DISSENTING OPINION OF JUDGE ZUPANČIČ JOINED BY JUDGE YUDKIVSKA

I

1. I regret that I cannot join my colleagues in the majority concerning this admittedly borderline case.

2. Superficially speaking, this case hinges upon the question whether retroactivity concerned a criminal sanction, i.e., punishment –, or whether it concerned only a therapeutic security measure imposed on the applicant. If the latter were true, the security measure would be , in its overall benevolence, akin to civil co mmitment (involuntary hospitalis ation) of ordinary mental patients –, which is why the issue of retroactivity would not have arisen.

3. The majority maintain that the retroactivity of the law enac ted on 25 February 2008 was acceptable, due to the fact that it concerned treatment rather than punishment –, although the law came into force four months and thirteen days after the nominally criminal act of the insane person had been committed. Seemingly, therefore, everything depends on the question whether this was, or was not, a criminal sanction.

4. The applicant maintains ( see paragraph 13 of the judgment ) that the prin ciple of legality enshrined in A rticle 7 of the European Convention on Human Rights should be seen as an obstacle to the sanction imposed on him given that the old law, valid at the time of the offenc e, did not foresee that direct route to incarceration in a mental hospital.

5. It is of course eminently acceptable that the legal system should, once it has established that an insane person had committed an act that is objectively imputable to him or her, react. In most other countries there are provisions for security measures to be applied in cases concerni ng the “ n ot guilty by reason of insanity” verdict . It is somewhat surprising that the French system did not have such provisions in place until 25 February 2008. Instead, it had a cumbersome system where the investigating judge would acquit the person in question. Only after the acquittal did the Prefect obtain the power to commit that person to a mental hospital.

6. In other words, I do not object at all to the parameters of the law of 25 February 2008. Seemingly, the only slight impediment in this case concerns the retroactivity of the said law. As usual the thorny questions in law arise only once we appreciate that the case is predicated on other tacit premises.

7. In the case of Achour v. France (as cited in the judgment) I concurred in the outcome. In that case, too, the problem was retroactivity , except that it concerned the retroactivity of a law on the consequences of the multi-recidivism. The French Cour t of Cassation in that case had very pertinently introduced a distinction between retroactivity in rem and retroactivity ad hominem . In other words, the Cour t of Cassation saw retroactivity concerning the status of being a multi-recidivist something that concerned him or her personally ( ad hominem ), and not only his or her criminal acts ( ad rem ). Therefore, being a multi-recidivist was something to be distinguished from having committed several criminal-recidivist acts. Thus it was possible for the Cour t of Cassation to maintain that there was no retroactivity given the fact that the applicant had, in the interim period, remained a multi-recidivist: once a recidivist , always a recidivist.

8. Therefore, we seem to be dealing with an underlying problem that has not, in criminal law doctrine, been addressed before let alone resolved. Namely, the question recurs as to whether a criminal actor is being punis hed only for the act that he has committed, or inversely, whether he or she is being punished for being a murderer, an arsonist, a rapist, etc.

II

9. Here, the question becomes noteworthy. There is an essential difference between the implications of criminal procedure on the one hand and the implications of an ordinary civil procedure on the other hand. In the latter case, it is easy to separate the object of the civil litigation from the subjectivity of the defendant. Even in tort , to take a more difficult case bordering on criminal law, the defendant may be liable to pay damages due to the fact that he had been negligent, reckless etc. –, b ut the damages that he must pay do not involve his whole personality. In other words, what is at stake in the civil litigation is clearly detachable from the personality of the defendant – the tortfeasor. In other instances of civil litigation, what is objectively at stake in the lawsuit has little or no connection to the subjectivity (personality) of the defendant.

10. However, this is not so in criminal procedure. Here, the criminal liability, the blame, the imputation of the crime, etc ., go directly to the defendant ’ s personality. Criminal law explores meticulously whether the act objectively imputable to the defendant is a genuine subjective expression of his overall personality: self-defence, defence of another, duress, insanity, mistake of fact , etc. – these are all reasons that break the cau sal link, as in the present case, between the personality of the defendant and the nominally criminal a ct. By comparison, even in tort the liability is more objective because there is no requirement of an intimate link between the damages and the personality of the tortfeasor. For this reason it is sometimes maintained that a criminal act is a tort in conjunction with sin, i.e., historically many acts that are criminal today were treated as torts. The damages for thes e acts were called wereg ild [1] .

11. Additionally, in civil litigation the tortfeasor may be found guilty and may be obliged to pay damages. If he has paid the damages, this is the end of the matter. Thus there is no recidiv ism or multi-recidivism in tort even though the tortfeasor may reiterate his objectionable behaviour. The inference never arises in law that he is a tortfeasor.

12. In criminal law and in criminal procedure, the personality of the defendant ( actor ) and the question of his or her criminal liability are inextricably linked. At its monocentric stage in criminal procedure, liability depends exclusively on the relevant subjective (personal) attitude of the criminal defendant (which is not even explored in civil litigation). Moreover, at the polycentric sentencing stage everything depends on the mitigating and aggravating circumstances and on other traits of the convicted person ’ s personality.

13. For this reason it is, in criminal law and criminal procedure, sometimes difficult to distinguish between the person ’ s criminal liability for the act as such ( per se ) on the one hand and his personality on the other . On the one hand we have the criminal act, which must be firmly connected to the personality of the actor and on the other hand we have the personality as such. Achour v. France demonstrates this elusive problem, as does the case at hand: in rem versus ad hominem .

14. The criminal sanction lato sensu , likewise , whether it is punitive or something else depends largely upon the very personality of the actor -defendant. If it is in fact punitive, the government cannot pretend otherwise ( see Blokhin v. Russia , no. 47152/06, 14 November 2013 , Chamber judgment pending before the Grand Chamber). And while in civil procedure the payment of damages represents the end of the case, in criminal procedure the convicted person – the defendant in person – must go to prison. Thus, the criminal sanction, too, whatever it is, cannot be detached from the convicted person ’ s personality.

15. The situation in this particular ca se is therefore ambiguous. The “objective imputation” of the material facts by the investigating judge depended on the sheer act of the defendant. In turn, once it ( the actus reus ) is objectively established, the circumstance that he is “ criminally insane ” ( the mens rea ) depends on his (mentally ill) personality. Yet this is only the exacerbation of the split in criminal procedure itself, where the finding of guilt is strictly separate from the criteria of sentencing.

16. There is no doubt in my mind that in criminal law and procedure the defendants are punished not only for what they have done ( an act) but also and primarily for who they are (being).

17. In the famous case of Robinson v. California , that I have dealt with in Achour v. France , the Los Angeles City ordinance incriminated the status ( ad hominem ) of being a drug addict. The Supreme Court of the United States then directed that the so-called “ status crimes ” we re inadmissible: there must always be an act of the criminal defendant, to which we attach criminal liability. Thus the issue of whether a criminal defendant is being punished for what he is (being) or for what he had committed (an act) has been settled , albeit superf icially . Neither the US Supreme Court nor other courts , nor criminal law doctrine , however, have ever succeeded in explaining why the principle of lega lity in criminal law, as enshrined in Article 7 of the Convention, is always dependent on the act of the defendant and seems to ignore the obvious fact that the personality of the actor is at the very centre of a ny criminal procedure.

18. The reason for this, as I have explained in Achour v. France , is pragmatic. The act of a defendant is determined in terms of place, of time and of modus ( operandi ). This is not true of the status of defendant. The status persists in time, has no abode, except hidden in the defendant ’ s personality, and may not have a consistent modus operandi .

19. In other words, the act is easily litigable, arguable, disputable , etc.; the status is not. (This comes to the fore even in the predetermined issue of insanity – mental illness as the cause of the act – , i.e., there may arise a battle between psychiatric experts, which may have no obvious conclusion.) Were it not for that reason, it would make much more sense to declare the defendant innocent or guilty strictly in terms, not of his instant act, but in terms of his enduring personality. After all it is the personality that is being punished. To put it otherwise, it is this “ enduring personality ” which goes to prison – the act is merely its symptom. Yet, the disease cannot be treated separately from the body of the patient.

III

20. The problem recurs in all cases that are somehow on the border line between the classical responsibility for the act on the one hand and those where the being of the actor is determinative of the case on the other. To name just a few, in Blokhin v. Russia (cited above) and in De Tommaso v. Italy (no. 43395/09), both now pending before the Grand Chamber, in M. v Germany (cited in the judgment) a nd prior to that in the above- mentioned Achour v. France case (cited in the judgment) , etc., the dangerousness of the actor was the de terminative factor. In juvenile delinquency cases the parens patriae doctrine, as in Blokhin , pretends that it is referring to the personality of the young offender in order to reform him (his personality, his being). In the case at hand, likewise, the S tate pretends that the indefinite forced incarceration in a mental hospital is for the benefit of the applicant. The gist of this beneficial assistance ( parens patriae ) theory is that there is no conflic t between the interests of the S tate and the interests of the person “ benefitting ” from this incarceration. In my opinion, the Court should look be hind these appearances. These cases also make it patent that the boundary between the act and the being of the actor is fuzzy.

21. This kind of uncertainty is not acceptable and thus the question arises as to how the human rights machinery must react in order to protect the applicant and the rule of law. One way or the other , all these cases are about Article 7 of the Convention , i.e., about the principle of legality. The latter, for the reasons we have explained above, concerning the time, place and modus of the offence, expressis verbis requires an act for the punishment to be legal and acceptable.

22. The position taken in M. v. Germany to the effect that retroactive prolongation of the security measu res wa s not acceptable under Article 7 of the Convention was perfectly correct. However, although they are essentially the same, the present judgment ( see paragraph 38 ) makes short shrift of distinguishing this case from M. v. Germany —, although this ought to have been the nucleus of the assessment. Thus, what is punitive on the other side of the Rhine is unexpectedly remedial on this side of the river.

23. Moreover, already subsequent to the 12 February 2008 Act and to the D ecree of 16 April 2008 (to the effect that the provisions of the law are applicable immediately), the Cour t of Cassation took the “ punitive ” position on 21 January 2009 — and then suddenly changed tack and reversed itself ( le revirement de jurisprudence ) nine months and twenty- two days later, on 16 December 2009.

24. This testifies to the chronic and widespread opacity of the legal situation. In the context of this fiction, clearly, the “ punishment ” is permitted for the criminal act of the defendant, whereas only “ treatment ” is permissible in so far as the being (the personality) of the defendant is concerned .

25. The reverse logic thus came into play , whereby the majority judgment maintains by legalistic fiat , i.e., on very formalistic grounds, that what affected the defendant in this case was not punishment but treatment. Since it was treatment, unlike the situation in M. v. Germany , retroactivity had been permissible. Thus in paragraph 38 of the majority judgment we read:

“ That distinction must nevertheless be used prudently, in view of the differences in the criminal laws of the member States that are enacted to protect society from the risks presented by dangerous criminals. The same type of measure may be classified as a penalty in one State and as a preventive measure, to which the nulla poena sine lege principle does not apply, in another (see M. v. Germany , cited above , §§ 74 and 126) .”

26. We would then expect a piece of persuasive reasoning as to why this dissimilarity , so difficult to explain and justify, does apply in France whereas it did not apply in Germany. However, in paragraphs 40, 41 and 42 we only read that “ the Court finds that the impugned measures imposed on the applicant, who was declared criminally insane, were not ordered following a conviction for a ‘ criminal offence ’ ”.

27. In other words, the Court ’ s doctrine according to which rights must not be illusory and theoretical and the Court ’ s intent must be to look behind mere appearances, has been sacrificed to this exquisite piece of legal formalism in which one notes the absence of a “ criminal offence ” even though we are told that the “ measures were ordered by the [Criminal!] Investigation Division after that court had declared the applicant criminally insane ” ( paragraph 40) . In view of this formalist ic distinction it is no wonder that the Cour t of C assation changed its mind within less than ten months.

IV

28. It remains to be explained why I assented to the majority judgment in Achour v. France , i.e., to the idea there , as opposed to the present case , that retroactivity was not a problem. In Achour the status of being (remaining) a multi-recidivist all along was seen as permitting the retroactive application of the law concerning multi-recidivists.

29. Why is it that here, in my opinion, the status of being a mental patient – since the applicant was apparently mentally ill from the outset – should not permit the Court to cover the retroactivity period of four months and thirteen days? Clearly, first, if it were a question of pure civil commitment (forcible hospitalisation of any mental patient), there would have been no problem as to the precise timing of the commitment. Secondly, the issue has never been raised, as it was in Achour , in terms of protraction (continuation) of the applicant ’ s being (argument ad hominem ) a dangerous mental patient. Thirdly, and most importantly, the status of continuous ly being a multi-recidivist in Achour was based on r ecurring criminal acts that had been legally proven since he had been, several times, finally convicted.

30. Ad primum , in the present case the commitment pronounced by the Criminal Investigation Division was not a civil commitment; it was imposed on the applicant by a division of the criminal court. Ad secundum , the Cour t of C assation has never reasoned in terms of civil commitment (forcible hospitalisation of any mental patient) because the commitment clearly came about in consequence of the “ objective imputation”, i.e., of a criminal-judicial finding that the applicant had in fact committed the act. Ad tertium , in Achour we had several proven criminal acts whereas in this case th ere were none: the applicant had acted once and had a mental il lness as an excuse, i.e., he had never been convicted of anything.

31. Furthermore , the mere findings by the investigating judge and Investigation Division did not amount to a conviction. The question of fair trial h ad never been raised, but since the “objective imputation” was determinative of the outcome (incarceration in a mental hospital) it, too, represents a major problem in the case at hand.

32. At any rate, in the absence of a fair trial in accordance with Article 6 of the Convention, the mere “objective imputation” cannot be considered as having established that the applicant had committed a criminal act. Even i f this is so, the presumption of innocence still prevails! Since such an act has not been fairly and legally established, it remains impossible to maintain that the applicant had been incarcerated in consequence of a criminal act.

33. Moreover, the French courts and this Court maintained that the sanction (to use a neutral term) was not a punishment, whereas if we go beyond appearances and if we grant the applicant a right that is not only theoretical and illusory, we must conclude that the incarceration in a mental hospital ward for the criminally insane is often much worse than ordinary imprisonment. There are a number of cases in our jurisprudence that testify to this (see for example the recent case of Zaichenko v. Ukraine (no. 2) , no. 45797/09, 26 February 2015 ). On top of that, the duration of this incarceration is, in contradistinction to ordinary p enal imprisonment, without the time- limit that otherwise applies to every criminal penalty. In view of this, to maintain that the applicant here was not punished is simply not true.

[1] I have borrowed this idea from the late Professor Harold Berman of Harvard Law School. See his work Law and Revolution: the Formation of the Western Legal Tradition , Harvard University Press, Cambridge, Massachusetts, 1985.