Detention for Protection: Searching for a ‘Fair Balance’ between the Restrictions on Preventive Detention and the Obligation to Protect Individuals

2.1 Article 2

The Court has developed a rich jurisprudence on positive obligations under ECHR Articles 2 and 3, which respectively protect the right to life and the freedom from torture.⁵ The existence of positive obligations under Article 2 was first recognised in 1998, when the Court explained that ‘the first sentence of Article 2 § 1 enjoins the State not only to refrain from the intentional and unlawful taking of life, but also to take appropriate steps to safeguard the lives of those within its jurisdiction’.⁶ This notion has been further developed in subsequent case law to include a large number of substantive and procedural aspects which apply in a wide range of situations.⁷ In fact, the Court has expressed in general terms that the positive obligation ‘must be construed as applying in the context of any activity, whether public or not, in which the right to life may be at stake’.⁸ Despite the fact that the positive obligation applies generally, the Court has expressly stated that ‘[n]ot every claimed risk to life … can entail for the authorities a Convention requirement to take operational measures to prevent that risk from materialising’, and that ‘the scope of the positive obligation must be interpreted in a way which does not impose an impossible or disproportionate burden on the authorities’.⁹

In the present context, the most relevant issue is that a State may have a positive obligation to protect an individual’s life against threats that stem from another individual. The original authority is the Osman case, which concerned a man who was killed by a person who had stalked the man’s family for a long time. Here, the Court held that there is ‘in certain well-defined circumstances a positive obligation on the authorities to take preventive operational measures to protect an individual whose life is at risk from the criminal acts of another individual’,¹⁰ and further:

In the opinion of the Court where there is an allegation that the authorities have violated their positive obligation to protect the right to life in the context of their above-mentioned duty to prevent and suppress offences against the person … it must be established to its satisfaction that the authorities knew or ought to have known at the time of the existence of a real and immediate risk to the life of an identified individual or individuals from the criminal acts of a third party and that they failed to take measures within the scope of their powers which, judged reasonably, might have been expected to avoid that risk.¹¹

The key element here is that a positive obligation would only arise in the case of ‘a real and immediate risk to the life of an identified individual’. Having been previously introduced in earlier cases, this requirement has persisted throughout the Court’s subsequent elaboration of the scope of the positive obligation to protect the life of individuals.¹² Yet, the Court has also opened the door to a positive obligation to protect the life of individuals who cannot be identified in advance. The Mastromatteo case concerned a man who was killed by a prisoner on leave from prison. The Court distinguished this case from the Osman case by observing that:

it is not a question here of determining whether the responsibility of the authorities is engaged for failing to provide personal protection to A. Mastromatteo; what is at issue is the obligation to afford general protection to society against the potential acts of one or of several persons serving a prison sentence for a violent crime and the determination of the scope of that protection.¹³

The Court did not rule out that a risk to life by unidentified individuals may trigger a positive obligation to avert that risk; the Court expressed that:

it must be shown that the death of A. Mastromatteo resulted from a failure on the part of the national authorities to “do all that could reasonably be expected of them to avoid a real and immediate risk to life of which they had or ought to have had knowledge” …, the relevant risk in the present case being a risk to life for members of the public at large rather than for one or more identified individuals.¹⁴

The Court found no violation of Article 2 in this case, but a violation was found in the relatively similar Maiorano case. Here, a prisoner on day-leave had killed two women without any apparent motive. The Court considered that the prisoner’s criminal record and behaviour in prison indicated that granting the prisoner leave would represent a risk to the life of an individual, even if it were not possible to identify the individual in advance.¹⁵

In summarising the key issue, it is useful to use the Court’s own words, namely that the Court has drawn:

a distinction between cases concerning the requirement of personal protection of one or more individuals identifiable in advance as the potential target of a lethal act … and those in which the obligation to afford general protection to society was in issue.¹⁶

At the same time, it is also necessary to keep in mind that a violation of the positive obligations under Article 2 has been found in both categories of cases.

2.2 Article 3

Similar principles have been developed with regard to Article 3. In 1997, the Court did ‘not rule out the possibility that Article 3 of the Convention may also apply where the danger emanates from persons or groups of persons who are not public officials’, if it is ‘shown that the risk is real and that the authorities of the receiving State are not able to obviate the risk by providing appropriate protection’.¹⁷ The Court found a violation of this positive obligation for the first time in A. v. the United Kingdom in 1998, which concerned the beating of a child by his stepfather. The Court explained that States have an obligation ‘to take measures designed to ensure that individuals within their jurisdiction are not subjected to torture or inhuman or degrading treatment or punishment, including such ill-treatment administered by private individuals’, and in that particular case the Court considered that the domestic law failed to provide adequate protection for the victim.¹⁸ In the case of Z and Others v. the United Kingdom, which concerned the parental abuse of four children, the Court added the limitation that the positive obligation applies to situations of ‘ill-treatment of which the authorities had or ought to have knowledge’.¹⁹ All of these general statements have been reiterated in subsequent case law.²⁰

However, in other cases the Court has expressed a standard which is identical to that under Article 2. In the Milanović case, the Court expressed that for:

a positive obligation to arise, it must be established that the authorities knew or ought to have known at the time of the existence of a real and immediate risk of ill-treatment of an identified individual from the criminal acts of a third party and that they failed to take measures within the scope of their powers which, judged reasonably, might have been expected to avoid that risk.²¹

Accordingly, the requirement that a threat must relate to ‘an identified individual’ has also been introduced under Article 3. The Court has not yet expanded this notion also to include a duty to protect unidentified individuals at risk of ill-treatment from other individuals, but it certainly cannot be ruled out that such an expansion will be made in future cases in line with what has been done under Article 2.

With these general principles as a backdrop, it is time to turn to the cases brought against Germany to illustrate some of the dilemmas that the States encounter.

3.1 Jendrowiak v. Germany

Since 1972, the applicant had been convicted of several counts of rape and attempted rape and he was convicted again in 1990 for attempted sexual coercion. He was sentenced to three years’ imprisonment and thereafter placed in preventive detention, as he was considered likely to repeat his offence. In 2002, after having served ten years in preventive detention, the detention was prolonged pursuant to Article 67d § 3 of the German Criminal Code which had been amended in 1998 and, accordingly, entered into force after he had committed his offence.²² The previous version of the provision stated that the first period of preventive detention could not exceed 10 years, and that when the maximum sentence duration is reached, the detainee should be released. The amended version provided that if a person has spent 10 years in preventive detention, the detention shall be terminated by a court only when it can determine that the detainee is not in danger of committing certain new serious offences upon release. The former maximum duration of the first period of preventive detention was abolished.²³ Jendrowiak complained that this retrospective prolongation of his preventive detention beyond 10 years violated Article 5.1 ECHR.²⁴

In its assessment, the Court built considerably on its earlier analysis in the case of M. v. Germany,²⁵ expressing that ‘there was not a sufficient causal connection between the applicant’s conviction by the sentencing court and his continued deprivation of liberty beyond the period of ten years in preventive detention’.²⁶ The continued detention was therefore neither justified under Article 5.1(a) nor under any other sub-paragraph.²⁷ Sub-paragraph (c) was deemed inapplicable considering that the applicant’s ‘potential further offences were not sufficiently concrete and specific … as regards, in particular, the place and time of their commission and their victims’.²⁸ Importantly, what distinguishes the Court’s assessment here from that in M. v. Germany and other previous cases, is that the Court continued to discuss the relevance of the State’s positive obligations under Article 3. It is appropriate to quote the Court’s statement in full:²⁹

36. The Court is aware of the fact that the domestic courts ordered the applicant’s preventive detention beyond a period of ten years because they considered that there was still a risk that the applicant might commit serious sexual offences, in particular rape, if released. They thus acted in order to protect potential victims from physical and psychological harm amounting to inhuman or degrading treatment which might be caused by the applicant. Under the Court’s well-established case-law, States are indeed required under Article 3 to take measures designed to ensure that individuals within their jurisdiction are not subjected to torture or inhuman or degrading treatment, including such ill-treatment administered by private individuals … These measures should provide effective protection and include reasonable steps to prevent ill-treatment of which the authorities had or ought to have had knowledge …

37. However, the Court has also repeatedly held that the scope of any positive obligation on State authorities to take preventive operational measures to protect individuals from the criminal acts of another individual must take into consideration the need to ensure that the authorities exercise their powers to control and prevent crime in a manner which fully respects the due process and other guarantees which legitimately place restraints on the scope of their action, including the guarantees contained, in particular, in Article 5 of the Convention … In other words, the Convention obliges State authorities to take reasonable steps within the scope of their powers to prevent ill-treatment of which they had or ought to have had knowledge, but it does not permit a State to protect individuals from criminal acts of a person by measures which are in breach of that person’s Convention rights, in particular the right to liberty as guaranteed by Article 5 § 1.

38. Consequently, the State authorities could not, in the present case, rely on their positive obligations under the Convention in order to justify the applicant’s deprivation of liberty which … did not fall within any of the exhaustively listed permissible grounds for a deprivation of liberty under sub-paragraphs (a) to (f) of Article 5 § 1. That provision can thus be said to contain all grounds on which a person may be deprived of his liberty in the public interest, including the interest in protecting the public from crime.

The Court thus recalled its previous case law on positive obligations under Article 3, but expressed the general limitation that the implementation of such positive obligations must comply with the State’s obligations under other Articles. An individual cannot be held in preventive detention in violation of Article 5, even if the State considers this necessary to protect others against ill-treatment in line with the positive obligation under Article 3. On this basis, the Court concluded that Article 5 had been violated.

3.2 The Case of O.H. v. Germany

This case is relatively similar to the previous one, and concerns the same 1998 legislative amendment. Since 1970, the applicant had been convicted on a number of occasions for theft, robbery, extortion, assault and dangerous assault. Between 1970 and 1987, he spent 14 years in prison. In 1987, he was convicted on two counts of attempted murder, sentenced to nine years’ imprisonment and thereafter placed in preventive detention. In 1996, it was decided that the applicant’s preventive detention should be spent in a psychiatric hospital, based on the findings of a psychiatric report drawn up at the request of the competent court, which concluded that the applicant suffered from ‘schizophrenia simplex characterised by autistic behaviour and from a serious personality disorder’.³⁰ He was placed in preventive detention in 1996, but transferred to prison in 1999 as detention in a psychiatric hospital was found not to help his rehabilitation. In 2001, the continuation of the preventive detention was ordered by the court. Similar to the situation in Jendrowiak, the applicant complained that the retrospective extension of his preventive detention from the original maximum period of ten years to an unlimited period of time violated the right to liberty under Article 5 ECHR.³¹

A remarkable feature of the Court’s assessment in Jendrowiak is that the reference to the positive obligations under Article 3 does not appear to be based on submissions by the parties; in any event, no submissions to this effect are cited in the judgment. However, in O.H. v. Germany, the German government made a submission along these lines. According to the Court’s summary of the submissions, the Government expressed doubts as to whether a narrow interpretation of Article 5 was necessary to protect individuals from arbitrary detention, and that the interpretation of the provision ‘had to take into account the States’ duty, originating in human rights and, in particular, in Articles 2 and 3 of the Convention, to protect victims from further offences’.³² The Government submitted further that the term ‘conviction’ in Article 5.1(a):

should not only comprise a finding of guilt in respect of an offence and the imposition of a measure involving deprivation of liberty by the sentencing court, but that it should also comprise the decisions of the courts responsible for the execution of sentences to extend a person’s detention depending on the danger he or she presented.³³

Unsurprisingly however, the Court simply reiterated its position in the Jendrowiak case, departing from Jendrowiak only in that the positive obligation under Article 2 was included side by side with the obligation under Article 3. Consequently, the Court considered that Article 5 had been violated.

3.3 Schwabe and M.G. v. Germany

This case concerned entirely different facts, but similar questions arose. From 6 to 8 June 2007, a G8 summit was held in Heiligendamm in the vicinity of Rostock in Germany. Prior to the summit, the police considered that there was a terrorist threat and a risk of riots and violent demonstrations. The two applicants intended to participate in demonstrations against the summit. They were arrested on 3 June, and a court ordered their detention until 9 June. The direct cause for their arrest was that they had been standing in a car park in front of a prison with folded banners in their van, which bore the inscriptions ‘freedom for all prisoners’ and ‘free all now’. Additionally, they had physically resisted an identity check by the police. On this basis, the national courts deemed the applicants’ detention lawful in order to prevent the imminent commission or continuation of a criminal offence, namely that of inciting others to free prisoners. The courts also considered that the detention was indispensable and proportionate.³⁴ In proceedings before the Court of Appeal on 7 June, the applicants submitted that this interpretation of the banners was far-fetched, and that the banners were not meant to incite others to attack prisons and to free prisoners. The banners were addressed to the police and the authorities, urging them to end arrests and detentions of demonstrators.³⁵ The Court of Appeal nevertheless upheld the assessments and the decision of the lower courts, adding that the police ‘had been entitled to assume that the applicants had intended to drive to Rostock and display the banners at the partly violent demonstrations there. As a result, a crowd which had been ready to use violence might have been incited to liberate people who had been arrested and detained.’³⁶ The Court of Appeal also considered that ‘in the tense situation in and around Rostock the police had been authorised to prevent ambiguous declarations which could have led to a risk to public security and order’.³⁷

To the European Court of Human Rights, the applicants complained that their detention for preventive purposes during the G8 summit had violated Article 5 ECHR, as it was not justified under any sub-paragraphs of that provision.³⁸ Of particular relevance to the case were sub-paragraph (c), which grants detention ‘when it is reasonably considered necessary to prevent his committing an offence’, and sub-paragraph (b), which permits detention ‘of a person for non-compliance with the lawful order of a court or in order to secure the fulfilment of any obligation prescribed by law’. Both parties presented a number of submissions, of which the present context calls for the description of only one. The German Government submitted that:

in Germany such detention for preventive purposes was necessary, as acts preparing criminal offences were, as a rule, not yet punishable, contrary to the criminal law applicable in other Contracting Parties to the Convention. This served to encourage potential offenders to give up their plans to commit an offence. Without the possibility to detain persons for preventive purposes, the State would therefore be unable to comply with its positive obligation to protect its citizens from impending criminal offences …³⁹

In its assessment, the Court initially considered that the detention of the applicants was not justified under either sub-paragraph (b) or (c), for reasons that need not be revisited here, but the Court continued to address the submission concerning positive obligations:

The Court further takes note of the Government’s argument that without the possibility of detaining individuals for preventive purposes, the State would be unable to comply with its positive obligation to protect its citizens from impending criminal offences. In the case at hand, however, even taking into account the general situation before and during the G8 summit, it has not been sufficiently demonstrated that a liberation of prisoners had been imminent. Therefore, the commission of that offence could not justify an interference with the right to liberty, especially as less intrusive measures could have been taken … The Court reiterates that, in any event, the Convention obliges State authorities to take reasonable steps within the scope of their powers to prevent criminal offences of which they had or ought to have had knowledge. However, it does not permit a State to protect individuals from criminal acts of a person by measures which are in breach of that person’s Convention rights, in particular the right to liberty as guaranteed by Article 5 § 1 …

Accordingly, the standard from Jendrowiak was applied also in this case, regardless of the different factual circumstances.

3.4 The case of B. v. Germany

In 2000, the applicant was sentenced to nine years’ imprisonment for sexual assault and rape. He had previously been convicted on several other counts of rape. He served his full sentence until July 2008, and was thereafter provisionally placed in preventive detention pending a decision from a competent court on whether to issue retrospective preventive detention.⁴⁰ This was ordered by the Regional Court in October 2008. The court considered it very likely that the applicant would again commit sexual offences, resulting in considerable psychological or physical harm to the victims, that preventive detention was the only suitable way to prevent further offences, and that other measures were insufficient to protect the public.⁴¹ The decision was subsequently upheld by higher courts. Again, the applicant complained that the retrospective preventive detention violated Article 5.1 ECHR. The Government reiterated its submissions delivered in the case of O.H. v. Germany, namely that the interpretation of Article 5.1 ‘had to take into account the States’ duty, originating in human rights and, in particular, in Articles 2 and 3 of the Convention, to protect victims from further offences’.⁴²Again, though, the Court was unimpressed, and concluded once more that Article 5 was violated. The Court considered that sub-paragraph (a) could not justify the detention, since ‘there was no sufficient causal connection between the applicant’s conviction in 2000 and his – retrospective – preventive detention since July 2008’.⁴³ Subsequently, the Court recalled its considerations from the cases of Jendrowiak and O.H., which may be reiterated here as a segue to the next section:⁴⁴

Articles 2 and 3 of the Convention do not permit a State to protect individuals from criminal acts of a person by measures which are in breach of that person’s Convention rights, in particular the right to liberty as guaranteed by Article 5 § 1. Consequently, the State authorities cannot, in the present case, rely on their positive obligations under the Convention in order to justify the applicant’s deprivation of liberty which … did not fall within any of the permissible grounds for deprivation of liberty exhaustively listed under sub-paragraphs (a) to (f) …

4.1 The strict interpretation of Article 5

The Court has, with good reason, always been reluctant to accept preventive detention. Two legal arguments are used to form the primary justification for this.

First, the Court has always insisted that Article 5.1(a)–(f) ‘contain an exhaustive list of permissible grounds for deprivation of liberty, and no deprivation of liberty will be lawful unless it falls within one of those grounds’, as the Court now prefers to phrase it.⁴⁵ The exhaustive character of the sub-paragraphs was previously established in the famous Engel case,⁴⁶ and the Court has never deviated from this position. Unless preventive detention in a particular context falls within one of the grounds listed in the sub-paragraphs, it will be considered unlawful. This does not mean, though, that preventive detention is unlawful per se, since it is beyond doubt that preventive detention under certain circumstances can satisfy the requirements in several of the sub-paragraphs. The Court has, for example, accepted some forms of preventive detention under sub-paragraph (a) if the detention is ‘directly linked to the applicant’s initial conviction … and can thus be regarded as “lawful detention … after conviction by a competent court”’,⁴⁷ and sub-paragraph (c) permits, per definition, some forms of preventive detention when this is ‘reasonably considered necessary to prevent his committing an offence’.

Secondly, the Court has developed a very strict interpretation of the sub-paragraphs. To consider detention lawful under sub-paragraph (a), there must be a clear causal connection between the conviction and the detention. Sub-paragraph (c), which would appear to allow for preventive detention to a rather considerable extent, has always been interpreted in a manner which limits its scope of application. Ever since the third Lawless judgment in 1961, the Court has maintained that ‘it is evident that the expression “effected for purpose of bringing him before the competent legal authority” qualifies every category of cases of arrest or detention referred to’ in sub-paragraph (c),⁴⁸ and this has been confirmed in very recent decisions.⁴⁹ Detention of someone in order to prevent him from committing an offence is lawful only if the purpose of the detention is to bring him before the courts or another competent legal authority, and the Court has later explained that sub-paragraph (c) ‘thus permits deprivation of liberty only in connection with criminal proceedings’.⁵⁰ Even more recently, the Court has explained that it ‘has long been established that the list of grounds of permissible detention in Article 5 § 1 does not include internment or preventive detention where there is no intention to bring criminal charges within a reasonable time’.⁵¹

However, the proper legal justification for this strict position is not clear. The argument in Lawless (No. 3) was only that this interpretation ‘is evident’, and that the interpretation is confirmed by the relationship with Article 5.3, which provides that anyone detained in accordance with Article 5.1(c) ‘shall be brought promptly before a judge … and shall be entitled to trial within a reasonable time’. So far, so good, but the more recent specification – i.e. that the provision ‘permits deprivation of liberty only in connection with criminal proceedings’ – was not explained. This is a more restrictive interpretation which needs further justification.

To illustrate the point, it is useful to play the devil’s advocate. From a linguistic point of view, while the Court’s interpretation from Lawless (No. 3) onwards is the most likely one, it is not ‘evident’. Sub-paragraph (c) could, in principle, be read in two ways, and the inclusion of some markers may illustrate the point.

The first reading is the Court’s interpretation: the lawful arrest or detention of a person effected for the purpose of bringing him before the competent legal authority [, either] 1) on reasonable suspicion of having committed an offence, or 2) when it is reasonably considered necessary to prevent his committing an offence. Here, it is linguistically clear that the requirement ‘purpose of bringing him before the competent legal authority’ applies to both situations.

The second reading is as follows: the lawful arrest or detention of a person 1) effected for the purpose of bringing him before the competent legal authority on reasonable suspicion of having committed an offence, or 2) when it is reasonably considered necessary to prevent his committing an offence. Here, the requirement would apply only to the first situation.

This linguistic point should not be exaggerated, but it is not merely a pedantic argument without practical implications. In an earlier draft of Article 5.1(c), the provision referred to ‘the lawful arrest or detention of a person effected for the purpose of bringing him before the competent legal authority on reasonable suspicion of having committed an offence or which is reasonably considered to be necessary to prevent his committing an offence or fleeing after having done so’.⁵² From a purely linguistic point of view, this wording – where ‘when it is’ is replaced with ‘which is’ – would have made the second interpretation above more prominent, because the first interpretation would result in the grammatically incorrect sentence ‘the purpose of bringing him before the competent legal authority … which is reasonably considered to be necessary to prevent his committing an offence’. It is unfortunate that the late amendment to the wording of the provision is not explained in the preparatory works.

The preparatory works do contain, however, an explanatory comment from the Conference of Senior Officials to the Committee of Ministers:

The Conference considered it useful to point out that where authorised arrest or detention is effected on reasonable suspicion of preventing the commission of a crime,⁵³ it should not lead to the introduction of a régime of a Police State. It may, however, be necessary in certain circumstances to arrest an individual in order to prevent his committing a crime, even if the facts which show his intention to commit the crime do not of themselves constitute a penal offence. In order to avoid any possible abuses of the right thus conferred on public authorities, [Article 18] will have to be applied strictly.⁵⁴

In sum, the preparatory works appear to suggest that the intention of the Convention drafters was to consider the detention of a person lawful when it is reasonably considered necessary to prevent his committing an offence, as long as the requirements in Article 5.3 are complied with. The statement also seems to contradict the position that Article 5.1(c) only permits detention when the intention is to bring criminal charges against the detainee.

Given that Article 5.3 requires the detainee to be brought promptly before a judge and be entitled to trial within a reasonable time, one might argue that it does not in practice matter whether the requirement in Article 5.1(c) of ‘bringing him before the competent legal authority’ also applies. But there are two possible implications. First, the requirement in Article 5.1(c) is that ‘bringing him before the competent legal authority’ shall be the purpose of the detention, and this is not the case with preventive detention, where the purpose is to prevent someone from committing an offence. Secondly, it is, presumably, the application of this requirement that has inspired the Court to consider that Article 5.1(c) ‘permits deprivation of liberty only in connection with criminal proceedings’, which, in the present writer’s opinion, is an erroneous interpretation.⁵⁵

4.2 A side-glance at other international human rights instruments

Before proceeding, it is useful to recall that the ECHR is unique in providing an exhaustive list of permissible grounds for deprivation of liberty. Other human rights treaties that protect the right to liberty and security all contain more open-ended provisions.

Article 9.1 of the International Covenant on Civil and Political Rights reads:

Everyone has the right to liberty and security of person. No one shall be subjected to arbitrary arrest or detention. No one shall be deprived of his liberty except on such grounds and in accordance with such procedure as are established by law.

Along the same lines, Article 7 of the American Convention on Human Rights provides that everyone ‘has the right to personal liberty and security’, that no one ‘shall be deprived of his physical liberty except for the reasons and under the conditions established beforehand by the constitution of the State Party concerned or by a law established pursuant thereto’ and, generally, that no one ‘shall be subject to arbitrary arrest or imprisonment.’

Article 6 of the African Charter on Human and Peoples’ Rights follow, more or less, the wording of the International Covenant on Civil and Political Rights, albeit with a different order:

Every individual shall have the right to liberty and to the security of his person. No one may be deprived of his freedom except for reasons and conditions previously laid down by law. In particular, no one may be arbitrarily arrested or detained.

Article 14 of the Arab Charter on Human Rights follow the same pattern: ‘Everyone has the right to liberty and security of person. No one shall be subjected to arbitrary arrest, search or detention without a legal warrant’ and no one ‘shall be deprived of his liberty except on such grounds and in such circumstances as are determined by law and in accordance with such procedure as is established thereby.’

All of these treaties provide for a more general prohibition against arbitrary detention, without providing an exhaustive list of grounds for detention that is not considered arbitrary. Given this wording, the assessment of preventive detention may stand in a quite different light. As long as preventive detention is not ‘arbitrary’, and as long as certain procedural requirements are complied with, the detention is lawful. It is worth noting at this stage that the initial proposals for Article 5 ECHR were formulated along the same lines, but the United Kingdom objected that this was unsatisfactory. In the UK Government’s view, ‘[t]he word “arbitrary” is vague and its force undefined’, and the exhaustive list was therefore eventually adopted instead.⁵⁶ However, with this context in mind, one might certainly discuss whether it is appropriate for the Court to go even one step further, by interpreting the sub-paragraphs very strictly. In the present writer’s opinion, the over-arching argument should still be that it is the arbitrariness that must be prevented.

4.3 The tension between positive and negative obligations: The development of a hierarchy?

Under ordinary circumstances, the Court’s approach is reasonable and well justified from a human rights perspective: by restricting the possibility of States to detain individuals for preventive reasons, the protection of those individuals against arbitrary detention is strengthened. The present author fully supports this position. However, the parameters for the assessment may change when an individual’s right to liberty comes into conflict with the rights and interests of other individuals or of the society, particularly when the conflicting interest is also a recognised human right. This brings us to the general tension between positive obligations under one provision and negative obligations under another.

Positive obligations have now been recognised under practically all provisions in the ECHR, and the duties that are imposed on States are quite far-reaching. It is not only a question of having an adequate legislative framework or of investigating and prosecuting wrong-doers; States also have a duty to prevent certain acts. As shown above, Articles 2 and 3 provide a positive obligation for States to take preventive, operational measures to protect an individual against certain criminal acts of another individual. More generally, the positive obligation to take active measures to protect the life of individuals against various risks is sometimes articulated as a duty for the State to take ‘all necessary measures’ to protect someone from being exposed to danger,⁵⁷ or as a duty to do ‘all that could have been required of it’.⁵⁸ The Court has, however, developed a number of standards that limit the scope of the positive obligations, and the whole notion of positive obligations is, in and of itself,(?) flexible and case-specific. States are required, for example, to take ‘measures within the scope of their powers’ or ‘appropriate steps’ to prevent human rights infringements of which they were or ought to have been aware, and the Court has consistently held that the scope of the positive obligation must be interpreted in a way which does not impose an impossible or disproportionate burden on the authorities. Yet, even with all these limitations, a State will still be faced with challenges with regard to how positive obligations may be implemented. If the State knows that someone represents a risk to the life or physical integrity of other people, how does it prevent that risk from materialising? This will, of course, depend on the circumstances in each particular case, but the general question of relevance in the present context is to what extent authorities may infringe on other rights in the implementation of their positive obligations.

As shown above, the Court has outright refused States to infringe on Article 5.1 in their implementation of positive obligations under Article 2 and 3. Recall from Jendrowiak and the subsequent case law that authorities cannot ‘rely on their positive obligations under the Convention in order to justify the applicant’s deprivation of liberty which … did not fall within any of the exhaustively listed permissible grounds for a deprivation of liberty under sub-paragraphs (a) to (f) of Article 5 § 1.’⁵⁹ Violating the negative obligations under Article 5 is simply not an option which is ‘within the scope of the powers’ of the State. This is a commendable position in its clarity and foreseeability: States must simply find other measures to implement their positive obligations, namely measures which do not infringe on other rights. But the devil’s advocate can also construct counter-arguments to the position: the position is not only clear – it is also inflexible and rigid. First, while the position may be commendable in itself, it brings with it the strict and narrow interpretation of the sub-paragraphs in Article 5.1, including sub-paragraph (c). While everyone, presumably, agrees that preventive detention should be restricted to a minimum, the present writer fails to see an inherent reason why preventive detention should be assessed in the exact same manner when its purpose is to protect others against threats to their life or physical integrity, in contrast to when no such interest is at stake. The strict interpretation of Article 5.1(c) is not ‘evident’, and the Court would not have to avail itself of any particular legal creativity to interpret the provision more flexibly when it stands in possible conflict with the rights and interests of others.

Secondly, the rigid position could – if one interprets it somewhat malevolently – suggest that the Court establishes a hierarchy of norms, where negative obligations are more important than positive obligations. It is commonly considered that the Convention contains no formal hierarchy of rights, and one might presume that the Court would be reluctant to develop a hierarchy between negative and positive obligations. In this regard, it is worth noting that the Court has taken a different approach to the relationship between positive and negative obligations when other rights are at stake. The Opuz case is a good illustration.⁶⁰ The case concerned serious domestic violence that had resulted in the ill-treatment of the applicant and the death of her mother, and one of the questions in the case was whether the national authorities had fulfilled their positive obligation to take preventive, operational measures to protect the right to life of the applicant’s mother.⁶¹ In its submissions, the Government argued that ‘each time the prosecuting authorities commenced criminal proceedings against [the perpetrator], they had to terminate those proceedings, in accordance with the domestic law, because the applicant and her mother withdrew their complaints. In their opinion, any further interference by the authorities would have amounted to a breach of the victims’ Article 8 rights.’⁶² Under Article 8, everyone has the ‘right to respect for his private and family life, his home and his correspondence’, and national authorities have a (negative) obligation to respect this right. Although in this case, the Court did not consider that the authorities had to implement their positive obligations under Article 2 without infringing on the right under Article 8; instead, the Court explained that it had to ‘examine whether the local authorities struck a proper balance between the victim’s Article 2 and Article 8 rights’,⁶³ and that there is a ‘duty on the part of the authorities to strike a balance between a victim’s Article 2, Article 3 or Article 8 rights in deciding on a course of action’.⁶⁴ The Court continued:

As regards the Government’s argument that any further interference by the national authorities would have amounted to a breach of the victims’ rights under Article 8 of the Convention, the Court recalls its ruling in a similar case of domestic violence …⁶⁵ where it held that the authorities’ view that no assistance was required as the dispute concerned a ‘private matter’ was incompatible with their positive obligations to secure the enjoyment of the applicants’ rights. Moreover, the Court reiterates that, in some instances, the national authorities’ interference with the private or family life of the individuals might be necessary in order to protect the health and rights of others or to prevent commission of criminal acts …⁶⁶ The seriousness of the risk to the applicant’s mother rendered such intervention by the authorities necessary in the present case.⁶⁷

While the Court only refers to the Article 8 rights of the victim, one might presume that the Article 8 rights of the perpetrator would have to be assessed in the same way. If so, the applicable principle is that if there is a conflict between a State’s negative obligations towards individual A under Article 8 and its positive obligations towards individual B under Article 2, then a fair balance must be struck between the two. Since Article 8.2 allows the State to interfere with the rights contained in Article 8 as long as conditions relating to legality, legitimacy and proportionality are complied with, this is not problematic. However, if there is a similar conflict between negative obligations under Article 5 and positive obligations under Articles 2 and 3, there is no call for a ‘fair balance’. Is this simply a coincidence, or is it a deliberate position by the Court? If it is deliberate, one may inquire whether the apparent difference between Articles 5 and 8 in this regard is sufficiently justified, though this line of inquiry is not pursued here. Instead, let us pursue the idea that a conflict also between negative obligations under Article 5 and positive obligations under Articles 2 and 3 could be approached in the search for a ‘fair balance’. What would this entail?

4.4 The search for a ‘fair balance’

Aharon Barak has, most eloquently, characterised human rights as ‘the jewels in democracy’s crown’; he has said that a ‘democracy has no raison d’être without human rights; extracting human rights from a democracy would leave it soulless, an empty vessel’. However, he used these words merely to set the scene for the converse point, namely that ‘the concept of a “right” derives from the concept of society; without society, rights have no meaning’.⁶⁸ In Barak’s words, ‘[h]uman rights are rights of humans as members of society vis-à-vis others whether collectively or individually’, and this relationship leads to a basic dilemma in all societies of determining what limitations to place on the will and interest, or on the rights and freedoms, of individuals.⁶⁹ It is undeniable that the protection of human rights is not the only relevant consideration to be taken into account in democratic societies, and that human rights must be subject to certain limitations. One category of limitations concerns the necessity to protect the rights and freedoms of other individuals. This limitation is expressly recognised in the Universal Declaration of Human Rights, Article 29, and it is inherent in all subsequent human rights treaties, even if not expressly recognised. Another category of limitations concerns constraints that are necessary to protect the general interests of the society. When combined, one ends up with the fundamental question of to what extent limitations of the human rights of individuals may be justified by the demands of the rights and interests of others or of general interests of the society. To respond to this question, one must search for a fair balance between the conflicting interests.

Even the most cursory investigations into the case law of the European Court of Human Rights reveal that this search for a fair balance is of fundamental importance. Already in the famous Belgian Linguistic case in 1968, the Court expressed how the Convention ‘implies a just balance between the protection of the general interest of the Community and the respect due to fundamental human rights while attaching particular importance to the latter.’⁷⁰ The scope of the principle was subsequently applied in cases concerning several different provisions, until the Court in the 1982 Sporrong and Lönnroth case found that the time had come to proclaim that the search for a ‘fair balance … between the demands of the general interest of the community and the requirements of the protection of the individual’s fundamental rights … is inherent in the whole of the Convention’.⁷¹ This articulation of the search for a fair balance is reiterated in a large number of decisions.⁷²

With regard to Article 5, the case law reveals a somewhat fragmented picture. In some cases, although they are rare, the Court refers explicitly to a ‘fair balance’ standard in its assessment of the lawfulness of a detention. In a case concerning sub-paragraph (a), the Court was ‘of the opinion that there is nothing to establish that at the time of the proceedings at issue the [national courts] had not struck a fair balance between the applicant’s interest to be released and the safety interest of the public’.⁷³ In cases concerning sub-paragraph (b), the Court has considered that the authorities ‘failed to strike a fair balance between the need to ensure the immediate questioning of the applicant and the importance of the right to liberty’,⁷⁴ and that ‘the domestic authorities must strike a fair balance between the importance in a democratic society of securing compliance with a lawful order of a court, and the importance of the right to liberty’.⁷⁵ Yet in another, more high-profile, case, the Court expressed generally:

The Court does not accept the Government’s argument that Article 5 § 1 permits a balance to be struck between the individual’s right to liberty and the State’s interest in protecting its population from terrorist threat. This argument is inconsistent not only with the Court’s jurisprudence under sub-paragraph (f) but also with the principle that paragraphs (a) to (f) amount to an exhaustive list of exceptions and that only a narrow interpretation of these exceptions is compatible with the aims of Article 5. If detention does not fit within the confines of the paragraphs as interpreted by the Court, it cannot be made to fit by an appeal to the need to balance the interests of the State against those of the detainee.⁷⁶

In a dissenting opinion in another case, three judges also expressed that ‘the wording of Article 5 in itself strikes the fair balance inherent in the Convention between the public interest and the individual right to liberty by expressly limiting the purposes which a deprivation of liberty may legitimately pursue’,⁷⁷ implying that there is no room for further balancing. This appears to be the prevailing view in the Court.

In his excellent book on ‘Fair Balance’, Jonas Christoffersen has discussed the role of the proportionality principle as a mechanism to limit human rights. In that context, he observes that ‘[f]ocusing on the interference in fundamental human rights often creates the impression that the measure affecting the right is of a dubious character and amounts to a prima facie violation’,⁷⁸ and that this is not a fruitful approach. In the present writer’s view, such an impression is generally unfortunate – a limitation of human rights is not per definition a prima facie violation of human rights, since many human rights provisions expressly permit limitations. All sub-paragraphs in Article 5 represent limitations to the general right to liberty and security of person, and their existence means that a detention does not amount to a prima facie violation of the right to liberty and security. Such an approach could create a presumption of violation, and there is no legal basis for this. A relevant inquiry in this regard could be whether the strict interpretation of the sub-paragraphs to a certain extent builds on such a notion of detention as a prima facie violation of Article 5, but the available materials do not suggest that this has been a deliberate approach.

If one were to open the door to striking a ‘fair balance’ between negative obligations under Article 5 and positive obligations under Articles 2 and 3, one would encounter the following general problem: any search for a fair balance calls for an assessment of what is being balanced against what. How can community interests be weighed and balanced against individual interests? How can a (certain) interference with one individual’s right to liberty be weighed against an (uncertain) interference with another individual’s right to life? The argument is often made that the search for a fair balance calls for the balancing of incommensurable values, but there are ways to solve the problem. Barak’s recommendation, for example, is to focus on the social marginal importance on both sides of the scale; the marginal increase in the goal compared to the marginal harm to the right.⁷⁹ If one also identifies the relative social relevance of these marginal aspects, one may reach a situation of balancing of commensurable values.