JOINT PARTLY DISSENTING OPINION OF JUDGES LAZAROVA TRAJKOVSKA, DE GAETANO AND GROZEV

PARTLY DISSENTING OPINION OF JUDGE PINTO DE ALBUQUERQUE

Table of contents

I. Introduction (§§ 1 ‑ 2)

First Part (§§ 3 ‑ 33)

II. Soft law in international law (§§ 3 ‑ 9)

A. The sources of international law in Article 38 § 1 of the ICJ Statute (§§ 3 ‑ 7)

B. Soft law where there is codification (§ 8)

C. Soft law where there is little or no codification (§ 9)

III. Soft law in European human rights law (§§ 10 ‑ 22)

A. The constitutional principle of evolutive interpretation (§§ 10 ‑ 13)

B. Deformalisation of sources of law (§§ 14 ‑ 20)

C. European consensus as the conceptual framework of normativity (§§ 21 ‑ 22)

IV. Soft law and the Council of Europe’s rule of recognition (§§ 23 ‑ 33)

A. The rule of recognition of a democratic international community (§§ 23 ‑ 26)

B. The deep-seated commitment to hardening soft law (§§ 27 ‑ 30)

C. The constitutional prohibition on softening hard law (§§ 31 ‑ 33)

Second Part (§§ 34 ‑ 59)

V. The hardening of prison law (§§ 34 ‑ 42)

A. Worldwide (§§ 34 ‑ 38)

B. In the Council of Europe (§§ 39 ‑ 42)

VI. The fight against prison overcrowding (§§ 43 ‑ 47)

A. The minimum living space in the ICRC standards (§§ 43 ‑ 45)

B. The minimum living space in the EPR standards (§§ 46 ‑ 47)

VII. The application of Convention standards to the present case (§§ 48 ‑ 59)

A. The majority’s contradictory cumulative approach (§§ 48 ‑ 53)

B. A coherent, pro persona cumulative approach (§§ 54 ‑ 59)

VIII. Conclusion (§§ 60 ‑ 63)

I. Introduction (§§ 1 ‑ 2)

1. Unlike the majority, I voted in favour of finding a violation of Article 3 of the European Convention on Human Rights (the “Convention”) with regard to the placement of the applicant in Bjelovar prison, during the entire period of time in which he disposed of less than 4 sq. m of personal space.

2. Since the majority assume that they are not bound by the standards set by the Committee of Ministers, the Committee for the Prevention of Torture (the CPT) and the Council for Penological Cooperation (PC-CP) of the European Committee on Crime Problems (CDPC) of the Council of Europe, I address, in the first part of this opinion, the underlying crucial issue of the legal nature of these standards [4] . After an introductory note on the role of soft law in general international law, I turn to a detailed analysis of its specific function within European human rights law, whereby I seek to prove that evolutive interpretation, European consensus and hardening of soft law compose the three pillars of the Council of Europe’s normative system. For the limited purposes of this opinion, I entertain a discussion on the Council of Europe’s rule of recognition, having regard to its long-standing, deep-seated commitment to the hardening of soft law in certain legal fields with a view to the “further realisation of human rights” and “economic and social progress” in Europe.

Subsequently, in the second part of this opinion, I will show that there has been a pan-European and worldwide trend towards hardening prison soft law in view of the phenomenon of prison overcrowding. I further demonstrate that the European Prison Rules (EPR) are the prototype of hardened soft law in the Council of Europe’s normative system. On the basis of the crystal-clear standards set out in this hardened soft law, I conclude that the majority are mistaken when they find that areas of 3 sq. m and even less of floor surface in multi-occupancy accommodation in prisons do not breach Article 3 of the Convention.

First Part (§§ 3 ‑ 33)

II. Soft law in international law (§§ 3 ‑ 9)

A. The sources of international law in Article 38 § 1 of the ICJ Statute (§§ 3 ‑ 7)

3. Soft law has been considered an inherently redundant and even pernicious contradictio in terminis . For some it is an empty catchphrase disguising an inflating conception of international law, an attempt to impose on States political engagements to which they did not wish to consent in the first place [5] . For others, soft law is a “fig leaf for power”, hiding the extensive power and influence of some States and non-State actors in the international arena, a problematic instrument which circumvents State consent and therefore the domestic democratic ratification process [6] . In both cases, the principle of sovereign and equal States is supposedly in jeopardy.

4. Admittedly, soft law is not included among the classical sources of international law listed in the Statute of the International Court of Justice (the “ICJ Statute”). Any jurisprudential or doctrinal attempt to extend that list carries with it a heavy burden of proof. This is all the more so, the critique goes, because deformalisation of sources of law would undermine the clarity and predictability provided by a binary approach to the definition of international law. Put simply, soft law would be nothing but a by-product of politics, causing an inexorable drift towards anarchy and randomness and thus a disservice to the essential function of international law [7] . Claims for a return to all-or-nothing, black-and-white, binary simplicity, which alone could face the everyday complexity with its simplifying dichotomic rigour, have been raised against the cataclysmic Leviathan of soft law.

But the argument drawn from the ICJ Statute is not decisive in the pitched battle between opponents and advocates of soft law.

5. Article 38 of the ICJ Statute, which corresponds to Article 38 of the Statute of the former Permanent Court of International Justice, is not exhaustive in itself, in view, for instance, of the existence of unilateral legal acts [8] or legal acts of international organisations. In the light of the Copernican change in international law after the Second World War [9] , it would be difficult to maintain that Article 38 is an immutable, enumerative provision, which petrified international law once and for all according to its stage of development in 1920.

6. Since the question is not closed by the ICJ Statute, it may usefully be reframed if soft law is confronted with hard law in order to verify their respective essential features and the added value of soft law, if any.

No unique substantive legal parameter is decisive for the purposes of distinguishing between hard and soft international law. Like hard law, soft law aims at setting a general rule of conduct for its addressees. Both exhibit a normative claim, with a command-like structure, which may be couched with more or less accurate terminology and precise content. Regardless of the multiple forms it assumes, soft law may appear as any other ready-to-apply norm of treaty or customary law. As a further complication, international law is not entirely hard in terms of its enforceability and justiciability. Traditionally, a lack of enforcement measures and judicial review makes international law soft.

7. Nevertheless, international soft law is distinct and distinguishable from hard law by its consequences. In spite of its normative claim, soft law may, in principle, be disregarded without the classical consequences of responsibility for internationally wrongful acts. But its non-observance may bring about adverse consequences. Hence, the addressees of soft law are not entirely free not to follow it, because they may have to endure other negative consequences of such a choice. These are not merely moral, political or reputational in nature.

B. Soft law where there is codification (§ 8)

8. Where treaty law exists, complementary soft law may reveal the intention of its authors [10] . While strengthening the normative commitments embedded within the binding aspects of treaty law, soft law adds to its normative density and coherence [11] . It also facilitates the application of binding instruments by resolving complex, technical issues that were not envisaged when they were approved or deadlocks that had not been anticipated. It adapts them more easily to the changing needs of organisations, institutions and societies [12] .

C. Soft law where there is little or no codification (§ 9)

9. Where there is no codification at all or insufficient codification of international law, soft law counts as a relevant practice of international organisations, States and non-State operators. When a legal issue becomes the object of international soft law, it is no longer part of the reserved domain of States and this paves the way for future binding international law based on State consent [13] . It may ultimately be formative of the opinio juris and State practice that generates customary international law [14] .

Significantly, no effort has been made until today to shed light on the way the Council of Europe – and especially the Court – uses soft law, in spite of the proliferation of all sorts of deformalised sources of law in its legal discourse and some criticism as to the muddying of the waters between law and politics in the Court’s reading of the Convention [15] . The following reflections seek to provide such conceptualisation.

III. Soft law in European human rights law (§§ 10 ‑ 22)

A. The constitutional principle of evolutive interpretation (§§ 10 ‑ 13)

10. The Convention cannot be interpreted in a vacuum, but must be interpreted in harmony with other international law and soft law. Ever since Golder , account must be taken of any relevant principles and rules of international law applicable in the relations between the parties, as indicated in Article 31 § 3 (c) of the Vienna Convention on the Law of Treaties of 1969 [16] . From the Court’s perspective, there is no methodological difference between the interpretation of international human rights law and other international law, or between contractual and law-making treaties, and therefore it assumes that the same interpretative methods can be applied in both fields of international law. Thus the European Court of Human Rights (the “Court”) departs from the contested position that there are “self-contained regimes” within international law [17] . As Judge Rozakis so elegantly put it, the judges of Strasbourg “do not operate in the splendid isolation of an ivory tower built with material originating solely from the [Court]’s interpretative inventions or those of the States part[ies] to the Convention” [18] .

11. This methodology is warranted by the Court’s cardinal principle of interpretation to the effect that the Convention must be interpreted in the light of present-day conditions [19] . It was in the seminal case of Tyrer v. the United Kingdom that the Court for the very first time used the leitmotiv of “the Convention as a living instrument”, whose interpretation has to take account of evolving norms of national and international law [20] . Deeply entrenched in American [21] and Canadian [22] constitutional law since the early twentieth century, this interpretation technique was introduced in European human rights law in 1978.

12. In Tyrer , confronted with the arguments advanced by the Attorney General of the Isle of Man under former Article 63 of the Convention that, “having due regard to the local circumstances in the Island”, the continued use of judicial corporal punishment on a limited scale was justified as a deterrent, the Court replied (§ 38):

“it is noteworthy that, in the great majority of the member States of the Council of Europe, judicial corporal punishment is not, it appears, used and, indeed, in some of them, has never existed in modern times; ... If nothing else, this casts doubt on whether the availability of this penalty is a requirement for the maintenance of law and order in a European country.”

By concluding that the Isle of Man must be regarded as sharing fully that “common heritage of political traditions, ideals, freedom and the rule of law” to which the Preamble to the Convention refers, the Court rejected the idea that there were local requirements affecting the application of Article 3 in the Isle of Man and, accordingly, found that the applicant’s judicial corporal punishment constituted a violation of that Article.

13. Accordingly, from the very beginning of the Court’s existence, the evolutive interpretation of the Convention was closely linked to the need for a consensual reading of the text, based on consideration of the domestic legal framework of the “great majority” of the member States of the Council of Europe and, ultimately, of the common heritage of political traditions, ideals, freedom and the rule of law, to which the Preamble makes reference.

B. Deformalisation of sources of law (§§ 14 ‑ 20)

14. In Strasbourg, soft law has provided, and still provides, the most important source of crystallisation of the European consensus and the common heritage of values. In fact, soon after Tyrer , the Court took the fundamental step of enlarging the array of sources of law in the light of which the European consensus may be established. In Marckx v. Belgium [23] the Court took into consideration the European shared values based on the domestic law of the “great majority” of member States of the Council of Europe, as well as the 1962 Convention on the Establishment of Maternal Affiliation of Natural Children, prepared by the International Commission on Civil Status, and signed but not ratified by the respondent State, the Council of Europe 1975 Convention on the Legal Status of Children born out of Wedlock, not even signed by the respondent State, and finally the Committee of Ministers Resolution (70) 15 of 15 May 1970 on the social protection of unmarried mothers and their children. To the argument that the 1962 and the 1975 conventions had only a small number of parties, the Court replied (ibid., § 41):

“Both the relevant Conventions are in force and there is no reason to attribute the currently small number of Contracting States to a refusal to admit equality between ‘illegitimate’ and ‘legitimate’ children on the point under consideration. In fact, the existence of these two treaties denotes that there is a clear measure of common ground in this area amongst modern societies.”

Mirroring the interpretative techniques of constitutional courts, the Court went even further and modulated the effects of its judgment in view of the principle of legal certainty, “which is necessarily inherent in the law of the Convention as in Community Law”, dispensing the respondent State from re-opening legal acts or situations that antedated the delivery of the judgment. For that purpose, it made reference to the fact that “a similar solution [was] found in certain Contracting States having a constitutional court: their public law limit[ed] the retroactive effect of those decisions of that court that annul[led] legislation.” [24] As if it were a European Constitutional Court, the Court resorted to the principle of legal certainty to accord itself the implied power of modulation of the temporal effect of its own judgments.

15. Later on, in Mazurek v. France [25] , the Court again invoked the Convention on the Legal Status of Children born out of Wedlock, which at that time had been ratified by only a third of the member States of the Council of Europe, but not by the respondent State, as evidence of the “great importance” attached by member States to the equal legal treatment of children born out of wedlock.

16. In the cases of Christine Goodwin [26] , Vilho Eskelinen [27] , and Sørensen and Rasmussen [28] , the Court was guided by the European Union’s Charter of Fundamental Rights, even though this instrument was not yet binding. Furthermore, in McElhinney [29] , the Court took note of the European Convention on State Immunity, which at the time had been ratified by eight member States, not including the respondent State. In Glor [30] , the Court referred to the Convention on the Rights of Persons with Disabilities as the basis of “a European and worldwide consensus on the need to protect people with disabilities from discriminatory treatment”, despite the fact that the relevant facts had taken place before the adoption of the Convention by the United Nations General Assembly and the respondent State had not ratified the Convention at the time of the Court’s judgment.

17. Finally, in the landmark case of Demir and Baykara v. Turkey , after reiterating the principle that:

“the Convention is a living instrument which must be interpreted in the light of present-day conditions, and in accordance with developments in international law, so as to reflect the increasingly high standard being required in the area of the protection of human rights, thus necessitating greater firmness in assessing breaches of the fundamental values of democratic societies”

and having regard to the developments in labour law, both international and national, and to the pertinent practice of Contracting States, the Court concluded that the right to bargain collectively with the employer had, in principle, become one of the essential elements of the right to form and to join trade unions for the protection of the interests set forth in Article 11 of the Convention. For that purpose, it cited the relevant ILO conventions, which the respondent State had ratified, the corresponding interpretations of the ILO Committee of Experts, as well as Article 28 of the European Union’s Charter of Fundamental Rights, Article 6 § 2 of the European Social Charter, which Turkey had not ratified, the European Committee of Social Rights’ interpretation of this Article, and Principle 8 of Recommendation No. R (2000) 6 of the Council of Europe’s Committee of Ministers on the status of public officials in Europe [31] .

18. In other words, for the purposes of interpreting the Convention, the legal relevance of human rights standards set out in other treaties and conventions depends neither on the number of their respective ratifying parties, nor on the number of Council of Europe member States bound by them, nor even on whether the respondent State itself has ratified them. Thus, under European human rights law, hard law is profoundly interwoven with soft law.

19. Evolutive interpretation of the Convention has also led the Court to support its reasoning by reference to norms emanating from other Council of Europe organs, even though those organs have no function of representing States Parties to the Convention, whether supervisory mechanisms or expert bodies. In order to interpret the exact scope of the rights and freedoms guaranteed by the Convention, the Court has made use, for example, of the work of the European Commission against Racism and Intolerance (ECRI) [32] and the European Commission for Democracy through Law (the Venice Commission) [33] .

C. European consensus as the conceptual framework of normativity (§§ 21 ‑ 22)

20. From the seminal formulation of the European consensus in Tyrer emanates a vision of a deliberative, international democracy in which a majority or representative proportion of the Contracting Parties to the Convention is considered to speak in the name of all and is thus entitled to impose its will on other Parties. As a matter of constitutional principle guiding the Council of Europe, consensus is decoupled from unanimity. Consensus as a volonté générale can still exist even if not all Contracting Parties concur in the same reading of the Convention [34] .

As has been shown above, it cannot be argued today that the founding fathers did not want this to happen, and that States have been trapped into engagements that they did not agree upon [35] . The now worn-out argument of lack of State consent is sometimes accompanied, as the other side of the coin, by the no less démodé critique about the Court’s lack of political legitimacy to interpret innovatively the Convention, still less to create law [36] , using soft law to circumvent the competent legislative bodies and to flout the principles of democracy, rule of law and subsidiarity [37] . Underlying this speech is almost invariably the sovereignist leitmotiv in dubio pro mitius .

21. The Preamble sets the Convention against the background of the Council of Europe’s general aims, with a view to creating “greater unity” between its member States, based on “a common understanding and observance of the Human Rights upon which they depend”. In the Statute of the Council of Europe, the language used makes reference not only to a “closer unity between all like-minded countries of Europe”, but also to an “organisation which will bring European States into closer association”. The very first Article of the Statute sets as the aim of the Council “to achieve greater unity between its members for the purpose of safeguarding and realising the ideals and principles which are their common heritage and facilitating their economic and social progress”. In the explicit terms of the Statute, the realisation of these ideals and principles warrants “agreements and common action” in all relevant areas of social life (economic, social, cultural, scientific, legal and administrative matters) and “in the maintenance and further realisation of human rights and fundamental freedoms”. No better words could proclaim the primacy of human rights obligations in all areas of governance. The principle in dubio pro persona could find no better formulation. Social and economic progress is thus intimately connected to the progress of human rights, representing two sides to the same coin.

22. This being so, evolutive interpretation, European consensus and hardening of soft law compose the three pillars of the European normative system within which State consent is relevant. Based upon these pillars from the very beginning, and animated by a common quest for “economic and social progress”, the Council of Europe legal order can no longer be confused with the traditional international accord of juxtaposed egoisms. Sovereignty is no longer an absolute given, as in Westphalian times, but an integral part of a human rights-serving community [38] .

In this context, the Convention cannot but be interpreted in the light of the formally binding “agreements” (i.e. treaties) [39] and the immense plethora of formally non-binding “common actions” performed by the political and technical bodies of the Council of Europe [40] , such as recommendations, guidelines and declarations of its Committee of Ministers [41] . Furthermore, the Convention itself calls for an open-minded approach to international law and soft law, since it is inspired by the Universal Declaration on Human Rights, as the Preamble states, and remains open to other legal instruments, both domestic and international, when these offer a better human rights protection (Article 53 of the Convention). In sum, this interpretative latitude is dictated by the letter and the very nature and purpose of the Convention itself.

IV. Soft law and the Council of Europe’s rule of recognition (§§ 23 ‑ 33)

A. The rule of recognition of a democratic international community (§§ 23 ‑ 26)

23. In European human rights law, a formal theory of sources of law is still prevailing. Based on the doctrine that the International Court of Justice expounds in its North Sea Continental Shelf and Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America) judgments [42] , the Court admits that a treaty provision may become customary international law, if the following conditions are fulfilled: the provision concerned must be of a fundamentally norm-creating character such as could be regarded as forming the basis of a general rule of law; there must be corresponding settled State practice and evidence of a belief that such practice is rendered obligatory by the existence of a rule of law requiring it ( opinio juris sive necessitatis ) [43] . The Court has also expressed its position on the existence of peremptory norms in international law, such as the prohibition of torture [44] or the prohibition of genocide [45] .

Yet, as has been demonstrated, there is no water-tight, binary distinction between hard law and non-law, since European human rights law evolves by means of a rich panoply of sources that do not necessarily share the classical, formal features of hard international law [46] . The “further realisation of human rights” calls for a certain degree of deformalisation, without abandoning the formal theory of sources of law. In fact, in Strasbourg there has never been a monopoly of formalism in the ascertainment of international law.

24. However, the inherent deformalisation of European human rights law is not a synonym of nihilist dissolution of legality, and there is a very fundamental reason for this. In the Council of Europe’s legal order, State consent is framed within the context of a cosmopolitan perspective of the universality of human rights and a dialogic understanding of the common heritage of values of European societies.

25. In the Council of Europe, the recognition rule is no more a Lotus -type [47] , State-centred, narrowly bilateral, exclusively voluntaristic, top-down norm-creation mechanism, but a democratic-type, individual-centred, broadly multi-lateral, purposefully consensual, bottom-up norm creation mechanism which involves European States and other European and non-European non-State actors. Distancing itself from an outdated jus inter gentes , the Council of Europe legal order has become a truly jus gentium , based on a participated, accountable and multi-level international law-making system which is not the preserve of States [48] .

Cosmopolitanism links the Council of Europe legal order with the world, as much as the dialogue with the European domestic legal orders, and notably with their apex domestic courts, links it with the values of European society. Since the Council of Europe normative order already meets standards of democracy, its recognition rule is not linked to democratic decision-making processes on the domestic level. Soft law, which does not require formal domestic ratification, is in any case not exempt from democratic control within the Council of Europe, performed by the Committee of Ministers, the Parliamentary Assembly and, ultimately, the Court, as a politically legitimised guarantor of the Convention (Articles 19 and 22 of the Convention). Furthermore, the involvement of States and grass-roots non-State actors in the exercise of law-making powers is not only testament to the principle of subsidiarity, but indeed reinforces the democratic nature of the process and the responsiveness of the international public policy-making system towards the European people.

Put differently, the European normative order exists beyond sovereign statehood, bilateralism and opposability. Since the State is not the sole possible subject capable of creating international law, a State-will theory of sources of law gives way to a democratic international community-based normativity theory.

26. The obvious higher degree of complexity of such a recognition rule should not be misunderstood as entailing a higher degree of legal uncertainty, and therefore as meaning that the value of the rule of law diminishes or entirely vanishes. There is no necessary correlation between the former and the latter. The degree of legal certainty will rather depend on the substance of the legal discourse [49] . In a world of decentralised and deformalised international law, legal certainty is more a question of substance than one of form and procedure, and certainly not a question of the more or less transparent intent of the stakeholders. The touchstone of legal certainty is neither the form which encapsulates the norm nor the procedure by which the norm is created, but definitely the substance of the norm. This is all the more so in a constitutional order, like that of the Council of Europe, where the rule of recognition is not neutral and value-free, but substantive and value-charged.

B. The deep-seated commitment to hardening soft law (§§ 27 ‑ 30)

27. In the continuum between hard law and soft law, several factors may harden the text. Like a degradé normatif [50] , the gradual normativity of the text increases with the number of these factors that are present and decreases with their absence [51] . In this gradualist logic, it is ultimately up to the Court to decide “how much weight” to attribute to these hardening factors of soft law [52] .

28. Soft European human rights law may be hardened by certain factors that relate either to the rule-making procedure or to the rule-application procedure. These are “building bricks in a wall of normativity” [53] .

First, the prescriptive language adopted in a text or the label attached to the instrument is indicative of its normative nature. A text with a prescriptive language or label must be read, in principle, as a standard-setting text, which goes beyond a mere declaratory statement or a purely programmatic assertion.

Second, the degree of linguistic accuracy and content precision of the text is a clear indication of its normative nature. The more accurate the terminology of the text and the more precise its content, the stronger its normative claim [54] . An extensive, detailed description of what is being regulated speaks in favour of a hardened law, which leaves no room for grey areas.

Third, the existence of travaux préparatoires , explanatory reports and commentaries, with a thorough discussion of the causes and consequences of the policy choices made, increases the text’s normative density.

Fourth, the complexity of the deliberation procedure, including the voting pattern, is an additional hardening factor [55] . Widespread acceptance of the text tends to legitimise its normative claim.

Fifth, wide publicity given to the normative text seeks to secure general awareness and effective compliance from its addressees.

Sixth, the delegation of authority for interpretation and conflict resolution to an independent third body and the existence of follow-up mechanisms strengthen the compliance obligation [56] . Norm inobservance may have not only a reputational or political cost, but also other negative consequences, such as an obligation to justify or even change the infringing conduct and provide remedies [57] . Accountability mechanisms reinforce the counter-factual force of the normative text.

Seventh, and finally, subsequent practice confirming or developing the standards set out in the text reinforce the standard-setting function of the text [58] . Even before reaching the point of crystallisation of customary law, the repetition of soft law by the same or different public authorities hardens its normative claim [59] . While the mere accumulation of non-law instruments cannot per se create international law, the emergence of an opinio juris , if accompanied by other hardening factors, can transmute such instruments into international legal norms, by sliding them up the scale of international normativity and thus integrating them into the Council of Europe’s binding normative system.

29. The Court considers these hardening factors of its own motion, even when the parties have not invoked them in their pleadings, or where they have occurred after the facts complained of [60] . It is also telling that, in the eyes of the Court, they are not a factum , to be regarded as a mere sociological reality, but a source of jura , subject to the principle jura novit curia .

30. Once hardened by one or more of the above-mentioned factors, soft law produces the same legal effects as hard law, regardless of the absence of the formal and procedural requirements of the latter. Hardened soft law is on a par with binding international law. First, hardened soft law has an entitling effect, so that any State acting in conformity with hardened soft law cannot thereby be committing an internationally unlawful act, and can invoke it before a court of law or arbitrator. Second, hardened soft law is an imperative constraint, the flouting of which constitutes an internationally unlawful act. Thirdly, hardened soft law also has an abrogatory effect on other conflicting law. Fourthly, hardened soft law may not be re-softened. Once it has passed the hard-law threshold, there is no turning back. The softening of law has its limit in the constitutional force of the Convention and its protocols. I will now take this line of argument one step further.

C. The constitutional prohibition on softening hard law (§§ 31 ‑ 33)

31. There is a major caveat to the acknowledgement of the role of soft law in European human rights law. Let there be no misunderstanding: the body of European human rights as posited in the Convention and the additional protocols is hard law, and the authority for their interpretation is delegated to an independent third body, the Court. This is very hard law and it does not bow gracefully to overriding political demands, regardless of the size of the demanding majority, prevailing even over conflicting constitutional law of the member States of the Council of Europe [61] .

32. Decidedly, the constitutionalisation of the European legal order puts an absolute bar on the softening of law. Hard European human rights law may not be softened. Softening of existing hard international law would be tantamount to circumventing binding international obligations. This would evidently constitute a fraudulent evasion of international law, defeating the purpose of the Convention and the aim of the Council of Europe. When implementing the Convention and the Court’s judgments, as any other agreement and common action of the Council of Europe, member States have an enforceable obligation to “collaborate sincerely and effectively in the realisation of the aim of the Council”, as set out in Article 3 of the Statute of the Council of Europe. Should it fail to fulfil its Article 3 obligation, any member State may be suspended from its rights of representation and requested by the Committee of Ministers to withdraw under Article 7 of the Statute; and if such member does not comply with this request, the Committee may decide that it has ceased to be a member of the Council as from such date as the Committee may determine.

33. Hence, soft law must not be a vehicle of political considerations to dilute or undermine the legal force of existing hard obligations. In practice, the obligatory law and non-obligatory non-law conceptual distinction still holds much water. In European human rights law, the drawing of the line between law and non-law is crystal-clear in order to preclude the downgrading of a binding norm of hard law to soft law. The boundary line is only blurred in order to upgrade a non-binding norm of soft law to hard law.

In a few words, the relationship between hard and soft law in European human rights law is a one-way street: where there is hard law, soft law may enrich it, but it may not weaken it. Were soft law to weaken existing hard law, it would be fraudulent, a breach of the European human rights normative threshold and thus a pathological phenomenon of international normativity. Soft law is one of the ways in which European law development may occur, not a way in which it regresses.

Where there is no hard law, there is evidently no relationship between two poles, and soft law may exercise alone its normative claim, in accordance with the relevant hardening factors that it puts forward.

Second Part (§§ 34 ‑ 59)

V. The hardening of prison law (§§ 34 ‑ 42)

A. Worldwide (§§ 34 ‑ 38)

34. No international organisation or body has done as much for the development of prison law as the Council of Europe and particularly its Committee of Ministers and its Committee for the Prevention of Torture. Composed of highly qualified independent experts, the Committee is tasked with the implementation of prison law standards in any place of detention in Europe and elsewhere, under the jurisdiction of any member State of the Council of Europe [62] . The Committee of Ministers’ resolutions and recommendations, the CPT General Reports and the CPT standards [63] are remarkable sources of Council of Europe soft law in this field. Their language is undoubtedly prescriptive (“standards”, “rules”) and technically rigorous. Their content excels in precision. Normally, explanatory reports or commentaries enrich the normative content of the rules set out in the text. Preparatory work by experts, providing a multidisciplinary vision of the pertinent issues, provides a solid basis for the policy choices made by a unanimous decision of all the stakeholders. Wide dissemination of these standards fosters effective compliance by member States of the Council of Europe. These are not virtual, de lege ferenda or in fieri standards, but truly normative instruments which “provide guidance as to the approach which should be taken to interpreting” the Convention [64] .

35. The Court itself has stated repeatedly that, in spite of their strictly non-binding nature, it attaches “considerable importance” or “great weight” to these normative instruments, “which are regularly taken into account by the Court in its examination of cases concerning ill-treatment” [65] . It did so with regard to, among many others, Resolution (73) 5 of the Committee of Ministers [66] , Resolution 76(2) of the Committee of Ministers on the treatment of long-term prisoners, Recommendation No. R (87) 3 on the EPR (revised and updated by Recommendation Rec(2006)2) [67] , Recommendation No. R (98) 7 of the Committee of Ministers to member states concerning the ethical and organisational aspects of health care in prison, Recommendation Rec (99) 4 on principles concerning the legal protection of incapable adults, Recommendation 2003(23) on the management by prison administrations of life sentence and other long-term prisoners, and Resolution 2010 (2014) of the Parliamentary Assembly of the Council of Europe on child-friendly juvenile justice: from rhetoric to reality [68] .

36. In paragraphs 113 and 141 of the judgment, the majority depart from this perspective. They no longer consider that the CPT standards have “considerable importance” or “great weight”, but only affirm that the Court “remains attentive to the standards developed by the CPT”. In blunt terms, the majority downgrade the importance of the CPT’s work. This is a regrettable step backwards in the protection of prisoners and other persons in detention in Europe. But that is not all.

37. The Committee of Ministers’ normative choice must be seen against the background of the hardening of prison soft law in Europe and worldwide, a point which the majority fail to consider.

Worldwide, there has been a persistent effort of the United Nations and other international organisations and bodies to extend the scope and reinforce the accuracy of the 1957 Standard Minimum Rules for the Treatment of Prisoners, namely by approving the 1985 UN Standard Minimum Rules for the Administration of Juvenile Justice (the Beijing Rules) [69] , the 1988 UN Body of Principles for the Protection of All Persons under any Form of Detention or Imprisonment [70] , the 1990 UN Rules for the Protection of Juveniles deprived of their Liberty (the Havana Rules) [71] , the 1990 UN Guidelines for the Prevention of Juvenile Delinquency (the Riyadh Guidelines) [72] , the 2007 Istanbul statement on the use and effects of solitary confinement [73] , the 2010 UN Rules for the Treatment of Women Prisoners and Non-Custodial Sanctions for Women Offenders (the Bangkok Rules) [74] , and, most recently, the new, much more detailed version of the Standard Minimum Rules (the 2015 Mandela Rules) [75] .

On 10 December 1984 the General Assembly of the United Nations adopted the Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment. The Committee against Torture, composed of 10 independent members and entrusted with broad powers of examination and investigation, was established pursuant to Article 17 of that Convention and began to function on 1 January 1988.

38. This effort has been emulated at regional level, with the 1996 Kampala Declaration on Prison Conditions in Africa [76] , the 1999 Arusha Declaration on Good Prison Practice [77] , the 2002 Ouagadougou Declaration and Plan of Action on Accelerating Prisons and Penal Reforms in Africa [78] , the revised 2006 EPR and the 2008 Principles and Best Practices on the Protection of Persons Deprived of Liberty in the Americas [79] .

B. In the Council of Europe (§§ 39 ‑ 42)

39. The hardening of prison soft law is particularly visible in Europe. The majority do not take into account the fact that the revised 1987 EPR and explanatory report did not deal with the appropriate personal space in individual cells and multi-occupancy accommodation in prisons [80] , but the 2006 EPR revision did fill that lacuna, by providing in the Commentary to Rule 18 a clear indication of the normative standard for European prisons along the lines of previous CPT work. The purpose of this exercise, let us not forget, was to “strengthen” the rules about accommodation, as the Commentary itself states [81] .

40. Worse still, the majority do not seem to care about the fact that the CPT position was confirmed politically at the highest level within the Council of Europe, by its own ruling body, the Committee of Ministers [82] . The majority ignore the fact that the Commentary to Rule 18 of the EPR took on board, to the letter, the CPT standards on the minimum space in multi-occupancy accommodation in prisons. It is worth recalling the tenor of that Commentary [83] :

“Rule 18 includes some new elements. The first, in Rule 18.3, is intended to compel governments to declare by way of national law specific standards, which can be enforced. Such standards would have to meet wider considerations of human dignity as well as practical ones of health and hygiene. The CPT, by commenting on conditions and space available in prisons in various countries has begun to indicate some minimum standards. These are considered to be 4m2 for prisoners in shared accommodation and 6m2 for a prison cell. These minima are, related however [sic], to wider analyses of specific prison systems, including studies of how much time prisoners actually spend in their cells. These minima should not be regarded as the norm. Although the CPT has never laid down such a norm directly, indications are that it would consider 9 to 10m2 as a desirable size for a cell for one prisoner. This is an area in which the CPT could make an ongoing contribution that would build on what has already been laid down in this regard. What is required is a detailed examination of what size of cell is acceptable for the accommodation of various numbers of persons. Attention needs to be paid to the number of hours that prisoners spend locked in the cells, when determining appropriate sizes. Even for prisoners who spend a large amount of time out of their cells, there must be a clear minimum space, which meets standards of human dignity.”

41. The EPR and the corresponding Commentary have the seal of the highest political body of the Council of Europe, its Committee of Ministers, which recommended that governments of member States “be guided in their legislation, policies and practice by the rules contained in the appendix to this recommendation” and “ensure that this recommendation and the accompanying commentary to its text are translated and disseminated as widely as possible and more specifically among judicial authorities, prison staff and individual prisoners”. Such a clear and unanimous expression of political will and legal understanding that “Council of Europe member states continue to update and observe common principles regarding their prison policy” should not so easily be discarded by the majority [84] . The majority cannot at one and the same time refer to the accommodation standards of the EPR and ignore what their purpose is, according to the relevant Commentary itself. Such superficial interpretation would only pay lip-service to the EPR. In the Commentary’s own very clear words, the EPR are intended to “compel Governments to declare by way of national law specific standards, which can be enforced” and these enforceable standards include certain European “minimum standards” in terms of accommodation: first and foremost, “there must be a clear minimum space”.

42. Moreover, by ignoring the “strengthened”, “compelling”, minimum standards on accommodation set out by the EPR and the Commentary thereto, the majority are also setting aside the penological work performed by the PC-CP of the CDPC which is at the root of the EPR [85] . Without any scientific background study to contradict the PC-CP experts’ opinion, the majority simply affirm ex cathedra their own different point of view. In fact, the majority do not adduce any expert opinions or scientific analysis that would be capable of gainsaying the unanimous view of international experts expressed in soft law instruments [86] . No consideration is given to the well-established scientific correlation between prison overcrowding, lack of proper living conditions and negative psychosocial effects on prisoners, including emotional instability, aggressiveness and self-harm [87] . In umbris est potestas .

VI. The fight against prison overcrowding (§§ 43 ‑ 47)

A. The minimum living space in the ICRC standards (§§ 43 ‑ 45)

43. Prison overcrowding, as a systemic problem of European criminal justice systems, has been on the agenda of the Court since 2009 [88] . The structural nature of the problem and the consequent need to address it in general terms were first acknowledged in respect of Polish prisons [89] and subsequently also Russian remand prisons [90] , Italian prisons [91] , Belgian prisons [92] , Bulgarian prisons [93] , Hungarian prisons [94] and Lithuanian prisons [95] .

44. Resocialisation is the primary purpose of imprisonment of human beings [96] . Prison overcrowding, with its physical, psychological and social consequences, is the first obstacle to the implementation of any resocialisation programme. Adequate personal living space is a sine qua non condition for the resocialisation of prisoners [97] . This absolute minimum space requirement is not essentially different for mentally fit or unfit prisoners, remand prisoners, prisoners serving sentences for the first time or recidivists, since there is no objective reason from an Article 3 perspective to subject mentally fit persons to a different standard of protection from that applicable to mentally unfit persons, still less to distinguish between mentally fit prisoners according to the harshness of their sentence or to whether they have been remanded or finally convicted.

45. In the absence of any universal standard, the International Committee of the Red Cross (ICRC) has recommended 5.4 sq. m per person in single-cell accommodation (excluding toilet facilities) and 3.4 sq. m per person in shared or dormitory accommodation (including toilet facilities) [98] . The ICRC has calculated the space needed for sleeping on the bed as 1.6 sq. m and toilet and shower space as 1.2 sq. m. In emergency situations [99] , the ICRC initially stipulated that the floor space in cells and dormitories must never be less than 2 sq. m per person, but more recently rescinded that recommendation and only provides guidance for returning a prison to normal conditions (including minimum space specifications) “as soon as possible”.

B. The minimum living space in the EPR standards (§§ 46 ‑ 47)

46. In the European penological context, the “strengthened” 2006 EPR standards on accommodation are more generous. As the Committee of Ministers and the CPT have stated [100] , the minimum living space of each prisoner must not be less than 6 sq. m in a single-occupancy space, excluding the sanitary facility, and 4 sq. m per person in a shared space, excluding a fully-partitioned sanitary facility, with at least 2 m between the walls of the cell and at least 2.5 m between the floor and the ceiling of the cell. For multiple-occupancy cells of up to four inmates 4 sq. m should be added per additional inmate to the minimum living space of 6 sq. m. Thus, this bare minimum of personal living space in prison facilities is an absolute condition whose non-fulfilment entails per se a violation of Article 3 of the Convention.

47. Moreover, the EPR require domestic law to set specific minimum requirements in respect of the accommodation provided to prisoners, with particular regard being had to the floor space, the cubic content of air, lighting, heating and ventilation (Rules 18.1-18.3). It is therefore appropriate to establish the maximum capacity ( numerus clausus ) for each prison through the definition of space per inmate as a minimum expressed in square and cubic meters. Hence, prison capacity must not be assumed to be a slippery concept whose elasticity can be used to manipulate prison reality and make overcrowding more or less apparent. Preventive remedies should be immediately available if and when the prison capacity minima are disregarded.

VII. The application of Convention standards to the present case (§§ 48 ‑ 59)

A. The majority’s contradictory cumulative approach (§§ 48 ‑ 53)

48. Quite contradictorily, the majority circumvent the “absolute” nature of the Article 3 prohibition and the “relevant minimum standard” for multi-occupancy accommodation in prisons (see paragraphs 110 and 113 of the judgment) by considering that the assessment of the level of severity required for any treatment to fall within the scope of that Article is “relative” and dependent on “a comprehensive approach to the particular conditions of detention” (see paragraphs 122 and 123 of the judgment).

49. In fact, the majority use the “cumulative effect” approach in two very different senses: on the one hand, the cumulative effect of “compensating factors” serves to attenuate the Article 3 obligations, in order to exonerate the respondent Government of any Convention liability (see paragraphs 137 and 138 of the judgment); on the other hand, the cumulative effect of “aggravating circumstances”, such as poor material conditions and lack of out-of-cell activities, can be considered inhuman or degrading, even in the case of sufficient cell space (see paragraph 140 of the judgment).

50. I agree with the latter argument, but disagree with the former.

Where the prisoner has had at his or her disposal sufficient personal living space, other negative aspects of the material conditions of detention may lead to the finding of a violation of Article 3. Whenever the adequate size of the personal living space is coupled with inadequate conditions of sleeping, lighting, ventilation, heating, sanitation and health care, the ill-treatment of the prisoner must still unequivocally be censured.

In the case of post-trial detention of mentally unfit prisoners, as well as mentally fit prisoners sentenced to a term of five years or more, the inexistence of an individual sentence plan or, where there is one, any serious shortcomings in its implementation will be major aggravating factors. Personal living space in the prison should be viewed in the context of the applicable resocialisation regime [101] . The inexistence of health, exercise, education and work programmes, or the existence of deficient programmes, will worsen the prisoner’s situation. Being closely linked to this aspect, any breach of the rules on the separation of prisoners is also a factor to be considered.

51. But I cannot agree with the submission that the lack of sufficient personal living space can be offset by the presence of other material conditions, such as personal sleeping space, access to natural light during the day and electric lighting at night, ventilation, heating, proper hygiene conditions and adequate food. Otherwise, a cumulative effect of “compensating” factors would water down the absolute Article 3 standard, inviting the prison authorities to go down a slippery slope with no objective limits [102] .

52. This is exactly the temptation to which the majority succumb. The majority refer to a rebuttable, “strong” presumption of a violation of Article 3 when 3 sq. m of personal space in the cell is not guaranteed, with the possibility for the Government of rebutting it if they can show that the periods of deprivation of such personal space were “short, occasional and minor”. But the majority do not provide for the slightest definition of these limits. In times of economic crisis, the duty to protect the dignity of detained persons is more relevant than ever, and for that purpose to be fulfilled clear rules are needed, as the European and the UN anti-torture bodies have recently stated [103] . In addition, the majority include space occupied by furniture in the available surface area (see paragraph 114), even though this may diminish considerably the capacity to circulate freely within the cell.

53. Furthermore, the offsetting factors referred to by the majority should already be part of the normal conditions within a prison, such as “sufficient freedom of movement outside the cell and adequate out-of-cell activities”, and even very broadly speaking the existence of “an appropriate detention facility”. There is a serious logical flaw in this reasoning. Here the majority’s criteria can hardly withstand Ockham’s razor. Pluralitas non est ponenda sine necessitate .

In an absolutely redundant way, the majority make use of what should be ordinary features of a prison facility in order to justify an extraordinarily low level of personal space for individuals in detention. For the majority, normal living conditions justify abnormal space conditions. Logic would require that extraordinary negative circumstances be offset only by extraordinary positive counter-circumstances. This is not the case in the majority’s logic. No extraordinary positive features of prison life are required by the majority to compensate for the deprivation of each prisoner’s right to adequate accommodation in detention.

B. A coherent, pro persona cumulative approach (§§ 54 ‑ 59)

54. The applicant spent 240 days in detention in the present case. According to the floor plans of Bjelovar Prison, which the Government provided to the Court and which are not disputed by the applicant, he was allocated 4 sq. m or more of personal space in the cells for a non-consecutive period of 70 days in total.

Cell no.

Period of

Detention

Total number of inmates

Overall surface area in

sq. m

Personal space in sq. m

Surface

minus

sanitary

facility

in sq. m

Personal space in

sq. m

8/O

03.05-05.05.2010

5

22.88

4.58

20.98

4.19

8/O

08.05-09.05.2010

5

22.88

4.58

20.98

4.19

8/O

26.05.2010

5

22.88

4.58

20.98

4.19

8/O

03.06-04.06.2010

5

22.88

4.58

20.98

4.19

8/O

17.06-19.06.2010

5

22.88

4.58

20.98

4.19

8/O

27.08-30.08.2010

5

22.88

4.58

20.98

4.19

8/O

07.09.2010

4

22.88

5.72

20.98

5.24

8/O

08.09-16.09.2010

5

22.88

4.58

20.98

4.19

8/O

18.09.2010

5

22.88

4.58

20.98

4.19

8/O

02.10-05.10.2010

5

22.88

4.58

20.98

4.19

8/I

06.10-07.10.2010

5

22.18

4.44

20.28

4.05

8/I

08.10-19.10.2010

4

22.18

5.55

20.28

5.07

8/I

20.10-21.10.2010

3

22.18

7.39

20.28

6.76

8/I

22.10-23.10.2010

4

22.18

5.55

20.28

5.07

8/I

24.10-25.10.2010

5

22.18

4.44

20.28

4.05

8/I

29.10-30.10.2010

5

22.18

4.44

20.28

4.05

4/O

06.11-09.11.2010

5

22.36

4.47

20.46

4.09

8/O

01.03-15.03.2011

5

22.88

4.58

20.98

4.19

55. Again according to the same data, the applicant spent 170 non-consecutive days with less than 4 sq. m of personal space in the cells.

Cell no.

Period of

Detention

Total number of inmates

Overall surface area in

sq. m

Personal space in sq. m

Surface

minus

sanitary

facility

in sq. m

Personal space in

sq. m

1/O

16.10-15.11.2009

6

19.7

3.28

17.8

2.96

1/O

16.11-19.11.2009

5

19.7

3.94

17.8

3.56

1/O

20.11.2009-05.02.2010

6

19.7

3.28

17.8

2.96

1/O

06.02-08.02.2010

5

19.7

3.94

17.8

3.56

1/O

09.02-10.04.2010

6

19.7

3.28

17.8

2.96

1/O

11.04.-20.04.2010

5

19.7

3.94

17.8

3.56

8/O

21.04.2010

8

22.88

2.86

20.98

2.62

8/O

22.04-29.04.2010

7

22.88

3.27

20.98

2.99

8/O

30.04-02.05.2010

6

22.88

3.81

20.98

3.49

8/O

06.05-07.05.2010

6

22.88

3.81

20.98

3.49

8/O

10.05.-25.05.2010

6

22.88

3.81

20.98

3.49

8/O

27.05-02.06.2010

6

22.88

3.81

20.98

3.49

8/O

05.06-16.06.2010

6

22.88

3.81

20.98

3.49

8/O

20.06-30.06.2010

6

22.88

3.81

20.98

3.49

8/O

01.07-02.07.2010

7

22.88

3.27

20.98

2.99

8/O

03.07-05.07.2010

8

22.88

2.86

20.98

2.62

8/O

06.07-17.07.2010

7

22.88

3.27

20.98

2.99

8/O

18.07-13.08.2010

8

22.88

2.86

20.98

2.62

8/O

18.08-26.08.2010

7

22.88

3.27

20.98

2.99

4/O

31.08-02.09.2010

8

22.36

2.80

20.46

2.55

4/O

03.09.2010

7

22.36

3.19

20.46

2.92

8/O

04.09-06.09.2010

6

22.88

3.81

20.98

3.49

8/O

17.09.2010

6

22.88

3.81

20.98

3.49

8/O

19.09-01.10.2010

6

22.88

3.81

20.98

3.49

8/I

26.10-28.10.2010

6

22.18

3.70

20.28

3.38

8/I

31.10-04.11.2010

6

22.18

3.70

20.28

3.38

4/O

05.11.2010

6

22.36

3.73

20.46

3.41

4/O

10.11-13.11.2010

6

22.36

3.73

20.46

3.41

4/O

14.11-18.11.2010

7

22.36

3.19

20.46

2.92

4/O

19.11-26.11.2010

8

22.36

2.80

20.46

2.55

4/O

27.11-30.11.2010

7

22.36

3.19

20.46

2.92

8/O

01.12-03.12.2010

6

22.88

3.81

20.98

3.49

8/O

04.12-09.12.2010

7

22.88

3.27

20.98

2.99

8/O

10.12-12.12.2010

8

22.88

2.86

20.98

2.62

8/O

13.12-21.12.2010

7

22.88

3.27

20.98

2.99

8/O

22.12-24.12.2010

8

22.88

2.86

20.98

2.62

8/O

25.12-31.12.2010

7

22.88

3.27

20.98

2.99

8/O

01.01-16.01.2011

6

22.88

3.81

20.98

3.49

8/O

17.01-25.01.2011

7

22.88

3.27

20.98

2.99

8/O

26.01-27.01.2011

6

22.88

3.81

20.98

3.49

8/O

28.01-23.02.2011

7

22.88

3.27

20.98

2.99

8/O

24.02-25.02.2011

8

22.88

2.86

20.98

2.62

8/O

26.02-28.02.2011

7

22.88

3.27

20.98

2.99

8/O

16.03.2011

6

22.88

3.81

20.98

3.49

56. Of the total amount of 170 days of detention with under 4 sq. m, the applicant spent fifty days with less than 3 sq. m.

Cell no.

Period of

detention

Total number of inmates

Overall surface area in

sq. m

Personal space in sq. m

Surface

minus

sanitary

facility

in sq. m

Personal space in

sq. m

1/O

16.10-15.11.2009

6

19.7

3.28

17.8

2.96

1/O

20.11.2009-05.02.2010

6

19.7

3.28

17.8

2.96

1/O

09.02-10.04.2010

6

19.7

3.28

17.8

2.96

8/O

21.04.2010

8

22.88

2.86

20.98

2.62

8/O

22.04-29.04.2010

7

22.88

3.27

20.98

2.99

8/O

01.07-02.07.2010

7

22.88

3.27

20.98

2.99

8/O

03.07-05.07.2010

8

22.88

2.86

20.98

2.62

8/O

06.07-17.07.2010

7

22.88

3.27

20.98

2.99

8/O

18.07-13.08.2010

8

22.88

2.86

20.98

2.62

8/O

18.08-26.08.2010

7

22.88

3.27

20.98

2.99

4/O

31.08-02.09.2010

8

22.36

2.80

20.46

2.55

4/O

03.09.2010

7

22.36

3.19

20.46

2.92

4/O

14.11-18.11.2010

7

22.36

3.19

20.46

2.92

4/O

19.11-26.11.2010

8

22.36

2.80

20.46

2.55

4/O

27.11-30.11.2010

7

22.36

3.19

20.46

2.92

8/O

04.12-09.12.2010

7

22.88

3.27

20.98

2.99

8/O

10.12-12.12.2010

8

22.88

2.86

20.98

2.62

8/O

13.12-21.12.2010

7

22.88

3.27

20.98

2.99

8/O

22.12-24.12.2010

8

22.88

2.86

20.98

2.62

8/O

25.12-31.12.2010

7

22.88

3.27

20.98

2.99

8/O

17.01-25.01.2011

7

22.88

3.27

20.98

2.99

8/O

28.01-23.02.2011

7

22.88

3.27

20.98

2.99

8/O

24.02-25.02.2011

8

22.88

2.86

20.98

2.62

8/O

26.02-28.02.2011

7

22.88

3.27

20.98

2.99

57. In other words, a simple statistical assessment of the available data shows that the applicant spent more than two thirds of his detention in overcrowded cells, according to the standard of the Committee of Ministers, the CPT and PC-CP of the CDPC. The data above show that the applicant spent 29.1% of his detention time with 4 sq. m or more of personal space in the cells and 70.9% with less than that. Of the total amount of days in detention, 20.8% were spent with less than 3 sq. m of personal space in the cells. It is beyond my understanding how the majority can argue that 20.8%, i.e., one fifth of the applicant’s detention days, represents a “short, occasional and minor” reduction of the required minimum personal space. With such a statistically odd evaluation, the majority totally dilute the common meaning of these adjectives. This is particularly grave in view of the additional flagrant breach of national law, which was aligned with the Council of Europe’s 4 sq. m standard. In fact, the same white-washing exercise had already been performed by the Constitutional Court, which used some Strasbourg case-law to bluntly disregard national law.

58. Finally, the majority have not considered two main complaints that the applicant constantly presented to the national authorities. First, he wanted to work and no work possibilities were provided in the prison facilities where he was detained. The majority confuse the purposeful, productive occupation of the prisoner’s time with watching TV most of the day and occasionally playing basketball or ping-pong for a couple of hours. Lack of productive occupation of those willing to work is no less degrading than unpaid or poorly paid work.

59. Second, he was not given any real possibility of meeting his relatives or of keeping in contact with his family. Bearing in mind that his family could not afford to travel to the prison, which is not disputed by the Government, arrangements should have been made by the prison authorities to provide the applicant with more telephone time than a mere twenty minutes per week, with an additional ten minutes on public holidays. Moreover, it is also unclear from the Government’s arguments why the applicant’s request for transfer to a prison facility closer to his family’s place of residence was first ignored and later repeatedly rejected.

Set against the backdrop of the overcrowded cells where he had to live, these shortcomings only aggravated the violation of Article 3 of the Convention.

VIII. Conclusion (§§ 60 ‑ 63)

60. The Convention and its protocols have full and undiluted legal force. Yet in European human rights law there is no straightforward “either/or” answer to the problem of the normative threshold. Once removed from the reserved domain of the States, soft law may slide up the scale of international normativity, depending on the presence of certain hardening factors. In the light of an evolutive interpretation of the Convention and the other “agreements” and “common action” instruments of the Council of Europe, the normativity threshold will be placed where the societal needs for “further realisation of human rights” and “economic and social progress” lie.

61. Hardened soft law is a source of law with “considerable importance” or “great weight” in European human rights law. The EPR and the Commentary thereto are the prototype of Council of Europe soft law. The majority’s decision seems to me to be out of step with the longstanding reform efforts made by the Council of Europe and in the rest of the world in the field of prison law. The EPR are intended to “compel Governments to declare by way of national law specific standards, which can be enforced”, and these enforceable standards include certain European “ minimum standards”: “there must be a clear minimum space”. The circular nature of the majority’s redundant argument is obvious: when the lack of personal space is justified by an impressionistic image of the general situation of the detention facility, the offsetting exercise becomes a question-begging masquerade to cover up a downgrading of the general level of human rights protection of prisoners.

62. The present judgment pays no heed to what the representatives of the forty-seven democratically elected European governments unanimously agreed upon in the Committee of Ministers of the Council of Europe, or indeed to the most laudable work of renowned legal, medical, psychological, sociological and penological experts such as those in the CPT and the Council for Penological Cooperation of the Council of Europe.

63. By failing to pay due attention to the Council of Europe’s own sources of law, and ignoring the hardening of the soft law concerning prison standards in Europe and worldwide, the majority set a standard that will lead to a strictly casuistic, fact-sensitive application of the Convention, leaving the door wide open to a slippery-slope regression of the human rights protection level already attained by the Council of Europe itself. With judgments of this kind, weakening as they do the Council of Europe’s human rights protection system from within, the Court not only discourages the work of other Council of Europe bodies, but, worse still, reinforces the impression of an incoherent European human rights protection system.

[1] The experience of federal courts in the United States in ordering injunctive relief, which started in the sixties and gradually developed to become an integral part of legal guarantees against inhuman treatment, is well described in the legal literature, see Michele Deitch, “The Need for Independent Prison Oversight in a Post-PLRA World”, Federal Sentencing Reporter , Vol. 24, No. 4, pp. 236-44.

[2] Elizabeth Alexander, “Watching the Watchmen after Termination of Injunctive Relief”, Pace Law Review , Vol. 24, Issue 2, Spring 2004.

[3] While federal legislation was enacted in 1996 to limit the involvement of the courts in ordering injunctive relief with respect to prison establishments, the approach as such was not challenged. Studies have noted that following those legislative amendments injunctive orders vis à vis prison establishments have become less broad and more focused, see Margo Schlanger, “Civil Rights Injunctions over Time: A Case Study of Jail and Prison Court Orders”, N.Y.U.L. Rev. 81, no. 2 (2006), pp. 550-630.

[4] I used the expression “soft law” in my opinion appended to the judgment in the case of Herrmann v. Germany [GC], no. 9300/07, 26 June 2012. I did not define the word then, but I find it necessary to do so now. This is the purpose of paragraphs 3-9 of this opinion.

[5] Prosper Weil, “Towards relative normativity in international law?”, in American Journal of International Law , volume 77 (1983), p. 441.

[6] For a caustic critique, see Klabbers, “The Undesirability of Soft Law”, in Nordic Journal of International Law 67 (1998), p. 391; and Koskenniemi, “Formalism, Fragmentation, Freedom: Kantian Themes in Today’s International Law”, (2007) 4 No Foundations , p. 18.

[7] See, among other critical writings, Jean d’Aspremont, Formalism and the Sources of International Law: A Theory of the Ascertainment of Legal Rules , Oxford, 2011, and “The Politics of Deformalization in International Law” , Goettingen Journal of International Law 3 (2011) 2, 503-550.

[8] The argument was admitted by the ICJ itself in Nuclear Tests (Australia v. France), Judgment, ICJ Reports 1974, § 46.

[9] On this change, see my opinion appended to the judgment in Al-Dulimi and Montana Management Inc. v. Switzerland [GC], no. 5809/08, 21 June 2016.

[10] I am not referring here to the category of cases in which a treaty rule makes reference to non-legal norms, respect for which is made obligatory by such reference, like Article 18 (1) (b) and (c) of the International Convention for the Suppression of Acts of Nuclear Terrorism. In this category of cases, incorporation of the non-legal standards into hard law changes the nature of those standards. These are, therefore, improper soft law instruments. In the text, I mean only proper soft law instruments, which are complementary to hard law, but have not been formally absorbed by it. This is the case, for instance, when hard and soft law instruments have an overlapping subject matter and the preamble to a treaty refers to a soft law instrument.

[11] See Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territory, Advisory Opinion, ICJ Reports 2004 , p. 171, § 86, which refers to “relevant resolutions adopted pursuant to the Charter by the General Assembly and the Security Council” as sources of rules and principles of international law which are relevant in assessing the legality of measures taken by Israel. These rules and principles can be found in the United Nations Charter, other treaties, customary international law and the resolutions are interpretative of these instruments. See even earlier, Gabcikovo-Nagymaros Project (Hungary/Slovakia), Judgment, ICJ Reports 1997 , pp. 77-78, § 140, which refers to “new norms and standards” that have been developed in the field of environmental law, set forth in a great number of instruments over the last two decades, stating that “[s]uch new norms have to be taken into consideration, and such new standards given proper weight, not only when States contemplate new activities but also when continuing with activities begun in the past.”

[12] The choice between hard and soft law is more often than not framed in terms of obligations codified in treaties, with no mention of obligations condensed into non-negotiated, unwritten, universal customary law. The claim has been made that soft law has rendered custom obsolete. The counter-claim that a certain acceleration of the custom-formation process, in the light of the ICJ’s case-law, could make soft law unnecessary has also been raised. Neither of those claims is borne out by the Court’s case-law, as will be explained below.

[13] One of the most telling examples is the United Nations International Convention on the Elimination of All Forms of Racial Discrimination, adopted on 21 December 1965, which drew on the United Nation Declaration on the Elimination of All Forms of Racial Discrimination, adopted by the General Assembly Resolution 1904 (XVIII), of 20 November 1963.

[14] See Legality of the Threat or Use of Nuclear Weapons, Advisory Opinion, ICJ Reports 1996 , p. 255, § 70, and Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America). Merits, Judgment. ICJ Reports 1986, p. 100, §§ 188 and 191.

[15] With the laudable exceptions of Luis Lopez Guerra, “Soft law y sus efectos en el ámbito del derecho europeo de los derechos humanos”, in Teoría y derechos , vol. 11 (2012), p. 150-67; Tulkens et al , “Le soft law et la Cour Europeénne des droits de l’homme : Questions de légitimité et de méthode”, in Revue Trimestrielle des Droits de l’Homme , 23 (2012), no. 91, pp. 433-89, and Tulkens and van Drooghenbroeck, “Le soft law des droits de l’homme est-il vraiment si soft? Les développements de la pratique interprétative récente de la Cour européenne des droits de l’homme”, in Liber amicorum Michel Mahieu , Brussels, 2008, pp. 505-26. Other more general studies, like the New York University School of Law project on Global Administrative Law, the Heidelberg project on International Institutional Law and the exercise of international public authority and the project on Informal International Lawmaking promoted by the Hague Institute for the Internationalisation of Law, did not grapple specifically with this topic.

[16] Golder v. the United Kingdom , 21 February 1975, § 29, Series A no. 18.

[17] The point was already made in my opinions in Al-Dulimi and Montana Management Inc. , cited above, § 71, Sargsyan v. Azerbaijan [GC], no. 40167/06, ECHR 2015, footnote 23, and Centre for Legal Resources on behalf of Valentin Câmpeanu v. Romania [GC], no. 47848/08, ECHR 2014, footnote 14.

[18] Rozakis, “The European Judge as Comparatist”, Tulane Law Review , 2005, p. 278.

[19] Tyrer v. the United Kingdom , 25 April 1978, § 31, Series A no. 26.

[20] Ibid., § 31, and later on repeated in many other leading cases, such as for example in Soering v. the United Kingdom , 7 July 1989, § 102, Series A no. 161.

[21] Missouri v. Holland 252 U.S. 416 (1920). Writing the majority’s opinion, Justice Holmes made this remark on the nature of the constitution: “With regard to that we may add that when we are dealing with words that also are a constituent act, like the Constitution of the United States, we must realize that they have called into life a being the development of which could not have been foreseen completely by the most gifted of its begetters. It was enough for them to realize or to hope that they had created an organism; it has taken a century and has cost their successors much sweat and blood to prove that they created a nation. The case before us must be considered in the light of our whole experience and not merely in that of what was said a hundred years ago.” The Supreme Court’s reference to “evolving standards of decency” is also understood as a clear mention of the “living constitutionalism” (see Trop v. Dulles , 356 U.S. 86 (1958): “The words of the [Eighth] Amendment are not precise, and ... their scope is not static. The Amendment must draw its meaning from the evolving standards of decency that mark the progress of a maturing society”).

[22] Henrietta Muir Edwards and others v. The Attorney General of Canada [1929] UKPC 86, [1930] A.C. 124 (18 October 1929). The case is not only memorable because it established that Canadian women were eligible to be appointed senators, but also because it introduced the “living tree doctrine” in Canadian constitutional law, according to which the constitution is organic and must be read in a broad and liberal manner so as to adapt it to changing times.

[23] Marckx v. Belgium , 13 June 1979, Series A no. 31.

[24] Ibid., § 58.

[25] Mazurek v. France , no. 34406/97, § 49, ECHR 2000-II.

[26] Christine Goodwin v. the United Kingdom [GC], no. 28957/95, ECHR 2002 ‑ VI.

[27] Vilho Eskelinen and Others v. Finland [GC], no. 63235/00, ECHR 2007 ‑ II.

[28] Sørensen and Rasmussen v. Denmark [GC], nos. 52562/99 and 52620/99, ECHR 2006-I.

[29] McElhinney v. Ireland [GC], no. 31253/96, ECHR 2001-XI.

[30] Glor v. Switzerland , no. 13444/04, § 53, ECHR 2009.

[31] Demir and Baykara v. Turkey [GC], no. 34503/97, §§ 146-54, ECHR 2008-VI. In paragraph 78, the Court stated, in a remarkably clear way, as follows: “The Court observes in this connection that in searching for common ground among the norms of international law it has never distinguished between sources of law according to whether or not they have been signed or ratified by the respondent State”. Another feature of this judgment worth mentioning is that the Court sometimes referred to the situation of domestic law in the “great majority” of States (§§ 76, 79) and other times in the “majority” of States (§§ 82, 85, 106, 165). In paragraph 151, it refers to both! This clearly shows that the Court did not accord much relevance to the quantitative weight of the majority relevant for the purposes of building a European consensus, admitting that such majority may well be a narrow one, when articulated with other sources of international law.

[32] The first case where the Court cited an ECRI General Policy Recommendation was Beard v. the United Kingdom [GC], no. 24882/94, § 70, 18 January 2001. The source quoted was Recommendation No. 3: Combating Racism and Intolerance against Roma/Gypsies.

[33] The first case where the Court cited the Venice Commission was Hirst v. the United Kingdom (No. 2) , no. 74025/01, § 24, 30 March 2004. The source quoted was the Code of Good Practice in Electoral Matters, adopted by the Venice Commission at its 51st Plenary Session (5-6 July 2002).

[34] It is highly symbolic that the Court resorts to this historically and philosophically much charged expression ( volonté générale ) in the French version of Demir and Baykara , cited above, § 84.

[35] See above, footnote 13.

[36] See on this critique my opinion appended to Centre for Legal Resources on behalf of Valentin Câmpeanu , cited above.

[37] It is interesting to note that a similar critique was addressed by the European Parliament to the use of soft law by the Commission, which has been accused of acting ultra vires and extending the competence of the Union beyond the principle of compétence d’attribution (European Parliament Resolution of 4 September 2007 on institutional and legal implications of the use of “soft law” instruments (2007/2028(INI)).

[38] On this point see my opinion appended to Sargsyan , cited above.

[39] There are two types of Council of Europe treaties: harmonisation treaties which seek to achieve harmonisation of national legislation and cooperation treaties which aim at facilitating and improving international cooperation between national law enforcement agencies (see Bartsch, “The Implementation of Treaties Concluded within the Council of Europe”, in Jacobs and Roberts (eds), The Effect of Treaties in Domestic Law , 1987, p. 197; “The specificity and added value of the acquis of the Council of Europe treaty law”, Working document prepared by Mr Jeremy McBride, AS/Jur (2009) 40, 17 September 2009.

[40] As the Court itself put it in Soltysyak v. Russia , no. 4663/05, § 51, 10 February 2011: “The Court reiterates its constant approach that it takes into account relevant international instruments and reports, and in particular those of other Council of Europe organs, in order to interpret the guarantees of the Convention and to establish whether there is a common European standard in the field.”

[41] On Council of Europe law, see Kleijssen, “Council of Europe standard-setting in the human rights field”, in NJCM-Bulletin : jaarg. 35, nr. 7 (nov.-dec. 2010), pp. 897-904; Benoit-Rohmer and Klebes, Council of Europe Law - Towards a pan-European legal area , Strasbourg: Council of Europe Press, 2005; Polakiewicz, Treaty-making in the Council of Europe , Strasbourg, Council of Europe Press, 1999. As put by Kleijssen, soft law instruments in the Council of Europe “are usually adopted by all member states and thus represent a common European position which refers to legally-binding standards (such as the case-law of the Court).” This means that “the relationship between the Court’s case-law and other Council of Europe standards is not circular, but it could be rather described as a spiral, or even as a symbiosis.”

[42] North Sea Continental Shelf, Judgment ( ICJ Reports 1969, pp. 41-44, §§ 71-78) and Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America), Merits, Judgment, ICJ Reports 1986, p. 98, § 186.

[43] See on the Court’s formal theory of sources of law, Vasiliauskas v. Lithuania [GC], no. 35343/05, §§ 165-75, ECHR 2015; Perinçek v. Switzerland [GC], no. 27510/08, §§ 266-68, ECHR 2015; Jones and Others v. the United Kingdom , nos. 34356/06 and 40528/06, §§ 88-94 and 202-215, ECHR 2014; Hirsi Jamaa and Others v. Italy [GC], no. 27765/09, § 75, ECHR 2012; Van Anraat v. the Netherlands (dec.), no. 65389/09, §§ 90-92, 6 July 2010; Kononov v. Latvia [GC], no. 36376/04, §§ 203, 211, 215 and 221, ECHR 2010, and the dissenting opinion in that case of Judge Costa joined by Judges Kalaydjieva and Poalelungi; Medvedyev and Others v. France [GC], no. 3394/03, §§ 65, 85 and 92, ECHR 2010; Cudak v. Lithuania [GC] , no. 15869/02, § 66, ECHR 2010; Stoll v. Switzerland [GC], no. 69698/01, § 59, ECHR 2007-V; Al‑Adsani v. the United Kingdom [GC], no. 35763/97, §§ 61-66, ECHR 2001-XI; and Banković and Others v. Belgium and Others (dec.) [GC], no. 52207/99, § 66, ECHR 2001-XII. In Jones and Others , cited above, § 198, the Court considered the ICJ’s case-law to be “authoritative as regards the content of customary international law”. It has been argued that a certain acceleration of the custom-formation process, in the light of the ICJ’s doctrine, could render soft law unnecessary, but the Court’s case-law does not confirm this point of view.

[44] Al-Adsani , cited above, § 61.

[45] Jorgić v. Germany , no. 74613/01, § 68, ECHR 2007‑III.

[46] The typology of these forms is immensely rich. They include non-conventional international agreements, like the Helsinki Final Act (see for example, Freedom and Democracy Party (ÖZDEP) v. Turkey , no. 23885/94, § 40, 8 December 1999); treaties not ratified by the respondent State (see for example, Marckx , cited above); declarations of international organisations, like the Universal Declaration on Human Rights and other General Assembly Declarations (see for example, K.-H. W. v. Germany [GC], no. 37201/97, § 95, ECHR 2001-II); resolutions and recommendations of international organisations, like those of the Parliamentary Assembly and the Committee of Ministers of the Council of Europe (see for example, Mosley v. United Kingdom , no. 48009/08, ,§§ 87, 119 and 124, 10 May 2011); General Comments of international organisations, like those adopted by the United Nations treaties bodies (see for example, Bayatyan v. Armenia [GC], no. 23459/03, § 105, ECHR 2007); and Codes of Conduct and Guidelines of international organisations, like those of the World Health Organisation (see for example, Oluic v. Croatia , no. 61260/08, § 60, 20 May 2010); commentaries and studies by ONGs, like the International Committee of the Red Cross study on customary international humanitarian law and commentaries on the Geneva conventions (see for example, Korbely v. Hungary , no. 9174/02, §§ 50, 51 and 90, ECHR 2008), and reports of individuals, like those of the United Nations Secretary-General ( Korbely , cited above, § 90). This latter case is also remarkable due to the high relevance given to scholarly opinion in paragraphs 82 and 87.

[47] “Lotus”, Judgment No. 9, 1927, PCIJ, Series A, No. 10 , at p. 18: “The rules of law binding upon States ... emanate from their own free will ...”

[48] This is not a new claim (see on the role of other subjects in international law, Reparation for Injuries Suffered in the Service of the United Nations, Advisory Opinion: ICJ Reports 1949, p. 178, and also Lauterpacht, “The Subjects of International law”, in Lauterpacht (ed.), International Law, The Collected Papers of Hersch Lauterpacht , volume I: The General Works , Cambridge, CUP, 1970, § 48).

[49] On this point see my opinion appended to Centre for Legal Resources on behalf of Valentin Câmpeanu , cited above, and its main message: legal discourse, and particularly judicial discourse, is not an instrument of interest-oriented Realpolitik . The present opinion on law sources must be read in the light of my reflections in the Câmpeanu case on legal argumentation.

[50] Pellet, “Le ‘bon droit’ et l’ivraie –plaidoyer pour l’ivraie”, in Mélanges offerts à Charles Chaumont, Le droit des peuples à disposer d’eux-mêmes. Méthodes d’analyse du droit international (1984), p. 488.

[51] The idea of a graduated normativity in international law has been acknowledged both in general international law (see the provisions on jus cogens of the Vienna Convention on the Law of Treaties, Articles 53 and 64), and in the Convention itself, Article 15.

[52] Tanase v. Moldova [GC], no. 7/08, § 176, ECHR 2010.

[53] The expression comes from Klabbers, “Reflections on Soft International Law in a Privatized World”, in Finnish Yearbook of International Law , volume XVI (2005), p. 322.

[54] The same could be said of some hard law provisions, like treaty provisions of a programmatic nature related to the obligation to undertake to take steps “to the maximum of its available resources” towards full realization of economic and social rights (see my separate opinion in Konstantin Markin v. Russia [GC], no. 30078/06, ECHR 2012), or the obligation to co-operate in good faith or to consult together ( Interpretation of the Agreement of 25 March 1951 between the WHO and Egypt, Advisory Opinion, ICJ Reports 1980 , p. 95, § 48).

[55] Aegean Sea Continental Shelf, Judgment. I CJ Reports 1978 , p. 39, § 96, and Maritime Delimitation and Territorial Questions between Qatar and Bahrain, Jurisdiction and Admissibility, Judgment, ICJ Reports 1994 , p. 121, § 23 : in order to ascertain if an international agreement has been concluded, “the Court must have regard above all to its actual terms and to the particular circumstances in which it was drawn up”.

[56] Kleijssen, cited above, p. 899: “such instruments may provide for a ‘follow-up’ mechanism in which the Committee of Ministers may ask member states to inform it about measures taken to implement that recommendation. This is a measure that has been used in all the most recent Council of Europe recommendations in the human rights field, in order to ensure that such instruments become a concrete source of reference for action in the member states”. In fact, he continues: “The first point to make is that the Council of Europe spends already far more resources on human rights monitoring than on human rights standard-setting, which reflects the high priority attached to monitoring and implementation of standards.” On the impact of these monitoring mechanisms, including the CPT, on member States, see “Practical impact of the Council of Europe monitoring mechanisms in improving respect for human rights and the rule of law in member states”, Directorate General of Human Rights and Legal Affairs of the Council of Europe, 2014, pp. 16-56, and especially pp. 32-35; and Benoît-Rohmer, “Mécanismes de supervision des engagements des États membres et autorité du Conseil de l'Europe”, in Haller et al , Law in Greater Europe , 2000; and de Vel and Markert, “Importance and weaknesses of the Council of Europe Conventions and of the Recommendations addressed by the Committee of Ministers to member states”, in Haller et al , cited above. Monitoring mechanisms that are not established by a treaty rely at least to some extent on treaty-based standards, like those of the Convention, or the more general commitment in Article 3 of the Statute of the Council of Europe (McBride, cited above, § 54).

[57] Such obligations have been acknowledged even for purely political commitments. In its Resolution on International Texts of Legal Import in the Mutual Relations of their Authors and Texts Devoid of Such Import, 29 August 1983, the Institute of International Law concluded as follows: “The violation of purely political commitments justifies the aggrieved party in resorting to all means within its power in order to put an end to, or compensate for, its harmful consequences or drawbacks, in so far as such means are not prohibited by international law.” A fortiori , remedies may be requested for harm caused by violation of soft law.

[58] This is not to say that law only exists to the extent with which it is complied. Such impact-based approach to law-ascertainment is too strict, since it puts the cart before the horse.

[59] See Demir and Baykara , cited above, §§ 48-52, and Bayatyan v. Armenia , cited above, §§ 46-49, and contrast Stummer v. Austria [GC], no. 37452/02, §§ 105-106, ECHR 2011.

[60] Both Judge Zagrebelsky in his opinion attached to the Demir and Baykara judgment, cited above, and Judge Gyulumyan in her opinion in Bayatyan , cited above, underscored the fact that the soft law instruments used by the majority were produced after the facts complained of.

[61] See my joint separate opinion, with Judge Dedov, appended to Baka v. Hungary [GC], no. 20261/12, 23 June 2016, and my separate opinion appended to Fabris v. France [GC], no. 16574/08, ECHR 2013.

[62] The CPT was set up under the Council of Europe’s European Convention for the Prevention of Torture and Inhuman or Degrading Treatment or Punishment (the CPT Convention), which came into force in 1989. The CPT Convention has been ratified by all the 47 member States of the Council of Europe.

[63] The CPT draws up a general Report on its activities, which is published once a year. In a number of its General Reports the CPT has described some of the substantive issues which it pursues when carrying out visits to places of deprivation of liberty. In this way, the CPT gives a clear advance indication to national authorities of its views regarding the manner in which persons deprived of their liberty ought to be treated. The sections drawn up to date deal with police custody, imprisonment, training of law enforcement personnel, health care services in prisons, foreign nationals detained under aliens legislation, involuntary placement in psychiatric establishments and juveniles and women deprived of their liberty. They have been brought together in a document called the “CPT standards”.

[64] See, mutatis mutandis , Manole and Others v. Moldova , no. 13936/02, §§ 102, 107, ECHR 2009.

[65] See Meier v. Switzerland , no. 10109/14 § 78, 9 February 2016: “ importance considérable” ; Harakchiev and Tolumov v. Bulgaria , nos. 15018/11 and 61199/12, §§ 204 and 264, ECHR 2014 (extracts): “considerable importance”, which also refers to the CPT’s eleventh general report and the twenty-first general report; Sławomir Musiał v. Poland , no. 28300/06, § 96, 20 January 2009: “importance of this recommendation”, Dybeku v. Albania , no. 41153/06, § 48, 18 December 2007: “importance of this recommendation”; Rivière v. France , no. 33834/03, § 72, 11 July 2006: “ grand poids” . In Shtukaturov v. Russia , no. 44009/05, § 95, 27 March 2008: “Although these principles have no force of law for this Court, they may define a common European standard in this area”. In Volkan Özdemir v. Turkey , no. 29105/03, § 39, 20 October 2009: “regularly taken into account by the Court.”

[66] In the seminal case of S. v. Switzerland , no. 12629/87, § 48, 28 November 1991, the Court decided that an accused’s right to communicate with his lawyer out of hearing of a third person was part of the basic requirements of a fair trial in a democratic society and followed from Article 6 § 3 (c) of the Convention, on the basis of Article 93 of the Standard Minimum Rules for the Treatment of Prisoners, annexed to Resolution (73) 5 of the Committee of Ministers, which set forth that right.

[67] See Enea v. Italy [GC], no. 74912/01, § 101, ECHR 2009: “Although this Recommendation is not legally binding on the member States, the great majority of them recognise that prisoners enjoy most of the rights to which it refers and provide for avenues of appeal against measures restricting those rights.”

[68] In addition, the Court has used the CPT national reports as reliable, impartial sources of factual information on the situation of prisons and centres of detention, since Amuur v. France (19776/92, § 28, 25 June 1996) and Aerts v. Belgium (30 July 1998, § 42, Reports of Judgments and Decisions 1998‑V). Although the CPT is not an investigative body, its reports are important elements in describing the situation on the ground in European States.

[69] Adopted by General Assembly Resolution 40/33 of 29 November 1985. The Court first cited them in V. v. the United Kingdom , no. 24888/94, § 73, 16 December 1999: “Rule 4 of the Beijing Rules which, although not legally binding, might provide some indication of the existence of an international consensus, does not specify the age at which criminal responsibility should be fixed but merely invites States not to fix it too low”.

[70] Adopted by General Assembly Resolution 43/173 of 9 December 1988. The Court first cited them in Brannigan and McBride v. the United Kingdom , no. 14553/89, § 61, 26 May 1993.

[71] Adopted by General Assembly Resolution 45/113 of 14 December 1990. Cited by the Court in Blokhin v. Russia [GC] , no. 47152/06, § 87, 23 March 2016. The first reference was made in a separate opinion in Ertuş v. Turkey , no. 37871/08, 5 November 2013.

[72] Adopted by General Assembly Resolution 45/112 of 14 December 1990. Cited by the Court in Blokhin , cited above, § 88. The first reference was made in a separate opinion in Ertuş , cited above .

[73] Adopted by a working group of 24 international experts on 9 December 2007, annexed to the interim report of the Special Rapporteur of the Human Rights Council on torture and other cruel, inhuman or degrading treatment or punishment, Manfred Nowak, of 28 July 2008. The Special Rapporteur considered it “a useful tool to promote the respect and protection of the rights of detainees”. It was recently cited by the Court in Babar Ahmad and Others v. the United Kingdom , nos. 24027/07, 11949/08, 36742/08, 66911/09 and 67354/09, § 120, 10 April 2012.

[74] Adopted by General Assembly Resolution 65/229 on 21 December 2010. It was recently cited in Korneykova and Korneykov v. Ukraine , no. 56660/12, § 91, 24 March 2016. The first reference was made in a separate opinion in Khoroshenko v. Russia [GC], no. 41418/04, ECHR 2015.

[75] Adopted unanimously by General Assembly Resolution 70/175 on 17 December 2015.

[76] Adopted by consensus in September 1996 by 133 delegates from 47 countries, including 40 African countries, which met in Kampala, Uganda. The President of the African Commission on Human and Peoples’ Rights, Ministers of State, Prison Commissioners, Judges and international, regional and national non-governmental organisations concerned with prison conditions took part in the meeting.

[77] Approved by the Prison Services in Central, Eastern and Southern Africa (CESCA), in Arusha, Tanzania, 23-27 February 1999.

[78] Approved by the second pan-African Conference on Prison and Penal Reform in Africa, held in Ouagadougou, Burkina Faso, 18-20 September 2002.

[79] OEA/Ser/L/V/II.131 doc. 26. Inter-American Commission on Human Rights Resolution 1/08, Principles and Best Practices on the Protection of Persons Deprived of Liberty in the Americas, 13 March 2008, No. 1/08.

[80] Recommendation No. R(87)3 adopted by the Committee of Ministers of the Council of Europe on 12 February 1987 and Explanatory memorandum. In the 1987 version, the only reference to the allocation of floor space to prisoners was in Rule 15 (“The accommodation provided for prisoners, and in particular all sleeping accommodation, shall meet the requirements of health and hygiene, due regard being paid to climatic conditions and especially the cubic content of air, a reasonable amount of space, lighting, heating and ventilation”), which was explained as follows: “it is desirable that standard specifications should be drawn up at national level to meet the requirements of this rule according to local circumstances and practice.”

[81] Quite eloquently, the Commentary to Rule 18 starts this way: “This Rule concerns accommodation. Developments in European human rights law have meant that rules about accommodation have to be strengthened.” The Commentary further adds that “The importance of ensuring appropriate accommodation is further strengthened in the new version of the rules by treating it in combination with issues of allocation …”

[82] As the Court itself put it, the Committee of Ministers is the best intermediary of the European consensus ( M.C. v. Bulgaria , no. 39272/98, § 162, 4 December 2003).

[83] Commentary on Recommendation Rec(2006)2 of the Committee of Ministers to member states on the European Prison Rules, p. 6.

[84] The sole dissenting voice was the representative of Denmark, on a very strict point. The Representative of Denmark reserved the right of his government to comply or not with Rule 43, paragraph 2, of the appendix to the recommendation with regard to the requirement that prisoners held under solitary confinement be visited by medical staff on a daily basis.

[85] Other than the European Convention on the supervision of conditionally sentenced or conditionally released offenders (1964), ETS N°51, the Convention on the transfer of sentenced persons (1983), ETS N°112, and the Additional Protocol to the Convention on the transfer of sentenced persons (1997), ETS N°167, the Council of Europe has had an immensely rich standard setting activity in prison law since its early days, as witnessed by, among others, CM/Rec (2014) 4 on electronic monitoring, CM/Rec (2014) 3 concerning dangerous offenders, CM/Rec (2012) 12 concerning foreign prisoners, CM/Rec (2012) 5 on the European Code of Ethics for Prison Staff, CM/Rec (2010) 1 on the Council of Europe Probation Rules, Rec (2008) 11 on the European Rules for juvenile offenders subject to sanctions or measures, Rec (2006) 13 on the use of remand in custody, the conditions in which it takes place and the provision of safeguards against abuse, Rec (2003) 23 on the management of life-sentence and other long-term prisoners, Rec (2003) 22 concerning conditional release (parole), Rec (2000) 22 on improving the implementation of the European rules on community sanctions and measures, R (99) 22 concerning prison overcrowding and prison population inflation, R (99) 19 concerning mediation in penal matters, R (98) 7 concerning the ethical and organisational aspects of health care in prison, R (97) 12 on staff concerned with the implementation of sanctions and measures, R (93) 6 concerning prison and criminological aspects of the control of transmissible diseases, including aids and related health problems in prison, R (92) 18 concerning the practical application of the Convention on the transfer of sentenced persons, R (92) 16 on the European rules on community sanctions and measures, R (89) 12 on education in prison, R (88) 13 concerning the practical application of the Convention on the transfer of sentenced persons, R (84) 11 concerning information about the Convention on the transfer of sentenced persons, R (82) 17 on the custody and treatment of dangerous prisoners, R (82) 16 on prison leave, R (79) 14 concerning the application of the European Convention on the supervision of conditionally sentenced or conditionally released offenders, Resolution (70) 1 on the practical organisation of measures for the supervision and after-care of conditionally sentenced or conditionally released offenders, Resolution (67) 5 on research on prisoners considered from the individual angle and on the prison community, and Resolution (62) 2 on electoral, civil and social rights of prisoners.

[86] This argument was used by the Court itself in Kiyutin v. Russia , no. 2700/10, § 67, 10 March 2011.

[87] See, for example, the White Paper on prison overcrowding adopted recently by the CDPC, PC-CP (2015) 6 rev 7, 30 June 2016, §§ 33-39; Criminal Justice Alliance, “Crowded Out? The impact of prison overcrowding on rehabilitation”, 2012; and ICRC, “Water…”, cited above, p. 35, which refers to “serious negative effects on the physical and psychological health of detainees”, as well as increased prison unrest. Overcrowding is highly disruptive to prisoners’ living routines, activities, and treatment. For example, a 2004 report by the Joint Committee on Human Rights raised concerns about the link between prison overcrowding and self-inflected deaths in custody.

[88] The expression “prison overcrowding” is used in the present opinion in its widest possible sense, including not only prison facilities, but all other publicly governed detention facilities, like police stations and prison hospitals. Likewise, I will refer to “prisoners” in order to include people detained on remand, serving a sentence or confined in prison hospitals.

[89] Orchowski v. Poland , no. 17885/04, 22 October 2009, and Norbert Sikorski v. Poland , no. 17599/05, 22 October 2009.

[90] Ananyev and Others v. Russia , nos. 42525/07 and 60800/08, § 239, 10 January 2012.

[91] Torreggiani and Others v. Italy , nos. 43517/09, 46882/09, 55400/09, 57875/09, 61535/09, 35315/10 and 37818/10, 8 January 2013. The new domestic remedies were assessed in Stella and Others v. Italy (dec.), nos. 49169/09, 54908/09, 55156/09, 61443/09, 61446/09, 61457/09, 7206/10, 15313/10, 37047/10, 56614/10, 58616/10, 16 September 2014.

[92] Vasilescu v. Belgium , no. 64682/12, § 128, 25 November 2014.

[93] Neshkov and Others v. Bulgaria , nos. 36925/10, 21487/12, 72893/12, 73196/12, 77718/12 and 9717/13, §§ 281, 292, 27 January 2015.

[94] Varga and Others v. Hungary , nos. 14097/12, 45135/12, 73712/12, 34001/13, 44055/13 and 64586/13, 10 March 2015.

[95] Mironovas and Others v. Lithuania , nos. 40828/12, 29292/12, 69598/12, 40163/13, 66281/13, 70048/13 and 70065/13, 8 December 2015.

[96] See my separate opinion in Oçalan v. Turkey (no. 2) , nos. 24069/03, 197/04, 6201/06 and 10464/07, 18 March 2014.

[97] As the UNODC Handbook on strategies to reduce prison overcrowding (2010) has indicated, prison overcrowding is “the root cause of a range of challenges and human rights violations in prison systems worldwide, threatening, at best, the social reintegration prospects, and at worst, the life of prisoners”.

[98] See the ICRC Water, Sanitation, Hygiene and Habitat in Prisons Supplementary Guidance , 2012. The ICRC adds that the appropriate amount of space cannot be assessed by a simple measuring of space alone, other factors having to be taken in account, such as the condition of the building, the amount of time prisoners spend in the sleeping area, the number of people in that area, the other activities occurring in the space, the ventilation and light, the facilities and services available in the prison, and the extent of supervision available.

[99] From the ICRC’s perspective, “Emergency situations are sudden events of short duration. They may be caused by a political crisis, natural disasters, fire, riots, health crises in which large numbers of detainees need to be separated from the others or events which require the transfer of detainees from a prison that has been damaged to another prison.”

[100] See the Commentary to Rule 18 of the EPR, the CPT’s “Living space per prisoner in prison establishments: CPT standards”, adopted in December 2015, and the CM White Paper on prison overcrowding, 23 August 2016, § 37.

[101] Varga and Others , cited above, §§ 15, 16 and 51.

[102] See already in this sense, my opinion in Miranovas and Others , cited above.

[103] See the declaration by Malcolm Evans, Chairperson of the United Nations Subcommittee on Prevention of Torture and Other Cruel, Inhuman or Degrading Treatment (SPT), and Mykola Gnatovskyy, President of the CPT, 24 June 2016.