DISSENTING OPINION OF JUDGE PETTITI

(Translation)

I have voted for finding a violation of Article 5 (art. 5), I dissociate myself entirely from the vote of the majority of the Chamber, and I regret that the case was not referred to the plenary Court.

In my opinion, the decision in the case of W. in fact departs from the traditional case-law of the European Court on the presumption of innocence and the restrictions to be imposed on pre-trial detention measures.

The decision which has been taken, even if it applies only to W. ’ s case, could be interpreted as approving pre-trial detention of four years in cases relating to economic legislation: bankruptcies, offences against company law, criminal mismanagement, etc.

The philosophy of Article 5 (art. 5) and of the European Convention leads me to adopt such a dissenting opinion. Article 5 (art. 5) is an article which protects personal freedom and limits pre-trial detention to specific cases, which must necessarily be given a restrictive interpretation.

Article 5 in combination with Article 6 (art. 5, art. 6) is an important provision of the Convention for protecting the presumption of innocence.

The Court, whether in plenary session or sitting as a Chamber, had always adopted a strict approach to reviewing compliance with Article 5 (art. 5), tending to limit the length of pre-trial detention. In the present case the reasoning of the Swiss Federal Court has prevailed. In my opinion the Federal Court wrongly adopted a criterion for assessment with respect to the risk of absconding which adds to the wording of Article 5 (art. 5). One cannot indeed reverse the burden of proof and require the detainee to prove that he will not abscond, a negative which is virtually impossible to prove. That would be to add a further exception to Article 5 (art. 5), as only imprisonment removes all danger of absconding.

In the present case the judicial authorities made no real attempt to require large sums as security (a normal practice in Switzerland ) or lay down conditions for strict judicial supervision, which would have warded off any danger of absconding, while the existence of residences abroad was not enough to make such danger certain. The cantonal judicial authorities appear to have been overburdened by the number of files relating to W.; but in bankruptcy cases liquidators ’ reports, civil actions by creditors, and expert reports produced in the bankruptcy procedure all provide precise evidence which may speed up the investigative procedures. The volume of documentation does not indicate any exceptional complexity, as it consisted primarily of account books and commercial documents.

In interpreting Article 5 (art. 5) and the nature of pre-trial detention, it must be borne in mind that liberty is the rule, detention the exception. Provisional or pre-trial detention must not damage the presumption of innocence. But that presumption is not only the fact of not being regarded by the judge as guilty, nor the fact of not being presented to third parties as guilty, but also the fact of making it possible for a defendant to cope with his position as an accused until his trial. As an extreme case, a person who knows he is guilty must be able, by remaining at liberty after being charged, to orientate his professional and family life and make arrangements for the future. In any event, as the European Court has often stated, pre-trial detention cannot be used to anticipate the sentence ( Letellier judgment of 26 June 1991; Tomasi judgment of 27 August 1992); it cannot reflect the judge ’ s feeling or opinion as regards the accused ’ s guilt.

Referring to the traditional case-law of the European Court , it will be seen that in cases such as Neumeister (judgment of 27 June 1968 ) and Stögmüller (judgment of 10 November 1969 ) the Court did not accept detention even for approximately two years.

Referring to comparative criminal law, it will be seen that the average length of pre-trial detention is less than two or three months and that with respect to economic offences and bankruptcies the average length is less than one year.

The Code of Criminal Procedure of the Canton of Berne provides in Article 111 that "during the preliminary investigation the accused shall as a rule remain at liberty", unless there is a risk of absconding, a risk that the discovery of the truth would be frustrated if the accused abuses his liberty, or a fear - if the accused has intentionally committed a further criminal offence during the proceedings - that he may commit other similar offences. These rules do not appear to have been applied strictly in W. ’ s case.

Criminologists are also mindful of the fact that every year thousands of persons accused of offences, who have been kept in pre-trial detention, possibly for a long period, have the charges against them dropped or are acquitted. Detention in such cases creates an injustice or an obvious social disorder which results in the judicial system being criticised.

The fact that the Swiss Federal Court delivered a very long judgment and upheld a severe sentence cannot in the circumstances justify the long period of detention, especially as the Swiss Federal Court had previously, with respect to the risk of collusion, partly quashed a decision to extend the detention which had been based on a criterion it considered open to criticism.

The European Commission originally put forward seven criteria for reviewing pre-trial detention. In the W. case the European Commission rightly found in its first report certain points of fact and procedure which induced it to vote by nineteen votes to one that there had been a violation of the Convention: bankruptcies between 1982 and 1984 (the facts went back to 1977); W. ’ s arrest on 17 March 1985; six out of twelve co-accused remained at liberty; eighteen searches between March 1985 and June 1986; first warrants issued from 3 April 1985; thereafter a series of rather confused stages:

"26. At the early stages of the proceedings difficulties arose with regard to the various accused persons ’ right to consult the case-file. At one stage, the investigating authorities considered that consultation of the case-file would have to be refused for some years. Altogether fourteen complaints and appeals were filed against various decisions of the investigating authorities. After April 1986 the accused were permitted to consult approximately 90% of the case-file, after 22 October 1986 virtually the entire case-file.

27. On approximately 350 occasions the investigating authorities questioned the applicant, the other accused and various other persons. However, as from 11 April 1986 onwards the applicant no longer replied to questions put to him by the investigating authorities. The latter nevertheless interrogated the applicant on altogether 36 occasions ...

...

40. The [Federal] Court noted [on 25 August 1986] that the two investigating judges charged with the investigations had so far worked very intensively, but that no assessment of the materials had yet commenced with a view to a subsequent indictment. Nor was it clear whether expert opinions should be ordered in respect of the company accounts and the psychiatric examination of the applicant. Finally, the court considered that the length of the applicant ’ s detention on remand did not yet come too close to the length of the applicant ’ s prospective prison sentence, even if in this respect the indictments chamber had probably gone too far when it assumed that an eventual sentence might be in excess of five years.

41. In July and October 1986 the investigating authorities ordered the preparation of two expert opinions concerning the company accounts, and a psychiatric examination of the applicant, respectively. The accountancy opinion was submitted on 10 April 1987 , the psychiatric opinion on 22 December 1986 . The latter confirmed the applicant ’ s full criminal responsibility ( Zurechnungsfähigkeit ).

...

52. The Federal Court considered in particular that the delay was justified by the additional work resulting from the investigations. Moreover, an excess of the maximum permissible duration of detention on remand would not be excluded as long as the investigating authorities had handled the investigations speedily. While its decision of 24 March 1987 had envisaged a maximum length of detention of two and a half years, the investigating authorities had meanwhile taken over proceedings instituted against the applicant in the Federal Republic of Germany. Nevertheless, the length of detention on remand should not come too close to the anticipated maximum duration of the prison sentence ...

...

65. The applicant ’ s further public law appeal was partly upheld by the Federal Court on 19 August 1988 . With reference to the Convention organs ’ case-law, the court found in particular that the applicant had not been granted the possibility in these proceedings to comment on statements of the public prosecutor ( Generalprokurator ) and the investigating judges.

...

67. The court found that a mere theoretical danger of collusion did not suffice to justify further custody. In the applicant ’ s case however there were concrete indications that such a danger existed. The court referred, inter alia, to the fear expressed by the Berne authorities that, if released from detention, the applicant would attempt to collude with his wife and various persons to fabricate exonerating evidence. The court also noted that on 2 September 1988 the applicant had been committed for trial ... and that therefore it could not be said that the authorities had disregarded the court ’ s instructions of 25 April 1988 ." [*]

It decided to commit for trial on 2 September 1988 and the applicant was convicted on 30 March 1989 for offences some of which dated back to 1977, civil and commercial proceedings having started in 1983 and the arrest dating back to 27 March 1985 .

The national decisions betray a certain embarrassment at the obstacles encountered by the defence in gaining full access to the case-file and the tendency to take account of the sentence which was likely to be passed in order to "justify" the extended detention. The Court could have drawn the consequences of such an assessment; it had always previously refused to accept the concept of detention anticipating the subsequent sentence ( Letellier judgment cited above and Kemmache judgment of 27 November 1991).

The argument adopted by the majority in the W. judgment does not seem to me to be adequate to a situation of four years ’ detention:

"Having regard to the intensive continuous review thus carried out by the highest national court, the Court agrees in substance with the Government ’ s arguments summarised in paragraph 41 above. It notes that the right of an accused in detention to have his case examined with particular expedition must not hinder the efforts of the courts to carry out their tasks with proper care (see among other authorities, mutatis mutandis, the Wemhoff and Tomasi judgments...)."

Such an interpretation of Article 5 (art. 5) might lead to the management of case-files being given priority over the right to liberty. It might be acceptable for swiftness to give way to judges ’ working requirements in the case of an accused who is at liberty or whose detention has just started, but not for detention for such a long period. The European Court had never accepted a duration of four years in earlier cases. Nor can the fact that the applicant allegedly committed other offences during his detention justify the extension of the detention: either this accusation was maintained and the judge was obliged to issue a separate arrest warrant on separate charges; or else the court was not able to use this fact as an argument for refusing release.

The fact that economic or financial criminal proceedings are very complex and require manifold investigations cannot justify extended detention. It is known that accounting reports in this field always take a very long time to produce, and this may prolong the investigation; but in bankruptcy cases specialist judges know how to make best use of the liquidators ’ reports so as to avoid prolonging their proceedings, which was not the case here, when W. ’ s bankruptcies dated back to 1982. It should be noted in addition that the accounting report was ordered by the court in July 1986, not in April 1985, although it was a measure which technically was necessary as from the arrest.

If one takes as a typical example the official statistics of the French Ministry of Justice, which could be transposed with similar results for other European States of similar population, the list of serious and less serious crimes by category for 1989 (similar figures in 1990-1992) shows: for bankruptcies, an average length of two months (seven cases of three months, one only in excess of eighteen months); for fraud, extortion and blackmail, an average length of four to eight months. Yet in France Parliament has often deplored the excessive length of pre-trial detention and has attempted to remedy this by reforming the Code of Criminal Procedure.

A reading of legal writing on criminal law and criminal policy shows that no academic specialist or practitioner in Europe justifies pre-trial detention lasting four years for economic offences, even multiple ones. In their writings the most eminent authors regret the excessive length of pre-trial detention. Thus Mr Vassalli , a former Italian Minister of Justice, a member of the Constitutional Court and the originator of criminal reform, in Droits de l ’ homme et durée de la détention , Giusto Processo 1989; similarly Mr Chiavario , Evolution du droit et procédure pénale , Vol. II, Politique criminelle , Giuffrè , 1991; similarly Belgian legal writing; similarly French legal writing: Delmas -Marty, Bouloc , Levasseur (see Revue de science criminelle et de droit comparé , and Mélanges Levasseur Ed. Litec ).

The study of pre-trial detention in English law by Professor L.H. Leigh, professor at the London School of Economics, makes observations to similar effect. In 1986 the Government had given an assurance to the House of Commons that the average period spent waiting for a hearing (for all summary and indictable offences taken together) was 57 days. In 1985 it was an average of 10.5 weeks, in 1987 12.9 weeks. For more serious offences the periods are still reasonable (see Home Office, Criminal Justice: a working paper, 1986).

The case-law on detention also has effects on the criminal policy of States confronted with serious crises and sometimes mutinies caused by the saturation of prisons and in part by excessive use of pre-trial detention. Judges ’ colleges are worried about such situations. The European Court therefore has a responsibility with respect to criminal policy by means of its case-law.

To justify four years of pre-trial detention is to step backwards in the history of criminal law, to regress to the "prehistoric" era of the Lombroso school of thought.

If the development of criminal law in Europe since the failure of the positivist school is considered as a whole, it will be found that:

(1) a number of States have enacted legislation laying down a maximum length for pre-trial detention (six months or one year, for example in Czechoslovakia );

(2) the case-law of the other States generally limits the length of pre-trial detention to about six months to two years;

(3) the teaching given at judges ’ training colleges on the European Convention on Human Rights and its Article 5 (art. 5) tends to persuade them to reduce pre-trial detention, in reliance on the case-law of the European Court, as some investigating judges have a tendency to prolong detention in order to put pressure on accused persons and induce them to make admissions or denunciations, which tends to abolish the right of silence;

(4) comparative law shows that no country (other than Switzerland ) practises detention for four years in the field of bankruptcy and fraud, even for criminal cases which are more serious than economic offences.

The history of the European Convention and the development of the case-law of its institutions are marked by serious concern to preserve individual freedom and limit pre-trial detention, at least for ordinary crimes.

The teaching given in the judges ’ training colleges and bar schools is inspired by the same principles. In the member States of the Council of Europe which have investigative proceedings, practitioners have noted that certain judges have a propensity to anticipate the sentence sometimes by pre-trial detention, or to press the accused to make admissions by postponing appearances for months while dismissing requests for release. In the present case W. ’ s refusal to co-operate may be explicable by the difficulties he experienced in having the documents in the case-file notified to him in full; the Federal Court noted this. In any event this refusal to "co-operate" was not capable of justifying prolongation of his detention for such a long period. An accused is entitled to take the risk that his negative attitude during the investigation may "handicap" him at the trial.

The perverse effects of prolonging pre-trial detention are well known to criminologists and criminal practitioners.

The use made of this by certain investigating judges may transform an investigation into a coercion to confess or a punishment for refusing to accuse oneself. It is known that for first offenders the exemplary and deterrent effect of detention operates from the first days or weeks; prolongation is therefore unnecessary and harmful. There are too many cases known of suicide or early death caused by illness during detention for one not to approve the tendency in European writing on criminal law to criticise the abuse of pre-trial detention.

The Court followed and - wrongly in my opinion - accepted the reasoning of the Swiss Federal Court without, however, adopting the Federal Court ’ s concept of assessment of the proportionality between the pre-trial detention and the future sentence likely to be passed, when the courts carry out their assessment. The Court has thus not overruled its earlier thinking and case-law which refused to accept that pre-trial detention could anticipate the sentence to be pronounced by the trial court.

The particular circumstances of the W. case admittedly relativise the scope of the European Court ’ s decision. The Court will have other cases before it which will permit it to give better expression to its "philosophy" of criminal policy with respect to pre-trial detention, as where liberty is concerned an overall concept of protection must be preserved.

For all these reasons I have concluded that there was a violation of Article 5 (art. 5).