

1. INTRODUCTION

1.1. This is my third annual report as Independent Reviewer and includes my review into Terrorism Prevention and Investigation Measures (TPIMs). My second annual report, The Terrorism Acts in 2019, was published in March 2021. In the Annex I set out which previous recommendations were accepted or rejected.

Is It Terrorism?

1.2. The fact that the definition of terrorism is broad and vague is not a new observation but bears repetition because it forms the basis for extended criminal liability[footnote 1], the use of strong and intrusive investigative powers including extended pre-charge detention[footnote 2], special civil powers such as TPIMs and, under recent legislation, longer sentences and more intrusive post-release management[footnote 3]. The countering of terrorism already commands huge resources in the United Kingdom, and new forms of ideological violence can stretch budgets and lead to difficult calls on competing priorities.

1.3. I reported last year that the definition remained appropriate for all its imperfections[footnote 4]. The flexibility of the definition requires continuing vigilance by officials, the police, courts, researchers, campaigners, and the media. There will continue to be hard cases involving unfamiliar ideologies, generally referred to as Mixed, Unclear or Uncertain.

1.4. A new form of social pressure suggests that terrorist legislation ought to be applied more liberally. The argument was made following the Plymouth shootings of August 2021[footnote 5] to the effect that labelling a particular use or threat of violence as terrorism was proof that society was properly invested in preventing its recurrence. It has always been part of the charm and danger of the legal edifice of terrorism law that the condemnation of individuals as terrorists brings political and legal dividends[footnote 6]. But allowing the treatment of offences or the exercise of powers to be politicised or used as tool to gain status or legitimacy or resources risks inconsistency, discrimination, and perverse outcomes.

1.5. In the case of the Plymouth shootings, the pressure to label them as terrorism made no discernible difference. My own observations over the past 3 years are that CT Police are conscious of the powers they exercise and thoughtful about their use, and well able to withstand this type of pressure. However, as discussed in Chapter 9, police in Northern Ireland are not immune to considering their powers through the lens of public perception.

Statistics

1.6. Intuition and anecdote are inevitable aspects of independent review; but statistics are better. The collection and analysis of data is important in helping to identify possibly hidden patterns of discrimination[footnote 7]. I have seen this in practice. CTP Borders have become aware of the high proportion of individuals who define themselves as “White Irish” who are examined under Schedule 7 Terrorism Act 2000 (considered in Chapter

6) without what the police assess to be a useful outcome. This calls into question whether the use of the power, principally at Common Travel Area ports, is excessive.

1.7. In my previous annual reports, I drew attention to statistical gaps and inconsistencies which the government has since agreed to address:

  • The number of biometric samples taken during Schedule 7 examination is now published in the annual and quarterly statistics on the operation of police powers under the Terrorism Act 2000 and subsequent legislation[footnote 8].

  • These annual and quarterly statistics also include the use of powers to delay or restrict access to a solicitor under Schedule 7 with further details to be included from 2022[footnote 9].

  • Cordon use is now published on an annual basis.

  • The number of and success rates of warrants of further detention under Schedule 8 Terrorism Act have been published since December 2021.

1.8. Five years after the Home Office said that it was working with CT Police to investigate whether statistics could be collected on the use of section 43 powers by forces other than the Metropolitan Police Service[footnote 10], I am pleased to report that since November 2021 the Home Office has published statistics on the use of section 43 by police forces across England and Wales [footnote 11].

1.9. The categories used for statistics on ethnicity require improvement. As I pointed out in my first report[footnote 12], the use of a catch-all “Chinese or other” to include individuals of Arab ethnicity in England and Wales risks distorting public perception. This category is currently used in respect of section 43 (stop and search) and Schedule 7 (ports and borders examination) powers[footnote 13]:

  • They show that individuals described as “Chinese or other” were the most requently detained under Schedule 7 Terrorism Act 2000[footnote 14].

  • However, CTP Borders unpublished data show that fewer than 1% of persons stopped described themselves as being of Chinese ethnicity.

  • It would be possible to rename “Chinese or other” category as “other”, as is the case in the arrest, charge, and conviction statistics. This would be crude but less misleading[footnote 15].

1.10. The size of the “Other” ethnicity category masks a deeper problem about self- defined ethnicity statistics. Many people who are presented with a range of ethnicity categories and asked to self-define appear to take the (entirely understandable) approach that none of the categories works for them and either specify “Other” or choose not to specify at all. Indeed, the “Not Stated” category is currently the third largest ethnicity category for Schedule 7, larger than “White”. For 2020, those in the “Other” or “Not Stated” categories represent 42% of those subject to Schedule 7 examination.

1.11. Given this rise of “Other” or “Not Stated” ethnicity categories, I recommend that the Home Office and CT Police give consideration as to how to ensure that statistics on the use of terrorism powers can continue to capture useful information about ethnicity.

1.12. One approach might be to require CT Police to record a person’s “ethnic appearance”, which currently applies to arrests, charges and convictions[footnote 16]. As is suggested by the position in Northern Ireland, police can themselves be reluctant to label people[footnote 17]. But reluctance to label people, however imperfectly, could lead to a loss of valuable opportunities to spot discrimination. At the very least it will allow a clear comparison of the use of terrorism powers against those who are white and those who are not. It may not be necessary to get too fine-grained about it: ultimately, most CT Police officers in post are white, and it is important (although not sufficient) to guard against the possibility that powers are disproportionately exercised against people who are not. A further option would be to bolster the ethnicity categories by taking data on nationality (recognising that this too might be imperfect in the case of dual nationals) and religion.

1.13. I further recommend:

  • The use of “Chinese or other” as an ethnicity category in CT statistics should be reconsidered so that it more accurately reflects the individuals within that category.

  • Consideration be given to publishing the statistics for “White Irish” individuals stopped under Schedule 7.

9.32 below.

2. REVIEW OF 2020

National Security Machinery

2.1. The response to national security threats evolved during 2020:

  • In March the government confirmed the existence of a joint state threats assessment team[footnote 18], underlining the increased attention being given to hostile state activity in addition to terrorism.

  • In April MI5 took primacy for Right Wing Terrorism, and Left, Anarchist and Single Issue Terrorism[footnote 19] meaning that terrorist threats emanating from these activities are now managed in the same way as International Terrorism.

  • In November funding was announced for a new counter-terrorism operations centre in West London, which became operational in 2021.

2.2. Officialdom continued to seek public engagement with counter-terrorism matters[footnote 20]. Prominent campaigns encouraged reporting of early signs of terrorist activity, or terrorist ideology in friends or family. Training was offered to the general public on protection against Terrorism in crowded spaces[footnote 21]. In August 2020 detailed guidance was published on responding to Marauding Terrorist Attacks[footnote 22]. ‘See it, Say it, Sorted’ announcements were prominent on trains.

2.3. In January 2020, the government undertook to consider whether new legislation was needed to require owners and operators to improve the protective security of public venues[footnote 23], leading to a formal consultation in the first part of 2021[footnote 24]. In policy terms, this is an aspect of Protect[footnote 25]. Clear eyes are needed about the trade-offs involved in the creation of positive duties affecting the general public. The human rights calculus through which the necessity and fairness of counter- terrorism laws are evaluated is generally focussed on the individual (and above all the right to life). But the collective impact of counter-terrorism measures on society matters greatly. Protecting against terrorism should always seek to avoid degrading “the cheerful spirit of security”[footnote 26] which allows for the spontaneous, the amateur, and the boisterous.

2.4. Victims are taking a more prominent role in the counter-terrorism landscape. In October 2020 funding for victim support services was provided by Home Office for those affected by terror attacks[footnote 27]. The government now supports a website for terrorist victims[footnote 28]. As terrorist attack inquests or inquiries become a settled feature, questions have arisen about the standing of those who are injured not killed[footnote 29], and access to funding. In 2020 the government consulted on its stated intention to create a standalone compensation scheme for victims of terrorism at home and abroad[footnote 30], and on reforms to ‘Pool Re’, the government-backed business insurance scheme against losses caused by certain forms of terrorism.

Threat Level

2.5. The threat level for Great Britain no longer distinguishes between different types of terrorist threat[footnote 31]. During 2020 it was set at ‘substantial’ (meaning that an attack is likely) until November when it was raised to ‘severe’ (attack highly likely)[footnote 32]. This was said to be a precautionary move in response to attacks in France and Vienna which showed that “the temperature of the threat in Europe is rising” with a fear that this, rather than any specific threat, might create a “galvanising effect in the UK”[footnote 33]. In the event there were no copycat attacks in the UK and the threat level was lowered to ‘substantial’ in February 2021.

2.6. The Manchester Arena Inquiry began its work in September 2020. Amongst its first set of published findings was the observation that if a threat level remains high for a long period of time, it becomes more and more difficult to ensure that people maintain the high level of alertness required in relation to potential dangers[footnote 34].

Events

UK Attacks

2.7. There were 3 completed attacks in the Great Britain during 2020, all incidents of Islamist terrorism:

  • In January, Brusthom Ziamani and Baz Hockton attempted to murder a prison officer inside maximum security HMP Whitemoor. They had prepared fake suicide belts and multiple weapons.

  • In February, Sudesh Amman stabbed two passers-by on a street in Streatham, and was shot dead by armed police. He was wearing a fake suicide vest.

  • In June, Khairi Saadallah stabbed and fatally injured 3 men in a park in Reading.

2.8. As with the Fishmongers’ Hall attack in November 2019, all these attackers were serving or released prisoners. Offender management and joint working was the subject of my review of Multi-Agency Public Protection Arrangements published in 2020[footnote 35], and my report Terrorism in Prisons, which was delivered to the government in October 2021.

2.9. Natural public curiosity exists about the extent of the balance between the Islamist and Right Wing terrorist threat. There is no doubt that the Islamist terrorist threat is the main threat faced by Great Britain and is the threat that continues to lead to the greatest number of terrorist deaths within Great Britain. At the same time, police investigation of Right Wing Terrorism has been given significant prominence, prompting some commentators to question whether the Right Wing threat has been exaggerated in order to achieve a politically-correct form of balance.

Arrangements’ (May 2020).

2.10. In my view it is implausible that investigative work is distorted in the interests of balance. Experience shows that CT Police and MI5 are ruthless in prioritising threat and it is far-fetched to believe that counter-terrorism investigative resources are deployed away from the most pressing threats to life.

2.11. The real issue is whether greater official publicity is given to Right Wing Terrorism than is warranted by the actual threat.

2.12. The starting point is the detail given by the Director General of MI5 in a speech in 2021; he stated that there have been 29 late-stage attack plots disrupted over the last four years, of which “fully 10 have been Extreme Right Wing”[footnote 36]. Late-stage attack plots naturally command police resources and figure highly when CT Police and officials consider the risk of actual violence posed by a particular ideology.

2.13. It can be seen from the figures given by the Director-General of MI5 that the number of Right Wing Terrorist late-stage attack plots is relatively high compared to the total of such plots. However, Right Wing Terrorism makes up about 20% of the overall terrorism casework: in other words, it comprises a smaller percentage of the overall terrorism casework and a higher percentage of late-stage attack plots.

2.14. At my request CT Police have identified 7 late-stage attack plots which led to criminal proceedings over the last 3 years. They are:

  • 2018: Steven Bishop (violent racist, amassed bomb-making materials)[footnote 37].

  • 2019: Jack Reed (teenage neo-Nazi, researched explosives)[footnote 38], Sam Imrie (plan to film arson attack on Islamic centre)[footnote 39], Paul Dunleavy (teenage neo- Nazi, obtained weapons)[footnote 40].

  • 2020: Matthew Cronjager (teenage, attempt to access firearms)[footnote 41].

  • 2021: ‘S’ (construction of firearms, trial ongoing), ‘W and others’ (construction of weapons, trial ongoing).

2.15. None of these late-stage attack plots resulted in any injuries.

2.16. Having discussed this issue at some length with CT Police, the point seems to be, as suggested by the 7 cases referred to above, that the possession, manufacture, discussion of weapons and explosives are often encountered in the context of Right Wing Terrorism investigations. Sorting between plots likely to result in violence and the noise and rhetoric is difficult. But where “weaponization” is in play, CT Police and MI5 are bound from a protective point of view to take such cases more seriously.

2.17. It follows that any imbalance in perception may be caused because the public is most aware of completed terrorist violence (generally Islamist and low sophistication) but CT Police and MI5 investigate a set of Right Wing Terrorist plots which, because of certain features, can hardly be ignored even though they have not moved - and indeed may not ever move - to actual violence.

Overseas attacks

2.18. Ten completed jihadist attacks are recorded as having been completed in the European Union in 2020, double the number in 2019: these were all carried out by lone individuals and, other than the attacks in Vienna in which firearms were used, all employed rudimentary methods[footnote 42]. Different European countries experienced notably different types of attack: in Germany most terrorist attacks were Right Wing, in Italy the attacks were mainly Left-Wing and Anarchist[footnote 43].

2.19. Preliminary global data for 2020 suggests fewer than 10,000 deaths from terrorism during 2020, down from a peak of 35,000 in 2014. The fall in the number of deaths from terrorism may reflect the military defeat of the so-called Islamic State in Syria and Iraq, and military interventions against Boko Haram in Nigeria. It has been suggested that the epicentre of terrorism has now shifted out of the Middle East and North Africa region and into sub-Saharan Africa[footnote 44].

Syria

2.20. In December 2020 a senior CT police officer referred to the government’s approach of “strategic distance” with respect to British nationals and residents who had travelled out to join the so-called Caliphate of Da’esh/ Islamic State: those who had chosen to leave should not in general be allowed to return, to avoid the risk of importing risk into the country[footnote 45].

2.21. The principal executive measure for furthering this policy has been deprivation of citizenship, a counter-terrorism power in all but name which falls outside the scope of the Independent Reviewer of Terrorism Legislation[footnote 46].

  • During 2020 the government’s approach suffered a reverse, when the Court of Appeal held that Shamima Begum should be allowed to return temporarily to the UK to ensure a fair hearing of her appeal against the deprivation of her citizenship[footnote 47]. The Court felt able to conclude that the national security concerns about her could be addressed and managed in the UK, potentially by use of Terrorism Prevention and Investigation Measures (discussed in Chapter 8).

  • The Supreme Court subsequently overturned this approach as being outside the Court of Appeal’s institutional and constitutional competence[footnote 48].

  • Because Shamima Begum’s case has only proceeded in the courts on issues of principle, the precise quantum of her risk to national security and the rationality of Home Secretary’s exercise of discretion, are yet to be the subject of independent scrutiny.

2.22. Future prosecution of terrorist travellers on return the UK remains largely aspirational although British information was ultimately used in the United States, subject to assurances against the use of the death penalty[footnote 49], to prosecute two former (now deprived) British nationals known as members of the ‘Beatles’. But uncertainty remains about the precise number of prosecutions for terrorism that have taken place of British nationals who travelled to Syria to join Da’esh/ Islamic State and then returned to the UK.

2.23. The most recent government estimate provided to me is that since 2013 over 900 UK-linked individuals of national security concern have travelled to engage with the Syrian conflict, with approximately 25% killed overseas and just under half known to have returned to the UK[footnote 50]. This equates to approximately 360 returned British nationals, of whom in the region of 10% are said to have been prosecuted, although this appears to cover returners also prosecuted for non-terrorist crimes[footnote 51].

2.24. No investigative or prosecutorial body in the United Kingdom keeps a record, either officially or unofficially, of the precise information that is needed to answer this question posed, that is the number of individuals who are known to have travelled to join Da’esh Islamic State in Iraq or Syria, and who have subsequently been prosecuted on their return for terrorist offences specifically in respect of their conduct in, or travel to, Syria.

2.25. However, Daniel De Simone of the BBC who has a longstanding involvement in reporting on the prosecution of foreign terrorist fighters has provided me with the following details which I reproduce with his permission:

As at October 2021, 10 returners have been convicted of Islamist terrorism offences in relation to their conduct in Syria (dates of conviction in brackets). They are:

A further 4 returners have been convicted of terrorism offences in relation to documents seized in the UK either after their departure or after their return, rather than their conduct in Syria. They are:

2.26. The position of British children, taken there or born to British dual-national mothers before their deprivation, raises fierce difficulties: there is the condition of the camps in which they live, and bleakness of their futures, the desirability or feasibility of separating them from their mothers, and the risk that such children may be brought up as career terrorists.

Coronavirus and Online

2.27. Living under the pandemic provides a strong indication of what it is like to be on the sharp end of mighty state restrictions, such as TPIMs and terrorist licence conditions. The pandemic provides a pungent lesson in balancing individual freedoms against the risk of social harm.

2.28. It is too early to assess the impact of the pandemic on terrorism. It is likely to have had a temporary suppressive effect because terrorists too have been affected by Covid restrictions and fewer crowds means reduced targeting opportunities. The pandemic has also affected the operation of the counter-terrorism machine resulting in fewer arrests and convictions[footnote 66], fewer targeted Schedule 7 examinations, and a temporary modification of the system of National Security Determinations covering the retention of biometrics.

2.29. Real world restrictions and the unprecedented role of online communication and entertainment during the lockdowns of 2020 may well have taken some terrorist activity online[footnote 67]. Statistics for 2021 show a sharp growth in the charges for disseminating terrorist publications[footnote 68].

2.30. The pandemic has been a boon for conspiracy theories, some of which feed pre-existing violent ideologies[footnote 69], and fringe beliefs. Recent research concerning risk assessments written between 2010 and 2017[footnote 70] on 235 radicalised individuals who had been “convicted of extremist offences” (including but not limited to terrorist offenders) found a dramatic increase in the role played by the internet: prior to 2005 the primary method of radicalisation was face-to-face in 83% of cases, but this figure fell to 17% in 2015-17[footnote 71], as the role of the internet has grown and grown.

2.31. The exhaust fumes of the internet include violent terrorist propaganda, whose possession does not necessarily amount to an offence but is referred to as “mindset material” by police and prosecutors. In my last annual report, I concluded that despite influential calls to extend the criminal law in this field, the scale of the problem was ultimately inconsistent with creating new terrorist offences[footnote 72]: the government has accepted this analysis[footnote 73]. Other traditional tools, such as the proscription or designation of terrorist organisations, are increasingly out of place in the unstructured world of online communication whose penetration, convenience and persuasiveness allow violent ideologues to advertise their brand and beliefs to millions. Proscription or designation is therefore unlikely to provide the clarity that tech companies cherish in support of automated takedown[footnote 74].

2.32. In December 2020 the government indicated that it proposed to tackle the problem at source by legislating for oversight of tech companies, leading in due course to the Online Safety Bill, and published a voluntary Interim Code of Practice[footnote 75]. The impact of the Code’s worthy principles of minimising access to terrorist material and supporting investigations remain to be seen and may depend on how service providers interpret the definitions in Appendix 1.

  • Online terrorist content is defined as “any content which, by uploading it or otherwise making it available to others online, a person is committing an offence under UK terrorism laws”.

  • But liability in this field does not depend purely on content: offences under section 12 Terrorism Act 2000 and section 2 Terrorism Act 2006 require at the least a reckless state of mind on the part of the encourager[footnote 76].

  • The question is whether, when a service provider is contemplating violent imagery, it concludes that a user is unlikely to have uploaded it or otherwise made it available without at least being reckless as to encouraging terrorism in others. This requires an assessment on mental state drawn from the content and circumstances of the posting[footnote 77].

  • Conversely, if the service provider considers its own liability for posting material, it is most unlikely (because of extensive free speech and commercial protections given to service providers[footnote 78]) to conclude that the content contravenes UK terrorism laws.

  • In any event, the Code is at present voluntary.

2.33. Tech companies frequently plead for leadership from government but resist regulation[footnote 79]. The Terrorist Content Analytics Platform developed by service providers operates at a scale that is modest[footnote 80] in comparison to the volume of material that is in circulation. In the second quarter of 2021, Facebook claims to have “taken action on”

7.1 million pieces related to “Dangerous Organisations: Terrorism”[footnote 81]. It is interesting to note that Facebook’s internal policy on dangerous organisations draws tiered distinctions between those who target civilians and those who only target state actors: these are distinctions which hardly correspond with international, let alone UK, approaches to terrorist harm[footnote 82].

2.34. I consider terrorist content and the new draft Online Safety Bill further in Chapter 3.

Brexit

2.35. The United Kingdom left the European Union on 31 January 2020 and the transition period, during which the UK remained in the EU single market and customs union, ended on 31 December 2020.

2.36. In February, a bomb was found in a lorry in an industrial estate in County Armagh, Northern Ireland. This bomb is assessed by officials to have been part of a plot to blow up a ferry whilst it was leaving Belfast Docks (on its way to Cairnryan, Scotland) on Brexit day.

2.37. Cooperation on criminal justice matters between the UK and the Republic of Ireland has until recently depended to a large extent on EU level agreements[footnote 83]. The same is true of many other areas of cooperation with other Member States and officials have been working to replicate agreements on the exchange of data, for example carrier data for passenger ferries arriving from the continent. As with the pandemic the impact of Brexit on day-to-day CT policing is probably too early to judge. To date, no police officer or official has identified a specific incident to me whose detection, investigation or prosecution has been adversely affected by Brexit. However, restrictions on access to EU data are probably not going to be felt with respect to specific incidents: the benefit of sharing data is often to discover previously unknown threats.

Legislation

Terrorist Offenders (Restriction of Early Release) Act 2020

2.38. In February 2020, the Terrorist Offenders (Restriction of Early Release) Act 2020 (“TORERA”) came into force.[footnote 84] The impetus for this emergency legislation was the attacks at Fishmonger’s Hall and Streatham, London. The Bill was introduced on 11 February, all stages in the House of Commons took place on 12 February (omitting a report stage), and the Bill completed its parliamentary stages in time to receive Royal Assent on 26 February.

2.39. The effect of TORERA was to toughen the release arrangements for certain terrorist prisoners so that even those not deemed to be dangerous within the meaning of the Sentencing Code would (in England, Wales, and Scotland[footnote 85]) be subject to the Parole Board release mechanism; and be liable (in England and Wales) to greater control on release through licence conditions set by the Secretary of State. Unlike a different amendment that came into force in April affecting the release of certain violent or sexual offenders, the TORERA changes were retrospective and therefore applied to serving terrorist prisoners[footnote 86]: indeed, the basis of bringing emergency legislation was that certain terrorist offenders were shortly due for automatic release.

2.40. Terrorism offenders[footnote 87] therefore emerged as a special class in sentencing terms, no longer capable of early automatic release.

2.41. A serving terrorist offender, Mohammed Zahir Khan[footnote 88], unsuccessfully challenged the legislation. The High Court held[footnote 89] that faced with the immediate threats to public safety demonstrated by the Fishmongers’ Hall and Streatham attacks, altering arrangements for the earlier release of terrorist prisoners was a logical and rational response[footnote 90]; that the new release provisions did not impose an additional penalty; and that the changes were in principle foreseeable (at least in the rather artificial sense that foreseeability has been interpreted by the European Court of Human Rights)[footnote 91].

Counter-Terrorism and Sentencing Bill

2.42. Hot on the heels of the emergency legislation, in May 2020 came the first reading of the Bill that was to become the Counter-Terrorism and Sentencing Act 2021[footnote 92]. Changes were proposed and made to length of sentence, release[footnote 93], and post- release monitoring (including the use of polygraph[footnote 94]), and the effect of the emergency legislation was extended to Northern Ireland[footnote 95]. The treatment of terrorist offenders as a separate class was therefore consolidated.

2.43. The Bill was subject to an exchange of correspondence between the government and the UN Special Rapporteur on the promotion and protection of human rights and fundamental freedoms while countering terrorism[footnote 96]. The government, under the heading of ‘Oversight’, relied on Parliamentary scrutiny and drew attention to the 5 notes I published on my website during the passage of the Bill. Whilst those notes were widely cited in debate, a general consensus that terrorist offenders should spend longer in prison meant that debate was most forceful at the margins (for example, on amendments to the TPIM regime that I consider in Chapter 8).

Other

2.44. Following a narrow win in the Investigatory Powers Tribunal[footnote 97], in September the government brought forward legislation to authorise the commission of crimes by human sources, subsequently the Covert Human Intelligence Sources (Criminal Conduct) Act 2021. The subject matter of the litigation was agent handling by MI5, whose non-statutory guidelines referred to the use of agents against sophisticated terrorist individuals or organisations.

2.45. As the Court of Appeal subsequently held, the offence of membership of a proscribed terrorist organisation (section 11 Terrorism Act 2000) is an obvious potential example of the criminality that undercover agents may be required to commit[footnote 98]. The Act extends to Northern Ireland where the past role played by human sources in murders was reviewed in the de Silva report[footnote 99] and is currently the subject of the police investigation “Operation Kenova”[footnote 100].

2.46. Further legal underpinning of a different sort was provided by an amendment to the Criminal Procedure Rules. In a small number of terrorist trials, prosecutors had exceptionally notified judges of sensitive material which, whilst not strictly discloseable under the Criminal Procedure and Investigations Act 1996, they considered ought to be made available to the Court in the interests of fairness to avoid the risk of inadvertent mismanagement of those trials. Following an appeal concerning this practice by the bomb-maker and attack-planner Khalid Ali[footnote 101], the rules now recognise this practice[footnote 102].

3. TERRORIST GROUPS

3.1. Proscription of an organisation under the Terrorism Act 2000 is a strong executive measure with immediate practical consequences.

3.2. Once an organisation has been proscribed membership becomes an offence carrying up to 14 years’ imprisonment[footnote 103], as is other conduct relating to meetings, flags and uniforms, support, and funding. Proscription is a gateway to the imposition of TPIMs, exclusion, deportation, digital takedown, sanctions, and non-jury trials in Northern Ireland[footnote 104]. It also has a powerful symbolic effect[footnote 105]. There are currently 78 groups banned under the Terrorism Act 2000 and 14 organisations in Northern Ireland who were proscribed under predecessor legislation[footnote 106].

3.3. The potency of proscription was apparent in the High Court’s decision in 2020 that the offence of displaying an article in a public place in such a way as to arouse suspicion that he is a member or supporter of a proscribed organisation (section 13 Terrorism Act 2000) was an offence of strict liability[footnote 107]. There was no need for the prosecution to prove that the defendant understood the import of the item or article he was carrying, or its capacity to arouse the requisite suspicion[footnote 108]. This followed from the need to prevent others being encouraged to support the organisation or view it as legitimate[footnote 109]. In other words, once an organisation has been banned by the Home Secretary, the criminal law comes down heavily on those who might, even inadvertently, solicit support in this way. The High Court’s decision was subsequently upheld by the Supreme Court[footnote 110].

Proscription Activity in 2020

3.4. Proscription depends on an assessment by the Home Secretary or the Secretary of State for Northern Ireland that the organisation “is concerned in terrorism”[footnote 111], and an exercise of discretion in favour of proscription as a justified response. In practice this assessment follows a recommendation by the Proscription Review Group (PRG), a meeting of officials from the Home Office, Foreign Commonwealth and Development Office, the intelligence agencies, and from the Department for levelling up, Housing, and Communities (on the question of community impact).

3.5. The Proscription Review Group also meets to consider whether existing proscribed organisations should be banned under an additional name. In 2020 this led to the addition of System Resistance Network as a further name for the proscribed organisation National Action[footnote 112] and to two further names being proscribed for the Kurdistan Workers’ Party (the PKK)[footnote 113].

3.6. Two new groups were proscribed during 2020:

  • Sonnenkrieg Division (‘SKD’), in February 2020[footnote 114]. SKD is a white supremacist group, formed in March 2018 as an offshoot of National Action/ System Resistance Network. Two teenage members of the group were sentenced for offences of encouraging terrorism and possession of documents useful to a terrorist in June 2019[footnote 115]. Its founder, 24-year-old Andrew Dymock, a politics student from Bath, was arrested in 2018 and convicted in 2021 of 13 terrorism offences.

  • Feuerkrieg Division (‘FKD’), in July 2020[footnote 116]. A white supremacist neo-Nazi group, FKD was assessed to have members across North America and Europe offering mainly online support to an apocalyptic race war. During 2019 various supporters in the United States and the United Kingdom were arrested. FKD was founded by a 13-year-old Estonian national[footnote 117]. In 2021 a 16-year-old boy was sentenced for terrorist publication offences connected to his role as “British head” of the organisation when he was just 14[footnote 118]. One of his recruits was Paul Dunleavy, 16 at the time, who was later convicted of attack planning[footnote 119]. It announced in February 2019 that it was dissolving[footnote 120] but was reportedly still recruiting in 2021[footnote 121].

3.7. There are some points of similarity.

(i) Both groups are of relatively recent formation.

(ii) They are principally active online (often on closed channels).

(iii) They attract young or very young members and organisers.

(iv) The proscriptions took place after a sequence of arrests.

(v) Both are adherents of ‘Siege Culture’, named after “Siege”, a collection of the writings of the neo-Nazi James Mason who advocated armed struggle in support of a race war[footnote 122]. The text is readily accessible via standard search tools and was available as an Audiobook on YouTube until July 2019[footnote 123]. It calls in numerous places for the murder of black people to fan the flames of total system revolution.

3.8. Applying the 5 published “discretionary factors” to online activity is not impossible but requires versatile thinking[footnote 124]. Two of these are “the nature and scale of an organisation’s activities” and the “extent of the organisation’s presence in the UK”. Where a group operates mainly online, the scale of its activities and the extent to which any organiser is physically present in the jurisdiction is perhaps less important than the extent to which its output is consumed and potentially acted on by UK-based individuals.

3.9. Proscription of online Right Wing Terrorist groups has proven to be an effective means of disrupting the consumption of their material, even if the organiser remains out of reach of the criminal law. It also provides a useful signal to service providers as to the need to remove terrorist content: whether by removing badged material, or restricting certain channels or accounts.

3.10. However, the use of proscription activity against online groups like SKD and FKD has led, understandably, to calls for the proscription of other analogous groups[footnote 125]. Public expectations and understanding would be better served by explaining how the discretionary factors operate in the online space. I therefore recommend that the Home Secretary should provide greater clarity over how the five public discretionary factors operate against predominantly online groups.

Texts and the Limits of Proscription

3.11. Inspirational to terrorists though books such as “Siege” undoubtedly are, proscription does not apply to texts. For one thing a book cannot be “concerned in terrorism” which is by definition a matter of human agency[footnote 126]. Nor is it possible to proscribe a book’s readership as an alternative method of banning a book’s circulation. Although ‘organisation’ is loosely defined to include any association or combination of persons[footnote 127] it requires more than individuals connected solely by common readership, or common agreement with an ideology such as the Great Replacement Theory[footnote 128].

3.12. On this view, the online age could see proscription becoming of less relevance in countering domestic terrorism. In the past, physical coordination between like- minded individuals was required in the form of meetings, weapons training, and the generation of printed material, in order to disseminate terrorist tradecraft, capability, and inspiration. Proscription of groups was an effective means of reducing this type of activity, as it has been for some online groups. But the decentralised nature of the internet and the ease of user-generated content means that individuals can find all the inspiration and tradecraft they need without joining or even interacting with any identifiable group.

Criminal offence alternatives

3.13. As the utility of proscription declines in this context, criminal offences directed at terrorist publications may become more relevant, both in terms of prosecution and, as discussed further below, as being integral to regulated online safety.

3.14. The starting point must be that criminalising a text merely because in the wrong hands it may encourage acts of terrorism is an unacceptable interference with the principle of personal autonomy[footnote 129]; and is unworkable because sensible distinctions cannot be made between texts that are inherently encouraging of terrorism and texts that are not. After all, extracts from mainstream religious texts are routinely used to justify terrorist violence.

3.15. The closest that terrorism legislation comes to criminalising texts is the offence of collecting or having possession of information likely to be useful to a person committing or preparing an act of terrorism[footnote 130]. However, as interpreted by the House of Lords[footnote 131], this refers to that technical subset of information that must “of its very nature, be designed to provide practical assistance to a person committing or preparing an act of terrorism”. It is therefore likely to capture only a small subset of information on the internet (I discuss the offence further in Chapter 7).

3.16. Terrorism offences generally require more. The key offence which could apply to texts like “Siege” is the offence of dissemination contrary to section 2 Terrorism Act 2006. Proof of the offence requires a potentially dangerous text (a “terrorist publication”) plus an act of knowing or reckless dissemination:

  • “Terrorist publications” contain matters not only that are likely to be useful to terrorists but also those which are likely to be understood by a reasonable person as a direct or indirect encouragement or other inducement to the commission, preparation, or instigation of terrorism[footnote 132]. This latter category can include ‘terrorist propaganda’ such as beheading videos[footnote 133], even though mere possession of this material does not and could not sensibly form the basis of criminal liability[footnote 134]. There has been a recent sharp growth in the use of this charge[footnote 135].

  • Proof of the offence is dependent on showing that the disseminator[footnote 136] intends, or is subjectively reckless, that his conduct will encourage or provide assistance to the commission of acts of terrorism[footnote 137]. This depends on the content of the publication and the particular circumstances in which the conduct, i.e. the dissemination, occurs[footnote 138]. It is therefore unnecessary to show that the individual intended that any act of terrorism would be carried out.

  • In practice the foundational requirements for prosecuting the offence are proof of attribution (that the defendant himself was responsible for the dissemination) and proof of knowledge (that the defendant knew of the terrorist nature of the publication). Once these two points are established prosecutors have the platform to establish that, in light of the content of the material and the circumstances of the dissemination, the defendant must at the very least have been reckless that the recipient would be encouraged to commit an act of terrorism.

3.17. Whether there is sufficient evidence to prosecute, and whether it is in the public interest to do so, are highly context-specific, but the availability of terrorist publications online and the ease of republication opens the door very widely to potential criminal liability. The section 2 offence may capture a very wide range of blameworthy conduct. A recent judgment on whether a foreign national offender had lost the benefit of the Refugee Convention illustrates the point. Though the offender had been convicted under section 2 Terrorism Act 2006 and sentenced to three and a half years’ imprisonment, there was no evidence as to who viewed the posts, no evidence that anybody of influence was commending the appellant, approving her actions, or using her materials, or that her actions led to any terrorist acts, or had any other impact on the behaviour of others, here or abroad[footnote 139].

3.18. This begs the question whether any posting of “Siege” would incur liability under section 2. If the post is to an online forum, particular a forum which at times celebrates violent racism, it is plausible that amongst these individuals will be one individual who will be encouraged by the dissemination of “Siege” to carry out an act of violence in order to intimidate a section of the public and advance a racist cause, i.e. to commit an act of terrorism. It is very unlikely that prosecuting the dissemination of “Siege” would interfere impermissibly with the right to freedom of expression as interpreted by the European Court of Human Rights[footnote 140]. If the post is on a gardening forum, liability under section 2 is less likely to be established because it will be difficult to prove that disseminator intended or knew the risk that one of the recipients would be encouraged to commit a terrorist act.

Online Safety

3.19. The section 2 offence is relevant to the government’s draft Online Safety Bill which proposes enforceable duties for service providers to minimise the presence online of “illegal content” which is content that “amounts to” a terrorism offence[footnote 141]. The draft Bill recognises content does not in itself amount to an offence, and that service providers cannot wait for a successful prosecution before they act: it therefore provides that content amounts to a relevant offence if the service provider has reasonable grounds to believe that the use or dissemination of the content constitutes a relevant offence[footnote 142] including the section 2 offence.

3.20. There are several observations to make about the proposed interaction between the criminal law and the online safety duty.

3.21. Firstly, service providers are not criminal law experts and, in many cases, will be physically headquartered in an entirely different criminal law jurisdiction. There is the risk of an overbroad assessment of what constitutes a terrorism offence

(restricting the legitimate free flow of content), as well as an assessment that is, in terms of restricting access to terrorist content, too narrow.

3.22. Secondly, in any event the question of whether dissemination of a text amounts to a terrorist offence is not clear-cut and depends on a range of circumstances, not all of which may be known to service providers.

  • Uncertainty could be about definition: for example deciding whether the violence promoted by certain incel publications falls under the category of “terrorism”.

  • Uncertainty could arise from information deficit: not knowing whether the dissemination in question is intended for violent obsessives or concerned academics.

  • Again, this could lead to an excess of caution or, conversely, to service providers standing their ground on the principle against doubtful penalisation, and the criminal standard of proof[footnote 143].

Online Public and Private

3.23. Section 2 refers to terrorist “publication” which implies some element of a self- contained text which is intended for repeated exposure to the public[footnote 144], even though the dissemination may be in a private setting. Conversely, section 1 Terrorism Act applies to all words which encourage terrorism, whether or not they are properly described as a “terrorist publication” – and would therefore apply to words that exhorted terrorist violence which merely made a reference to or paraphrased “Siege”.

3.24. The recipients of the encouragement must, for the offence to be established, be members of the public[footnote 145]. It is open to question to what extent members of a closed Telegram channel, for example, could be described as members of the public for the purposes of section 1.

3.25. This is an aspect of a larger issue about the meaning of public in the online space in connection with terrorism offending[footnote 146]. The definition of public in the Terrorism Act 2006 includes “references to a meeting or other group of other persons which is open to the public (whether unconditionally or on the making of a payment or the satisfaction of other conditions)”[footnote 147].

3.26. The fundamental difficulty with applying pre- or early internet notions of privacy that it is very easy to create ‘technical’ privacy online in a way that has no real world analogue. Technical privacy may be secured by accreditation, encryption and anonymous or pseudonymous participation, for very large numbers of people, who may be spread right across the globe, to establish a private and protected space that is not “open to the public”. The costs of establishing a private group in the real world– traveling to meet, organising a venue, physically excluding the general public – simply do not apply online.

3.27. It follows that whereas the definition of public would extend to an online group open to any internet user who can satisfy a particular condition, such as having a user name beginning with the letter ‘P’, the definition does not sit easily with an invitation- only group that is not publicly visible, even if it happens to contain one hundred thousand members. Dissemination of “terrorist texts” within such a group could, depending on the nature of the group, result in the commission of an offence under section 2; but encouragement of terrorism in such a group is unlikely to lead to criminal liability under section 1.

Section 3 Terrorism Act 2006

3.28. The progress of the Online Safety Bill and its application to online terrorist content is a matter for future reports. However, it is worth drawing attention to an existing power under terrorism legislation. Under section 3 Terrorism Act 2006, a constable may require an internet service provider to remove material that is “unlawfully terrorism-related” (meaning in broad terms, the same as terrorist publication[footnote 148]), which opens the door to the service provider and not just the user- generator being prosecuted for the section 2 offence[footnote 149]. This power has in fact never been used. Under Home Office Guidance which is no longer publicly available, voluntary dialogue is to be encouraged.

3.29. A risk with voluntary dialogue, as with the Online Safety Bill, is that views about what is “unlawfully terrorism-related” may differ. The fact that the full text of “Siege” is still readily available online suggests either that voluntary dialogue has not been attempted, or has been unproductive. It is to be expected that service providers will rightly challenge policemen who tell them what they can and cannot do in the field of free expression.

3.30. Clarity comes from experience. Strange as it may seem[footnote 150], beheading videos have in the real world been found to constitute an encouragement to terrorism[footnote 151]. Forging a connection between criminal cases and decisions on whether publications are “unlawfully terrorism-related” can only assist in dispelling doubt or disagreement, particularly if and when new terrorist ideologies are encountered.

3.31. It is not the role of criminal courts to determine whether service providers should take down terrorist content. However, by loose analogy with the power of the criminal court to recommend deportation[footnote 152], a criminal court which on a prosecution for a terrorism offence receives evidence of online content which, on analysis, is “unlawfully-terrorism related” is well-placed to recommend that the section 3 power be exercised. This could ensure a continuing link between material that is encountered in connection with terrorist offending, and the section 3 power.

3.32. I therefore recommend that legislation should enable a court sentencing an individual for a terrorism or terrorism-connected offence to make a content recommendation. A content recommendation would be a recommendation by the court that a section 3 direction should be given by a constable in respect of what is, in the court’s view, “unlawfully terrorism-related” material. It would then be for the police to determine whether to use the section 3 power or seek to achieve the same result by voluntary dialogue or, if appropriate, take no further steps.

Deproscription

3.33. The Proscription Review Group advises on applications for deproscription made under section 4 Terrorism Act 2000, and would in theory provide advice if the Secretary of State wished to consider deproscription of her own motion. If the Secretary of State agrees to deproscribe, an order is laid before Parliament for its approval[footnote 153].

3.34. No applications for deproscription were received during 2020, but in October the Proscribed Organisations Appeal Commission handed down its judgment on the Home Secretary’s refusal to deproscribe the Liberation Tigers of Tamil Eelam (the LTTE) [footnote 154]. As I reported in last year’s report, the Commission found that the refusal was flawed because the views of the Proscription Review Group were materially misstated in the submission put up to the Home Secretary[footnote 155].

3.35. It was procedurally flawed decision-making, which in my view reflected at least to some extent the lack of periodic review of proscription decisions[footnote 156], that led to the appeal being allowed. The government will have taken comfort from the fact that the Commission did not find that the Secretary of State’s decision was unsupported by evidence or Wednesbury unreasonable[footnote 157].

3.36. The outcome of the appeal was that the Home Secretary undertook to reconsider the LTTE deproscription application in 90 days, later varied to allow the Appellants to make further representations before that period began.

3.37. It should be noted that the Commission has no power to order that a group is deproscribed[footnote 158]. The most the Commission can do where allows an appeal is to make an order following which the Secretary of State shall as soon as reasonably practicable lay a removal order before Parliament[footnote 159], although the Commission declined to exercise that power in the LTTE case[footnote 160].

  • Because Parliament is sovereign on issues of proscription, the theoretical possibility exists that Parliament could simply refuse to make an order that it disagreed with, however irrational the Commission concluded a group’s continuing proscription to be.

  • However, on the single occasion that the Commission found a proscription was perverse (the People’s Muhajideen of Iran), an order was duly agreed[footnote 161]. In the extremely unlikely event that Parliament refused to desproscribe in these circumstances, it probable that the other constitutional arrangements would eventually come to the rescue[footnote 162].

3.38. The first and principal deproscription consideration is whether the group is still “concerned in terrorism”. Total contemporaneity between information on involvement and the point of determination is unlikely. Inferences of continuing involvement are made based on comparatively recent information; in practice, the Joint Terrorism Assessment Centre (JTAC) has been instructed to identify evidence within the last 12 to 18 months[footnote 163]. Older information is relevant for historical context[footnote 164], and sustained past involvement in terrorism by an organisation can put a sinister gloss on more recent events. Like individuals who are placed under special civil orders (TPIMs, considered in Chapter 8), groups may find it hard to escape the implications of their past activity.

3.39. Short of a shift in international relations leading for example to a dramatic need to rehabilitate or accommodate a current terrorist group (perhaps as part of a peace process) [footnote 165], it would take a brave act for the Home Secretary or Secretary of State for Northern Ireland to conclude that the time had come to deproscribe a group that was concerned in terrorism. Once continuing terrorism is established, the public interest will almost inevitably favour the curtailment of individual rights of expression and association which, it may be said with some force, can be enjoyed effectively without involvement in the organisation in question.

3.40. But particularly when dealing with foreign groups, a cosmopolitan approach may be needed on the issue of impact. In the LTTE case, evidence was given that the legitimate flag of Tamil Eelam (the putative separate Tamil state) was so similar to the flag of the proscribed LTTE that demonstrators were afraid to use it, and social media companies were likely to remove it. Where a proscribed organisation is closely associated with, or the dominant means, of expressing a political view or ideology, then it may be difficult to protest without attracting suspicion. Similar points may arise in the Northern Ireland context. Continuing proscription may be justified but should not be considered as cost-free.

3.41. The maximum penalty for membership of a proscribed group was raised from 10 to 14 years by the Counter-Terrorism and Sentencing Act 2021. I reiterate the view, echoing Lord Anderson QC, that this level of jeopardy should not exist for groups that are no longer concerned in terrorism, and maintain that some form of systematic review should ensure that groups that are no longer concerned in terrorism should not remain proscribed.

Aid Sector

3.42. The tension between counter-terrorism and humanitarian imperatives is on full display after the fall of the Afghan government. The Taleban’s deputy leader and appointee as interior minister, Sirajuddin Haqqani, is identified by the Home Office as one of the leaders of a proscribed organisation, the Haqqani Network. The Network is assessed to have “long established links” with Al Qa’ida[footnote 166]. Contact and interaction between aid or capacity/ peace-building agencies and Afghan ministries is necessary if work, often funded by government departments to avert or mitigate humanitarian crises, is to be undertaken.

3.43. The United States government has shown commendable responsiveness in publicly declaring its commitment that its sanctions against the Taleban and the Haqqaqi Network should not limit the ability of civilians in Afghanistan to receive humanitarian support from the US government and international community, alongside the issuance of general licences[footnote 167].

3.44. However, UK government communications have been less clear.

3.45. On 1 November 2021, in response to events in Afghanistan, the Office of Financial Sanctions Implementation (‘OFSI’) issued revised Charity Sector Guidance, principally directed at sanctions but also concerned with terrorist financing offences under the Terrorism Act 2000[footnote 168].

  • The guidance states, under the heading ‘Afghanistan’ that sanctions “will not automatically preclude all forms of contact or engagement with a designated person…”. The effect of this message is to put any contact or engagement with a designated person under the microscope.

  • However, the sanctions in question (the Afghanistan (Sanctions) (EU Exit) Regulations 2020) amount to financial and trade sanctions only. Designations under this regime could not purport to limit mere contact and engagement.

  • If the reference to “contact or engagement” was intended as a reference to the offence of attending a meeting to further the activities of a proscribed organisation that is not a matter of sanctions at all, but a potential offence under section 12(2) Terrorism Act 2000 in relation to proscribed organisations not designated individuals.

  • Moreover, if the intention had been to refer to section 12(2), then the guidance is incomplete without reference to the ‘For Information Note’ published by the Home Office and OFSI which expressly and helpfully refers to the defence that exists for genuinely benign meetings “…for example, a meeting designed to encourage a proscribed organisation to engage in a peace process or facilitate delivery of humanitarian aid”[footnote 169].

3.46. OFSI is responsible for the implementation and enforcement of financial sanctions but inevitably its guidance will wish to refer to terrorist financing under the Terrorism Act 2000 and potentially wider offences under that Act. Guidance of this nature is an important public-facing document that will be perceived as an expression of government policy in its interpretation and application and enforcement of relevant legislation. Official communication can be a valuable source of confidence the financial sector on whom aid delivery often depends, whereas uncertainty is anathema to the private sector[footnote 170].

3.47. To my mind, there is a question as to whether OFSI, and the relevant government departments and bodies with which it consulted in the drafting of the above guidance, fully appreciated the wider impact of this wording. The goal may have been to remind aid sector organisations of the risks of involvement in Afghanistan, but the terminology used may well have been more inhibiting than intended.

3.48. To date I have made two recommendations in connection with aid agencies, both of which were accepted.

3.49. Firstly, the Home Secretary agreed to meetings between Home Office officials, National Crime Agencies officers and aid agencies within the Tri-Sector Group to consider (and ‘workshop’) certain identified scenarios[footnote 171] with a view to formulating guidance on the use of section 21ZA in connection with humanitarian assistance[footnote 172]. A meeting took place in April 2021 and a further meeting took place in October 2021.

3.50. The effectiveness of section 21ZA – by which National Crime Agency officers may authorise transactions which would otherwise put aid agencies and others at risk of committing terrorist funding offences – has therefore for the first time been discussed in a meaningful way to test whether it provides a practical option. The exercise has, as hoped, led officials to consider more holistically how extra-territorial terrorism legislation should operate, that is, by reference to the aid ambitions of the interests of the FCDO as well as the security interests rightly articulated by the Home Office.

3.51. It is to be hoped that progress is maintained in considering factors such as the size of the amount given to a proscribed organisation (for example where a small toll is paid at a crossing point) and the nature of the payment (for example, paying for the travel of a member of a proscribed organisation to a peace conference). In principle, section 21ZA is available where the overall balance of risk (including the risk that aid will not achieve its intended purpose) favours the transaction[footnote 173].

3.52. In contrast to my slightly lukewarm references in last year’s report, I am now able to point more enthusiastically to progress by the Tri-Sector Group, the body that was set up in 2017 on Lord Anderson QC’s recommendation, and which comprises officials from government, the banking industry, and NGOs.

3.53. Other work by the Tri-Sector Group has led to the recent publication of a revised For Information Note, the Home Office guidance on operating within counter- terrorism legislation to which I have already referred[footnote 174]. This is a restructured, expanded, and welcome development of earlier guidance and is a tangible benefit of the Tri-Sector process. NGOs are also participating in a February 2021 international initiative to study and mitigate the unintended consequences (de-risking, financial inclusion, undue targeting of non-profit organisations, and the curtailment of human rights) of anti-money laundering and anti-terrorist financing standards[footnote 175]. In May 2021 a different initiative was launched to examine “the dilemma of finding a balance between ensuring that [aid agencies] receive the funds they need to work and preventing finances from being diverted for terrorists’ use”[footnote 176].

3.54. Overall, having observed and commented on the work of the Tri-Sector Group for almost 3 years, I can say that it has now grown into a valuable presence, and one that has some impact internationally. It does not, regrettably, have an official web presence but it is to be hoped that this is merely a matter of time. Transparency and understanding will be served by explaining its approach, publishing outcomes, provide and signposting relevant resources.

3.55. Away from the Tri-Sector Group, NGOs are seeking to coordinate better on seeking common legal advice and assistance. I have commented on this in previous reports. There are two key issues – firstly, getting access (something that smaller charities struggle with); secondly, trying to minimise divergent advice being given to different NGOs about the same legislation.

3.56. Secondly, the Home Secretary agreed (in October 2020) to invite the creation of prosecutorial guidance on overseas aid agencies and proscribed groups[footnote 177], and wrote to the Director of Public Prosecutions (‘DPP’) accordingly. In August 2021 a detailed paper on potential guidance was sent to the DPP by NGO members of the Tri-Sector Group.

3.57. No guidance has yet been published, although I am pleased to report that draft guidance has now been produced with a view to eventual publication in 2022. This matters because, as I wrote in my first report[footnote 178], the UK, considered a world leader in aid, should not be tripped up by its own laws.

4. INVESTIGATING TERRORISM Introduction

4.1. This chapter is concerned with the investigative powers conferred by the Terrorism Act 2000 and other counter-terrorism legislation. These powers have greater reach than ordinary PACE powers for two broad reasons. Firstly, they are concerned with identifying “terrorists” and “terrorism” which do not require officers to have specific offences in mind when exercising the power. Secondly, they include powers (such as cordons, search warrants, and production orders) that are directed at the expediency of or value to a “terrorist investigation”[footnote 179] without being limited to a search for evidence of criminal offending.

4.2. Digital evidence is key to terrorist investigations and last year I considered the sufficiency of police powers when dealing with encrypted data. The government has accepted my recommendation to consider whether new or amended powers are needed for CT Police to compel disclosure of encryption keys in CT investigations[footnote 180].

4.3. In summary, the stop and search powers under the Terrorism Act 2000 are:

  • Section 43, a power to stop and search a person reasonably suspected to be a terrorist to discover whether he has in his possession anything which may constitute evidence that he is a terrorist.

  • Section 43A, a power to stop and search a vehicle which it is reasonably suspected is being used for terrorism, for evidence that it is being used for such purposes.

  • Section 47A, a no-suspicion power that can only be used in extremely limited circumstances.

Section 43 and 43A

London

4.4. Figures for the use of section 43 were previously published only for the Metropolitan Police Service area but have now been published annually from November 2021 for all forces in England and Wales.

4.5. In 2020 524 people were stopped and searched by the Metropolitan Police Service (compared to 663 in 2019). This represents a decrease of 21%. There were 57 arrests (not necessarily for a terrorism offence) following a section 43 stop and search, down from 65 in 2019 (a 5% decrease)[footnote 181]. The arrest rate was 11%, which is comparable with previous years. There has been a long-term decline in the use of section 43 (there were 1052 stops and searches under section 43 in the year ending 31 Dec 2011).

4.6. The self-defined ethnicity of those stopped under section 43 in London since 2010 is as follows[footnote 182]:

Year ending Dec White Asian Black Chinese/ Other Mixed/not stated Total
2010 43% 30% 11% 7% 9% 999
2011 35% 37% 9% 8% 11% 1052
2012 39% 31% 12% 7% 11% 614
2013 34% 32% 14% 9% 10% 491
2014 41% 22% 12% 9% 16% 394
2015 30% 27% 13% 10% 21% 521
2016 29% 27% 11% 12% 21% 482
2017 30% 27% 14% 7% 22% 776
2018 25% 26% 16% 13% 19% 643
2019 29% 23% 11% 10% 27% 663
2020 26% 21% 11% 7% 34% 524

4.7. There was an increase in stops of those people in the category “Mixed/not stated” in the latest year. This may reflect an increasing reluctance for individuals who are stopped and searched to define themselves by reference to their ethnicity. There was a reduction in stops of people who self-identified as “Asian”.

Northern Ireland

4.8. In Northern Ireland in 2020[footnote 183]:

  • 22 people were stopped and searched under section 43 of the Terrorism Act 2000, down from 26 in the previous year.

  • A further 1 person was stopped under section 43A, down from 4 in the previous year.

  • 4 people were stopped and searched under sections 43 and 43A (8 were stopped in 2019), and 3 under sections 43/43A in combination with the special security powers available in Northern Ireland under the Justice and Security (Northern Ireland) Act 2007. As with previous years, by far the most stops in Northern Ireland are under the 2007 Act[footnote 184].

Section 47A

4.9. The no-suspicion power under section 47A was not authorised for use in the year 2020. It has only been used five times in the United Kingdom: once in Northern Ireland (2013) and four times in England (2017), in the circumstances described in Terrorism Acts in 2018[footnote 185]. The government has accepted my recommendation for improvements in the way section 47A is used in response to an act of terrorism, with better supervision and training aiming at greater consistency and has committed with CT Police to carry out a review of the 2012 Code of Practice.

Cordons

4.10. Section 33 of the Terrorism Act 2000 gives police officers of at least the rank of superintendent the power to authorise the use of a cordon in an area where it is considered expedient to do so for the purposes of a terrorist investigation. A police officer may order a person to leave cordoned areas, and prohibit pedestrian or vehicle access, and it is an offence to fail to comply with such a requirement.

4.11. Statistics for cordons are now reported by calendar year for both Great Britain[footnote 186] and Northern Ireland[footnote 187]:

Police force 2019 2020
Avon & Somerset - -
Cheshire - -
Cumbria - -
Dyed-Powys - -
Gloucestershire - -
Greater Manchester 4 2
Lancashire 2 1
Leicestershire - -
City of London 1 1
Merseyside - -
Metropolitan Police Service 1 1
North Yorkshire - -
Nottinghamshire - -
South Wales - -
Thames Valley - -
West Midlands - -
England and Wales 8 5
British Transport Police - -
Scotland - -
Northern Ireland 12 18
United Kingdom 20 23

4.12. In my first annual report[footnote 188], I recommended that the power to authorise searches within cordons should only be exercised in urgent cases. The government agreed with this recommendation and committed to amending the legislation at the next available opportunity. This change has yet to take place.

Search warrants

4.13. Paragraph 1 of Schedule 5 to the Terrorism Act 2000 provides a power for a magistrates’ court to authorise entry, search, and seizure of anything likely to be of substantial value to a terrorist investigation. This power can be exercised without the need for suspicion of a specific offence. A search can be authorised by a Superintendent in cases of urgency. Other parts of Schedule 5 deal with access to excluded or special procedure material where there are reasonable grounds for believing that there is material on the premises that is likely to be of substantial value (whether by itself or taken together with other material) to the investigation.

4.14. There are no published statistics for the use of this power in Great Britain. In Northern Ireland in 2020, 114 premises were searched under warrants granted pursuant to Schedule 5 (down from 201 in 2019)[footnote 189].

Production Orders

4.15. Paragraph 5 of Schedule 5 enables a court to require the production of material of substantial value (whether by itself or taken together with other material). There is no urgency provision for authorisation to be granted by a police officer. Other production order powers exercised in connection with terrorist exist under PACE and the Proceeds of Crime Act 2002.

4.16. In 2019, 497 production orders were obtained by CT Police under all statutory powers. In 2020, the total was 342.

4.17. In last year’s annual report I considered production orders made against journalists in connection with terrorist investigations[footnote 190]. The government is considering practicalities of my recommendation that all first instance judgments on applications for journalistic material under Schedule 5 Terrorism Act 2000 should be published.

Post-charge questioning

4.18. Power is conferred by sections 22 to 26 of the Counter-Terrorism Act 2008 to question a suspect post-charge, in exceptional circumstances, in relation to persons in detention charged with terrorism offences. Failure to answer questions may give rise to adverse inferences being drawn at trial. I reported on detail in this power, which has been exercised only 4 times to date[footnote 191]. It was not exercised in 2020.

Financial Investigations

4.19. Financial investigators within the National Terrorist Financial Investigation Unit play an important supportive role in most terrorist investigations.

Disclosure Orders

4.20. Powers exist to apply for disclosure orders in cases of “terrorist financing investigations” under Schedule 5A to the Terrorism Act 2000, and in respect of disclosures already made by the regulated sector, for further information orders under section 22B of the Terrorism Act 2000. Once a disclosure order is made, CT Police may serve individual notices to provide information under the authority of that order. A requirement to disclose financial information is a routine condition imposed by TPIMs and could be a requirement of a Serious Crime Prevention Order.

4.21. No official statistics are published on the use of disclosure orders. Approximately 5 disclosure orders were made in 2019; 4 orders were made in 2020, leading to 8 notices to provide information. A further 14 disclosure orders were made under the Proceeds of Crime Act 2002 which had some connection to a counter- terrorism investigation, with 41 notices served[footnote 192].

4.22. In Dec 2020 government published a new terrorist-financing risk assessment which stated[footnote 193] that

  • Terrorist finance activity in the UK remains varied and typically low-level in scale. There is no one method of financial activity associated with terrorism.

  • The raising and movement of funds are not considered to be the primary aim for terrorists.

  • Instead, terrorist finance activity continues to be for the purposes of sending small amounts to associates located abroad or for funding low-cost attacks.

  • Recent attacks in the UK have not required external fundraising, using low- cost, low-sophistication methodologies.

Customers Information Orders, Explanation Orders and Account Monitoring Orders

4.23. Customer information orders may be granted under paragraph 1 of Schedule 6 to the Terrorism Act 2000 in connection with financial information. There are no statistics available to me as to how often this power is used. The same is true of explanation orders under paragraph 13 of Schedule 5, which may require a person to provide an explanation for material seized under warrant or produced in response to a production order. National CT Police Headquarters have however been gathering these statistics since 2021.

4.1. As I have reported in previous annual reviews, account monitoring orders under paragraph 2(1) of Schedule 6A to the Terrorism Act 2000, which require financial institutions to supply bank account information for a specified period, appear to be widely used. I have been informed by National Counter Terrorism Policing Headquarters that in 2020, 238 account monitoring orders were made under the Terrorism Act 2000 and the Proceeds of Crime Act 2002. Disaggregated figures will be available next year. On average 69 account monitoring orders were active in any one month during 2020. A total of 29 TACT production orders were made with added account monitoring orders.

Suspicious Activity Reports

4.24. A source of information which companies and individuals are duty bound to provide are Suspicious Activity Reports (known as SARs) under the Terrorism Act 2000. Failure to comply with this duty to report is a criminal offence, punishable by up to 5 years’ imprisonment. The NCA publishes statistics on a financial year basis[footnote 194] which shows a sectoral breakdown of those making reports which are then considered by the National Terrorist Financial Investigation Unit or Counter-Terrorism Units: banks are by far the most likely to report suspicions which are subsequently disseminated for consideration based on a potential link to terrorism[footnote 195].

4.25. For 2020 I have been provided with the following statistics by National CT Policing Headquarters (figures from last year’s report in brackets[footnote 196]):

  • Terrorism Act 2000 SARs disseminated for assessment: 943 (955).

  • Defence against Terrorist Financing SARs disseminated for assessment: 378 (368).

  • Proceeds of Crime Act 2002 SARs identified as potentially relevant to terrorism and disseminated: 408 (502).

  • Defence against money-laundering Proceeds of Crime Act 2002 SARS disseminated for assessment: 116 (no figures for 2019).

5. ARRESTING AND DETAINING Introduction

5.1. In England, Wales and Scotland, the vast majority of arrests of suspected terrorists continue to be made under ordinary arrest powers: the Police and Criminal Evidence Act 1984 in England and Wales, and the Criminal Justice (Scotland) Act 2016 in Scotland.

5.2. There is also a special terrorist arrest power under section 41 of the Terrorism Act 2000, which may be exercised by police throughout the United Kingdom. Unlike ordinary arrest powers, it does not require the arresting officer to have a specific offence in mind and enables the suspect to be held for a much longer period of time in pre-charge detention (up to 14 days)[footnote 197].

Arrests in 2020 under Section 41

5.3. In Great Britain there were 26 arrests made under section 41 of the Terrorism Act 2000[footnote 198]. This is a decrease of 18 compared with the 44 arrests made in 2019. Arrests made under section 41 represented 14% of the total “terrorism-related arrests”, of which were 185 in 2020 (97 fewer than in 2019, a decrease of 48%). This is the lowest number of arrests for terrorism-related activity in the last ten calendar years. Of the 185 terrorism-related arrests, I have been informed by CT Police that roughly 30 were arrests on suspicion of terrorism offences by Right Wing Terrorists.

5.4. In Northern Ireland there were a total of 79 arrests made under section 41 of the Terrorism Act 2000 (a reduction of 73 from the previous year)[footnote 199]. I consider further in Chapter 9 the question of why Northern Ireland continues to make so many more arrests under section 41 than England, Wales, and Scotland.

Periods of detention in 2020

5.5. In Great Britain, of the 26 people arrested under section 41 of the Terrorism Act 2000[footnote 200]:

i. Just 12% were held in pre-charge detention for less than 48 hours (after which time a warrant for further detention is required from the court). This compares to 16% in 2019, 11% in 2018, 33% in 2017, and 14% in 2016.

ii. 85% were held for less than a week, which is comparable with the average of 88% since 2001.

iii. 4 people were detained beyond a week (down from 11 last year): these detentions were for 11 days (3 people) and 12 days (1 person).

5.6. As with 2018 and 2019, the maximum period of 14 days was not reached, reinforcing the view I have previously expressed that there is no indication that the current maximum period is proving insufficient.

5.7. The government has accepted my recommendations in last year’s report that section 41 should be amended so that where a person is first arrested under PACE and then the detention is ‘flipped’ to section 41 detention, the clock for TACT detention should start to run, in most cases, from the time of first arrest; and that the detention clock should not start to run where an individual is arrested under section 41 whilst being treated in hospital, and not questioned.

5.8. As I explained last year, in England and Wales all applications for warrants for further detention are currently made before a “judicial authority” under Schedule 8 Terrorism Act 2000, in practice designated District Judges at Westminster Magistrates’ Court in London via video link from TACT suites. In 2019 there were 40 individuals made subject to warrants of further detention. For the year under review, I have been provided with a slightly different set of statistics: there were 26 individuals detained under section 41 (as set out above) and the overall number of warrants of further of detention sought and obtained was 28 (not necessarily for each individual; some individuals were not subject to an warrant of further detention, and some were subject to more than one)[footnote 201]. There were no refusals.

5.9. Information in support of applications for a warrant of further detention is given by a police superintendent who is not directly involved in the investigation. As a result of recommendations made by the former Chief Magistrate during 2020, steps have been taken to ensure that the relevant police superintendent is better informed about its progress. The purpose is to enable speedier and more complete evidence on the statutory conditions for further detention: that there are reasonable grounds for believing that further detention is necessary and that the investigation is being conducted diligently and expeditiously[footnote 202].

5.10. As I reported last year, there are no statistics on the success rates for warrants for further detention in England and Wales. I recommended that this data should be published. This recommendation was accepted by the government.

5.11. In Northern Ireland of the 79 people detained under section 41 of the Terrorism Act 2000, 66 were held for 48 hours or less[footnote 203]. The remainder (13) were held for over 48 hours following applications for warrants for further detention, all of which were successful[footnote 204]. No one was held for more than 7 days[footnote 205].

Conditions of pre-charge detention under Schedule 8

5.12. Detention under Schedule 8 is governed by a special Code of Practice, Code

H. In practice, individuals arrested under section 41 are taken to TACT suites which have an unusually high staff-to-prisoner ratio. Reflecting the importance of terrorist investigations, the desirability of admissions not being excluded because of unfairness or oppression, and the very lengthy periods of effectively solitary detention that Schedule 8 allows (up to 14 days), great care is taken to assess and meet the needs of detainees.

5.13. Since January 2017 the Independent Reviewer of Terrorism Legislation has been officially designated as part of the United Kingdom’s National Preventive Mechanism. As a signatory to the Optional Protocol to the Convention Against Torture and other Cruel, Inhuman or Degrading Treatment or Punishment, the United Kingdom is obliged to have a mechanism in place to review conditions of detention including terrorism detention.

5.14. As part of my role, I review all the visit reports filed by Independent Custody Visitors, attend periodic meetings with those who organise the TACT custody visitors schemes, and routinely speak to custody officers at TACT suites. Although the pandemic inevitably had an impact on the ability of custody visitors to visit TACT suites in person, by May 2020 all TACT custody visiting schemes in England and Wales were either visiting in person or had set up remote monitoring to allow direct contact between the detainee and the custody visitor. TACT custody visiting continues to play an important role in ensuring high standards in the treatment of detainees.

Young detainees

5.15. As the statistics below demonstrate, more and more juvenile detainees are being arrested for terrorism. It has been drawn to my attention that, paradoxically, the care and flexibility shown by TACT suites results in practical dilemmas if the juvenile is charged, bringing detention under Schedule 8 to an end, but not granted bail to attend court.

  • In this situation, the general rule is that the detained juvenile shall be moved to non-secure local authority accommodation, at which point the juvenile is formally detained by a person acting on behalf of the authority[footnote 206]. In practice, the local authority must then decide what accommodation is appropriate which could mean the juvenile being accommodated with friends or family, foster families or in a children’s home[footnote 207].

  • The circumstances in which secure local authority accommodation, shamefully lacking in England and Wales[footnote 208], will be made available are very limited. The custody officer must believe that this child poses a risk of serious harm to the public between being charged and appearing at court. This is a described as

“a very high bar” for a child to meet[footnote 209]. It is unlikely to be satisfied in many of the most common cases involving possession of terrorist information.

  • I am confident, having spoken to a range of police officers involved in TACT detention of juveniles, that TACT detention can provide a calm and predictable environment in which, for example, favourite foods and DVDs are provided. Sadly, the overall standards of care can contrast favourably with home life. Assuming secure accommodation is not appropriate, leaving TACT detention for temporary accommodation with a foster family – perhaps involving lengthy travel, settling in, eating, sleeping, and then more travel to court the following day – may amount to an unwelcome and potentially distressing change of scene.

  • Officers do have a narrow option to keep a child in TACT detention following charge, where they are able to certify that transfer to local authority accommodation is “impracticable”[footnote 210]. This is interpreted as meaning that “exceptional circumstances render movement of the child impossible” or that the child is due at court in such a short space of time that transfer would deprive them of rest or cause them to miss a court appearance, to be judged on a case- by-case basis[footnote 211]. Although Code C does not apply post-charge, the general rule is that a detainee must be allowed 8 hours continuous rest within a 24 hour period[footnote 212].

  • Custody officers may sometimes be able to make use of this narrow statutory exemption. However, this does not amount to a general welfare provision. Neither law nor principle allows police officers to determine that continuing detention in a TACT suite as opposed to local authority accommodation is in the child’s best interests.

  • I will continue to monitor this situation: I make no recommendation but invite those involved in TACT detention of juveniles to draw any difficult cases to my attention so that I can continue to review the adequacy of the law in this field.

Arrest Outcomes

5.16. Officials refer to the rate of terrorism arrests which lead to terrorism charges as the “conversion rate”. Given that the special and highly restrictive arrest and detention regime created by section 41 and Schedule 8 of the Terrorism Act 2000, particular attention should be given to the extent to which arrests under section 41 led to people being prosecuted, or merely released without charge, although as noted above, the majority of terrorism-related arrests in England, Wales and Scotland are carried out using general arrest powers.

5.17. A lower conversion rate may justifiably result in the wake of large scale attacks, such as the bombing of the Manchester Arena in 2017, where a large number of suspects who are believed to be linked to the plot or the attacker may be arrested in urgent circumstances[footnote 213].

Numbers charged in 2020

5.18. In 2020 of the 185 “terrorism-related arrests” in Great Britain 56 people (30%) were charged with an offence[footnote 214]. This compares to a charge rate of 38% for “terrorism- related arrests” in 2019, 48% in 2018 and 39% in 2017. Of the 26 people arrested under section 41 of the Terrorism Act 2000, 13 were charged (50%). The significantly higher charge rate no doubt reflects the intelligence-led nature of many such arrests.

5.19. Of the 185 people who were arrested, 40 (22%) were released without being charged and 73 (39%) were bailed and returned and released under investigation[footnote 215]. Alternative action is recorded as having been taken in 13 cases (7%), which included 11 recalls to prison. This is the second highest number of prison recalls since the Terrorism Act 2000 was enacted.

5.20. Of the 56 people charged with an offence having been arrested on suspicion of committing a terrorism-related offence in 2020, 46 were charged with an offence under terrorism legislation. 2 people were charged with terrorism-related offences

(other than those contained in terrorism legislation) and 8 were charged with non- terrorism related offences[footnote 216].

5.21. The principal offence[footnote 217] for which persons were charged under the Terrorism Acts in Great Britain was the collection of information useful to an act of terrorism (16 persons). The other offences were dissemination of terrorist publications (8 persons), failure to notify changes (6 persons), failure to comply with a port examination (4 persons), breach of a TPIM (3 persons, the joint highest figure since the reporting period beginning 2002), preparation for terrorist purposes (3 persons), encouragement of terrorism (2 persons), membership of a proscribed organisation (2 persons), possession of an article for terrorist purposes (1 person), breach of a foreign travel restriction order (1 person).

5.22. As I explained last year, because Home Office statistics only refer to the principal offence with which an individual is charged, these figures must be approached with caution because it is likely that the figures do not reflect the number of “lesser” charges. For example, if an individual is charged with a serious offence such as preparation for terrorist acts, but also with collection of terrorism information found on his computer at the time of his arrest, only the former offence will be recorded.

5.23. Of the 79 arrests made under section 41 of the Terrorism Act 2000 in Northern Ireland only 14 people were charged with an offence[footnote 218]. This is a charge rate of 18% following section 41 arrest.

Gender, age, ethnicity and nationality

5.24. The Home Office publishes detailed figures for the gender, age, ethnicity and nationality of those subject to terrorism-related arrest, charge and conviction in 2020. No such figures are published in Northern Ireland.

5.25. Women comprised 10% of “terrorism related arrests” in 2020, 8% of those charged with terrorism-related offences and 10% of those convicted after charge[footnote 219].

5.26. In terms of age, the figures for Great Britain in 2020 are as follows (previous year’s figures in brackets)[footnote 220]:

2020 Under 18 18-20 21-24 25-29 30 and over
% of terrorism-related arrests 10% (4%) 11% (10%) 11% (11%) 18% (17%) 49% (57%)
% of terrorism-related charges 6% (8%) 10% (15%) 17% (15%) 23% (16%) 44% (46%)
% of terrorism- related convictions 0% (10%) 10% (15%) 20% (12%) 35% (23%) 35% (40%)

5.27. The number of terrorism-related arrests of juveniles continued to rise. During 2020 the arrest figures for all age groups fell compared to 2019, with one exception: the number of juveniles arrested actually rose from 12 (2019) to 19 (2020). The phenomenon is not confined to Great Britain – the Intelligence and Security Committee noted the sustained attachment of young people to dissident republican causes in Northern Ireland[footnote 221]. CT Police have informed me that roughly half of the 19 juvenile arrests were for suspected Right Wing Terrorism offences.

5.28. As for ethnic appearance the figures (based upon officer-defined data) for Great Britain in 2020 are as follows[footnote 222] (figures from 2019 in brackets):

2020 White Black Asian Other Not known
% of terrorism-related arrests 48% (41%) 8% (8%) 34% (39%) 10% (11%) 0% (0%)
% of terrorism-related charges 46% (45%) 15% (10%) 33% (33%) 6% (13%) 0% (0%)
% of terrorism-related convictions 25% (48%) 15% (13%) 50% (23%) 10% (15%) 0% (0%)

5.29. In terms of self-defined nationality, British citizens (including those with dual nationality) comprised 81% of those arrested for terrorism-related offences in 2020 and 81% of those charged with such offences[footnote 223].

5.30. As I reported last year, no statistics are published for the self-defined, or officer- defined, religion of those arrested.

6. STOPPING THE TRAVELLING PUBLIC Introduction

6.1. Schedule 7 to the Terrorism Act 2000 allows police officers to stop and question (“examine”) members of the travelling public at ports and borders to determine if they are terrorists; to search them; to detain them; to require them to hand over their electronic devices for examination and copying; and to take their fingerprints and DNA. Failure to cooperate with an examination is a criminal offence.

6.2. How Schedule 7 operates is driven by world events – in 2020, the global pandemic which dramatically curtailed ordinary travel, plus shifting focus on different theatres of international conflict. Until recently the focus was Da’esh controlled territory; future years may show greater attention to travel to Afghanistan.

6.3. Because this exceptional and intrusive power may be exercised without the need for reasonable suspicion, my approach has been to review the Schedule 7 power with some granularity. In my exchanges with CTP Borders, the part of the CT network whose officers are trained to use the power, I have pushed on various points of detail: in this year’s report, particular reference is made to the obtaining and retaining of digital data (generally from phone downloads) whose governance remains imperfect.

6.4. Schedule 7 is also a counter-terrorism power that elicits particular concern amongst a significant minority[footnote 224]. In this context perceptions of unfair targeting or victimisation extend not just to formal exercise of the Schedule 7 power but to the interactions that often precede it[footnote 225].

6.5. Having carried out my review, I remain of the overall view, as expressed in earlier reports, that the Schedule 7 power is exercised conscientiously and effectively. I have been particularly impressed at the self-critical approach of senior management within CTP Borders.

6.6. A new Code of Practice on the operation of Schedule 7 was brought into effect in

August 2020: I considered the relevant changes in last year’s report[footnote 226].

6.7. At the time of writing the government has proposed amendments to Schedule 7 to enable the power of examination to be exercised away from the place of arrival for those who arrive irregularly by sea and are detained under the Immigration Acts[footnote 227]. The proposed power would extend to those who arrive by sea (whether small boats, or hidden on a container or lorry transported by ferry) but not via the Channel Tunnel.

6.8. In principle, people arriving irregularly in the United Kingdom should be liable to counter-terrorism examination as much as those arriving at sea ports and airports. In 2020, an individual who had previously been excluded from the UK on grounds of national security arrived back in the UK on a small boat[footnote 228]. Promisingly, the proposed power to examine is strictly tethered to the time of arrival (meaning that the person must have been apprehended within 24 hours of arrival) rather than being a free- floating power that could be used to examine anyone who may have arrived irregularly at any time. But attention will need to be given to the personal condition of migrants on arrival and in detention[footnote 229], and to the interplay between the offence of illegal entry[footnote 230] – during the investigation of which the suspect has a right to silence – and the obligation to answer questions under Schedule 7 at risk of criminal penalty[footnote 231].

Frequency of use

Great Britain

6.9. Before 2020 there had already been a significant decline in the number of Schedule

7 examinations in Great Britain. In the period 2010/11, the year data was first published, there were 65,684 stops[footnote 232]. In 2018 there were 11,876 stops and in 2019 there were 9,540 stops. Since the year ending December 2012, this represented a decrease of 84%.

6.10. Unsurprisingly, there was a dramatic falling-off in 2020: 3,315 examinations, a 65% decrease from the previous year, and undoubtedly attributable to the restrictions on international travel resulting from the pandemic. It is impossible to draw wider conclusions from 2020, and further analysis of long-term downwards trend will probably not be possible until data is available for a full 12 months in which travel is uninhibited.

6.11. It is however likely that the introduction of hand-held devices which allow examining officers quick and easy access to relevant information are increasingly allowing officers to screen out some individuals, so that interactions do not lead to formal examinations.

Northern Ireland

6.12. In Northern Ireland in 2020 there were 119 examinations. This represents a 79% decrease from the previous year. There were 8 detentions, down from 31 in 2019.

Intra-UK

6.13. The government now publishes statistics on the number of Schedule 7 examinations which involved intra-UK travel (that is travel between one UK port and another UK port, so journeys between or within England, Wales, Northern Ireland and Scotland)[footnote 233]. In 2020 the number of intra-UK examinations was 591[footnote 234].

Utility

6.14. A month-long survey of CTP Borders on the utility of Schedule 7 closed in May 2021, with evaluation underway at the time of drafting this report. I was informed that since April 2020, 10 terrorism arrests have been directly attributable to Schedule 7 examinations[footnote 235], and that ‘cold’ or intuitive stops have led to the identification of important Subjects of Interest.

6.15. As I reported last year, CT Police HQ (‘CTPHQ) now have an enviable set of data from ports around the United Kingdom, enabling some early evaluation of how often a Schedule 7 examination leads to an arrest, detention or seizure. Of course, as I have stressed to officials, arrest for a non-terrorism offence (for example, possession of drugs) does not prove the utility of a counter-terrorism power, whilst the fact that a person is detained may simply reflect the rule (which I have criticized in an earlier report[footnote 236]) that any examination that lasts longer than an hour must result in detention.

6.16. It follows that CTPHQ must strive to identify the most robust measures of success, for example by analysing the percentage of examinations that lead to intelligence reports with relevance to national security. As I reported last year, analysis of less thorough data from 1 July 2018 to 30 June 2019 indicates that relevant intelligence reports were filed in roughly half of tasked, and one fifth of untasked, Schedule 7 examinations. Analysis for the calendar year 2020 that was conducted at my request shows that tasked stops led to relevant intelligence reports in four-fifths of cases, and untasked stops to relevant intelligence reports in one third of cases: on its face, a substantial increase in the intelligence yield for both types of examinations.

6.17. Analysis also shows that the numbers examined naturally vary widely within the United Kingdom (for example it is used far more in the London region than the South-West) and that there are significant differences between regions in terms of the detention of those who are examined, and the percentage of data downloads and biometric captures that are carried out, and national security intelligence reports generated.

  • In some instances, CTPHQ have been able to explain apparent discrepancies because of recording failures (for example, based on a misunderstanding as to when a Schedule 7 examination begins); other discrepancies may be explicable on the basis of the very different conditions under which sea ports and airports operate[footnote 237].

Advance Information

6.18. Schedule 7 yields the most intelligence and occasionally evidence when exercised on the basis of or in conjunction with advance information.

6.19. During 2020 there were some modest improvements made in connection with the obtaining of Advance Passenger Information (API) and Passenger Name Records (PNR) across the United Kingdom, including for travel within the Common Travel Area (CTA)[footnote 238]. API is useful for watch listing known terrorist suspects; PNR provides a depth of information which is useful to identifying new ones.

6.20. In an earlier report I warned against the temptation of using statutory powers conferred for different purposes[footnote 239] to make up for the absence of a duty to show passports in the CTA[footnote 240]. Facial recognition at borders is becoming more routine and the technology offers one solution to knowing who is entering and leaving the country.

Detentions

6.21. In the year under review, there were 1,191 detentions under Schedule 7[footnote 241]. This is 43% drop from the figure in 2019 (2,082), but, as the table below shows, the rate of detention has now increased to 36% of those examined. Whilst it is important not to draw too many conclusions from an unusual year, every indication is that detention is becoming so frequent that, at the present rate, it will become the norm within a few years. Only 20 individuals were detained who were examined for less than one hour[footnote 242].

6.22. Internal statistics show that tasked stops are far more likely to lead to detentions than untasked stops.

2016 2017 2018 2019 2020
Examinations 19,355 16,349 11,876 9,543 3,315
< 1 hour 17,857 14,703 10,131 7,548 2,144
> 1 hour 1,498 1,646 1,745 1,995 1,171
Detained 1,539 1,700 1,836 2,082 1,191
% detained 8% 10% 15% 22% 36%

6.23. Under Article 4 of the Optional Protocol to the Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, to which the United Kingdom has been a party since 2003, visits must be allowed by inspection bodies (known collectively as the National Preventive Mechanism) to any place under a state’s jurisdiction and control where persons are or may be deprived of their liberty[footnote 243].

  • A place specially designated for detention at a port[footnote 244] certainly falls within the scope of Article 4, and work is currently being done to determine how detention can be monitored by the Independent Custody Visiting Scheme (or some other body) whether routinely or, more likely, on an ad hoc basis.

Conduct of Examinations

6.24. The impact of Schedule 7 on innocent members of the travelling public puts a premium on the power being exercised with consideration and transparency. Research suggests, and I agree, that using “procedural justice” can mitigate some of the negative effects of perceptions and experiences of Schedule 7 stops[footnote 245]. Two years after my first report, I am told that schedule 7 training still uses the mnemonic ‘PARTICIPATE’ as a way of teaching procedural justice[footnote 246]; this is to be commended. The government accepted my recommendation that training materials should remind officers that, despite the additional powers for hostile state examinations under Schedule 3 Counter-Terrorism Borders and Security Act 2019 (which I consider below), Schedule 7 does not authorise the use of journalistic or legally privileged material[footnote 247].

6.25. Although the government rejected my earlier recommendation[footnote 248] that the Code of Practice should be amended to require officers to consider whether inbound examinations may be as effective as outbound examinations, the Code explicitly requires officers to seek to minimise travel disruption wherever possible[footnote 249], and the need to consideration the effectiveness of inbound/ outbound examinations is now contained in internal operational guidance.

6.26. Another rejected recommendation concerned the possibility, disconcerting to the person stopped, of multiple examinations over a period of time[footnote 250]. I am informed that rather than recording the number of previous examinations on Port Circulation Sheets, CTP Borders have put in place mechanisms, involving the joint team of MI5 and CT Police officers[footnote 251], to ensure that details of previous examinations are always provided to frontline officers. Whilst previous examinations should not prevent subsequent examinations as circumstances change, repeat examinations should not happen through ignorance.

6.27. There is still no systematic monitoring of complaints, and I recommend that information on complaints about the exercise of Schedule 7 is routinely captured from all police forces across the United Kingdom. Whether or not complaints are upheld, at the very least they provide useful concrete jumping-off points for my own analysis of the use of the power.

6.28. I was able to establish that there were 4 complaints made to the Metropolitan Police Service, whose border officers cover the busy London airports, during 2020 which concerned allegations of discriminatory questioning, unlawful taking of photograph and biometrics and abuse of the power of search. There were no court cases relating to the conduct of Schedule 7 examinations during 2020.

6.29. The power to restrict or postpone the right to consult with a solicitor was not exercised in 2020[footnote 252].

Ethnicity of those examined

6.30. The collection of ethnicity data for Schedule 7 stops has been carried out on a self-definition basis since April 2010. In last year’s report I provided the figures for Great Britain for the past 7 years. So that a comparison may be made, below are the figures for the years 2018 to 2020[footnote 253]:

Great Britain

Total examinations

2018 2019 2020
White 24% 23% 18%
Mixed 6% 5% 5%
Black 7% 7% 8%
Asian 25% 26% 26%
Chinese or other 23% 27% 26%
Not stated 15% 12% 18%

6.31. These figures are remarkably stable save for a drop in those self-identifying as White and an increase in those preferring not to say[footnote 254]. It follows that, as was also noted in last year’s report:

  • It is a reasonable inference from these ethnicity figures that the main use of Schedule 7 powers is to detect Islamist terrorism which continues to be the principal threat within Great Britain.

  • The increasing number of arrests for Right Wing Terrorism is not reflected in an increase in Schedule 7 examinations of white people. This is despite research showing that Right Wing “extremists” (a term that does not mean terrorist or potential terrorist) are more globally networked than ever[footnote 255], although I would hesitate to conclude from this that the Schedule 7 power is being misused, and without further understanding of how the scale international travel by potential Right Wing terrorists and travel for potential Right Wing terrorist purposes (for example to a ‘Blood & Honour’ type event overseas) compares to the scale of travel by potential Islamist terrorists and travel for potential Islamist terrorist purposes (for example, in its heyday, to territories controlled by Da’esh)[footnote 256].

6.32. The greatest risk of unlawful discrimination arises in context of intuitive stops, particularly, but of course not only, at seaports in the Common Travel Area where patchy advance information makes targeting more difficult. Examining officers may have little to go on whilst watching a flow of traffic disembarking a busy vehicle ferry after dark.

6.33. In all circumstances, officers must ensure, as required by the Code of Practice[footnote 257], that ethnicity is not used as a criterion for selection simply because the presence of a passenger of colour is more unusual on a particular route. Under the Code, ethnicity can only be used as a criterion for selection to the extent that it is used in considerations that relate to the threat from terrorism: for example, where a travel route is believed to be exploited by terrorists from a particular part of the world.

6.34. To safeguard against the risk of unchecked unconscious bias occurring on the part of examining officers where the Schedule 7 power is used intuitively, I recommend that the National Counter-Terrorism Policing HQ should analyse ethnicity categories for those subject to tasked examinations compared to untasked examinations.

6.35. A further refinement, noted in Chapter 1, is the possibility of publishing “White Irish” as a separate category. The fact that CT Police HQ spotted that examinations of White Irish people were underproducing intelligence dividends is an excellent example of how statistics can provide an opportunity for self-challenge. But it is even better if those using the Schedule 7 power can be publicly accountable as well, hence my recommendation in Chapter 1 not just to keep collecting, but to publish these figures.

Detentions

6.36. The ethnicities of those detained are broadly comparable to the two previous years[footnote 258].

2018 2019 2020
White 11% 13% 12%
Mixed 7% 6% 5%
Black 11% 7% 9%
Asian 28% 30% 26%
Chinese or other 29% 29% 28%
Not stated 14% 15% 20%

Northern Ireland

6.37. The self-defined ethnicity of those examined and detained in Northern Ireland provided to me by PSNI is as follows:

2019 2020
White 55% 38%
Mixed 10% 8%
Black 4% 13%
Asian 13% 16%
Chinese or other 10% 17%
Not stated 7% 8%

Detentions

2019 2020
White 13% 13%
Mixed 19% 13%
Black 6% 13%
Asian 26% 25%
Chinese or other 23% 0%
Not stated 13% 38%

Digital Downloads

Obtaining

6.38. New internal CTP Borders guidance on taking digital downloads was circulated in January 2020.

6.39. Figures on the number of media downloads are not published, but internal CTP Borders statistics show that:

  • a much higher percentage of tasked stops lead to media downloads than untasked stops. So even though the number of untasked stops is far greater than the number of tasked stops, in 2020 tasked stops led to three-quarters of all digital downloads.

  • There is a strong correlation between the percentage of Schedule 7 examinations that lead to digital downloads and those that lead to detention: this may reflect the time it takes to carry out a download coupled to the statutory obligation to detain once the examination has lasted longer than one hour[footnote 259].

6.40. Despite the importance of digital downloads, examining officers must avoid the risk of ‘conveyor-belt decision-making’. The fact that untasked stops in 2020 were less likely to lead to digital downloads does indicate that thought is being applied to whether the separate power to copy should be deployed. However, recent history[footnote 260] suggests that when budgetary pressures are applied to officer numbers, perverse incentives can be created to exercise powers in order to demonstrate that examining officers are productive. It is important that officers are not encouraged to use powers on a ‘use it or lose it’ basis.

6.41. Current legislation must continue to be measured against the evolving ways that modern technology stores information. This is particularly relevant to remote data.

6.42. There are a number of categories of remote data may be accessible from searched devices: (a) data that is pulled into apps from a third party data source (for example banking apps) (b) basic information which is held remotely as a design setting (such as on the Google Chromebook device[footnote 261]) (c) data that has been deliberately removed from the device to a remote location, typically for storage reasons.

6.43. Under Schedule 7:

  • Digital downloads are obtained using the power of search found in paragraph

8. By paragraph 8(1) in addition to the searching the person, an examining officer may search “anything which he has with him”. The power to copy under paragraph 11A arises in respect of anything found under paragraph 8.

  • The Code of Practice states (at paragraph 58) that “…The power to search anything which the person has with them includes the power to search electronic devices, such as mobile phones…Searching such a device may result in information being accessed which is stored other than on the device itself.” So, Code recognises that the functionality of modern devices may include data being accessible from the device which is stored permanently or temporarily elsewhere.

  • Qualifications to this general position on remote access are made in the Code (also at paragraph 58):

    • Firstly, access to remote data cannot be obtained using a device that is not the subject of the search. So, whilst an officer “…may access emails using an application on the phone being searched” he “…may not obtain the email username and password and log onto the email account from another computer”

    • Secondly, the Code provides that the power to search electronic devices cannot be used to intercept communications in the course of their transmission within the meaning of the Investigatory Powers Act 2016 with the important technical rider that this prohibition does not extend to the obtaining of stored communications by use of the Schedule 7 power[footnote 262].

6.44. Legally, the source of the power to search, and therefore to carry out digital downloads, is Schedule 7 itself and not the Code. Paragraph 58 of the Code is in effect an attempt to interpret and apply the statutory words “has with him” to the digital age. It is a possible, albeit contestable, interpretation that a person with a mobile phone has at least some data “with him” even though it is held remotely[footnote 263]. Such data may be accessible as part and parcel of the everyday use of the device: for example, the contents of emails which are pulled off the cloud when the user presses the header shown in the reading pane of the mail app.

6.45. The position is less straightforward for old data that is deliberately stored remotely and not routinely consulted if at all from the device, for example old digital photos which are held in the cloud for storage reasons. On one view, the cloud is simply an extension to the device’s memory; and since a laptop with terabytes of local storage could be searched under Schedule 7, the power of search should equally apply to data held on the cloud. But the words are “has with him” not “has available to him”. If the power applies to data that is merely available from a device then the nexus between the data and the device becomes less clear, since such data (for example, documents on Microsoft OneDrive) are presumably available from any device used by that person.

6.46. The fact that the Code addresses remote access is commendable from a transparency perspective, in addition to its function of telling examining officers what they can and cannot do[footnote 264]. It would be possible to suggest amending the Code to give further and better guidance on what “has with him” means in the remote storage context. This would require consideration being given to (i) data that is believed to be routinely accessed as part of the day to day functioning of the device and (ii) data that is not believed to be routinely accessed as part of the day to day functioning of the device, but which is nonetheless accessible from the device.

6.47. However, this would be to ignore the point that a power to obtain remote digital material in the absence of any suspicion is a strong and intrusive power, and strong and intrusive powers are not authorised by vague statutory wording[footnote 265]. The fact that the legislation does not permit a meaningful distinction between routine data and long- term stored data illustrates why the language is too vague in this context.

6.48. A clearer reading of paragraph 8 is that the power of search is limited to searching articles that physically accompany the person being examined and that it does not apply to remote data. A distinction can also be drawn between the Schedule 7 and the Police and Criminal Evidence Act 1984: only the latter contains reference to any sort of remote access (the power to require the provision of information in computer form that is “contained in a computer and accessible from the premises”)[footnote 266]. I therefore recommend that paragraph 8 of Schedule 7 should be amended to enable the proportionate searching and copying of remotely held data, to be accompanied by an amended Code of Practice which gives practical guidance on the limits of this power (for example, limiting searches to remote data that is believed to be routinely accessed from the phone).

6.49. In its report on search warrants, the Law Commission observed that remote storage was considered by a number of its consultees to be one of the biggest issues for the law of search warrants, and itself found that the treatment of remotely stored data was one of the most difficult aspects of its project[footnote 267]. It declined to reach a definitive conclusion on a new model for access for the search warrant regimes it was considering without further technical and cross-sectional input[footnote 268]. Whilst I acknowledge that remote access is an issue that is relevant to many powers of search, I do not recommend that the government should postpone the exercise I recommend until a solution can be found that fits all statutory regimes. Paragraph 8 of Schedule 7 is a power that needs to be considered on its own terms, and there is an imperative to act because of both its utility in protecting national security and its potential to reach deeply into the digital lives of the travelling public.

Retention

6.50. Members of the travelling public would be entitled to conclude that terrorism legislation, and the public policy documents created by CTP Borders, fail to provide an answer to this question: you have downloaded the entire contents of my phone – what are you keeping it for and for how long[footnote 269]? I have raised this issue in both of my previous annual reports[footnote 270] and noted that public transparency about the retention of digital downloads fulfils three purposes:

  • Legitimacy, by ensuring that the public understand the powers exercised in their name;

  • Fairness, so affected individuals know that they have been dealt with according to non-arbitrary standards and can hold officers to account;

  • Fair processing under the Data Protection Act 2018.

6.51. To their credit, CTP Borders recognise the importance of this issue. It is heartening that officers are conscious of this aspect of the impact of the Schedule 7 power, self-critical as to how they exercise it, and open about the difficulties of using it in the digital age.

6.52. The power to retain digital data for intelligence purposes derives from paragraph 11A of Schedule 7[footnote 271]. An examining officer may copy the contents of a phone or laptop and “the copy” may be retained “…for so long as necessary for the purposes of determining whether a person falls within section 40(1)(b)” (paragraph 11A(3)(a)). A person falls within section 40(1)(b) if they are a terrorist, i.e. they are or have been involved in the commission, preparation or instigation or acts of terrorism.

6.53. The retention power broadens the consequence of examination considerably: data may be examined for the purpose of determining whether “a person” (not necessarily the person examined, but possibly a family member or associate) is a terrorist; the examination need not be done by the examining officer or even a police officer, opening the door to the use of intelligence analysts.

6.54. The retention power is not unconfined: it must not be retained under this power for any other purpose; and “the copy” may only be retained for “as long as necessary” for the permitted purpose. Inherent in paragraph 11A(3)(a) is that some assessment is performed on the necessity of retaining that particular data.

6.55. Given this statutory power, it is doubtful that a common law power of retention exists in parallel; even if it did, it could hardly operate to the prejudice of the statutory constraints imposed by Parliament. In other words, a digital download could not be retained under common law powers for longer than was necessary for the purpose of determining whether a person was a terrorist.

6.56. This is reinforced by the Code of Practice which provides that the powers under paragraph 11A must be “…exercised in a manner which is proportionate to the legitimate aim”[footnote 272]. The reference to proportionality recognises that retaining copied data has an impact on the privacy of individuals, whether the owner of the phone, or the individual whose personal data (in the form, for example, of text messages or photos) happen to appear on that phone.

6.57. In the leading case of Beghal[footnote 273], the effect of paragraph 11A(3)(a) was not argued but Lord Hughes distinguished, obiter, between:

  • retaining data for the duration of the stop, and for a short period afterwards to compare records. This “would not appear to be disproportionate”.

  • indefinite retention “so as to provide a bank of data”. This “would seem to be a different matter”.

6.58. On this analysis, between these outer limits – retention for a short period after the examination, and indefinite retention – there is a period during which retention may or may not be proportionate/ retained for no longer than is necessary, depending on the particular facts.

6.59. The reality is that data is useful when retained in the medium and long term, as well as the short term. The value of intelligence obtained from Schedule 7 is often in the accumulation of individually small pieces of intelligence which, combined, may inform both particular and general responses to the terrorist threats confronting this country[footnote 274]. That value may not crystallise until many years after it was obtained, meaning that its value may not be apparent for some time. For example,

  • a non-resident’s travel from the UK may not be recognised as being for terrorist purposes until he is detained many years later returning from Syria;

  • an individual may be part of a sleeper cell whose activities only become apparent many years after first contact with the authorities[footnote 275];

  • its importance may only be apparent when combined with information from an overseas authority which is only made available many years later, for example in the wake of an overseas attack.

6.60. The recent overthrow of government of Afghanistan by the Taleban further illustrates the point. In the early 2000s travel to Afghanistan, often via Pakistan, was frequently linked to terrorist plots in the UK[footnote 276]. Information obtained in the early 2000s may prove to be very relevant to travel in the 2020s.

6.61. There is therefore a plausible argument that it is necessary to construct a bank of data for indefinite retention. Alternatively put, in order to find the needle, it is necessary to keep building the haystack.

  • The starkness of indefinite retention, leaving aside whether it could ever be lawful under the Data Protection Act 2018, would require the strongest of safeguards, some of which are already available to CT Police: for example, CT intelligence is held on separate national systems to which only trained and vetted police and staff have access for national security purposes.

  • It would be possible to improve the regime so that information is not processed any further than necessary, for example by establishing a system of prior authorisation, akin to the regime proposed under clauses 36 to 42 Police, Crime, Sentencing and Courts Bill and the draft Code of Practice on extraction of information from electronic devices[footnote 277]. Some elements of independent oversight could be achieved[footnote 278].

  • This has something in common with bulk data sets where the retention of data is not targeted at known individuals but enables the identification of subjects of interest at some stage in the future. The European Court of Human Rights has commented that for this type of data “Safeguards are therefore pivotal and yet elusive”[footnote 279].

6.62. But indefinite detention is not what paragraph 11A(3)(a) contemplates; otherwise, the words “as long as necessary” would be entirely surplus. In any event, retention of personal data for longer than is necessary is contrary to the Data Protection Act 2018[footnote 280].

6.63. The Code of Practice does not provide any clarity to the interested reader as to how the assessment of necessity takes place. It merely states that copies of information obtained during an examination must be managed in compliance with the requirements of “Management of Police Information guidance” (known as MOPI), General Data Protection Regulation provisions and the Data Protection Act 2018[footnote 281]. Anyone familiar with the latter two instruments will know that they are often impenetrable, even for lawyers, and therefore if there is any clarity is must be found in MOPI.

6.64. MOPI is a publicly available document and contains a detailed section entitled “Information management: Retention, review and disposal”[footnote 282]. In its current incarnation, this policy sets out three different review schedules for “…information related to people convicted, charged, arrested, questioned or implicated with an offence”, namely Group 1 (Serious offences and public protection matters), Group 2 (Other sexual and violent offences) and Group 3 (All other offences). For Group 1, material is to be retained until “the subject” has reached 100 years of age, followed by a manual review, with reviews every 10 years to ensure adequacy and necessity, with tighter controls or retention for information in Groups 2 and 3.

6.65. The previous iteration of MOPI had a Group 4 (intelligence products) but this category no longer exists[footnote 283]. Shorn of this group, the current version of MOPI is even more inapt for dealing with counter-terrorism intelligence. Information copied from phones examined under Schedule 7, a power that does not depend on reasonable suspicion, is not readily described as information relating to people “convicted, acquitted, charged, arrested, questioned or implicated with an offence”. Information is initially obtained to determine whether the person examined is a “terrorist” within section 41(1)(b) (which is not in itself an offence) and then retained in order to determine whether “a person” is a “terrorist”. MOPI by contrast provides that the review of whether to retain information “focuses on the offender”. There is no reference in MOPI to national security as a ground for retention.

6.66. CT Police currently regard Schedule 7 data as falling within Group 1 but recognise that MOPI is an uncomfortable fit. More worryingly, CT Police have been able to give me very little assurance that CT intelligence is subject to any kind of real world review and disposal so that if a phone number and name, for example, are extracted from a phone, that phone number and name will in time be deleted from all databases unless its retention has been identified as necessary. Given the volume of data, this sort of process could only be achieved, if at all, on an automated basis.

6.67. These issues are not unique to CT Policing. The Information Commissioner has reported on mobile phone extraction, for example in connection with the investigation of sexual offending, and concluded that many police forces are “simply unaware of the nature and extent of the material they are holding” meaning that urgent work is required on obtaining, interrogating, and retaining this data[footnote 284]. Post-Brexit, the Home Office is required to implement a new, indeed unique, process for the conditional retention of Passenger Name Records under the Trade and Cooperation Agreement, including for counter-terrorism purposes[footnote 285].

6.68. As things currently stand, the combination of paragraph 11A(3)(a), the Code of Practice and MOPI are insufficiently clear as to how copied information will be managed. I therefore recommend that

  • CT Police immediately establish a new standalone public policy on CT intelligence management which explains, as far as is consistent with national security, how data obtained from Schedule 7 is managed, reviewed, retained or deleted. The policy should explain what controls there are on access to this data and what, if any, oversight there is on the integrity of the retention regime.

  • the Code of Practice be amended to refer, at paragraph 66, to that new policy.

Biometrics

6.69. According to internal figures given to me, the percentage of examinations leading to the taking of biometric markers (defined for the purpose of internal statistics as a fingerprint, photograph or DNA) taken during 2020 is slightly lower than the percentage for digital downloads.

6.70. Under Schedule 7 biometrics are confined to fingerprints and DNA profiles. The retention of these are subject to a specialised regime of National Security Determinations overseen, presently, by the Biometrics Commissioner[footnote 286]. During 2020 the scheme was temporarily altered because of the effect of the pandemic[footnote 287]. In August 2020 the length of National Security Determinations were increased to a maximum of 5 years[footnote 288].

6.71. The Biometrics Commissioner, Professor Paul Wiles, published his final report during 2020[footnote 289]. In March 2021 Professor Fraser Sampson was appointed to the separate statutory roles of both the Biometrics Commissioner and the Surveillance Camera Commissioner. In September 2021 the government began a consultation on data reform which included a question on whether the functions of the Biometrics Commissioner should be rolled into the office of the Information Commissioner. My response and strong disagreement with this course of action is available online[footnote 290].

Freight

6.72. Schedule 7 contains powers enabling non-covert searches of freight[footnote 291]. The practicalities of selecting freight for examination is quite different from selecting people for examination. Freight powers are not widely understood even within CTP Borders.

6.73. The potential importance of freight powers to counter-terrorism is shown by the example of Ukrainian far-right badges. Identifying shipments of these badges to the UK can assist in identifying potential Right Wing Terrorism at home.

6.74. In Great Britain in 2020, there were 2,832 examinations of unaccompanied freight (518 air freight and 2,314 sea freight), as compared with 5,232 the previous year. This represents a decrease of 46%, a smaller decrease than examinations of person which may reflect the fact that Covid has impacted less on the flow of freight than of people.

6.75. According to statistics provided to me by the PSNI, there were 19 examinations of unaccompanied freight in Northern Ireland in 2020, compared with 255 in 2019. This represents a decrease of 93%. It remains to be seen what the final shape of post- Brexit non-terrorism border examination powers is going to be in relation to goods passing between Northern Ireland and Great Britain.

6.76. The government accepted my recommendation in last year’s report that guidance (now published) should clearly delineate between the power of examination and the “screening” power to enter buildings and vehicles to decide whether to exercise the power to search[footnote 292].

Hostile State Activity powers

6.77. During 2020 new powers to examine persons and freight to detect hostile state activity, not terrorism, came into force.

6.78. I described the powers under Schedule 3 to the Counter-Terrorism and Border Security Act 2019 in last year’s report[footnote 293], which are subject to review by the Investigatory Powers Commissioner[footnote 294] and noted that there could be some potential overlap between these powers and Schedule 7. Indeed, the Schedule 3 Code of Practice notes the possibility of switching regimes[footnote 295].

6.79. There are subtle but important differences between the two powers (in particular, that Schedule 3 does allow access to privileged and journalistic material) which means that it is important to know which power is in play.

6.80. Having seen training materials and discussed the Schedule 3 power with officials and CTP Borders, it is unlikely in principle that there will be any muddling up of powers. But it is not impossible in the real world that examining officers, making difficult decisions potentially on the spur of the moment, will need to pick a careful path between the two powers. I have been informed that there has already been one example of switching between the two statutory regimes.

6.81. I will continue to monitor this situation and will remain in contact with the Investigatory Powers Commissioner’s Office (IPCO) on this topic. It is unnecessary to make a recommendation, but it would be sensible for CTP Borders to make a record of every example of switching, so that either I or IPCO can consider the topic further in due course.

7. TERRORISM TRIALS AND SENTENCING Generally

7.1. In England and Wales, statistics from December 2011 to 2020[footnote 296] reveal 944 charges under terrorism legislation, of which 151 related to information collection, 84 to membership, 82 to possession of articles, and 192 to attack-planning; and 392 terrorist-related charges under non-terrorism legislation including 47 relating to homicide and 31 relating to explosives.

7.2. The principal offence-making statutes are the Terrorism Act 2000 and Terrorism Act 2006. These Acts of Parliament are constantly being weathered by a stream of amendments, recent examples of which include: liability for streaming information likely to be useful to a terrorist, the as yet unused Designated Area offence, and recklessness as an element of inviting support for a proscribed organisation[footnote 297].

7.3. Generally, amendments are put before Parliament when quite specific gaps are identified by CT Police or the Home office. Many amendments are modest and reasonable but there is always a temptation for government to propose and Parliament to enact changes on a something-must-be-done basis, informed by projections as to utility that in hindsight fall wide of the mark.

7.4. For this reason I proposed in my last report that a mechanism should be found for evaluating the utility of changes to counter-terrorism legislation, specifically by inviting prosecution authorities to make a record of whether amended or new offences are being charged for a period for 5 years from the relevant legislation[footnote 298]. The government has accepted this recommendation, and the three prosecution agencies (Director of Public Prosecutions (in England and Wales), Lord Advocate (in Scotland) and Director of Public Prosecutions (in Northern Ireland) have agreed to make such a record. The effectiveness of this will depend on this information being both recorded and separately presented, and I look forward to reporting on this data in due course.

7.5. The Designated Area offence could not have been prosecuted in 2020 because the Home Secretary did not, and still has not at the time of writing this report, designated any part of the world[footnote 299]. There are outstanding questions about what designation is designed to achieve:

  • Is about banning entry by United Kingdom residents or nationals to certain areas or, given the long list of specific defences capped by a general defence of reasonable excuse[footnote 300], is it really concerned with criminalising those who travel with terrorist intent[footnote 301]?

  • Is its purpose dissuasive as to future travel (which may be more effective for waverers than hardcore travellers), or to assist in prosecuting returners where battlefield evidence may be hard to come by?

  • What is the purpose of penalising British nationals or residents who, being in the area at the time of its designation, fail to return within a month[footnote 302]? This question would apply with particular force were Syria to be designated, where the government’s policy is one of maintaining those who travelled out at a “strategic distance”[footnote 303], rather than encouraging them to return.

Prosecutions in 2020

7.6. In 2020, 54 persons were proceeded against by the Crown Prosecution Service for terrorism-related offences[footnote 304]. This is the same figure as the previous year. Previous high conviction rates were maintained. 49 defendants were convicted (91%), of whom 30 (61%) pleaded guilty and 19 (39%) entered a not guilty plea.

7.7. Conduct in Great Britain leading to convictions in 2020 included:

a. murder, (R v Hashem Abedi, brother of Manchester Arena attacker), R v Saadallah, the Reading attacker)

b. attempted murder, (R v Ziamani and Hockton, HMP Whitemoor attacker).

c. attack-planning, (R v Sheikh, suicide bomb plot against St.Paul’s Cathedral, R v Mohiussanath Chowdhury, Pride parade attack plot[footnote 305], R v Paul Dunleavy, teenage attack-planner and follower of Feuerkrieg Division, R v Jack Reed, teenage neo- Nazi attack-planner)

d. attempting to possess or control an explosive substance with intent, (R v Chowdhury).

e. encouraging terrorism, (R v Vaughan, R v Nimmo, R v Hunter).

f. dissemination of a terrorism publication, (R v Vaughan, R v A Youth).

g. Membership of a proscribed organisation (R v Jones)

h. possession of a document containing information likely to be useful to a person committing or preparing an act of terrorism, (R v Chowdhury, R v Yanaouri, R v Vaughan, R v Dunleavy, R v A Youth, R v Nimmo, R v Conroy).

i. failing to inform the police of a terrorist attack plot (R v Sneha Chowdhury, sister of Pride parade attack plotter above[footnote 306]).

j. failure to comply with a Schedule 7 examination, (R v Golding).

7.8. The CPS publishes a very useful summary of the terrorism prosecutions on its website.

7.9. As with previous years, the large majority of terrorism convictions in 2020 related to Islamist extremism. No returning Foreign Terrorist Fighter was convicted of any terrorism offence during 2020. Greater availability of material and information from the vacated battlefields in Syria does not necessarily translate into admissible and contextualised evidence of criminal offending[footnote 307].

7.10. A notable number of teenage Right Wing terrorist offenders were also convicted of child sex offences: Jack Reed (sexual touching of a child under 13)[footnote 308]; R v Vaughan (indecent images of children)[footnote 309], R v Tchorzewski (indecent images of children)[footnote 310].

Sentences in 2020

7.11. Of the 45 people tried and convicted of terrorism-related offences[footnote 311]:

  • 2 received life sentences.

  • 1 received a sentence of between 10 and 20 years.

  • 14 received a sentence of between 4 and 10 years.

  • 11 received a sentence of between 1 and 4 years.

  • 9 received a sentence of under 1 year.

  • 1 received a hospital order.

  • 7 received a non-custodial sentence.

7.12. 2020 saw the most defendants convicted of TACT offences receiving a custodial sentence of imprisonment of less than 12 months and the second highest number of non-custodial sentences was also imposed. These most likely result from breaches of notification orders, TEOs or TPIMs, or failing to comply with the requirement of Schedule 7 Terrorism Act 2000. Recent terrorism convictions of youths have attracted non-traditional disposals such as referral orders[footnote 312] and conditional cautions.

7.13. In last year’s report I observed that no Serious Crime Prevention Order (‘SCPO’) had yet been imposed post-conviction in relation to terrorist offending[footnote 313]. In 2020 an SCPO was imposed against Zakaria Yanaouri who was convicted of possession of information likely to be useful to a terrorist[footnote 314], with a term of 5 years.

7.14. The sentencing reforms of the Counter-Terrorism and Sentencing Act 2021 are likely to result in longer sentences for some, and greater periods of post-release supervision for most. The legislation reflects a trend, begun in 2008[footnote 315] and driven forward by the attacks of 2019 to 2020, towards regarding terrorist offenders as a special class of offenders whose risk, potentially enduring, must be monitored closely for years after release. As I separately discuss in my ‘Terrorism in Prisons’ report (2021), monitoring must also be improved in prison.

7.15. This makes Parliament’s decision to remove the ability of the Parole Board to consider the risk posed by terrorist (including juvenile) offenders found dangerous by the sentencing court where the offence carries a maximum of life imprisonment[footnote 316], the more striking. It runs contrary to the need to keep risk under review and is inconsistent with trust placed on the Parole Board in other respects[footnote 317]. I suggested at the time that there were three potential consequences: loss of a spur to good behaviour, loss of an opportunity to understand current and future risk at Parole Board hearings and, with respect of juveniles, a failure to recognise that the risk presented by young offenders may be more susceptible to change than adults.

7.16. Although it is a nostrum of sentencing for the most serious terrorism offences that the object of the Court is to punish, deter and incapacitate and that “rehabilitation is likely to play a minor (if any) part”[footnote 318], the public interest is served by effective management of future risk, which includes (particularly in the case of young offenders) release at the right time.

Prison in 2020

7.17. At the end of 2020 209 individuals were in prison for terrorism-related offences (down from 231 individuals in the previous year)[footnote 319].

7.18. Of these, 156 were Islamist extremists (down from 177 in the previous year and the lowest number since the year ending 2015), 42 were identified as adhering to right-wing ideologies (the highest number ever recorded, up from 41 in the previous year) and 11 were categorised as “other” (this includes prisoners not classified as holding a specific ideology).

7.19. Of the 209 individuals identified as terrorists, 154 declared themselves as being Muslim[footnote 320]. Twenty-four of the prisoners self-identified as Christian, 1 as Buddhist, 2 as Jewish, and 1 as Sikh. Of the remainder, 18 declared themselves as having no religion and 9 belonged to “other religious groups”.

Releases from prison

7.20. Given that most terrorism sentences have, to date, been measured in years rather than in decades, it follows that most convicted terrorists will at some stage be released.

7.21. In Great Britain in the year ending December 2020, 42 individuals in prison for terrorism-related offences were released[footnote 321]. Of these, 1 was serving a life sentence, 14 were serving a sentence of 4 years or more, 15 were serving a sentence of between 12 months and 4 years, 1 was serving a sentence of between 6 months and a year, and 11 had not been sentenced (this includes individuals held in custody on remand prior to charge or sentencing). There are no official statistics for Northern Ireland.

7.22. The Fishmongers’ Hall and Streatham attack have heightened the readiness of officials to manage the risk posed by released offenders by recalling them to prison or amending their licence conditions[footnote 322]. During 2020, an obscure power was exercised by the Youth Court to delay the release of a young Right Wing terrorist by 2 months: a new convert to Islam he had started associating with Sudesh Amman, who was to become the Streatham attacker[footnote 323].

Collecting or possessing information

The offence

7.23. Under section 58(1) Terrorism Act 2000 it is an offence for a person to collect, possess or view online:

“a record of information of a kind likely to be useful to a person committing or preparing an act of terrorism”.

7.24. Although everyday information may well be useful to terrorists – for example a map of the rail network – the offence has been authoritatively interpreted as applying only to information that is, of its very nature, designed to provide practical assistance to a person committing or preparing an act of terrorism[footnote 324].

7.25. In recent years, the offence has been applied to: the Anarchy or Anarchist Cookbook, Islamist and Right Wing terrorist handbooks on weapons, explosives, military or terrorist tactics including assassinations, knife attacks, vehicle attacks, and poisons, Da’esh articles on “Just terror tactics” and instructions on 3-D handguns[footnote 325]. The information is generally gathered online but has also been found in the form of handwritten notes.

7.26. The maximum sentence is 15 years, increased from 10 years in 2019[footnote 326]. According to statistics published by Sentencing Council, the average sentence for adult offenders convicted of the section 58 offence in period 2008-2018 was roughly 3 and a half years’ imprisonment[footnote 327]. Sentencing guidelines[footnote 328] indicate a starting point of 12 months’ custody for the least serious offending. A panoply of restrictive measures apply on release[footnote 329].

7.27. Probably no other offence confers such a degree of discretion on CT Police and prosecutors because so little needs to be proven.

  • All that needs to be established is that the information is of the required type, that the defendant was in possession of it and was aware of the nature of its contents[footnote 330]. Unlike the offence under section 57, it is not necessary that the defendant had any intention to use it himself to commit an act of terrorism or pass it to someone else to do so[footnote 331]. It covers the defendant who has gathered and stored the information without having any clear idea of what he intends to do with it[footnote 332].

  • There is a defence of “reasonable excuse” [footnote 333] but the defendant must show an objectively reasonable excuse for having the information, not just that his purpose was not to commit an act of terrorism[footnote 334].

7.28. However conservatively prosecutors approach the question of whether there is sufficient evidence, with all due attention to factors relevant to any defence[footnote 335], many individuals will fall into the net of liability.

7.29. It follows that great importance is shifted to the public interest component of prosecutorial decision making by the Crown Prosecution Service. As the Code for Crown Prosecutors recognises, it has never been the rule that a prosecution will automatically take place once the evidential stage is met, and that whether to prosecute depends on a balance of public interest factors against and in favour of prosecution.

7.30. Those public interest factors will include:

  • whether there a link (including any link established by sensitive intelligence) to terrorist activity or a terrorist mindset.

  • Whether criminal justice intervention is going to make things worse.

Purpose of the offence

7.31. Section 58 is one of the ‘precursor’ offences created by terrorism legislation which seek to capture the risk of future acts of terrorism rather than awaiting the outcome.

7.32. Originally found in legislation applying to Northern Ireland and later extended to Great Britain in 1994[footnote 336] the offence was found to be useful by Lord Lloyd in his 1996 Inquiry into Legislation against Terrorism. Lord Lloyd referred to the well-recognised need for the police to intervene against the terrorist at an early stage, explaining that the offence “…is designed to catch possession of targeting lists and similar information, which terrorists are known to collect and use.”[footnote 337]

7.33. In enacting section 58 Terrorism Act 2000, Parliament must therefore have proceeded on the view that, in fighting something as dangerous and insidious as acts of terrorism, the law was justified in intervening to prevent these steps being taken, even if events were at an early stage or if the defendant’s actual intention could not be established[footnote 338].

7.34. Professor Clive Walker QC analyses this type of offences as a risk mitigation measure which applies where the prospect of harm is uncertain and where “the only immorality has been the imagining of wickedness rather than its infliction, or a fair imputation that it will occur”[footnote 339].

7.35. For my part I would qualify the reference to “immorality” since a fair imputation that harm will occur is an objective assessment of risk which does not necessarily require any blameworthiness on the part of the defendant.

  • For example, an individual may commit the section 58 offence with no foresight as to when or how the information might ever be used even though, objectively analysed, the information could fall into the wrong hands. Contrast the offences of encouraging terrorism, disseminating terrorist publications, or inviting support for a proscribed organisation, each of which require subjective recklessness that a person will be encouraged[footnote 340].

  • The fact that the offence does not penalise intrinsically harmful behaviour[footnote 341], but merely conduct that only may or may not be linked to future acts of terrorism has led some to the conclusion that section 58 penalises a form of conduct that should not form the basis of criminal liability[footnote 342]. Less persuasive is the argument that the offence has chilled legitimate activity.

7.36. This justification for the section 58 offence puts considerable weight on the fair imputation that acts of terrorism are somewhere down the line, whether committed by the defendant or some other person - in other words the link between the prohibited conduct and the undesired harmful result[footnote 343].

7.37. The fact that it is the risk that the information could be used for future harm that is relevant to section 58 rather than moral blameworthiness is illustrated by the penalisation of information that is practically useful to a terrorist, on the one hand, and the non-penalisation of terrorist mindset material, on the other.

  • Mindset material could include terrorist propaganda such as beheading videos

  • the type of information found on the electronic devices of Khuram Butt the year before he carried out the 2017 London Bridge and Borough Market attacks with Rachid Redouane and Youssef Zagba, but which did not fall within section 58[footnote 344].

  • Downloading this type of material for personal pleasure may well suggest a clearer link to terrorist intent than possession of a single document which contains useful information, meaning that intuitively arbitrary distinctions can arise between an individual whose cache of violent Islamist or Right Wing terrorist propaganda happens to contain a document with useful information, and an individual whose cache of material does not.

  • However, as I reported in my previous annual report[footnote 345], and as the government accepts, to prosecute those in possession of mindset material would be to spread the net of terrorist offending too widely.

Modern use

7.38. Current statistics show that section 58 was the most charged principal[footnote 346] offence in 2019 and 2020, and the second most charged principal offence in the years 2012, 2014, 2017 and 2018[footnote 347]. It was the most common principal offence of which defendants were convicted in 2007, 2019 and 2020, and the second most common principal offence in 2009, 2012, 2013, 2014 and 2018.

7.39. This frequency is not surprising. The widespread presence of this type of material online means the offence is easy to commit. The ever-increasing ease of access to the internet and time spent online[footnote 348] suggests that this offence will continue to be committed at high levels whether the availability of this material genuinely reflects a growth in the number of individuals willing to carry out terrorist attacks, or not. This was not in contemplation when the Terrorism Act 2000 was enacted.

7.40. The more widely available a document is, and the more it becomes popular or attracts cachet for unpleasant, hateful, fashionable but non-terrorist reasons[footnote 349], then the less there is a link to ‘real’ terrorism. Prosecutors are conscious of this point and recognise the danger of forcing the issue: even if a defendant is in possession of a document which is intrinsically useful to terrorists, so that liability is made out, the fact that it is widely available may tip the balance on public interest grounds against prosecution.

Analysis

7.41. The validity of section 58 depends on the link between the possession of information that is useful to a terrorist and a risk of acts of terrorism. Put shortly, section 58 begins to lose its validity as a risk measure if individuals are prosecuted for possession of information where there is no sensible risk that they or any other person will use that information.

7.42. As terrorism evolves, that link must continue to be evaluated. The scope of the offence should be no wider than necessary.

7.43. The evolution of terrorism, at least outside Northern Ireland, is affected by the following factors, at least:

  • A decline in the role of proscribed organisations and the growth of online self- initiators.

  • An increasingly young cohort coming to the attention of CT Police[footnote 350], although it remains the case that completed Right Wing Terrorist attacks tend to be carried by older men[footnote 351].

  • More individuals coming to the attention of CT Police who are diagnosed with or appear to have mental health conditions, or neurodivergent conditions such as autism[footnote 352] which may affect the manner in which defendants gather or retain material[footnote 353].

  • The spread of Right-Wing Terrorist ideology which often glorifies technical aspects of weaponry[footnote 354]. The Court of Appeal has recently held that no defence of reasonable excuse could arise if the purpose of possessing the material was simply to be able to boast to anyone that the defendant has it and has derived knowledge from it as to how to break the law by making home-made weapons and ammunition[footnote 355]. That sort of boasting may be attractive to some.

7.44. It is not possible to say, at this stage, that the justification for the section 58 offence has fallen away. There remains a strong public interest in allowing the police to arrest and detain, and the CPS to prosecute, in circumstances where little or no evidence can be obtained of attack-planning. I am not aware of any prosecutions brought where there was neither evidence of terrorist intent[footnote 356] or terrorist sympathy. In other words, section 58 is responsibly used by prosecutors.

7.45. However, despite the sensitivity of its use to date, it remains an offence at the outer edges of legitimacy. Attracting up to 15 years’ imprisonment, it is capable of being committed with limited if any moral blameworthiness and therefore, as the

Sentencing Council has suggested[footnote 357], non-custodial options ought to be available, as should other options in the case of youths (referral orders, conditional cautions). Wise exercise of prosecutorial discretion is not a complete cure for overbroad liability[footnote 358] and it is always preferable that conduct involving an allegation of terrorism is prosecuted using offences with an element of terrorist fault.

8. SPECIAL CIVIL POWERS TPIMS

8.1. This part contains my discretionary annual review of the operation of the Terrorism Prevention and Investigation Measures Act 2011 for 2020. For the calendar year 2022 until 2026, the Independent Reviewer of Terrorism Legislation will be obliged to carry out an annual review[footnote 359].

Generally

8.2. Terrorism Prevention and Investigation Measures (TPIMs) are special civil measures imposed on individuals to limit the capability, and improve the monitoring of, individuals whose terrorist risk cannot be effectively managed by criminal prosecution. They were introduced in 2011 as less severe alternatives to control orders previously made under the Prevention of Terrorism Act 2005. They are anticipatory, in that they seek to forestall future terrorist harm but require some proof of involvement in terrorism-related activity. As with the former control order regime, their greatest drawback is that morally culpable behaviour is addressed outside the criminal process and subject to closed procedures from which the TPIM subject is excluded.

8.3. The total number of individuals who have been served a TPIM Notice since the TPIM Act 2011 received Royal Assent up to 31 December 2020 is 24[footnote 360]. 52 were subject to control orders under the Prevention of Terrorism Act 2005 in a shorter period[footnote 361].

8.4. An illustration of the level of restrictions that may be imposed by TPIMs is given by the case of JM. The TPIM imposed in 2019, and still in force in 2020, contained the following measures which I reproduce in full[footnote 362]:

i. A requirement to reside in a particular city away from his family home and to remain in that residence overnight between 21:00 and 07:00.

ii. A travel measure requiring JM to surrender travel documents and prohibiting him from leaving Great Britain without permission. This measure also prevented JM from leaving a specified area of the particular city.

iii. An exclusion measure that prevented JM from entering specified areas or places unless the Home Office has given him permission.

iv. A movements and directions measure requiring JM to comply with any directions given to him by a police officer.

v. A financial services measure.

vi. A property measure requiring him to take certain steps in relation to any property that he owns or rents.

vii. A weapons and explosives measure.

viii. An electronic communication device measure that sets out restrictions on JM’s use and possession of communications and electronic devices and that of others living at or visiting his residence.

ix. An association measure that restricts JM’s ability to meet and communicate with listed individuals.

x. A work or studies measure.

xi. A reporting measure that requires JM to report in person to a specified police station, and by telephone to the electronic monitoring service from the monitoring unit in his residence on days and at times notified by the Home Office.

xii. An appointments measure that requires JM to attend appointments with persons notified by the Home Office.

xiii. A photography measure requiring him to permit the police to take photographs of him.

xiv. A monitoring measure that requires JM to wear an electronic tag which uses satellite tracking technology and to keep the tag charged.

8.5. Firstly, taken cumulatively these measures pare back the ordinary freedoms of the TPIM subject in a way that is ordinarily only possible for adults in the context of criminal proceedings, immigration proceedings, and restrictions under the Mental Health Act 1983.

8.6. Secondly, the functioning of these measures imports a high degree of overt control and supervision by the authorities: in establishing the initial operation (moving to a new address, handing in unapproved electronic devices and travel documents, providing information about property, being fitted with a tag); through requiring cooperation with ongoing positive obligations (reporting in person to a police station and by telephone, attending appointments); and through requiring permission for any occasional additional freedoms (for movement outside defined areas or meetings with unapproved individuals). This is in addition to any covert monitoring that may take place.

8.7. Thirdly, both the restrictions on the TPIM subject and the degree of control exercised by the authorities impinge very significantly on any family life that the TPIM subject has, and are particularly impactful on family members living in the same house: to take one example, the Secretary of State may impose requirements on the individual in relation to the possession or use of electronic communication devices by other persons in the individual’s residence[footnote 363].

8.8. Fourthly, each measure is backed up by criminal sanction, placing TPIM subjects in a net of potential criminal liability which in practice results in a sizeable proportion of TPIM subjects being prosecuted for breaches. Last year I reported that there had been 4 prosecutions of the TPIM subjects in 2019. During 2020:

  • In February KG was sentenced to two years’ imprisonment after pleading guilty to 12 breaches of the financial services measure of a TPIM notice[footnote 364].

  • In May, LF was charged with two counts of breaching his TPIM notice[footnote 365].

  • During the second half of 2020 QT was charged with breaching his TPIM notice and was subsequently sentenced in June 2021 to a community order.

  • In September LF cut off his electronic tag and took a taxi to London, sparking a massive manhunt and huge delays at ports. In 2021 he was sentenced to 3 years and 4 months imprisonment[footnote 366].

TPIMS in 2020

8.9. Basic information about TPIMs is contained in quarterly written Ministerial statements setting out the number of individuals subject to TPIMs, whether they are British nationals, whether they have been relocated, and summarising any criminal proceedings resulting from breaches of TPIMs during that period[footnote 367].

8.10. These statements show that during 2020 the number of TPIMs in force fluctuated between 3 and 6 TPIMs, reflecting the fact that a number of TPIMs expired during 2020, whilst others were revoked on account of breaches.

8.11. The reports also contain information about nationality. Considering the genesis of the TPIM regime, it is interesting to note that all TPIM subjects during 2020 were British nationals. Control orders, which are the immediate predecessors of TPIMS, were created by the Prevention of Terrorism Act 2005 to replace flawed and discriminatory detention legislation which only applied to foreign nationals[footnote 368]. It might therefore be thought that at least some of those caught by TPIMs would be foreign nationals who presented a threat to national security but who could not, like the original Belmarsh detainees, be deported because of a risk of death or torture in their home countries.

8.12. The application to British nationals only could reflect a potential fall in the threat posed by resident foreign nationals. Analysis of the statistics on arrests, charges and convictions shows an overall decline in the number of foreign nationals who were arrested, charged and prosecuted for terrorism-related activity in the years ending 2017 to 2020[footnote 369]. This may in turn reflect the number of foreign or former dual nationals who travelled to Syria and Iraq to join Islamic State/ Da’esh during the first part of the last decade. No longer resident, their threat is capable of being addressed by immigration powers: for example, the power to deprive dual nationals of their British citizenship was exercised 104 times in 2017 meaning they could no longer lawfully return[footnote 370].

8.13. The more likely explanation is that, as in 2019, all but one of the TPIMs in force during 2020 were against members of the proscribed organisation, Al-Muhajiroun (‘ALM’), whose members are on the whole British nationals.

8.14. Few could conclude that this targeted response against a single organisation was anything other than a reasonable and proportionate tactical response. ALM’s direct or indirect impact on UK terrorism includes:

  • Attacks by ALM members Adebolajo and Adebowale (2013 murder of Fusilier Lee Rigby in Woolwich), Khuram Butt (2017 London Bridge attack), Usman Khan (2019 Fishmongers’ Hall attack).

  • Attack-planning by individuals influenced, encouraged or given tacit approval by ALM members: Brusthom Ziamani (2014 knife and hammer attack plot; later convicted of 2020 attempted murder of prison officer in HMP Whitemoor), Nadir Syed (2014 knife attack plot), 15-year old S (convicted of inciting another to commit terrorism), Lewis Ludlow (pleaded guilty in 2018 to vehicle attack plot), Safiyya Shaikh (member of all-female suicide cell, pleaded guilty in 2020)[footnote 371].

8.15. In August 2020, MI5 concluded that there was a significant risk that ALM-linked attacks would take place against the UK and UK nationals.[footnote 372]

8.16. During 2020 the following ALM members were subject to TPIMs during 2020:

  • HB: TPIM imposed 2018, revoked and remanded in custody 2018, revived March 2020 (following the end of HB’s post-release licence conditions), extended for a further year in December 2020.

  • QT: TPIM imposed 2018[footnote 373], extended for a further year in March 2020.

  • HC: TPIM imposed 2018, extended 2019, expired August 2020.

  • JD: TPIM imposed 2018, extended 2019, expired September 2020.

  • JM: TPIM imposed 2019, extended for a further year in November 2020.

  • LF: TPIM imposed 2019, revoked and remanded in custody 2020.

8.17. Only one TPIM was made against an individual in 2020 who was not assessed to be a member of Al Muhajiroun. This was KG, whose TPIM was imposed in January 2020 but revoked within weeks on account of the breaches which led to his imprisonment in February.

8.18. The TPIMs imposed against JM and LF were each second TPIMs. JM and LF were assessed to be senior leaders of Al Muhajiroun whose first TPIMs had expired in 2018. The TPIM Act 2011 allows for a further TPIM if new ‘terrorism-related activity’ can be identified which occurs after the earlier TPIM came into force[footnote 374]. The second TPIMs contained sterner measures than in the first TPIMs.

  • In the case of JM, MI5 assessed that, after the imposition of the 2016 TPIM, JM continued to act for the benefit of ALM. As a member and senior leader of ALM, he was believed to have participated in activities which served to further the aims and ambitions of ALM. This included giving encouragement to the commission, preparation or instigation of acts of terrorism including while subject to the 2016 TPIM[footnote 375].

  • The case against LF was that he also continued to act for the benefit of ALM, had been in possession of Islamist material which included material condoning acts of violence, and had encouraged terrorism[footnote 376].

  • In upholding each TPIM, the High Court noted that since the original TPIMs had been upheld, the activity of Al Muhajiroun members had continued, notably in the London Bridge, Fishmongers’ Hall and HMP Whitemoor attacks[footnote 377].

8.19. The Overnight Residence Measure (in practice, relocation) continues to be used and during 2020 at least half of TPIM subjects were relocated[footnote 378]. The power to relocate was added by the Counter-Terrorism and Security Act 2015 following recommendations by Lord Anderson QC in this third annual review of the TPIM regime. According to post-legislative scrutiny of the 2017 Act published in 2021, since the Act received Royal Assent in February 2015, 12 individuals have been subject to forced relocation [footnote 379].

8.20. During 2020 I attended the majority of TPIM Review Group (‘TRG’) meetings[footnote 380]. At these, officials from the Home Office, CT Police and MI5 review the necessity and proportionality of TPIM measures, consider variations, discuss exit strategy, are updated on the prospects of criminal prosecution, and consider the outcome of practical and ideological mentoring sessions. I read all relevant papers and evidence.

8.21. The standard of consideration at TRG meetings remains high and adequate time was allowed for a meaningful review of each TPIM.

  • The Home Office official chairing the meeting injected a proper degree of challenge to the ongoing management of the TPIM subject, including on the possibility of relaxing certain measures, and impact on family members.

  • The TRG is conducted using a draft agenda which now requires consideration of each measure in turn: this is a clear improvement over the previous practice of considering the measures as a whole.

  • Following my observations in previous reports, I am pleased to say that there is greater analysis of whether prosecution for terrorism offending is a reasonable alternative to a TPIM.

  • However, I remain of the view that insufficient attention is given to the passage of time as noted in last year’s report[footnote 381].

The TPIM Catch-22

8.22. TPIMs offer a degree of control over dangerous individuals which it can be hard to relinquish. Quite reasonably, investigators are wary of concluding that terrorist risk, once established, has gone away absent a real change of heart or disavowal of former terrorist intentions[footnote 382]. Mere compliance with TPIM measures, an absence of terrorist- related behaviour whilst strictly monitored, or plausible-sounding denials to official interlocutors (police, mentors), are unlikely to be sufficient.

8.23. This gives rise to the TPIM catch-22:

  • On the one hand, in order to test whether an individual would revert to terrorism-related activity in the absence of TPIM measures, there may be no alternative but to reduce or remove measures; for example, by allowing an individual to associate or move more freely.

  • On the other hand, association and movement measures have been imposed precisely to counter the risk of terrorist-related activity. In the absence of evidence of risk reduction, to do so might put members of the public at risk of harm.

  • The conundrum operates even more strongly where the TPIM contains only a limited number of measures: removing one measure may remove the essence of the TPIM and raise the question of whether the remaining measures remain necessary at all.

  • Section 9 reviews provide useful scrutiny and can result in amendments to measures, and provide an opportunity for the TPIM subject, if they yield evidence demonstrating a change of heart. However, this is undermined by the occasional unavailability of legal aid which I consider below.

8.24. On this basis, future TPIMs which can now be renewed for up to 5 years[footnote 383], may go into a third, fourth or fifth year simply on the basis that no change in risk can be confidently identified. TPIM subjects may be left with the sole hope that competition for resources means that a TPIM in their case is no longer seen as cost and resource effective.

8.25. I know from my discussions with officials and attendances at TRGs that the Home Office is aware of the general desirability of reducing TPIM measures where possible. I also recognise that all TPIM cases are different. Whilst I do not claim that there are easy solutions applicable to all cases, I recommend that the Home Office and MI5 formulate general internal guidance on evaluating risk reduction during the currency of a TPIM which could include:

  • Learning from past TPIMs and control orders where risk reduction was successfully tested or observed. Officials move on, and institutional knowledge fades.

  • identifying measures which could be relaxed or removed over a period of time, potentially replaced with alternative measures to provide additional assurance during the period of testing.

Risk

8.26. Like terrorist offenders, each TPIM subjects is unique. Unsurprisingly, some TPIM subjects have been diagnosed with, or appear to have, a range of psychological or neuro-divergent conditions.

8.27. As part of reviewing the necessity of a TPIM and its individual measures, TRGs require officials to evaluate risk by reference to observed conduct on the part of the TPIM subject, and an assessment of how an individual’s level of risk is likely to react in connection with future events, for example the relaxation of a particular measure. Where a TPIM subject is atypical in the sense described above, that process may be less robust in the absence of specialist psychological input.

8.28. Secondly, the effectiveness of measures in reducing risk for atypical individuals may not be understood without specialist input. For example, the delivery of practical or ideological mentorship may require adjustment. Certain measures, for example measures that provide a high degree of structure in the day, may be particularly effective and may exclude the need for other intrusive measures.

8.29. I therefore recommend that the Home Office should, in cases involving atypical individuals in the sense described above, consider whether the attendance of a psychologist at TRGs may be useful when evaluating risk and measures to reduce risk.

8.30. Of all the points raised in last year’s report, the failure of the government to secure that legal aid is available to all individuals on whom a TPIM has been imposed (subject to means) has caused me most concern. I will not repeat my reasoning that this failure undermines the statutory regime as a whole[footnote 384]. In correspondence with a UN Special Rapporteur, the government made a virtue of the judicial oversight and the “automatic review” carried out under section 9[footnote 385]. When the TPIM scheme was presented to Parliament, it was never suggested that the availability of legal aid would depend on the type of TPIM imposed[footnote 386]. It is foreseeable that in some if not all cases an inability of secure representation in these complex types of proceedings will lead to real unfairness as well as practical difficulties for the court.

8.31. The Ministry of Justice does not suggest that legal aid is not available in principle for light-touch TPIMs. For completeness, I do not share officials’ legal analysis that paragraph 45 of Schedule 1 to the Legal Aid, Sentencing and Punishment of Offenders Act 2012, which establishes the prima facie entitlement for legal aid for TPIM cases, was qualified by secondary legislation in 2013[footnote 387].

8.32. However, the Ministry of Justice points to the manner in which the Director of Casework of the Legal Aid Agency applies relevant regulations made under section 11 of the 2012 Act.

  • The Legal Aid Agency is an executive agency of the Ministry of Justice. Its director must comply with directions and guidance given by the Lord Chancellor about their functions but cannot be given a direction or guidance about an individual case[footnote 388].

  • The regulations in question[footnote 389] require the Director to consider any application for legal funding against certain merit factors including prospects of success, the importance to the individual and the public interest. It is open to the government to exempt certain proceedings from the ambit of the regulations[footnote 390].

  • The suggestion appears to be that, applying the regulations, the Director has decided that there are certain TPIM cases where funding of a section 9 review hearing does not merit the grant of legal aid. However, officials have been unable to confirm categorically that this is the explanation for the withholding of legal aid.

8.33. The position is deeply unsatisfactory.

8.34. Since the automatic review of TPIMs is a matter of principle, the numbers affected by the absence of legal aid are not material. But there are, in truth, no more than a handful of affected cases (no more than three at the time of writing). Given the possibility of legal challenge, the failure to ensure legal aid for all, subject to means, not only frustrates Parliament’s expectations but appears to be a false economy. Moreover, as I intend to report more fully in next year’s report, there is a noticeable change in the TPIM cohort towards the more chaotic, neurodiverse and/or mentally unwell.

8.35. In these circumstances, although my recommendation in last year’s annual reported has been rejected by the government, I reiterate my recommendation that subject only to means legal funding should swiftly be made available to all TPIM subjects for the purpose of participating in section 9 reviews and recommend that this is achieved by making an order under section 11(6) Legal Aid, Sentencing and Punishment of Offenders Act 2012, exempting TPIM proceedings from the criteria referred to in that section.

Changes to the Regime

8.36. Five amendments have now been made to the TPIM regime by the Counter- Terrorism and Sentencing Act 2021. The most important changes (to time limits and standard of proof) only apply to new TPIMs made on or after 29 June 2021.

8.37. Firstly, the standard of proof for past terrorist-related activity has been changed from “is satisfied, on the balance of probabilities” to “reasonably believes”. It is doubtful how much of an impact this change will have: certainly less than the original proposal to change the standard to “reasonable cause to suspect”[footnote 391].

8.38. Secondly, and more significantly, TPIMs can now be renewed without proof of new terrorism-related up to a period of 5 years, an increase from 2 years. The initial proposal was for TPIMs of potentially indefinite duration.

8.39. Thirdly, more changes were made in relation to possible measures relating to residences (to allow a variation of the overnight curfew condition if considered necessary for resource reasons), curfews (to allow a curfew to be imposed at any time of the day), the provision of information (regarding the TPIM subject’s place of residence and electronic communications devices), and drug-testing.

8.40. Fourthly, a power to add a polygraph measure was added. No regulations have yet been made for the conduct of TPIM polygraph sessions[footnote 392]. Evidence from TPIM polygraph sessions is expressly excluded from criminal proceedings[footnote 393], but, although the government stated that the provision “is not designed to allow for information derived from a polygraph examination to be used as evidence in proceedings for breaching a TPIM (which is a criminal offence), to extend the duration of a TPIM notice, or to impose a new TPIM”[footnote 394], and indeed that “any attempt to use information derived from a polygraph examination to extend the duration of a TPIM notice would be unlawful”[footnote 395] there is no statutory bar as such. I expect the Home Office to draw to my attention any case in which polygraph evidence obtained under compulsion is sought to be introduced (in any manner) into TPIM proceedings, so I can consider the position in next year’s report.

8.41. Fifthly, the Secretary of State is empowered to vary a relocation measure by substituting a new place of residence where it is necessary for resource reasons[footnote 396]. The stated purpose of this amendment was to allow a TPIM subject to be moved to a new police area if, for example, there is a sudden loss of counterterrorist policing capacity in the original area.

TEOs

Generally

8.42. Temporary Exclusion Orders (TEOs) were introduced by the Counter- Terrorism and Security Act 2015 and are used to control the return to the United Kingdom of British nationals[footnote 397] who are outside the country at the point of imposition via a ‘permit to return’ system. Their name is therefore something of a misnomer – they are not about exclusion, but about managing return, typically when there is some indication that the individual intends to travel to the UK. The threshold for imposition is reasonable suspicion of involvement in terrorism-related activity[footnote 398], on permission from the High Court.

8.43. Once returned, the individual may be subject to one or more of the following measures for up to 2 years from the date of imposition: a duty to report to a police station, a duty to attend appointments (in practice, an ideological and/or practical mentor, and a duty to notify police of their place of residence and any change). These are three of the obligations that may be imposed under TPIMs, but TEOs are administered separately by a dedicated team which is responsible for national security immigration matters. Quarterly TEO Review Group meetings are held, many of which I have attended. I found these meetings impressive which included a very clear focus on the need to reintegrate the TEO subject. As with all counter-terrorism measures, control is one thing, but the most desirable outcome is no need to control at all.

8.44. The TEO regime was created with those returning from Da’esh-controlled territories in mind, but its use is not limited to those aligned or formerly aligned with Da’esh.

8.45. Uncertainties about when an individual will return, if at all, generates special pressures. Firstly, the need to renew TEOs where the individual does not return as anticipated to avoid the clock running down on the 2-year maximum[footnote 399]; last year I recommended that TEO obligations should only come into force on return, which the government has accepted but is yet to implement[footnote 400]. Secondly, the shortness of time to prepare a TEO application if return is imminent, although emergency provisions (as yet unused) do exist[footnote 401].

TEOs in 2020

year Number of TEOs imposed Number of returnees
2016 0 n/a
2017 9 (3 males, 6 females) 4 (1 male, 3 females)
2018 16 (14 males, 2 females) 5 (2 males, 3 females)
2019 6 (2 males[footnote 402], 2 females) 3 (2 males, 1 female)
2020 0 1 (female)

8.46. At the end of July 2020 there were 4 individuals in the UK subject to TEOs. Three were subject to both reporting and appointment obligations. One individual had been arrested and remanded into custody. Unlike under the TPIM regime[footnote 403], there is no scope to revoke and revive TEOs where individuals are held in custody in the UK in order to extend the maximum 2-year period. This means that whilst TEO obligations are suspended for an individual in custody, the clock continues to run.

  • This difference between the TPIM and TEO regimes may be explained by the purpose of the TEO regime, described in the explanatory notes to the Counter- Terrorism and Security Act 2015 as one imposing “temporary restrictions on travel”[footnote 404]; with the emphasis on managing return through a ‘permit to travel’ and then providing a degree of assurance through the imposition of a very limited number of obligations in circumstances where there may only be a reasonable suspicion of involvement in terrorism-related activity. In these circumstances, if the individual is detained, the purpose of assurance is satisfied. As with all potential terrorists who move in and out of custody, intelligence needs to be passed between those responsible for terrorism in prisons, and those responsible for monitoring TEOs.

8.47. Breach of a TEO is a criminal offence. In May 2020 a TEO subject known as QQ was given a suspended sentence for failing to meet his reporting obligations[footnote 405]. This was the first TEO breach resulting in criminal proceedings.

TEOs and Passport cancellation

8.48. By section 4(9) of Counter-Terrorism and Security Act 2015, when a TEO comes into force any British passport held by the excluded individual is “invalidated”. A TEO therefore has the effect of withdrawing the passport facilities of those subject to an order, without any further action by the authorities using the royal prerogative.

8.49. In 2017 the Court of Appeal considered whether the royal prerogative was no longer available to cancel passport facilities (‘abrogated’ in legal parlance) and that the authorities were limited to using the powers contained in the TPIM Act 2011[footnote 406]. The argument failed because the TPIM Act 2011 neither expressly or by necessary implication says anything about passport cancellation[footnote 407]. But because 2015 Act does contain statutory passport removal powers, the question (not addressed in any authorities) arises whether the royal prerogative can be used against a person for whom a TEO is available. Unlike the Sanctions and Money-Laundering Act 2018, which provides[footnote 408] for immigration sanctions but expressly provides that nothing affects the power to exclude by virtue of royal prerogative, the royal prerogative is not referred to in the 2015 Act.

8.50. The approach taken by officials is that the use of royal prerogative is excluded only where the circumstances arise in which it is appropriate to control a British national’s return to the UK on grounds of their terrorist risk, i.e. where a TEO is appropriate. The royal prerogative may be used in a wider range of circumstances than are material to the 2015 Act, for example to cancel the passport of someone believed to have been killed, or who is involved in serious organised crime, or hostile state activity.

Disclosures

8.51. Unlike TPIMs[footnote 409], TEO obligations do not extend to controlling the nature of employment. There is therefore no power for the authorities to veto employment obtained by returning TEO subjects. Separate police powers do exist for the police to disclose information where it is necessary to protection of, for example, children or critical national infrastructure, and it is possible that the circumstances leading to the imposition of a TEO might lead the police to speak to an employer or potential employer.

8.52. However, it is important to keep these powers separate to avoid the risk of legal confusion and practical frustration. Officials accept that there is a need to avoid loose language when referring on the one hand to the obligations that exist under TEOs, and on the other to the quite separate power to disclose information to employers or potential employers that may only arise in particular circumstances. The ultimate goal of the TEO regime is reintegration and this includes getting and keeping a job.

Passport Seizure and Retention

8.53. The power to seize and retain passports under Schedule 1 to the Counter- Terrorism and Security Act 2015 was exercised only once in 2019 and not at all during 2020[footnote 410]. In the years 2015 to 2018 it was exercised 58 times[footnote 411]. This pattern no doubt reflects the tailing off of travel by UK individuals to Da’esh controlled-areas, followed by the pandemic. There is however no reason to disagree with the post-legislative assessment that this is a “critical power”[footnote 412] for use in emergency situations where individuals, including children, are suspected of travelling for terrorist purposes.

High Court SCPOs

8.54. No Serious Crime Prevention Orders have yet been obtained against suspected terrorists in High Court proceedings. Additional powers were conferred on the police to apply such orders under the Counter-Terrorism and Sentencing Act 2021[footnote 413].

Money Measures

8.55. I have been supplied by National CT Policing Headquarters with the following figures on the use which was made of various types of financial measures in 2020 (comparable 2019 figure in brackets) by CT Policing using both terrorism and non- terrorism powers:

Type of Order Total Number of Orders/Seizures/Proceedings during 2020 Value (if applicable) during 2020
Account freezing orders 123 (53) £6.8 million
Account forfeiture orders 27 (11) £194,702
Anti-Terrorism, Crime and Security Act ongoing cash forfeiture proceedings (average per month) 4 (1)  
POCA ongoing cash forfeiture proceedings (average per month) 64 (72)  
Value new cash seizures   £2.54 million
Cash forfeiture orders granted at court 15 (26) £221,000
Confiscation orders granted at court 7 (9) £2.14 million
Restraint orders granted at court 19 (4)  

8.56. Many of these statistics were not available last year and I will refrain from making any direct comparisons. It will be interesting to see whether the comparatively high values of interim account freezing orders (£6.8 million) and cash seizures (2.54 million) translate into final orders. No figures are provided for the rate of recovery for confiscation orders.

9. NORTHERN IRELAND

Introduction

9.1. As with my two previous reports, this separate chapter on Northern Ireland is intended to ensure that issues which are particular to Northern Ireland are given the scrutiny they merit[footnote 414].

9.2. In last year’s report, I made a single recommendation: greater transparency over the use of terrorism legislation in Northern Ireland. The Secretary of State for Northern Ireland has accepted my recommendation that he should take steps to increase public understanding of the approach to countering Northern Ireland-related terrorism in Northern Ireland.

  • The need for greater public understanding is evident from the 2020 report into local policing arrangements in South Armagh[footnote 415]. On the one hand, local police representatives referred to a severe terrorist threat from violent dissident republicans which required stronger security measures than elsewhere. On the other hand, local elected representatives considered that this style of counter-terrorism policing was inappropriate because the threat was not from terrorism but criminal gangs. This likely speaks to different views in Northern Ireland as to what constitutes terrorism, and what constitutes non-terrorist criminality.

  • Greater transparency over official policy on countering terrorism allows the public to seek accountability in terms of explanations and assessments[footnote 416].

  • Publicity for counter-terrorism policy is well-established in Great Britain (in the form of CONTEST); greater transparency in Northern Ireland is consistent with the process of normalisation since the Good Friday/ Belfast Agreement[footnote 417].

9.3. In Northern Ireland, there are two other Independent Reviewers whose work overlaps with mine. In the year under review, David Seymour CB was the Independent Reviewer of the Justice and Security (Northern Ireland) Act 2007. This Act provides Northern Ireland-specific powers in connection with munitions and wireless apparatus and are the principal powers of stop and search used by the PSNI. These powers are broader than those under the Terrorism Acts, as they are concerned with preventing any risk arising from the use of munitions and not just risks arising from terrorism. David Seymour’s thirteenth and last report, covering the period 1 August 2019 to 31 July 2020, was published in April 2021.

9.4. His successor, Professor Marie Breen-Smyth, was appointed with effect from 1 February 2021 and I look forward to benefiting from her reports as I have from the clarity and rigour of Mr Seymour’s.

9.5. His Honour Brian Barker CBE QC’s term as Chair of the Northern Ireland Committee on Protection and Independent Reviewer of National Security Arrangements in Northern Ireland was extended for a period of up to one year in February 2021. The main findings of his report, dealing with the period from 1 January 2020 to 31 December 2020, were set out in a written statement from the Secretary of State for Northern Ireland in July 2021[footnote 418].

9.6. In April 2020, the Policing Board appointed a new chair and vice chair. In October 2020, John Wadham’s first report as the Policing Board’s Human Rights Adviser was published. This addressed some of the observations I have made in my earlier reports.

The Northern Ireland Security Situation

9.7. In 2020, despite the pandemic, there was no diminution in the threat level in Northern Ireland from Northern Ireland-related terrorism, which remains at “severe” (meaning that an attack is highly likely)[footnote 419]. In terms of the United Kingdom as a whole, once again Northern Ireland-related terrorism was responsible for the vast majority of ideological violence in the United Kingdom. Of the 62 terrorist incidents reported to Europol in 2020, 56 were “security-related incidents” in Northern Ireland. Unlike terrorism in Great Britain, ideological violence in Northern Ireland continues to be perpetrated with munitions and explosives: the 56 incidents included 39 shootings and 17 bombings[footnote 420].

9.8. The principal terrorist threat in Northern Ireland emanates from two groups – the new IRA (nIRA) and the Continuity IRA (CIRA). Other smaller groups, such as Arm na Poblachta (ANP) and the Irish Republican Movement (IRM) continue to have the intent to carry out attacks but lack the capability to do so. Óglaigh na hÉireann declared a cessation of attacks against the British state in January 2018.

9.9. So far as 2020 was concerned:

  • There were no “national security attacks” during 2020[footnote 421].

  • Two civilians were killed as a result of “deaths attributable to the security situation”[footnote 422].

  • There were 39 shooting incidents (the same figure as for 2018 and 2019) and 22 bombing incidents, in which 25 bombing devices were used in connection with the “security situation” (the bombing figures being higher than 2018 and 2019)[footnote 423].

  • There were a total of 57 casualties as a result of “paramilitary-style attacks”[footnote 424].

  • There paramilitary attacks were made up of 13 “paramilitary style shootings” (4 committed by Loyalist groups and 9 by Republican groups) and 44 “paramilitary style assaults” (33 committed by Loyalist groups and 11 by republican groups)[footnote 425].

  • The PSNI recovered 18 firearms, 2265 rounds of ammunition and 3.24kg of explosives[footnote 426].

9.10. The attack methodologies of CIRA and nIRA mostly involved firearms or small improvised explosive devices, such as pipe bombs. However, larger and more destructive devices such as vehicle-borne improvised explosive devices and explosively formed projectiles were also deployed in the year under review[footnote 427].

9.11. The first Covid lockdown led to an overall decrease in terrorist activity. However, dissident republicans continued to target police officers, prison officers and members of the armed forces to advance their political aims, undermining the normalisation process within Northern Ireland[footnote 428].

9.12. On 4 February 2020, an improvised explosive device was recovered attached to a commercial goods vehicle at Belfast docks. The device was discovered following a claim by CIRA that it had been planted to coincide with the United Kingdom’s departure from the EU (the ‘Brexit day bomb plot’). A command wire-initiated explosive device was recovered in June 2020 by the PSNI during a search conducted in Londonderry. The handgun used in the murder of journalist Lyra McKee (who was shot dead during rioting in April 2019) was also found during this search. A number of hoax devices were also deployed during the course of 2020 (believed to be used to test police response and tactics).

9.13. During 2020 a major joint agency operation known as Operation Arbacia was mounted against the nIRA leading to terrorism and terrorism-connected charges against eight men and two women[footnote 429]. Little can be said because criminal proceedings are currently ongoing, but the operation reveals:

  • That the authorities believe that the nIRA continues to present a significant terrorist risk over 20 years since the Good Friday/ Belfast agreement.

  • Significant joint working between PSNI and MI5 within Northern Ireland, and coordination with Police Scotland, An Garda Síochána and the Metropolitan Police.

  • A willingness to use the product of MI5 surveillance devices in criminal proceedings.

The Northern Ireland Proscribed Organisations

9.14. There was no change to the list of 14 proscribed organisations in Northern Ireland, a list that has remained unaltered since before the enactment of the Terrorism Act 2000. The failure to weed out inactive heritage groups such as Cumann na mBan despite the passage of years demonstrates that, in this respect, the proscription regime is wanting. The potency of the proscription regime demands rigour in its application, not passive tolerance of anomaly.

9.15. The Secretary of State for Northern Ireland, the minister responsible for proscription under the Terrorism Act in Northern Ireland, has never exercised his power to proscribe or deproscribe, although as I explained in my first report, the Red Hand Commando made an application for deproscription in 2018. However, this application failed to satisfy the relevant procedural criteria and so the Secretary of State never had to make a decision[footnote 430].

9.16. Admittedly, the devolution settlement calls into question how effective decisions can be made:

  • Under Schedule 2 of the Northern Ireland Act 1998, the Secretary of State has political responsibility for national security which is an “excepted matter” falling outside the competencies of the devolved administration.

  • However national security matters cannot be kept entirely separate from areas of business (such as policing, criminal justice, prisons, housing, community relations) which do fall within devolved competencies.

  • PSNI have a foot in both camps, being responsible for the investigation of all criminal conduct (including, with MI5, terrorism offences), whilst “security interface meetings” provide a bridge between the Northern Ireland Executive and those bodies with responsibility for national security (such as the Northern Ireland Office and MI5).

  • However, the devolved administration has no formal role in the Proscription Review Group that would consider proscription or deproscription with a view to advising the Secretary of State for Northern Ireland. This means, in particular, that the Secretary of State does not have formal advice from the devolved administration on community impact – whereas in Great Britain, the Home Secretary is advised by the Department for Levelling up, Housing and Communities.

  • The absence of the devolved administration may be traced to two factors in particular: political reluctance on the part of the devolved administration to engage directly with the Northern Ireland office on matters relating to terrorism, and a restrictive approach to information-sharing that appears to go further than any requirements imposed under the Northern Ireland Act 1998 on the treatment of excepted measures (which include “special powers and other provisions for dealing with terrorism or subversion”[footnote 431]).

9.17. The practical consequence of this state of affairs is that the Secretary of State for Northern Ireland may be asked to make a decision about proscription (and indeed other matters which relate to national security) without access to the fullest range of information, particularly on community impact.

9.18. This is, in truth, a further aspect of transparency: not public transparency (which I referred to in last year’s report) but transparency between bodies which exercise powers for the common good in the same part of the United Kingdom. Mechanisms need to exist which enable the relevant devolved institutions to convey their views to the Northern Ireland Office in a systematic fashion to ensure the Secretary of State has the range of information necessary to make decisions about national security (such as proscription) which are fully informed. If there are insurmountable objections about the devolved institutions engaging directly with the Northern Ireland Office on matters which pertain to national security, for example by attending a proscription review group meeting, then the PSNI may be best placed to act as a bridge between national and local bodies.

9.19. The baleful influence of proscribed organisations to many aspects of life in Northern Ireland is apparent in statistics quoted by the government in its November 2020 consultation on non-jury trials. In the context of the harm caused by “paramilitary groups” (or in other words, proscribed organisations) the government referred to statistics from the Northern Ireland Housing Executive showing that 2,643 people were driven out of their homes between 2014 and 2020 (to date) due to paramilitary and sectarian intimidation [footnote 432]. Arguably this comprises a magnitude of harm that ought to be reflected in the security situation statistics published by PSNI.

9.20. A bespoke project, Base 2, is funded by the Executive to assess the housing needs of those experiencing threats from their community or from paramilitary organisations. According to its March 2020 report, the source of alleged threats recorded by Base 2 in 2017-2018 (the latest figures shown) were loyalist (over 600), republican (over 300) and community (over 100)[footnote 433]. The flags of proscribed organisations can be used to threaten members of one community or another[footnote 434].

Investigations

9.21. In this part, I consider stop and search powers, and the use of police cordons, in Northern Ireland. Other terrorism powers which are available in Northern Ireland are considered in Chapter 4.

Stop, Search and Question

9.22. The powers of stop and search in sections 43, 43A, and 47A of the Terrorism Act 2000 exist alongside the more widely used powers in the Justice and Security (Northern Ireland) Act 2007. In summary, the most relevant powers in the 2007 Act are:

  • Section 21 – A power to stop a person for so long as is necessary to question them to ascertain their identity and movements. There is also a power to stop a person for so long as is necessary to question them to ascertain— (a) what they know about a recent explosion or another recent incident endangering life;

(b) what they know about a person killed or injured in a recent explosion or incident. It is an offence for a person to fail to stop; to fail to answer a question; or to fail to answer to the best of their knowledge and ability a question which has been addressed to them. This power includes a power to stop vehicles.

  • Section 23 – A power to enter any premises if it is considered necessary in the course of operations for the preservation of peace or the maintenance of order. An authorisation from an officer of at least the rank of superintendent must be obtained before this power can be exercised, unless it is not reasonably practicable to obtain authorisation.

  • Section 24/Schedule 3, paragraph 2 – A power to enter any premises for the purpose of ascertaining whether there are any munitions unlawfully on the premises, or whether there is any wireless apparatus on the premises. An officer may not enter a dwelling unless he is an authorised officer and they reasonably suspect that the dwelling unlawfully contains munitions or contains wireless apparatus.

  • Section 24/Schedule 3, paragraph 4 – A power to stop and search a person whom a constable reasonably suspects to have munitions unlawfully on them or to have wireless apparatus on them.

  • Section 26/Schedule 3 – These provisions extend the power to search premises to stop vehicles and to take a vehicle to any place for the purposes of carrying out a search. It is an offence to fail to stop a vehicle.

9.23. In his latest Report, David Seymour CB points out that the year under review was the fourth financial year in a row that the use of the stop, search and question powers in the Justice and Security (Northern Ireland) Act 2007 have fallen. He reports that, since his previous report, there has been:

  • a 38% decline in the use of stop and question;

  • a 20% decline in the use of stop and search of a person without reasonable suspicion;

  • a 55% decline in the search of premises; and

  • a 41% decline in the search of vehicles.

9.24. Compared with the position 9 years ago, Mr Seymour reports that the use of the power to stop and question is 77% lower and the use of the power to stop and search a person is 57% lower.

9.25. Mr Seymour explains that the decline in the use of these powers is only partly explained by the lockdown, which started on 23 March 2020. Even if there had been no pandemic, Mr Seymour reports that the use of powers in question would still have declined.

9.26. As far as the powers in the Terrorism Act 2000 are concerned, the table below shows how frequently the stop and search powers in section 43, 43A and 47A of the

Terrorism Act have been used in Northern Ireland since 2013, by calendar year[footnote 435]. It also shows the frequency with which the comparable powers in the Justice and Security (Northern Ireland) Act 2007 have been used. The reference to “TACT in conjunction with other powers” refers to the use of powers under the Terrorism Act 2000 together with powers under various other legislative provisions, such as the Misuse of Drugs Act 1971.

Section 43 Section 43A Sections 43/43A Section 47A TACT in conjunction with other powers Section 21 JSA Section 24 JSA Sections 21/24 JSA
2013 77 10 29 70 30 1497 5609 863
2014 77 4 15 0 18 1301 3660 563
2015 105 13 78 0 38 1307 4384 619
2016 91 11 92 0 34 1783 7285 986
2017 65 3 29 0 13 1163 6109 610
2018 41 2 9 0 13 1023 6052 323
2019 26 4 8 0 5 920 5003 189
2020 22 1 4 0 3 361 3519 128

9.27. Unlike in Great Britain, the self-defined ethnicity of those stopped in Northern Ireland is not published.

9.28. In 2020 there was a small decline in the number of stops carried out under section 43 of the Terrorism Act 2000. This is in the context of a 76% decline in the number of stops carried out under section 43 since 2016. Use of the other stops and search powers in the Terrorism Act 2000 has also declined significantly since 2016. The same trend is evident in the use of the powers in the Justice and Security (Northern Ireland) Act 2007. The decline in the use of the powers in the Terrorism Act 2000 in recent years may be attributable to the fact that the Paramilitary Crime Task Force, which tends to use non-Terrorism Act powers, is now fully operational.

9.29. As I reported last year, in Ramsey (No 2) the Northern Ireland Court of Appeal considered the legality of the non-suspicion stop and search powers in the Justice and Security (Northern Ireland) Act 2007. The Court of Appeal also made a number of observations on the need to monitor community background to avoid the risk of profiling people from certain ethnicities or religious backgrounds[footnote 436]. Whilst the Code of Practice which accompanies the stop and search powers did not specify any particular methodology by which the monitoring should take place, it did create a legal duty to do so and in particular a requirement “…that some proportionate measure is put in place in order to ensure that there can be adequate monitoring and supervision of the community background of those being stopped and searched[footnote 437].

9.30. Although the Justice and Security Act powers fall outside my remit, community monitoring has been advocated by the Northern Ireland Policing Board for the use of Terrorism Act powers. In a paper published in October 2013 the Policing Board recommended that the PSNI should consider how to include within its recording form the community background of all persons stopped and searched under both Terrorism and Justice and Security Act powers[footnote 438]. For that reason, I believe it is appropriate for me to consider how the PSNI has responded to the judgment in Ramsey (No 2).

9.31. In his latest report, David Seymour sets out the progress which has been made in implementing the aspect of the court’s judgment which deals with community monitoring[footnote 439]. The PSNI informed Mr Seymour on 24 November 2020 that a working group had been established to consider various methodologies and explore practical ways of capturing community background information. Mr Seymour notes that, while a good deal of work has been undertaken by this working group, given the lack of progress on this subject over the past 7 years, a sceptical observer “might view this programme of work as an attempt to ‘kick the can down the road’”[footnote 440]. It could be argued that this programme of work is unnecessary, as all that is required is a separate assessment, after the event, based on intelligence, of existing information and officer perception of the individual’s background. Mr Seymour argues that this should not be difficult for the following reasons:

  • the PSNI stress that the powers are used, almost exclusively, on an intelligence led basis, against those who present the greatest threat;

  • it would be anonymised and generic data – an overarching set of percentages indicating broad categories;

  • it would be similar to the information referred to by the Lord Chief Justice in paragraph 26 of his judgment which referred to statistics for the 2013/2014 period in relation to repeat stop and searches - 81% Dissident Republicans, 7% criminal associations, 3% loyalist associations, 1% related interface disorder[footnote 441] and 8% unspecified;

  • the PSNI’s own security statistics are broken down in this way into

Republican/Loyalist categories

9.32. As I said last year, I suspect that, however the PSNI chooses to go about it, community monitoring will likely reveal that the powers in the Justice and Security (Northern Ireland) Act 2007 are directed towards dissident republicans who come from a particular part of the community in Northern Ireland. As the court stated in Ramsay (No 2), “…in light of the nature and threat from [dissident republicans] it would come as no surprise to anyone in Northern Ireland that the impact on exercise of this power was more likely to be felt by the perceived catholic and/or nationalist community”[footnote 442]. The powers in the Terrorism Acts are overwhelmingly used against dissident republicans, so community monitoring would reveal a similar, if not greater, trend. So community monitoring may prove to be a distraction, but I will reserve further comment until the PSNI begins to gather the relevant data.

Cordons

9.33. The following table sets out the number of designated cordons in place in each year since the Terrorism Act 2000 was enacted[footnote 443]. There has been a significant decline in the use of cordons in Northern Ireland. The figures for 2019 and 2020 are set out on a calendar year basis alongside the figures for England and Wales in Chapter 4.

Year Number of designated cordons
2001 62
2002 239
2003 175
2004 126
2005 72
Year Number of designated cordons
2006 38
2007 29
2008 59
2009 102
2009/10 128
2010/11 120
2011/12 87
2012/13 57
2013/14 55
2014/15 45
2015/16 43
2016/17 29
2017/18 16
2018/19 18
2019/20 17
2020/21 20

Arrest and Detentions

9.34. The powers of arrest in section 41 of the Terrorism Act 2000 are set out in Chapter 5. In Northern Ireland, there were a total of 79 arrests made under section 41 of the Terrorism Act 2000 in 2020 (73 less than in the previous year)[footnote 444] relating to the security situation. According to HHJ Brian Barker QC, the pandemic caused a decrease in police activity which fed this sharp decline in arrests under terrorism legislation, although there was an increase in the recovery of ammunition and explosives[footnote 445].

9.35. It will be noted that the official PSNI statistics for the use of section 41 relate to the “security situation” only, therefore to Northern Ireland-related terrorism, whereas persons who have been arrested under section 41 for other reasons are excluded[footnote 446]. The section 41 arrest power is a key aspect of the Terrorism Act 2000, and it is not possible to understand the operation of section 41, or indeed to review it (as I am obliged to do under section 36 Terrorism Act 2006) without official statistics that are all-inclusive. Terrorism statistics ought, as in Great Britain, to be ideology-neutral and I recommend that the PSNI’s published statistics should include all arrests under section 41.

9.36. By comparison in Great Britain, there were 26 arrests during 2020[footnote 447]. Once again, Northern Ireland has accounted for a high proportion of the arrests made under section 41 of the Terrorism Act 2000. This year the figure was 75% of all section 41 arrests in the United Kingdom (last year it was 77%).

  • There is no legal requirement for the police to use section 41 arrest powers when arresting for terrorism-related activity. In fact, in Great Britain the vast majority (86%) of arrests for terrorism-related activity are carried out under non-terrorism arrests powers, i.e. under PACE[footnote 448].

  • PACE powers do not enable the long periods of pre-charge detention available under section 41 Terrorism Act 2000 (up to 14 days) but do allow release on bail, which may be a useful accommodation between the demands of the investigation and the protection of the public in lower threat investigations.

  • However, PSNI take the view that arrests for terrorist-related activity ought to be carried out using terrorism powers for reasons relating to public perception: that terrorism is being taken seriously (because more serious arrest powers are being used) and that there is nothing underhand in the manner of the investigation (which might be suggested by the use of non-terrorism powers in relation to suspected terrorism). The recent Pitt Park arrests were followed by a PSNI announcement that the arrests were “conducted under the Terrorism Act”[footnote 449].

9.37. Whilst this does offer some explanation for the rate of section 41 use in Northern Ireland, it is dismaying that the choice of arrest power is affected by reasons relating to public perception.

9.38. Firstly, if a less intrusive investigative measure can be used against an individual who is after all only suspected of criminal wrongdoing, it should be. Section 41 is a more restrictive detention regime than detention under PACE (the Police and Criminal Evidence (Northern Ireland) Order 1989) because it permits deprivation of liberty for up to 14 days. Reasons relating to public perception should not be used to interfere with fundamental rights.

9.39. Secondly, it may be sensible for PSNI to have the option of releasing under investigation on conditional bail. Conditional bail may provide greater protection to the public if further investigation is required before charge.

9.40. Thirdly, it is a requirement of section 31A Police (Northern Ireland) Act 2000 that officers shall carry out their functions with the aim, inter alia, of securing the support of the local community. Community confidence is unlikely to be served if the PSNI use arrest powers based on public perception rather policing need. Nor does securing community support mean pandering to the irrational, and it is irrational to suggest that police in Great Britain are taking terrorism any less seriously because they mainly arrest under PACE. Policing to public perception also carries the risk of using police powers in order to “even out” in order to avoid the oft-cited suggestion of two-tier policing[footnote 450].

9.41. Of the 79 people detained under section 41 of the Terrorism Act 2000, there were 13 applications for warrants of further detention and no refusals[footnote 451].

9.42. The 79 arrests made under section 41 resulted in 14 people being charged with an offence (4 less than last year). This represents a charge rate of 18% (which is marginally better than last year’s figure of 12%)[footnote 452]. In 2020, 13 people were convicted of terrorism offences[footnote 453].

9.43. Both and I my immediate predecessors have consistently made the point that the charge rate following a section 41 arrest in Northern Ireland appears to be anomalous compared to the comparable figure in Great Britain. This year the charge rate in Northern Ireland following arrest under section 41 was 18%, while in Great Britain it was 50%.

9.44. I explored this issue with the PSNI in the preparation of last year’s report and, as I have already explained, I was informed that the default position in Northern Ireland when an arrest for a terrorism related offence is being planned is to rely upon section 41 of the Terrorism Act 2000. Bearing in mind that in Great Britain the vast majority of arrests for terrorism-related activity are carried out using non-section 41 powers, the better comparison may be between, in Northern Ireland, the charge rate following section 41 arrest and, in Great Britain, the charge rate following both section 41 and non-section 41 arrest. The latter rate, for the 2020, is 30% although in previous years it has been between 37% and 55%[footnote 454]. The use of section 41 is therefore unlikely to provide an explanation for the low charge rate in Northern Ireland[footnote 455].

9.45. I have been informed that the PSNI is considering commissioning a working group to review current practices on the use of section 41 of the Terrorism Act 2000. This is a welcome development but does not preclude me from making the following recommendation: PSNI should not take account of public perception when deciding on the appropriate arrest power for terrorist-related activity.

Conditions of detention

9.46. Independent Custody Visitors in Northern Ireland are trained and coordinated by the Northern Ireland Policing Board. Unlike in Great Britain, there is no statutory requirement in Northern Ireland for custody visitors’ reports to be sent to me, but in practice they are.

9.47. In my previous reports, I remarked that the forms used by custody visitors in Northern Ireland differ from those which are used in Great Britain. In my first report I recommended that the Northern Ireland Policing Board ensure that independent custody visitors all use the recommended form. In its response to my first Report, the government informed me that the Northern Ireland Policing Board is currently undertaking a detailed review of the capture and reporting of custody visiting statistics, which would include the feasibility of independent custody visitors in Northern Ireland using the form recommended by the Independent Custody Visitors’ Association for those detained under the Terrorism Acts.

9.48. During the course of 2020, I was informed by the Northern Ireland Policing Board that the form which is used in Northern Ireland has been amended to ensure that it replicates the key sections of the form which is used in Great Britain. I welcome this development. This amended form was piloted during the course 2021. I hope to be in a position to report on its use in next year’s report.

9.49. The table below sets out information provided to me by the Policing Board of Northern Ireland about the independent custody visits which took place in Northern Ireland in 2020. All detainees were arrested under section 41 Terrorism Act 2000.

2020 Detainees Valid visits Invalid visits Seen by ICVs CCTV reviews Unsatisfactory visits
- 79 34 4 15 1 0

9.50. I have remarked in previous reports that the number of detainees who consent to being visited by an independent custody visitor in Northern Ireland is low. The Policing Board has acknowledged that the rate remains low, despite a change in policy to allow self-introduction to individuals who are detained under section 41 of the Terrorism Act 2000. The Policing Board has informed me that it intends to revisit this issue in the next Human Rights Annual Report. This is a development I welcome and I look forward to working with the Board in its efforts to address these consistently low rates.

9.51. In 2020 a new Code of Practice was promulgated enabling the use of new technologies to record interviews following detention under section 41 and Schedule 7 Terrorism Act 2000[footnote 456], bringing the position into line with England and Wales and Scotland.

Stopping the Travelling Public

9.52. Schedule 7 of the Terrorism Act 2000 allows officers to examine those travelling through ports or borders to determine if they are terrorists; to search them; to detain them; to require them to hand over electronic devices for examination; and to take their fingerprints. Failure to cooperate with an examination is a criminal offence.

9.53. As I explained last year, a Schedule 7 examination can take place at a port or in “the border area”. Paragraph 4(1) of Schedule 7 provides that a place in Northern Ireland is within the border area if it is no more than one mile from the border between Northern Ireland and the Republic of Ireland. By virtue of paragraph 4(2) if a train goes from the Republic of Ireland to Northern Ireland, the first place in Northern Ireland at which it stops for the purpose of allowing passengers to disembark is within the border area for the purposes of conducting a Schedule 7 examination.

  • This latter paragraph exists to accommodate the direct train route which runs between Belfast and Dublin. The first place in Northern Ireland at which the train stops is Newry, a town approximately 8 kilometres from the border with the Republic of Ireland. In addition, authorisations have been in force on a continuous basis in recent years under Schedule 3 to the Justice and Security (Northern Ireland) Act 2007, which enable officers throughout Northern Ireland, including border areas, to stop people and vehicles to look for munitions and wireless apparatus on a no-suspicion basis.

9.54. In my previous two reports, I committed to considering whether retaining a power to examine at a land border under Schedule 7 is justified. However, I felt unable to fulfil this commitment until the outcome of Brexit was known. Although the United Kingdom has now left the EU, the transition period did not end until 31 December 2020. As a result, I remain of the view that it would be premature to draw any conclusions about whether the continued existence of the power in paragraph 4(1) of Schedule 7 is justified.

9.55. As in Great Britain, there has been a decline in the number of Schedule 7 examinations in Northern Ireland.

9.56. This is both part of a long-term trend and a consequence of restrictions on travel during the pandemic[footnote 457].

Year Number of stops
2016 2082
2017 1248
2018 717
2019 559
2020 119

9.57. In terms of detentions, in 2017, 11 people were detained. In 2018, 6 people were detained. In 2019, 31 people were detained. In 2020, 8 people were detained.

9.58. The pandemic means that it is impossible to draw any conclusions from these statistics.

9.59. As with previous years, I obtained the figures on self-defined ethnicity directly from the PSNI as they are not published.

Total examinations

2019 2020
White 55% 38%
Mixed 10% 8%
Black 4% 13%
Asian 13% 16%
Chinese or other 10% 17%
Not stated 7% 8%

Detentions

2019 2020
White 13% 13%
Mixed 19% 13%
Black 6% 13%
Asian 26% 25%
Chinese or other 23% 0%
Not stated 13% 38%

9.60. I have no reason to question the impression I have formed in previous years that the PSNI is careful in its use of Schedule 7. I have, however, noted the potential problems with relying upon intuitive stops and, in last year’s report, regretted the fact that ethnicity and community background data is not published. The Policing Board has recommended to the PSNI that it should review its collection and publication of Schedule 7 data[footnote 458].

9.61. I am pleased to report that PSNI Schedule 7 coordinators are encouraging greater openness and transparency in the exercise in Schedule 7 powers[footnote 459], and seeking to minimise the duration of phone seizures (under paragraph 11, property may be detained for the purposes of examination for up to 7 days). Like their colleagues in Great Britain, PSNI officers are struggling with the demands of retention, review and disposal of digital data which I detail in Chapter 6.

9.62. In last year’s report I referred to 6 outstanding cases of failing to comply with Schedule 7 examinations[footnote 460]: the outcomes were: 1 caution, 2 convictions, 2 cases withdrawn, 1 acquittal[footnote 461].

9.63. As I explained last year, the judgment of the Northern Ireland Court of Appeal in Ramsay (No 2) may have implications for the use of Schedule 7, as both powers can be exercised without reasonable suspicion[footnote 462]. As the PSNI has yet to implement the community monitoring required by the Northern Ireland Court of Appeal, I am not yet in a position to consider what, if any, relevance this may have to the exercise of the power in Schedule 7.

9.64. In terms of freight, in the year under review there were 19 examinations of unaccompanied freight (down from 255 in 2019).

Brexit

9.65. The transition period did not end until 31 December 2020, so the impact of the United Kingdom’s departure from the EU upon Northern Ireland was not fully felt in the year under review.

9.66. The Northern Ireland Protocol came into force at the start of 2021 and requires certain goods travelling between Great Britain and Northern Ireland to be checked upon entering Northern Ireland. This so-called “sea border” between Northern Ireland and the rest of the United Kingdom has inflamed tensions within certain loyalist communities. For example, in February 2021 the media reported that (as yet unverified) threats had been made against staff making Brexit-related checks at

Northern Ireland’s ports[footnote 463]. This is something for next year’s report.

Terrorist Trials, Sentencing, and Criminal Justice

9.67. TPIMs have never been used, or proposed for use, in Northern Ireland. Viewed purely in instrumental terms, the criminal justice system is therefore the only means by which medium or long-term disruption of terrorists can be achieved. In evidence to the Intelligence and Security Committee (“ISC”) of Parliament in October 2019[footnote 464], MI5 stated that criminal justice outcomes were their “preferred course of action wherever achievable” and as “the critical tool to successful and long-term disruption” of the threat posed by Northern Ireland-related terrorism.

9.68. For the third successive year, I regret to report that aspects of the Northern Ireland criminal justice system continue to undermine that aspiration. The ISC confirmed in its report of October 2020 that criminal trials suffer from systemic delays, a point illustrated by the ongoing trial of Colin Duffy[footnote 465]. The ISC concludes that shorter sentences are imposed for terrorism offences than in the rest of the United Kingdom[footnote 466] and that the criminal justice system is “in urgent need of an overhaul”[footnote 467]. I agree that the slow pace and procedural heaviness of criminal proceedings in Northern Ireland has a deleterious impact on the use of terrorism legislation[footnote 468]. In its third report published in November 2020, the Independent Reporting Commission repeatedly drew attention to the importance of criminal justice reforms as part of bringing paramilitarism to an end in Northern Ireland[footnote 469].

Delay

9.69. In last year’s report I drew attention to the issue of oral committal hearings and proposals for their reform[footnote 470]. The Criminal Justice (Committal Reform) Bill was introduced into the Northern Ireland Assembly in November 2020. The Bill successfully completed Final Stage on 14 December 2021 and Royal Assent is now being sought.

9.70. The new legislation abolishes committal hearings for all offences that, in the case of an adult, can only be tried on indictment[footnote 471]. Additional offences exempted from committal hearings may be added by way of secondary legislation. The difficulty with this phased approach is that some offences of the sort typically committed by terrorists, such as collecting information likely to be of use to a terrorist (section 58 Terrorism Act 2000), weapons training (section 54 Terrorism Act 2000) or encouraging terrorism (section 1 Terrorism Act 2006), will not immediately be impacted by the changes. As a result, committal hearings will be abolished for some terrorism offences, but not immediately for others which will include those frequently linked to paramilitary groups[footnote 472].

9.71. I remain of the view that this is a regrettable state of affairs. It would be preferable for committal hearings to be abolished for all terrorism offences at the same time. In Northern Ireland terrorism trials tend to suffer even more inordinate delay than trials for other types of offence, a problem which is at least partly attributable to the continued existence of committal hearings.

9.72. Oral committal hearings are not the only cause of criminal justice delay. Even allowing for the impact of the pandemic during 2020 and early 2021, the current progress of a high-profile terrorism prosecution is telling: charged in late 2013, the accused were released on bail in early 2016, the trial began in March 2019 and, as at the end of October 2021 was still ongoing. Significant cultural change to the way the criminal justice system operates is still needed: whilst a new case management practice direction was issued by the Office of the Lord Chief Justice in November 2019[footnote 473] it provides that “It does not change, but rather builds upon, the practice introduced in the Crown Court since 2011”.

Sentencing

9.73. In November 2020 seven members of CIRA pleaded guilty in what PSNI described as “one of the most significant terrorism cases in recent times”[footnote 474] (another case involving delay: the main defendant Patrick “Mooch” Blair was arrested in 2014 and was on bail since July 2015). The offences included being a member of a proscribed organisation, providing weapons and explosives training, conspiracy to possess explosives, firearms, and ammunition with intent to endanger life, and preparing acts of terrorism. The starting points apparently taken by the sentencing judge (14, 4 and half, 10, 14, 8 and a half, 5, 5 and a half years[footnote 475]) were very significantly lower than those that apply in England and Wales[footnote 476].

9.74. Although the unduly lenient scheme has now been extended to terrorism offences[footnote 477], there will be cases where, in the absence of clear sentencing guidelines or guideline cases, it is difficult for an appellate court to conclude that such a sentence is not just lenient but unduly lenient.

9.75. The fundamental question which arises is how greater consistency can be achieved.

  • The outcome of the Northern Ireland Department of Justice’s Sentencing Review, which commenced in 2019, was that there was insufficient evidence to support the case for a sentencing guideline body akin to those established in Scotland and England and Wales[footnote 478].

  • A sentencing guideline body would at least be forced to choose whether to adopt the starting points for terrorism sentences in England and Wales, or make a conscious decision to treat terrorist offending less seriously in Northern Ireland.

  • The Review’s recommendation that legislation be enacted to enable the Northern Ireland Court of Appeal to provide guideline judgments does nothing to address the disparity between sentencing in Northern Ireland and the rest of the United Kingdom, and in any event the Northern Ireland judiciary already issues guideline cases including some limited ones for terrorism[footnote 479].

9.76. The sentencing guidelines that apply in England and Wales do not apply in Northern Ireland. They were not formulated with the input of the judiciary or legal profession or other interested persons in Northern Ireland. But, as the Recorder of Belfast observed in a recent case[footnote 480], terrorism offences apply throughout the UK and there is a strong presumption that there should be consistency of sentencing principles throughout the UK. The possibility that linked defendants should be given different sentences depending on whether they are arrested and prosecuted in Northern Ireland or Great Britain[footnote 481] is unacceptable. The Recorder therefore held that the leading English authority on sentencing offences under section 5 Terrorism Act 2006[footnote 482] was “of assistance”[footnote 483].

9.77. Sentencing Council Guidelines from England and Wales on attempted murder were also referred to by the Court of Appeal with approval in the terrorist sentencing case of Christine Connor in 2021[footnote 484]. There appears to be no reason in principle why guidelines from England and Wales, whilst in no way binding, should not be taken into account when sentencing terrorism cases.

9.78. In a limited number of cases a further spur towards greater consistency may be the requirement for minimum mandatory “serious terrorism sentences” for certain dangerous terrorists under the Criminal Justice (Northern Ireland) Order 2008 (as amended the Counter-Terrorism and Sentencing Act 2021). However, there are reasons to question the extent to which this legislation will be applied in Northern Ireland in practice: firstly, the Probation Service of Northern Ireland has not traditionally carried out assessments of dangerousness in cases of terrorist or politically-motivated violence[footnote 485], meaning that judges must make their own assessments unassisted; secondly, the blurred lines between terrorism and ‘paramilitary activity’ complicate the assessment of whether a non-terrorism offence, such as attempted murder, is nonetheless connected to terrorism under the Counter Terrorism Act 2008.

9.79. An additional consideration arises from the Counter-Terrorism and Sentencing Act 2021, and the current direction of sentencing policy generally, which mandates longer sentences, with fewer opportunities for early release, and more onerous post- release commitments. Severe measures like these call for strong public confidence that they are being deployed fairly and consistently. In correspondence with the UN Rapporteur on the Counter-Terrorism Sentencing Bill, the government referred, incorrectly, to Northern Ireland having an equivalent body to the Sentencing Council in England and Wales; but did correctly recognise that guidelines are needed to “…support both the Courts and the public in understanding the changes made”[footnote 486].

9.80. In order to achieve greater consistency in terrorism sentencing between Northern Ireland and the rest of the United Kingdom, I recommend that the Director of Public Prosecutions for Northern Ireland seeks an authoritative ruling from the court, at the earliest opportunity, on whether the terrorist sentencing guidelines issued by the Sentencing Council in England and Wales or the Scottish Sentencing Council should be considered (not followed) for the purpose of sentencing terrorism cases in Northern Ireland. Considering guidelines places no constraints on the discretion or independence of sentencing judges to achieve just outcomes[footnote 487]. It is open to the Northern Ireland judiciary to provide much-needed clarity on whether sentencing levels in England and Wales are relevant, and if so how.

Other matters

9.81. In 2020 the Northern Ireland Office published a consultation on whether the non-jury trial provisions in the Justice and Security (Northern Ireland) Act 2007 should be renewed until 2023. My response[footnote 488] echoed that of David Seymour CB, who recommended the continuation of non-jury trials at the present time. The Justice and Security (Northern Ireland) Act 2007 (Extension of Duration of Non-jury Trial Provisions) Order 2021 was made on 21 July 2021, and extends non-jury trials in Northern Ireland until 31 July 2023.

9.82. I also referred to the fact that the Justice Minister stated that commencing the provisions in the Criminal Finances Act 2017 which would empower the Northern Ireland High Court to make unexplained wealth orders was one of her key priorities. Although some of the provisions in the Criminal Finances Act 2017 have been commenced (from 17 June 2021), this does not include the provisions in the Act which deal with unexplained wealth orders[footnote 489], despite earlier indications to the Independent Reporting Commission[footnote 490]. As I explained last year, these orders have been used against a London woman suspected of being involved in paramilitarism and should be available for use in Northern Ireland.

10. SCOTLAND

10.1. The Terrorism Acts apply to Scotland because national security and special powers for dealing with terrorism are reserved matters under the Scotland Act 1988[footnote 491], although their operation is modified somewhat by Scotland’s different legal system[footnote 492].

Legislation

10.2. Unlike Northern Ireland, the Terrorist Offenders (Restriction of Early Release) Act 2020 extended to Scotland. Legislative consent from the Scottish Parliament was sought and granted[footnote 493].

10.3. The position was different with respect to the use of polygraphs against terrorist offenders provided for by the Counter-Terrorism and Sentencing Bill. The Scottish government indicated that it would not promote a legislative consent motion for polygraphs in licence conditions and the relevant clauses amending procedure in Scotland were withdrawn. Consent was granted for the remainder of the Bill which became an Act in 2021[footnote 494].

10.4. The Lord Advocate accepted the recommendation in my previous report[footnote 495] that a Code of Practice on the Detention of Individuals Under Section 41 and Schedule 8 of the Terrorism Act 2000, or its equivalent, should be issued in place of collection of earlier guidance which lacked detail and was not openly available[footnote 496]. I can report that the Lord Advocate will be issuing Lord Advocate’s Guidelines under section 12 of the Criminal Procedure (Scotland) Act 1995 which will provide equivalent protections to those found in Code H in England and Wales. Such Guidelines are legally binding on Police Scotland[footnote 497].

Arrests

10.5. 7 individuals (6 male and 1 female) were arrested by Police Scotland in 2020 in connection with counter-terrorism investigations.

  • All arrests were under section 1 Criminal Justice (Scotland) Act 2016, save one arrest carried out on an arrest warrant from Northern Ireland on suspicion of attempting to murder a member of the Police Service of Northern Ireland. It follows that no arrests were made using the special arrest power of section 41 Terrorism Act.

  • All but two of the arrests were on suspicion of Northern Ireland Related Terrorism. One individual was arrested on suspicion of Islamist terrorism: Firoz Madhani who subsequently pleaded guilty to Twitter posts encouraging terrorism contrary to section 1 Terrorism Act 2006[footnote 498]. Another individual was arrested in connection with suspected Right Wing Terrorism.

  • Unlike the position in England and Wales there were no juvenile arrests.

10.6. A further 2 individuals were arrested by Police Scotland’s Border Policing

Command for failing to comply with Schedule 7 Terrorism Act 2000 examinations.

Criminal trials

10.7. The Crown Office and Procurator Fiscal Service (COPFS) pursued 4 prosecutions in 2020 resulting in:

  • The conviction of Gabrielle Friel for possessing articles (including a machete and crossbow) for a purpose connected with the commission, preparation, or instigation of an act of terrorism (section 57 Terrorism Act 2000). The defendant was subsequently sentenced to an extended sentence of 15 years (10 years imprisonment with a 5-year extension period) together with a serious crime prevention order of 5 years to commence on his release from prison[footnote 499]. A charge of attack-planning (section 5 Terrorism Act 2006) was found “not proven”[footnote 500]. Friel had a previous (non-terrorist) conviction for stabbing a police officer. The sentencing judge noted that the defendant spent 12 hours or more each day surfing the internet and that this had “been extremely damaging for you”.

  • An acquittal in respect of publishing images in a way likely to arouse suspicion that the individual is a member of a proscribed organisation: the defendant had uploaded an image of himself to Facebook wearing a headband bearing the words “up the RA”.

  • A plea of guilty to failing to comply with a Schedule 7 examination.

  • An acquittal in respect of failing to comply with a Schedule 7 examination: the defendant pleaded guilty to other (non-terrorist) charges.

10.8. The trial and conviction of Gabrielle Friel is notable because the sole “political, religious, racial or ideological cause” which the COPFS sought to establish that Friel had the purpose of advancing to demonstrate a terrorist purpose was inceldom[footnote 501]. An expert gave evidence before the jury about the nature of this ideology.

  • To convict the defendant of the offence contrary to section 57 Terrorism Act 2000, the jury must have been sure that Friel had the weapons in circumstances which gave rise to a reasonable suspicion that his possession was for a purpose connected with the commission, preparation or instigation of an act of terrorism[footnote 502].

  • This did not require proving the precise subjective intention of the defendant[footnote 503] but did require COPFS to establish that carrying out an act of serious violence to advance the incel cause could in principle amount to terrorism. This is the first and only occasion in which incel ideology has led to a terrorist conviction in the UK.

  • The jury must have also been sure that Friel’s defence – that he had the weapons not for terrorist purposes but so that the police would kill him on arrest (sometimes referred to as “suicide by cop”) – was untrue[footnote 504]. The defence did not dispute that inceldom could amount to a terrorist cause.

  • Because the jury acquitted the defendant of the offence contrary to section 5 Terrorism Act 2006, the jury cannot have been sure that Friel went further and engaged in preparatory conduct for giving effect to an intention to commit acts of terrorism.

  • Overall, the Friel case illustrates the special anticipatory reach of terrorism legislation. A person who did store weapons that are not firearms or explosives with an intention to carry out an atrocity in a public place would not commit an offence under the general criminal law unless another person was involved by way of an agreement (conspiracy) to commit an offence. An offence is only committed when he steps outdoors with the weapons[footnote 505]. But if the circumstances give rise to a reasonable suspicion that the object is held, even at home, for terrorist purposes, or if it can be shown that the weapon has been obtained preparatory to carrying out a terrorist attack, offences are committed under section 57 Terrorism Act 2000 or section 5 Terrorism Act 2006,.

10.9. During 2020, the Appeal Court dismissed an appeal against sentence by David Dudgeon who had been convicted in 2019 of possessing material likely to be useful to a terrorist (section 58 Terrorism Act 2000). The Appellant, who suffered from poor mental health, had a sustained interest in far-right material and had frequently expressed violent and extremist views. The terrorism investigation resulted from disclosures made to his treating psychiatrists. A sentenced of 2 years imprisonment with a supervised release order of 12 months[footnote 506] was upheld[footnote 507].

Search warrants

10.10. In August 2020, a search warrant under the Terrorism Act 2000 was executed in Edinburgh in connection with the PSNI’s investigation into the activities of the New IRA, Operation Arbacia.

Stop and Search

10.11. I have been provided with figures on the use of stop and search powers under section 43 Terrorism Act 2000. Whilst these are not published separately, they are now to be fed into national statistics[footnote 508]. I can say that the section 43 power is used only very occasionally by Police Scotland.

Ports

10.12. Schedule 7 statistics are not published separately for Scotland.

10.13. Publicity was given to the 2020 examination under Schedule 7 of a local human rights campaigner, leading to public observations by the First Minister[footnote 509]. I have asked Police Scotland for more details of this to see whether the stop raises any individual or thematic points about the use of Schedule 7, and will report on this in next year’s report.

10.14. Police Scotland have drawn my attention to a potential anomaly caused by the interaction between Scottish procedure and the law governing persons detained under Schedule 7. Whereas paragraph 10 of Schedule 8, which applies to England, Wales and Northern Ireland, enables fingerprints to be taken at port with consent and without consent at a police station, paragraph 20, which applies to Scotland, appears to limit the taking of fingerprints to police stations, whether consent is given or not. This paragraph provides that subject to immaterial modifications:

“(1)…section 18 of the Criminal Procedure (Scotland) Act 1995 (procedure for taking certain prints and samples) shall apply to a person detained under Schedule 7 or section 41 at a police station in Scotland as it applies to a person arrested or a person detained under section 14 of that Act.”

10.15. Although it is just about possible to read “under Schedule 7” as a separate qualifier from “[under] section 41 at a police station in Scotland”, the more natural reading of this section is that the power to take fingerprints only arises at a police station, whether detained under Schedule 7 or section 41. This interpretation is reinforced by the fact that the provisions relating to Scotland do not distinguish between taking of fingerprints with or without consent.

10.16. Given the great distances that may apply in Scotland, and the desirability of avoiding long periods of detention where a person is willing to have their fingerprints taken at port, I recommend that paragraph 20 of Schedule 8 is amended so that the power to take fingerprints applies with consent at a port.

11. RECOMMENDATIONS Introduction (Chapter 1)

11.1. The Home Office and CT Police should give consideration as to how to ensure that statistics on the use of terrorism powers can continue to capture useful information about ethnicity (1.11).

11.2. The use of “Chinese or other” as an ethnicity category in CT statistics should be reconsidered so that it more accurately reflects the individuals within that category (1.13).

11.3. Consideration be given to publishing the statistics for “White Irish” individuals stopped under Schedule 7 (1.13).

Terrorist Groups (Chapter 3)

11.4. The Home Secretary should provide greater clarity over how the five public discretionary factors in favour of or against proscription under section 3 Terrorism Act 2000 operate against predominantly online groups (3.10).

11.5. Legislation should be enacted to enable a court sentencing an individual for a terrorism or terrorism-connected offence to recommend that the power in section 3 Terrorism Act 2006 be exercised by a constable (3.32).

Ports and Borders (Chapter 6)

11.6. Information on complaints about the exercise of Schedule 7 should be routinely captured from all police forces across the United Kingdom (6.27).

11.7. National Counter-Terrorism Policing HQ should analyse ethnicity categories for those subject to tasked examinations compared to untasked examinations (6.34).

11.8. Paragraph 8 of Schedule 7 to the Terrorism Act 2000 should be amended to enable the proportionate searching and copying of remotely held data, to be accompanied by an amended Code of Practice (6.48).

11.9. Counter-Terrorism Police immediately establish a new standalone public policy on CT intelligence management which explains, as far as is consistent with national security, how data obtained from Schedule 7 is managed, reviewed, retained or deleted. The policy should explain what controls there are on access to this data and what, if any, oversight there is on the integrity of the retention regime (6.68).

11.10. The Code of Practice should be amended to refer to the above new policy (6.68).

Civil Powers (Chapter 8)

11.11. The Home Office and MI5 should formulate general internal guidance on evaluating risk reduction during the currency of a TPIM (8.25).

11.12. The Home Office should, in cases involving neurologically atypical individuals, consider whether the attendance of a psychologist at TRGs may be useful when evaluating risk and measures to reduce risk (8.29).

11.13. An order should be made under section 11(6) Legal Aid, Sentencing and Punishment of Offenders Act 2012, exempting TPIM proceeds from the criteria referred to in that section (8.35).

Northern Ireland (Chapter 9)

11.14. PSNI’s published statistics should include all arrests under section 41, not just those related to the ‘security situation’ (9.35).

11.15. PSNI should not take account of public perception when deciding on the appropriate arrest power for terrorist-related activity (9.45).

11.16. The Director of Public Prosecutions for Northern Ireland should seek an authoritative ruling from the court, at the earliest opportunity, on whether the terrorist sentencing guidelines issued by the Sentencing Council in England and Wales or the Scottish Sentencing Council should be considered (not followed) for the purpose of sentencing terrorism cases in Northern Ireland (9.80).

Scotland (Chapter 10).

11.17. Paragraph 20 of Schedule 8 should be amended so that the power to take fingerprints applies with consent at a port in Scotland (10.16).

12. ANNEX: RESPONSE TO PREVIOUS RECOMMENDATIONS

In 2022, the Home Secretary responded to my Terrorism Acts in 2019 report.

Terrorist Groups (Chapter 3)

12.1. Home Office officials and National Crime Agency officers should meet with aid agencies within the Tri Sector Working Group to consider (and ‘workshop’) the situations identified at 3.26 with a view to formulating guidance on the use of section 21ZA in connection with humanitarian assistance (3.34). Accepted

Terrorist Investigations (Chapter 4)

12.2. Consideration should be given by the Home Secretary to whether new or amended powers are needed for police to compel encryption keys in counter-terrorism investigations (4.30). Accepted

12.3. The government should make arrangements, in consultation with the judiciary, to publish all first instance judgments on applications for journalistic material under Schedule 5 Terrorism Act; and, where publication has to be delayed on the grounds of prejudicing a forthcoming trial, to ensure that judgments are available for use in other cases (4.51). Under consideration

Arrest and Detention (Chapter 5)

12.4. CT Police and the Home Office should consider whether section 41 Terrorism Act 2000, and the time at which the detention clock starts to run, deals adequately with persons arrested for terrorism offences in hospital (5.13). Accepted

12.5. CT Police Headquarters should modify the forms completed by arresting officers so that any use by police superintendents of the power under paragraphs 8 and 9 of Schedule 8 Terrorism Act 2000 is clearly recorded, and the data gathered (5.27). Accepted

Ports and Borders (Chapter 6)

12.6. CTP Borders should draw up a policy in which the distinction between “screening” (using the power to enter under paragraph 9(4) of Schedule 7), and formal examination of goods, is clearly delineated (6.50). Accepted

12.7. CT Police training materials on the revised Schedule 7 Code should make it clear that Schedule 7 does not authorise the use of journalistic or legally privileged material (6.58). Accepted

Terrorism Trial and Sentences (Chapter 7)

12.8. The Home Secretary should invite the Director of Public Prosecutions (in England and Wales), the Director of Public Prosecutions (in Northern Ireland) and the Lord Advocate (in Scotland) to ensure that their prosecution services make a record of whether amended or new offences are charged for a period of 5 years from the relevant amending or creating legislation (7.9). Accepted

Civil Powers (Chapter 8)

12.9. The Secretary of State should keep under review the question of whether there either currently exists or might reasonably be obtained evidence that gives rise to a realistic prospect of conviction of a TPIM subject (8.52). Accepted

12.10. In considering the proportionality of a TPIM and its measures, the TPIM review group should expressly identify the passage of time since the previous TPIM review group meeting as a factor weighing against continuation (8.58). Accepted

12.11. The government should ensure that, subject only to means, legal funding is swiftly made available to all TPIM subjects for the purpose of participating in section 9 review hearings (8.70). Rejected

Northern Ireland (Chapter 9)

12.12. The Secretary of State for Northern Ireland should take steps to increase public understanding of its approach to countering Northern Ireland-related terrorism in Northern Ireland (9.41). Accepted

Scotland (Chapter 10)

12.13. Lord Advocate should issue a Code of Practice on the detention of individuals detained under section 41 and Schedule 8 Terrorism Act 2000 (10.14). Accepted

In 2022, the Home Secretary also responded to some recommendations made in the Terrorism in 2018 Report which were previously under consideration.

Investigations (Chapter 4)

The Home Office and CT Police should consider whether the 2012 Code of Practice on section 47A, which is now several years old, requires revision (4.18). Accepted

Arrest and Detention (Chapter 5)

Section 41 Terrorism Act 2000 should be amended so that the “relevant” time includes the time of arrest under the Police and Criminal Evidence Act 1984 for specified terrorist offences (5.29). Accepted

Civil Powers (Chapter 8)

The Home Secretary should consider whether Temporary Exclusion Orders should be available for individuals other than British citizens (8.61). Rejected

E02743840 978-1-5286-3308-6

  1. Zedner, L., ‘Terrorizing Criminal Law’, Crim Law and Philos (2014) 8:99-121 (2014). 

  2. Section 41 and Schedule 8 Terrorism Act 2000. 

  3. Counter-Terrorism and Sentencing Act 2021. 

  4. For a very radical rethink by the former UN Special Rapporteur on Human Rights and Counter-Terrorism, see Scheinin, M., ‘A Proposal for a Kantian definition of Terrorism: Leading the World Request Cosmopolitan Ethos’, European University Institute, Law 2020/15. For Sheinin the key principle of terrorism is “instrumentalisation” of other human beings. 

  5. Jake Davison killed 5 people, then himself, on 12 August 2021. 

  6. I am grateful to Professor Clive Walker QC for this observation. 

  7. Amnesty, ‘A Human Rights Guide for researching racial and religious discrimination in Counter-Terrorism in Europe’, 2021. 

  8. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Table S.05. 

  9. Ibid., Table S.04a. 

  10. Terrorism Acts in 2018 at para 4.8. 

  11. Home Office, ‘User guide to operation of police powers under the Terrorism Act 2000 and subsequent legislation’, updated 9 September 2021, at para 7. 

  12. Terrorism Acts in 2018 at para 1.36. 

  13. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Tables S.02, S.03 and S.04a. 

  14. Ibid., Table S.04a. 

  15. Ibid., Table A.11. 

  16. Ibid., Table A.11. 

  17. In relation to the monitoring of powers under the Justice and Security (Northern Ireland) Act 2007, see 

  18. JSTAT was created in 2017 but not publicly avowed until 17 March 2020. 

  19. For an explanation of these terms see Terrorism Acts in 2019 at 2.20 to 2.26. 

  20. Cf. Jusué, I.R., ‘Counter-terrorism training “at your kitchen table”: the promotion of “CT citizens” and the securitisation of everyday life in the UK’, Critical Studies on Terrorism (2021). 

  21. ACT Awareness e-Learning, https://act.campaign.gov.uk. 

  22. National Counter Terrorism Security Office, Marauding Terrorist Attacks (June 2020). 

  23. Home Office, London Bridge and Borough: Prevention of Future Death (PFD) Report, 10 January 2020. 

  24. Home Office, Protect Duty Consultation (2021). 

  25. There are 4 aspects of the government’s CT strategy: Prevent, Prepare, Protect and Pursue. 

  26. Waldron, J., ‘Terrorism and the Uses of Terrorism’, Journal of Ethics 8: 5-35 (2004). 

  27. Home Office announcement (13 October 2020), https://www.gov.uk/government/news/new-funding-awarded-to-improve-support-for-victims-of-terrorism. 

  28. HM Government, ‘Support for victims of terrorism’, https://victimsofterrorism.campaign.gov.uk. 

  29. Statement of Sir John Saunders, Chairman of the Manchester Arena Inquiry, on the application for Core Participant status by 56 survivors of the Manchester Arena attack, 21 April 2020. 

  30. Ministry of Justice, Criminal Injuries Compensation Scheme Review (2020), at para 82. 

  31. The previous threat level distinguished between the threat from International Terrorism and the threat from Northern-Ireland Related Terrorism in Great Britain. 

  32. M15, ‘Threat levels’, [https://www.mi[^5.gov.uk/threat-levels.](https://www.mi[^5.gov.uk/threat-levels) 

  33. Hansard (HC) Vol.683 Col.521 (5 November 2020). 

  34. Sir John Saunders, Report of the Public Inquiry into the Attack on Manchester Arena on 22nd May 2017, Volume 1: Security for the Arena, at para 8.12. 

  35. ‘Terrorist Risk Offenders: Independent Review of Statutory Multi-Agency Public Protection 

  36. Annual threat update, 14 July 2021. During 2021, Right Wing Terrorism has been referred to a ‘Extreme Right Wing Terrorism’. 

  37. https://www.theguardian.com/uk-news/2019/apr/10/man-jailed-four-years-plot-bomb-south-london-mosque. 

  38. https://www.bbc.co.uk/news/uk-england-tyne-55618747. 

  39. https://www.bbc.co.uk/news/uk-scotland-59069363. 

  40. https://www.bbc.co.uk/news/uk-england-coventry-warwickshire-54843050. 

  41. https://www.bbc.co.uk/news/uk-england-essex-58441466. 

  42. European Union Terrorism Situation and Trend Report, Europol (2021). 

  43. Burleigh, M., ‘Blood and Rage: a cultural history of terrorism’, Harper Collins (2009) contains a powerful reminder of the terrorism practiced by the Red Brigades in the 1970s. 

  44. Global Peace Index 2021 (June 2021). 

  45. ‘Shamima Begum must be treated as a threat, anti-terror commander Neil Basu says’, Evening Standard, 15 December 2020: https://www.standard.co.uk/news/uk/shamima-begum-threat-neil-basu-antiterrorism- b[^320306.html. 

  46. The government rejected has rejected both Lord Anderson QC and my recommendations that where deprivation is used for counter-terrorist purposes it should be subject to annual review and report. 

  47. R (on the application of Begum) v Special Immigration Appeals Commission; R (on the application of Begum) v Secretary of State for the Home Department; Begum v Secretary of State for the Home Department [2020] EWCA Civ 918. 

  48. [2021] UKSC 7. 

  49. Following the decision of the Supreme Court in Elgizouli v Secretary of State for the Home Department [2020] UKSC 10. 

  50. An update on the estimate referred to in Allen, G., Harding, M., ‘Terrorism in Great Britain: the statistics’, House of Commons research briefing CBP7613 (October 2021) at page 34. 

  51. In 2016, 35 prosecuted with 30 involved in ongoing prosecutions; in 2018, 40 prosecuted; in 2019, 40 charged: see Daniel De Simone, 29 July 2021 ‘Man admits sharing Islamic State beheading videos’, BBC: https://www.bbc.co.uk/news/uk-58012430. 

  52. https://www.bbc.com/news/uk-27488006. 

  53. https://www.bbc.com/news/uk-30213771. 

  54. Ibid. 

  55. https://www.bbc.com/news/uk-30348153. 

  56. Ibid. 

  57. https://www.bbc.com/news/uk-31166062. 

  58. https://www.bbc.com/news/uk-england-35460697. 

  59. https://www.bbc.com/news/uk-england-london-35542054. 

  60. https://www.bbc.co.uk/news/uk-42268116. 

  61. https://www.bbc.com/news/uk-50019778. 

  62. https://www.bbc.co.uk/news/uk-32142038. 

  63. Ibid. 

  64. https://www.bbc.com/news/uk-england-london-35103690. 

  65. https://www.bbc.co.uk/news/uk-58012430. 

  66. Europol, European Union Terrorism Situation and Trend report 2021, page 12 (“The UK cautioned that the decline in terrorism-related arrests and convictions can also be attributed to the operational changes necessary under government restrictions imposed in March 2020, as a result of the COVID-19 pandemic”). 

  67. UN CTED, ‘The Impact of the COVID 19 Pandemic on Terrorism, CT and CVE’ (2020). Council of EU, ‘Draft Council Conclusions on the impact of the COVID-19 pandemic on the threat posed by terrorism and violent extremism, including the impact on CT and CVE authorities and their activities’, 8633/21 12 May 2021; ‘Conclusions on the impact of the COVID-19 pandemic on internal security: threats, trends, resilience and lessons learned for EU law enforcement’, 9546/21, 7 June 2021. 

  68. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation’, June 

  69. Commission for Countering Extremism, ‘How hateful extremists are exploiting the pandemic’ (July 2020). 

  70. Some individuals may have been convicted many years earlier than the reports were completed. 

  71. Kenyon, J., Binder, J., Baker-Beall, C., ‘Exploring the role of the Internet in radicalisation and offending of convicted extremists’ (Ministry of Justice, 2021). 

  72. Terrorism Acts in 2019 at paras 7.61 et seq. 2021, annual data tables, A.05a. 

  73. In September 2020 the Law Commission launched a project on the Reform of the Communications Offences https://www.lawcom.gov.uk/project/reform-of-the-communications-offences/. However, terrorist offences committed online are outside its scope. 

  74. Tech Against Terrorism, Handbook, July 2021, page 9. 

  75. Government response to consultation on the Online Harms White Paper; Home Office, ‘Interim code of practice on terrorist content and activity online’. 

  76. Contrast the definition of ‘terrorist content’ in Article 2(7) of Regulation (EU) 2021/784 of the European Parliament and of the Council of 29 April 2021 on addressing the dissemination of terrorist content online which is defined by reference to whether the material itself incites, etc, the commission of an offence. 

  77. For an illustration of how difficult tech companies are likely to find this, see the decision of Facebook’s Oversight Board, Case decision 2021-006-IG-UA, 8 July 2021, on the removal of content discussing the treatment of the PKK’s imprisoned leader. The Board concluded that the post should not have been removed because the user did not advocate violence in their post and did not express support for the PKK or its ideology. 

  78. Section 230 of the Communications Decency Act 1996 in the US; E-Commerce Directive 2000/31/EC in the EU; the Electronic Commerce (Amendment etc) (EU Exit) Regulations 2019 (SI 2019/87) in the UK. 

  79. Tech Against Terrorism—written evidence (FEO0036), House of Lords Communications and Digital Committee inquiry into Freedom of Expression Online, 15 January 2021. 

  80. In June 2021 identifying 1,088 URLs containing terrorist content, with 711 alerts being sent to 36 tech companies with 81% of the material being removed, TCAP newsletter (June 2021), https://www.terrorismanalytics.org/blog/tcap-newsletter-june-2021. 

  81. Facebook Transparency Centre, Recent Trends, https://transparency.fb.com/data/community-standards-enforcement/dangerous-organizations/facebook/. 

  82. Patel, F., Dwyer, M.P., ‘Facebook’s New Dangerous Individuals and Organizations Policy Brings More Questions Than Answers’, Just Security, 20 July 2021. 

  83. Davies, G., Facilitating Cross-Border Criminal Justice Cooperation Between the UK and Ireland After Brexit: ‘Keeping the Lights On’ to Ensure the Safety of the Common Travel Area (2021) 85 Journal of Criminal Law 77. 

  84. Section 10(4). 

  85. A legislative consent memorandum was lodged with the Scottish Parliament on 13 February 2020. 

  86. Estimated to number around 50: Ministry of Justice press release (26 February 2020), https://www.gov.uk/government/news/automatic-early-release-of-terrorists-ends. The Release of Prisoners (Alteration of Relevant Proportion of Sentence) Order 2020/158 only applied to newly sentenced violent or sexual offenders. 

  87. Specifically, those convicted of the offences listed in Sch.19ZA to the Criminal Justice Act 2003 (in England and Wales) and Sch.1A to the Prisoners and Criminal Proceedings (Scotland) Act 1993 (in Scotland). Concern over retrospectivity led to Northern Ireland being initially excluded from the scope of the Act. 

  88. Convicted of encouraging terrorism, dissemination of terrorism publications, and stirring up religious hatred in relation to online conduct between December 2016 and March 2017. 

  89. R. (Khan) v Secretary of State for Justice [2020] EWHC 2084 (Admin); [2020] 1 W.L.R. 3932. 

  90. Para 78. 

  91. Para 122. 

  92. Given royal assent on 29 April 2021. 

  93. Including by abolishing the role of the Parole Board for dangerous terrorist offenders. 

  94. See Ministry of Justice, ‘Polygraph Examinations – Instructions for Imposing Licence Conditions for Polygraph on People Convicted of Sexual Offences (PCoSOs), Terrorist and Terrorist Connected Offences’ (August 2021). 

  95. Section 30 inserted a new article 20A into the Criminal Justice (Northern Ireland) Order 2008 (S.I. 2008/1216 (N.I. 1). Section 30 was commenced on 30 April 2021. 

  96. Letter, 22 July 2020 (OL GBR 7/2020), Note Verbale, 12 October 2020 (No.318). 

  97. Privacy International and others v Secretary of State for Foreign and Commonwealth Affairs and others [2019] UKIPTrib IPT 17 186 CH, upheld by the Court of Appeal [2021] EWCA Civ 330. 

  98. At para 63. 

  99. Sir Desmond de Silva QC, The Report of the Patrick Finucane Review, December 2012. 

  100. Concerning the alleged agent “Stakeknife”. The investigation has been independently reviewed by Alyson Kilpatrick BL, formerly my special adviser on Northern Ireland and now the Chief Human Rights Commissioner, in a published opinion dated 26 August 2021. 

  101. [2019] EWCA Crim 1527. The examination of Ali under Schedule 7 Terrorism Act 2000 (see Chapter 6) played a material part in identifying him as a manufacturer of IEDs in Afghanistan. 

  102. Criminal Procedure (Amendment) Rules 2020, SI 2020/32, Part 3, r.3.29. 

  103. Following amendment made by the Counter-Terrorism and Sentencing Act 2021, section 26(1)(a) and (2). Although no distinction is made in the elements of the offence between prominent members and foot soldiers, the Sentencing Council’s Guideline requires sentencers to make this distinction. 

  104. I describe the proscription “footprint” in Terrorism Acts in 2018 at paras 3.17 et seq. 

  105. For details of the symbolic effect see Terrorism Acts in 2019 at paras 3.11 et seq. 

  106. Dawson, J., Proscribed Terrorist Organisations, House of Commons Library (26 August 2021). 

  107. Pwr and others v Director of Public Prosecutions [2020] EWHC 798. The appeal raises the question, inter alia, of the liability of a person who waves a flag in public in genuine ignorance of its association to a proscribed organisation. 

  108. Para 50. 

  109. At para 52. Another, perhaps more compelling, mischief at which the offence is aimed is the terrorising of populations by flag-waving terrorist supporters. 

  110. [2022] UKSC 2. The Supreme Court confirmed, as was common ground, that the defendant must know that he is wearing or carrying or displaying the relevant article, and to that extent a limited mental element was contained within the offence (para 26). 

  111. Section 3(4) Terrorism Act 2000. 

  112. Proscribed Organisations (Name Change) Order 2020, SI 2020/169. 

  113. The Terrorism Act 2000 (Proscribed Organisations) (Amendment) Order 2020. One of the names (Teyrebazene Azadiye Kurdistan (TAK)) belongs to an organisation that was previously proscribed in its own right; it is now assessed that TAK is simply an alias for the PKK. 

  114. Terrorism Act 2000 (Proscribed Organisations) (Amendment) Order 2020. 

  115. https://www.bbc.co.uk/news/uk-48672929. 

  116. The Terrorism Act 2000 (Proscribed Organisations) (Amendment) (No. 2) Order 2020. 

  117. https://www.bbc.co.uk/news/uk-53392036. 

  118. https://www.cps.gov.uk/cps/news/youngest-british-terrorist-sentenced-neo-nazi-manuals-stash. 

  119. R v Dunleavy [2021] EWCA Crim 39. 

  120. Explanatory memorandum to The Terrorism Act 2000 (Proscribed Organisations) (Amendment) (No. 2) Order 2020. 

  121. Observatorio Internacional de Estudios sobre Terrorismo, June 2021 (PDF, 247 KB) 

  122. Johnson, B., Feldman, M., ‘Siege Culture After Siege: Anatomy of a Neo-Nazi Terrorist Doctrine’, ICCT Research Paper (July 2021). 

  123. Counter Extremism Project, ‘In Policy Reversal, YouTube Removes Augiobook Copies of Siege, the Turner Diaries’ (July 2019), https://www.counterextremism.com/press/policy-reversal-youtube-removes-audiobook-copies-siege-turner-diaries. 

  124. The 5 factors are the nature and scale of an organisation’s activities, the specific threat that it poses to the UK, the specific threat that it poses to British nationals overseas, the extent of the organisation’s presence in the UK, the need to support other members of the international community in the global fight against terrorism: Home Office, ‘Proscribed terrorist groups or organisations’ (updated 16 July 2021). 

  125. For example, the Order of the Nine Angles. 

  126. Section 3(5) Terrorism Act 2000. 

  127. Section 121 Terrorism Act 2000. Professor Clive Walker QC notes that the definition is wide enough to encompass an affinity group, an “anarchistic disorganisation”, diffuse networks such as Al-Qa’ida and self- generating combinations. 

  128. Which is frequently used to justify attacks on Muslims. Davey, J., Ebner, J., ‘The Great Replacement: The Violent Consequences of Mainstreamed Extremism’, Institute for Strategic Dialogue (2019). 

  129. Or responsible agency: Duff, R.A., ‘Answering for Crime: Responsibility and Liability in the Criminal Law’ (Hart 2007) 165. 

  130. Section 58 Terrorism Act 2000. 

  131. R v G [2009] UKHL 13. 

  132. Section 2(3). 

  133. For example, R v Aristidou, reported at https://www.bbc.co.uk/news/uk-58012430. 

  134. At least for a terrorism offence. See Terrorism Acts in 2019 at 7.61 et seq. for why creating criminal liability for possession would extend the boundaries of terrorism legislation too far. 

  135. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, June 2021, annual data tables, Table A.05a. 

  136. I refer to disseminator, although section 2 captures a wider range of conduct including possession with a view to dissemination: section 2(2)(f). 

  137. Section 2(1) Terrorism Act 2006. 

  138. Section 2(5). 

  139. AE (Iraq) v Secretary of State for the Home Department [2021] EWCA Civ 948 at para 30. 

  140. See Erkizia v Spain, App.No.5869/17, 22 June 2021 at 39: free speech protections do not apply to words directly or indirectly advocating the use of violence or “hate speech”. 

  141. Clauses 41, 42 and Schedule 2. 

  142. Clause 41(3)(c). 

  143. The test under clause 41(3) is whether the service provider has reasonable grounds to believe that the dissemination of the content “constitutes a relevant offence”, which does not exclude the role of the criminal standard of proof in determining whether reasonable grounds exist. Since the test is not “might constitute” the test appears higher than the test of no case to answer (R v Galbraith 73 Cr.App.R. 124). 

  144. Section 20(1): assuming that publication is a cognate (see section 20(1)) of “publish”. 

  145. The encouragement must be to members of the public (plural): ss 1(1), 1(2)(b). According to assurances given to Parliament, the encouragement must be directed at members of the public rather than “persons”, and is not intended to capture private conversation: Hansard HL vol 676 col 435 (5 December 2005), Baroness Scotland. 

  146. I leave aside the difficult public/private divide in the context of online surveillance under the Regulation of Investigatory Powers Act 2000. As noted by Ward, R. and Jones, R., National Security (OUP, 2021) para.4.19, there may be a reasonable expectation of privacy even in public places, and private information may be gathered in public. The topic is dealt with in the Home Office’s Covert Surveillance and Property Interference Revised Code of Practice (2018). 

  147. Section 20(3)(b). 

  148. Section 3(7). 

  149. Section 3(4). 

  150. Although not unprecedented in the annals of terrorism: Burleigh, M., op.cit. 

  151. R v Aristidou, supra. 

  152. Section 3(6) Immigration Act 1971. 

  153. Section 3(3) and section 123(4) Terrorism Act 2000, or in the case of urgency section 123(5). 

  154. Arumagam and others v Secretary of State for the Home Department PC/04/2019, 21 October 2020. 

  155. Terrorism Acts in 2019 at para 3.21. 

  156. Ibid. 

  157. Arumugam v Secretary of State for the Home Department PC/04/2019, 18 February 2021, at para 21. 

  158. Home Office, ‘Proscribed terrorist groups or organisations’ (updated 16 July 2021) was amended following the appeal. However, it still erroneously refers to the Commission having a power to deproscribe (“following a successful appeal [the Commission] makes an order for the organisation to be deproscribed…”). 

  159. Section 5(4). 5(5) also allows an order to be made without laying before Parliament in a case of urgency. It is difficult to see how urgency would arise in this context. Although the Act does not provide a power for the Commission to set aside a flawed decision, this was inherent in the scheme: para 14. 

  160. Arumugam, supra, 18 February 2021. 

  161. The history of the proscription of the The People’s Muhajideen of Iran is set out in Smith, B., House of Commons Briefing Paper Number CBP 5020, 7 March 2016. 

  162. For example, by seeking a remedy from the High Court to vindicate rights protected by the Human Rights Act 1998 (in the context of defective secondary legislation, as in RR v Secretary of State for Work and Pensions [2019] UKSC 52). 

  163. Arumugam, supra, 21 October 2020 at para 29. 

  164. Ibid. 

  165. The Biden Administration took the decision to de-list Ansar Allah (the Houthis) as a terrorist group because of the control exercised by the group over large parts of Yemen and the potential humanitarian consequences of keeping the designation in place: Testimony of Adam M. Smith before the United States Senate Committee on Banking, Housing, and Urban Affairs, “Afghanistan’s Future: Assessing the National Security, Humanitarian, and Economic Implications of the Taliban Takeover” (October 5, 2021). 

  166. ‘Proscribed terrorist groups or organisations’, Home Office (updated 16 July 2021). 

  167. US Treasury, Press Release, 24 September 2021 accompanying issue by OFAC of GL. 14 and 15. Unlike the US, which goes beyond UNSCR obligations in this respect, the UK does not apply financial sanctions against the Taleban as a whole. 

  168. See page 4. Terrorist financing is referred to 6 times in this document. 

  169. Home Office, OFSI, “For information note: operating within counter-terrorism legislation, counter- terrorism sanctions and export control” (updated 11 October 2021). 

  170. Testimony of Adam M. Smith before the United States Senate Committee on Banking, Housing, and Urban Affairs, supra. 

  171. Terrorism Acts in 2019 at 3.26. 

  172. Terrorism Acts in 2019 at 3.34. 

  173. There is wider scope to provide de facto exemptions under proscription legislation than under sanctions regimes which implement UN Security Council Resolutions. UNSCR regimes, such as UNSR 2255, may not provide any express provision for humanitarian licences. 

  174. ‘For information note: operating within counter-terrorism legislation” (Updated 11 October 2021), to which the CPS also contributed. 

  175. Financial Action Task Force (FATF), Mitigating the Unintended Consequences of FATF Standards, http://www.fatf-gafi.org/publications/financialinclusionandnpoissues/documents/unintended-consequences-project.html. 

  176. Global Counter-Terrorism Forum, ‘Safeguarding Civic Space and CFT Measures’ (May 2021), https://www.thegctf.org/What-we-do/Initiative-Activities/ArtMID/815/ArticleID/160/Safeguarding-Civic-Space-and-CFT-Measures. 

  177. Terrorism Acts in 2018 at 3.66. 

  178. Terrorism Acts in 2018 at 3.70. 

  179. See Terrorism Acts in 2018 at 4.19. 

  180. Murphy, C., ‘The Crypto-Wars myth: The reality of state access to encrypted communications’, Common Law World Review 2020, Vol 49(3-4) 245-261 rightly points to two practical limitations associated with a power to compel disclosure of encryption keys which any user of the power would need to consider; the suggested legal limitation (that it may engage an individual’s right not to incriminate themselves) is less convincing (see R v S and another [2008] EWCA Crim 2177). 

  181. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Table S.01. 

  182. Ibid, Table S.02. 

  183. Calendar year data provided by PSNI. 

  184. See Chapter 9 for further details. 

  185. At 4.15 et seq. 

  186. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Table S.05. 

  187. Northern Ireland Office, ‘Northern Ireland Terrorism Legislation: annual statistics 2020/21’, Table 10.1. As with other Northern Ireland Office statistics, which are compiled on a financial year basis, I have obtained a calendar year figure by taking account of the final quarter of the previous year’s figures. 

  188. Terrorism Acts in 2018 at 4.27. 

  189. Northern Ireland Office, ‘Northern Ireland Terrorism Legislation: annual statistics 2020/21’, Table 2.2. 

  190. Terrorism Acts in 2019 at 4.36 et seq. 

  191. It was used in 2014, 2016 and 2019, as detailed in Terrorism Acts in 2019 at 4.57. It was also used in 2021: I will report on this in next year’s annual report. 

  192. Source, National CT Policing Headquarters. 

  193. National Risk Assessment of Money-Laundering and Terrorist Financing 2020, HM Treasury and Home Office (December 2020). 

  194. UK Financial Intelligence Unit, Suspicious Activity Reports Annual Report, National Crime Agency (2020). 

  195. Ibid, page 19. 

  196. These are the figures given in last year’s report, Terrorism Acts in 2019. However, caution is required because I have subsequently been provided with the following figures for 2019: 1,141; 240; 308; 51. 

  197. For the differences between ordinary arrest powers and section 41 see Terrorism Acts in 2018 at 5.1 to 5.4. 

  198. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Table A.01. 

  199. Northern Ireland Office, ‘Northern Ireland Terrorism Legislation: annual statistics 2020/21 and 2019/20’, Table 3.1. 

  200. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Table A.02. 

  201. Source: CTPHQ. 

  202. Para 31 Schedule 8. 

  203. Northern Ireland Office, ‘Northern Ireland Terrorism Legislation: annual statistics 2020/21’, Table 4.1. 

  204. Ibid., Table 3.1. 

  205. Ibid., Table 4.2. 

  206. Section 38 PACE. 

  207. Concordat on Children in Custody, Home Office, at page 14. 

  208. In the matter of T (A Child) [2021] UKSC 35 at para 166, Lord Stephens. 

  209. Concordat, supra, at page 12. 

  210. Section 38(6) PACE. 

  211. Condordat, supra, at page 12. According to this guidance it does not relate to the availability of Local Authority accommodation or transport; the nature of the accommodation offered by the Local Authority; the child’s behaviour or the nature of the offence; and does not mean ‘difficult’ or ‘inconvenient’. 

  212. Code C para 12:2. 

  213. Max Hill QC gave details of the 23 persons arrested and released without charge in Operation Manteline in Terrorism Acts in 2017 at 4.22 et seq. 

  214. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Table A.03. 

  215. Ibid., Table A.02. 

  216. Ibid., Table. A.03. 

  217. Ibid., Table A.05a. 

  218. Police Service of Northern Ireland, Policing Recorded Security Situation Statistics Northern Ireland, table 7. 

  219. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Table A.09. 

  220. Ibid, Table A.10. 

  221. ISC, Northern Ireland-related Terrorism, HC 844 (5 October 2020) at paras 12 to 20. 

  222. Ibid., Table A.11. 

  223. Ibid., Table A.12a. 

  224. Lewis, J. and Marsden, S., Public Experiences of the UK Counter-Terrorism System (CREST, 2020). 

  225. Ibid. Sometimes referred to as ‘screening questions’. 

  226. Terrorism Acts in 2019 at 6.33. 

  227. Nationality and Immigration Bill 2021. 

  228. R (on the application of C1) v Secretary of State for the Home Department [2021] EWHC 242 (Admin). 

  229. The conditions at Tug Haven, the main processing centre for Dover, were found to be inadequate and unsuitable: Annual Report of the Independent Monitoring Board at Dover Short-Term Holding Facility (2019-2020) (published September 2021). 

  230. Strengthened by Clause 37 of the Bill to capture those who arrive without valid entry clearance. 

  231. The fact that answers are automatically excluded from criminal proceedings by paragraph 5A of Schedule 7 is not sufficient protection against violation of the right to a fair trial under Article 6. This is because Article 6 can be violated without any prosecution taking place. The Strasbourg authorities to this effect were reviewed by Lord Reed in the Privy Council decision Volaw Law Trust and Corporate Services Ltd and others v Her Majesty’s Attorney General for Jersey [2019] UKPC 29, see paras 47-48, 50-51, 72. 

  232. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation’, Great Britain 2010/11 (13 October 2011, HOSB 15/11). 

  233. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Table S.04. 

  234. For reasons I do not understand, the total number of examinations in Table S.04 (3,434) is greater than the total in Table S.03 (3,315, which I understand to be the correct figure for 2020). 

  235. Zakaria Yanaouri, later prosecuted for possession of information likely to be useful to a terrorist, came to the attention of the authorities on 11 January 2020 when he was examined under Schedule 7: https://www.cps.gov.uk/crime-info/terrorism/counter-terrorism-division-crown-prosecution-service-cps-successful-prosecutions-2016. 

  236. Terrorism Acts in 2018 at 6.83 et seq. 

  237. See Terrorism Acts in 2019 at 6.17 to 6.19. 

  238. A new Authority to Carry Scheme was published in January 2021 by the Home Office under the Counter- Terrorism and Security Act 2016, although the scheme does not itself contain a requirement to provide information. Most progress is made through government engagement with individual carriers (although Brexit is a complicating factor). The government has announced an intention to create a UK equivalent of the US ESTA travel scheme for air travel. 

  239. Terrorism Acts in 2019 at 6.20. 

  240. Recent guidance on how the CTA rules apply in travel between the UK, Ireland, the Isle of Man, Guernsey and Jersey was published in August 2021: https://www.gov.uk/guidance/travelling-between-the-uk-and-ireland-isle-of-man-guernsey-or-jersey. 

  241. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Table S.03. 

  242. Comparing the figures of 1,191 and 1,171 in the table below. 

  243. By Article 4.2 deprivation of liberty means any form of detention or imprisonment or the placement of a person in a public or private custodial setting which that person is not permitted to leave at will by order of any judicial, administrative, or other authority. 

  244. Under paragraph 1 of Schedule 8 as a place of detention for the purposes of Sched 7 through the relevant Home Office Circular. 

  245. Lewis, J. and Marsden, S., op cit, at page 10. 

  246. Training is biennial on a pass/fail basis. Officers who fail are offered another chance to pass the test. 

  247. Terrorism Acts in 2019 at 6.50. 

  248. Terrorism Acts in 2018 at para 10.14. 

  249. Para 24. 

  250. Terrorism Acts in 2018 at para 10.16. 

  251. Described at paras 6.59 to 6.65. 

  252. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Table S.03. 

  253. Ibid. 

  254. I referred to the need to consider how to ensure that ethnicity data remains available in Chapter 1. 

  255. Y. Musharbash, ‘The Globalization of Far-Right Extremism: An Investigative Report’ (2021) 16 CTC Sentinel 6. 

  256. In 2020 Paul Golding, the leader of the far-right ‘Britain First’, was convicted under Schedule 7 of failing to hand over his devices for examination: https://www.cps.gov.uk/crime-info/terrorism/counter-terrorism-division-crown-prosecution-service-cps-successful-prosecutions-2016. 

  257. Para 25. 

  258. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Table S.02. 

  259. Para 6A Schedule 7 Terrorism Act 2000. 

  260. In particular, the Demand Risk and Resource review referred to in Terrorism Acts in 2018 at 27. 

  261. Law Commission, ‘Search Warrants’, Law Com No 396, HC 852 (2020), at para 16.121. 

  262. “…other than where the obtaining of stored communications is authorised by section 6(1)(c)(ii) of that Act.” The Schedule 7 power is a statutory power falling within section 6(1)(c)(ii). 

  263. It was common ground in R (on the application of KBR, Inc) (Appellant) v Director of the Serious Fraud Office (Respondent) [2021] UKSC 2 that the statutory power to compel provision of information could be used against a UK company in respect of documents overseas (para 26). This suggests, although the Supreme Court itself expressed no view on the concession, that the presumption against extraterritorial effect applies weakly or not at all in these circumstances. 

  264. Under paragraph 5 of Schedule 14 Terrorism Act 2000, examining officers are to perform their functions in accordance with the Code of Practice. 

  265. In human rights terms, where they authorise interference with fundamental rights (here, privacy under Article 8), laws must be sufficiently precise and clear: Gillan v United Kingdom (2010) 50 EHRR 1105, paras 76-77. 

  266. No doubt originally intended to enable access to information held on local office servers, the Law Commission concluded that it probably did extend to remote data held overseas: op.cit. at para 18.51. 

  267. Op.cit., at paras 16.1 and 16.2. 

  268. Op.cit., at para 16.195. 

  269. The issue of data retention also arises in connection with the data in Ports Intelligence Reports which are created by officers after examinations under Schedule 7. 

  270. Terrorism Acts in 2018 at 6.104; Terrorism Acts in 2019 at 6.41 to 6.43. 

  271. Other purposes under paragraph 11A(3) are retention for use as evidence in criminal proceedings or for use in connection with a deportation decision. 

  272. Para 66. 

  273. Beghal v DPP [2015] UKSC 49 at para 57. 

  274. Lord Hughes, ibid, at para 22, citing the analysis of former IRTL, Lord Anderson QC. 

  275. For example, the sleeper cell led by Dhiren Barot, who was convicted in 2006. 

  276. For example, Usman Khan who carried out the attack at Fishmongers’ Hall in 2019. 

  277. Home Office, ‘Extraction of Information from Electronic Devices Draft Code of Practice’ (July 2021) at para 48. 

  278. As presently occurs for biometrics under the Biometrics and Surveillance Camera Commissioner, a role which I consider further below. 

  279. Rattvisa v Sweden, App.No.35252/08, Grand Chamber at para 236. 

  280. Part 3, section 39. 

  281. Para 66. 

  282. Available at https://www.app.college.police.uk/app-content/information-management/management-of- police-information/retention-review-and-disposal-of-police-information/. 

  283. Referred to in Terrorism Acts in 2019 at 6.41. 

  284. Information Commissioner, ‘Mobile Phone Data Extraction by police forces in England and Wales, Investigation Report’ (July 2020) at para 3.8. Mobile phone extraction was considered in R v Bater-James and another [2020] EWCA Crim 790. 

  285. Under Article 552.4, “…the United Kingdom shall delete the PNR data of passengers after their departure from the country unless a risk assessment indicates the need to retain such PNR data. In order to establish that need, the United Kingdom shall identify objective evidence from which it may be inferred that certain passengers present the existence of a risk in terms of the fight against terrorism and serious crime.” 

  286. Described in Terrorism Acts in 2018 at 6.121 to 6.123. 

  287. Biometrics Commissioner, ‘Biometrics Commissioner statement on the Coronavirus Act and the Protection of Freedoms Act’ (7 September 2020). 

  288. Following amendments made by the Counter-Terrorism and Border Security Act 2019. 

  289. Office of the Biometrics Commissioner, Interim Report (December 2020). The annual report for 2020 was published in November 2021. 

  290. “Data: a new direction”: public consultation, Response by the Independent Reviewer of Terrorism Legislation(PDF, 107 KB) 

  291. Under paragraph 9. 

  292. Paragraph 9(4). 

  293. Terrorism Acts in 2019 at 6.52 et seq. 

  294. Schedule 3 para 62. 

  295. At paras 130-1, 162. 

  296. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Tables A.03 and A.05b. 

  297. Sections 12(1A), 58(1)(c) and 58B Terrorism Act 2000 all added by the Counter-Terrorism and Border Security Act 2019. Section 58(1)(c) resulted in conviction in R v Gregory, Winchester Crown Court (2021), https://www.bournemouthecho.co.uk/news/19284192.bournemouth-man-robert-gregory-jailed-terrorism-offences/. 

  298. Terrorism Acts in 2019 at 7.9. 

  299. Under section 58C Terrorism Act 2000. 

  300. Section 58B(2) to 6). 

  301. In which case section 5 Terrorism Act 2006 would be available for some. It is more difficult, as I reported in Terrorism Acts in 2018 at 7.28, where a person, typically female, travels to provide only moral or household support to a terrorist fighter. 

  302. Section 58B(1) and (3). 

  303. [https://www.standard.co.uk/news/uk/shamima-begum-threat-neil-basu-antiterrorism-b[^320306.html.](https://www.standard.co.uk/news/uk/shamima-begum-threat-neil-basu-antiterrorism-b[^320306.html) 

  304. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Table C.01. 

  305. Chowdhury had been acquitted in 2018 of terrorist changes after attacking policemen outside Buckingham Palace with a sword. 

  306. Her suspended sentence was upheld as not being unduly lenient, owing to her family’s controlling behaviour: Att.-Gen.’s Reference (R. v Sneha Chowdhury) [2020] EWCA Crim 1421. 

  307. For a sense of scale, in ‘The EU’s Work on Battlefield Information: Stocktaking and possible next steps’ (Council doc. 9481/21, 2021), it is reported that the US FBI has provided EU Member States and Europol with data about 2 700 possible FTFs held in custody in North-East Syria after the Turkish incursions in the latter part of 2019; UNITAD (The United Nations Investigative Team for Accountability of Daesh/ISIL) has been able to extract facial profiles from image data across its archives, resulting in a dedicated repository of over 175,000 of such profiles. 

  308. https://www.bbc.co.uk/news/uk-england-tyne-55618747. 

  309. https://www.cps.gov.uk/crime-info/terrorism/counter-terrorism-division-crown-prosecution-service- cps-successful-prosecutions-2016. 

  310. https://www.bbc.co.uk/news/uk-england-beds-bucks-herts-53528343. 

  311. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Table C.04. 

  312. https://www.gloucestershire.police.uk/news/gloucestershire/news/gloucestershire-boy-sentenced-for-terrorism-offences/. 

  313. Terrorism Acts in 2019 at 8.99. SCPOs are described in Terrorism Acts in 2018 at 7.49. 

  314. https://www.cps.gov.uk/crime-info/terrorism/counter-terrorism-division-crown-prosecution-service-cps-successful-prosecutions-2016. 

  315. Counter-Terrorism Act 2008 Part 4 created a special registration scheme for certain terrorist offenders. 

  316. For example, attack-planning, terrorist training, or serious offences such as possession of explosives which are found to be connected to terrorism. 

  317. A role that is set to be increased under the Police, Crime, Sentencing and Courts Bill with respect to dangerous offenders. 

  318. R v Martin [1999] 1 Cr. App. R. (S.) 477, per Lord Bingham LCJ. 

  319. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Table P.01. 

  320. Ibid., Table P.04. 

  321. Ibid., Table P.05. 

  322. In R (Latif) v SS for Justice [2021] EWHC 892 (Admin), the High Court upheld emergency changes to the Claimant’s licence conditions based on a fear of copycat attacks after Fishmongers’ Hall. 

  323. Section 102(5) Power of the Criminal Courts (Sentencing) Act 2000. The decision was upheld by the High Court in The Queen (on the application of X) v Ealing Youth Court [2020] EWHC 800 (Admin). 

  324. R v G; R v J [2009] UKHL 13, at para 43. 

  325. By reference to date of sentence: R v Chowdhury (2021), R v Yanaouri (2020), R v Vaughan (2020), R v Dunleavy (2020), A Youth (2021), R v Nimmo (2020), R v Ashfaq (2019), R v Ghani (2019), R v Fletcher (2019), R v Bishop (2019), R v Adam (2019), R v Aweys (2019), R v Hussain (2019), R v Amman (2018), R v Morrice (2021), R v John (2021), R v Brock (2021), R v Nugent (2021), R v Alcharbati (2018). 

  326. By the Counter-Terrorism and Border Security Act 2019. 

  327. Sentencing Council, Terrorism offences: Data tables (21 October 2019). 

  328. Sentencing Council, Guideline: Collection of terrorist information (April 2018). The Council is currently consulting on revised guidelines in light of the increased maximum penalties in the Counter-Terrorism and Sentencing Act 2021. 

  329. Monitoring under Part 4 Counter-Terrorism Act 2008, and enhanced licence conditions. 

  330. R v G; R v J [2009] UKHL 13, at para 46. 

  331. Strictly speaking, section 57 does not require proof of a terrorist purpose. However, it is a defence that the article was not held with such a purpose. This is an important distinction between the seriousness of section 57 and section 58 offences: see for example, Secretary of State for the Home Department v NF [2021] EWCA Civ 17 at para 36. 

  332. R v G, supra, at 49. Cf. Sentencing Council guidelines for section 58 (effective 27 April 2018), part of which expressly apply to an offender with no terrorist connections or motivations. 

  333. Section 58(3). This defence has special application where a person is merely charged with viewing: section 58(3A). 

  334. So it would cover a person who picked up a document by accident, looked at it, and was going to hand it to the police; but not a person who had information that was useful to a terrorist in order to assist him in committing a bank robbery; nor for the purpose of winding up prison officers: R v G, supra, at para 79, 87. 

  335. Such as the defendant’s age, his background, his associates, his way of life, the precise circumstances in which he collected or recorded the information, and the length of time for which he possessed it: R v G, supra, at para 81. 

  336. By the Criminal Justice and Public Order Act 1994 

  337. At paras 14.5 and 14.8. A classic example of offending in the Northern Ireland context would be a person who collects number-plates of police officers as targeting information. 

  338. R v G, supra, at 49. 

  339. Walker, C., Blackstone’s Guide to Terrorism Legislation (3rd ed, 2014) at para 6.03. 

  340. Sections 1(2)(b)(ii) and 2(1)(c) Terrorism Act 2006, section 12(1A)(b) Terrorism Act 2000. 

  341. Although it could be said that looking at a graphic torture manual could have indirectly harmful consequences, by harming the mental health of the viewer, or, potentially, sustaining a market for the production of such materials (and therefore the carrying out of atrocities to provide further illustrations). 

  342. For example, Zedner, L., “Countering terrorism or criminalizing curiosity? The troubled history of UK responses to right-wing and other extremism” Common Law World Review 2021, Vol 50(1) 57-75, referring to the principle of responsible agency. 

  343. de Coensel, S., ‘Self-study, obtaining or viewing terrorist material over the internet’ (2020) 28 European Journal of Crime, Criminal Law and Criminal Justice 379. 

  344. Terrorism Acts in 2019 at 7.61. 

  345. Terrorism Acts in 2019 at 7.61 et seq. 

  346. Identified as the sole or, where multiple offences are charged, the most serious. In many terrorism cases, section 58 will not be the principal offence. 

  347. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Table A.05a. 

  348. Statista (1 September 2021), https://www.statista.com/topics/3246/internet-usage-in-the-uk/. 

  349. For example, ‘Panini card’ reasons: a desire to collect for its own sake. 

  350. Terrorism Acts in 2019 at 2.15. 

  351. For example, Thomas Mair who murdered Jo Cox MP in 2016, Darren Osborne who carried out the 2017 attack at Finsbury Park mosque, and Vincent Fuller who stabbed a teenager in a Surrey car park in 2019 (the day after the Christchurch attacks in New Zealand). 

  352. Jonathan Hall QC, ‘What is the threat to the UK today’, speech to Bright Blue, 7 June 2021. 

  353. In sentencing the Right Wing terrorist and former police officer Benjamin Hannan the judge observed that the defendant’s autism “…explains why you kept material which others may have discarded and why you were meticulous about holding it in appropriate folders and sub-folders on your computer….”: R -v- Hannam Sentencing Remarks (4 May 2021). See also, Little, R., Peter Ford, P., Girardi, A., Online self-radicalisation: a case study of cognitive vulnerabilities for radicalization to extremism and single actor terrorism, Journal of Intellectual Disabilities and Offending Behaviour (2021); Walter, F., Leonard, S., Miah, S., Shaw, J., Characteristics of autism spectrum disorder and susceptibility to radicalisation among young people: a qualitative study, The Journal of Forensic Psychiatry and Psychology (2020). 

  354. Cf the creation or use of 3D-printed guns as in the 2019 attack in Halle, Germany (https://www.independent.co.uk/news/world/europe/3d-gun-print-germany-synagogue-shooting-stephan-right-wing-terrorism-deny-possessing-parts-3D-printed-gun.html). 

  355. R v Dunleavy [2021] EWCA Crim 39 at para 48. 

  356. As demonstrated, for example, by the commission of other terrorist offences as in the case of Mohammed Chowdhury who was arrested in 2020 for possession of terrorist manuals and attempting to possess a hand grenade with intent: https://www.cps.gov.uk/crime-info/terrorism/counter-terrorism-division-crown- prosecution-service-cps-successful-prosecutions-2016. 

  357. Terrorism Offences Guideline, Consultation (October 2019). 

  358. R v Gul [2013] UKSC 64 at para 36. 

  359. Under section 41 Counter-Terrorism and Sentencing Act 2021. 

  360. HM Government, Transparency Report: Disruptive Powers 2019/20. 

  361. Allen, G., Harding, M., ‘Terrorism in Great Britain: the statistics’, House of Commons research briefing CBP7613 (October 2021) at para 8.2. 

  362. Secretary of State for the Home Department v JM and LF [2021] EWHC 266 (Admin) at para 57. 

  363. Paragraph 1 Schedule 7 TPIM Act 2011. 

  364. Quarterly statement, 28 April 2020, HCWS203. 

  365. Quarterly statement, 16 July 2020, HCWS374. 

  366. https://www.telegraph.co.uk/news/2021/02/22/terror-suspect-jailed-breaching-tpim-removing-electronic-tag/. The charges from May 2020 were left on the file and not proceeded with. 

  367. Quarterly reports to 29 February 2020: Hansard (HC) Vol.675 Col.23WS (28 April 2020); to 31 May 2020: Hansard (HC) Vol.678 Col.76WS (16 July 2020); to 31 August 2020: Hansard (HC) Vol.681 HCWS497 (8 October 2020); to 30 Nov 2020: Hansard (HC) Vol.687 HCWS698 (12 January 2021). 

  368. Hansard (HL) Vol.669 Col.1102 (22 February 2005), Baroness Scotland (Minister of State, Home Office). 

  369. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Tables A.12a – A.12c. 

  370. Transparency Report 2018: Disruptive and Investigatory Powers, July 2018, Cm.9609 at section 5.9, page 27. 

  371. Secretary of State for the Home Department [2021] EWHC 266 (Admin), at para 42. 

  372. Ibid. 

  373. Upheld by the High Court in QT v Secretary of State for the Home Department [2019] EWHC 2583 (Admin). 

  374. TPIM Act 2011 section 3(6)(b), (c). 

  375. JM and LF, supra, at para 62. 

  376. Ibid., at para 169. 

  377. Ibid. at 159. 

  378. TPIM Quarterly reports, supra. 

  379. Post-legislative Scrutiny Memorandum to the Home Affairs Committee, CP455 (June 2021). 

  380. One consequence of the pandemic was that some meetings could not take place: the review process was limited to an exchange of draft minutes over email. 

  381. Terrorism Acts in 2019 at 8.53 to 8.58. 

  382. Terrorism Acts in 2019 at 8.56. 

  383. Following amendments made by the Counter-Terrorism and Sentencing Act 2021. 

  384. Terrorism Acts in 2019 at 8.62 to 8.70. 

  385. Correspondence with UN Special Rapporteur on the promotion and protection of human rights and fundamental freedoms while countering terrorism on the Counter-Terrorism and Border Security Bill, Letter, 22 July 2020 (OL GBR 7/2020), Note Verbale, 12 October 2020 (No.318). 

  386. Indeed, the original position was that legal aid would be available on a non-means tested basis: James Brokenshire MP (Parliamentary Under-Secretary of State for the Home Department), Fourth Delegated Legislation Committee (1 February 2012). 

  387. The Legal Aid, Sentencing and Punishment of Offenders Act 2012 (Amendment of Schedule 1) Order 2013/748 which inserted paragraph 18 of Part 2 of Schedule 1 of the 2012 Act containing an extended definition of excluded “judicial review” proceedings. The analysis presented to me was that since section 9 review hearings require the judge to apply the judicial review test, review hearings were therefore excluded from the scope of the TPIM entitlement to legal aid but are eligible for legal aid on the basis that they are in effect justice review proceedings under paragraph 19. On my contrary analysis, the purpose of the amendments was purely technical, designed to put beyond doubt that legal aid for judicial review proceedings was only available as set out in paragraph 19 of Part 1: see Explanatory Memorandum, at para 7.9. The prima facie right to legal aid under paragraph 45 therefore remains for review hearings. 

  388. Section 4 Legal Aid Sentencing and Punishment of Offenders Act 2012. 

  389. Civil Legal Aid (Merits Criteria) Regulations 2013. 

  390. Section 11(6) Legal Aid Sentencing and Punishment of Offenders Act 2012. 

  391. The imposition of the balance of probabilities test was said by MI5 to have made the TPIM process more onerous and requires increased disclosure of sensitive material: Memorandum to the Home Affairs Select Committee, CP455, Home Office (June 2021) at para 59. 

  392. Under section 10ZA TPIM Act 2011 as inserted by section 38 of the Counter-Terrorism and Sentencing Act 2021. 

  393. By section 10ZA(4). 

  394. Letter, Lord Parkinson to Baroness Hamwee, 26 February 2021 (PDF, 108 KB) 

  395. Lord Parkinson, Hansard (HL) Vol.810 Col.1219 (3 March 2021). 

  396. Under section 12(1A) TPIM Act 2011. 

  397. The government rejected my recommendation to make TEOs available for non-British nationals on the basis that they did not foresee any net operational benefit. 

  398. The regime is described in Terrorism Acts in 2018 at 8.33 to 8.43. 

  399. If a TEO is made but not served, the practice is to review its suitability after 6 months. To avoid the clock being run down on the two year maximum, the Home Office may revoke a TEO and remake it at a later stage. 

  400. A reform also proposed by MI5, Post-legislative Scrutiny Memorandum to the Home Affairs Committee, CP455 (June 2021) at para 39. 

  401. Ibid at para 38. 

  402. A total of three orders was made in respect of one male: two were revoked and one was served. 

  403. Section 13(6) TPIM Act 2011. 

  404. Explanatory Notes at paras 6, 17. 

  405. https://www.bbc.co.uk/news/uk-england-leicestershire-52673463. 

  406. XH and (2) AI v Secretary of State for the Home Department [2017] EWCA Civ 41. 

  407. Paras 89 to 93. 

  408. At section 53. 

  409. Paragraph 9 Schedule 1 provides for ‘work or studies’ measures. 

  410. Transparency Report, supra, at para 5.6. 

  411. Terrorism Acts in 2018 at 8.73. 

  412. Post-legislative Scrutiny Memorandum to the Home Affairs Committee, CP455 (June 2021). 

  413. Section 43 and Schedule 12, enabling applications under the Serious Crime Act 2007 to be made by chief officers of police. 

  414. I am particularly grateful to Karl Laird, one of my special advisers, for his work on this Chapter. 

  415. PSNI, South Armagh Policing Review Findings and Recommendations (17 December 2020). 

  416. Committee on the Administration of Justice, ‘Covert policing and the accountability gap: Five years on from the transfer of ‘national security’ primacy to MI5’ (November 2012). 

  417. Whilst recognising the differences between, say, Islamist terrorism in Great Britain and Northern Ireland-Related Terrorism whose pull is, according to Ferguson, N. and McAuley, J.W., Radicalization or Reaction: Understanding Engagement in Violent Extremism in Northern Ireland (2020) 41 Political Psychology 215, based more on collective identity, community, peer and family relationships and pressures than ideology. 

  418. Hansard (HC) Vol.699 Col.36WS (19 July 2021). 

  419. As I reported last year, the threat posed specifically by Northern Ireland-related terrorism to Great Britain, as opposed to other forms of terrorism, is no longer published separately. 

  420. Europol, European Union Terrorism Situation and Trend report 2021, pages 12, 23. 

  421. As deemed by PSNI. 

  422. PSNI, Security Situation Statistics, information up to and including March 2020, Table 3. 

  423. Ibid., Table 5. 

  424. Ibid., Table 4. 

  425. Ibid, Table 4. 

  426. Ibid, Table 6. 

  427. Europol, TE-SAT Report 2021, page 23. 

  428. David Seymour CB, Thirteenth Report, supra, at para 4.2. 

  429. Allegations include directing terrorism and seeking to obtain Semtex explosive. 

  430. Terrorism Acts in 2018 at 3.56. 

  431. Schedule 2 para 17. 

  432. At para 14. 

  433. Base 2, ‘Working to support individuals under threat, An inspection of the role of Base 2 in threat verification’ (March 2020), at page 20. 

  434. For example, the use of UVF flags in Cantrell Close, Belfast, in 2019. 

  435. PSNI, Security Situation Statistics, information up to and including March 2021, Table 7. 

  436. In the matter of an application by Stephen Ramsey for judicial review (No 2) [2020] NICA 14 at para 54. 

  437. At para 58. 

  438. Policing Board of Northern Ireland, ‘Human Rights Thematic Review on the use of police powers to stop and search and stop and question under the Terrorism Act 2000 and the Justice and Security (Northern Ireland) Act 2007’ (2013). 

  439. Paras 5.2 to 5.9. 

  440. Ibid., para 5.8. 

  441. Interface refers to common boundary lines between blocks of nationalist and unionist households. 

  442. At para 31. 

  443. Northern Ireland Office, ‘Northern Ireland Terrorism Legislation: annual statistics 2020/21’, Table 10.1. The figures for 2017/18, 2018/19, 2019/20 were revised. 

  444. PSNI, ‘Policing Recorded Security Situation Statistics Northern Ireland’, Table 5. PSNI statistics form the basis of Northern Ireland Office’s official statistics. 

  445. Summary of main findings of 2020 Report, supra. 

  446. PSNI, User guide to Security Situation Statistics Northern Ireland (September 2020), page 4. 

  447. Home Office, ‘Operation of police powers under the Terrorism Act 2000 and subsequent legislation, year ending December 2020’, Table A.01. 

  448. Ibid. 

  449. Independent, 17 February 2021, [https://www.independent.co.uk/news/uk/home-news/three-men-arrested-loyalist-gathering-belfast-pitt-park-b[^1803443.html?r=34485.](https://www.independent.co.uk/news/uk/home-news/three-men-arrested-loyalist-gathering-belfast-pitt-park-b[^1803443.html?r=34485) 

  450. See for example, Let’s Talk Loyalism, Loyalist Engagement Survey, Protocol, Policing and Politics (August 2021) 

  451. Northern Ireland Office, ‘Northern Ireland Terrorism Legislation: annual statistics 2020/21’, Table 3.1. 

  452. PSNI, ‘Policing Recorded Security Situation Statistics Northern Ireland’, Table 5. 

  453. Northern Ireland Office, ‘Northern Ireland Terrorism Legislation: annual statistics 2020/21’, Table 7.1. 

  454. 56 charged out of 185 arrests. 

  455. The explanation provided by Lord Anderson QC following an analysis by Alyson Kilpatrick BL was that there were particular difficulties in Northern Ireland in converting intelligence into evidence: Terrorism Acts in 2015 at 8.16 to 8.21. An addition explanation is the need for disruptive arrests on public safety grounds: Terrorism Acts in 2018 at 9.69, ft. 60. 

  456. Terrorism Act 2000 (Video Recording with Sound of Interviews and Associated Code of Practice) (Northern Ireland) Order 2020, SI 2020/860. 

  457. Figures provided by PSNI. 

  458. PBNI, Human Rights Annual Report 2019/20, recommendation 13. 

  459. By using the procedural justice model, Participate. 

  460. Terrorism Acts in 2019 at 9.79. 

  461. Source: Public Prosecutor Service of Northern Ireland. 

  462. Although the Code of Practice which governs the use of Schedule 7 of the Terrorism Act 2000 differs from the Code considered in Ramsay (No 2). 

  463. https://www.bbc.co.uk/news/uk-northern-ireland-55901429. 

  464. ISC, Northern Ireland-related Terrorism, HC 844 (5 October 2020). 

  465. I referred to the Colin Duffy trial in Terrorism Acts in 2019 at 9.85. 

  466. In his 2020 Report, supra, the Independent Reviewer of National Security Arrangements referred to some areas of the community in Northern Ireland being subject to unacceptable criminal acts and attitudes “at a level which has almost come to be regarded by many as normal”. Once criminal behaviour linked to terrorist groups is normalised, it is a short step to normalising their terrorist activity too, which is one implication of comparatively shorter terrorism sentences. 

  467. Ibid at paras 21 to 26. 

  468. Terrorism Acts in 2019 at 9.87. 

  469. Independent Reporting Commission, Third Report, HC 911 (17 November 2020). 

  470. Ibid at 9.89 to 9.93. Preliminary proceedings for terrorism cases in the magistrates can be costly as well as delaying: it was reported that legal aid of over £116,352 was spent on the magistrates proceedings in the case of Christine Connor, later convicted in the Crown Court of attempting to murder a police officer (out of a total of over £1m legal aid: https://www.belfastlive.co.uk/news/belfast-news/christine-connor-belfast-woman-jailed-19423062). 

  471. It also abolishes the use of oral evidence in committal proceedings for all remaining offences, which will now proceed by way of a preliminary inquiry (i.e. on the papers only). 

  472. Independent Reporting Commission, supra, at A11. 

  473. Practice Direction No.2/ 2019. 

  474. https://www.bbc.co.uk/news/uk-northern-ireland-54934138. 

  475. The Queen v Morgan and others [2020] NICC 14. Two defendants were sentenced to indeterminate terms and I have therefore doubled the custodial term specified by the judge to reach a notional determinate term starting point. 

  476. For example, R v Ciaran Maxwell, Central Criminal Court (31 July 2017) where the judge took a starting point of 27 years’ imprisonment. 

  477. Terrorism Acts in 2019 at 9.101. 

  478. DoJ, Sentencing Policy Review Consultation Way Forward, at para 72. 

  479. https://www.judiciaryni.uk/sentencing-guidelines-northern-ireland. 

  480. R v Lehd [2021] NCC 4 at para 30. 

  481. As in R v Lehd itself, linked to R v Ciaran Maxwell, supra. 

  482. R v Kahar [2016] EWCA Crim 568. 

  483. R v Lehd, supra, at para 31. 

  484. [2021] NICA 3, applying the non-terrorist case R v Loughlin [2019] NICA 10 in which the Lord Chief Justice held, at para 19, that the aggravating and mitigating factors referred to in the Sentencing Council Guidelines on Attempted Murder should be taken into account in determining the correct sentence. 

  485. Nash’s (Thomas) Application for leave to apply for judicial review [2015] NICA 18. 

  486. Note Verbale No.318 (12 October 2020). 

  487. McEvoy, K., Schwartz, A., ‘Judges, Conflicts and the Past’, Journal of Law and Society Vol.42 no.4(December 2015), comments on the strength of the independence of the Northern Ireland judiciary. 

  488. Non-Jury Trials: Northern Ireland Office Consultation (November 2020), Response by the Independent Reviewer of Terrorism Legislation (PDF, 116 KB) 

  489. Criminal Finances (2017 Act) (Commencement) Regulations (Northern Ireland) 2021/167. 

  490. Independent Reporting Commission, supra, at A16. The Department of Justice was informed by the Home Office that commencement was likely to be in the first quarter of 2021. 

  491. Schedule 5 Part II para B8. 

  492. Described in Terrorism Acts in 2019 at 10.6 to 10.15. 

  493. Legislative Consent Memorandum (February 2020); motion agreed to by Scottish Parliament on 20 February 2020. 

  494. Legislative Consent Memorandum (January 2021); motion agreed to by Scottish Parliament on 10 March 2021. 

  495. Terrorism Acts in 2019 at 10.14. 

  496. Terrorism Acts in 2019 at 10.12. 

  497. Section 17(3)(b) of the 1995 Act requires the Chief Constable to comply with any lawful instruction given by the Lord Advocate under section 12. 

  498. https://www.dailyrecord.co.uk/news/scottish-news/pensioner-admits-glorifying-acts-terrorism-25081539. He has now been sentenced to a restriction of liberty order for a period of 8 months and a supervision order for a period of 1 year. 

  499. Lord Beckett’s sentencing remarks (January 12 2021) are at https://www.judiciary.scot/home/sentences-judgments/sentences-and-opinions/2021/01/12/hma-v-gabrielle-friel. 

  500. A “not proven” verdict is unique to Scotland. It is available as an alternative to “not guilty” but equally results in an acquittal. 

  501. See Terrorism Acts in 2019 at 2.28 to 2.31. 

  502. And have rejected any attempt by the defendant to show that the possession of the article was not for a purpose connected with terrorism. 

  503. R v G and J [2009] UKHL 13, para 55. 

  504. This defence arises if the defendant adduces sufficient evidence to raise as an issue that he did not have the article for a terrorist purpose: sections 57(2) and 118 Terrorism Act 2000. 

  505. Possession of an offence weapon in a public place contrary to section 47 of the Criminal Law (Consolidation) (Scotland) Act 1995. In England and Wales the equivalent offence is section 1 of the Prevention of Crime Act 1953. 

  506. A judge can impose a supervised release order for individuals convicted on indictment: it requires the offender to be put under the supervision of a criminal justice social worker (the equivalent to a probation officer in the rest of the UK) and follow any conditions set. 

  507. Dudgeon v Her Majesty’s Advocate [2020] HCJAC 6. 

  508. See Chapter 1 on the (belated) publication of national section 43 statistics. 

  509. Glasgow Times, ‘Glasgow human right’s campaigner’s ‘racial profiling’ nightmare’ (28 October 2020), https://www.glasgowtimes.co.uk/news/18826184.glasgow-human-rights-campaigners-racial-profiling-nightmare/. The National, ‘Nicola Sturgeon: Mohammed Asif’s ‘unacceptable’ treatment shows change needed’ (29 October 2020).