DISSENTING OPINION OF JUDGE MARTENS JOINED BY JUDGE KURIS

I - INTRODUCTION

- A -

1.   I have found it impossible to convince myself that in this case the United Kingdom has violated Mr Saunders ’ s rights under Article 6 of the Convention (art. 6). Nor has the Court ’ s judgment so persuaded me.

2.   What is at stake is a knotty, but important question relating to a topic which is not only very controversial but also appears prone to arousing rather strong emotions.

Assume that the "right to silence" and the "privilege against self ‑ incrimination" are not absolute (see paragraphs 7 - 12 below) but - like other rights implied in Article 6 (art. 6) - allow for limitations; assume, further, that such limitations cannot be taken into account unless they are in accordance with the law, pursue a legitimate aim and are proportionate to that aim [9] ; then the question becomes: Are these requirements fulfilled in the present case?

3.  Unlike the majority, I have come to the conclusion that this question is to be answered in the affirmative.  In order to elucidate that opinion I find it necessary to start with some general considerations with respect to both immunities in issue.

- B -

4.   In its judgment of 8 February 1996 in the case of John Murray v. the United Kingdom (Reports of Judgments and Decisions 1996-I, p. 49, para . 45) the Court has proclaimed that the notion of a fair procedure under Article 6 of the Convention (art. 6) comprises two immunities: the "right to remain silent" and the "privilege against self-incrimination".

The wording of this paragraph in the John Murray judgment - especially if compared to that in paragraph 44 of the Funke v. France judgment of 25 February 1993 (Series A no. 256-A, p. 22, para . 44) - clearly suggests that, in the Court ’ s opinion, two separate immunities are involved. From a conceptual point of view it would, however, seem obvious that the privilege against self-incrimination (= roughly speaking, the right not to be obliged to produce evidence against oneself) is the broader right, which encompasses the right to silence (= roughly speaking, the right not to answer questions).

The present judgment makes it less certain that the Court really makes a distinction. I will come back to that aspect of the present judgment later on (see paragraph 12 below). Here it suffices to note that in my opinion two separate, but related, rights are involved, of which the privilege against self-incrimination, as I have just indicated, is the broader one.

5.   In paragraph 45 of the John Murray judgment the Court further noted that these rights are not specifically mentioned in Article 6 (art. 6). Of course it was, moreover, aware of the fact that the Universal Declaration ignores both rights and that the International Covenant on Civil and Political Rights only contains "the right not to be compelled to testify against himself or to confess guilt" (Article 14 para . 3 (g)). It nevertheless felt entitled to hold as indicated in paragraph 4 above on the ground that these two immunities "are generally recognised international standards". Thus it furnished, albeit subsequently, both some motivation for and clarification of a similar finding in paragraph 44 of the above-mentioned Funke judgment, a finding which has been widely criticised as being unmotivated and unclear.

6.   In paragraph 45 of the John Murray judgment the Court even ventured to go into the difficult and highly controversial question of the rationale of these two immunities. It said that:

"By providing the accused with protection against improper compulsion by the authorities these immunities contribute to avoiding miscarriages of justice and to securing the aims of Article 6 (art. 6)."

- C -

7.   In the John Murray case I belonged to the majority. As the Court noted, what was at stake in that case was whether these two immunities were absolute. I was - and I still am - convinced that this question was rightly answered in the negative. Consequently, I found it neither necessary nor opportune to express my disagreement in respect of the Court ’ s observation that these two immunities "lie at the heart of the notion of fair procedure". I now do: I feel that this high-strung qualification - which is repeated in paragraph 68 of the present judgment - is somewhat exaggerating the weight of both rights, more particularly that of the privilege against self-incrimination.

8.   I think that, historically, both rights must be seen as the very negation of the old, inquisitorial notion that a confession is an indispensable condition for conviction and therefore must, if need be, be extorted. These immunities thus served the purpose of preventing suspects from being subjected to improper [10] physical or psychological pressure. I accept that both rights - and more especially the right to remain silent - still serve this purpose. Also today it remains necessary to protect suspects under custodial police questioning against such pressure.

I also accept that since there is a not negligible chance that statements made under pressure may be unreliable, the rationale of the immunities under discussion comprises - as the Court put it – the avoidance of miscarriages of justice.

Furthermore, I accept that there is a certain link between these immunities and the presumptio innocentiae as enshrined in Article 6 para . 2 (art. 6-2) in so far as they allow an accused not only to keep silent during police interrogation but also to refuse to answer questions of investigating or trial judges as well as to give evidence himself [11] .

9.   However, these rationales hardly justify the Court ’ s qualification of these two immunities as lying "at the heart of the notion of fair procedure". I therefore suspect that other, not explicitly mentioned, rationales have contributed to that qualification.

In this context I note that legal writers and courts have frequently accepted a further rationale [12] . Its formulations vary, but they all essentially boil down to the proposition that respect for human dignity and autonomy requires that every suspect should be completely free to decide which attitude he will adopt with respect to the criminal charges against him. On this view it would be improper, because incompatible with such respect, to compel an accused to cooperate in whatever way in bringing about his own conviction. This rationale often seems to be the main justification for the broader privilege against self-incrimination.

The present judgment strongly suggests that the Court now has embraced this view. A first argument for this interpretation is that in the second sentence of paragraph 68 it repeats the rationale given in John Murray (see paragraph 6 above) but - by prefacing its quotation by the words "inter alia" - underlines that this is only part of the rationale of the two immunities. A second and still more telling argument is the stress laid, both in the last sentence of paragraph 68 and in paragraph 69, on the will of the accused: the Court now underlines that the privilege against self-incrimination is primarily concerned "with respecting the will of an accused person". That comes very near to the rationale outlined above which allies both immunities to respect due to human dignity and autonomy.

10.   I do not, of course, deny that there is an element of truth in this view, but I am inclined to think that its weight should not be exaggerated. "Human dignity and autonomy" have an absolute ring, but in our modern societies it must remain possible to protect the community against forms of crime, the effective combating of which makes it imperious to compel (specific categories of) suspects to cooperate in bringing about their own conviction. I believe that especially the broader privilege against self-incrimination may be restricted by law in order to protect legitimate interests of the community. In my opinion it is, in principle, open to the national law to compel (specific categories of) suspects by threat of punishment to contribute passively or actively to the creation of evidence, even decisive evidence, against themselves. Suspects may be compelled to allow or even to cooperate in the taking of fingerprints, in the taking of blood for alcohol tests, in the taking of bodily samples for DNA tests or to blow in a breathalyser in order to ascertain whether they are drunken drivers. In all such and similar cases national legislatures are, in my opinion, in principle free to decide that the general interest in bringing about the truth and in bringing culprits to justice shall take precedence over the privilege against self-incrimination [13] .

11.   I fear that the impugned qualification of both immunities (as lying "at the heart of the notion of fair procedure") as well as the newly advanced rationale tying them to respecting "human dignity" imply that, in the Court ’ s opinion, the privilege against self-incrimination is far more absolute than it is in my view.

But for paragraph 69 of the present judgment - to be discussed in paragraph 12 below - I would have added that this difference of appreciation might also be illustrated by the above-mentioned Funke judgment. What was at stake in that case was not so much the right to remain silent as the privilege against self-incrimination, for Funke refused to hand over (possibly) incriminating documents. The Commission had, in my opinion rightly, concluded that the legitimate interests of the community overrode the privilege [14] , but the Court curtly refused to follow, which strongly suggests that in its opinion there was no room for a balancing exercise at all. It is worth while noting that the reasoning subsequently given in the John Murray judgment, namely its reliance on generally recognised international standards [15] , does certainly not justify this decision: both under the Fifth Amendment to the United States Constitution [16] and under the case-law of the Court of Justice of the European Communities [17] there is a right to remain silent, but in principle not a right to refuse to hand over documents, let alone an absolute right to do so.

12.   It is, however, at least open to doubt whether the Court in paragraph 69 of its present judgment has not - implicitly, without saying so openly, let alone without adducing cogent reasons for doing so - overruled Funke and essentially converted to the more restricted doctrine adopted inter alia by the Court of Justice of the European Communities. In this context it is significant that paragraph 69 refers to how the privilege against self-incrimination is understood "in the legal systems of the Contracting States and elsewhere". What is more important is that, whilst the first sentence of paragraph 69 seems to amalgamate the privilege against self-incrimination with the right to remain silent, the second sentence seems to imply that - contrary to Funke - the privilege does not comprise the power to refuse to hand over incriminating documents nor that to prevent the use of such documents, obtained under compulsion, in criminal proceedings.

I confess that I fail to see any other possible construction of paragraph 69 so that I presume that the above interpretation is correct. On that assumption two points should be made.

The first is that the merging of the conceptually broader right not to incriminate oneself with the right to remain silent reduces the scope of protection afforded to suspects. In my understanding of the privilege against self-incrimination the Court retains the power to control national legislation and practice (see paragraph 10 above), which it has forfeited under its present doctrine.

The second point is that it is - to put it mildly - open to grave doubt whether the distinction made between the licence to use in criminal proceedings material which has "an existence independent from the will of the suspect" and the prohibition of such use of material which has been obtained "in defiance of his will" is a sound one. Why should a suspect be free from coercion to make incriminating statements but not free from coercion to cooperate to furnish incriminating data? The Court ’ s newly adopted rationale does not justify the distinction since in both cases the will of the suspect is not respected in that he is forced to bring about his own conviction. Moreover, the yardstick proposed is not without problems: can it really be said that the results of a breathalyser test to which a person suspected of driving under the influence has been compelled have an existence independent of the will of the suspect? And what about a PIN code or a password into a cryptographic system which are hidden in the suspect ’ s memory?

In sum: I cannot accept the new doctrine. I stick to the notion of the privilege against self-incrimination and the right to remain silent being two separate, albeit related, immunities which allow for limitations.

II - FURTHER DELINEATING THE ISSUE

- A -

13.   It is high time, after these introductory remarks of a general nature, that I turn to the case at hand.

In doing so a first point to make is that it is of the utmost importance to keep in mind that in this case two stages should be clearly and carefully distinguished: at a first stage Mr Saunders had to appear before the inspectors of the Department of Trade and Industry (DTI) and only at a second stage did he have to face trial.

14.   It is important to carefully distinguish both stages since Article 6 (art. 6) applies to the second stage only. The reason is that during the first stage Mr Saunders was not yet "`charged with a criminal offence ’ , within the autonomous meaning of this expression in Article 6 (art. 6)" [18] .

Although, strictly speaking, paragraph 67 of the Court ’ s judgment only reminds us that neither Mr Saunders nor the Commission have alleged otherwise, its wording and especially the reference to the Deweer judgment [19] make it clear that a majority within the majority subscribes to this proposition.

15.   Why was Mr Saunders not yet "charged" during this first stage? Simply because he had not yet received an "official notification given to an individual by the competent authority of an allegation that he has committed a criminal offence" [20] .

Admittedly, a charge "may in some instances take the form of other measures which carry the implication of such an allegation and which likewise substantially affect the situation of the suspect" [21] . It might, therefore, be argued that the inquiry before the DTI inspectors was a form of such "other measures" which: (1) in view of the purpose of the inquiry and the circumstances of the case, carried the implication that Mr Saunders, who was a director of Guinness at the time of the bidding, was suspected of having committed a criminal offence and (2) was affecting his situation of suspect as substantially as if a criminal investigation had been ordered against him.

There is, however, no merit in this argument since, just as there is no "charge" if the "official notification" is not given by "the competent authorities" - that is, by the competent prosecuting authorities -, so there is no "charge" if the "other measures" do not emanate from the competent prosecuting authorities. It is common ground, however, that the inquiry before the DTI inspectors – apart from being, essentially, investigative [22] - did not emanate from nor was taken over by the prosecuting authorities [23] (15).

16.   It follows, firstly, that the fact that Mr Saunders was not entitled during this first stage to refuse to answer incriminating questions did not give rise to any violation of his rights under Article 6 (art. 6), especially not the right to remain silent or the privilege against self-incrimination.

Secondly, it follows that the essential issue in the present case is whether, where someone has made incriminating statements in the context of an inquiry during which he is obliged to answer any and all questions on pain of fine or imprisonment, it would be compatible with the right to silence and the privilege against self-incrimination to allow the use of these statements as evidence against him at his trial [24] (16).

- B -

17.   As already indicated in paragraph 2 above this is an important issue.

Our modern societies are "information societies", also in the sense that all of us, government agencies as well as citizens, to a large extent depend on various kinds of information, notably on information provided by (other) citizens. This applies in particular to government agencies: countless administrative decisions – whether imposing a duty or granting a favour - are based on such information. Information which simply cannot always be verified beforehand. One must, therefore, be able to rely on the veracity of such information. However, the old virtue of truthfulness has largely disappeared from modern morals. We have become "calculating citizens", putting our own interests above those of the community. No wonder that fraud in multiple forms is the bane of our societies: fraud in the field of taxes [25] and social security, fraud in the field of governmental subventions, fraud in the environmental sphere (illegal disposal of dangerous waste), fraud in the sphere of the arms trade and drugs (money laundering), fraud in the corporate sphere which may imply any species of the other frauds just mentioned. Frauds that are all the more tempting since our computerised world with its manifold cryptographic devices makes it much easier to effectively hide them [26] .

It is generally accepted, therefore, that the mere threat of penal and other sanctions does not suffice, but that regular random as well as special proactive audits, inspections and investigations by highly specialised agencies are indispensable for effectively combating such frauds. The auditors not only need expert knowledge, they also cannot do without "appropriate special powers" [27] . That normally includes not only the right to inspect correspondence and files, to verify books and accounts, but also to require a certain degree of active cooperation [28] by those under investigation, to be informed of passwords and other secret information, to secure the handing over of documents and replies to questions. Normally, such rights are enforceable by threat of punishment.

Hence - and because, obviously, such audits may imperceptibly evolve into a criminal investigation - there is an inherent conflict with the right to silence and the privilege against self-incrimination.

18.   This conflict may be solved in various ways and I think that we should realise that, even in a given legal system, different solutions may coexist.

Legislatures whose starting-point for such audits is the idea that ascertaining the truth is the weightier interest and, consequently, deny those under investigation the right to silence and the privilege against self ‑ incrimination by making it an offence not to answer questions or otherwise to refuse cooperation, have several options concerning the use in evidence of the material thus acquired in subsequent criminal proceedings against those who have been under investigation. Sometimes it is provided that such material cannot be used in evidence at all; sometimes, that such material may only be used in evidence in case of a prosecution for perjury; sometimes it may also be used when a person who has been under investigation, in criminal proceedings against him, gives evidence which is incompatible with the material in question; sometimes the material may be used in evidence unless the trial court finds that under the circumstances such use would be unfair. The present case is an example of the latter type of solution: whilst subsections (1) and (3) of section 434 of the 1985 Act [29] leave no doubt that this provision concerns the type of investigation discussed in paragraph 16 above, subsection (5) of that provision explicitly allows the use in evidence of answers given to the DTI inspectors.

19.   It follows that the main issue in the present case is whether subsection (5) of section 434 of the 1985 Act is incompatible per se with the right to silence or the privilege against self-incrimination. In view of what I have said in paragraph 17 above on audits it is difficult to deny that this issue is of general importance. The more so since holding that subsection (5) of section 434 of the 1985 Act is incompatible per se with the right to silence or the privilege against self-incrimination may, as a matter of inherent logic, entail that (since no use may be made of the answers of those who have been under investigation by DTI inspectors) the very same prohibition affects facts discovered in consequence of such answers, such as the existence of a foreign bank account or of a secret file!

III - DECIDING THE ISSUE(S)

20.   In trying to decide the main issue my starting-point is that I accept that the United Kingdom legislature - which in such matters should be allowed a certain margin of appreciation - could reasonably conclude that, where there are serious rumours of fraudulent management, the public interest of protecting society against such fraud demands that the truth comes out and that this justifies the system of inquiry as set up under the 1985 Act, a system under which officers and agents of the investigated company are obliged to cooperate with the DTI inspectors as laid down in section 434 of that Act, without enjoying the immunities under discussion.

21.   A first point to make is that subsection (5) of section 434 of the 1985 Act presupposes that the answers are incriminating: it allows their use in evidence "against him".

A second point to make is that, although at first blush it may appear that what is at stake is not the right to silence, but rather the privilege against self-incrimination, further analysis suggests that both rights are equally implied: on this view what is in issue is whether it is permissible to use in evidence answers obtained in an investigation in respect of which the legislator has deliberately set aside both the right to silence and the privilege against self-incrimination (see paragraphs 17 and 18 above and the text of sections 434 and 436 of the 1985 Act).

22.   I confess that I have hesitated somewhat in deciding the main issue. However, in the end I have come to the conclusion that I had not been persuaded that subsection (5) of section 434 of the 1985 Act is incompatible per se with the right to silence or the - broader - privilege against self ‑ incrimination. It is only fair to say that in this decision the serious consequences of the latter view indicated at the end of paragraph 19 above have played their part.

As already indicated in paragraphs 7 to 12 above, I see neither right as absolute and I therefore fundamentally disagree with the sweeping statement in paragraph 74 of the Court ’ s judgment according to which: "The public interest cannot be invoked to justify the use of answers compulsorily obtained in a non-judicial investigation to incriminate the accused during the trial proceedings."

I accept that at the trial of a driver accused of driving when intoxicated the outcome of a breathalyser test to which he has been compelled may be used in evidence against him, although it was obtained as a result of the legislature ’ s setting aside the privilege against self-incrimination. Why, then, should it per se be inadmissible to use in evidence statements obtained as a result of a similar setting aside both of the right to silence and the privilege against self-incrimination?

As far as the rationale for these immunities is to provide the accused with protection against improper compulsion by the authorities and, thereby, to contribute to avoiding miscarriages of justice, their rationale does not require to hold that subsection (5) of section 434 of the 1985 Act is inadmissible per se: the impartiality of the DTI inspectors who merely seek to establish the truth, their professional qualities - generally speaking senior barristers and accountants -, the procedure before them, based as it is on natural justice under control of the courts, and, finally, the circumstance that those under investigation are given advance notice in writing of what is required of them and may be accompanied by their lawyers [30] , seem to offer sufficient guarantees against improper physical or psychological pressure, whilst the power of the trial judge under section 78 of the Police and Criminal Evidence Act 1984 (PACE) [31] constitutes a further guarantee against unfairness arising from the inquisitorial nature of their inquiry as well as against any residual danger of miscarriage of justice.

Nor does the argument as to "human dignity and autonomy" compel one to conclude that the use in evidence of the answers given in the inquiry is per se inadmissible. As I have already indicated, this rationale is also to be relativised (see paragraph 10 above) and there is special justification for doing so in the present context. After all, under the hypothesis we are discussing, the answers are "incriminating" (see paragraph 21 above). That means that they to a certain extent disclose both the offence and its perpetrator. The question therefore comes down to asking whether that disclosure may be used when trying to bring that perpetrator to justice. Would it not be stretching the respect for his human dignity and autonomy - or, in the terminology of the Court, the respect for his will - too far to hold that the mere fact that he has made these "disclosing" statements in the context of an inquiry during which he enjoyed neither immunity necessarily results in making any and all use of such answers in evidence against him inadmissible per se?

I think that this question should at any rate be answered in the affirmative as making any and all use of such answers in evidence inadmissible per se would imply that there is a good chance that - although to a certain extent it had already been disclosed that he was the perpetrator - he would nevertheless go unpunished which - as I am prepared to accept [32] - in practice would be the effect in a good many of these complicated fraud cases. I find it rather difficult to accept that once the result of the investigation is that such frauds are exposed, the right to silence and the privilege against self-incrimination should, nevertheless, make it to all practical purposes impossible that those whose responsibility is to a certain extent already disclosed are brought to justice and punished. This would lead to undermining the deterrent function of the criminal law in an area where it is particularly needed (see paragraph 17 above) and, furthermore, seriously offend the public ’ s sense of justice.

23.   There is one further argument against the admissibility of legislation of the present type that must be discussed separately since it apparently impressed the Commission and, accordingly, has been - successfully [33] - urged by counsel for the applicant in his pleadings before the Court.

24.   I recall that we are discussing a two-tier type legislation, characterised by (1) establishing investigation proceedings in which those under investigation are obliged to cooperate with the investigators and to answer their questions without enjoying the two immunities under discussion (the first tier) and (2) making it, moreover, possible that answers obtained in those investigation proceedings be used in evidence at a subsequent trial against someone who has been under investigation (the second tier). The overall justification of this type of legislation is to protect the public against serious forms of fraud. That public interest justifies, firstly, setting aside the immunities under discussion in the first tier (the investigation stage) and, secondly, using the answers in the second tier (the trial stage) in order to make sure that, where the first tier has disclosed a perpetrator, that perpetrator gets his punishment in the second tier. Seen from the latter ’ s viewpoint, however, the whole process nullifies the right to silence and the privilege against self-incrimination.

The argument to be discussed claims that if a two-tier process, which amounts to nullifying these immunities for the sake of protecting the public against serious forms of fraud, is to be condoned, such a two-tier process must also be accepted if on similar arguments – that is: on the argument that public interest in being protected against robbery, violence, murder, etc., outweighs these very same immunities - it is established in respect of ordinary crimes. Which would, obviously, be the end of the two immunities under discussion.

25.   The argument is flawed in that it disregards a difference between the various forms of serious fraud and such ordinary crimes as robbery, violence and murder which in the present context is essential: in ordinary crime cases discovery of the crime nearly always precedes the investigation which, consequently, as a rule merely aims at establishing "who did it"; whilst in fraud cases the investigation generally has as its first and main purpose to establish whether or not a cr ime has been committed at all. This difference is essential since it explains why investigations into ordinary crimes as a rule come within the ambit of Article 6 (art. 6), whilst investigations in the field of possible fraud do not: those who are targeted by investigations of the former type as a rule ipso facto become "charged with a criminal offence" within the autonomous meaning of this expression in Article 6 (art. 6), whilst those who are under an investigation into possible fraud do not, and, therefore, may, without violation of Article 6 (art. 6), be denied the privileges under discussion in the first tier.

As to the second tier, where a charge has been brought so that Article 6 (art. 6) ipso facto applies always, there are sound arguments for distinguishing corporate fraud from other species of crime. First, there are important typological differences between the often well bred, highly sophisticated corporate wrongdoer and other criminals and, secondly, there is the essential feature which these forms of fraud share and sets them apart from other species of crime, namely that generally they are only detectable after an investigation of the type referred in paragraph 17 above and, moreover, may only be successfully prosecuted when the results of that investigation may be used in evidence against the wrongdoer (see paragraph 22 above).

For these reasons the argument fails. So does a similar objection from the applicant which puts the same argument in the form of alleged discrimination between "corporate criminals" who are deprived of the two immunities under discussion and "ordinary criminals" who enjoy them. The differences just mentioned imply that the cases are not similar, whilst the argument moreover fails to take into account the essential difference between the very urbane proceedings before the DTI inspectors and custodial police questioning.

26.   My conclusion that subsection (5) of section 434 of the 1985 Act is not incompatible per se with the two immunities does not, of course, exempt me from examining whether under the specific circumstances of the case the use of the applicant ’ s answers to the DTI inspectors was nevertheless unfair. This is what the Government called the "factual issue".

In this respect I recall: (1) that under section 78 of PACE [34] it was for the trial judge to see to it that use of these answers did not have "such an adverse effect on the fairness of the proceedings that the court ought not to admit it"; (2) that the trial judge held two extensive voir dires on the subject at the end of which he gave rulings which demonstrated preparedness to use his powers under the provision as well as sensitivity for the interests of the defence - inter alia by excluding the evidence from the eighth and ninth interviews [35] ; and (3) that the trial judge, in his summing-up - which was qualified by the Court of Appeal as "in the main a masterly exposition" [36] -, compared and contrasted what the applicant had said in court with his answers to the DTI inspectors, thereby demonstrating that he did not consider such use unfair [37] ).

Apparently, Mr Saunders has not been able to persuade the Court of Appeal that the trial judge has been remiss in guarding the fairness of the proceedings. Nor have I been so persuaded.

In paragraph 72 the Court attaches much weight to the fact that at a certain stage of the trial - days 45 to 47 - the transcripts of the interviews [38] were read to the jury. I recall, however, that right from the beginning of the trial Mr Saunders ’ s defence was essentially that, whatever fraud had taken place, he was innocent because ignorant. He maintained that he knew nothing about giving of indemnities or the paying of success fees and that he had not been consulted on such matters. The transcripts made it possible to refute this defence and were used to do so [39] .

I therefore find that - whatever is the exact meaning of the Court of Appeal ’ s rather approximate remark that the interviews "formed a significant part of the prosecution case" [40] (32) - Mr Saunders ’ s answers to the DTI inspectors were only used in evidence against him, essentially, in order to demonstrate that the evidence which he had chosen to give at his trial was not reliable in that it was incompatible on certain points with those answers. I do not consider that to have been an unfair use of those answers.

27.   For the above reasons I have voted against finding a violation of the applicant ’ s rights under Article 6 (art. 6).

[1] The case is numbered 43/1994/490/572. The first number is the case's position on the list of cases referred to the Court in the relevant year (second number). The last two numbers indicate the case's position on the list of cases referred to the Court since its creation and on the list of the corresponding originating applications to the Commission.

[2] Rules A apply to all cases referred to the Court before the entry into force of Protocol No. 9 (P9) (1 October 1994) and thereafter only to cases concerning States not bound by that Protocol (P9). They correspond to the Rules that came into force on 1 January 1983, as amended several times subsequently.

[3] For practical reasons this annex will appear only with the printed version of the judgment (in Reports of Judgments and Decisions 1996-VI), but a copy of the Commission's report is obtainable from the registry.

[4] See paragraph 67 of the judgment.

[5] See paragraph 48 of the judgment.

[6] See paragraph 67 of the judgment.

[7] See paragraph 48 of the judgment.

[8] See paragraphs 49 and 50 of the judgment.

[9] See the Ashingdane v. the United Kingdom judgment of 28 May 1985, Series A no. 93, pp. 24-25, para . 57; and my concurring opinion in the De Geouffre de la Pradelle v. France judgment of 16 December 1992, Series A no. 253-B.

[10] Term taken from paragraphs 45 and 46 of the John Murray judgment previously cited; if the Court's terminology in that judgment implies, as I think it does, that not every form of compulsion violates these rights, I agree; if, however, it implies that every form of compulsion is "improper" - which is a possible reading, the more so since it squares with the rationale to be discussed in paragraph 10 below – I disagree also on this point.

[11] See paragraph 47 of the above-mentioned John Murray judgment.

[12] I pass over - as in my view defective - such "rationales" as that these immunities prevent a suspect from being subjected to "cruel choices", or that it is unethical to compel somebody to collaborate in bringing about his own doom. Such "rationales" cannot justify the immunities under discussion since they obviously presuppose that the suspect is guilty, for an innocent suspect would not be subjected to such choices nor bring about his own ruin by answering questions truthfully! Innocent suspects are, therefore, not treated cruelly or unethically, whilst guilty suspects should not complain that society does not allow them to escape conviction by refusing to answer questions or otherwise hiding evidence.

[13] It remains, of course, for the European Court of Human Rights to control whether the restrictions of the privilege are in accordance with the law, pursue a legitimate aim and are proportionate to that aim - see paragraph 2 above.

[14] Loc. cit ., pp. 33 et seq ., paras. 63-65.

[15] See paragraph 5 above.

[16] There is only a valid Fifth Amendment claim if, due to the particular facts and circumstances of the case, the "act of producing" is both "testimonial" and "incriminating". To be noted: when a custodian of a collective entity produces the corporate records and documents his act does not constitute testimonial self-incrimination; he is, however, protected from condemning himself by his own oral testimony. In the context of my dissent in the present case it is of interest to note one of the Supreme Court's arguments for this restrictive doctrine:

"We note further that recognising a Fifth Amendment privilege on behalf of the records custodians of collective entities would have a detrimental impact on the Government's efforts to prosecute `white-collar crime', one of the most serious problems confronting the law-enforcement authorities. `The greater portion of evidence in wrongdoing by an organisation or its representatives is usually found in the official records and documents of that organisation. Were the cloak of the privilege to be thrown around these impersonal records and documents, effective enforcement of many federal and State laws would be impossible.' ... If custodians could assert a privilege, authorities would be stymied not only in their enforcement efforts against those individuals but also in their prosecutions of organisations." (Braswell v. US 487 US 99 (1988)).

[17] See its Orkem /Commission judgment of 18 October 1989 (374/87 [1989] ECR I-3343 et seq.) and its Otto/ Postbank judgment of 10 November 1993 (C-60/92 [1993] ECR I-5683 et seq.).

[18] See the above-mentioned Funke judgment, p. 22, para . 44.

[19] See note 15 below.

[20] See the Corigliano v. Italy judgment of 10 December 1982, Series A no. 57, p. 13, para . 34.

[21] Ibid.

[22] See the Court's judgment of 21 September 1994 in the case of Fayed v. the United Kingdom , Series A no. 294-B, pp. 47 et seq., paras . 61 and 62. See also paragraph 47 of the present judgment.

[23] See the Court's judgment of 27 February 1980 in the case of Deweer v. Belgium , Series A no. 35. In that case there was no official notification of impending prosecution. There was an inspection which was not performed within the context of the repression of crime. Nevertheless, as from a certain moment, the inspection got to a point that Deweer was deemed to be "charged" and that was when the procureur du Roi - the prosecuting authority par excellence – offered Deweer a means to avoid prosecution (see paragraph 46 in conjunction with paragraph 43). Similarly: application no. 4517/70, report of the Commission, Decisions and Reports 2, p. 21, paras . 68-72.

[24] So far I am in agreement with the Court: see paragraph 67 of its judgment.

[25] See Aronowitz , Laagland and Paulides , Value-Added Tax Fraud in the European Union ( Kugler Publications, Amsterdam, 1996) and the addendum thereto, which on a comparative basis gives data on the methods of combating this kind of fraud in the Netherlands , Belgium , the UK and Germany .

[26] "These shadow-accounts were maintained in special files, shielded by an impressive battery of passwords, software `bombs' and other defence mechanisms, and in theory at least could not be accessed through the main computer." Salman Rushdie, The Moor's Last Sigh. The committee which elaborated Recommendation No. R (95) 13 – see following note - noted "the expanding misuse by offenders of new telecommunication technologies and facilities, including cryptography".

[27] Term borrowed from Committee of Ministers Recommendation No. R (95) 13 concerning problems of criminal procedural law connected with information technology. See on this recommendation: P. Csonka , Computer Law and Security Report 1996 (vol.12), pp. 37 et seq. The introductory paragraph of this very helpful article shows that the problem has been studied within the framework of the OECD, the EU and the UN: the relevant reports and recommendations are quoted.

[28] See Chapter III of the appendix to Recommendation No. R (95) 13 referred to in the previous note.  Paragraph 10 of that chapter stipulates that "the investigating authorities should have the power to order persons who have data in a computer system under their control to provide all necessary information to enable access to a computer system and the data therein". Paragraph 10 refers to the obligation to cooperate in a criminal procedure "subject to legal privileges or protection". I quote it here since it demonstrates the necessity of a specific duty of cooperation with regard to modern technology.

[29] See paragraph 48 of the Court's judgment.

[30] See paragraphs 42 and 43 of the Court's judgment.

[31] See paragraph of the Court's judgment.

[32] I recall that Mr Saunders submitted that the prosecution case against him would have been in serious difficulties but for the use of the interview transcripts (see paragraph 83 of the Court's judgment).

[33] See the second sub-paragraph of paragraph 74 of the Court's judgment.

[34] See paragraph 52 of the Court's judgment.

[35] See paragraphs 28 and 29 of the Court's judgment.

[36] See paragraph 38 of the Court's judgment.

[37] See paragraph 33 of the Court's judgment.

[38] In paragraph 72 the Court says "part of the transcript", but paragraph 31 clearly implies that the complete transcripts were read which makes it understandable that the reading took three days. That the transcripts were read in full is the more probable since reading only parts could have been unfair towards the defence. However, if one accepts - as I do - that the transcripts were read in full, this reading can hardly be styled as "such extensive use".

[39] See paragraph 31 of the Court's judgment. I fail to understand why the Court, in paragraph 72 of its judgment seems to find it material that the prosecution "must have believed that the reading of the transcripts assisted their case in establishing the applicant's dishonesty". Of course they did and, as the outcome of the trial shows, rightly so. But what has that to do with the issue whether that reading made the trial unfair? Does the Court suggest that the prosecution had improper motives? Is that why it furthermore tries to base an argument on the prosecution's wish to avail itself also of the transcripts of the eighth and ninth interview, although in paragraph 29 of its judgment it has established that these transcripts had been ruled out by the trial judge? These uncertainties make the Court's reasoning on this important issue, which at any rate seems rather the more unpersuasive.

[40] See paragraphs 40 and 72 of the Court's judgment.