JUDGMENT

Justice E.E. Levy

Introduction

1. Can the silence of a defendant who is called to testify in his trial satisfy the requirement of ‘something extra’ for confessions that he made during his interrogation, in the course of which he took responsibility for the offences that were attributed to him? That is the question before us in this appeal.

On 24 Adar II 5765 (4 April 2005) leave was granted to appeal on this issue. Later, because of the importance of the issue, the attorney-general was joined as an additional respondent in the appeal. The time has come to make a decision.

The main facts

2. In an indictment that was filed against the appellant, Corporal Itai Milstein, in the Central District Court Martial, it was alleged that during his military service he made use of a dangerous drug of the cannabis type (‘grass’), an offence under s. 7(a) and 7(c) of the Dangerous Drugs Ordinance [New Version], 5733-1973.

During the trial the appellant denied the offences attributed to him, but when he was asked to testify, he exercised his right to remain silent. Consequently the military prosecutor applied to submit as evidence two confessions that he made during his interrogation (prosecution exhibits 2 and 3), in which he confessed to several incidents in which he used cannabis-type drugs.

In his confession (prosecution exhibit 2), the appellant said that from the age of 16 he has been in the habit of using drugs. The appellant also admitted that after he was conscripted into regular military service he once made use of a drug (‘grass’) together with a friend. The appellant even described his feelings after smoking the drug (‘I felt “high,” I was thirsty’) and also his feeling of regret that he felt as a result of this act (‘after that use, I said to myself: Why did I do that? I am in the army, in a combat unit. I hope to become a tank commander’). Moreover the appellant said that he used the drug after he had not left the base on leave for 21 days, and he went on to give various details about the method of smoking and the appearance and smell of the drug.

After it examined this confession, the District Court Martial came to the conclusion that it should be given considerable weight, in view of the many indications of truth that it revealed. In its words:

‘A reading of the defendant’s confession (prosecution exhibit 2) shows that we are dealing with a well-ordered confession from a chronological point of view, and one that is intelligible, complete and logical. The defendant includes his descriptions of events from the time when he was 16 years old until the time of the interrogation… Moreover, his confession is very detailed; it gives a detailed description of the manner in which he used drugs, including an explanation of the various methods of using cannabis. The defendant gives details of his feelings after he used the drug, the reasons that led him to use the drug during his military service, and finally he expressed regret at the deed… The concern that the confession may be false is also reduced because the defendant chose in his confession not to mention the names of the persons who supplied him the drugs or the name of the person with whom he made use of the drug during his military service. The defendant also knew how to say, in his defence, that he never bought drugs himself or had possession of drugs at home… From this we can see that the defendant’s replies in his interrogation were well thought out and it is clear that they were not given lightly or because of a hidden ulterior motive or pressure’ (p. 4 of the verdict).

The court martial discussed how the concern that a confession may be false is also not significant in this case because of an additional reason, and that is a memorandum (prosecution exhibit 5) that was made by the interrogator, in which it says that ‘the interrogation took place in a good atmosphere with full cooperation on the part of the suspect.’ The court martial also emphasized that the appellant did not make any claim or complaint in his interrogation and that ‘he did not act in an unusual manner during his interrogation, even though during the interrogation he shed a tear or two’ (p. 4 of the verdict).

Later the court martial turned to consider whether there was ‘something extra’ that might provide the support required for this confession. It was held that the appellant’s silence in the trial could satisfy this requirement, since he did not take the trouble to give any reason for his refusal to testify, and not even the smallest piece of evidence was brought before the court martial to indicate that the confessions were made under any pressure that he experienced. Therefore it was held that ‘It was to be expected that a defendant who wishes to deny his confession will take proper steps to persuade the court martial of the truth of his later story’ (ibid.), something that the appellant did not do. Later, in view of his conviction, the court martial sentenced the appellant to 20 days actual imprisonment and a two month suspended sentence.

3. The appellant, who was not reconciled to his conviction, brought his case before the Appeals Court Martial, and he claimed that the support required for his confessions in the interrogation should not be inferred from his silence. With regard to this, counsel for the appellant raised a broader argument. According to him, the very possibility of reaching incriminating inferences from the silence of a defendant is likely to undermine the presumption of innocence and the right of defendants to remain silent, and it may also impose on them the burden of proving their innocence.

The Appeals Court Martial rejected these arguments. It held that according to what was stated in s. 162 of the Criminal Procedure Law [Consolidated Version], 5742-1982 (hereafter: ‘the Criminal Procedure Law’), if a defendant refrains from testifying in his defence, this may serve as support for the prosecution’s evidence and even serve as corroboration for it. Therefore, a fortiori it was held that there was nothing to prevent the silence of the defendant also constituting ‘something extra’ for his confession in the interrogation. Moreover, the Appeals Court Martial held that a defendant who chooses to remain silent during his trial expresses consent to the charges brought against him by his conduct, and therefore there is nothing wrong in the possibility of attributing to this choice an incriminating probative significance. Finally it was held that drawing incriminating inferences from the silence of a defendant in a criminal trial does not derogate from the right of silence that the law gives him, nor does it impose upon him the burden of proving his innocence, since this burden remains the duty of the prosecution authorities, who are required to prove the guilt of the defendant beyond all reasonable doubt.

The Appeals Court Martial relied in its judgment on a series of rulings it had made in the past, in which it was held that the silence of a defendant may serve, in appropriate circumstances, as ‘something extra’ for a confession that was made out of court. In one of those rulings (A 41/01 Chief Military Prosecutor v. Levy [75], which was cited at length in the judgment of the Appeals Court Martial, the following was said:

‘The concern that there may be an “internal pressure” that leads a defendant to refrain from testifying in the court is not consistent with his pleading not guilty to the charges, since the significance of pleading not guilty is, in practice, that the defendant retracts his confession that he made out of court.

It may be imagined that someone who “commits suicide by his confession” will continue with his “suicidal” approach and also plead guilty to an offence that he did not commit. And if the defendant decides to change his policy and defend himself against the charge brought against him, there is, as a rule, no reason why he should not take his place on the witness stand and tell his story…

If it really was a case of “internal pressure” arising from the inner world of the defendant or from other external causes, it is hard to believe that the defendant would have been able to retract his confession in the interrogation and deny the charges in court’ (ibid. [75], at p. 21 of the judgment).

For other decisions in the same vein, which testify to the strong hold that this ruling has in the courts martial, see also A 85/80 Cohen v. Chief Military Prosecutor [76]; A 190/01 Lahav v. Chief Military Prosecutor [77]; A 238/02 Britchock v. Chief Military Prosecutor [78]; A 59/03 Chief Military Prosecutor v. Schulman [79]; AA 28/03 Chief Military Prosecutor v. Al-Dema [80].

For the sake of completeness I will add that the appellant’s appeal against his sentence was allowed, and it was held that he would serve the sentence that was handed down to him in the form of military labour and not behind bars.

The arguments of the parties

4. The parties set out at length their positions on the question under discussion. They attached to their pleadings many appendices and references to Israeli case law and foreign case law, learned articles and provisions of statute. The following, in a nutshell, is a summary of the arguments.

Learned counsel for the appellant, Adv. Or Bassok, is of the opinion that the additional probative requirement of ‘something extra’ for a defendant’s confession — as opposed to the additional requirement of ‘support’ — should be external and separate from the confession, and therefore the silence of a defendant in his trial cannot satisfy this requirement. He argued that the defendant’s silence at the trial, which took place after he confessed during the interrogation that he committed the offences attributed to him, does not necessarily indicate that he was involved in those offences, but it may derive from a series of other reasons, which do not indicate his guilt. For all these reasons learned counsel for the defence expressed his opinion that the assumption that a person who is innocent will defend himself in a positive manner against the charges levelled at him by giving testimony in court is a speculative and unfounded assumption. Moreover he argued that the rulings made by the courts martial are based on an erroneous assumption that only defendants who are actually guilty exercise their right to remain silent, whereas innocent defendants will not hesitate to make a positive statement of their innocence during the trial.

He also emphasized that the provisions of s. 162 of the Criminal Procedure Law, which state that if a defendant refrains from giving testimony, this may serve as ‘support’ or ‘corroboration’ for the prosecution evidence, do not address the additional requirement of ‘something extra’ at all. Consequently, according to the outlook of learned counsel for the defence, it is not possible to extend s. 162 by means of interpretation, since had the legislature intended to include ‘something extra,’ it would have said so expressly. He also argues that the outlook that regards ‘support’ as a stricter probative requirement than ‘something extra’ is erroneous, since the difference between ‘support’ and ‘something extra’ is a qualitative difference and not a quantitative one. On the basis of all this, counsel for the appellant is of the opinion that the conclusion reached by the Appeals Court Martial, according to which the refusal of a defendant to testify may serve as ‘something extra’ for his confession, is an erroneous conclusion, both because it negates the difference between ‘support’ and ‘something extra’ and makes them into one, and also because it undermines the purpose for which the courts require a confession of a defendant to be supplemented by ‘something extra.’ Such an interpretation, so counsel for the defence believes, is also required by the Basic Laws and by the desire to protect the basic rights of defendants in a criminal trial.

An additional line of argument presented by counsel for the appellant concerns the alleged undermining of the presumption of innocence and the right to silence of defendants in criminal proceedings caused by the ruling which is the subject of this appeal. He argues that a finding that a refusal of a defendant to testify may serve as evidence against him undermines the presumption of innocence and the rule that the burden of proof rests with the prosecution authorities, since they are no longer required to discharge the burden of proving guilt independently, and from now they can also rely on the defendant for this. Such a situation, in the opinion of counsel for the defence, is also likely to make the right to remain silent meaningless, since the very threat of the possibility of making use of silence as incriminating evidence will compel the defendant to waive this right and to testify even when he does not wish to do so.

5. In the opinion of the attorney-general — which was submitted in consultation with the chief military prosecutor — the following position was presented: the provisions of s. 162 of the Criminal Procedure Law, which make it possible to infer from the defendant’s silence ‘support’ or ‘corroboration’ for the weight of the prosecution’s evidence, can serve as a legal source for regarding this silence also as ‘something extra’ for a confession that the defendant made during his interrogation. The respondents also think that the possibility of regarding the defendant’s silence as ‘something extra’ for his confessions does not undermine either the presumption of innocence or the right to remain silent, since it does not compel the defendant to abandon the right to remain silent and testify in his trial.

Notwithstanding, the respondents did not hide their opinion that the possibility of regarding silence as ‘something extra’ for a confession may give rise to a difficulty because of the concern of false confessions, and in their arguments they even expressly admitted that ‘there may be cases in which a defendant, who took responsibility in a police interrogation for carrying out an offence that he did not commit, will deny the charges but choose not to testify in his own defence’ (para. 44 of the attorney-general’s arguments). In view of this recognition, the attorney-general also issued a guideline to the prosecution authorities to act with caution before filing indictments that are based solely on the confession of a suspect in an interrogation. This guideline, which was attached to the respondent’s arguments, says the following:

‘a. As a rule, indictments should not be filed on the basis of out of court confessions if the evidence collected during the investigation does not contain “something extra” to strengthen those confessions. In exceptional cases it will be possible to file an indictment in such circumstances, but only with the approval of the state attorney or the chief military attorney as applicable.

b. In so far as the question will arise in the trial, because the ‘something extra’ that was contained in the investigation material was undermined during the trial, as a rule the prosecution should not request the court to regard the refusal of the defendant to testify as “something extra,” except in special cases, in which the confession itself is complete, detailed and convincing, and all the circumstances surrounding it remove all reasonable doubt that it was not made freely and willingly or that it was made as a result of some kind of “internal pressure,” and even then it requires the approval of the district attorney or the attorney of the relevant command as applicable’ (para. 4 of the attorney-general’s arguments).

Notwithstanding the aforesaid, the respondents are of the opinion that there is no basis for introducing a rule that will create a sweeping prohibition against treating a defendant’s silence as ‘something extra’ for his confession, and in their opinion the courts should be given discretion to determine on a case by case basis — in view of the nature of the confession — whether the silence of the defendant can be regarded as ‘something extra’ or not.

Deliberations

6. The question before us in this appeal requires us to consider the importance of defendants’ confessions as a means of discovering the truth and doing justice, the concerns that accompany such confessions and the nature of the probative requirement of ‘something extra’ that is needed where the conviction of a defendant is based solely on a confession that was made by him during his interrogation.

The probative significance of the right to remain silent

7. It will be recalled that a considerable part of the appeal was devoted to the argument that the possibility of giving the silence of defendants an incriminating probative significance undermines the right to remain silent, the presumption of innocence and the rule that provides that in criminal law the burden of proof rests with the prosecution and not with the defendant.

There is no need to elaborate on the importance of the presumption of innocence and the right to remain silent in a criminal trial. According to the presumption of innocence, which is one of the basic principles of criminal law, we assume that ‘a person — every person — has a presumption of innocence from felonies and misdemeanours as long as the contrary has not been proved’ (CrimA 4675/97 Rozov v. State of Israel [1], at p. 369; see also CrimApp 8087/95 Zada v. State of Israel [2], at p. 145). The practical significance of this presumption is that the burden of rebutting it rests with the prosecution authorities, and it will succeed in doing this only when it proves the defendant’s guilt beyond reasonable doubt (s. 34V(a) of the Penal Law). Nothing less will suffice. In this sense it is possible to see the right to remain silent also as one of the aspects of the presumption of innocence. According to this right, which is enshrined in s. 161 of the Criminal Procedure Law, a defendant cannot be compelled to defend himself in a positive manner against the charges brought against him, and therefore he is not obliged to testify during his trial or answer questions that are asked in the course of an investigation that is being conducted against him, when the answers to these questions may incriminate him (see LCrimA 8600/03 State of Israel v. Sharon [3], at pp. 756-757; LCA 5381/91 Hogla v. Ariel [4], at p. 381). This shows another aspect of the strong status of the presumption of innocence, because legal systems that give defendants a right to remain silent do not compel them to disclose incriminating information that is in their possession, and the burden of proving guilt remains with the prosecution authorities alone.

8. Thus we see that the presumption of innocence and the right to remain silent are two of the foundations on which our criminal law is based. They also constitute an integral part of the right of a defendant to a fair trial (see CrimA 5121/98 Yissacharov v. Chief Military Prosecutor [5], at para. 66; CrimA 4596/05 Rosenstein v. State of Israel [6], at para. 53; HCJ 11339/05 State of Israel v. Beer-Sheba District Court [7], at para. 24). They are directly connected with the principles of fairness, justice and liberty (HCJ 6972/96 Association for Civil Rights in Israel v. Attorney-General [8], at p. 783). They reflect a recognition of the huge disparity of forces between the state, in its capacity as prosecutor, and the defendant standing trial. They impose on the state the burden of justifying the violation of human rights that is caused as a result of the conviction and sentencing of defendants. They reduce the risk of mistakenly convicting an innocent person. Therefore some authorities regard them as constitutional rights that are derived from the Basic Law: Human Dignity and Liberty (CrimApp 2169/92 Suissa v. State of Israel [9], at p. 342; HCJ 6319/95 Hachmi v. Justice of Tel-Aviv-Jaffa Magistrates Court [10], at p. 755; A. Barak, ‘The Constitutionalism of the Legal System following the Basic Laws and its Ramifications on (Substantive and Procedural) Criminal Law,’ 13 Bar-Ilan Law Studies (Mehkarei Mishpat) 5 (1996), at p. 23; D. Bein, ‘The Constitutional Protection of the Presumption of Innocence, 22 Tel-Aviv University Law Review (Iyyunei Mishpat) 11 (1999); M. Lindenstrauss, ‘The Presumption of Innocence’ in Israeli Law and American Law — Selected Topics (1999), at pp. 5-7, 112-113; R. Kitai, ‘The Importance of a Positive Presumption of Innocence, its Role and Nature in the Proceedings Prior to the Verdict in Criminal Trials,’ 3 Alei Mishpat 405 (2004), at p. 444; R. Kitai, ‘Protecting the Guilty,’ 6 Buff. Crim. L. Rev. 1163 (2004).

Prof. Dennis discussed the relationship between the presumption of innocence and the right to remain silent, on the one hand, and the values of human liberty and dignity and the relationship between the individual and the state, on the other:

‘It is for the prosecution to prove the accused’s guilt and not for the accused to prove innocence. According to the theory the fundamental rule concerning the burden and standard of proof in criminal cases expresses more than a bare rule of decision for the court in situations of uncertainty, and more than a rule about the allocation of the risks of misdecision. In addition it makes a political statement about the relationship between the state and the citizen’ (I.H. Dennis, ‘Instrumental Protection, Human Right or Functional Necessity? Reassessing the Privilege against Self-Incrimination,’ 54 Cambridge L.J. 342 (1995), at p. 353).

The appellant is of the opinion that the possibility of attributing probative significance to the silence of a defendant undermines the status of the right to remain silent and the presumption of innocence, in that it allows the prosecution authorities to use the defendant as a means of obtaining incriminating evidence. Thereby, in the appellant’s opinion, the burden of proof is moved from the prosecution to the defendant.

9. This question, which has proved fertile ground for legal literature, is broader that the scope of the dispute addressed by this appeal and it is not limited merely to the question of the status of a defendant’s confession in an interrogation and the nature of the additional probative requirement that it needs. There is no doubt that the possibility of drawing incriminating inferences from the silence of a defendant may induce defendants to waive this right and encourage them to testify with regard to the charges against them, but there are important considerations, both for and against, with regard to the question whether this policy should be adopted and the defendant should pay a price for his silence which will happen if we allow the courts to draw incriminating inferences from the use of this right.

Those who oppose the drawing of such inferences hold that the mere threat that hovers over the defendant that his silence may strengthen the prosecution’s evidence will place him under pressure and thereby compel him to waive the right to remain silent. Those who hold this position claim that such a situation moves the burden of proof to the defendant and even turns him into a source of incriminating evidence. This was what G.W. O’Reilly meant when he claimed that drawing inferences from the defendant’s silence also limits his freedom of choice when he comes to decide how to conduct his defence:

‘It also diminishes the accusatorial system’s protection of individual autonomy and free choice because, when suspected of a crime, individuals are no longer free to choose whether or not to provide the government with evidence to aid in securing their own conviction; they are bound to do so or face an inference of their guilt’ (G.W. O’Reilly, ‘England Limits the Right to Silence and Moves Towards an Inquisitorial System of Justice,’ 85 J. Crim. L. & Criminology 402 (1994), at p. 451).

(See also the position of Dr S. Easton in her book The Case for the Right to Silence (second edition, 1998), at pp. 180-181; C.A. Chase, ‘Hearing the “Sounds of Silence” in Criminal Trials: A Look at Recent British Law Reforms With an Eye Toward Reforming the American Criminal Justice System,’ 44 Kan. L. Rev. 929 (1996), at pp. 942-946). Another opinion holds that the right to silence is a means of doing justice and not a value in itself. According to this approach, the possibility of drawing incriminating inferences from the defendant’s silence will make it difficult to achieve the main goal of arriving at the truth, because it will provide an incentive for defendants to give false testimony instead of availing themselves of the right to silence, which will require the courts to distinguish between true testimony and false testimony, despite the risk of error to which this gives rise. An absolute right to remain silent, without any possibility of drawing inferences that are unfavourable to the defendant from his use of this right, is therefore, according to this approach, a tool that will make it possible to distinguish between innocent defendants and guilty defendants (D.J. Seidmann & A. Stein, ‘The Right to Silence Helps the Innocent: A Game-Theoretic Analysis of the Fifth Amendment Privilege,’ 114 Harv. L. Rev. 430 (2000)).

On the basis of these outlooks, which have been adopted in American law, the United States Supreme Court has held in the well-known case of Griffin v. California [81], at p. 615, and in a host of later cases that no probative inferences of any kind should be drawn from the silence of a defendant in a trial. The court there also emphasized the lofty status of the privilege against self-incrimination as a constitutional right that is enshrined in the Fifth Amendment to the Constitution, and it held that drawing inferences from the silence of the defendant is tantamount to imposing a sanction on his exercising this constitutional right. The Supreme Court of Canada has also held, on the basis of the same outlook, that no adverse inference should be drawn from the silence of the defendant, because this would violate the right to remain silent that is enshrined in section 11 of the Canadian Charter of Rights and Freedoms (R. v. Noble [83]).

10. But this outlook has not been accepted in the Israeli legal system. According to the outlook prevailing in Israel, the right to remain silent is not violated even if the court is permitted to draw inferences from the defendant’s decision to realize this right. Admittedly the defendant is free to decide whether to testify or to remain silent. We do not compel the defendant to testify. ‘The defendant who remains silent — as opposed to the witness who remains silent — acts within the framework of the law; but the court has the power to interpret his conduct in accordance with its impression and understanding’ (CrimA 1497/92 State of Israel v. Tzubari [11], at p. 203). This outlook is also consistent with the general provision of s. 53 of the Evidence Ordinance [New Version], 5731-1971, that ‘The reliability of witnesses is a matter for the court to decide in accordance with the witnesses’ conduct, the circumstances of the case and the indications of truth that are revealed during the trial,’ since the silence of the defendant, like all other conduct, may also be a source for the court forming an impression of the defendant. My colleague, President Barak, has also expressed his outlook that the possibility of drawing inferences that are unfavourable to a defendant who remains silent does not violate his constitutional right:

‘The right to remain silent is a part of human dignity, in the sense that a person should not be compelled to testify, but it would appear that human dignity is not violated if we draw an unfavourable inference from his remaining silent’ (A. Barak, ‘Human Dignity as a Constitutional Right,’ 41 HaPraklit 271 (1994), at p. 285).

With regard to the English legal system, which also allows adverse inferences to be drawn from the defendant’s silence, Prof. Ingraham has expressed the opinion that the possibility of drawing inferences from the defendant’s silence is consistent with the presumption of innocence and does not transfer the burden of proof to the defendant:

‘The jury is not compelled to draw the inference of guilt; the law does not create a presumption of guilt, which becomes conclusive on the failure of the defendant to offer rebuttal evidence’ (B.L Ingraham, ‘The Right of Silence, the Presumption of Proof, and a Modest Proposal: A Reply to O’Reilly,’ 86 J. Crim. L. & Criminology 559 (1996), at p. 591).

Prof. Dennis expressed a similar opinion in his article cited above. According to his outlook, the possibility that the silence of the defendant will give rise to probative inferences is also dictated by logic, since where there is incriminating evidence against the defendant and he refuses to offer an innocent explanation for that evidence, it may be assumed that such an explanation simply does not exist. Dennis is also of the opinion that such an outcome is also consistent with the burden of proof required in criminal cases:

‘The legal burden of proof is not reversed by the restriction of the right to silence; if the tribunal of fact is left with a reasonable doubt after consideration of all the evidence the accused must be given the benefit of it. It is not for the accused to “prove” innocence’ (Dennis, ‘Instrumental Protection, Human Right or Functional Necessity? Reassessing the Privilege against Self-Incrimination,’ supra, at p. 355).

See also in this context the position of E. Gross, ‘The Right Not to Incriminate Oneself — Is It Really a Landmark in the Struggle of Enlightened Man for Progress?’ 7 Bar-Ilan Law Studies (Mehkarei Mishpat) 167 (1989), at pp. 188-190. This outlook has also been expressly recognized in statute, and in 1976 the Criminal Procedure Law was amended by adding the provision now found in s. 162, which provides the following:

‘Silence of the defendant

162. (a) A refusal of the defendant to testify may serve as support for the weight of the prosecution’s evidence and as corroboration for the prosecution’s evidence where it requires corroboration, but it shall not serve as corroboration for the purpose of section 11 of the Rules of Evidence Amendment (Protection of Children) Law, 5715-1955, or for the purpose of section 20(d) of the Interrogation Procedures and Testimony of Persons with Disabilities Law.

(b) …’

This court has held in a host of cases that this provision is consistent with experience, since ‘an innocent man is not only willing to testify, but he is happy for the opportunity to enter the witness box and refute the incriminating evidence, which he claims is false’ (CrimA 196/85 Silberberg v. State of Israel [12], at p. 525). The source of this approach can be found in the time of one of the greatest legal thinkers, Jeremy Bentham, and his famous statement —

‘Innocence claims the right of speaking, as guilt invokes the privilege of silence’ (J. Bentham, A Treatise on Judicial Evidence (London, 1825), at p. 241).

This gives rise to the outlook that the defendant’s decision to remain silent rather than to try and make a positive declaration of his innocence ‘may imply that his conscience recognizes that he is guilty’ (CrimA 139/52 Attorney-General v. Keinan [13], at p. 644).

11. It is not superfluous to point out that the provisions of s. 162 do not make it possible to regard silence as evidence with an independent weight of its own. It is clear that a conviction of a defendant cannot be based solely on the silence of the defendant: silence is not the same as a confession. The provision was also not designed to allow the prosecution to extract from the defendant new incriminating evidence ex nihilo. All that the provision was intended to do is to allow the court to regard silence, in the appropriate cases, as additional support for the existing evidence that was assembled by the investigation authorities (see the remarks of Justice Sussman in CrimA 112/52 Gabuv v. Attorney-General [14], at p. 254). Moreover, the provisions of s. 162 do not require the trial court to regard the silence of the defendant as support or corroboration for the prosecution evidence. All that the provision says is that in appropriate circumstances the silence of the defendant may corroborate or support the prosecution’s evidence, and there may certainly be cases — as has happened on more than one occasion — where it was found that it would not be right for the defendant’s silence to strengthen the incriminating evidence assembled in his case, since we recognize that there may be cases in which the defendant’s silence is the result of innocent motives and does not indicate his guilt. Thus, for example, in several cases it was held that where the defendant’s silence is intended to shield a person close to him that was also involved in the criminal enterprise, it is possible that his silence has no probative value (see, for example, CrimA 115/82 Muadi v. State of Israel [15], at p. 234; CrimA 7293/97 Jafar v. State of Israel [16], at pp. 474-475; CrimA 1888/02 State of Israel v. McDaid [17], at p. 231). There may of course be additional cases, with their own special circumstances.

This outlook, which makes it possible to regard silence as support for the evidence presented by the prosecution, was also adopted by English law more than a decade ago in the Criminal Justice and Public Order Act 1994. Section 35 of that law, which introduced a major normative change in the attitude of the courts to the silence of a defendant, also provided that the silence of a defendant may have probative value in determining his guilt:

‘(3) Where this subsection applies, the court or jury, in determining whether the accused is guilty of the offence charged, may draw such inferences as appear proper from the failure of the accused to give evidence or his refusal, without good cause, to answer any question.’

Immediately thereafter, in the same section, it was clarified that this provision does not oblige defendants to testify during the trial, and that a defendant who chooses to remain silent does not commit any offence by doing so:

‘(4) This section does not render the accused compellable to give evidence on his own behalf, and he shall accordingly not be guilty of contempt of court by reason of a failure to do so.’

Lord Taylor also held in R. v. Cowan [84], at p. 378, which was cited at length by the Appeals Court Martial, that the possibility of drawing incriminating inferences from the defendant’s silence does not violate the right to remain silent:

‘It should be made clear that the right of silence remains. It is not abolished by the section; on the contrary, subsection (4) expressly preserves it.’

The argument that this arrangement undermines the presumption of innocence was also rejected in that case, at p. 379:

‘Thus the court or jury is prohibited from convicting solely because of an inference drawn from the defendant’s silence… the burden of proving guilt to the required standard remains on the prosecution throughout. The effect of section 35 is that the court or jury may regard the inference from failure to testify as, in effect, a further evidential factor in support of the prosecution case. It cannot be the only factor to justify a conviction and the totality of the evidence must prove guilt beyond reasonable doubt.’

12. If I have discussed these matters at some length, it was because of the elevated status of the right to remain silent in our legal system. But from reading the aforesaid, there cannot be any doubt that the question whether it is possible to draw adverse inferences from the silence of a defendant has been decided in Israeli law, in s. 162 of the Criminal Procedure Law. This provision is unambiguous. It makes it possible to regard silence as corroboration or support for the prosecution’s evidence, and therefore it also constitutes the premise for our deliberations. Within the framework of this appeal we were not even asked to examine the legality or constitutional of this provision. Therefore I also see no reason to make a decision regarding the various approaches that exist on this issue. The question that we need to examine is therefore a narrow one. We need to examine whether it is right to interpret the provisions of s. 162 so that it allows us to regard the silence of a defendant not only as ‘corroboration’ and ‘support,’ but also as ‘something extra’ for a defendant’s confession, even though the provision does not expressly address this type of probative requirement.

As we shall clarify below, there are in my opinion good reasons for holding that the defendant’s silence should not be capable of constituting ‘something extra’ for his confession, unless the conditions that I shall discuss below are satisfied.

On defendants’ confessions

13. A defendant’s confession that he committed the offence attributed to him plays a central role in our laws of evidence. The reason for this is clear, since, as a rule, the defendant is the person who knows better than anyone else whether the charges levelled at him are true. Moreover, the assumption is that in the ordinary course of events a person does not portray himself as a criminal, in the sense that a person does not confess to an offence that he did not commit, and therefore there is a nucleus of truth in what he says. This is why the legal system attributes prima facie credibility to defendants’ confessions. This approach has led to a situation in which the interrogations of defendants have often focused on attempts to extract confessions that implicate them in criminal acts, and these were subsequently, during the trial, submitted as incriminating evidence against them. This has also led to the fact that in our legal system it is possible to base a conviction of a defendant on a confession that he made during his trial alone (see s. 154 of the Criminal Procedure Law).

Moreover, the legal system practised in Israel makes it possible to submit in evidence confessions that were made by the defendant even during his interrogation, as an exception to the rule prohibiting hearsay. The assumption is that such a confession goes against the interests of the person making it, and it may be assumed that he would not accept responsibility for an act that he did not commit (see also P. Murphy, Murphy on Evidence (Oxford, ninth edition, 2005), at p. 258). Notwithstanding, the status of these confessions is different from that of confessions made by the defendant during his trial, and they must be supported by ‘something extra’ in order to bring about the conviction of a defendant. The idea underlying this requirement is that confessions that come before the trial court in written form were not made before the watchful eyes of the courts and in such circumstances it is not possible to form an impression of the circumstances in which they were made, nor is it possible to examine the person who made the confession, which makes it difficult to evaluate the weight that should be given to such evidence (CrimA 556/80 Ali v. State of Israel [18], at pp. 185-186). This difficulty is further increased in view of the experience that has been accumulated in our legal system, as well as in foreign legal systems, which has shown that in reality there are indeed cases in which persons under interrogation have admitted committing offences that they never committed at all. The outlook that may have been accepted in the past — that a person never accepts responsibility for acts that he did not commit — is no longer accepted as absolutely correct. Regrettably, even in Israel there have been several cases in which convictions that were based on confessions made by suspects during their interrogations have been found to be erroneous (RT 1966/98 Harari v. State of Israel [19]; RT 3032/99 Baranes v. State of Israel [20]). In order to deal with these phenomena, a commission was appointed, with Justice E. Goldberg as chairman, and this commission published its conclusions in 1994 (see the report of the Commission concerning Convictions Based Solely on a Confession and concerning the Grounds for a Retrial (1994) (hereafter — ‘the Goldberg Commission report’)). On this subject we have also been blessed with thought-provoking literature (see M. Kremnitzer, ‘Conviction on the Basis of a Confession — Is There a Danger in Israel of Convicting Innocents?’ 1 HaMishpat 205 (1993); U. Struzman, ‘Protecting the Suspect against False Confessions,’ 1 HaMishpat 217 (1993); A. Bendor, ‘Taking a Defendant’s Confessions and its Admissibility — Purposes, Methods and What Lies In-Between,’ 6 Israel Journal of Criminal Justice (Plilim) 245 (1996); D. Dorner, ‘The Queen of Evidence v. Tarak Nujeidat — On the Danger of False Confessions and How to Deal with It,’ 95 HaSanegor 5 (2005); B. Sangero, ‘The Confession as a Basis for a Conviction — “Queen of Evidence” or Empress of False Convictions,’ 4 Alei Mishpat 245 (2005)).

14. There are many factors that lead to false confessions, but they may be classified into two categories. The main concern relates to false confessions made by suspects who were subjected to improper interrogation practices, such as violence, physical coercion or unfair psychological pressure, and who were induced by these to confess to something they had not done. As is well known, the arrangement that is designed to deal with circumstances such as these is found in s. 12 of the Evidence Ordinance, which provides that a confession made out of court will be admissible ‘only if the prosecutor brings testimony concerning the circumstances in which the confession was made and the court sees that the confession was free and voluntary.’ This section 12, which has been considered extensively in the case law of this court, is not relevant in the present case, in which it has not been claimed that the appellant’s confessions were not made of his own free will.

But that is not all. There is an additional concern that the defendant will make a false confession during his interrogation of his own free will, even in cases where the interrogation was conducted in a proper manner and without any of those external coercions that are addressed by the rule of admissibility in s. 12 of the Evidence Ordinance being used against him. This court has recognized in a whole host of cases the possibility that even subjective pressure and internal tension may result in the person under interrogation breaking down and ‘committing suicide’ in his confession (CrimA 48/54 Irshid v. Attorney-General [21], at p. 691). ‘It is the law that a confession is always accompanied by a concern that perhaps the person making the confession had a reason for taking responsibility for an act that he did not commit, even when the concern is unclear and not apparent’ (CrimA 6289/94 Janshvili v. State of Israel [22], at p. 176; see also CrimA 715/78 Levy v. State of Israel [23], at p. 234; CrimA 124/87 Nafso v. Chief Military Prosecutor [24], at p. 635 {266}; CrimA 3967/91 Mazon v. State of Israel [25], at p. 171; FH 3081/91 Kozali v. State of Israel [26], at p. 448).

The Goldberg Commission report listed three categories of factors that cause persons who are under internal pressure to make false confessions:

‘The first category contains those false confessions that are made because of the personality type of the person under interrogation. The person under interrogation does not distinguish between fantasy and reality, he thinks that by confessing he will “atone” for improper conduct in the past (whether true or imaginary), or he has a tendency towards self-destruction because of general and unfocused feelings of guilt, and he is one of those “troubled depressed persons that look forward to death, who stick knives in themselves and throw themselves from the rooftops’ (Maimonides, Hilechot Sanhedrin, 18, 6)…

The second category contains those false confessions that are made because of the influence of the interrogation or the arrest on the person under interrogation. This group includes persons under interrogation who, because of their inability to withstand the pressure of the interrogation from an emotional viewpoint, wish to put an end to it, sometimes in the belief that in the trial they will prove their innocence. There are cases where a person under interrogation is prepared for these reasons to confess to a less serious offence than the one of which he is suspected, out of considerations of short-term relief…

The third category contains those false confessions that are made because of social considerations and pressures: the desire to shield the true offender, family reasons (‘family honour’), solidarity with the true offender, belonging to the criminal classes and sometimes pressure from the real offender; the desire to become famous or notorious, or to be admitted into a criminal organization; taking the whole blame in order not to be labelled as an informer’ (ibid., at pp. 9-10).

The commission recognized that every person who is interrogated has his own personality type. Therefore persons who are interrogated respond and conduct themselves differently during the interrogation; ‘every person under interrogation has a personal “breaking point,” according to his personality and his ability to call upon emotional strengths in order to withstand conditions of pressure’ (ibid., at p. 8). The commission also warned that the danger that a defendant would accept responsibility for an offence that he did not commit was greater precisely for someone who was being interrogated for the first time, did not belong to the criminal classes and had no experience of arrest (ibid.). Legal and psychological literature has suggested a whole host of other possible reasons for false confessions that are unrelated to the use of improper interrogation practices, including the defendant’s distorted perception of reality; attempts to please the interrogators and win their sympathy; emotional or mental disabilities, etc. (R. Kitai-Sangero, ‘Silence as Admission: On the Erroneous Approach to Silence in the Court as Something Extra for a Confession in an Interrogation,’ 18 Law and Army 31 (2005), at pp. 38-39).

15. Admittedly, a false confession appears prima facie to be an irrational form of conduct. But we must reiterate that persons under interrogation are not homogeneous. Not all persons under interrogation act in the manner that we would expect a reasonable and thinking person to act. An interrogation places the person under interrogation in a threatening situation with which he is usually unfamiliar. This is inherently a coercive situation, which subjects the person under interrogation to many pressures, particularly because of the threat hovering over him that he may be found guilty and punished. It is not without justification that it was held that an interrogation, ‘even if it does not involve the use of physical measures, violates the liberty of the person under interrogation. It sometimes violates his dignity and his privacy’ (HCJ 5100/94 Public Committee Against Torture v. Government of Israel [27], at p. 831 {584}); ‘granting the authority to conduct a criminal investigation gives power, and it therefore involves a danger of a violation of the privacy, dignity, liberty and property of persons under interrogation’ (CrimA 4855/02 State of Israel v. Borovitz [28], at para. 48); elsewhere it was said that ‘any investigation, even if it is the most reasonable and fairest investigation of all, places the person under interrogation in embarrassing situations, burdens him, pries into his private concerns, invades his innermost recesses and places him under serious emotional pressure’ (Y. Kedmi, On Evidence (part 1, 1999), at p. 38). In view of all this, some authorities have gone so far as to claim that a false confession is in certain circumstances a normal reaction to an abnormal situation in which persons under interrogation are placed:

‘The false confession is not the product of a diseased mind, different in kind from a normal mind, but is simply an extreme manifestation of quite “normal” and understandable behavior’ (C.J. Ayling, ‘Corroborating Confessions: An Empirical Analysis of Legal Safeguards against False Confessions, 1984 Wis. L. Rev. 1121, at p. 1157).

For further discussion of the factors that cause false confessions, see G.H. Gudjonsson, The Psychology of Interrogations, Confessions and Testimony (Chichester, 1992); W.S. White, ‘False Confessions and the Constitution: Safeguards against Untrustworthy Confessions,’ 32 Harv. C.R.-C.L. L. Rev. 105 (1997), at p. 108. A similar position was expressed by Justice D. Dorner, in a minority opinion. According to her, a confession that was made in an interrogation is suspect evidence that should be regarded with great caution:

‘A defendant’s confession is suspect evidence, even if it was made without any external pressure being exerted on the defendant. The reason for this is that in the absence of other solid evidence that would prove the guilt of the defendant even without a confession, making a confession is in many cases an irrational act, and taking the irrational step of making a confession gives rise to a suspicion as to whether the confession is true. This suspicion is not merely theoretical, but it has been proved on more than once occasion by experience’ (CrimFH 4342/97 El Abeid v. State of Israel [29], at p. 836).

But as I have said, this position of Justice Dorner remained a minority opinion on that point, and confessions have always been and remain a very important tool for discovering the truth. A confession is ‘primus inter pares in the realm of evidence’ (per Justice M. Cheshin in El Abeid v. State of Israel [29], at p. 833). ‘Take confessions and testimonies from the law and you have removed the heart and soul of justice’ (per Justice T. Strasberg-Cohen, ibid. [29], at p. 855). ‘… the confession has a place of honour in the hierarchy of evidence in criminal law, and it is an important and accepted tool in those cases where the court is persuaded that the confession is a true one’ (per Justice E. Arbel in CrimA 6679/04 Steckler v. State of Israel [30], at para. 23). But it is important to clarify that we do not ignore the possibility that defendants will make false confession during their interrogations. On the contrary, we are aware of this danger and therefore we are enjoined to examine with great care the content of the confession before we base the defendant’s conviction on it. The outlook that ‘a person does not incriminate himself if he is innocent cannot be accepted as a legal axiom’ (in the words of the Goldberg Commission report, at p. 6). For this purpose case law has also developed two mechanisms that seek to prevent miscarriages of justices that may be caused as a result of relying on false confessions. We will now turn to consider these mechanisms.

Not relying on false confessions — the internal and external tests

16. There are two tests — an internal test and an external test — that are designed to contend with the fear that the defendant made false confessions in his interrogation. The first test is internal to the confession; according to this we evaluate the weight of the confession itself in view of the indications of truth that it reveals. The nature of this test was discussed by my colleague President Barak in one case, and among the ‘indications of truth’ he listed the logic or lack of logic in the version of events recounted by the defendant in his confession, ‘the clarity or confusion of the details contained in it and other similar indications of common sense that lead a rational person to trust what someone says,’ and also whether the story is ‘coherent, contains an internal logic of its own and is complete, or whether the story is confused, fragmented and has no logical consistency’ (Levy v. State of Israel [23], at pp. 234-235; see also Kozali v. State of Israel [26], at p. 458). In cases where the trial court finds, in reliance on these criteria, that the confession does not contain indications of truth, and that it should be given negligible weight, then it has been held that the defendant should not be convicted on the basis of the confession, and in the absence of other incriminating evidence this is sufficient to bring about the defendant’s acquittal. This situation did not exist in the appellant’s case. The District Court Martial found, and I do not think this conclusion should be changed, that the appellant’s confession should be given considerable weight:

‘We are dealing with a consistent confession from a chronological viewpoint; it is comprehensible, complete and logical. The defendant incorporates his descriptions of events that happened from the time he was 16 until the time of the interrogation… Moreover, his confession is very detailed; it gives a detailed description of the manner in which he used the drugs, including an explanation about the various methods of using cannabis. The defendant explained his feelings after using the drugs, the reasons that led him to use them during his military service, and finally he even expressed regret for what he did… The fear of a false confession is also lessened because the defendant chose in his confession not to mention the names of the persons who supplied him the drugs as well as the name of the person who used them with him… From this we learn that the defendant’s replies during his interrogation were well thought out and there is nothing in them to indicate that they were made casually or as a result of any hidden internal motive or as a result of pressure’ (p. 4 of the verdict).

17. But this determination is insufficient, since, as I have said, there is another mechanism for examining the confession. This other mechanism is external to the confession, and according to it we are required to strengthen the defendant’s confession with an additional probative requirement of ‘something extra.’ This requirement was introduced in our legal system in its earliest days (CrimA 3/49 Andelersky v. Attorney-General [31]), and over the years this court has reiterated its importance as a means of contending with the fear of false confessions. In addressing the nature of this requirement of ‘something extra,’ Justice S.Z. Cheshin held that it imposes upon the prosecution authorities ‘a duty to show why we should rely on the confession, i.e., what are the circumstances that require the conclusion that the confession is correct’ and also that it is designed to show ‘that the defendant is a person who had an opportunity to do the felonious act to which he confessed before he was brought to trial’ (ibid. [31], at p. 593). It was also held that the requirement of ‘something extra’ did not need to prove the fact that the offence itself was committed or to identify the defendant as its perpetrator, and any evidence, whether direct or circumstantial, would suffice to authenticate the confession (CrimA 290/59 A v. Attorney-General [32], at p. 1499; CrimA 543/79 Nagar v. State of Israel [33], at pp. 141-142). In this way our legal system differs from the American legal system. Admittedly, there too the courts require confessions made out of court to be supported by additional authenticating evidence, but the requirement there is that this support should be independent and substantial evidence that also relates to the corpus delicti of the offence and nothing less than this (see Opper v. United States [82]; see also Ayling, ‘Corroborating Confessions: An Empirical Analysis of Legal Safeguards against False Confessions, supra, at pp. 1126, 1145-1152).

18. The requirement of ‘something extra’ is therefore intended to authenticate the version of events that the defendant told in his interrogation. It is intended to remove the doubt that ‘perhaps for hidden reasons known only to the person making the confession, he chose to admit to an act that he did not do’ (CrimFH 3391/95 Ben-Ari v. State of Israel [34], at pp. 450-451). For this reason we do not satisfy ourselves with an examination of the weight of the confession. We therefore require an ‘external’ addition to the confession, since the ‘something extra’ is ‘the only objective criterion available to the court for examining the truth of the “subjective” confession of the defendant’ (ibid. [34], at p. 451). This objective criterion also emphasizes the great importance of the ‘something extra’ as a means of protecting the defendant against false confessions. After all, examining the weight of the confession (‘the internal test’) is hardly unique to the fear of false confessions, since the courts are required to determine the proper weight of every piece of evidence that comes before them, which they do on a regular basis (s. 53 of the Evidence Ordinance). From this we see the great importance of the requirement of ‘something extra’ (‘the external test’) as a mechanism that is unique to examining the credibility of confessions not made under the watchful eye of the courts. The Goldberg Commission report also recognized the great importance of the requirement of ‘something extra,’ and this was reflected in the commission’s recommendation to enshrine this requirement in statute, in the following language:

‘(a) A statement of a person outside the court may be admitted as evidence against him in a criminal trial.

…

(e) A defendant shall not be convicted on the basis of a statement admitted under subsection (a) unless the evidence before the court contains additional external evidence that constitutes corroboration, support or something extra which in the circumstances of the case is capable, in the opinion of the court, of removing any doubt as to the credibility of the statement in so far as the commission of the offence is concerned’ (Goldberg Commission report, at pp. 21-22; emphasis supplied).

Prof. Mordechai Kremnitzer, who was one of the members of that commission, was of the minority opinion that ‘something extra’ should not be sufficient for a defendant’s confession, and this requirement should be replace by the stricter requirement of ‘corroboration’ for the confession (see p. 64 of the Goldberg Commission report). Later a private bill was tabled with the aim of adopting the position that requires the defendant’s confession to be supported by corroborating evidence, but this did not become law (see the draft Evidence Ordinance Amendment (Requirement of Corroboration for a Conviction on the Basic of a Confession) Law, 5764-2004).

19. By contrast, there were some authorities who questioned whether the requirement that the confessions of persons under interrogation are supported by ‘something extra’ is justified, since in their opinion it is possible to base the conviction on this confession only, without any need for authenticating evidence (see the article of E. Harnon, ‘The Need for “Something Extra” to Convict a Person in accordance with a Confession Made out of Court,’ 28 HaPraklit (5732) 360, and his position in his book The Law of Evidence (vol. 2, 1977), at pp. 282-286; a similar position was expressed also by Justice Ben-Porat in Ali v. State of Israel [18], at p. 182). I will also add that even according to the legal system practised in England it is possible to base a conviction of the defendant solely on a confession that he made during interrogation, and there is no express requirement for any additional authenticating evidence for this purpose (D.B. Griffiths, Confessions (Edinburgh, 1994), at p. 116; see also J.H. Wigmore, Evidence in Trials at Common Law (Boston and Toronto, vol. 7, rev. by J.H. Chadbourn, 1978), at pp. 508-510). Learned counsel for the appellant believes that this credibility that the English legal system gives to confessions made in interrogations, and its willingness to rely on them alone, is also related to the fact that the legal system practised in England has recognized the right of the defendant to receive legal advice at every stage of the investigation and has introduced requirements to make audio recordings of interrogations and other such rules that are intended to protect persons under interrogation and are enshrined in the Police and Criminal Evidence Act of 1984 and regulations enacted thereunder (for a comprehensive survey of the rights and arrangements enshrined in that law, see M. Zander, The Police and Criminal Evidence Act 1984 (London, fifth edition, 2005)). This argument is logical. It is possible that these rules, which introduced close supervision of the manner in which interrogations are conducted, allow the courts to form a better impression during trials of the confessions made by persons under interrogation, which reduced the fear of false confessions (see, in this regard, Easton, The Case for the Right to Silence, at pp. 105-127). Whatever the case, the approach that regards a confession on its own as sufficient has not been adopted in our legal system, and the requirement of ‘something extra’ for a confession that the defendant made in his interrogation remains intact.

20. The requirement of ‘something extra’ is flexible and open-ended. What will satisfy it varies from case to case and depends also on the credibility of the confession itself. The greater the weight of the confession, the smaller the weight of the ‘something extra’ that is required to authenticate the confession, and vice versa, the smaller the weight of the confession, the greater the weight required for the ‘something extra.’ It has also been held, therefore, that it is possible that there may be cases in which the ‘something extra’ will only need to be ‘as light as a feather’ (CrimA 178/65 Usha v. Attorney-General [35], at p. 156; CrimA 428/72 Ben-Lulu v. State of Israel [36], at p. 270; El Abeid v. State of Israel [29], at p. 834). On the other hand, it has been held that there may be cases in which the weight required for the ‘something extra’ will be so great that it will turn into a requirement of ‘corroborative evidence’ (Ben-Ari v. State of Israel [34], at p. 449). For this reason my colleague President Barak has said that ‘evidence that is fit to be used as “something extra” in one case may not be considered fit to be used as “something extra” in another case, since it all depends upon the circumstances of the case’ (Levy v. State of Israel [23], at p. 234).

The evidence that has been found, in the specific circumstances of individual cases, to be capable of satisfying the requirement of ‘something extra’ include the following: the lies of the defendant on a major matter; implicating conduct of the defendant after the criminal act; the failure of an alibi argument (Janshvili v. State of Israel [22], at pp. 175-176); the knowledge of secret details — details of which the person making the confession could not have had any knowledge unless he was involved in the criminal act; in this regard it was held that:

‘The greater the number of secret details that the defendant incorporated in his confession, the smaller the fear that his knowledge of these does not derive from the commission of the offence but from hints suggested to him unconsciously by his interrogators when he was interrogated and that were picked up by him’ (CrimA 4769/92 Nijam v. State of Israel [37], at para. 12).

The requirement of ‘something extra’ may also be satisfied by one of the following: expressions that indicate a feeling of guilt on the part of the defendant (CrimA 7595/03 A v. State of Israel [38], at pp. 11-12); proof that shows he was present at the scene of the crime when it was committed, when there is no satisfactory explanation for this (CrimA 5225/03 Habbas v. State of Israel [39], at p. 32); a reconstruction of the crime by the defendant, when he describes in detail the sequence of events in a manner that is consistent with the findings from the scene of the crime (CrimA 3338/99 Pakovitz v. State of Israel [40], at para. 19). It need not be said that when there is additional and independent incriminating evidence against the defendant, the requirement of ‘something extra’ is also satisfied.

We therefore need to consider the question whether the silence of the defendant during the trial may be added to this list and may also constitute ‘something extra’ for his confession in his interrogation. In my opinion, as I have already implied above, the answer to this question is no, unless conditions that I will set out below are satisfied. I will now explain my reasons for this.

The silence of the defendant in a trial as ‘something extra’ for his confession during interrogation

21. The question whether it is right to regard the silence of the defendant as ‘something extra’ for his confession during interrogation has not yet been decided by this court. But the question arose more than twenty years ago in Ali v. State of Israel [18]. In that case the justices hearing the case expressed their opinion on this issue briefly, but their remarks were made in obiter dicta since in the circumstances of that case there was additional evidence against the defendant that was capable of satisfying the requirement of ‘something extra’ for the confession. Justice Ben-Porat expressed the opinion that the silence of the defendant in his trial constitutes ‘something extra’ for his confession, since ‘when the legislature determined [in s. 162 of the Criminal Procedure Law] that the silence of the defendant amounts to “corroboration,” it thereby made silence an independent and separate piece of evidence in support of the prosecution’s case’ (ibid. [18], at p. 182). The other justices hearing the case, Justices Elon and Shiloh, disagreed with this position. This is how Justice Elon expressed the matter:

‘It is right and proper that the court should not regard the silence of the defendant in court as support for his confession that he made out of court. Admittedly a confession made by the defendant in court is sufficient on its own in order to convict him, but an express and clear confession made in court before the judge cannot be compared to the silence of the defendant and his refusal to testify. In the former case the judge can be presumed to understand and hear from the defendant’s statements that his confession is a real one and that it is not being made by him in order to take responsibility for something he did not do; but it is otherwise in the latter case, where the defendant remains silent, since the judge does not as a rule have any indication or basis from which he may infer the reason and background for the silence’ (ibid. [18], at p. 185).

In later cases that came before the court the issue under discussion in this appeal did not arise again, and it remained undecided (see Muadi v. State of Israel [15], at p. 234).

22. When we seek for a solution to the question that arises in this case, we cannot merely refer to the provisions of s. 162 of the Criminal Procedure Law. We have seen that this section, which allows the silence of the defendant to be regarded as support or corroboration for incriminating evidence, does not expressly mention ‘something extra.’ One possible interpretation of this, which was adopted by the courts martial, is to extend the scope of the provision and to allow it to regard silence also as ‘something extra.’ I will admit that this interpretation is logical: the law is that the requirement of corroboration is satisfied only when three cumulative conditions are fulfilled — that it is independent and separate from the evidence requiring corroboration; that the corroborating evidence implicates the defendant in the commission of the offences with which he is charged; and, finally, the corroboration relates to a significant point that is in dispute (CrimA 387/83 State of Israel v. Yehudai [41], at p. 203; CrimA 2949/99 Cohen v. State of Israel [42], at p. 645; CrimA 1538/02 A v. State of Israel [43], at p. 598). As I have shown, the requirement of ‘something extra’ may be satisfied even when these conditions are not fulfilled. In case where the confession has consideration weight and it contains many indications of truth, the trial court may be satisfied with authenticating evidence that has very little weight. This gives rise to the claim that the stricter requirement includes the more lenient requirement, and what may satisfy the requirement of corroboration can also constitute ‘something extra’ for the defendant’s confession. But this interpretation is not free of doubt, since if you say that the legislature wanted to regard the silence of the defendant also as ‘something extra,’ you are compelled to wonder why it did not state this expressly within the framework of s. 162 of the Criminal Procedure Law? Indeed, the possibility that the silence of the legislature in this matter did not arise from an omission and was not caused by inadvertence is a very reasonable possibility in view of the fact that at the time when the law was amended and this provision was added to it in 1976, the requirement of ‘something extra’ already existed. This also leads to the conclusion that the question whether the silence of the defendant should be regarded as ‘something extra’ for his confession cannot be determined solely on the basis of s. 162. It is an ethical question, which also requires us to consider the weight of confessions in our legal system and the reason why they need to be supported by additional authenticating evidence. Moreover, the legal answer to the question that arises in the case before us cannot be derived only from the technical definitions that were given to the terms ‘support’ and ‘corroboration’ in the aforesaid s. 162, and before we reach the correct result we should take into account the purpose underlying the arrangements and the practical ramifications of the possible legal outcomes. This is what my colleague President Barak meant when he said that ‘the law is a social tool. Legal concepts are intended to realize social goals. They are an instrument for achieving social objectives. They are an expression of the proper balances between conflicting values and interests’ (HCJFH 4601/95 Serrousi v. National Labour Court [44], at para. 7). Elsewhere my colleague the president added that —

‘We should distance ourselves from a jurisprudence of concepts, according to which the theoretical concept forces itself upon interests and values that require a normative arrangement. We should aspire to a jurisprudence of values, according to which the theoretical balance is the result of balancing and arranging the interests and values that require a normative arrangement’ (CrimFH 4603/97 Meshulam v. State of Israel [45], at para. 18).

23. I have discussed how the requirement of ‘something extra’ is intended to help the court identify false confessions. The main concern that the requirement is intended to address is the concern of false confessions that are made because the defendant was under internal pressure or had a hidden motive during his interrogation. Admittedly, external pressure or some other improper conduct that was directed at the defendant in his interrogation may also lead to false confessions, but in order to deal with these concerns there are other special arrangements, some that have been provided in statute — s. 12 of the Evidence Ordinance — and others that have been developed in case law, such as the rule that was determined in Yissacharov v. Chief Military Prosecutor [5].

We have seen that the motives and factors that lead to the creation of internal pressure may be hidden and to a bystander the course of action of making a confession may seem unreasonable. It is precisely because of this that I am of the opinion that it would not be right to rely on the silence of the defendant — the meaning of which is uncertain — in order to dispel the concerns that arise. The court martial was of the opinion that the defendant’s plea of not guilty in his trial shows that he is no longer subject to the internal pressure that led him to confess during his interrogation to an act that he did not commit. In such circumstances the court martial held that the defendant can be expected to explain what led him to confess to the acts and to defend himself in a positive manner against the charges levelled at him. I cannot accept this outlook. Just as we cannot accept the argument that a person does not confess to offences that he did not commit as a legal axiom, so too we cannot rely on the argument that a defendant who denies the charges is not subject to any hidden pressures. On the contrary, it is possible that precisely because such internal pressures also exist during the trial, the defendant remains silent and does not enter the witness box. Who can guarantee that the same hidden motive that caused the defendant to take responsibility during his interrogation for an act that he did not commit will not return and determine the line of defence that he will adopt during the trial, when he chooses the right to remain silent? It will be remembered that even the respondents in their reply recognized this possibility that ‘a defendant who took responsibility during a police interrogation for committing an offence that he did not commit will deny the charges but choose not to testify in his own defence’ (para. 44 of the attorney-general’s reply). Admittedly it is possible that such a situation will occur only in exceptional circumstances. It is also possible that a reasonable defendant, who acts rationally, will try to prove his innocence by giving positive testimony. But in criminal law we are not dealing with the reasonableness or unreasonableness of the defendant’s conduct, but with the question whether he committed the offences attributed to him or not. There is no need to reiterate that in criminal cases we are dealing with matters of life and death, and that the danger of convicting an innocent person is great. This also leads to the fundamental idea that it is better to acquit ten guilty persons than to convict one innocent one. The requirement that the defendant’s confessions should be supported by ‘something extra’ is intended to prevent such consequences. Relying on the silence of the defendant as ‘something extra’ for his confession — when the motive for the confession and the motive for the silence are unclear — may undermine the purpose of that requirement, namely the desire to authenticate the defendant’s confession. Prof. E. Kamar expressed these concerns well when he said:

‘In practice, even that depressed defendant, who wishes to “commit suicide” in his confession, may be subject to an internal pressure that will prevent him from explaining to the court why he confessed. We cannot of course expect someone, who has just now regretted his intention to “commit suicide” in his confession, to stand before the court and announce the exact reasons why he made a false confession. The claim that the defendant’s plea of not guilty to the charges in court rules out the possibility that he is still under the influence of the factor that induced him to confess in the interrogation is logical in ordinary cases; but the requirement of something extra is intended to contend precisely with the exceptional cases, when the conduct of the defendant differs from the norm and is difficult to explain. In such cases it is hard to rely on the logical assumption that someone who denies the charges in the court is entirely free of the pressures that were sufficiently strong to induce him earlier to confess to something that he did not do’ (E. Kamar, ‘Corroborating Confessions by the Defendant’s Own Statements,’ 5 Israel Journal of Criminal Justice (Plilim) 277 (1996), at pp. 292-293).

It also cannot be denied that in those cases where the legislature allowed the defendant’s silence to be regarded as support and corroboration for the prosecution’s evidence, the silence serves as an addition to other testimonies that do not originate with the defendant. This is the case when the silence is an addition to the testimony of a state’s witness, an accomplice, a minor who testified before a child interrogator or a witness whose statement was submitted by the prosecution under s. 10A of the Evidence Ordinance. The situation is different when we are speaking of the defendant’s confession, since them his silence is an addition to his own testimony that he gave during the interrogation, and in such circumstances I am of the opinion that the silence cannot authenticate the confession to the required degree.

24. I have also reached this conclusion for another reason. In my opinion it is not desirable, as a matter of policy, to allow the prosecution authorities to file indictments when all they have in their files is the confession of the defendant, in the hope that when the defence case is presented, the defendant will choose to remain silent, thereby providing them with the ‘something extra’ that is required for supporting the confession. Allowing this might result in greater efforts being made to obtain a confession from the defendant where the investigation should also focus on other evidence that is external to the confession, whether material evidence or testimony, which also has an important role in revealing the truth. This court has emphasized in a whole host of cases the great importance of carrying out a thorough investigation, which exhausts the possible lines of investigation thoroughly, and in which all the relevant evidence is assembled, so that the truth will come to light (see CrimA 721/80 Turgeman v. State of Israel [46], at pp. 471-472; CrimA 10596/03 Bashirov v. State of Israel [47], at paras. 19-20; CrimA 5386/05 Alhorti v. State of Israel [48]). Prof. Kremnitzer has described these concerns well, even though his remarks were directly mainly at interrogations in which coercive measures and prohibited pressure are directed at the person under interrogation:

‘A legal system that allows a conviction on the basis of a confession, is tolerant of interrogation methods that involve pressure and usually allows defendants’ confessions encourages the investigative authorities to resort to and rely on investigation methods that revolve around the person under interrogation, instead of directing the investigative effort at assembling other evidence. This phenomenon has serious repercussions upon the general level of the police investigation, the image and character of the interrogators, their education and training and the interrogation ethos’ (M. Kremnitzer, ‘Conviction on the Basis of a Confession — Is there a Danger in Israel of Convicting Innocents?’ supra, at p. 215).

It may also be assumed that the greater the willingness of the legal system to rely on defendants’ confessions as the sole evidence for a conviction, the greater the efforts to obtain a confession during interrogations, despite the negative repercussions that may result from such a situation (see also in this context G.H. Van Kessel, ‘The Suspect as a Source of Testimonial Evidence: A Comparison with the English Approach,’ 38 Hastings L.J. 1 (1986), at pp. 111, 122; Kamar, ‘Corroborating Confessions by the Defendant’s Own Statements,’ supra, at p. 294; Sangero, ‘The Confession as a Basis for a Conviction — “Queen of Evidence” or Empress of False Convictions,’ supra, at pp. 265-274; O. Bassok, ‘ “Hard Cases Make Bad Law and Bad Law Makes Hard Cases” — On the Ramifications of the Yosef Ruling,’ 18 Mishpat uTzava (Law and Army) 77 (2005), at pp. 126-136).

25. At the same time, it is possible to conceive of cases in which the defendant’s confession is a true one, but additional external evidence supporting this confession does not exist or cannot be found. In these cases, when it becomes clear to the trial court that the confession was not made as a result of any internal pressure, there is a concern that it is the defendant’s acquittal — for the reason that his silence does not constitute ‘something extra’ for his confession — that will lead to a failure to arrive at the truth, and this will be also undermine the need to fight the increasing rate of crime and to protect the public from the danger of lawbreakers. This danger, it need not be said, increases when we are speaking of serious offences. In view of this, we should consider whether there are circumstances in which the trial courts have a means of effectively supervising the way the interrogation is conducted, so that it will be possible to obtain a proper impression of the defendant’s conduct and the actions of his interrogators. Here we should emphasize that the more transparent the interrogation, the easier it will be to form an impression of the circumstances in which the confession was made, and the less justification there will be for making a distinction between a confession made during the investigation and a confession made in court, which, it will be remembered, is sufficient on its own for convicting the defendant. Because of this argument there have been calls for the need to make video or audio recordings of the interrogations of suspects, and not merely to make a verbal record of them, in order to create a way in which the courts can see what is happening in the interrogation rooms, despite the (mainly budgetary) difficulties that this creates. The Goldberg Commission also recommended introducing a requirement that video recordings of interrogations should be made and it emphasized that such recordings would also result in defendants no longer being able to raise unfounded complaints about the manner in which the interrogation was conducted (see pp. 30-35 of the report). Others have emphasized that a video recording of the interrogation will not only allow the courts to supervise the interrogation practices being used, but it will also allow them to receive an impression of the interaction between the interrogator and the person under interrogation and other factors that are not contained in a verbal record (see White, ‘False Confessions and the Constitution: Safeguards against Untrustworthy Confessions,’ supra, at pp. 153-156). The importance of the video recording was discussed by Justice Goldberg more than twenty years ago:

‘The recording does not merely preserve accurately the verbal part of the statement but also the factors that accompany it (the tone in which the statement was made, hesitation, anger, etc.) and also what was said by the person speaking to the person making the statements, and in this way it “speaks for itself.” The court is thereby given additional tools to see for itself the credibility of the witness and the correct meaning and truth of what he says’ (CrimA 323/84 Shriki v. State of Israel [49], at p. 517).

In 2002 these calls were answered by the legislature, which introduced, in s. 7 of the Criminal Procedure (Interrogation of Suspects) Law, 5762-2002, a requirement that video recordings of interrogations should be made when there was a suspicion that serious offences carrying a sentence of ten years imprisonment or more had been committed. This provision, which because of various delays has not yet come into effect, should in the future significantly improve the ability of the courts to assess the nature of the confessions brought before them. It will allow the courts to form the closest possible impressions of the true circumstances in which the confession was made, and it will provide better protection for defendants. The great importance of this provision can also be seen from what is stated in the explanatory notes to the draft law:

‘The proposed amendment will also allow the court to discover what part of the confession was said by the defendant of his own free will and of his own initiative, and what was said to the interrogators in response to their questions, how it was said and how the suspect appeared when making the confession. The court will also be able to decide whether the defendant was “led” by the interrogator’s questions to answer as he did or whether he was only asked clarifying questions, and it will also be possible to obtain an impression from the defendant’s conduct and the way in which he speaks of his state of mind and his understanding of his interrogators’ questions. The video recording will prevent a situation in which the interrogator records an erroneous reply because of a misunderstanding and the suspect signs his confession without understanding it or without it having been read to him.

The proposed amendment constitutes an improvement in the protection of the rights of the defendant, and it prevents or reduces the possibility of mistakes, misunderstandings or deliberate errors, and thereby improves the ability of the court to arrive at the truth on a question that is so central to the criminal trial’ (Explanatory Notes to the draft Evidence Ordinance Amendment Law (no. 15) (Confession of a Defendant on Serious Offences), 5761-2000 (Draft Laws 2928, 30 October 2000, at p. 54).

It need not be said that the fear of a miscarriage of justice caused as a result of a conviction based on a false confession is smaller when a video recording of the interrogation is made and the recording is brought before the trial court, so that it can obtain an almost direct impression of the manner in which the interrogation was held and of the conduct of the defendant and his interrogators.

26. In view of this, I am of the opinion that whereas it would be right to determine that as a rule the silence of a defendant should not constitute ‘something extra’ for a confession that he made during his interrogation, it would not be right to determine such a sweeping rule with regard to interrogations where a video recording was made, so that the court is given the possibility of watching the interrogation and the defendant’s confession. This conclusion is also capable of striking a proper balance between the desire to protect the rights of the defendant and the desire to arrive at the truth, bring lawbreakers to justice and protect the public from them.

Here we should emphasize that this result does not lead to the conclusion that in every case where a video recording of the defendant’s interrogation is made, his silence should be regarded as ‘something extra’ for the confession that made during his interrogation. The question of the significance that should be attributed to the silence of the defendant is a matter for the trial court to decide in accordance with the special circumstances of the case before it and in accordance with its discretion. The video recording of the interrogation is therefore an essential but not a sufficient condition for regarding the silence of a defendant in a trial as ‘something extra’ for his confession in the interrogation. Thus, for example, where an examination of the confession itself shows that the indications of truth in it are not many, or where the silence of the defendant is given a satisfactory explanation that is consistent with his innocence, then the silence of the defendant ought not to be regarded as ‘something extra’ for his confession. In addition, where there is a disparity between facts stated by the defendant in his confession and objective findings from the scene of the crime and where, despite the video recording of the interrogation, the trial court for various reasons has difficulty in forming an impression of the circumstances in which the confession was made, it would be right to require for such a confession some external authenticating evidence and not merely the silence of the defendant in order to dispel the concerns that arise. This list is not exhaustive. There may be other examples. The important thing is that the court should exercise great caution before it reaches the conclusion that the silence of the defendant may be regarded as the additional evidence that is required for the confession he made during his interrogation.

Conclusion

27. The silence of the defendant may — but does not necessarily — constitute ‘something extra’ for the confession he made during his interrogation, provided that a video recording of the interrogation was made and the court that saw the recording of the interrogation received the impression that it has before it a true confession. When one of these two conditions is not satisfied, the silence of the defendant cannot constitute ‘something extra’ for his confession.

In my opinion this outcome sets out the ideal approach in this matter. Notwithstanding, I should emphasize that the practical ramifications of this decision will not lead to a far-reaching change in the criminal law or to any difficulty in convicting guilty persons in so far as the civil courts are concerned, since, as the respondents admit, in the prosecution of civilians the practice is not to file indictments against someone where the only incriminating evidence against him is his confession in the interrogation. The source for this practice is the attorney-general’s guideline no. 53.000 of 1970. It is not clear to me why the military prosecution failed to comply with this and I see no reason why the military courts should be allowed discretion to regard the defendant’s silence as ‘something extra’ for his confession where the attorney-general himself saw fit to rule out such a possibility.

In view of the aforesaid, and in view of the fact that no video recording of the appellant’s interrogation was made, I would recommend to my colleagues that we order his acquittal on the offence of the use of a dangerous drug.

Justice E. Arbel

A confession that was made by a defendant out of court is admissible evidence, if it satisfies the conditions set out in s. 12(a) of the Evidence Ordinance [New Version], 5731-1971 (hereafter: ‘the Evidence Ordinance’), namely that it was made ‘freely and willingly.’ Once the confession has overcome the admissibility ‘barrier,’ it may be used to convict someone as the sole evidence against him, if it satisfies two tests: the confession’s internal indications of truth test (‘the internal test’) and the ‘something extra’ test, which is capable of authenticating the content of the confession (‘the external test’).

1. At the heart of the appeal lies the question whether a defendant’s silence in his trial may constitute ‘something extra’ for the confession that he made during his interrogation, namely the extra evidence that is required in order to support the truth of the confession.

Deciding this question requires a delicate balance between several values. First and foremost we have before us the value of discovering the truth in a criminal trial, which realizes the main goals of the criminal law, namely fighting crime, safeguarding the public and protecting the rights of actual and potential victims of crime. The need to further these goals becomes more urgent as the rate and sophistication of crime increase. Of no less importance than the value of discovering the truth is the right of the defendant to a fair and proper trial, which guarantees that justice is done and false convictions are avoided. A proper balance between these serves to further the supreme purpose of criminal law — the punishing of the guilty and the acquittal of the innocent — and thereby ensures public confidence in the ability of the judiciary to do justice to the individual and to society as a whole (see the remarks of Justice Beinisch in CrimA 5121/98 Yissacharov v. Chief Military Prosecutor [5], at para. 43 {383} and the remarks of Justice Procaccia in HCJ 11339/05 State of Israel v. Beer-Sheba District Court [7], at para. 11. In this balance the interest of exhausting probative methods of discovering the truth competes with the right of the defendant to a fair trial and the desire to avoid a miscarriage of justice by convicting the innocent or acquitting the guilty. We do not have a ‘magic solution.’ In each case we need to consider the competing values and find a proper and fair solution according to its concrete circumstances.

2. The various rules that require additional evidence in certain circumstances for a criminal conviction — in our case the ‘something extra’ for a defendant’s confession — further the purposes of discovering the factual truth and protecting the right of the defendant to dignity and liberty. In this respect both of these purposes complement one another — they ‘both serve the supreme purpose of the law that concerns the doing of justice and preventing miscarriages of justice in their broadest sense’ (Yissacharov v. Chief Military Prosecutor [5], at para. 45 {385}). But in certain circumstances these purposes may tip the scales in opposite directions, and there is therefore a need to strike a balance between them.

3. My colleague Justice Levy examined the fundamental question before us thoroughly and comprehensively, and he reached the conclusion that the answer that should be given to this question is no, unless the conditions set out in his opinion are satisfied. Justice Levy concludes his opinion with the following words:

‘The silence of the defendant may — but does not necessarily — constitute “something extra” for the confession he made during his interrogation, provided that a video recording of the interrogation was made and the court that saw the recording of the interrogation received the impression that it has before it a true confession. When one of these two conditions is not satisfied, the silence of the defendant cannot constitute “something extra” for his confession’ (para. 27 of Justice Levy’s opinion).

I agree with my colleague’s position that the silence of a defendant may — but does not necessarily — constitute ‘something extra’ for a confession that he made in his interrogation, and like him I too am of the opinion that as a rule we should refrain from regarding the defendant’s silence in his trial as ‘something extra.’ But my position differs from my colleague’s position on two main issues: one is the statutory premise for examining the question, and the other concerns the exceptions to the aforesaid rule which should be recognized within the scope of judicial discretion and on the basis of proper judicial policy.

4. My position in essence is that s. 162 of the Criminal Procedure Law [Consolidated Version], 5742-1982 (hereafter: ‘the Criminal Procedure Law’ or ‘the law’) makes it possible to hold that the silence of a defendant in his trial may serve as ‘something extra’ for a confession that he made during his interrogation. At the same time, by virtue of proper judicial policy that is derived from a whole range of considerations and interests that are relevant to the issue, I agree with my colleague’s position that in general the silence of a defendant in his trial should only constitute ‘something extra’ for a confession that he made in his interrogation in exceptional cases. But whereas my colleague is of the opinion that the exception should only apply to interrogations where a video recording was made, my position is that the trial court should be allowed a broader margin of discretion, by determining criteria for the restrained and careful manner in which it should be exercised.

My colleague set out in his opinion a comprehensive examination of fact and law, and therefore I will restrict myself to explaining my position by clarifying how my position differs from his.

Background — on defendants’ confessions in general and on the fear of false confessions in particular

5. The weight that should be given to defendants’ confessions, among the other evidence that is recognized for the purpose of convicting someone in a criminal trial, has arisen repeatedly over the years in legal deliberations and academic and social debate. This important debate, like the experience and the knowledge that have been accumulated over the years with regard to the various factors that cause defendants to make false confessions, have gradually led to changes and fluctuations in accepted outlooks concerning the status of the confession in our legal system (report of the Commission concerning Convictions Based Solely on a Confession and concerning the Grounds for a Retrial (1994), which was chaired by Justice E. Goldberg (hereafter — ‘the Goldberg Commission report’); see also the various opinions that were expressed on this issue in CrimFH 4342/97 El Abeid v. State of Israel [29]; M. Kremnitzer, ‘Conviction on the Basis of a Confession — Is there a Danger in Israel of Convicting Innocents?’ 1 HaMishpat 205 (1993); B. Sangero, ‘The Confession as a Basis for a Conviction — “Queen of Evidence” or Empress of False Convictions,’ 4 Alei Mishpat 245 (2005)). I am aware of the various concerns that accompany the making of confessions. They cannot be ignored. But I am also of the opinion that they should not be exaggerated. Confessions were and remain a main tool of the legal system in arriving at the truth and in bringing criminals to justice, if only for the simple reason that, as President Barak said: ‘Sometimes only the defendant who confesses can provide a coherent and detailed version of the sequence of events’ (El Abeid v. State of Israel [29], at p. 865). The probative importance of the confession has been recognized in case law that has described its centrality in the process of discovering the truth: ‘Take confessions and testimonies from the law and you have removed the heart and soul of justice’ (ibid. [29], at p. 855). In this regard I shall mention once again what I said in a previous case:

‘… Since the Goldberg Report the splendour of the confession as the “queen of evidence” has indeed lost its sheen, and a spotlight has been directed at the concern that innocent people may be convicted. It remains my opinion that the confession has a place of honour in the hierarchy of evidence in criminal law, and it is an important and accepted tool in those cases where the court is persuaded that the confession is a true one. I also reject the approach that, as a premise, every confession contains an inherent “defect” (as Justice Strasberg-Cohen describes the approach of Justice Dorner in El Abeid v. State of Israel [29], at p. 854, an approach that she did not accept) of being a false confession and inherently suspect evidence, since it is prima facie tainted with irrationality’ (CrimA 6679/04 Steckler v. State of Israel [30], at p. 1664).

6. This position is mainly based on experience that teaches us that a voluntary confession made by a defendant during his interrogation with regard to an offence that he did not commit is the exception and not the rule, and that although it cannot be said that there are no cases in which people confess to deeds that they have not committed, when confessions are not the result of improper means that were used to obtain them, usually people confess to deeds that they have committed rather than deeds that they have not committed (see El Abeid v. State of Israel [29], per Justice Or at p. 819, per Justice Strasberg-Cohen at p. 856, per Justice Mazza at p. 857 and per President Barak at p. 865). The remarks of Justice Cheshin in this regard are apt:

‘The confession of a defendant in a police interrogation has sometimes been called the “queen of evidence.” Even if it not a reigning sovereign, it is a “princess of evidence,’ one of the most important types of evidence in the kingdom. Who know this better than we, who sit in judgment on a daily basis? And we have long known also that experience and practical knowledge guide us on our path. Oliver Wendell Holmes, one of the wisest of our community — the community of legal scholars — taught us that “The life of the law has not been logic; it has been experience” (O.W. Holmes, The Common Law (Boston, 1881), at p. 1). Usually a person does not portray himself as a criminal, i.e., a person does not confess to committing a certain offence unless he has committed that offence. Wisdom teaches us that a person does not portray himself as a criminal when he is a saint. This is particularly true when we are speaking of the most serious offences, those offences that send a person to prison for many years. This is not a legal assumption. Nor is it an absolute presumption. It is a supposition of common sense, an assumption that is based on human nature’ (El Abeid v. State of Israel [29], at pp. 833-834).

7. Moreover, experience also teaches us that there are weighty reasons for making a confession that is not a false confession. These reasons were described, albeit not in full, by Justice Or in that case:

‘A weighty reason is the basic emotional need of someone who committed an offence to divest himself of the burdensome feeling of guilt that he can no longer carry inside him. Another possible reason is the rational recognition of someone suspected of an offence — for example, when there is other evidence such as testimony against him or physical evidence — that his failure to cooperate may not help prevent him from being brought to trial, whereas cooperation with the authorities may result in a certain credit being given to him at a later date, such as at the sentencing stage. And I have not mentioned all of the reasons’ (ibid. [29], at p. 819).

8. Although false confessions are an exception to the rule, the legal system has long recognized that various pressures will lead a defendant to confess to an offence that he did not commit, ‘in the sense of “committing suicide” in his confession’ (CrimA 48/54 Irshid v. Attorney-General [21], at p. 691):

‘There are two concerns that relate to a confession that a defendant made out of court: one is the concern that “external” pressure was exerted on the defendant and this led him to confess to committing an act that he did not do. This pressure is addressed by the statutory barrier of admissibility, which is provided in s. 12 of the Evidence Ordinance [New Version] and which renders inadmissible any evidence that was not given “freely and willingly.” The other concern is that the defendant acted as a result of “internal” pressure that led him to take responsibility for committing an act that he did not do at all, thus causing him to “commit suicide in the confession” (Irshid v. Attorney-General [21], at p. 691). There is no admissibility barrier that guards against this internal pressure, but the court will examine this possibility when considering the weight that it will give the confession’ (CrimA 715/78 Levy v. State of Israel [23], at p. 234).

With regard to the concern that ‘internal’ pressure may lead the defendant to make a false confession, I will cite the remarks of Justice Goldberg, who said that even if improper pressures were not exerted on the defendant during his interrogation, thereby rendering the confession inadmissible, there is still a concern that his confession is false:

‘The court should still be concerned that other factors may have led the person making the confession to “commit suicide in the confession.” This “committing suicide” may be conscious (such as the desire to protect someone else, or the desire to bring about the end of the interrogation because of internal psychological pressure) or unconscious (because the person under interrogation believes his confession is true because of emotional defects and a distorted perception of reality). With regard to “committing suicide in a confession” the rabbis said (against the background of the principle of Jewish law that a person cannot incriminate himself) that:

“… the Sanhedrin does not pass a capital or corporal sentence on someone who confesses to an offence, in case he is mentally disturbed in this matter. Perhaps he is one of those troubled depressed persons that look forward to death, who stick knives in themselves and throw themselves from the rooftops. Perhaps in this way he will come and say something that he did not do, so that he will be killed…” (Maimonides, Hilechot Sanhedrin, 18, 6)’ (CrimA 4769/92 Nijam v. State of Israel [37], at p. 2186).

9. This fear of false confessions has led to the creation of ‘barriers and brakes’ for examining the admissibility and weight of confessions (in the words of President Barak in El Abeid v. State of Israel [29], at p. 865). These ‘barriers and brakes’ include the rule that a person may not be convicted solely on the basis of a confession that he made out of court, even when it was made lawfully without any external pressure (see for example CrimA 290/59 A v. Attorney-General [32], at p. 1495; CrimA 556/80 Ali v. State of Israel [18], at p. 184). It would seem that the aspiration to create a ‘barrier’ against the danger of a false confession for motives that are hidden in the soul of the person making the confession (to use the expression of Justice Landau in A v. Attorney-General [32]) is what led to the development of the requirement of ‘something extra’ in the case law of this court, which also constitutes a main and important ‘brake’ against false convictions. This requirement is intended to protect the defendant from being convicted on the basis of a false confession that he made during his interrogation, but no less importantly it aims to help the court arrive at the truth, and therefore it constitutes a means of protecting society as a whole from the conviction of innocent persons and the acquittal of guilty ones.

10. These remarks form the background for deciding the question that arises in the current appeal. As we have said, deciding this question is a result of a balance between values. Our journey begins with the provisions of s. 162 of the law, which constitutes the normative framework for examining the question, and it is on this basis that we must formulate the proper legal policy on this issue. Let us therefore turn to consider the questions that require a decision.

The normative framework — s. 162 of the law as a legal source for regarding silence as ‘something extra’

11. As I have said, the legal framework for examining the issue is s. 162 of the Criminal Procedure Law. For the sake of completeness, I will cite the section as it was amended in 2006:

‘Silence of the defendant

(b) A refusal of the defendant to testify shall not serve as evidence against him if an expert opinion is received to the effect that the defendant is a person with a mental disability or a person with an emotional disability as defined in the Interrogation Procedures and Testimony of Persons with Disabilities Law, and because of his aforesaid disability he is refusing to testify.’

It can be seen that subject to the exceptions listed expressly in it, the section clearly provides that the silence of a defendant in his trial has probative significance, i.e., it may constitute ‘support’ for the weight of evidence brought by the prosecution and ‘corroboration’ for this evidence. Is it right and proper to interpret the section as allowing the silence of a defendant to be regarded not only as ‘corroboration’ and ‘support,’ as the section states, but also as ‘something extra’ for the purpose of convicting him on the basis of his confession during interrogation, even though the provision does not expressly relate to this probative requirement? This is the first question that we should address.

12. My colleague Justice Levy is of the opinion that the interpretation adopted by the courts martial according to which, since a defendant’s silence can satisfy the requirement of corroboration, it can certainly also constitute ‘something extra’ for the confession of a defendant, is not free from doubt. According to him, if this was indeed the position of the legislature, it is surprising that it chose not to state this expressly in the section, especially in view of the fact that when this provision was added to the law, the requirement of ‘something extra’ already existed. From this my colleague infers that the answer to the question before us cannot be derived from the provisions of s. 162 of the law alone or from the technical definitions that were given to the terms ‘support’ and ‘corroboration’ in this section (para. 22 of his opinion).

13. My opinion is that even if the decision in this matter cannot be derived solely from the narrow interpretation of s. 162 of the law, the position that the silence of a defendant can also be regarded as ‘something extra,’ and thereby act against the defendant, should be based on some statutory source, and therefore this interpretive question cannot be left without an express decision. On its merits, a reading of the provisions of s. 162 and the interpretation of the provision from literal and purposive viewpoints show that it is possible to apply the section also to the probative requirement of ‘something extra,’ and therefore the section may be a statutory source for regarding the silence of the defendant in his trial as ‘something extra’ for a confession that he made out of court, as argued by the respondents. As I shall show later, my position is based on reasons arising from the literal interpretation of the section, as well as reasons concerning its legislative purposes and legislative history and reasons arising from the nature of the probative requirement of ‘something extra’ in comparison to the probative requirements of ‘corroboration’ and ‘support.’

On the purpose of s. 162 of the law

14. The enactment of s. 162 in 1976 was the result of a recommendation of the Public Commission for Criminal Procedure chaired by Justice M. Landau. The background to the legislation was the rise in scope and level of crime in Israeli society and the outlook that the legal system should be given stronger and more effective tools than in the past in order to deal with the rise in crime (Knesset Proceedings 71, 2115 (1974); 73, 2888 (1975)). The legislation process was accompanied by fierce debates among members of the Knesset, particularly because of the concern of a violation of the right to remain silent and the presumption of defendants’ innocence, but ultimately the section was passed despite the objections. It should be noted that whereas in the original wording of the section, as presented in the draft law by the government, it was proposed to regard the refusal of a defendant to testify solely as ‘corroboration,’ in the final wording, as it remains until today, it was stated that this refusal could also constitute ‘support for the weight of the prosecution’s evidence,’ for the purpose of ‘strengthening the section’ and not limiting it solely to cases where statute, from a technical perspective, requires a probative addition of ‘corroboration’ (see the remarks of Knesset Member A. Ankorin, Knesset Proceedings 77, 3506 (1976)).

The explanatory notes to the government draft law contain the following:

‘According to the consistent rulings of the courts, the silence of the defendant is not in itself evidence against him, nor can it serve as a substitute for prosecution evidence nor can it serve as grounds for his conviction…

This leads to the conclusion that the defendant’s silence cannot serve as a substitute for evidence that is required in certain circumstances (for example as support for the testimony of someone involved in an offence… or as support for a confession made to the police, and it certainly cannot serve as a substitute for proving corroboration when it is needed, such as in sex offences or to support the testimony of an accomplice in crime…).

The proposed section is intended to introduce a law that allows the refusal of a defendant to testify to be regarded as actual corroboration where it is needed; in other words, it allows the silence to be given a positive value. This is in fact a means of encouraging defendants to speak in the court, in order to prevent miscarriages of justice that may arise from the acquittal of defendants (sic!) no less than from the conviction of the innocent. It should be remembered that we are speaking about the stage after the prosecution evidence, i.e., when the facts already indicate that the defendant is involved in the criminal act, where common sense and experience tip the scales against the defendant who insists on the right to remain silent. However, it is not proposed that silence should constitute automatic corroboration, and the matter is left to the discretion of the court’ (Explanatory notes to the draft Criminal Procedure Law (Amendment no. 5), 5734-1974, Draft Laws 1103, 84, at p. 87).

It can thus be seen that the section under discussion was intended to reduce the likelihood of acquitting guilty defendants, by giving a positive value to silence and encouraging defendants to tell their version of events to the court.

In my opinion, even if the probative requirement of ‘something extra’ was not considered by the legislature when it enacted the section, a reading of the minutes of the debates in the Knesset at the various stages of the legislation shows that it can also not be inferred that there was an intention to distinguish between the probative requirements of ‘support’ and ‘corroboration,’ which are enshrined in the section, and the probative requirement of ‘something extra’ that was not included in it but was recognized by case law at that time. In my opinion, the purpose of the section, as it arises from the explanatory notes, is valid even when the probative addition required for a conviction is ‘something extra,’ since even when the only evidence against the defendant is his own confession, after which he denied the charges, there is a clear need to encourage him to testify before the court, so that the evidence on which the court bases its verdict as to his innocence or guilt is as complete as possible.

15. From the remarks of my colleague Justice Levy it may be inferred that the fact that the legislature did not expressly address the probative requirement of ‘something extra’ implies a ‘negative arrangement,’ since had the legislature wanted to regard the silence of a defendant also as ‘something extra,’ it may be assumed that it would have said this expressly in s. 162 of the law (para. 22 of his opinion). It is possible to disagree with this argument: first, the requirement of ‘something extra’ is, it will be remembered, the product of case law, and it is not enshrined in statute (see CrimA 3/49 Andelersky v. Attorney-General [31], at pp. 592-593, in which the requirement was introduced, and which this court has affirmed on countless occasions). For this reason it is doubtful whether it is right to attribute far-reaching significance to the fact that the legislature did not address this probative requirement expressly. Second, originally it was accepted that the meaning of the concept of ‘support’ was identical to the meaning of the concept ‘something extra,’ in view of the clear recognition that it was not a requirement of ‘corroboration’ but a lesser requirement, in a desire to refrain from creating an additional term that would necessitate making fine distinctions between it and the term similar to it (Y. Kedmi, On Evidence (first part, 2003), at p. 297; cf. also CrimA 735/80 Cohen v. State of Israel [50], at p. 99).

16. In this context, it is interesting to refer to remarks that were uttered during the debates in the Knesset before the enactment of the Evidence Ordinance Amendment Law (no. 6), 5742-1982, in which the requirement for ‘support’ was introduced for the conviction of a person on the basis of the testimony of an accomplice (s. 54A(a) of the Evidence Ordinance). When presenting the proposed amendment to the law, the Minister of Justice at the time, M. Nissim, said the following:

‘The guideline for the degree of caution required will from now on be given in the requirement of “something to support the testimony.” This is a requirement that corresponds to the requirement of ‘something extra’ as formulated in case law [emphasis supplied]. The purpose of the requirement of “something extra” is to direct the attention of the judge to the danger of a conviction on the basis of the sole testimony before him, and to caution him to ascertain, with the aid of a probative addition that is small from the viewpoint of its nature, quantity and quality, that the testimony under consideration is true. The addition to the testimony may arise either from the testimony itself or from other evidence, provided that it persuades the judge of the credibility of the testimony being supported’ (Knesset Proceedings 90, 309 (1981)).

It should be noted that, like s. 162 of the Criminal Procedure Law, the aforesaid s. 54A(a) of the Evidence Ordinance was enacted as a result of a recommendation of the Public Commission for Criminal Procedure, and it is interesting to note that its proposal in this regard was to cancel the requirement of ‘corroboration’ for the testimony of an accomplice, and to replace it with the requirement of ‘something extra.’ As is well known, ultimately the section was amended so that the requirement of ‘corroboration’ was replaced by the requirement of ‘support’ (see the explanatory notes to the draft Evidence Ordinance Amendment Law (no. 6), 5740-1980 (Draft Laws 1477, 397; CrimA 6147/92 State of Israel v. Cohen [51], at p. 69). It would appear that these events can indicate a relative conceptual kinship between the two terms (‘something extra’ and ‘support’), and in practice, despite the distinctions that exist between the two aforesaid probative requirements, sometimes that have been mentioned by case law ‘in one breath,’ as if they were an identical requirement (Kedmi, On Evidence, supra, at p. 360; cf. CrimA 190/82 Marcus v. State of Israel [52], at p. 294; CrimA 1242/97 Greenberg v. State of Israel [53], at p. 91). It follows that although s. 162 was added to the Criminal Procedure Law when the rule concerning the requirement of ‘something extra’ already existed, in my opinion we should not infer from the fact that there is no express mention of this probative requirement that the legislature intended s. 162 not to apply to it.

The nature of the various probative requirements — ‘corroboration,’ ‘support’ and ‘something extra’

17. The aforesaid interpretation of s. 162 of the law is also required, in my opinion, by an analysis of the substance of the section, of the nature of the probative requirement of ‘something extra’ and the relationship between it and the other probative requirements. In my opinion, the fact that the legislature determined in this provision of statute that a failure of a defendant to testify may act as ‘support’ or ‘corroboration’ implies necessarily that it may also serve as ‘something extra.’ This position of mine derives from the qualitative and quantitative ‘hierarchy’ that exists between these probative requirements, as they have been shaped and formulated in legislation and case law, where the probative requirement of ‘something extra’ stands at the bottom of the hierarchy, so that it is included and incorporated in the other probative requirements.

Before I begin to discuss this matter thoroughly, we should restate, in a nutshell, the basic principles concerning the various probative requirements.

18. The requirement that additional evidence should be produced for certain incriminating evidence, as a condition for convicting a defendant of a criminal offence, is found in the Israeli legal system in the form of three different concepts: ‘corroboration,’ ‘support’ and ‘something extra.’

In cases where additional evidence is required in the form of ‘corroboration,’ the intention is that we require evidence that ‘implicates’ (or tends to implicate) the defendant in the commission of the offence that is attributed to him. Inter alia, ‘corroboration’ is required for the testimony of a state’s witness who incriminates the defendant and the testimony of a minor who is the complainant in an indictment involving a sex offence and whose testimony is brought before the court through a child interrogator. In these cases, the corroboration evidence should be ‘independent evidence that points to the involvement of the defendant in the offence’ (CrimA 238/89 Askapur v. State of Israel [54], at p. 411; CrimA 378/74 Messer v. State of Israel [55], at pp. 695-698; CrimA 85/80 Katashvili v. State of Israel [56], at pp. 68-71). In view of this, it has been held that for a certain piece of evidence to constitute ‘corroboration,’ it should satisfy three cumulative requirements: it should have a separate and independent origin to the testimony that requires corroboration; it should ‘implicate’ the defendant in the commission of the act that is the subject of the indictment; and it should relate to a significant point that is the subject of dispute between the parties (see, for example, CrimA 387/83 State of Israel v. Yehudai [41], at p. 203). The reason for these strict requirements is the need to ensure the truth of the statements made by a witness against the defendant and to dispel the concern that perhaps he is making up an untrue story about the defendant, where experience shows that the testimony may lack credibility and therefore it cannot serve on its own as a basis for convicting a defendant (CrimA 290/59 A v. Attorney-General [32]; CrimA 389/73 Ben-Lulu v. State of Israel [57], at p. 492; CrimA 169/74 Kadouri v. State of Israel [58], at p. 403; CrimA 5544/91 Moyal v. State of Israel [59], at p. 1357; CrimA 6147/92 State of Israel v. Cohen [51], at p. 72; Kedmi, On Evidence, supra, at p. 126).

19. Unlike ‘corroboration,’ the requirement of additional evidence of the type known as ‘support’ is a requirement of ‘authenticating’ evidence: ‘the requirement of support is satisfied if the support adds a degree of truth to the witness’s statement, and it does not need to “implicate” the defendant in the criminal act’ (CrimA 241/87 Cohen v. State of Israel [60], at p. 746; Kedmi, On Evidence, supra, at p. 298). Inter alia, ‘support’ is required for the testimony of an accomplice (s. 54A(a) of the Evidence Ordinance), and for a statement made to the police by a witness who later gives contradictory testimony in court, when the statement is admitted as evidence (s. 10A(d) of the Evidence Ordinance). ‘Support’ is ‘intended to indicate that the witness’s testimony, when taken on its own, is credible. Therefore there is no requirement that the support should relate to the offence or the criminal conspiracy between the defendant and the witness-accomplice’ (Askapur v. State of Israel [54], at p. 411). Like the requirement of ‘corroboration,’ ‘support’ is needed because of the concern that the witness gave false testimony that implicates the defendant, and therefore evidence is required to strengthen the credibility and reliability of this testimony (CrimA 6147/92 State of Israel v. Cohen [51], at p. 80).

20. Last, the requirement of ‘something extra’ is, like ‘support’ and unlike ‘corroboration,’ a requirement of additional ‘authenticating’ evidence. Therefore, unlike ‘corroboration,’ the ‘something extra’ does not need to indicate the guilt of the defendant; any direct or circumstantial evidence that is external to the defendant’s confession and that can confirm to some degree the content of the confession and indicate its truthfulness is sufficient (CrimA 290/59 A v. Attorney-General [32], at p. 1499; CrimA 6936/94 Awad v. State of Israel [61], at p. 848). The requirement of ‘something extra’ is intended to remove the concern that the defendant is taking responsibility for an act that was done by someone else or that was not done at all, and therefore, in principle, very little evidence is required to satisfy it (CrimA 6147/92 State of Israel v. Cohen [51], at p. 72). Consequently case law has pointed out on more than one occasion that this evidence ‘can be very small indeed’ (CrimA 178/65 Usha v. Attorney-General [35], at p. 156), and even ‘as light as a feather’ (El Abeid v. State of Israel [29], at p. 834). It is sufficient that the court is satisfied that the confession is not ‘merely a fabrication,’ and that it is persuaded that the version of events told by the defendant in his confession is indeed a possible one (CrimA 312/73 Matzrawa v. State of Israel [62], at p. 809).

21. These differences, both in terminology and in substance, between the three probative requirements have led to the creation of a kind of ‘hierarchy’ between them. This hierarchy is based on the assumption that, as a rule, ‘a person does not portray himself as a criminal’ (El Abeid v. State of Israel [29], at p. 833). It follows that most confessions are true confessions, whereas it is a more reasonable possibility that someone else who testifies against the defendant will present false testimony, for various reasons such as anger, jealousy, a desire to prevent himself from being found responsible or to protect a third party, a desire to obtain some benefit, etc.. In other words, the need for additional evidence for a defendant’s confession to dispel the fear that his confession may be false is smaller than the need for additional evidence that is required for the testimony of someone else against the defendant, and therefore ‘corroboration’ and ‘support’ should satisfy stricter criteria than ‘something extra.’ In this vein, it was said in one case with regard to evidence that satisfied the requirement of ‘something extra’ that ‘it is therefore accepted that this kind of evidence should be regarded as “lower” on the ladder of “secondary” evidence (“corroboration” and “support”)’ (Nijam v. State of Israel [37], at p. 2186). The aforesaid hierarchy has been clearly expressed in the case law of this court, which has repeatedly held over the years that various kinds of evidence that may serve as ‘corroboration’ automatically satisfy the requirements of ‘support’ and ‘something extra.’ Thus, for example, Justice H. Cohn said that:

‘The aforesaid statements of the deceased could have served as corroboration for the appellants’ confessions, had there been a need for any real corroborating evidence. The argument is an a fortiori one: if these statements could have constituted actual corroboration for the appellants’ confessions, they certainly constitute “something extra” for the confessions’ (CrimA 399/72 Menahem v. State of Israel [63]).

In another case Justice Or said that ‘If such corroboration is found, this corroboration will also automatically satisfy the requirement of support for a witness’s statement, on which the court relies by virtue of the provisions of s. 10A of the Ordinance, since what satisfies the greater requirement also satisfies the lesser requirement’ (CrimA 450/82 Tripi v. State of Israel [64], at p. 597, and similar remarks were made by President Shamgar: ‘If the testimony of the accomplice could have served as corroboration, whose weight and scope are broader than evidence that constitutes “support”… and “something extra,” it is obvious that such testimony could also serve as support’ (Askapur v. State of Israel [54]; see also CrimA 282/75 Karki v. State of Israel [65]; CrimA 34/78 Algul v. State of Israel [66]; CrimA 949/80 Shuhami v. State of Israel [67], at p. 72; CrimA 146/81 Al-Sena v. State of Israel [68], at p. 503; CrimA 777/80 Beinashvili v. State of Israel [69], at p. 472; CrimA 533/82 Zakkai v. State of Israel [70], at p. 73).

To summarize the law in this matter, Kedmi says that:

‘Evidence that is capable of serving as “corroboration,” where such additional evidence is required (i.e., additional “implicating” evidence), can also serve as “something extra,” since the greater requirement includes the lesser one: if it is capable of “implicating” the person making the confession, it can certainly “authenticate” his confession.’

And with regard to the relationship between ‘corroboration’ and ‘support,’ Kedmi says that:

‘Evidence that can serve as corroboration will satisfy the requirement of “support”.’

(Kedmi, On Evidence, supra, at pp. 141, 303).

This is also true in my opinion with regard to the relationship between ‘support’ and ‘something extra.’ Just as the support that is required does not need to amount to actual corroboration in its character and weight (Shuhami v. State of Israel [67], at p. 72), so too ‘something extra’ does not need to amount to actual support in its character and weight (in this matter, see the different position of Kedmi, On Evidence, supra, at pp. 184, 203, 297-298).

22. In our case, with regard to the probative significance that can be attributed to the silence of the defendant in his trial, I do not think that it would be right to determine that the aforesaid hierarchy does not apply. Given that the defendant’s silence can provide support for the testimonies of others against him and can constitute corroboration when it is required, I do not think that we should hold that it cannot support his confession or corroborate it, or at the very least, serve as ‘something extra,’ when it is needed. The reason for this is that, as a rule, it is easier for a person to refute or explain his own confession than to refute an accusation levelled at him by someone else (see and cf. E. Kamar, ‘Corroborating Confessions by the Defendant’s Own Statements,’ 5 Israel Journal of Criminal Justice (Plilim) 277 (1996), at p. 292, who agrees with this solely on a prima facie basis).

23. My colleague Justice Levy pointed to the fact that in cases where the legislature allowed the defendant’s silence to be regarded as ‘support’ and ‘corroboration’ for the prosecution evidence, the silence is an addition to testimonies that do not originate with the defendant (the testimony of a state’s witness, the testimony of an accomplice, etc.), whereas when we are speaking of a confession of the defendant, the silence is an addition to testimony that he himself gave in the interrogation. In such circumstances, Justice Levy believes that the silence cannot authenticate the confession to the required degree (para. 23 of his opinion). I am not persuaded that this conclusion is necessary. I discussed above the distinctions that exist between the various probative requirements. These distinctions are valid. But I do not think that they, together with the aforesaid argument that was raised by my colleague, lead to the conclusion that a defendant’s silence in his trial cannot serve as ‘something extra.’ According to the rule that has been formulated in the case law of this court, there is nothing to prevent the source of the ‘something extra’ being the defendant himself. Thus, various evidence that originates with the defendant, such as the defendant’s incriminating conduct, his lies on major issues, a confession of the defendant to major issues during his testimony in court and statements that he made before a judge during arrest proceedings have all been recognized as satisfying the requirement of ‘something extra’ (see, for example, Matzrawa v. State of Israel [62], at p. 809; CrimA 788/77 Bader v. State of Israel [71], at p. 831; CrimA 6289/94 Janshvili v. State of Israel [22], at pp. 171 and 176; CrimA 5825/97 Shalom v. State of Israel [72], at p. 958; CrimA 6613/99 Smirk v. State of Israel [73], at pp. 557-559; Kedmi, On Evidence, supra, at p. 129). I am not persuaded that there is a sufficiently strong reason for determining a different rule for the defendant’s silence.

24. In view of all of the aforesaid, my position is that on the basis of s. 162 of the law, a refusal of a defendant to testify can also satisfy the requirement of ‘something extra’ for a confession made by him out of court. Admittedly this is not evidence that is ‘external’ to the defendant, as some authorities think it should be (see, for example, Kamar, ‘Corroborating Confessions by the Defendant’s Own Statements,’ supra, at p. 280), but we are speaking of additional evidence that is ‘external’ to the confession, and in my opinion it is capable of authenticating, to the required degree, the defendant’s confession to the acts attributed to him. Therefore, in accordance with the stipulation of the legislature and in accordance with the rule in force today as I have explained it, it is sufficient. Although I do not take lightly the various concerns raised by my colleague, as well as the basic right of a defendant to a proper and fair trial, it would not be right in my opinion to derive from these a normative conclusion by means of a strict interpretation that is not required by the wording or the purpose of the statute.

25. The fact that there is a legal basis for regarding the refusal of a defendant to testify as ‘something extra’ for his confession during the interrogation does not mean that this should be adopted as a legal policy. From the wording of s. 162 of the law it can be clearly seen that the legislature left the court with discretion as to whether to regard the refusal of a defendant to testify as having the aforesaid probative significance and whether to give it any weight. How should this discretion be exercised? Is it desirable that the silence of a defendant in his trial should serve as ‘something extra’ for his confession, and in what circumstances? What, then, is the proper judicial policy that should be adopted in this matter?

The proposed legal policy — the rule

26. My colleague Justice Levy described in his opinion important reasons why he believes that the silence of a defendant should not be regarded as ‘something extra’ for his confession, unless the conditions that he stated in his opinion are satisfied, namely a video recording of the interrogation and the impression of the court that the confession is a true confession. As I have already explained, I agree with my colleague with regard to the rule, but with regard to the exception, which I shall discuss below, in my opinion the trial court should be given a greater margin of discretion than that proposed by my colleague.

I propose therefore that as a rule the silence of the defendant should not constitute ‘something extra.’ There are several reasons for this. First and foremost, my position is based on the basic aspiration of the legal system to arrive at the truth and on the innate fear of convicting an innocent person of a crime on the basis of a false confession that he made. Even though, as we have said, false confessions are the exception to the rule, we must be wary of them. We should mould our legal system in a manner that will restrict to the absolute minimum the possibility that they will be used to convict an innocent person.

My colleague mentioned in his opinion three categories of factors that cause a person under internal pressure to make a false confession, just as these were enumerated in the Goldberg Commission Report (at pp. 8-10). The reasons that may lead a certain defendant to make a false confession are many and varied. Some of them will always remain somewhat unclear. This difficulty is no less relevant, and maybe even more relevant, to the silence of a defendant in his trial. This silence appears contrary to logic and common sense that tell us that an innocent person should not only be prepared to tell his version of events in court, but he should eagerly await the possibility of entering the witness box to refute the suspicions against him and to clear his name, and that a defendant who decides to remain silent is not prepared to tell his version of events or to be cross-examined on the version of events that he told in the interrogation (see, for example, CrimA 196/85 Silberberg v. State of Israel [12], at p. 525).

Reality, it would appear, is sometimes different. In practice we cannot rule out the possibility that there are cases where a defendant in his interrogation will take responsibility for committing an offence that he did not commit, but later he will deny the charges, even though he chooses not to testify in his own defence. The attorney-general also recognizes this possibility and he said in his response that experience shows that there are persons who make false confessions out of court (not necessarily because improper pressure was exerted on them by the police) and that some of them — even after they have denied the charges — may during their trial choose the right of silence. According to him, even though it is possible that someone who made a false confession out of court will act consistently and repeat his confession when he enters the witness stand, it is not possible, a priori, to rule out the possibility that he will not do this but will choose to remain silent (paras. 42-44 of the attorney-general’s arguments).

27. In a situation such as this, where there is an innate difficulty in understanding the motive for false confessions, as well as the motive for refusing to testify, I agree with my colleague’s position that there are grounds for concern that relying on the defendant’s silence as ‘something extra’ for his confessions may undermine the purpose that led to the requirement of ‘something extra’ in the first place, namely the desire to authenticate the defendant’s confession (para. 23 of my colleague’s opinion). Notwithstanding the probative significance that it is possible to attribute to silence, it on its own is limited in its ability to dispel the uncertainty and to provide the court with the explanations that it lacks. When the defendant refuses to talk, it is not possible to cross-examine him in order to discover the truth of what was recorded in his confession at the police station, and therefore there is a difficulty in bridging the gap between his confession in the interrogation and his denial of the charges. Moreover, just as there is a certain conceptual difficulty in saying that lies uttered by the defendant are a guarantee that his confession is not a false confession, when it is the only evidence against him (El Abeid v. State of Israel [29], at p. 721), attributing probative significance to the defendant’s silence may in my opinion suffer from a similar difficulty. For who can guarantee that those lies, or that silence, actually imply a feeling of guilt and are not the result of the same motive that led the defendant to make a false confession. Therefore, just as the court is required to take great care not to err in convicting an innocent person by relying on his lies, it is required a fortiori to take great care not to err in convicting an innocent person by relying solely on his silence as ‘something extra’ for his confession. This caution requires us, in my opinion, to adopt a judicial policy whereby relying on the silence of a defendant as ‘something extra’ for his confession will be done in exceptional cases only.

28. Moreover, an additional reason to refrain, as a rule, from regarding the silence of the defendant in the court as ‘something extra’ for his confession to the police arises from the significant concern that the investigation and prosecution authorities should not be given a ‘green light’ to regard a confession on its own as sufficient for the purpose of filing indictments. This may deprive these authorities of any incentive to try and locate additional evidence apart from the confession, and thereby undermine the basic interest of clarifying and discovering the truth. The attorney-general’s guideline no. 53.000 of 1970 provides, admittedly, that as a rule an indictment should not be filed on the basis of confessions made out of court without there being ‘something extra’ in the evidence that was accumulated in the investigation to authenticate those confessions, and that it may only be done in exceptional cases with the approval of the state attorney or the chief military prosecutor. I believe, however, that when formulating the proper policy on this issue and in order to ensure an additional aspect of the defendant’s right to a fair trial, it is important to encourage the prosecution to act in order to obtain as much evidence as possible against the defendant, apart from his confession, and where possible to base the indictment on a body of evidence that is comprehensive and solid.

29. Finally, and more indirectly, my position is also based on the importance that I attach to preventing the right to remain silent and the presumption of defendants’ innocence from being undermined to a greater extent than what the legislature has expressly recognized in s. 162 of the law. In his arguments counsel for the appellant emphasized that the question that arises in our case, despite its technical appearance as simply a question of evidence, has constitutional importance, since it lies at the crossroads between two constitutional rights: the right of defendants to remain silent and the presumption of innocence enjoyed by every person before he has been convicted in a criminal trial, which are both regarded as basic rights. It is well known that there may be different approaches on the question of which procedural rights in criminal trials are indeed included within the scope of the constitutional right to dignity and liberty, as it is enshrined in the Basic Law: Human Dignity and Liberty, and similarly also with regard to the question of the scope of the constitutional protection that should be given to rights that are not expressly mentioned in the Basic Law (see, for example, Yissacharov v. Chief Military Prosecutor [5], at p. 1107 {351}). In view of what I regard as the clear provisions of s. 162 of the law, I do not think that there is any need for us to decide between the aforesaid approaches in the current appeal. In any case, it would appear that there can be no dispute that the aforesaid rights are relative rights and not absolute rights. This can be seen by the provisions introduced by the legislature in s. 162 (see also HCJ 6319/95 Hachmi v. Justice of Tel-Aviv-Jaffa Magistrates Court [10], at pp. 765-766). I do not accept the argument that regarding the silence of a defendant as ‘something extra’ for his confession violates these rights beyond what the legislature has permitted in s. 162 of the law. But I am aware of the concern that was expressed by defence counsel that these rights may suffer too great a violation, beyond what was permitted by the legislature, and for this reason also I am of the opinion that this use should be made sparingly and in moderation, solely as an exception.

In view of all of the aforesaid reasons, I too am of the opinion that we should conclude that, as a rule, we should not regard the defendant’s silence as ‘something extra’ for the confession that he made in his interrogation. Notwithstanding, it would not be right in my opinion to determine on this issue a sweeping rule, and there is a basis for recognizing exceptions to the aforesaid rule and for granting discretion to the trial court. This legal policy is what will guarantee the proper balance between our commitment to the basic value of discovering the truth, which is based on the goal of ensuring the conviction of the guilty and the acquittal of the innocent, and our duty to protect the right of the defendant to a proper and fair trial. As I said at the beginning of my remarks, this balance is also required by the need to act forcefully against the rise in crime, in order to protect the safety and security of the public, including the victims of crime, while at the same time preserving the fairness and ethical and moral strength of society in general and the legal system in particular. The remarks of Justice Strasberg-Cohen are particularly apt in this regard:

‘The function of a civilized society is to find the proper balance between all of these in such a way that the legal system will have the proper tools that will allow the holding of a fair and efficient trial, prevent the conviction of the innocent and ensure that the guilty do not escape justice. The tools for achieving these goals within the framework of the judicial system are, inter alia, criminal procedure and the rules of evidence. It is through these that the legal system plots its course to discover the truth and to administer justice’ (Hachmi v. Justice of Tel-Aviv-Jaffa Magistrates Court [10], at p. 756).

30. In our case, following a rule that denies the trial court any discretion whatsoever with regard to the probative significance that should be attributed to the defendant’s refusal to testify and that obliges it to acquit a defendant even when it is perfectly clear to the court that his confession is a true one, solely because no external evidence was found that supports it, is likely to give the aforesaid rule of evidence precedence over the value of arriving at the truth. Such a position is undesirable. In this regard I agree with the apt and instructive remarks of my colleague, who pointed to the possibility that there may be cases in which the acquittal of the defendant for the reason that his silence does not constitute ‘something extra’ for his confession may undermine the search for the truth, as well as the fight against crime and protecting the public from it (para. 25 of his opinion). In this regard it is not superfluous to mention that —

‘An erroneous acquittal, and certainly a false conviction, harm both the doing of justice and the appearance that justice is being done, and it may undermine public confidence in the ability of the judicial authority to do justice to the individual and to society’ (Yissacharov v. Chief Military Prosecutor [5], at p. 1124 {383}).

To summarize what we have said so far: my position is that although s. 162 of the law makes it possible to regard silence as ‘something extra,’ as a matter of policy we should not, as a rule, give the silence of the defendant the aforesaid probative significance. This significance will be given to silence only in limited cases, which will constitute the exception rather than the rule. Within the scope of this exception, it is important to give the court discretion in accordance with the circumstances of each case and in accordance with the guidelines that we will set out below.

The proposed legal policy — the exceptions to the rule

31. It will be remembered that the position of my colleague on this matter is that the court should be allowed to regard the silence of the defendant as ‘something extra’ only in cases where a video recording was made of the interrogation and the court that saw the recording obtained the impression that it has before it a true confession. The centre of gravity of the disagreement between me and my colleague revolves around these exceptions. I agree with my colleague that the decision concerning the significance that should be attributed to the silence of a defendant is a matter for the trial court to decide according to the special circumstances of the case before it and in accordance with its discretion (para. 26 of his opinion). But it is precisely for this reason that I am of the opinion that the court should be given a broader margin for exercising its discretion than the one that my colleague is prepared to recognize, and that the exercising of this discretion should not be conditional upon making a video recording of the interrogation.

32. Confessions for which a video recording is made, which my colleague requires as a basis for the exception proposed by him, do of course have a clear advantage that cannot be disputed. A video recording allows the court to form a better impression of the defendant’s interrogation and of the circumstances in which his confession was made, and also to form an impression as to whether his confession was made because of pressures that he was under as a result of the conditions of the interrogation or the arrest. But it seems to me that it would not be right to give too much weight, and certainly not exclusive weight, to the fact that a video recording of the interrogation was made. We should be aware of the possibility that even when the court has a video recording of the interrogation, there is still a possibility that various factors that may have led the defendant to make a false confession were not necessarily expressed in this video recording. Therefore we should take into account that there may be cases in which the fact that a video recording of an interrogation was made will not necessarily be sufficient for the court to regard his silence as ‘something extra’ for his confession (see also what my colleague said in para. 26 of his opinion), and also that there may be cases in which, despite the lack of such a recording, the refusal of the defendant to testify may satisfy the requirement of ‘something extra,’ all of which at the discretion of the court that formed an impression of the confession and the circumstances in which it was made in the specific case that is before it.

33. In my opinion, the court should be allowed this judicial discretion in every case on its merits and in accordance with its circumstances. Notwithstanding, it should be exercised in moderation and with care, in a limited number of cases, and in accordance with the following three criteria, which result from everything that I have said in my opinion thus far:

First, an essential condition for regarding the refusal of the defendant to testify as ‘something extra’ for his confession is that the confession itself is of very great internal weight, so that the court is persuaded that it is a ‘confession that stands on its own in view of its independent weight’ (Levy v. State of Israel [23], at p. 234). It will be remembered that the internal test of a confession ‘examines the confession in accordance with the indications of truth that it reveals, such as its logic or lack of internal logic, the clarity or confusion of the details contained in it and other similar common sense indications that lead a rational person to trust what someone says’ (ibid. [23]). This test considers the content of the confession, and it seeks, on the basis of logic and common sense, to find indications of truth in it, which can be seen from the wording of the confession, the sequence of events, the amount of detail, its internal reasonableness and other such tests (FH 3081/91 Kozali v. State of Israel [26], at p. 458; on the importance of the amount of detail in a confession, see also El Abeid v. State of Israel [29], at p. 771). There is a rule in our case law that the greater the independent weight of the confession, as determined by the indications of truth that can be seen in it, the smaller the need for resorting to the external test of ‘something extra’ (Levy v. State of Israel [23], at p. 234; El Abeid v. State of Israel [29], at p. 795). As a rule, the silence of a defendant is not in itself evidence of great weight. Therefore, in view of the reciprocal relationship that exists between the internal weight of the confession and its external weight (Steckler v. State of Israel [30], at p. 1665), the refusal of a defendant to testify may constitute ‘something extra’ for his confession only in those cases where the court is persuaded, according to the internal test of the confession, that it is a true confession, i.e., when the confession before it is a detailed, coherent and persuasive confession from the viewpoint of its nature, its internal logic and the indications of truth that can be seen in it.

Second, an additional condition that is required in order to make use of the defendant’s silence as ‘something extra’ concerns the ability of the court, on the basis of all the external circumstances of the confession, to rule out — to the required degree of certainty — the possibility that the defendant acted as a result of ‘internal’ pressure that led him to take responsibility for committing an act that he did not do or that was not done at all. In this context, the court should consider the various reasons that may lead a defendant to confess to committing an offence that he did not commit, as set out in the Goldberg Commission report, and it should ask itself whether all reasonable doubt has been removed as to whether one of these reasons exists in the case before it. Thus, for example, it should examine whether there are grounds for a concern that the defendant’s confession was the result of social or other pressures that influenced him. This matter requires a detailed examination especially when we are dealing with an offence that was committed by several accomplices, when we are dealing with a defendant who belongs to a criminal organization, or when the circumstances of the case show a possibility that the confession was made out of a desire not to incriminate someone else, such as a family member. In addition, the court should consider whether to rule out the possibility that the defendant is a member of one of the ‘risk groups’ that give rise to a greater concern that internal pressure may cause them to confess to committing acts that they did not do, such as someone who is mentally disabled (Levy v. State of Israel [23], at p. 235) or intoxicated (Steckler v. State of Israel [30], at pp. 1664-1665). It is well known that in cases such as these there is greater concern that the factors that caused these defendants to confess to an act that they did not do also caused them to refuse to testify in their trial, and therefore the court should not regard the defendant’s silence as sufficient, but should examine and demand a more substantial and significant authentication of the defendant’s confession.

It should be emphasized that we are not speaking of a closed set of criteria. These are merely examples of the kinds of indications that the court should seek in order to rule out the possibility that the confession is a false one.

Last, the final criterion is actually the converse of the rule described in the second criterion. In this context, the court should ascertain that there are no indications that may show that the silence of the defendant in the court, and not merely his confession in the interrogation, is a result of any internal or external pressure; alternatively, it should examine whether there is any evidence at all with regard to a motive for the defendant’s silence and with regard to the circumstances of this silence, which, although it may not serve on their own as ‘something extra,’ allows the court to dispel the clouds of uncertainty in this matter. The court should therefore be persuaded, after considering all the circumstances of the case, that the silence of the defendant is not the result of an innocent motive, and therefore the logical conclusion is that he really does not have an answer to the charges leveled at him, to which he confessed in his interrogation.

34. When these three criteria, which are mainly intended to serve as guidelines, are all satisfied, the court may determine that the silence of the defendant in his trial may constitute the ‘something extra’ that is required for his confession. As we have said, this should be done only in exceptional cases. Because of the complexity of the issue and the large number of factors that need to be taken into account, I do not think that it is possible or proper to determine more rigid criteria with regard to the nature of the exceptional circumstances in which the court will decide that silence may constitute ‘something extra.’ The decision on this matter will be made by the trial court, which will exercise the discretion given to it by the legislature in accordance with the criteria outlined above and after considering the circumstances of each case on its merits. This discretion should be exercised sparingly, carefully, reasonably and wisely, so that the exception is not allowed to become the rule. The remarks of Justice Barak in another case are illuminating in this respect:

‘… The criminal trial should discover the truth, and this is its main goal. Of course, following rules and discovering the truth are not two conflicting goals. On the contrary, the rules are intended to determine a standard for holding a trial, which, as experience shows, can lead to the discovery of the truth, and thereby the two goals coincide. Notwithstanding, there are cases where a formal insistence on rules in a special case will result in a miscarriage of justice, whether in the form of a conviction of the innocent or the acquittal of the guilty. In these circumstances we should aspire to the goal that the rules themselves will give the court power and discretion to do justice’ (CrimA 951/80 Kanir v. State of Israel [74], at p. 516).

35. Finally, before concluding, we should mention once again the guideline of the attorney-general according to which, as a rule, an indictment should not be filed based on a defendant’s confession only, without there being ‘something extra’ in the evidence. It is to be hoped that this guideline, which applies to both the civilian and military prosecutors, will lead to a result whereby the cases in which the court will be required to decide whether to regard the silence of the defendant as ‘something extra’ for his confession will be few and far between.

From general principles to the specific case — the conviction of the appellant in this case

36. With regard to the appellant before us, I agree with the conclusion of my colleague Justice Levy, although not with his reasoning. In my opinion, without making any hard and fast determination on this matter, it would prima facie appear that this case is precisely one of those cases that may fall within the scope of those exceptional cases in which the trial court was entitled to find ‘something extra’ in the refusal of the appellant to testify. I should point out that the findings of the court martial — and like my colleague I too see no reason to depart from them — were that the appellant’s confession has great weight; his interrogation was conducted in a good atmosphere, with full cooperation on his part; no claim was raised that pressure had been exerted on him; in addition, prima facie there was no indication that the appellant’s confession, as well as his choice not to testify, were the result of any pressure to which he was subject. In any case, as I have said, this is a matter for the trial court to decide. Therefore, even though according to the different legal outlook that I have proposed it might have been proper to return the case to the court martial, so that it might reconsider and reexamine the matter in accordance with the criteria that we have outlined above, I do not propose to do this. In view of the fact that the position of the chief military prosecutor, as presented by the representative of the attorney-general, is that he no longer insists upon the appellant’s conviction, and in view of the circumstances of the case, including the fact that the offence is on the lowest level of criminal offences, the circumstances in which the offence was committed and the large period of time that has passed since the offence was committed, I too am of the opinion, like my colleague, that we should order the appellant’s acquittal.