DISSENTING OPINION OF JUDGES ZUPANČIČ AND MARUSTE

A. Introduction

We regret that we cannot join the majority in finding a violation in this case.

Principally, to put into operation the full powers of Article 6, including the right to legal representation under Article 6 § 3 (c), in the context of post-conviction remedies (parole), presupposes a premise concerning the legal nature of the original sentence. With this premise we cannot wholly agree. On the other hand, if the situation were reversed, that is, if a member State were to introduce full formalisation of the in-prison disciplinary procedures, we would perhaps question the wisdom of such a solution – but we could not a priori maintain that this violates any aspect of the Convention. That is, unless dire consequences were to derive from such formalism – as they in fact did in Mastromatteo v. Italy ([GC], no. 37703/97, ECHR 2002-VIII).

If it were true that hard cases make bad law, we feel this is one of those situations.

1. In what we consider the key paragraphs ( 123 and 124) of the Grand Chamber judgment, it is said:

“123. As noted by Lord Chief Justice Woolf in R. v. the Secretary of State for the Home Department , ex parte Carroll, Al-Hasan and Greenfield ..., the award of additional days did not increase a prisoner ' s sentence as a matter of domestic law. The applicants ' custody during the additional days awarded was thus clearly lawful under domestic law. Nevertheless, the Court does not consider that this goes to the heart of the question of the precise nature of the penalty of additional days. ... [T]he Court ' s case-law indicates that it may be necessary to look beyond the appearances and the language used and concentrate on the realities of the situation. The reality of awards of additional days was that prisoners were detained in prison beyond the date on which they would otherwise have been released, as a consequence of separate disciplinary proceedings which were legally unconnected to the original conviction and sentence.

124. Accordingly, the Court finds that awards of additional days by the governor constitute fresh deprivations of liberty imposed for punitive reasons after a finding of culpability ...” (emphasis added)

2. We do not concur with the purported realism of the majority judgment – basically relying on the distinction between the so-called “ realities of the situation ” as opposed to mere “ appearances ” – especially since it is contradicted by the very last sentence of the next paragraph, where it is said:

“It is true that in Campbell and Fell the Court concluded that the penalty imposed ' came close to, even if it did not technically constitute ' a deprivation of liberty. However, the Court was constrained to so frame its finding since it was examining a ' loss of remission ' as opposed to an ' award of additional days ' for which the later 1991 Act provided.”

Either the distinction is only apparent or it is real. Its authenticity, as we shall demonstrate, cannot depend just on semantic modifications in the domestic law.

3. Our dissenting opinion thus turns on the legal nature of the three or respectively four concurrent original sentences : for rape, possession of an imitation firearm and attempted murder in the case of the first applicant, and for rape and robbery in the case of the second. Before we grow sentimental about the human rights of the two gentlemen, let it be noted that these are not traffic offences. Formalistic assessment of the risk personified by unreformed and potentially recidivist prisoners may grievously endanger the human rights of other people. This is just what we saw in Mastromatteo , cited above, where we examined the other side of the coin, namely the positive obligations of the State under Article 2.

4. With reference to paragraphs 36 and 37 of the Chamber judgment, and more specifically to section 33(2) of the 1991 Criminal Justice Act, the real question is whether, under this legal regime, the “original sentence” – at least in the established criminal-law sense (and in its entirety) – is stricto sensu still a sentence. Has a final judgment which ex lege entails the automatic right to be released, after only two-thirds of it have been served and in so far as the remaining third is concerned, become a superfluous legal fiction? Moreover, since the execution of a final penal judgment necessitates an exchange between the two branches of power (the judiciary and the executive branch), does this not raise a fundamental constitutional issue?

With reference to paragraph 116 of the Grand Chamber ' s judgment, it is clear that it has applied the Engel criteria – despite the Government showing how different these situations are and that no direct comparison is possible. We agree with the Government. We would like to point out, however, that the main difference between the situations does not lie in the fact that someone who is in the course of fulfilment of his civi c duty is under a certain (military) discipline. We consider as essential the fact that military service is not served as a consequence of a final criminal conviction by a court of law – whereas a prison sentence is. In principle, the final criminal conviction and sentence are, at least in the theory of penal and penitentiary law, res judicata . Consequently, what happens within the context of this conviction and sentence (in the course of serving the sentence) is related to the original penal judgment. Were there no original conviction, neither would there have been any disciplinary proceedings against the applicants. It is thus not accurate to argue, as the Grand Chamber does in paragraph 108, that disciplinary proceedings are “unrelated to the original conviction”.

B. Constitutional implications

5. The constitutional-law aspect arises indirectly as it did for Lord Woolf in R. v. the Secretary of State for the Home Department, ex parte Carroll, Al-Hasan and Greenfield (judgment of the Court of Appeal of 19 July 2001 ). There, the appellants argued that Article 6 of the Convention should apply to prison disciplinary proceedings, referring, inter alia , to the changes brought about by the 1991 Act. Lord Woolf, in order to resolve the issue, cogently reverted to the crucial preliminary question concerning the true legal nature of the “original sentence”.

“Section 42(1) of the 1991 Act provided a power to make prison rules which included provision for the award of additional days but section 42(2) makes it clear that where additional days are awarded to a prisoner the additional days are aggregated with the period which would otherwise have to be served before the prisoner is released on licence. ...

The new statutory framework properly understood is not fatal to the cases advanced by the appellants. Section 42 merely gives their case its proper perspective. The awards of additional days to be served by each of the appellants did not have the effect of adding to their sentence . It was not a fresh sentence of imprisonment. Their effect was to postpone the appellant ' s release on licence. The awards clearly had a practical effect so far as the appellants were concerned and that practical effect was to postpone their release. But there was no question of their sentence being increased as a matter of law . Additional days could not be imposed so that they extended the actual sentence, which the appellants were serving, and the sentence passed by the court was the justification for the appellant ' s detention for the purposes of Article 5(1) [of the European Convention on Human Rights].” (emphasis added)

We do not know how much clearer one can be. Taking such a view, with which we fully agree, on the “original sentence”, of course implies that the early release, the remission, the conditional release, the parole or whatever one chooses to call it, cannot be a prisoner ' s right. It may be a factual “expectation”, even a reasonable one, but at bottom it is still a privilege. The privilege may or may not be granted.

6. The accepted meaning of the res judicata doctrine in criminal law implies that the final judgment pronounced by the judicial branch is to be executed by the executive branch. In criminal law, the doctrine entails the finality of the sentence pronounced by the trial judge. The judicial branch of power, while applying the substantive criminal-law norm, finally determines the required term of imprisonment. When taking into account the retributive, the preventative, the reformative and other decisive factors for his legally binding sentencing decision, the judge, and especially so in the Anglo-Saxon legal tradition, enjoys a certain margin of discretion (teleologically, in terms of comparative justice etc.). The resulting pronouncement of a criminal sentence to a particular defendant is therefore a final act of an independent judicial power.

7. The execution of the judgment in criminal law may not be a single event. The due execution of the sentence of imprisonment may take many years. It requires the unremitting involvement of the executive branch (the prison authorities). At the end of the trial, the retributive element of the sentence may be finally determined but it is impossible for the trial judge to foresee the evolution of the desired improvement in the prisoner ' s personality (re-socialisation). On a day-to-day basis, it is hence the executive branch (the prison authorities) who must deal with the prisoner.

In consequence, for example, two-thirds of the sentence continue to depend on the retributive, reformative, preventative , etc. , criteria contemplated at the outset by the sentencing judge, whereas the latter two criteria are now, through the monitoring of the behaviour of the prisoner, under the discretionary power of the prison authorities. The secondary nature of this assessment derives from the fact that the “just desert” and other determinate aspects of punishment have been unyieldingly fixed in the original sentence. The early release is therefore by its very legal nature a less determinate matter of the prospective in-prison assessment of the prisoner ' s response to re-socialisation endeavours by the prison authorities and of mercy, clemency, leniency, etc.

8. Into this constitutional division of labour between the branches of power falls the judicial proviso that, for example, one-third of it may at the discretion of the executive branch possibly be subtracted from the original sentence if and only if the prisoner has demonstrated a real and personal capacity to be re-socialised. The import of parole (prospective conditional release), or the release earlier than foreseen in the final judicial sentence, is simply that it provides (a) for the required flexibility in the application of a certain fraction of punishment and (b) for the prisoner ' s direct motivation to improve his attitude vis-à-vis society.

9. This does not mean that the judicial branch of power, fully governed by the “fair trial” requirements of Article 6 as it should be, has thus surrendered its sovereignty (the finality of its judgment) to the executive branch. Judicial discretion and the power it implies have, as it were, already been exhausted. The final result, however, is the full original sentence.

A “window of opportunity” has been left open in case the prisoner serving his full original sentence behaves well. This should be seen as an exception to the rule implying the prisoner ' s duty to serve the full original sentence. The prison authorities are therefore assessing the situation within the legally binding context of the original sentence. It is that sentence and that sentence alone which empowers them to use their own discretion as to whether they will fully (or only partially) execute it.

10. In terms of constitutional law, therefore, the judicial discretion is primary, whereas the executive branch ' s discretion is secondary and derivative.

11. How e ver, under the current English regime, what was formerly a judicial proviso is now mandated by the legislature. In terms of comparative law this is not at all unusual. Most national criminal laws provide for the prospect of parole after two-thirds of the sentence have been served. The idiosyncrasy of the English legal system is the prisoner ' s automatic “right”, unless he breaches the prison rules, to be released earlier than foreseen by the original sentence.

This “automatic right” represents a further legislative incursion into the definiteness of the judicial decision. Certain legislative mandatory sentencing schemes were for this reason found to be unconstitutional by the various national supreme and constitutional courts – mostly on the grounds of the checks and balances doctrine. All such legislative intercession takes place in the context of a binding original sentence. The legislature cannot constitutionally set up a system in which every prisoner would automatically acquire the right to be released once he has served two-thirds of his sentence.

12. It is thus logically compelling that the denial of early release cannot be interpreted as a “fresh deprivation of liberty” (see paragraph 124 of the majority judgment).

C. The right and the privilege

13. The “proper perspective” (or, in the language of the majority judgment, “ the precise nature of the penalty of additional days”) Lord Woolf is referring to stems from the clear jurisprudential distinction between a “right” and a “privilege”. This differentiation has many decisive legal implications [13] . Rights, especially in criminal law, require restrictive substantive criteria ( lex certa, lex clara , principle of legality, etc.) and strict procedural formalism – whereas privileges (clemency, rewards, awards, prizes, honours, etc.) do not [14] . Rights and duties lend themselves to legal remedies and regulation, whereas privileges do not. To confuse the two, in other words to say that the prisoner now has ex lege the right (or the enforceable “legitimate expectation”) [15] to be released, rather than a privilege resulting from his morally desirable “good behaviour”, makes the law defeat precisely what it is intended to defend, namely the accepted wisdom of parole. If the law makes the conditional release a right rather than a privilege, it effectively deprives the prisoner of his motivation to improve.

14. We suspect that the established custom of awarding early release led the English legislature pragmatically to enshrine the status quo into law. It thus suddenly made a prisoner ' s release appear as a right – unless he breached the prison rules. It was this pragmatic “nomotechnical” approach which created the present mystification in the first place.

D. The philosophy of parole

15. One should keep in mind that probation and parole in criminal law have, ever since their inception in the nineteenth century, been predicated upon the positive and flexible – that is, non-rigid and non-formalistic – prospect of rewarding prisoners ' good behaviour. The historic success of both parole (conditional release) and probation (conditional sentence) is explained by this positive and lasting influence the rewarding of good behaviour has on the personality of the convicted criminal [16] .

16. E arly release in England is now semi-automatic and mandated by law. The prison governor may prolong the imprisonment for 42 days (for each breach of the prison rules). Yet, in comparison with the classic early release systems , this is simply the reversal of the method. Grosso modo and in advance , this regime promises to reward the prisoner ' s “good behaviour” unless he breaches the prison rules. In the classical parole system , the duty to serve the full sentence comes first , while the privilege of the reward of early release is secondary . Here, the privilege is promised beforehand and the original duty to serve a longer fraction of the sentence ensues only if there is a breach of the prison rules. This should dispel the false impression that reverting to the initial duty to serve a larger portion of the original sentence constitutes the new “ reality of the situation ” and that “ awards of additional days by the governor constitute fresh deprivations of liberty imposed for punitive reasons after a finding of culpability ” (see paragraphs 123 and 124 of the Grand Chamber judgment). The denial of a conditionally promised privilege (reward) cannot be construed as a new punishment. That the refusal of the provisional reward may now appear to be a (new) punishment does not at bottom change the nature of the parole system. The “new reality” of the situation, in other words, is simply the mirror image of the old reality.

17. To be clearer, let us take this one step further. Were the legislature minutely to regulate – say , in various “prison rules” – all the preconditions for early release, surely it would include in this regulation (since this is the principal purpose of parole, conditional release , etc.) the criterion of “good behaviour”. The assessment of what constitutes “good behaviour,” even if “bad behaviour” were to be exhaustively itemised in the prison rules, would inevitably require some discretionary judgment on the part of somebody, say , prison authorities. It should not perplex us that the situation here is reversed – since the question becomes whether the prison governor has the right to impose up to another 42 days for each instance of “bad behaviour” (breach of prison rules).

18. In addition, it would be illogical to infer from this alone that the prisoner ' s privilege to be released has for this reason become a right. Of course, once a privilege is granted, it does become a right. But the decision whether it should be granted is a decision about privilege, not about a right. This misunderstanding, we think, is part of the perplexity of this case.

19. Another pragmatic discrepancy (with fundamental procedural consequences) between judicial sentencing, on the one hand , and the early release assessment by the prison authorities, on the other hand, stems from the two very different kinds of reasoning required by the two different appraisals.

The sentencing phase of the criminal trial is wholly retrospective , whereas the monitoring of the prisoner by the prison authorities with a prospect of early release is mostly prospective . The retrospective assessment performed by the sentencing judge derives from the facts established beyond reasonable doubt during the main trial. The prospective assessment, however, since it deals with the probabilistic imponderables of the prisoner ' s future dangerousness – we have seen the tragic consequences of this in Mastromatteo , cited above, where the assessment was perfunctorily performed by the Milan giudice delle pene – simply does not lend itself to the same “fair trial” requirements as do the hard facts in the retrospective criminal trial.

The speculative imponderables concerning the future probability that the particular prisoner will , upon (conditional or early) release , relapse into criminal conduct , make unfeasible the regular adversarial (“fair trial”) give and take based on hard facts prove d beyond reasonable doubt. A measure of exploratory and provisional “arbitrariness”, it being a prognostic exercise, is unavoidable in this prospective assessment procedure.

Are we really prepared to make the early release procedure a mini “fair trial”, this being consequent to calling an early release a “right” requiring the application of Article 6 procedural minimal standards , if the formalistic prospective assessment will result in such appalling consequences?

20. Yet herein precisely lay the ingeniousness of the whole idea of parole. It let the judicial branch pronounce the full “original” sentence – with the proviso that the potential early release be subject only to the anticipated, because unavoidable, uncertainty of the subsequent decision by the executive branch.

As Lord Woolf saw it, the “original sentence” legally covers the prisoner ' s duty to serve his term until the last day of mandated imprisonment. Should he behave well and thus justify a benevolent expectation concerning his future civility, this cannot mean that the conditional release has become his right, his “legitimate expectation” , etc. A higher form of pragmatism tells us that the conditional release of a prisoner can only be a privilege authorised by the judge or the legislature and bestowed upon him by the executive branch of power (the prison authorities) .

21. Are the 42 days, which at a maximum the prison governor may impose, part of this prospective assessment, or are they simply a retrospective punishment for the breach of prison rules? If they are a punishment, do they not require a separate trial? If they do not, are they not somehow connected to the original sentence? If they are justified as an extrapolation from the original sentence, are they not so because they imply the prospective assessment we described above? If they do not, that is, if they are a fresh punishment for the breach of the prison rules, then they require a separate and “fair trial” – not just certain elements of it.

The Grand Chamber judgment either goes too far, or it does not go far enough.

[1] . See Engel and Others v. the Netherlands , judgment of 8 June 1976 , Series A no. 22.

[2] . See, for example, Weber v. Switzerland , judgment of 22 May 1990 , series A no. 177, p. 18, § 33 (“Disciplinary sanctions are generally designed to ensure that the members of particular groups comply with the specific rules governing their conduct”).

[3] . In Campbell and Fell v. the United Kingdom (judgment of 28 June 1984, Series A no. 80), the “especially grave” nature of the offences and related considerations, “whilst not of themselves sufficient to lead to the conclusion that the offences with which the applicant was charged have to be regarded as ‘criminal’ for Convention purposes, do give them a certain colouring which does not entirely coincide with that of a purely disciplinary matter” (p. 36, § 71). Mr Campbell was found guilty on charges concerning mutiny and gross personal violence (pp. 10-12, §§ 13-14).

[4] . See Tyler v. the United Kingdom (no. 21283/93, Commission decision of 5 April 1994, Decisions and Reports 77-A, p. 81) in which the applicant had been “found guilty” on charges of adultery by English ecclesiastic courts in proceedings characterised as “disciplinary” by the Commission.

[5] . Campbell and Fell , cited above, pp. 11-12, §§ 14-16, and pp. 37-38, § 72.

[6] . “The changes are not, however, as significant as the claimants contend. The most important common feature between sentences before the 1991 Act came into force and sentences after it came into force are that in both cases the sentences announced by the court are the actual period of the sentence” (see paragraph 41 of that domestic judgment). Lord Woolf also emphasised that he had to deal with Article 6 of the Convention, as Mr Greenfield had committed an offence against prison rules after the entry into force of the Human Rights Act 1998 (see paragraphs 24 and 33).

[7] . In paragraph 42 of Weeks (quoted in paragraph 64 of Stafford , to which a reference is made in paragraph 123 of the present judgment) it is stated, inter alia , as follows: “Furthermore, the word ‘after’ in sub-paragraph (a) does not simply mean that the detention must follow the ‘conviction’ in point of time: in addition, the ‘detention’ must result from, ‘follow and depend upon’ or occur ‘by virtue’ of the conviction ... In short, there must be a sufficient causal connection between the conviction and the deprivation of liberty at issue ...”

[8] . Indeed, Stafford , more than just taking into account the domestic legal developments, shows that the Court’s earlier approach to the situation of mandatory lifers (as reflected in Wynne v. the United Kingdom , judgment of 18 July 1994, Series A no. 294-A) had even been regarded by some domestic judges as a hindrance to desirable developments (see Stafford , §§ 46-47).

[9] . The majority judgment (paragraph 122) accepts that the original conviction and sentence continued to be the legal basis for the deprivation of liberty during the additional days.

[10] . See, for example, Winterwerp v. the Netherlands , judgment of 24 October 1979 , Series A no. 33, pp. 22-23, §§ 53 and 55.

[11] . See Van Droogenbroeck v. Belgium , judgment of 24 June 1982 , Series A no. 50, pp. 23-26, §§ 44-48.

[12] . See, for example, Weeks , pp. 28-29, §§ 56-58.

[13] . The distinction derives from the fundamental dissimilarity between the morality of duty and the morality of aspiration as elucidated by Lon L. Fuller in The Morality of Law (1965).

[14] . This should not be interpreted to mean that the privilege of early, temporary, conditional, etc., release may be granted arbitrarily, in a discriminatory fashion, etc. (see paragraph 9 above).

[15] . The term “legitimate expectation” (in French, e spérance légitime ) is what might be called the legitimatio ad causam activa relevant in terms of the first paragraph of Article 1 of Protocol No. 1. We find its use in the present context misleading.

[16] . See, for example, http://www.appa-net.org/media2003/parolehistory.htm .