JOINT PARTLY DISSENTING OPINION OF JUDGES SAJÓ AND VUĊINIĊ

1. We voted with the majority in finding a violation of Article 5 §§ 1, 2 and 4, but, to our regret, we are unable to follow them on two points.

2. First, we would conclude that, with respect to the applicants, the conditions at the CSPA (first reception centre) on Lampedusa did not reach the threshold of Article 3.

3. It is well-established case-law that the duration of ill-treatment is an important factor in determining whether the threshold of Article 3 has been reached, particularly with respect to conditions of detention. Citing extensive case-law, the Grand Chamber recently reiterated as follows (see Tarakhel v. Switzerland [GC], no. 29217/12, § 94, ECHR 2014):

“The Court has held on numerous occasions that to fall within the scope of Article 3 the ill-treatment must attain a minimum level of severity. The assessment of this minimum is relative; it depends on all the circumstances of the case, such as the duration of the treatment and its physical or mental effects and, in some instances, the sex, age and state of health of the victim.”

4. In the past, the Court has found that conditions of detention did not reach the threshold of Article 3 on the grounds that the duration of the detention was too brief (see Slivenko and Others v. Latvia (dec.) [GC], no. 48321/99, ECHR 2002 ‑ II; Sakkopoulos v. Greece , no. 61828/00, 15 January 2004; and Hinderberger v. Germany (dec.), no. 28183/06, 30 March 2010). Indeed, the Court has found no violation or has declared inadmissible a number of applications related to conditions of detention of much longer duration than that of the applicants in the present case on the grounds that the short stay in question did not reach the Article 3 threshold (see Gorea v. Moldova , no. 21984/05, 17 July 2007 (14 days); Terziev v. Bulgaria , no. 62594/00, 12 April 2007 (10 days); and Karalevičius v. Lithuania , no. 53254/99, 7 April 2005 (6 days)).

5. Although the Court has in some cases found violations of Article 3 notwithstanding the fact that the duration of the treatment was short, those cases contained other significant factors (not present in the instant case) which outweighed that fact. The aggravating factors which have allowed the Court to find that the severity of the harm outweighed its short duration include the following: the presence of a particularly vulnerable individual, such as a sick or mentally ill detainee (see Brega v. Moldova , no. 52100/08 , § 42-43, 20 April 2010 , and Parascineti v. Romania , no. 32060/05 , § 53-55, 13 March 2012 ); exceptionally egregious conditions, such as overnight stays in confined areas with no place to lie down or lack of access to sanitary facilities (see T. and A. v. Turkey , no. 47146/11 , § 95-99, 21 October 2014 ; Gavrilovici v. Moldova , no. 25464/05 , § 42-44, 15 December 2009 ; Aliev v. Turkey , no. 30518/11 , § 81, 21 October 2014 ; and Burzo v. Romania , no. 75240/01, § 99-100, 4 March 2008); or confinement in a locked cell unsuited for habitation or otherwise dangerous (see Koktysh v. Ukraine , no. 43707/07 , § 93-95, 10 December 2009 ; Căşuneanu v. Romania , no. 22018/10 , § 61-62, 16 April 2013 ; Ciupercescu v. Romania (no. 2) , no. 64930/09 , 24 July 2012 ; Tadevosyan v. Armenia , no. 41698/04, § 55, 2 December 2008; and Neshkov and Others v. Bulgaria , nos. 36925/10, 21487/12, 72893/12, 73196/12, 77718/12 and 9717/13 , § 198, 27 January 2015 ). It should be noted that many of these cases involved more than one aggravating factor, such as confinement in a cell and lack of sanitary facilities. The situations at issue in those cases (in which the Court specifically noted the short duration) lasted between two and twelve days. Even with their aggravating factors, only three cases ( Brega , T. and A. v. Turkey , and Neshkov ) involved detention periods as short as those of the present applicants.

6. In the present case the applicants are healthy young men; they did not allege ill-treatment by the Italian officials, and during their stay they were able to move relatively freely within the Centre: they had access to food and necessary facilities; and their stays in the Centre lasted less than four days. Furthermore, it should not be overlooked that the applicants ’ temporary placement in the CSPA was in response to an emergency situation in which an unexpected influx of migrants had arrived on an island that lacked the infrastructure to properly accommodate all the newcomers.

7. To be sure, the reports cited by the majority (see paragraph 131) show that the conditions on Lampedusa at the time were distressing and unfit for long-term accommodation. However, in situations where the conditions of detention might otherwise reach the threshold of Article 3 for anyone subjected to long-term confinement, a short duration may minimise the harm caused by those poor conditions, with the result that the measure falls short of the Article 3 threshold. Such was the case here. Therefore, we conclude that there has been no violation of Article 3.

8. Second, we find that we are unable to join the majority ’ s conclusion that there has been a violation of Article 4 of Protocol No. 4.

9. The term “collective expulsion” has a specific meaning in international law, stemming as it does from the historical roots of mass expulsions [1] . As Jean-Marie Henckearts wrote in his seminal book on the topic of mass and collective expulsions, when Article 4 of Protocol No. 4 was drafted in 1963, it was the first international treaty to address collective expulsion. Its explanatory report defines the meaning of “collective expulsion” with reference to the recent past. As Henckearts notes, “[t]he phrase ‘ collective expulsions of aliens of the kind which have already taken place ’ alludes to the expulsion of Germans and others in the wake of World War Two and of the forced movement of peoples in Europe during the Interbellum.” [2]

10. International law prohibits targeting a group for removal from a territory without regard for the individual members of that group. As the International Law Commission has explained, “[t]he decision concerning expulsion is made with respect to the group of aliens as a whole. The procedure is conducted with respect to the group of aliens as a whole. The implementation of the decision is carried out with res pect to the group of aliens as a whole.” [3] Although Article 4 of Protocol No. 4 is not limited to mass expulsion of an entire ethnic community, it retains the core principle of individual treatment. Therefore, in order to understand what is and is not covered by the prohibition in Article 4 of Protocol No. 4, it is vital to distinguish between the expulsion of a large number of individuals in similar situations (which is permitted) and the expulsion of a group qua group (which is prohibited).

11. It is extremely rare for the Court to find a violation of Article 4 of Protocol No. 4. It is with good reason that only four violations of this Article have ever been found. The Court has, primarily, adhered to the historically rooted concept of collective expulsion in international law. Therefore, prior to this case, the Court has only found a violation of Article 4 of Protocol No. 4 where expulsion took place on the basis of group removal rather than on an individual basis.

12. In this connection, the Court has addressed two sets of circumstances in which Article 4 of Protocol No. 4 applies, neither of which are present in this case. First, there are cases in which members of a group are targeted for expulsion from a State ’ s territory purely on the basis of their membership of that group. The second situation is one in which an entire group of people are “pushed back” from a territory without consideration of the individual identities of the group members.

13. In the cases of ÄŒonka v. Belgium (no. 51564/99, § 62-63, ECHR 2002 ‑ I) and Georgia v. Russia (I) ([GC], no. 13255/07, § 175, ECHR 2014), the Court found that there were official policies of targeting a certain minority group for removal (Roma and Georgian nationals respectively). The Court determined that those removals had been ordered on the basis of group membership, rather than individual factors, notwithstanding the judicial approval of the deportations. In both cases the expulsions involved the forced removal of a minority population living within the member State ’ s territory.

14. The only other two cases in the Court ’ s history in which a violation of Article 4 of Protocol No. 4 has been found involve the return of an entire group of people without adequate verification of the individual identities of the group members. In Hirsi Jamaa and Others v. Italy ([GC], no. 27765/09 , § 185, ECHR 2012) , the Court found that “the transfer of the applicants to Libya [had been] carried out without any form of examination of each applicant ’ s individual situation”, adding: “It has not been disputed that the applicants were not subjected to any identification procedure by the Italian authorities, which restricted themselves to embarking all the intercepted migrants onto military ships and disembarking them on Libyan soil” (emphasis added). The Court found similar problems with identification in Sharifi and Others v. Italy and Greece , no. 16643/09 , 21 October 2014 , where the applicants ’ identities had not been individually verified before they were returned to Greece.

15. The situations in Hirsi Jamaa and Sharifi correspond to the ILC ’ s definition of collective expulsion: “The decision concerning expulsion is made with respect to the group of aliens as a whole. The procedure is conducted with respect to the group of aliens as a whole. The implementation of the decision is carried out with res pect to the group of aliens as a whole.” Indeed there can be nothing but collective treatment when a group is returned or deported without so much as identification of the individuals (such as passengers on a boat). It is also important to note that, in both of the above-cited cases, at least some i ndividuals on board were asylum- seekers, who were unable to state their asylum claims before they were returned.

16. The case-law therefore shows that there are two ways in which a collective expulsion can be carried out. First, individuals may be identified for expulsion based on their membership of a group; such was the case in Čonka and Georgia v. Russia . Alternatively, a group – by virtue of being physically together – can be targeted for expulsion without consideration of the individual people who make up the whole; this was the case in Hirsi Jamaa and Sharifi . Conversely, in the case of M.A. v. Cyprus (no. 41872/10, § 254, ECHR 2013), the Court found that there had not been a collective expu lsion where unsuccessful asylum- seekers had received identical deportation orders. The Court found that “the fact that the deportation orders and the corresponding letters were couched in formulaic and, therefore, identical terms and did not specifically refer to the earlier decisions regarding the asylum procedure is not itself indicative of a collective expulsion”.

17. In the present case the applicants were not expelled on the basis of membership of an ethnic, religious, or national group. They were returned to a safe country and were not, in any event, asylum- seekers; thus there was no issue of non-refoulement . Asylum- seekers, unaccompanied minors, and other vulnerable individuals were treated differently, as their status so required (the treatment of such individuals is not at issue in the present case). The applicants in this case fit into none of these categories. The fact that the applicants were not eligible for leave to enter Italy rendered unnecessary any further examination besides establishing their identity, nationality, and the existence of a safe country of return. That examination was carried out individually. Each applicant was identified upon arrival in Italy and later by the Tunisian consular authority and was provided with an individual removal order written in a language that he could understand. In the applicants ’ cases, there were no other individualised factors to be considered. Whether there had been one or one thousand migrants, the process would have been the same. After having their identities and nationalities verified by the Italian and Tunisian authorities, they were returned home in accordance with a treaty between the Italian and Tunisian g overnments. The return was ordered for each individual applicant by an Italian judicial authority. The streamlined process created by the two countries to deal with the sudden change in migratory flow did not disregard the individual migrant, but rather took into account the necessary considerations when deciding on removal.

18. By labelling as “collective expulsion” Italy ’ s attempts to police its borders during an unforeseen emergency, the majority do a grave disservice to an intentionally focused and narrow concept in international law, which is meant to apply only in the most severe of circumstances. To find a violation here misrepresents the reality of the situation faced by the Italian authorities and by the migrants in question. It necessarily dilutes a clear prohibition under international law that has its roots in the national homogenisation and genocidal policies of the twentieth century. Article 4 of Protocol No. 4 has no place in the present case of non-discriminatory and procedurally regular removal.

In view of our position regarding both Article 3 of the Convention and Article 4 of Protocol No. 4 (no violation), we conclude that there has been no violation of Article 13 either; in fact an effective remedy was available . For the reasons set out above we also conclude that the just satisfaction award is excessive.