H.C.J 428/86
H.C.J 429/86
H.C.J 431/86
H.C.J 446/86
H.C.J 448/86
H.C.J 463/86
M.A.A 320/86
Y. BARZILAI, ADV.
v.
1 GOVERNMENT OF ISRAEL
2. ATTORNEY-GENERAL
3-6. A-D H.C.J 428/86
1. Y. SARID, M.K.
2. D. ZUCKER, SECRETARY-GENERAL OF CITIZENS RIGHTS AND PEACE MOVEMENT
3. CITIZENS RIGHTS AND PEACE MOVEMENT
4. S. ALONI, M.K.
5. R. COHEN, M.K.
v.
1. MINISTER OF JUSTICE
2. ATTORNEY GENERAL
3. INSPECTOR-GENERAL OF POLICE
4. DEPUTY PRIME MINISTER ANDMINISTER OF FOREIGN AFFAIRS
5-8. A-D. H.C.J 429/86
1. M. MAROZ, ADV.
2. D. YIFTAH, ADV.
v.
1. MINISTER OF POLICE
2-5.A-D. H.C.J 431/86
A. ZICHRONI, ADV.
v.
1. INSPECTOR-GENERAL OF POLICE
2. ATTORNEY-GENERAL
3. MINISTER OF JUSTICE
4. HEAD OF THE GENERAL SECURITY SERVICE (G.S.S.)
5. ASSISTANT NO. 1 TO HEAD OF G.S.S.
6. ASSISTANT NO. 2 TO HEAD OF G.S.S.
7. ASSISTANT NO. 3 TO HEAD OF G.S.S.
8. DEPUTY PRIME MINISTER AND MINISTER OF FOREIGN AFFAIRS H.C.J 446/86
1. DR. J.M. EDREY
2. DR. H. BEN-MENAHEM
3. DR. B. BRACHA
4. DR. M. GUR-ARIEH
5. DR. K. MANN
6. DR. A. MAOZ
7. DR. C. PASBERG
8. DR. M. KREMNITZER
9. PROF. D. KRETZMER
10. DR. A. ROZEN-ZVI
11. DR. Y. SHACHAR
12. PROF. M. SCHIFMAN
v.
1. INSPECTOR-GENERAL OF POLICE
2. MINISTER OF POLICE
3. HEAD AND THREE OTHER MEMBERS OF THE G.S.S.
H.C.J 448/86
1. M.A. ABU GAM'A
2. S.H. ABU GAM'A
v.
1. GOVERNMENT OF ISRAEL
2. MINISTER OF POLICE
3. ATTORNEY-GENERAL
H.C.J 463/86
A. BARAK
v.
1. Y. SARID
2 .D. ZUCKER, SECRETARY-GENERAL OF CITIZENS RIGHTS AND PEACE MOVEMENT
3. CITIZENS RIGHTS AND PEACE MOVEMENT
4. S. ALONI, M.K.
5. R. COHEN, M.K.
6. MINISTER OF JUSTICE
7. ATTORNEY-GENERAL
8. INSPECTOR-GENERAL OF POLICE
M.A.A 320/86
In the Supreme Court sitting as the High Court of Justice
[6 August 1986]
Before: Justice Meir Shamgar, President
Justice Miriam Ben-Porat, Deputy-President
Justice Aharon Barak.
Constitutional and Administrative Law - Pardon of offenders by President of the State - Presidential power to pardon unconvicted suspects - Basic Law: The President of the State. sec. 11(b) - Interpretation of Statutes -Meaning of the expression "to pardon offenders" - A "spacious interpretation" of constitutional provisions - Attorney-General's power to stay criminal proceedings - Presidential power of pardon and the powers of other State authorities relating to criminal justice -Pardon and Amnesty-High Court of Justice - Locus Standi - Amenability of State President to jurisdiction of the courts - Indirect judicial review of Presidential functions - Rule of Law.
The petitions to the court related to the decision of the President of the State to pardon the Head of the General Security Service (G.S.S.) and three of his assistants in respect of all the offences attributed to them connected with the incident known as "bus no. 300". The pardons were granted by the President under sec. 11 (b) of the Basic Law: The President of the State, by which he is empowered "to pardon offenders and to lighten penalties by the reduction or commutation thereof." The principal issue raised in the petitions was whether the President had the power to pardon persons before conviction. The court was also asked to order the competent authorities to investigate the incident referred to.
1. In regard to the interpretation of sec. 11(b) of the Basic Law, the court examined the import of the terms "offence" and "offender," and reviewed the legislative background to the enactment of sec. 11(b) above, in particular Article 16 of the Palestine Order in Council of 1922 and sec. 6 of the Transition Law, 1949. The court also considered the influence on the interpretation of sec. 11(b) of the Basic Law of the Anglo - American Practice in regard to pardon, as well as the place of the Presidential power of pardon in Israel in relation to the powers of other State authorities charged with the administration of criminal justice.
Held, per Shamgar P. (Miriam Ben-Porat D.P. concurring):
Having regard to the legislative purpose in the light of the above considerations, to the need for a "spacious" interpretation in matters of constitutional content, and to the accepted construction of the pardoning power since enacted in its original form, sec. 11(b) was to be interpreted as empowering the State President to pardon offenders both before and after conviction.
Per M. Ben-Porat D.P.:
The grant of a pardon involves a conflict between two very important interests: one - equality before the law, which requires that every offender against the law should answer for his conduct; the other - the safeguarding of a vital public interest. The proper balance between the two is the determining factor and the State President was faced with the same predicament when making his pardoning decision.
Minority opinion in A. v. The Law Council [2] and the decision in Attorney-General v. Matana [3] followed:
Per Barak J., dissenting:
Upon a proper interpretation of sec. l l(b) of the Basic Law: The President of the State, the Israel Legislature cannot be presumed to have favoured Presidential intervention in criminal proceedings before these have run their full course. Under the Israel "constitutional scheme" the Presidential power of pardon must not be construed as a paramount power, or as rivaling the powers of other State authorities, such as the police, the prosecution, the courts. It must be construed as a residual or a "reserve" power to be exercised only after the other authorities concerned have exhausted their own powers, i.e. after conviction of the accused. The pardons granted in the present case therefore are void and of no effect.
2. Held by the court (per Shamgar P.):
(a) The absence of a real personal interest on the part of any of the petitioners, does not justify the immediate dismissal of the petition. The Supreme Court will take a liberal view on this aspect and grant access to petitioners where the question that arises is "of a constitutional character" or of "public interest related directly to the advance of the rule of law". This entails no general recognition of the actio popularis, only a general guideline that enables the court to open its doors in suitable cases of a public-constitutional character.
(b) In granting the pardons, the State President was acting in a matter "connected with his functions and powers" as provided in sec 13 of the Basic law: The President of the State. Hence he is not amenable to the jurisdiction of the courts in connection therewith, including the Supreme Court's powers of direct review - its authority to demand of the president himself an explanation of his decisions. This immunity relates to the direct challenge of any Presidential act, but there is no obstacle to indirect judicial review of the President's discharge of his functions in proper cases and when the proceedings are directed against some other respondent.
Israel cases referred to:
[1] H.C. 73/85, "Kach"Faction v. Chairman of the Knesset 39(3) P.D. 141.
[2] H.C. 177/50, A. v. Chairman and Members of the Law Counci15 P.D. 137.
[3] F.H. 13/60, Attorney-General v. Matana 16(1) P.D. 430; S.J., vol. IV, p.112.
[4] H.C. 249/82, Wakhnin v. Military Appeals Tribuna1 37(2) P.D. 393.
[5] Cr. A. 224/85, Alba Pharmacy Ltd. v. State of Israel 39(4) P.D. 798.
[6] H.C. 156/56, Schor v. Attorney-General 11 P.D. 285; 21 P.E. 227.
[7] H.C. 329/81; M.A. 217/82, 376, 670/83, Nof v. Attorney-General 37(4) P.D. 326.
[8] Cr. A. 117/50, Haddad v. Attorney-General 5 P.D. 1413; I P.E. 318.
[9] H.C. 171/69, Filtzer v. Minister of Finance 24(1) P.D. 113.
[10] H.C. 228/84, unpublished.
[11] H.C. 270/85, unpublished.
[12] H.C. 659/85, Bar Yosef (Yoskovitz) v. Minister of Police 40(1) P.D. 785.
[13] H.C. 297/82, Berger v. Minister of the Interior 37(3) P.D. 29.
[14] H.C. 483/77, Barzilai v. Prime Minister of Israel et al. 31(3) P.D. 671.
[15] H.C. 652/81, Y. Sarid M.K. v. Knesset Chairman Savidor 36(2) P.D. 197.
[16] H.C. 40/70, Becker v. Minister of Defence 24(1) P.D. 238.
[17] H.C. 217/80, Segal v. Minister of the Interior 34(4) P.D. 429.
[18] H.C. 1/81, Shiran v. Broadcast Authority 35(3) P.D. 365.
[19] E.A. 23/84, Neiman et al v. Chairman of the Eleventh Knesset Elections Central Committee 39(2) P.D. 225.
[20] H.C. 186/65, Reiner v. Prime Minister of Israel et al. 19(2) P.D. 485.
[21] M.A. 838/84, Livni et al. v. State of Israe1 38(3) P.D. 729.
[22] H.C. 58/68, Shalit v, Minister of the Interior 23(2) P.D. 477; S.J., Spec. Vol. (1962-1969), 35.
[23] H.C. 390/79, Diukat v. Government of Israe134 (1) P.D. 1.
[24] H.C. 561/75, Ashkenazy v. Minister of Defence 30(3) P.D. 309.
[25] Cr.A. 185/59, Matana v. Attorney-General 14 P.D. 970.
[26] H.C. 742/84, Kahana v. Chairman of the Knesset 39(4) P.D. 85
[27] H.C. 94/62, Gold v. Minister of the Interior 16 P.D. 1846; S.J., vol. IV p. 175.
[28] C.A. 165/82, Kibbutz Hatzor v. Rehovot Tax Assessment Officer 39(2) P.D. 70.
[29] C.A. 481/73, Administrator of Estate Late E. Bergman v. Stossel 29(1) P.D. 505.
[30] H.C. 246,260/81, Agudat Derekh Eretz v. Broadcast Authority 35(4) P.D.1.
[31] H.C. 306/81, Flatto Sharon v. Knesset Committee 35(4) P.D. 118.
[32] H.C. 547/84, "Of Ha-emek" Registered Agricultural Cooperative Society v. Ramat Yishai Local Council 40(1) P.D. 113.
[33] H.C. 98/69, Bergman v. Minister ofFinance 23(1) P.D. 693.
[34] M.A. 67/84, Haddad v. Paz 39(1) P.D. 667.
[35] H.C. 507/81, Abu Hatzeira M.K. et al. v. Attorney-General 35(4) P.D. 561.
[36] C.A. 507/79, Raundnaf (Korn) v. Hakim 36(2) P.D. 757.
[37] H.C. 73,87/53, "Kol Haam"Co. Ltd. et al. v. Minister of the Interior 7 P.D. 871; 13 P.E. 422; S.J., vol. I, p. 90.
[38] C.A. 150/50, Kaufman v. Margines 6 P.D. 1005; 5 P.E. 526.
[39] C.A. 214/81, State of Israel v. Pahima 39(4) P.D. 821.
[40] H.C. 732/84, Tzaban v. Minister of Religious Affairs 40(4) P.D. 141.
English cases referred to:
[41] Reg. v. Boyes (1861) 9 Cox C.C. 32.
[42] R. v. Foster (1984) 2 All E.R. 678 (C.A.).
[43] McKendrick et al. v. Sinclair (1972) S.L.T. 110 (H.L.).
[44] Jennings v. United States (1982) 3 All E.R. 104 (Q.B.).
[45] Church Wardens & C. of Westham v. Fourth City Mutual Building Society (1892) 28 Q.B. 54.
[46] Thomas v. The Queen (1979) 2 All E.R. 142 (P.C.).
[47] Mistry Amar Singh v. Kulubya (1963) 3 All E.R. 499 (P.C.).
[48] Godden v. Hales (1686) 89 E.R. 1050 (K.B.).
American cases referred to:
[49] Ex parte Grossman (1925) 267 U.S. 87; 45 S.Ct. 332; 69 Law Ed. 527.
[50] M'Culloch v. Maryland (1819) 4 Law Ed. 579; 17 U.S. 316.
[51] Youngstown Sheet and Tube Co. v. Sawyer (1952) 26 A.L.R. 2d. 1378; 343 U.S. 579; 96 Law Ed. 1153.
[52] Ex Parte Garland(1866) 71 U.S. 333.
[53] Burdick v. United States (1915) 236 U.S. 79; 59 Law Ed. 476.
[54] Murphy v. Ford (1975) 390 F. Supp. 1372.
[55] United States v. Wilson (1833) 32 U.S. 149.
[56] Biddle v. Perovich (1927) 274 U.S. 480.
[57] Montgomery v. Cleveland (1923) 32 A.L.R. 1151; 98 So. III.
[58] Schick v. Reed (1974) 419 U.S. 256.
[59] Osborn v. United States Bank (1824) 22 U.S. 738.
[60] New York v. United States (1951) 342 U.S. 882.
[61] Ex Parte Wells (1855) 15 Law Ed. 421; 18 How. 307.
The petitioner in H. C. 428/86 appeared in person; A. Gal- for the petitioners in H.C. 429/86; The petitioners in H.C. 431/86 appeared in person; A. Zichroni and I. Hanin - for the petitioners in H.C. 446/86; M. Soaked - for the petitioners in H.C. 448/86; H Langer- for the petitioners in H.C. 463/86; D. von Wiesel and A. Barak - for the petitioner in M.A. 320/86; Y. Harish, Attorney-General and Y. Ben-Or. Senior Assistant State Attorney and Director of the Criminal Department of the State Attorney's Office - for the respondents; Y. Arnon and D. Weisglas for A-D.
SHAMGAR P.
The Matter in Issue
1.(a) On 25 June 1986 the State President granted the Head of the General Security Service and three members of that Service a pardon in respect of
all the offences connected with the so-called bus no. 300 incident, and committed from the time of the incident on the night between 12 April and 13 April until the date of this Warrant.
The pardon was preceded by certain events which became the subject of debate for a period of several months, both in the Government and among the general public, centering mainly on the proper steps to be taken by the authorities in consequence of the stated offences, which had meanwhile become known collectively as the "bus no. 300 incident".
The President was apprised of the details of the matter in two conversations with the aforementioned Head of the Service, and the decision to grant the pardon followed formal requests to that end. The pardon was granted before any legal proceedings had been instituted in respect of the matters mentioned in the Warrant of Pardon. On the day the pardon was granted, the President made a public statement in which he explained the reasons for his decision, inter alia as follows:
Acting under the power vested in me by law, I have today granted Avraham Shalom, Head of the General Security Service, and three of his assistants, a full pardon in respect of every offence prima facie committed in connection with the "bus number 300 incident". I have so acted with a view to putting a stop to the "devils' dance" raging around the incident and so preventing further grave harm to the General Security Service. In so exercising my power, I have acted upon the recommendation of the Minister of Justice following a cabinet meeting held last night with the participation of the Attorney-General.
My decision was based on the deep conviction that it was for the good of the public and the State that our security be protected and the General Security Service spared the damage it would suffer from a continuation of the controversy surrounding the incident. This Service is charged with waging the difficult war against terrorism, and the remarkable work of its members saves us tens of casualties every month. Last year alone the Service uncovered some 320 terrorist bands who were responsible for 379 outrages and attempted assaults in all parts of the country. So far this year the Service has exposed the perpetrators of 255 terrorist acts, apprehending among them also those who had committed murder. I wish to mention another aspect of the war against terrorism, which relates to the security arrangements for the protection of Israel's diplomatic missions and other agencies abroad. It may be recalled that only recently a murderous assault on an El-Al aircraft was prevented when members of the Service foiled the attempt to smuggle aboard a bomb in a suitcase in London. The public in Israel does not really know what debt we owe to all those anonymous heroes of the General Security Service, and how many lives have been saved thanks to their efforts."
As President of the State, I feel it my duty to rally to the support of members of the Service; knowing as I do the vitally important and arduous task fulfilled by them, devotedly and in secrecy, daily and hourly. I do so in the hope of preventing moral harm to the intelligence organisation and the security network, and to the war against terror.
In the special conditions of the State of Israel we cannot allow ourselves any relaxation of effort, nor permit any damage to be caused to the defence establishment and to those loyal men who guard our people.
The effect of the Attorney-General's unequivocal intimation at the cabinet meeting, that there was no alternative but to open a police investigation into the incident, was to create a situation which requires members of the Service to submit to the investigation without being left any possibility of defending themselves, short of divulging security information of the utmost secrecy. In this situation I saw it as my primary duty to act as I have done in protecting the interests of the public and the security of the State.
(b) Two principal issues have been raised in these petitions. The first concerns the President's power to pardon an offender before his trial and conviction; the second relates to a demand for the holding of an investigation into the events known as the "bus no. 300 incident". Concerning the latter issue, on 15 July 1986 we received the Attorney General's intimation, confirmed by the Inspector General of Police, that the police would investigate the complaints lodged in connection with that incident.
We accordingly have to deal here with the scope of the presidential power of pardon, and shall refer also to two related matters, namely: the reasons for our decision on 30 June 1986 concerning joinder of the President as a respondent in three of the petitions (H.C. 431/86, H.C. 446/86 and H.C. 463/86); and our ruling on the question of locus standi.
2. The President's power of pardon is defined in sec. 11(b) of the Basic Law: The President of the State (1964) (hereinafter "the Basic Law"), thus:
The President of the State shall have power to pardon offenders and to lighten penalties by the reduction or commutation thereof.
This legal provision is formulated to deal with two subjects, i.e. the pardoning of offenders and the lightening of penalties. The latter subject is not in issue in the present matter, so what remains to be decided here is whether the power to pardon offenders extends also to someone who has not yet been convicted.
The Approach in the Case Law
3. (a) For the reasons I shall set out below, it has to be concluded that in the case now before us the State President based the exercise of his power under the above section upon a legal construction in accordance with the accepted approach of this court for the past thirty-five years, which sees the President as empowered to grant a pardon also before conviction. I am not unaware of efforts made over the years by a few distinguished jurists to challenge that interpretative approach. That even they, however, have had to take the view enunciated by this court as the starting point of their analysis, is understandable and clear, for the reason succinctly stated by my learned colleague, Barak J., in "Kach"Faction v. Speaker of the Knesset [1], at p. 152:
In a democratic regime, based on the separation of powers, the authority to construe all legislative enactments - from Basic Laws to regulations and orders - is entrusted to the court.... Inherent in every statutory provision, naturally and axiomatically, is a delegation of the interpretative authority to the court. It is true that every state organ - and in the present context also every individual - will seek to interpret the law in order to plan ahead. In the case of certain organs, it is sometimes customary for the interpretative authority to be entrusted to a particular functionary. Thus, for instance, the Executive's interpretative function is entrusted to the Attorney-General, and his construction binds the Executive internally. But where the question of interpretation arises in court, this authority rests with the court and its construction will bind the parties. And where the construction is that of the Supreme Court, it will bind everyone (by virtue of the doctrine of stare decisis - sec. 20(b) of the Basic Law: The Judicature). In this regard Shamgar J. remarked as follows:
"Every governmental authority must on occasion contend with the interpretation of a legislative enactment, for the application of statutory law frequently (in theory invariably) entails the formulation of an attitude as to its substance and content. But the final and decisive interpretative word respecting the law in force at any given time, rests with the court; and in respect to issues raised for deliberation within the legal system, this last word rests with the supreme judicial tribunal" (H.C. 306/81, at p. 141).
Any other approach would strike at the very heart of the judicial process and completely undermine the doctrine of the separation of powers, and the checks and balances between them. Hence, relations both between the Judiciary and the Executive and between the Judiciary and the Legislature, are governed by the principle that the binding interpretation is that given by the court, no matter what other interpretations may be given.
I might add that the reservations expressed about this court's approach have related, as we shall see presently, to the recognition not only of a pardon before conviction, but also a pardon after serving the punishment and in other cases (see Prof. S.Z. Feller, "Rehabilitation", Mishpatim, 113 [1969], 497, 507). In fact, a complete alternative system of pardon has been prepared, and this proposed legislative revision is deserving of deeper study and deliberation than is possible in the framework of this judgment. (See the proposed Bills in the appendix to Prof. Feller's abovementioned article, which also include a proposed rephrasing of sec. 11(b) of the Basic Law by replacing the words "to pardon offenders" with an expression connoting remission of punishments.) Such further study of the subject would accordingly be important for shaping the desirable law in the future, if indeed it is sought to depart from the existing arrangement.
(b) I shall now review the pronouncements of this court on the subject of pre-conviction pardoning under the prevailing law. The subject was first mentioned in A. v. The Law Council [2]. The petitioner asked for the restoration of his name to the Roll of Advocates following upon a presidential pardon granted him (after having served his sentence of imprisonment) in respect of the offence for which his name had been removed from the Roll. The petition was dismissed by a majority of the Court, for reasons relating both to the powers of the Law Council and to the Supreme Court's modes of exercising its discretion.
Justice Agranat was the only member of the court to address the question of the pardoning power. The statutory provision underlying the President's power of pardon at that time (sec. 6 of the Transition Law, 1949) was phrased, so far as is relevant here, in language identical to that in sec. 11(b) of the Basic Law. It appears from the judgment of Agranat J. (as he then was) that he saw the power of pardon conferred on the President of the State of Israel as generally parallel to that vested in the King of England or in the President of the United States, whether in underlying perception, in nature and scope, or in the consequences of its exercise. In this connection the learned Judge referred, inter alia, to a statement in Halsbury's Laws (2nd. ed., Hailsham, vol. 6, p. 477) that "Pardon may, in general be granted either before or after conviction." Also referred to was the decision of the U.S. Supreme Court in Ex Parte Grossman (1925) [49], where it was held with regard to the Constitutional power to grant pardons for offences, that the Executive could grant a pardon for an offence at any time after its commission. Summing up his opinion on the scope of the power of pardon in Israel, Agranat J. clearly held that the President has the power to pardon offenders either before or after conviction (at p. 751; my italics - M.S.).
It is true, of course, that the question of the power of pardon before conviction was not part of the ratio on which Justice Agranat founded his decision in that case. Nevertheless, the wider question of principle involving the substance and scope of the pardoning power, the matter of its historical roots and its present day construction, on which the decision of Agranat J. was founded, encompassed also this specific aspect of the exercise of the power before conviction. This aspect arose directly out of and became an integral part of the interpretative method adopted. That is to say, inherent in Justice Agranat's adoption of the view that the President's pardoning power was the same as that of the British King or the American President, was the conclusion that the definition of that power likewise derived from the interpretative process on which the learned Judge had founded his decision, as he himself in fact noted.
Justice Agranat's abovementioned opinion has come to be recognized as representing the prevailing and commonly accepted interpretation of this court, whether this be due to the fact that no contrary judicial opinion on the matter has been expressed or whether this be attributed to the Further Hearing in the Matana case, a landmark decision in our constitutional law to which I shall presently return.
(c) This subject arose again in the rehearing in Attorney-General v. Matana [3], representing the leading and most comprehensive decision so far on the power of pardon. Once again the substance of the power was analysed, this time in the court's full consciousness that the decision which had occasioned the rehearing amounted to a rejection of the minority opinion of Agranat J. in A. v. The Law Council, insofar as he had found a parallel between the power of the President of Israel and that of the British Monarch. In his judgment in the Further Hearing, Deputy President Agranat (as he then was) reiterated his view expressed in A. v. The Law Council that the President's power of pardon was exercisable also before conviction. He noted that while there was indeed no room for an equation of the President's power of pardon with that of the High Commissioner of Palestine (under Art.l6 of the Order in Council, 1922), he also had no hesitation in reaffirming his approach in A. v. The Law Council as regards the scope of the President's power and its comparison with that of the Executive under the corresponding Anglo-American constitutional law. The power under sec. 6 of the Transition Law, 1969 (which for our present purpose is the same as that set forth in sec. 11(b) of the Basic Law) was termed by Agranat D.P. an "original" power forming part of a "Constitution in miniature of an independent State." Hence it was not comparable to the pardoning power instituted under the Mandatory legislation, and the model for comparison was the power of the British or the American Head of State.
In this regard the learned Deputy President added, by way of an interpretative guideline, that in view of its constitutional content the statutory provision concerned did not need a restrictive interpretation ( M'Culloch v. Maryland (1819) [50] at p. 602; Youngstown Sheet & Tube Co. v. Sawyer (1952) [51] at 1399, per Frankfurter J.). From the Youngstown case Agranat D.P. cited the statement of Jackson J. that because the American President enjoyed only those powers mentioned in the constitution it "does not mean that the mentioned ones should be narrowed by a niggardly construction."
In short, it emerges from the Deputy President's judgment that while the relevant provision did indeed relate to a new and independent legislative enactment, for the proper understanding of its substance it was nevertheless permissible to refer also to the corresponding powers that existed in the countries looked upon as the principal models for comparison, and which had nourished and shaped our own legislation.
Cohn J. (as he then was) - who together with Silberg J. concurred with Agranat D.P. in forming the majority opinion of the court - wrote a separate opinion stressing certain matters which appear to be particularly relevant in relation to the background of the problem now before the court, inter alia holding as follows (at p. 462):
Under sec. 6 of the Transition Law, 1949, the President of the State is empowered to pardon offenders and to reduce punishments. The Presidents of the State have exercised this power from 1949 until the day on which judgment was delivered in Matana v. Attorney-General (23 June 1960), in the manner laid down for them in the judgment of this court (per Agranat J.) in A. v. The Law Council (at p. 745 et seq.). That is to say, both the President of the State and the Minister of Justice, whose countersignature of the President's decision is required by the Law, and also the wide body of citizens who have had need of the President's grace, have always regarded this power of the President as equal and parallel in nature and scope to the power of pardon and reduction of punishments possessed by the Queen of England, and which was possessed by the High Commissioner of Palestine. It has already been said more than once by this court (both during the Mandate and after the establishment of the State) that the court will hesitate very much to reverse a particular practice which has taken root during the years, and if this was said in respect of matters of practice which did not rest upon the authority of judicial precedent, how much more is it applicable to a matter of practice which rests upon a specific decision of the Supreme Court. As for myself, even if I were inclined to agree with the opinion held by my colleagues Berinson J. and Landau J. that the practice followed by the President of the State year after year is based upon too wide an interpretation of sec. 6 of the Transition Law, 1949, even then I would not venture today to change this practice which has received the seal of the Knesset at least by its silence, and more especially since the practice followed by the President of the State adds the "grace" extended by him to its citizens.
I have no doubt, however, that the said provision in sec. 6 should be given a wide and not a narrow interpretation. But for the principles laid down in A v. The Law Council, which the Deputy President has again adopted in his instructive judgment in this Further Hearing, I would perhaps have gone further and interpreted the said provision even without reference to the powers of the King of England under the common law, which were also given to the High Commissioner of Palestine by virtue of the Order in Council, 1922. For the purposes of the decision in the present case, however, the principles laid down in the judgment referred to are sufficient for me too, and I arrive at the same conclusions as those reached by my colleague, the Deputy President, but without resort to the English and American authorities which he cites in his judgment.
(d) In a dissenting judgment Berinson J. disputed the abovementioned interpretative theses. In essence, however, and notwithstanding the divergence between the minority view (of himself and Landau J.) and the majority view as to the President's power of substituting one sentence for another, even he was expressly of opinion that the President's pardoning power extended also to an act for which the offender had not yet been tried and convicted. In this sense, as Berinson J. expressly pointed out, the President's power was wider than that of the High Commissioner at the time:
Moreover the President's power of pardon is in a certain sense wider than that possessed by the High Commissioner. Whereas the High Commissioner was unable to pardon a crime before the offender was tried and convicted unless he turned King's evidence and led to the conviction of his accomplice (the first part of Article 16 of the Order in Council), the President is not bound by this condition and, so it seems to me, may pardon any offender even before he is brought to trial (ibid. p. 469).
4. Recently Justice Cohn has had further occasion to express his opinion on the subject ("Symposium on Pardon," hereinafter "Symposium," Mishpatim 15/1 [1984], 14). It was decided law, in his view, that it was never intended by the Israel lawgiver - whether in the Transition Law, 1949,or in the Basic Law - to curtail the scope of the pardoning powers vested in the King of England under the constitutional conventions; it followed that the power to pardon offenders before their conviction availed also in Israel.
5. (a) In view of the reference in our decisions to the Anglo-American comparative model, it is fitting that we supplement our above remarks with a brief review of the law of those countries on our present subject. It is consistently asserted in the literature of English constitutional law, that the King is empowered under the common law to grant also a pre-conviction pardon. It is so stated in Blackstone's Commentaries on the Laws of England (San Francisco, 1916; vol. II, p. 400). In Halsbury's Laws (4th ed., vol. VIII, 8, par. 949, p. 606) it is stated:
In general, pardon may be granted either before or after conviction.
S.A. De Smith opines that "a pardon may be granted before conviction" (Constitutional and Administrative Law, 5th ed., Street and Brazier, 1985, p. 150, note 121). He holds that this prerogative power, though not exercised today, has not become abrogated by disuse and, like Sleeping Beauty, "it can be revived in propitious circumstances" (p. 143). In other words, in exceptional circumstances which so justify, the King may conceivably have renewed recourse to this power. A like view is expressed by O. Hood Phillips - "A pardon may generally be granted before or after a conviction" ( Constitutional and Administrative Law, 6th ed., 1978, p. 378). English decisions and treatises on the subject are replete with statements to the same effect and one need not repeat them all here (see Reg. v. Boyes, [41] ).
(b) In the U.S.A. the pre-conviction pardoning power is clearly enunciated in the classical work on the U. S. Constitution prepared by the Research Service and the Library of Congress: The Constitution of the United States of America, Analysis and Interpretation (Washington, 1973), p. 474. In Am. Jur. 59, 2d (Rochester & San Francisco, 1971) par. 25, the presidential power of pre-conviction pardoning is explicitly asserted, and with regard to the separate States it is added:
if the constitution does not expressly prohibit the exercise of the power until after conviction, it may be exercised at any time after the commission of an offense before legal proceedings are taken.
That is to say, the customary interpretation is that any State wishing to preclude the grant of a pre-conviction pardon has to make express constitutional provision to that effect, and a power of pardon mentioned without such a reservation means that it may be exercised also before conviction of the offender. See also W.W. Willoughby, The Constitutional Law of the United States (New York, 2nd ed. 1929), vol. III, at p. 1491; B. Schwartz, A Commentary on the Constitution of the United State (New York, 1963), vol. II at p. 87; B. Schwartz, Constitutional Law (N.Y. and London, 2nd ed. 1979) at p. 198; L.H. Tribe, American Constitutional Law (Mineola, 1978) at p. 191.
In the American precedents the power of pre-conviction pardoning is constantly reiterated. In the celebrated case of Ex Parte Garland (1866) [52], it was held (at p. 380) that the pardoning power
….extends to every offence known to the law and may be exercised at any time after its commission either before legal proceedings are taken, or during their pendency or after conviction and judgment.
See also L.B. Boudin, "The Presidential Pardons of James R. Hoffa and Richard M. Nixon: Have the Limitations on the Pardon Power been Exceeded?" Un. Colo. L. Rev. 48 (1976/77), p. 1.
6. In Attorney-General v. Matana [3] the majority agreed with Agranat D.P. as regards the substance of the pardoning power. In the wake of this decision came a legislative development which also throws some light on the attitude of the authorities at the time to the abovementioned legal questions. When the Bill for the Basic Law: The President of the State came before the Knesset, it was decided to vary the text of the provision concerning pardon in a manner clearly enunciating the President's power to commute sentences, this being a matter on which divided opinions had been expressed in the Matana rehearing. However, no attempt at all was made to vary the existing statutory arrangement so far as its interpretation in the rehearing was concerned - both by Agranat D.P. who expressed the majority opinion and by Berinson J. - as empowering the President also to grant pre-conviction pardons. The amendments pertaining to the commutation of sentences clearly stemmed from the wish to eliminate possible doubt resulting from the divergent interpretations on this point in the Matana case. Thus, for instance, Dov Joseph, Minister of Justice at the time, had this to say in support of the proposed amendment:
As to the presidential power to pardon offenders dealt with in sec. 6 of the Transition Law, the matter is now regulated in sec. 12 of the proposed new Law. The change in relation to the existing Law is that the new provision expresses the commutation of punishments to be, along with their reduction, a presidential power of pardon. This is no fundamental change, for even under the existing Law, which mentions the reduction of punishments but not their commutation, the latter was held to fall within the purview of the pardoning power vested in the President. However, since the Supreme Court arrived at this conclusion after much toil, with a minority of the Justices holding otherwise, we thought it desirable to clarify in the proposed new provision that such was the lawgiver's intention from the start (Minutes of the Knesset, 36 (1963/4), 964).
Also of interest in this connection are the comments made in the same debate by Prof. Y. H. Klinghoffer:
In a decision of the Supreme Court a year ago, it was decided - as already indicated by the Minister of Justice in his opening remarks - that sec. 6 embodied a power of substitution of a lighter punishment for the one imposed, and in particular to commute a sentence of imprisonment to one of conditional imprisonment" ( ibid., p. 966).
A year later (in 1965) Professor Klinghoffer rested on the same foundation his argument that if difficulties were to arise in consequence of the adoption of a certain proposal raised by him for debate, recourse could be had to the pre-conviction pardoning power in order to solve hard personal problems of punishment for which no other solution was available:
Another unconvincing argument advanced is that if a suspect be very ill, it would be an act of cruelty to put him on trial. Unconvincing, because in rare cases of this kind the President of the State would be able to grant a pardon before the trial commenced. The President is empowered to pardon offenders either before or after conviction. That was laid down by Justice Agranat in the case of A. v. The Law Council, H.C. 177/50 ( Minutes of the Knesset, 43(1965), 2319):
Statements made in the course of Knesset debates do not, of course, bind the court when construing the law, let alone the fact that here we are seeking mainly to draw an inference from the non-amendment of the existing statutory arrangement on the subject. We do accept, however, that a particular enactment's legislative history may be a valuable aid in its interpretation (Wakhnin v. Military Appeals Tribunal [4] at p. 424), and we may be so guided here too. An inference may properly be drawn from the fact that at the time when the Knesset debated the implications of the decision in A. v. The Law Council, not a single voice was raised in favour of narrowing the President's power in 'respect of pre-conviction pardoning. Legislatively speaking, it has so far not been seen fit to disturb the wording of the pertinent provision nor, indeed, its manner of interpretation - by Justice Agranat in A. v. The Law Council and by the Justices of the majority as well as the minority opinion in A. G. v. Matana- as empowering the President to pardon offenders also before conviction. The opportune time for having effected an amendment in curtailment of the pardoning power, should anyone have disapproved of its judicial interpretation, was surely at the stage when the Law was amended anyway to clarify operation of the pardoning power in a different respect, as already mentioned. The fact that the power as it stood was left intact in relation to the matter of our present inquiry, is proof that neither the Legislature nor the Executive saw fit to alter the legal situation that emerged from the expansive interpretation given the pardoning power in A.G. v. Matana.
Incidentally, though at the time there may have been room for debating whether or not the provisions of the Transition Law, 1949, including sec. 6, were endowed with any permanent constitutional standing, there could certainly have been no doubt as to the constitutional content and standing of the Basic Law: The President of the State. Yet sec. 11(b) thereof repeats verbatim the part of sec. 6 of the Transition Law that is pertinent to our present inquiry, and which the court construed as it did in the Law Council and Matana cases.
7. The fact that the Knesset did not vary the court's construction of the power "to pardon offenders" as embracing also pre-conviction pardoning, has contributed to a general recognition of the approach in the two precedents cited as the accepted approach on this subject. Prof. A. Rubinstein, for instance, writes as follows ( The Constitutional Law of Israel, 3rd ed., at p. 394; in Hebrew):
(e) Pardoning of offenders before conviction. The Law does not restrict the President's power to pardon offenders, and he may do so even before they have been convicted. English law is the same as regards the prerogative of pardon of the British Crown. The High Commissioner, however, was delegated only a part of the stated royal power and, in terms of Art.16 of the Order in Council, 1922, was able to pardon offenders only upon their conviction. In this respect the President's power is like that of the British Crown. Even Justice Berinson who interpreted the presidential power narrowly in the Matana case, held that "he has power to pardon any offender also before he is brought to trial."
A more restrictive construction of the expression "to pardon offenders" - even if centering more on the meaning of the term "offenders," which point I shall presently discuss in greater detail - was proposed by Prof. Klinghoffer at a symposium on this subject, though with express acknowledgment that his own view differed from the interpretation given by the Supreme Court (see Prof. Y. Klinghoffer, "Pardon's Constitutional Framework," Lectures at the Symposium "Amnesty in Israel", held in Jerusalem on 13-14 May 1968, Publications of the Hebrew University Institute of Criminology, 2, 5; hereinafter "Lectures on Amnesty"). A similar view was expressed by Prof. S.Z. Feller in his abovementioned article, "Rehabilitation" (at p. 507, note 28). Also present at the symposium was the then incumbent Attorney-General, M. Ben-Zeev, who made these observations:
Prof. Klinghoffer's constitutional analysis of the pardoning power is undoubtedly comprehensive and interesting. I feel it necessary, however, to mention one matter on which I disagree with him - if only to ensure that a different opinion also be heard on this important point. Prof. Klinghoffer interpreted the President's power "to pardon offenders" as applying only to convicted offenders and not to unconvicted suspects, a conclusion felt by him to flow from the very expression here used. Since in our law a person is presumed innocent until convicted he cannot be an "offender" until he is convicted; therefore, in Prof. Klinghoffer's opinion, the President has no power to pardon any person before he has been convicted. In this connection I might mention the case of A. v. The Law Council, cited by Prof. Klinghoffer, in which the wording of the relevant provision in the Transition Law was construed - and from which wording there was no departure, in the instant context, in sec. 11(b) of the Basic Law: The President of the State. This identical expression in both the above enactments was interpreted by Justice Agranat, after lengthy analysis in his abovementioned judgment, to mean that the President "has the power to pardon offenders either before or after conviction." This authority accordingly contradicts the approach of Prof. Klinghoffer and I myself, in my capacity as Attorney-General, have relied on this authority in giving my opinion that the President may pardon offenders also before conviction. The term "offender" obviously cannot be understood here to mean someone who has been duly convicted, but rather someone who comes to the President saying: "I have committed an offence and I ask you to pardon me." It is inconceivable that a person should come before the President and say: "I have not in fact committed an offence, but if I have, please grant me a pardon." Such alternative kind of pardon naturally finds no place in our law. But if a person should come and say that he has committed an offence for which he asks to be pardoned, then he is among the offenders whom the President has the power to pardon ( ibid., p. 53).
Another participant in the abovementioned symposium was Dr. Leslie Sebba of the Criminological Institute of the Hebrew University, Jerusalem, who made this comment on the legal situation as portrayed there ("Summary of the Lectures at the Symposium 'Amnesty in Israel' " p.x):
Finally, there was some doubt as to the proper interpretation of the phrase "to pardon offenders." Did this include persons not yet convicted? In the opinion of the Government, which based its view on judicial opinion, such persons could be regarded as offenders for the purpose of the pardon, for the request for a pardon could in itself be regarded as an admission of the offence .
This leads us to Dr. Sebba's illuminating work, On Pardon and Amnesty: Juridical and Penological Aspects (Ph.D. dissertation, Faculty of Law of the Hebrew University, Jerusalem, 1975; hereinafter On Pardon and Amnesty), in which the writer examined the scope of the pardoning power in Israel, inter alia making this comment (at pp. 152-153):
Sometimes pardon before conviction or "advance pardon" is treated as an independent form of pardon. This power, which has a legal foundation in both English and American law (but not in France), is generally attributed also to the State President. This view is challenged, however, by Prof. Klinghoffer on the ground that everyone is presumed innocent until duly convicted: "Hence no person is an 'offender' until a final convicting judgment be given against him."
In our view, the law does indeed enable a pardon to be granted without prior conviction. For certain purposes the Legislature has seen fit to describe an unconvicted suspect as an offender, and the Supreme Court has also held that an unconvicted suspect may be deemed to have a "criminal past" under sec. 2(3) of the Law of Return, 1950 (see H.C. 94/62, Gold v. Min. of Interior, 16 P.D. 1846). Finally, Art. 16 of the Order in Council conferred express power to pardon a person who turned "King's evidence" and whose evidence led to conviction of the principal offender.
The introduction to Article 16 of the Order in Council clearly related to unconvicted suspects who were willing to give evidence for the prosecution. This was envisaged as the main area for application of this provision, which, as in English law for the past centuries, has been the main justification for preserving the power of "advance" or pre-conviction pardon.
8. We have so far described the ruling interpretative thesis. Having regard, however, to the arguments advanced during the hearing, it is necessary to examine the reservations and doubts raised as to the President's power to pardon unconvicted offenders under sec. 11(b) of the Basic Law. We shall pursue this examination, and the formulation of our interpretative standpoint regarding sec. 11(b), along three principal lines of inquiry: first, the linguistic import of the statutory provisions under consideration; second, the contention as to abrogation of the disputed power following the enactment of other, new powers; and third, the juridical-constitutional substance of the power.
The Language of the Section
9. Sec. 11(b) of the Basic Law speaks of the power "to pardon offenders." There is no definition of the term "offender" in either the Basic Law or the Penal Law of 1977. As already mentioned, Prof. Klinghoffer founded his narrow interpretation of the presidential power on the perception that the term "offender" applied only to someone duly tried and convicted (see "Lectures on Amnesty",supra; "Symposium," at p. 5). The like opinion was expressed by H. Zadok, former Minister of Justice (ibid., p. 9) and by Prof. S.Z. Feller (ibid., p. 10; and see also "Rehabilitation," at p. 507). Disagreeing with these views, Justice H. Cohn argued that the existence of a pre-conviction pardoning power was also indicated in the language of the Law. He commented that an enactment aimed at expanding civil rights and benefits called for a spacious and liberal interpretation. For purposes of the President's power, an "offender," in his opinion, was anyone who testified to himself as being such “ Symposium," at p. 14).
I do not think the term "offender" must be understood as referring only to someone who has been tried and convicted. It is accepted in this court that an expression in a particular Law must be interpreted in the light of its legislative context, as was held by my colleague the Deputy President in Alba Pharmacy Ltd. v. State of Israel [5] at p. 802:
Expressions and directives in a Law must be interpreted in the light of the purpose it is intended to achieve. Hence it will sometimes happen that the identical expression appearing in different enactments is differently construed, all in accordance with the inherent purpose and intent of the enactment (C.A. 480/79, Treger v. Customs Collector, at p. 306).
According to its plain meaning, the term "offender" relates to someone who has committed an act defined as an offence, and from the word offence or offender itself one can hardly learn that it has no other legislative application than to someone proved, in final criminal proceedings ending in a conviction, to have committed an offence. Fundamental to our perception of criminal justice is the presumption that a person is innocent until duly proven guilty, but this presumption is an incident of the individual's rights and obligations in confrontation with the judicial process, or with any other authority, or individual. It does not necessarily reflect upon all the possible linguistic nuances of a descriptive term employed in a variety of legislative contexts and conjoined to a variety of eventualities in the penal law and related area. Not infrequently one finds mention in enactments of the term "offence" or "offender," when it is intended to refer simply to a criminal act or omission, or to the person to whom such is attributed, even though not yet convicted in criminal proceedings. And this is so even though the circumspection needed when a person's status may be affected in criminal proceedings, would seem to indicate the use of other expressions such as "a person accused of..." or "charged with..." or "alleged to have committed an offence," or like language. The Penal Law of 1977, for example, makes frequent mention of the term "offence" in a variety of contexts. But when it speaks in sec. 4 of bringing an offender to trial, the reference is clearly to someone charged with, and not already convicted of, the offence. (In like vein see also secs. 7, 8 and 10(d) of the Law.) In this connection Dr. L. Sebba refers ( On Pardon and Amnesty, at p. 153) to the Criminal Procedure (Arrest and Search) Ordinance (New Version), sec. 3(3) of which empowers a police officer to arrest a person without warrant if he "has committed in the police officer's presence, or has recently committed" a certain kind of offence; here too one is clearly dealing with someone suspected of committing the offence rather than someone already convicted thereof.
It is provided in sec. 3 of the Police Ordinance (New Version) that "the Israel Police shall be employed for the prevention and detection of offences, the apprehension and prosecution of offenders." There undoubtedly cannot be any reference here to already convicted offenders. The definition of the term offender in the Interpretation Ordinance (New Version), stresses the element of the sanction but nowhere mentions a finality of legal proceedings. And so one could without difficulty quote many more examples.
We might, for the purpose of our linguistic inquiry, also examine other provisions of law on matters which may be said to be in pari materia. In this respect the wording of sec. 6 of the Transition Law, 1949 sheds no additional light on the meaning of the term "offender" in the Basic Law. However, besides the individual pardon provided for in the Basic Law, two other Laws were enacted dealing with the subject of general amnesty. The first was the General Amnesty Ordinance of 1949, which in see 2 provided that a person who prior to a specified date "committed an offence... shall not be arrested, detained or prosecuted for it, or if he is already being prosecuted... the proceedings shall be discontinued and he shall not be punished." Clearly the words "...committed an offence" extended the benefit of the amnesty also to offenders who had not yet been tried and convicted. The wording of sec. 2 spoke for itself, and a statement to the same effect was made by the then Minister of Justice, Mr. Y.S. Shapiro, when introducing the Bill for the Amnesty Law of 1967 before the Knesset:
This is the second occasion on which a general amnesty is extended by the State directly through the legislature. The first time the general amnesty was granted by the Provisional Council of State in its final session, prior to the convention of the elected assembly - the First Knesset. In the Law passed at the time by the Provisional Council of State, it was laid down that any person who had committed an offence, other than one entailing sentence of death or life imprisonment, should receive a pardon, whether already tried and convicted or not (Minutes of the Knesset, 49, p. 2484).
In sec. 5 of the Amnesty Law of 1967, the second enactment of its kind, mention was again made of a "discontinuance of proceedings" taken in any court for "any offence committed" before a specified date. Thus the amnesty was once more extended in respect of "offences" for which the offender had not yet been tried or the proceedings concerning which had not yet been completed.
The manner of use of the term "offence" in a Knesset enactment dealing with a general amnesty has implications for the construction of the same term in an analogous Knesset enactment dealing with individual pardons.
In sum, it may be learned from a linguistic examination of pertinent statutory provisions, that the terms "offence" and "offender" may, according to the subject matter and context, simply import a criminal act or the person accused or suspected of having committed that act, and not necessarily a conviction, or a convicted offender. By analogy, the same term in sec. 11(b) of the Basic Law was intended to embrace also a person to whom a criminal act, attempt or omission is attributed, and not only someone already convicted of the same.
By way of comparison it may be noted that the same term mentioned in the constitution of the U.S.A., in the context of "pardons for offences" (art. II, sec. 2, clause 1), has also not been interpreted as applying solely to criminal conduct which is followed by trial and conviction.
Parallel Statutory Powers
10. It was a central argument of the petitioners that the power of pardon before conviction was abrogated by the effects of later, as it were, superseding legislation. This argument assumed diverse forms and I propose to deal with its different aspects. Since, for purposes of our present inquiry, it first found expression in a directive of the Attorney-General included in one of the petitions now before us, I shall start therewith.
11. (a) In his capacity as Attorney-General, Prof. Y. Zamir published a directive (no. 21.333) concerning the President's power to pardon offenders before completion of the trial. The learned writer first referred to the opinion of Justice Agranat in the case of A. v. The Law Council, contending that the equation there of the power of the High Commissioner with that of the British Crown was erroneous, as the former was not competent to pardon any person before his conviction. It followed that if the High Commissioner was not so empowered, no power of that nature could possibly have been conferred under sec. 6 of the Transition Law of 1949, when it was enacted.
The above conclusion as to an equality of pardoning power displayed indeed a certain inaccuracy, for the power delegated to a colonial Governor or to the High Commissioner of a Mandated Territory did not coincide with the prerogative power of the King. But this point was clarified in the Matana case and, I might add, the Attorney-General himself fell victim to an inaccuracy when writing that the High Commissioner had no power to pardon unconvicted offenders. For it was expressly provided in the first part of art. 16 of the Order in Council that the High Commissioner might pardon "any accomplice in such crime or offence who shall give such information and evidence as shall lead to the conviction of the principal offender." This empowerment therefore did not relate specifically to already convicted persons, and further proof is to be found in the continuation of art. 16, where express reference is made to convicted offenders in quite a different context.
(b) The stated directive was further predicated on the premise that the abovementioned statements of Justices Agranat and Berinson (on the instant issue) did not amount to binding precedent. I do not accept this reasoning, since it overlooks the connection between the court's overall decision, as already described above, and the specific conclusion concerning the power of pre-conviction pardoning. The existence of this nexus has not only been acknowledged in extra-judicial commentaries, but was also expressly mentioned by Justices Agranat and Berinson in their respective judgments in the Matana case, both clearly having regarded the power of pardon before conviction as flowing integrally from their underlying legal perception of the wider issue before them. The fact that the learned Justices saw fit to recognise the possibility of pre-conviction pardoning, is evidence that such recognition was a natural corollary of a viewpoint shared by the Judges of the majority as well as the minority opinion in the Matana rehearing. Moreover, the fact that the question of a pardon before conviction was directly addressed in the abovementioned decisions, even though the question was not directly in issue on the facts in either of the two cases concerned, is further evidence of a clear and patent connection seen between the essential pardoning power - as interpreted by the court - and the possibility of a pardon granted before conviction. What I am saying is that one has to examine the judgments of the majority opinion in the Matana rehearing according to their essential legal rationale, rather than merely answer the question whether the judgments dealt directly with the power to pardon before conviction. The ratio of the majority opinion in the Matana rehearing is to be found in the conclusion that the presidential power, although original and autonomous by virtue of an Israel enactment, was nevertheless shaped by and for its legislative purpose according to the Anglo-American model. At the same time the court added its conclusion that the Presidential pardoning power in Israel was equal in scope to that of the King of England, or of the President of the U.S.A. The details of the power, also in the pre-conviction contingency, were but a derivative legal consequence. It was the constitutional analogy with the corresponding Executive powers in the above two countries - whose legal systems, far more than others, have inspired and nourished our own legal and constitutional notions and doctrines - that gave birth to the conclusion that is now the subject of our deliberation.
A like opinion was expressed by Prof. C. Klein ("Symposium," at p. 17):
The source of the pardoning power is the royal prerogative. There is a clear connection between the method of pardon in Israel and the corresponding English method, from which one can learn about the scope of the presidential power of pardon in Israel (a divergence of opinion on this matter is echoed in the Matana case).
The power of pardon is not everywhere the same and, as we shall presently see, a variety of methods are followed in other countries. At the time, however, it was not the constitutions of such other countries that served as the models for shaping our own powers of pardon, so that no conclusion whatever can be drawn from any comparison with them, and their situation cannot now reflect on our own, except as an exercise in the desirable.
It would also be wrong to conclude from the analysis of principles in the abovementioned precedents that we are, as it were, held captive by our legal heritage and that we lack the vigour to fashion our own constitutional doctrines. Not so! Our essential constitutional form has throughout been autonomously our own, and remains so today. What is at stake is a historical-interpretative question that is concerned with the legal perspectives adopted at the time, with the constitutional result distilled from and founded on the same, and with the tenor of our precedents - representing, for some considerable time now, the accepted legal interpretation.
Of course, there always remains the possibility that the Legislature may be disposed to replace the existing order with a new arrangement considered more suited to our time. Interesting proposals to this effect have been made, some of them ranging in substance far beyond the limited question of our immediate inquiry. Only in an appropriate manner, however, should we abandon a chosen path of the Knesset and the legislative purpose enshrined in the relevant provisions of the Transition Law and the Basic Law, especially when the powers conferred thereunder are of known scope after lengthy judicial analysis and circumscription. We should take care that any material change contemplated be not impelled by passing events, however stormy their nature, but result from orderly constitutional research and discussion. Any change resolved upon should be effected in a manner showing proper deference to a constitutional norm followed for a comparatively long time, that is to say, it should be done by way of legislative enactment.
12. In his directive the Attorney-General founded his conclusions as to the scope of the presidential pardoning power largely upon its comparison with his own power to issue a nolle prosequi:
A wider use of the power to order a stay of criminal proceedings has always been made in Israel, and in recent years thousands of requests for such a stay have been lodged annually with the Attorney-General. In practice, therefore, the power to pardon accused persons before completion of their trial needs less to be exercised in Israel than in England....
A presidential power to intervene in criminal proceedings pending before the court, in a manner permitting termination of such proceedings at any time, is undesirable in principle. The pardoning power of the President bears no comparison with the Attorney-General's power to intervene in criminal proceedings by way of staying the same. The Attorney-General functions from the start as an integral factor in criminal proceedings, for he is empowered by law to prefer the charge on behalf of the State.... The President, on the other hand, is an extraneous factor in criminal proceedings. In this situation, his grant of a pardon in the course of a trial might be seen as an unwarranted intrusion into the domain of the court....
Any interpretation that would empower the President to pardon also unconvicted suspects, suffers in addition the practical disadvantage of a concurrence between this power and the power of the Attorney-General to stay the proceedings against such suspects.
In this connection it was contended that even in England the prerogative of pardon before conviction was no longer exercised. Accordingly, it was concluded in the directive that the President was competent to pardon only convicted persons, for the reason that his power to pardon unconvicted suspects had been replaced by the Attorney-General's power to order a stay of criminal proceedings.
13. The above argument is complex and involves, as we shall see presently, not only the matter of a nolle prosequi and its effects, but also other legal processes and their ramifications, including problems of interpretation. The full import of the argument is that a whole array of new penal laws enacted over the years have served to abrogate the power of pardon before conviction. It is true that the argument was not presented to us in precisely this form, but this was clearly its substance. It would be helpful, therefore, for us to dissect the argument into its component parts and different legal aspects, and to examine each in turn, namely:
(a) First, what is the nature of the order staying a criminal prosecution, and what are the points of similarity and difference between this step and the power of pardon before conviction?
(b) Second, what ground is there for the contention that the power of pardon before conviction is no longer existent in Anglo-American law?
(c) Third, what other relevant statutory provisions exist on the issue before us, even if not mentioned in the Attorney-General's directive, or in the arguments addressed to us?
(d) Fourth, can a constitutional directive deriving from statute or from the common law (where it exists) be considered to have been implicitly repealed or abrogated by later legislation dealing with the same subject?
(e) Fifth, does the emergence, in practice, of a pragmatic legal substitute for an existing constitutional arrangement, implicitly repeal the latter, and is there any difference for this purpose between a constitutional or legal arrangement deriving from express statutory provision, and one which is solely the creation of judicial interpretation?
14. (a) As regards the issue of a nolle prosequi, it is provided in sec. 231 of the Criminal Procedure Law (Consolidated Version) of 1982, that the Attorney-General may stay the proceedings by reasoned notice to the court at any time after lodging of the information and before judgment; upon such notice the court shall discontinue the proceedings in that trial. The Attorney-General may also delegate to his deputy - either generally or in respect of a particular matter or classes of matters - his power to stay any criminal prosecution except in cases of felony. In a recent Bill for the amendment of the above Law (no. 1703, p. 34) it is further envisaged that the Attorney-General may delegate this power - in charges other than felonies, preferred by a prosecutor who is not an attorney of the State Attorney's Department (e.g. a police prosecutor) - also to the State Attorney or his deputy, to the District Attorney, or to any other attorney of the State Attorney's Department given the power of a District Attorney under see. 242 of the above Law. In the explanatory notes to that Bill, it was stated that the Attorney-General and his deputies were experiencing difficulty in handling the greatly increased number of requests made for a stay, and hence the proposed widening of this delegatory power.
This proliferation of requests is apparently attributable, inter alia, to the proportion of such requests acceded to, and it is clear from the cited passage in the Attorney-General's directive, that thousands of requests are made annually in what seems latterly to have become a regular and widespread practice. I see no need to express any detailed opinion on the question (which is not in issue here) of the proper use of the power to stay a prosecution, and but for it having become interwoven with the argument now before the court, would have preferred not to deal with it at all. I shall merely state my lack of conviction that the wide use of the staying power, as described in the directive, is in accord with the lawgiver's underlying intentions and the character of the staying procedure. Any misgivings one might have with regard to the implications of a pre-trial exercise of the pardoning power, must also to a large extent accompany this phenomenon of a stay of criminal proceedings before the trial has run its course. We appear to be dealing here, not with the rare and exceptional exercise of a given power, but with a rapidly widening process which entails no public deliberation, no participation of any other authority, and which, by its very nature, allows little opportunity for judicial or any other kind of scrutiny (cf. K.C. Davis, Discretionary Justice, Baton Rouge, 1969, pp. 211-212). And, as already indicated, there are proposals afoot for a further, vertical, diffusion of this power.
(b) Sec. 232 of the same Law provides that, following a stay of proceedings under sec. 231, the Attorney-General may upon written notice to the court renew the proceedings, provided no more than a specified period has elapsed from the time of the stay. Upon such notice, the court renews the proceedings and may commence them again from the start or continue from the stage of their discontinuance. Upon a second stay of the same proceedings, they may not again be revived. This means that the first stay does not finally close the matter, for it does not preclude the revival of the proceedings within a specified period, and only thereafter is finality reached.
(c) The power to order a stay of proceedings is not to be seen as an institution of later legislative vintage than the pardoning power. The Attorney-General's power of stay did not first come into being in the Criminal Procedure Law of 1965, but existed before that under the Mandatory Art.16 of the Order in Council 1922. It continued to exist after the establishment of the State when the pardoning power was later re-enacted, first under sec. 6 of the Transition Law of 1949, and then under sec. 11(b) of the Basic Law of 1964 (see sec. 59 of the Criminal Procedure [Trial upon Information] Ordinance of 1924, and sec. 18 of the Magistrates Courts' Jurisdiction Ordinance of 1939). It follows that the theoretical parallel between the power to pardon and the power to order a stay of proceedings was there from the start - i.e. from the very inception of the pardoning power in its new constitutional guise after the establishment of the State - and that the power to stay a prosecution indeed antedated the Knesset's enactments on the power to pardon offenders.
This fact alone should suffice to controvert the proposition that the presidential power of pardon was abrogated or curtailed by a later conferment of power on the Attorney-General to order a stay of criminal proceedings. The latter power coexisted with Art. 16 of the Order in Council 1922, and was still operative when the power to pardon offenders was widened in the Transition Law of 1949 and in sec. 11(b) of the Basic Law. And the construction of the power conferred under these enactments, in Matana and in A. v. The Law Council, did not precede, but followed the creation of the Attorney-General's power of stay.
(d) On the relationship between the two powers, Justice Cohn, for instance, has said:
There is no similarity or parallel between the stated presidential power and the power of the Attorney-General to order a stay of proceedings: the one is a prerogative power, the other purely administrative; the one is subject to revocation and change at the Attorney-General's wish, the other is an act of much solemnity and by its very nature of rare and exceptional exercise ("Symposium," at p. 15).
I myself am not inclined to view the Attorney-General's power as being administrative. It relates to a criminal procedure involving the exercise of a quasi-judicial discretion (see Schor v. Attorney-General [6]; Nof v. Attorney-General [7]). The purpose of the staying function was to reserve for the chief prosecution authority the power to halt criminal proceedings, without this entailing the consequences set forth in sec. 93 of the consolidated version of the Criminal Procedure Law (withdrawal of the charge), but retaining the possibility of resuming the proceedings within a given period. However, I do recognize differences between this power and the power of pardon, which I shall summarise presently.
(e) There can be no full parallel between the power of stay and the power of pardon, since the former comes into play only after the suspect has been charged (sec. 231 of the Criminal Procedure Law [Consol. Version]). An immunity from prosecution promised a state witness who has not yet been charged, cannot be founded on the power of stay under see. 231, but only upon an Executive commitment or, if deemed fit, a pardon.
(f) To sum up, the points of difference between the two powers are the following:
(1) A stay of proceedings is inconclusive until expiry of the statutory prescribed period. A full and unconditional pardon, on the other hand, cannot be withdrawn (see Killinger, Kerper and Cromwell, Probation and Parole in the Criminal Justice System, St. Paul, 1976, p. 318).
(2) A stay of proceedings under sec. 231 is possible only after the suspect has been charged.
(3) A pardon (according to the decision in Matana ) acts to remove the stain of guilt utterly (in contrast, for example, to the prevailing approach in Britain, as expressed in R. v. Foster [42] and holding the pardon to wipe out only the consequences of the conviction; and see, in the U.S.A., Ex Parte Garland [52], and cf. Burdock vs. U.S. [53]; see also Killinger, Kerper and Cromwell, Probation and Parole, p. 322). A stay of proceedings is merely a trial procedure which, under the Criminal Procedure Law (Consolidated Version) of 1982, calls a halt on further activities from the time the stay is ordered, without any retroactive effect.
(4) As a trial procedure acting to halt the proceedings, the stay of a criminal prosecution is not unique, as appears from sec. 93 of the abovementioned Law concerning withdrawal of a charge by the prosecutor.
(5) It is necessary for the Attorney-General to give his reasons for issuing a stay of proceedings, whereas no reasons need be given for the issue of an instrument of pardon.
15. The comparison made with English law and the contended disuse of the pre-conviction pardoning power, as advanced in the Attorney-General's above directive no. 21.333, seems to show a confusion between the continued existence of a power and the frequency of its exercise. The fact of an abrogation of the royal prerogative to grant a pardon at any time after commission of the offence, is nowhere postulated in English legal writings. One view, stated for example by Hood Phillips (Constitutional and Administrative Law, p. 378) and by R.F.V. Heuston ( Essays in Constitutional Law, 2nd ed., London,l964, at p. 69), takes the form of a mere recital of the power as existing and valid, without any comment or reservation. Another view, advocated by De Smith, holds the prerogative power to be valid but slumbering, and capable of reawakening in special circumstances of need (Constitutional and Administrative.Law, at p. 150, n. 121):
It would seem that a pardon may be granted before conviction; but this power is not exercised.
Also (at 143):
In a Scottish appeal to the House of Lords ( McKendrick v. Sinclair [43] at pp. 116, 117 - M.S.), Lord Simon of Glaisdale said that "a rule of the English common law, once clearly established, does not become extinct merely by disuse"; it may "go into a cataleptic trance", but, like Sleeping Beauty, it can be revived "in propitious circumstances."
It is noteworthy that under the heading "Pardon" it is provided in sec. 9 of the English Criminal Law Act of 1967, that "nothing in this Act shall affect her Majesty's royal prerogative of mercy." As formulated, the section makes no distinction between classes of free pardon. It is at all events clear that the exercise of this prerogative power has greatly diminished in England. Already in 1926 Sir Edward Troup wrote ( The Home Office, 2nd. ed., 1926, p. 57) that the prerogative was not exercised before conviction except in rare cases where the pardon would enable an important witness to testify without incriminating himself in respect of a minor offence. There is reason to believe that since then the power has come to be even less frequently exercised. But, as I have already said, the existence of the power and the measure of its use are two separate matters.
The question of the continued existence of the prerogative power of pardon, alongside and notwithstanding the power to order a stay of proceedings, is discussed in an article written by A.T.H. Smith in which he states this conclusion ("The Prerogative of Mercy, the Power of Pardon and Criminal Justice," Pub. L. [Autumn 1983], 416-417):
Whether or not the power continues to exist is a matter of some conjecture, but the better view would seem to be that it does. It has certainly not been abrogated by statute, and although it is true that prerogative powers can be lost or modified merely by disuse, as in the case of the royal power to sit as a judge, the criteria for deciding whether or not a power has become "obsolete" are far from clear. As a general principle, the rules of the common law (of which the prerogative is undoubtedly part) do not lapse through desuetude or obsolescence. Even though the power does not at present seem to serve any identifiable constitutional purpose, the prerogative has proved itself to be a remarkably enduring power, and one that can reappear at unexpected moments, and until the advance pardon is expressly abrogated by statute, the possibility that its use will revive at some future time cannot be discounted.
In other words, the accepted view is that the prerogative power, which in England emanates from the common law and not from statute as does the power of pardon in Israel, has not been abrogated by disuse but continues to exist; moreover, neither in theory nor in practice is there anything to prevent its renewed use in special circumstances, and only an express statutory directive can extinguish its efficacy.
As for the situation in the U.S.A., it will be recalled that the presidential power of pardon was exercised on two recent, well-known occasions. On the first occasion it was exercised in favour of President Nixon (39 Fed. Reg. 32601-02 [1974]). (In this connection see Murphy v. Ford [54] in which the grounds for exercise of the power were discussed; see also Mark P. Zimmett, "The Law of Pardon," Annual Survey of American Law, 1974/5.) On the second occasion, in 1977, the power was exercised by President Carter in favour of evaders of conscription in the Vietnam War, i.e. a form of pardon for a class of persons and a class of offences, bearing the character of a partial "general" amnesty. The pardon was formulated to extend, inter alia, to "all persons who may have committed any offence between August 4, 1964 and March 28, 1973 in violation of the Military Selective Service Act or any rule or regulation promulgated thereunder" (42 Fed. Reg. 4391 [1977]). Bernard Schwarz writes of this pardon of 1977, "the legality [of which] was never questioned" (Constitutional Law, at p. 198). (On this pardon, see also G.S. Buchanan, "The Nature of a Pardon under the U.S. Constitution," Ohio St. L. J. 39 [1978], 36, 61, where the writer comes to the same conclusion.)
The English system, so far as it is of comparative significance, seems accordingly to invite a conclusion that is the opposite of the one we have been asked to draw. If in England a power stemming from the common law has not lapsed or become obsolete notwithstanding the lack of its use, how much less so in our own case involving a statutory power construed by the Supreme Court (in 1960, in the Matana rehearing) as a valid and existing power in the opinion of all.
As in England, so with us, the situation is one of a practical non-exercise of the debated power (see e.g. par. 5a of the Attorney-General's directive no. 62.100), rather than its explicit repeal. That is to say, the power itself has been reserved for use in exceptional cases, and its use knowingly restricted. The legal situation in the U.S.A. reflects a similar sparing use of the power of pardon before conviction, but its validity is clearly accepted there.
16. Our next assignment is to trace such other legal directives as may have a bearing on the aspect of pardon with which we are now concerned. In this regard Prof. Klinghoffer observed ("Lectures on Amnesty," at p. 7):
Showing mercy is not a monopolistic power of the State President. Other authorities too are competent to extend grace and clemency - as does the Attorney-General when issuing a nolle prosequi, or the Minister of Police when authorizing the early release of prisoners, with or without the recommendation of the competent board. The Military Justice Law likewise provides for the functioning of a penalty review board, with power to mitigate or substitute punishments, without derogation from the Presidential power of pardon. Already at the beginning of the lecture it was hinted that the contemporary trend is to permit the courts an increasing measure of mercy along with the doing of justice. Other matters affecting pardon and having constitutional implications, come into play when the quest for a pardon is pursued along the lines of a retrial.
As already indicated, the statutory creation of an alternative legal framework for some of the processes for which the pardoning power is now used - in its various forms, mainly after conviction but also before - was proposed by Prof. Feller in his abovementioned article, "Rehabilitation." His proposal included a draft Rehabilitation Law, a supplemented and expanded rehearing facility, and an express narrowing of the provisions of sec. 11(b) of the Basic Law so as to encompass only respite or remission of punishments not yet served (as distinct from a pardon in respect of the conviction). Prof. Feller proposed a synthesis between new legislation and amendments to existing enactments, aimed at supplementing the existing arrangements on matters such as a stay of proceedings, review of punishment, mitigation of punishment and retrial.
Some years ago the trend embodied in the above proposals gained momentum with the enactment of the Crime Register and Rehabilitation of Offenders Law of 1981, which provides, inter alia, for the automatic deletion from the register in certain cases of a person's previous convictions, for restricting the availability of information concerning the same, and like directives. In other words, we witness here the completion of part of the legislative program designed to establish new legal machinery that will give expression to and aid in the practical attainment of an equality of rehabilitative opportunity and rights.
In sum, we see in existence today a number of statutory provisions to reach many of the same results as are attained through exercise of the power of pardon. Some of these provisions precede the Knesset's enactment of the existing power of pardon and its judicial construction, for instance those concerning the prosecution's discretion as to charging a suspect, or the power to stay criminal proceedings. Other such provisions have sprung up contemporaneously and in coexistence with the statutory directives concerning pardon, such as the provisions relating to retrial and review of punishment, while more recently provision was made, as already mentioned, for expunging a person's criminal record. Thus some of the new provisions apply in the pre-conviction stage of the trial, while others - and these form the bulk - are applicable in the post-conviction stage, i.e. the stage where most of the decisions affecting exercise of the pardoning power are in practice made today.
17. We must now give attention to the fourth of our questions posed above, namely, the nature of the reciprocal tie between existing legislation and new legislation on the same subject or, more specifically: does the emergence of a new statutory arrangement alongside and overlapping an existing provision entail any abrogation of the latter?
The prevailing Anglo-American interpretative approach is to start on the premise that the lawgiver intends no tacit repeal of earlier enactments, particularly not when the enactments are all of modern date (see F.A.R. Bennion, Statutory Interpretation j London, 1984, p. 433, with reference to the decision in Jennings v. United States [44]; the same view is taken by R. Cross, Statutory Interpretation, London, 1976, 3). Incidentally, according to Cross. English law also does not recognize the possibility of abrogation of a law through desuetude, so that a statute will not cease to be valid merely on account of obsolescence. Generally speaking, express legislative direction is required for such invalidation.
It is interesting that a similar approach was advocated by Prof. Klinghoffer, speaking at the time in a Knesset debate (Minutes of the Knesset, 43 (1965), 2319):
It is not the function of the prosecution to determine whether certain provisions of the penal enactments have become a dead letter. As long as they remain inscribed in the statute book they must be observed, and if their further observance be undesirable, it is up to the lawgiver - and not the prosecution - to repeal them.
English law does not regard the mere concurrent existence of earlier and later legislative enactments on the same subject as warranting the inference of an implied repeal. This consequence flows only from contradiction between two enactments (see E.A. Driedger, Construction of Statutes, 2nd ed., Toronto, 1983, p. 226; also W.F. Craies, On Statute Law, 7th ed., London, 1911, p. 366). Cross succinctly states the situation thus ( Statutory Interpretation, p. 13):
The test of whether there has been a repeal by implication by subsequent legislation is this: are the provisions of a later Act so inconsistent with, or repugnant to, the provisions of an earlier Act that the two cannot stand together.
In support the writer quotes the decision in Westham Church Wardens v. Fourth City Mutual Building Society (1892) [45], adding, "The fact that two provisions overlap is therefore not enough."
18. (a) The question of a repeal by implication was deliberated by this court in Haddad v. Attorney-General [8], which involved the relationship between different penal enactments prescribini different measures of punishment for one and the same offence. In a majority decision it was held, inter alia, as follows (at p. 1421):
If the contradictions and inconsistencies between the two Laws cannot be aligned and reconciled with each other, the conclusion cannot be avoided that the earlier enactment has been implicitly repealed by the new one. In keeping with that rule, it has been laid down that even though the new Law deals with an offence that is also dealt with in the earlier Law, both Laws may continue to exist together if the new Law is found to have a different purpose and for that reason prescribes a substantially different penalty for the offence concerned; in such event, the offender may be charged under either Law. If, however, the penalty varies in degree only, that is to say, the one enactment prescribes a heavier or a lighter penalty than does the other, the earlier enactment will be deemed repealed by the subsequent one (see Maxwell, pp. 193-194; also Henderson v. Sherborne [1837].
Reliance upon the interpretative rule concerning repeal by implication, was founded in the above case on the doctrine that an accused person is entitled to be held to account under the less stringent of two penal directives applicable to him. Therefore, the directive of earlier date, which differs from the later one only in the heavier punishment it prescribes for the same criminal act, will be deemed repealed by implication. This interpretative rule operates only in respect of punitive criminal directives which are laid down in two separate enactments, each dealing with the identical act, omission or attempt. In other words, the contradiction finds expression, in the situation described above, in the different measure of punishment prescribed. That situation has little bearing on the problem now before us, and certainly the above rule of interpretation has no application to the situation described in the Attorney-General's directive, namely, a concurrence of the power of staying criminal proceedings and that of pardoning offenders. There is neither a contradiction nor an identity between the two.
(b) The question of the rule to apply when the same power is extended in two overlapping Laws arose directly before this court in Filtzer v. Minister of Finance [9]. The issue was the effect on certain powers conferred under the Land (Acquisition for Public Purposes) Ordinance of 1943, of other powers to achieve the same objective conferred subsequently under the Rehabilitation Zones (Reconstruction and Evacuation) Law of 1965.
For our present purpose, the above case is directly in point, since there too it was contended that the existence of parallel powers of different legislative vintage (in our own case the presidential pardoning power and the Attorney-General's power of stay) implied an abrogation of the earlier power. More specifically, it was argued in Filtzer that the Finance Minister's power of land expropriation under the Land Ordinance had been abrogated by the subsequent conferment of a parallel power on the rehabilitation authority constituted underthe later statute. Landau J. (as he then was) rejected the contention as to an invalidation of the power under the Ordinance of 1943, holding that even if the same purpose could be achieved under two different Laws, that did not preclude application of the earlier Law, though its provisions were less favourable to the citizen than those of the later Law. He noted that the two enactments were of equal status, and the Law of 1965 did not serve to deprive the Minister of his powers under the Ordinance of 1943, merely because the rehabilitation authority could achieve the same objective under the Law of 1965. Yet this conclusion had been said by the petitioners to be self-evident, in reply to which the learned Justice said (at pp. 119, 120):
The gist of the petitioner's argument is that the Law had effected a pro tanto repeal by implication of the earlier Ordinance in respect of all the eventualities covered in the Law of later date. Were it not for such an implicit repeal, it would anyhow be impossible to attribute to the Minister of Finance an abuse of his power under the Ordinance, when the exercise of such power is competent under the Ordinance as it stands. In H.C. 5/48 there arose a similar question in relation to the application of regulation 48 of the Defence Regulations of 1939. It was argued that this regulation had been implicitly repealed by regulation 114 of the Defence (Emergency) Regulations of 1945. This argument was rejected, the learned President (Smoira), quoting the following passage from Maxwell:
Repeal by implication is not favoured. A sufficient Act ought not to be held repealed by implication without some strong reason. It is a reasonable presumption that the Legislator did not intend to keep really contradictory enactments on the Statute book, or, on the other hand, to effect so important a measure as the repeal of a law without expressing an intention to do so. Such an interpretation, therefore, is not to be adopted unless it be inevitable. Any reasonable construction which offers an escape from it is more likely to be in consonance with the real intention.
The reason for such extra caution in declaring a statute implicitly repealed, is clear: to act otherwise is to introduce an undesirable element of uncertainty into the interpretation of laws which derive their validity directly from the will of the Legislature.
Landau J. referred also to the Haddad case [8], mentioning that ground for holding an earlier enactment to be implicitly repealed by later overlapping legislation, existed in the area of penal statutes, with specific reference to differences in the mode of trial or the punishment prescribed in the later legislation. Landau J. then quoted this passage from the judgment of Smoira P. in the above criminal appeal (at p. 1426):
Great importance attaches, in my opinion, to the principle of such an implied repeal specifically in the area of the penal law. One finds the theory as to a possible coexistence between two penal Laws dealing with the same offence, to be accompanied by the routine statement that the public prosecution may choose as it sees fit to prosecute either under the more stringent or the more lenient law. However, this statement has no foundation other than its routine recital, and in my view violates the fundamental penal law rule: nulla poena sine lege. A plurality of penal statutes from among which the public prosecution has a right of selection, is tantamount to a situation of having no defined and certain law.
The abovementioned statements are particularly instructive in the context of our present inquiry, as they show the court frowning in that case on the development of a situation in which the citizen who has committed an offence is dependent upon the prosecution's discretionary choice of the penal provision under which he be charged, whether the more onerous or the more lenient provision. "The penal law can affect the citizen' s freedom and reputation" Landau J. pointed out, adding that there was no like reservation with regard to other legislation in the public field.
These views expressed by Smoira P. and Landau J. have a bearing on the contention that the Attorney-General's power of stay operates as an implied pro tanto repeal or curtailment of the presidential pardoning power. At stake is the repeal or abrogation of a recognized constitutional power, the power of pardon of such scope and substance in our law as fundamentally perceived and construed in the decisions of our courts. Any interpretation that seeks to narrow the hitherto understood scope of this power, would normally require to be unequivocally established, for, as already indicated, the tendency is to give constitutional powers a wide interpretation. It is all the more difficult, therefore, to adopt the perspective that a power entrusted to the prosecution, at the exercise of its sole discretion, should oust a constitutionally endowed presidential power. It would seem difficult enough to accept the proposition advanced even if we were talking about a power of the prosecution that was new, rather than one already in existence when the pardoning power was enacted. I must reiterate what I have already stressed, that the question here is the existence of the power and not the manner of its exercise.
Our inquiry accordingly leads to the conclusion that the rule as to an implied repeal of a statutory provision by later overlapping legislation, is not applicable in the matter now before us. This is so not only for the reasons stated in the Filtzer case [9], but because the question of an implied repeal by later legislation actually fails to arise at all in our present matter, inasmuch as the Attorney-General's power of stay antedated the modern enunciation of the presidential pardoning power.
19. The fifth question we posed was whether the pragmatic development of defined processes in our current legal reality can curtail the operational scope of an existing constitutional arrangement which is essentially the product of statutory interpretation. This question entails here the notion of a pro tanto repeal of the concurrent part of an earlier enactment, a notion which was explained by Justice Landau in the Filtzer case to have no application in circumstances of the kind now before us. The answer to the question is negative.
In the first place, the answer to the question would normally depend upon the substantive nature of the processes at work, as weighed against the degree to which the constitutional arrangement evolved from statutory interpretation, and upon which the stated processes would impinge, has taken root. If this arrangement is the outcome of a wide, basic constitutional perspective, its efficacy will not be diminished by processes which are not contradictory thereto.
Second, we are dealing here with a contention that is in fact predicated upon a change in the rate and frequency of exercise of the power of stay. This change, so the argument runs, should be seen as justification for viewing the presidential power of pardon as having been curtailed. That is tantamount to saying that the Attorney-General, by the number of nolle prosequi's he issues, determines whether or not the presidential power continues to exist. I do not believe that this proposition finds any existing legal foundation. Jurisprudence does not yet recognise a biological process by which, within a complex of existing interrelated statutory provisions, a kind of law of natural selection functions as a mechanism for the abrogation of Laws for which there is abated need because they have, as it were, fallen into disuse.
Third, there will be scant inclination in a democratic regime espousing the rule of law and individual rights, to adopt an interpretation that suffers the whittling down, and even negation, of a power destined mainly to serve the individual, and it matters little that there exist other parallel institutions exercising like powers.
20. We might conveniently summarise our above reasoning as follows:
(1) It is our accepted view that a legislative overlapping or even duplication does not in itself abrogate an existing enactment or power.
(2) The validity of a statutory directive is not annulled by the fact of its disuse or rare use.
(3) An implied repeal of statute law may result either from directives which are contradictory in content or, in the penal field, from the prescription of a lighter punishment in a later enactment. The presence of such contradictory directives was not argued in the matter before us, nor have we perceived it to exist. That the existence of two concurrent competent authorities does not in itself amount to a contradiction is clear from the ruling in Filtzer [9].
(4) The mere fact of an overlapping between the power of stay and the power of pardon before conviction, does not invoke the rule of interpretation that would negate one of the two arrangements. The one is a procedural power, whereas the other is among the powers vested in the person who functions as the formal Head of State. The latter powers were fundamentally, by their very nature, intended to produce similar consequences to those resulting - to one extent or another, whether by legal design or in practice - from the acts of other governmental authorities. The two powers are of different juridical substance and the one does not negate the other.
(5) The crucial question is whether the power of pardon before conviction was ever actually created. Once it transpires that this court has recognised the existence of such a constitutional power, and regarded it as an element of the wider presidential pardoning power, the same can no longer be amenable to an inadvertent or implied repeal. It is proper that the repeal of a constitutional power be effected only after due consideration and in a patent and advertent manner, as befits the subject of the repeal. Moreover, the rule of law is fortified when we show respect for our constitutional directives, inter alia, in the way we set about their amendment or repeal.
21. Were we to hold that the very enactment of new legislation can curtail the scope of an existing statute, or even implicitly repeal an express constitutional provision, then surely even the presidential power to pardon after conviction should be overtaken by the same consequence. A retrial, or the different punishment review boards, or the operation of the abovementioned Law concerning the rehabilitation of offenders, all serve purposes which overlap, partly at least, those of a pardon after conviction. As already indicated, this situation lately assumed added significance with the enactment by the Knesset of the Crime Register and Rehabilitation of Offenders Law of 1981, incorporating the ideas of Professors Klinghoffer and Feller.
The argument as to legislative duplication and overlapping mechanisms, can hardly be confined to comparison of a stay of proceedings with the power of pardon before conviction (and substitution of the one for the other), but should properly embrace all the pertinent parallel mechanisms in the field of the constitutional as well as the penal law. An interesting illustration of the possible co-existence of parallel powers in the post-conviction stage, is provided by the Privy Council decision in Thomas v. The Queen [46], where the power of pardon of the New Zealand Governor-General was not considered invalid in relation to the class of cases in which the law permitted a retrial.
There is no logical basis for a mode of interpretation that would differentiate, for the purpose of determining the scope of validity of the pardoning power, between the various new statutory provisions and their effects, and single out precisely those pertaining to the pre-trial stage. The power of pardon has been interpreted in our law in relation not only to the post-conviction stage, but also the pre-trial as well as the trial stages, and the fact of a gradual evolution of overlapping and parallel mechanisms provides no justification for a selective kind of interpretation.
In fact we have here no implied repeal, nor any other phenomenon of an extinction without trace. When dealing with a constitutional directive such as sec. 11(b) of the Basic Law, we cannot sanction the elimination of any part thereof except by an explicit statutory provision which, after all, is the product of methodical study and preparation and is founded upon tried and tested legal concepts rather than chance eventualities. It is important that objectives of constitutional import be attained in a seemly manner.
In recapitulation, it seems clear that an enacted constitutional power is not repealed except upon express statutory directive, and that the statutory conferment on the prosecution of a power which is similar in content to that exercised by the President, does not act as an implied repeal of the constitutionally bestowed presidential power. It follows that the presidential power as construed in the cases of A. v. The Law Council and Matana, continues to exist and remain valid so long as not repealed by the Knesset.
I would not disparage the view that the power of pardon needs to be reconsidered in a manner leading perhaps to revisory legislation. Any deliberation towards this end should, in the nature of things, encompass also a solution of the problem that arises, not infrequently, from the contradiction between the judicial decision and the pardon that follows it. The search for a proper balance and separation between the different Executive organs, and the coordination of their separate activities, does not come to an abrupt halt at the chance limits set by the legal dispute in a particular matter.
Juridical Substance of the Pardoning Power
22. Our next matter for inquiry, as I have already indicated in paragraph 8 of this judgment, is the juridical substance of the pardoning power.
The power of pardon has ancient roots, and has for thousands of years been so interwoven with the ruler's status, as to induce an opinion that it finds no place in a democracy (see e.g. Blackstone's Commentaries, p. 397). The view that pardon was a feature of autocratic rule also found expression at the time of the French Revolution, when the power was abolished for the first time and left without trace for some years. Beccaria (On Crime and Punishment, New York 1963, pp. 58-59) saw an unbridgeable gap between his own penological perspectives and the power of pardon. Yet the power has survived in an overwhelming majority of world legal systems, although in a rich variety of forms so far as concerns its scope and the authority in whom it is vested (see Dr. L. Sebba, "The Pardoning Power - A World Survey," J. Crim. L. and Criminology 68 [1977], 83). The prevailing constitutional perspective is that the pardoning power now reposes in the people who, by the force of legislation, confer it in turn on a defined authority ( Am. Jur., supra, at p. 10).
As to the variety of constitutional arrangements, I might briefly mention that sometimes the power of pardon vests in the Head of State, i.e. the President or the King; sometimes it entails the functioning of an advisory board representing all or some of the connected governmental authorities, or consultation with the court or a special judicial tribunal, or a judge (see J. Monteil, La grace en droit francais moderne, Paris, 1959, p. 22). In other countries the actual power is wielded by the legislature as such (for instance, in Switzerland and Uruguay), or by the judiciary. Sometimes the power is vested in the Council of State (for instance, in a number of Eastern European countries and in South Korea), or in the Presidium or a council specially constituted for this purpose. In Sweden the power rests with the Government and in the U.S.S.R. with the Supreme Soviet Presidium.
These examples illustrate, without exhausting, the range of pardoning powers, which vary also in their prescribed procedures, such as the manner of lodging the request, of consultation with judicial bodies or other agencies, and of arrival at the decision. In Australia, for instance, an inquiry is conducted in all cases by a Justice of the Peace appointed by the Governor-General or by a Judge of the Supreme Court.
In some countries the power of clemency is confined to the reduction of punishment alone (as in France, but there one finds also the special power of la grace amnistiante, which enables the grant of a full pardon to certain classes of persons; see Monteil, La grace, at p. 207). Pardon before conviction is possible in numerous countries, inter alia, the U.S.A., Britain, New Zealand, Singapore, Malawi, Sri Lanka, Iceland, Czechoslovakia, Lichtenstein, and the State of Queensland in Australia (see Dr. L. Sebba, On Pardon and Amnesty, at p. 291). In many other countries, however, there is pardon after conviction only (for instance in India, where the restriction is statutorily prescribed). The legal consequences of a pardon also vary greatly from country to country.
In some countries the actual decision may be directly or indirectly challenged in the courts, whereas elsewhere, for instance in France, the decision offers no ground for recourse to the courts, whether as to the legality of the decision or as to its substance within the national framework. F. Luchaire and G. Conac phrase the situation thus: tant au niveau de leur legalité qu'au plan de la responsibilité de l'Etat (La constitution de la republique francaise, Paris, 1979, p. 351). The writers rely in this connection on a resolution of the Conseil d'Etat (30.6.1892; Gugel, Dalloz Periodique, 1894, III p. 61; 28.3.1947, Gombert, Sirey, 1947, III p. 89).
Sometimes pardon is granted for political offences alone (for instance in Colombia), and at other times these are specifically excluded as a type of offence for which a pardon may be granted.
Our purpose in sketching the abovementioned varieties and possibilities of pardon, is to illustrate the lack of any uniform model and the fact that virtually every legal system has fashioned its own peculiar perspective on the subject, in harmony with its other governmental institutions. For comparative purposes, it is of no moment that in the U.S.A. the President, in whom the power of pardon is vested, serves to head the executive, whereas in Israel the President fulfills the function of a titular and formal Head of State - much like the arrangement in England, adopted also in many of the European democracies after World War I. There is no uniform tie between the nature and general status of the executive office filled by the holder of the pardoning power, and the power itself, since it is sometimes vested in authorities other than the President or King. When this court made reference in Matana and in A. v. The Law Council to the constitutional situation in Britain or the U.S.A., it did so, not in order to link the Israel arrangement to one or another foreign complex of powers, but to indicate the source and substance of the viewpoints we ourselves adopted. These the court found reflected in what was taken at the time to be the prototype for our own constitutional mould when our initial autonomous directives to this end came to be enacted. Once domestically fashioned, the powers became independent of any influence other than our own perspectives and concepts. Processes in other countries may be of instructive and comparative interest, but cannot deflect us from what is customary and accepted here until such time as we ourselves decide to change the approach, and do so in the appointed manner, having regard to the character of the subject and the substance of the power concerned. For this reason, too, there is little logic in seeking guidance from other systems structured upon essentially different perspectives. If, for instance, French law decrees that grace, in the case of an individual, shall relate only to the punishment and not to the conviction (except in the case of grace amnistiante ), there is little we can learn from it as regards the possibility of pardon before conviction. The French method of grace, incidentally, seems to differ also from our own method of remission of punishment, for instance in relation to a mandatory death sentence. (On the reservations of former French President Giscard d'Estaing in this connection, see Luchaire et Conac, La constitution, at p. 348.) The opposite applies in Belgium, where remission Of a mandatory minimal punishment is possible (see Dr. Sebba's article "The Pardoning Power," at p. 86). In short, the lack of a power of pardon before conviction in France or Germany, for instance, has no bearing on the present situation in our law since the models in those countries played no part in the shaping of our constitutional framework of pardon. Furthermore, for a proper evaluation of standards, we should put the emphasis on the substance of the pardoning power, and not on the functionary who exercises it, or the manner of its exercise. In our law it has been held that the President is invested with the widest form of the power of pardon (as regards offences) and clemency (as regards punishment), being empowered to obliterate even the stain of the offence and not only its consequences. That is our existing legal situation and it is in the light of this conclusion that we have to draw further inferences as to specific aspects of the power. The fact that proposals have been made to change the legal situation - and I certainly am not opposed to the discussion of these ideas, and even the adoption of some of them - does in no way affect the substance of the existing law.
23. It accordingly transpires that in the present case the decision to pardon came within the formal scope of the State President's power. In this regard it should be noted that the lawgiver has made provision for the preliminary ascertainment of this court's views on matters of pardon. Thus sec. 32 (a) of the Courts Law (Consolidated Version) of 1984, provides that upon a request for a pardon or reduction of sentence lodged with the President, any question which arises and in the opinion of the Minister of Justice deserves to be dealt with by the Supreme Court, but which provides no ground for retrial under sec. 31 of the Law, may be referred by the Minister to that Court.
Exercise of the Power
24. Having concluded that the President has a valid power to pardon before conviction, we might now inquire as to the occasion for its exercise. In fact the power has so far remained virtually unused, such a pardon having been granted until now in only a small number of exceptional cases, some of which were brought to our notice in the course of our deliberations here. It is only right that the power be used sparingly, for only the most exceptional circumstances of paramount public interest or personal plight - for which no other reasonable solution can be envisioned - will justify such anticipatory intervention in the normal course of the trial proceedings. It would be wrong, therefore, to attempt to classify in advance the proper cases for the exercise of this power.
The decision to pardon was held by Justice Marshall of America to be generally motivated as an act of grace ( United States v. Wilson [55], at pp. 160-161), but the prevailing American approach is to rest the decision on considerations of the public welfare (see: Biddle v. Perovich [56] at p. 486; and see also C.C. Joyner, "Rethinking the President's Power of Executive Pardon," Federal Probation 43 (1979) 16).
As the general starting point for examining an exercise of the pardoning power, I am disposed to accept the approach enunciated in Montgomery v. Cleveland [57] at p. 1157:
While a pardon is a matter of grace, it is nevertheless the grace of the State, and not the personal favor of the Governor. It is granted out of consideration of public policy, for the benefit of the public as well as of the individual, and is to be exercised as the act of the sovereign state, not of the individual caprice of the occupant of the executive office as an individual. He is supposed to act in accordance with sound principles and upon proper facts presented to him.
Normally, a pardon is not a natural further progression in the course of judicial proceedings, but should properly come into play only in exceptional circumstances which involve a material change in the situation after completion of the trial proceedings, and warrant an alteration of the judicial decision. All the more rarely and exceptionally, therefore, should the power of pardon be exercised before conviction, this being a reserve or residual constitutional power left with the President - something in the nature of a "safety valve."
A theoretical example of circumstances warranting the grant of a pre-conviction pardon, was outlined by Prof. Klinghoffer in his abovementioned statement before the Knesset, the relevant passage from which I shall repeat below for the sake of convenience:
Another unconvincing argument advanced is that if a suspect be very ill, it would be an act of cruelty to put him on trial. Unconvincing, because in rare cases of this kind the President of the State would be able to grant a pardon before the trial commenced. The President is empowered to pardon offenders either before or after conviction. That was laid down by Justice Agranat in the case of A. v. The Law Council, H.C. 177/50.
As a further example one might mention that reasons of state, involving arrangements with hostile elements, have been recognized in the past as legitimate grounds for the early release of prisoners from custody, even before completion of the trial and, implicitly at least, as warranting also the grant of a pardon (cf. H.C. 228/84 [10]; H.C. 270/85 [11] and Bar-Yosef v. Min. of Police [12]). Of course, even in the stated circumstances every case would still require to be examined independently and the situation would vary from one concrete set of circumstances to another.
25. Just as it would be inconsistent with the purpose of pardon for it to become converted into a kind of instance of appeal from judicial proceedings, so too pardon before conviction ought not to become a mode of appeal against the decisions (to prosecute) of the public prosecution. This slumbering, residual kind of power has been preserved for sole use in the exceptional situation of a risk of serious harm which the holder of the power may legitimately take into account, which is incapable of being warded off by other means, and thus warrants relaxation of the essential restraint on the exercise of the power.
In concluding my remarks on the instant point, I should like to recall, and endorse with such changes as may be necessary, the recommendation made by Dr. L. Sebba (On Pardon and Amnesty). He proposes that even upon adoption and completion of the comprehensive legislative program proposed by him for the creation of machinery to deal effectively with all matters and foreseeable problems connected with or likely to result from punishment under the criminal law, there should still be left with the President a reserve or residual power, as he put it, to deal with exceptional cases (at p. 267):
However, even if all the proposed solutions be accepted, we do not recommend the complete abolition of this power. Even if the parole arrangement be instituted, even though it embody regulation of the penalty of life imprisonment, and even upon the abolition or qualification of prescribed minimal punishments, there will always remain special cases in which the offender will not find salvation unless the President be empowered to come to his aid. It is true that the flexibility contributed by the pardoning power to the process of meting out the punishment, has largely become redundant in view of the increased freedom allowed the courts over the years in this regard. It is also customary nowadays to enable the Executive to intervene in the more advanced stages of implementation of the punishment, so as to maintain flexibility in these stages as well. But in the end it is still necessary to leave an opening for intervention on the part of some additional authority, in the event that the other two authorities be unable to effect the desired solution. The proper authority for this purpose is indeed the State President, who ranks in status above the other two authorities, and especially since there is sometimes involved a departure from the policy laid down by the third authority, i.e. the Legislature. In these residual cases there remains room, therefore, for entrusting the President with a power that will function as a kind of "safety valve" in the event the customary processes provide no solution.
And now, arising out of the hearing of the instant petitions, there are some additional observations I have to make.
The State and the Rule of Law
26. (a) The rule of law is not an artificial creation. It is to be observed in a concrete day-to-day manner in the maintenance of binding normative arrangements and their actual application to one and all, in the upholding of the basic freedoms, in the insistence upon equality and the creation of an atmosphere of trust and security. The rule of law, the public welfare and the approach of the State to problems are not opposing conceptions but complement and sustain each other.
The court is specially charged with the practical realisation of these expectations, but all of the State organs are committed to the attainment of the stated objectives. One cannot conceive of a sound administration without maintenance of the rule of law, for it is a bulwark against anarchy and ensures the State order. This order is essential for the preservation of political and social frameworks and the safeguarding of human rights, none of which can flourish in an atmosphere of lawlessness. National security also leaned on the rule of law, both in protecting internal policy measures, and in aiding the creation of means to combat hostile elements. There can be no organized activity of any body of persons, or any discipline, without norms based on binding legal provisions.
(b) Sound government requires that the authority concerned be in full possession of the relevant facts before acting. It is not necessary that the information be known to all, and the confinement thereof to a few persons is sometimes not only desirable but also legally imperative. Yet the need for the responsible authority fully to acquaint itself with the facts increases as the subject takes on greater importance. It must be remembered that the "leaking" of classified information does not happen by itself, but by its deliberate or accidental disclosure by some person involved.
(c) Sound government is founded upon the faculty of sound decision making, which there can never be without prior knowledge of the relevant particulars, no matter the subject of the decision. The matter was discussed by this court in Berger v. Minister of the Interior [13], in the context of the Minister's duties with regard to the introduction of summer-time or "daylight saving." In background importance the subject, of course, did not match that which is now under deliberation, but the principle enunciated there is equally applicable elsewhere. In the above case the court formulated rules affecting the manner of ministerial decision making, reiterating the obvious proposition that this should result from and be structured upon knowledge of the factual situation.
Sound administrative procedures will ensure diverse facilities for obtaining information, maintaining constant supervision and overseeing the implementation of directives. The process of gathering information or holding an investigation, when necessary, may also assume different forms. Here one golden rule has to be observed, valid for purposes of administration as well as inquiry, namely: the sooner a matter calling for investigation is examined, the better from all points of view. A particular authority may perhaps confine information departmentally, or otherwise restrict its dissemination and ensure that no harm result from the disclosure or obtaining of information. But there are no circumstances that allow an administrative authority to refrain totally from investigating a matter which may bear upon its capacity, and that of its subordinates, to function properly, and to decide issues within the scope of its immediate responsibility, or perhaps affecting its responsibility to the public at large. There is a world of difference between a decision to hold a controlled and protected investigation, and a decision not to conduct one at all. The latter option would be like trying to cross a busy road with one's eyes shut.
(d) There are different ways to conduct a confined or departmental inquiry or investigation into any subject - including recourse to whatever legal proceedings be considered necessary - without prejudicing the national security. Such problems have been dealt with before, and I shall say no more on the subject on the assumption that the processes mentioned by the Attorney-General in his intimation of 15 July 1986, have been set in motion.
The President as a Respondent
27. On 30 June 1986 we ruled to delete the President's name as a respondent in petitions H.C. 431/86 and H.C. 446/86, and ruled likewise on 20 July 1986 in petition H.C. 463/86. Our reason for so doing is set forth in the Basic Law: The President of the State, sec. 13(a) of which reads as follows:
The President of the State shall not be amenable to any court or tribunal, and shall be immune from any legal act, in respect of anything connected with his functions or powers.
When he granted the instant pardon, the President was acting in a matter "connected with his functions and powers," so that he is not amenable to the jurisdiction of the courts in connection therewith, including this court's powers of direct review - its authority to demand of the President himself an explanation of his decisions. This immunity relates to the direct challenge of any presidential act, but there is no obstacle to indirect judicial review of the President's discharge of his functions - in proper cases and when the proceedings are directed against some other respondent, as happened, for instance, in the case of Bar-Yosef v. Minister of Police [12].
28 (a) The question of the legality of the pardon granted is of wide range, embracing as it does both the power itself and the manner of its discretionary exercise. As regards the power itself, we have dealt extensively with the matter and sought to provide the correct answer above. With regard to the exercise of the presidential discretion, this court has had occasion to comment as follows, in connection with a ministerial recommendation for a pardon referred to the President:
Even if the President was misadvised, or even if he himself erred in the exercise of his discretion, the legal validity of his decision remains unaffected thereby and this court does not sit in appeal from the President's decision" (Barzilai v. The Prime Minister [14] at p. 672).
The matter calls for a measure of clarification and qualification. It is accepted that in exercising judicial review, the court does not assume the role of the functionary whose conduct is under challenge (even if indirect) but examines whether the functionary acted as one in his position should have done ( Nof v. Attorney-General [77] at p. 334). The court does not seek to project and substitute its own decision but intervenes only when convinced that no reasonable authority in a similar situation could have arrived at that same conclusion. The degree of reasonableness required depends upon the status of the authority and the nature of its powers. That is to say, in exercising its jurisdiction the court will also have regard to the identity of the constitutional authority whose conduct is under review. The norms for the judicial review of discretionary power will in any event incorporate reference to the functional character and nature of the authority concerned (cf. Sarid v. Knesset Chairman [15] at pp. 203-4).
(b) The petitioners' criticism of the President's exercise of his discretion extended also to the paucity of the information made available to him prior to his decision, as well as the haste, so it was further contended, with which the different pardons were deliberated and granted, and like contentions. I find none of them to provide any ground for intervention by this court. First, as regards the facts, there is no reason to dispute the declaration before us that the President was fully informed and had also met twice with one of the persons later granted a pardon. The fact that he did not meet with the other three applicants can hardly be regarded as an impropriety, as in fact the President normally deals only with written requests for a pardon and it is exceptional for him to meet with the applicant (see E. Abramovitz and D. Paget, "Executive Clemency in Capital Cases," N.Y.U.L. Rev. 39 [1964], 136, 137; and see Dr. Sebba, On Pardon and Amnesty, at p. 194). Once it is established that there was evidence before the President of the commission of offences as set forth in the pardon applications referred to him, whether verbal or in writing, and also that the applicants admitted having committed the criminal acts for which they asked to be pardoned, then clearly the President had before him sufficient particulars upon which to decide, thus leaving no ground for the court's intervention.
29. A further argument concerning the presidential pardoning power, focused on the distinction between amnesty and individual pardon, was addressed to us by Adv. Michal Shaked, learned counsel for the petitioners in matter H.C. 448/86. She contended that the circumstances of the grant of the pardons indicate them to have been in the nature of an amnesty, whereas the President enjoyed no such power, but the power to grant individual pardons alone. In support of her contention counsel quoted the following statement (extract from The Attorney-General's Survey of Release Procedures, Department of Justice, Washington, 1939 vol. III):
In an attempt to classify the institution of amnesty, we may state that it belongs to the upper concept of pardon. It is a plurality of pardoning acts, and its main feature is that the amnesty determines the conditions and the extent of the pardon by groups of persons or groups of crimes or by certain general attitudes of the individuals concerned. There is a pronounced predilection to lay stress on the motive. Even the exceptions and limitations in an amnesty are generally given by groups, regardless of the merits of the single case.
It indeed appears from the decision in Matana [3] (at p. 445) that the President enjoys the power of individual pardon only (as is the case in England). But that exactly was the power exercised by the President in the instant case. It is true that he issued four different warrants of pardon, but each of them related solely to the individual named in that warrant and to the offence therein stated. The warrants did not define the right to the pardon according to a class of persons, or offences, or qualifying conditions. The fact that a number of pardons are granted simultaneously to several individuals involved in the same act or incident, does not serve to convert each separate warrant, or all of them together, into an amnesty (see Dr. Sebba, On Pardon and Amnesty, at p. 61).
Locus Standi
30. At the commencement of the hearing learned counsel for the respondents asked for dismissal of the petitions in limine, on the ground that the petitioners had no legal standing to contest the validity of the pardons granted. It was argued that these were in the nature of an individual act of the President and of concern to the recipients of the pardon alone. It was contended that the petitioners could not point to any real and direct personal interest in the invalidation of the pardons, as these operated solely for the benefit of the individuals pardoned (certain of the respondents in these proceedings), so that the petitioners, far from seeking any relief for themselves, were motivated merely to deprive others of a benefit (see Becker v. Minister of Defence [16], at p. 147).
The absence of a real personal interest, even if this be true of the petitioners in the present case, does not, however, justify the immediate dismissal of the petition. This court has already held that it would take a liberal view on this aspect and grant access to petitioners where the question that arose was "of a constitutional character" ( Segal v. Minister of the Interior [17] pp. 429, 433), or "of public interest related directly to the advance of the rule of law" ( Shiran v. Broadcast Authority, [18] at 374; see also Dr. Zeev Segal's illuminating book, Standing Before The Supreme Court Sitting as the High Court of Justice, Papyrus Publishing, 1986). Needless to say, there is no general recognition here of the actio popularis, a "public petition" to the court, only a general guideline that enables the court to open its doors in suitable cases of a public-constitutional character.
Guided by the above rule I find the petitions now before us, which centre on the scope of the presidential pardoning power under the Basic Law: The President of the State, to disclose sufficient petitioner interest for recognition of their standing.
The Approach of Justice Barak
31. I have meanwhile had the opportunity of reading the interesting opinion of my learned colleague Barak J., and I am prompted to make several further observations in elucidation of our divergent approaches.
(a) I naturally take no issue with the fundamental doctrine that we must decide according to our best knowledge and understanding of the law, regardless of the surrounding influences of the time and the subject concerned. That standpoint has always been customary with this court, and nothing new has happened in this generation to change the court's perspective.
(b) A perusal of Justice Barak's opinion may lead one to think that our present subject has no acknowledged legal starting point founded in precedent, and that one is being referred (in Matana and in A. v. The Law Council) to nothing more, as it were, than some forgotten obiter dictum raised here from oblivion for the first time and elevated - without legal justification - to the standing of a recognized legal thesis. One might further gain the impression that even Justices Agranat and Berinson intended no differently in the above precedents. I must reject this approach because it does not accord, with all due respect, with the factual situation. The legal proposition that the President is endowed with the power of pardon before conviction, was clearly demonstrated first in the case of A. v. The Law Council and later, even more emphatically, in the Matana majority decision. Incidentally, even Landau J., at the end of his dissenting opinion in Matana. noted his complete agreement with the opinion of Berinson J. (at p. 461), whose remarks on the presidential power to pardon before conviction have already been quoted in full above.
In brief, the ruling in Matana has become known and accepted as faithfully reflecting, for some decades now, the prevailing law on the subject. Confirmation thereof is to be found in the written commentaries and in all academic discussion of the subject. This situation has been so clear to all as to have prompted the two distinguished jurists who advanced a different perspective on the subject (Professors Klinghoffer and Feller), to acknowledge that their view was not in accord with the approach of the Supreme Court - which they interpreted substantially as I have understood and set it out above. One of them, moreover, relied on the very existence of the pre-conviction pardoning power for a proposed solution to other legal problems discussed by him at the time (see Prof. Klinghoffer's abovementioned remarks in the Knesset - Minutes of the Knesset, 43, p. 2319). It will be recalled that one of these jurists (Prof. Klinghoffer) based his approach upon a construction of the language of the pardoning directive, while the other (Prof. Feller) argued on the basis of the working of a complex of new (overlapping) statutory enactments, but I gather from the remarks of Justice Barak that his own viewpoint is founded on neither of the above two perspectives.
There is no escaping the fact that Justices Agranat, Berinson and Cohn (to whose clear statements on the subject Justice Barak has not referred) all unequivocally expressed their opinion on the power of pardon before conviction within the general framework of pardon. That opinion has held sway until now. It was on the strength of an identical opinion that a past Minister of Justice, P. Rosen, acting upon the Attorney-General's advice, referred recommendations to the President for certain pre-conviction pardons which were subsequently granted. Our task here is not to search for the desirable constitutional framework, but rather to ascertain the existing legal situation concerning pardon in Israel, just as it was in fact enunciated by this court many years ago, without so far having undergone any change.
(c) The constitutional development towards the existing situation was clearly traced in the Matana decision, from which one can gather the court's reasons for construing as it did the scope of the pardoning power under sec. 6 of the Transition Law and sec. 11(b) of the Basic Law: The President of the State. It is not possible to ascertain the meaning of an expression in a Law by seeking to unravel the true wishes of Knesset committee members from the surviving summaries of their statements in minutes of proceedings never published. In my recognition, the answer lies in an understanding of the legislative purpose. This is to be derived from the "spacious" interpretation to be given to constitutional provisions; from the construction of expressions according to their manifest purpose; and from factors such as legal background and development, constitutional analogy, the characteristics of our legal system and our own constitutional notions as given expression, inter alia, in the very determination of the presidential office, its object and functions. All these were dealt with in the Matana case and I shall not cover the same ground again.
Justice Barak has sought to point out a divergence between the interpretative approach in Matana, and my own approach. Little substantiation of this has been provided, however. There is no substantial difference between the "historical-interpretative approach" said to have been adopted by me here, and the so-called legal-constitutional approach ascribed to Justice Agranat, and the difference in title is but a semantic one. Substantively speaking, the two approaches are alike: that followed on the one hand by Justices Agranat, Berinson and Cohn and - on the question of pardon before conviction - also adopted without reservation by Justices Silberg and Landau, and on the other hand, my own approach here. My learned colleague has commented thus:
Justice Agranat accordingly did not construe the Transition Law on the basis that its legislative purpose "was fashioned in the Anglo-American mould, which served as its prototype."
In support of this connection he quotes the following observation of Agranat J.:
The result is that the ground of the absence of any similarity or comparison between the status of the President of our country and that of the British thrown (or of the President of the United States) is erroneous.
This observation speaks for itself and, with all due respect, refutes my learned colleague's contention in indicating the opposite conclusion.
(d) The legal situation in France, Germany and Italy was not fully portrayed in the Matana case, and I should like to clarify some additional facets. As far as I am aware, pardon before conviction is known in Italy too, but the pardon only comes into operation if the suspect is later convicted. This arrangement does not preclude putting the suspect on trial, and allows for an acquittal on the merits without recourse to the pardon. Briefly, in Italy and in Germany there has evolved the duality of a judicial pardon side by side with an extra-judicial one justizgebundener Gnadenakt and justizfreier Gnadenakt, see Mario Duni, Il Perdono Giudiziale, Milan, 1957; Richard Drews, Das Deutsche Gnadenrecht, Cologne, 1971; Klaus Huser, Begnadigung und Amnestie als Kriminalpolitisches Instrument, Hamburg, 1973).
Judicial pardon or clemency, I believe, should be seen as a convincing reason for gradual curtailment of the Executive pardon. This process, which is also discussed by Prof. Feller within the wider framework of his proposed legislative program, has acted to shift the focal centre of the pardoning decision from the King, or President, to the judicial tribunal or special statutory bodies created to deal directly with the review of conviction and punishment (retrial, release and parole boards, and the like). The comparison of our system with those applied on the Continent is therefore questionable and premature in the existing state of affairs.
(e) As to the pre-conviction pardoning power in England, concerning which too Barak J. has expressed reservations, I need only reiterate that there is not a single English constitutional text that fails to mention the continued legal validity of this power, though it be reserved for use in exceptional cases. Even the post-conviction pardoning power would seem to be somewhat less frequently exercised in England nowadays.
I must also contest Justice Barak's endeavour to distinguish the American constitutional situation from our own on the basis of the President's status there as Head of the Executive. In fact, the power of pardon was originally conferred on the U.S.A. President as part of the legal continuity adopted there, with the concomitant imitation of the English model of the King's prerogative power (see the majority decision in Schick v. Reed [58], per Burger C.J.). The view that the U.S.A. President holds the pardoning power in his capacity as Executive Head, runs counter to authority:
Our government is established upon the principle that all governmental power is inherent in the people. Hence, crime is an offense against the people, prosecuted in the name of the people, and the people alone can bestow mercy by pardon. As subsequently is noted, the people may confer the pardoning power upon any officer or board that they see fit ( Am. Jur., at p. 10).
The import of the passage is that the U.S.A. President holds his power as representative of the people and it is not his executive capacity, designation or status that determine its scope.
In our own constitutional framework the President stands outside the political arena, and this neutrality should allay at least one of the apprehensions expressed by my learned colleague. Moreover, the conferment and exercise of all power can and should properly be subjected to supervision and review, as was indeed noted by Justice Agranat in Matana (at p. 461):
Nor have I overlooked the fact that to endow the power in question with its "full" content may lead to its excessive use, which in turn involves the danger that the authority of the law in the eyes of the public will be weakened. My reply to this point, however, is that every instrument of pardon by the President requires the countersignature of the Prime Minister or one other Minister (sec. 7 of the Transition Law, 1949). This means that even if the decision to pardon or to reduce a sentence must be the personal decision of the President, it is also conditional upon the recommendation of the Minister concerned. This Minister will ordinarily be the Minister of Justice who has the means of conducting a precise investigation into the circumstances of the case before submitting his recommendation to the President. It is clear that this recommendation, and therefore the decision to pardon as well, are subject to review by the Knesset and it is this possibility which must be regarded as the guarantee laid down by law against the danger referred to.
And Justice Cohn had this to say on that same point (at p. 465):
I have no fear whatsoever of any supposed impairment of the courts power to judge and to punish by the extent that the President of the State is empowered to change or set aside the results of their judicial acts. I could see some slight ground for fear and some small degree of impropriety if the power in question were possessed by the government or one of its organs, or even by the Knesset, for that would perhaps involve some confusion of the boundaries between the judicial on the one hand and the executive or legislative authorities on the other. The President of the State, however, stands above all these three authorities. He embodies in his person the State itself.
So far as I am concerned, the existing Israel form of the pardoning power is not a sine qua non for the maintenance of orderly constitutional government. The variety of arrangements made on this subject in different countries is indicative of more than one solution to a universal problem. Our own arrangement is hardly, therefore, to be seen as the sole possibility. The central feature of the pardoning power wielded is a personal, selective decision which is dependent, inter alia, on the recommendation and countersignature of an authority of a political character, i.e. the Minister. My own inclination is to prefer some new legislative arrangement that will introduce appropriate statutory mechanisms free to function, as regards judicially decided matters, without recourse to the decisions of political organs. At the same time, however, one has to reject the view that the full pardoning power presently prevailing is inconsistent with the rule of law. It should be remembered that we are dealing here with legally valid constitutional arrangements of the kind found today in countries of recognised democratic character, and to say that the existence of an effective rule of law is negated by reason of a pardoning power of full scope, where it exists, is an extreme proposition lacking any real foundation.
I must also refer to the contention that the very overlapping of the presidential power with like administrative powers is inconsistent with the maintenance of good government. I have already pointed out that there is no complete parallel between the two kinds of power. Pardon in all its existing forms represents an institution which by its very nature and working contradicts the rulings of other authorities, just as it does whether it is extended before or after conviction. Such overlapping is therefore an inherent feature of the entire pardoning process and in this respect its exercise before conviction is not exceptional.
It is only right that the abandonment of the existing arrangement in favor of newly devised systems should be preceded by a comprehensive study of the subject - of the kind undertaken by Prof. Feller - and be followed by orderly legislation embracing all aspects of pardon and clemency. But until the fundamental constitutional perspective underlying our recognition of the pardoning power be revised in the appointed way, there is no room for the abrogation - in a sporadic manner, by the method of interpretation - of one of the facets of that power which has been recognised for many years now, and is rooted in the fundamental judicial understanding of the pardoning institution in our legal system. The ad hoc erosion of an existing legal arrangement in answer to the needs of the hour, weakens rather than strengthens the rule of law. This was the kind of situation I had in mind when I remarked thus in Neiman v. Central Knesset Elections Committee [19] (at 260):
When constitutional matters are under review, their import and implications have to be considered in the long term, and proper weight has to be given to their influences on the political and social frameworks within which they operate. If these be subjugated to the needs of the hour and we adopt a casuistic approach in matters of constitutional content, we shall miss the mark and deal less than justly with the subject.
(f) It is, in sum, an inescapable conclusion that the Matana precedent adopts a wide and embracing interpretation of the presidential pardoning power. It was explicitly decided there by Justices Agranat, Berinson and Cohn that it also encompassed pardon before conviction. Though the product of autonomous Israeli legislation, the power cannot be divorced from its repeated comparison and equation, in the Matana case, with the parallel power held in the Anglo-American legal system by the King or President, as the case may be. This equation had a direct bearing on the reach of the constitutional power unfolded in the above precedent. Much as I try, I find no evidence in the Matana decision to support the suggestion of Barak J., that at that time the origin and substance of the power in England and in the U.S.A. had not been properly understood. I also find no evidence that this court had overlooked, as it were, differences of constitutional structure between those countries and Israel or, for that matter, the prosecution's own powers and independence in Israel, or the clash of the presidential power with other overlapping, frequently exercised powers - both before and after conviction. This suggestion is in entire disaccord with the long-accepted Matana ruling.
The constitutional situation is, therefore, that enunciated in Matana, by which precedent we have to be guided - as regards the scope of the pardoning power until the lawgiver sees fit to intervene. We have to contend with the legal and factual circumstances as we find them unfolded before us, rather than with hypothetical or desirable situations, and without circumventing or bypassing the decisions of this court and their consequences. It is our judicial task, in the present context, to give a principled, normative decision, structured upon existing legal foundations. In the pursuit of this objective we should do well to apply Chief Justice Marshall's well-known dictum in Osborn v. United States Bank [59] (at p. 866):
Judicial power is never exercised for the purpose of giving effect to the will of the judge; always for the purpose of giving effect .... to the will of the legislature.
The Investigation
32. We understand from the State's reply, as intimated by the Attorney-General, that the police will conduct an investigation into the events forming the subject of these petitions. This leaves no room, in my opinion, for intervention by this court in connection with the holding of an investigation.
Following the Attorney-General's intimation, the petitioners in file H.C. 431/86 gave notice of withdrawal of their petition, and the petitioner in file H. C. 428/86 advised that he was confining his petition to the sole issue of the legality of the pardons granted.
As to the petition in file M.A. 320/86, I see no reason to question the decision on the investigation as intimated by the Attorney-General, the nature of which I find acceptable in principle.
33. I would accordingly dismiss the petitions and discharge the order nisi.
MIRIAM BEN-PORAT D.P.
1. The divergence of opinion between my learned colleagues, President Shamgar and Barak J., persuaded me to await their written judgments before giving my own decision on the important question under consideration here, namely: is the President of Israel empowered to grant a pardon to a person before trial and conviction? I find my learned colleagues to have unfolded in their judgments a wide and colourful tableau of concepts, precedents and scholarly comment, which have aided me greatly in formulating my own opinion. Their painstaking and comprehensive analyses leave me free to concentrate mainly and briefly on my reasons for concurring in the judgment of the learned President - more particularly, my reasons for agreeing that the pardons granted by the President are legal and valid and, primarily, my reason for holding that the stated presidential power of pardon before conviction effectively exists.
2. In see. 11(b) of the Basic Law: The President of the State, it is provided that the President
shall have power to pardon offences (and to lighten penalties by the reduction or commutation thereof).
I have put the latter part of the directive in parentheses since the first part is the focus of our deliberation here, although I shall of course deal with the whole in substantiation of my viewpoint. As already clarified by my learned colleagues, the power "to pardon offenders" was previously vested in the President under sec. 6 of the Transition Law of 1949, and remained so vested until the repeal of this provision by sec. 26(a) of the above Basic Law. A comparison of the language of the two sections shows only a slight difference in wording, of no material significance. We may accordingly treat anything stated or decided on the basis of sec. 6 of the Transition Law as equally applicable to sec. 11(b) of the Basic Law, with which we are now concerned. For the better understanding of my exposition below, it should be recalled that until the enactment of the Transition Law, the President's power was anchored (pursuant to sec. 14 of the Law and Administration Ordinance of 1948), in Article 16 of the Palestine Order in Council, 1922, which provided as follows:
When any crime or offence has been committed within Palestine, or for which the offender may be tried therein, the High Commissioner may, as he shall see occasion, grant a pardon to any accomplice in such crime or offence who shall give such information and evidence as shall lead to the conviction of the principal offender or of any such offenders if more than one; and further may grant to any offender convicted of any crime or offence in any court.... a pardon either free or subject to lawful conditions, or any remission of the sentence.
I have stressed, in the above passage, the word "offender" and the phrase "a pardon.... subject to lawful conditions," for purposes which I shall presently elucidate. At this stage, however, I merely wish to summarize the President's power at that time (from the transfer to him of the High Commissioner's powers under the above Ordinance) as embracing a pre-conviction pardon granted an accomplice who was willing to give information leading to the conviction of the principal offender or any such offenders, as well as a pardon granted any convicted offender. The pardoning power, as already indicated, was formulated differently in the Transition Law, and this version was later repeated without material change in the Basic Law.
3. Our search for the proper interpretation of the statutory provision in issue here, hardly breaks new ground. Justice Barak is aware of this fact, but attaches little weight to the precedents cited, for two reasons. In the first place, he holds the statements made in these precedents to have been obiter, and secondly, he considers certain passages therein actually to support his own view. Thus he mentions, for instance, that Justice Agranat saw the power conferred under sec. 6 of the Transition Law as an "original" one, and therefore offering no basis for analogy with the corresponding power in English law. Justice Barak also attaches no significance to the practice that has evolved out of those precedents.
I accept Justice Agranat's determination, in the Matana case [3] (at p. 443), that the language of sec. 6 of the Transition Law- and likewise of sec. 11(b) of the Basic Law - was not comparable with that of Art.16 of the Order in Council, since the Transition Law provision represented an "original" Israel power of constitutional content, in contrast to the class of powers delegated by the English King to colonial Governors. The latter were much narrower than the King's own powers, and required a restrictive interpretation.
However, I disagree with Barak J., that in the Matana case Agranat D.P. (as he then was) did not view the presidential power of individual pardon under the Transition Law (as opposed to a general amnesty), as being basically the same as that of the English King or the American President. In other words, Justice Agranat's remarks on the original nature of the power set forth in sec. 6 of the Transition Law, and on the universality of the pardoning concept, were only intended, I believe, to explain why the non-repetition in sec. 6 of the Transition Law of certain parts of Art.16 of the Order in Council, could properly be ignored in construing that section. For the purposes of the issue in Matana, Agranat D. P. was not prepared to regard the non-repetition in sec. 6 of the Transition Law, of the words "pardon.... subject to lawful conditions" (appearing in Art.16 of the Order in Council and stressed by me in the above citation), as being in derogation or restriction of the presidential power. On the contrary, his opinion was that the general language used in sec. 6 was characteristic of a constitutional directive and called for a wide interpretation, and he saw the Anglo-American legal sources as prompting the proper interpretation of our own statutory provisions (see the Matana case [3], at pp. 453, 454). It was his opinion (which became the majority opinion of the court) that notwithstanding the absence of an express empowerment of the President to grant a pardon subject to conditions, the general wording ("the power to pardon offenders") sufficed, by virtue of the wide interpretation, to invest the President with this power as well (i.e. to pardon conditionally).
Any remaining doubt as to Justice Agranat's recognition (in Matana) of the link between Israel and England as regards the power of individual pardon, is surely dispelled upon reading his judgment in the earlier case of A. v. The Law Council [2]. While the learned Justice erred there with regard to the power of colonial Governors (i.e. the High Commissioner of Palestine), an error he subsequently corrected, his basic standpoint has nevertheless prevailed. This standpoint he expressed in the following terms, and in other statements to the same effect in his judgment:
I am of the opinion that the power of pardon of the President of Israel is the same, generally speaking, as the power of pardon of the King in England, in its nature and in respect of the consequences which flow from its exercise" (ibid., p. 750).
Agranat J. (as he then was) was indeed alone in considering the full arguments of counsel as to why the name of A., the petitioner, was to be restored to the Roll of Advocates following upon the full pardon granted him (after he had served his full sentence). Yet the related comments of Agranat J. were not mere obiter dicta - and as is known, these too can carry considerable weight - but were made in the deliberation, on its merits, of what he considered to be the real question underlying the dispute in that case:
The real dispute being waged today between the petitioner and the community, has its origin in the former's argument that the pardon gave him an absolute right to the restoration of his name to the Roll of Advocates.... It is clear that this court alone is competent to adjudicate upon this dispute between the petitioner and the public.... The fact that the petitioner, for the reason of having misconceived the powers of the Law Council, turned to that body for the enforcement of his right, does not negate the possibility that the petitioner's abovementioned argument may be finally disposed of in the present proceedings.... If we find the petitioner's argument to be well founded, and declare him entitled to renew practice as an Advocate, such a declaration will bind everyone, and the petitioner should experience no difficulty in having his name restored to the Roll. If, on the other hand, we decide that the pardon does not bring about the desired result, that ruling will equally resolve this dispute between the citizen and the public. One way or the other, I believe it is required of us to decide this whole question.... which is what I now proceed to do (my italics-M.B.P.).
It seems to me that Justice Agranat's attitude reflected his clear perception that the power of pardon in Israel required to be widely interpreted, as in England and America, as embracing also the pardon of an offender before his conviction. This attitude is to be gathered from his judgments, in A. v. The Law Council and in Matana. It so transpires from his citation and adoption of a statement in Halsbury's Laws of England that "pardon may, in general, be granted either before or after conviction," and especially from his own conclusion (in A. v. the Law Council [2], at p. 751):
from which I learn that the President has the power to pardon offenders both before and after conviction, either unconditionally, or with qualifications.
Justice Agranat gave practical implementation to his above perception by interpreting the consequences of the pardon in issue there in accordance with the customary approach in England and in the U.S.A. (ibid., p. 751).
A perusal of the two precedents reveals that none of the other Justices dissented from the interpretation according to which the President of Israel was competent to pardon offenders also before conviction; indeed, most of the Justices explicitly took the same view. Thus the difference of opinion between Berinson J. and Agranat J. in Matana, as to the comparison of sec. 6 of the Transition Law with Art.16 of the Order in Council (with Berinson J. refusing to recognize a presidential power to grant a pardon subject to conditions, owing to the absence in sec. 6 of such express provision), did not prevent Berinson J. from holding (perhaps on account of the first part of Art. 16) that the presidential power of pardon was exercisable also before conviction ( ibid., at p. 469):
the President's power of pardon is in a certain sense wider than that possessed by the High Commissioner. Whereas the High Commissioner was unable to pardon a crime before the offender was tried and convicted unless he turned King's evidence and led to the conviction of his accomplice (the first part of Article 16 of the Order in Council), the President is not bound by this condition and, so it seems to me, may pardon any offender even before he is brought to trial.
Landau J., at the end of his opinion in Matana, expressed his "complete agreement" with the judgment of Berinson J., from which it follows that he agreed also with the content of the above passage, or at least had no reservations about it.
Justice Cohn fully supported Justice Agranat's interpretative approach, and emphasized his view that the Presidential power was to be widely construed.
The general opinion, therefore, was that there was in Israel an existing, valid presidential power of pre-conviction pardoning. There was, however, a divergence of opinion in Matana on the question of equating the power of the President of Israel with that of the British Monarch.
It may be noted that Justice Agranat's approach has been followed in practice ever since the decision in A. v. The Law Council. This fact is confirmed in the judgment of Cohn J. in Matana (at p. 461):
Under sec: 6 of the Transition Law, 1949, the President of the State is empowered to pardon offenders and to reduce punishments. The Presidents of the State have exercised this power from 1949 until the day on which judgment was delivered in Matana v. Attorney-General (June 23, 1960, in the manner laid down for them in the judgment of this court (per Agranat J) in A. v. The Law Council, at 745 et seq.). That is to say, both the President of the State and the Minister of Justice, whose countersignature of the President's decision is required by the Law, and also the wide body of citizens who have had need of the President's grace, have always regarded this power of the President as equal and parallel in nature and scope to the power of pardon and reduction of punishments possessed by the Queen of England.
This practice (with which few were as familiar as Justice Cohn, who had held office as Attorney-General for a lengthy period) had been followed for some ten years when the decision was given in Matana, and in this context the learned Justice went on to comment as follows (at p. 462):
It has already been said more than once by this court (both during the Mandate and after the establishment of the State) that the court will hesitate very much to reverse a particular practice which has taken root during the years, and if this was said in respect of matters of practice which did not rest upon the authority of judicial precedent, how much more is it applicable to a matter of practice which rests upon a specific decision of the Supreme Court. As for myself, even if I were inclined to agree with the opinion held by my colleagues Berinson J. and Landau J. that the practice followed by the President of the State year after year is based upon too wide an interpretation of sec. 6 of the Transition Law, 1949 [and it will be recalled that the opinions differed there on the question of a pardon subject to conditions, and not as regards a pardon before conviction - M.B.P.], even then I would no venture today to change this practice which has received the seal of the Knesset at least by its silence (my italics - M.B.P.).
We are, therefore, talking about a practice that has now been followed for some decades. Initially the practice was founded on the single opinion of Agranat J, that is, on the ratio of his decision in A. v. The Law Council, and later also on the Matana decision.
That even distinguished jurists treated the decision in A. v. The Law Council as laying down a rule to be accepted, may be gathered from the following extract from a statement made by Prof. Y. H. Klinghoffer in a Knesset debate on 29 June 1965 ( Minutes of the Knesset, 43, p. 2319):
Another unconvincing argument sometimes advanced is that if a suspect be very ill, it would be an act of cruelty to put him on trial. Unconvincing, because in rare cases of this kind the President of the State would be able to grant a pardon before the trial commenced. The President is empowered to pardon offenders either before or after conviction. That was laid down by Justice Agranat in the case of A. v. The Law Council (my italics-M.B.P.).
The above remarks were made with reference to the question whether it were better to render it obligatory for the prosecution to charge a suspect and put him on trial when the evidence so warranted, or to leave the decision to the discretion of the prosecution. Prof. Klinghoffer clearly favoured the former option, reasoning thus:
In expressing my reservations I would recommend we abandon this method in favour of one that obliges the prosecution to put a person on trial when it is in a position to substantiate the charge upon sufficient evidence. When a particular act or omission is defined by statute as a criminal offence, the matter should properly be submitted for judicial determination, and it is not the concern of the prosecution, which is a part of the Executive authority, to relieve the suspect of responsibility for his act or omission by not putting him on trial, and precluding the competent court from judging him according to law. To entrust the prosecution with the option to decide on its own whether or not there be any public interest in holding a particular trial, and accordingly whether or not to institute criminal proceedings against the suspect, is to invite dangers of a political nature. This arrangement would amount to a conferment of power to pardon someone in advance, and convert the prosecution into a kind of pardoning institution, something that is not in keeping with its essential function (my italics - M.B.P.).
We accordingly see that Prof. Klinghoffer drew a clear distinction between an unqualified discretion given the prosecution whether or not to charge a suspect - according to its perception of the public interest in the matter - and the exercise of the pardoning power. The distinguished jurist saw such an option as holding out certain dangers, and undesirably conferring a power of advance pardon, whereas he accepted as a matter of fact the presidential power to grant a pardon, in rare cases, even "before the trial commenced."
In the end the Knesset took the middle path, but I shall come back to this aspect later.
4. Two conclusions may be drawn from what I have said so far:
(a) As regards the scope of the presidential power to pardon offenders before conviction, we are not without guidance, for the existing judicial pronouncements on the subject to the effect that the President does possess such power, cannot be said to be purely obiter;
(b) We are confronted with a practice that has taken root in Israel ever since the decision in A. v. the Law Council, that is to say, for some decades now.
There can be no doubt as to the importance of these two considerations in the determination of our attitude.
(c) Also carrying weight, purely as an interpretative indicator for me (and not as a source of legislation), is the fact of the lawgiver's silence on the instant point when the content of sec. 6 of the Transition Law was reenacted in sec. 11(b) of the Basic Law. This silence was maintained despite the clear trend of the precedents and the practice evolved and based thereon, and it stands out against the express addition in the Basic Law of a presidential power to "commute" sentences, the lack of which had been established in the precedent cited, together with the court's unanimous opinion that the President was empowered to pardon also before conviction. This silence and its implications are fully elucidated in the judgment of Shamgar P.
5. It is true, however, that a later legislative development sometimes does dictate a change in interpretative approach. Attitudes also change with the passage of time, and these changes come to be reflected in the decisions of the courts - in the manner of new wine poured into an old flask - if weighty reasons be found for departing from an existing rule, deeply rooted though it may be. If, for instance, the customary interpretation be found necessarily to misconceive the purpose of the provision concerned, or that it has not even the slightest foundation in the language of the provision, or that its implementation in the exigencies of new reality poses a real threat to the maintenance of good government, then I should be inclined to construe the pardoning power restrictively as being confined to the post-conviction stage alone.
I have come to the conclusion, however, that there are no weighty reasons for disturbing the existing precedents and practice. My reasons for so concluding are the following:
(a) The wide interpretation given the term "offender" finds ample justification in the language of the statutory provision in question. My colleague, Barak J., acknowledges that, linguistically speaking, the term "offender" could embrace also a suspect who has yet to be tried and convicted; hence, in his opinion, the wording of the provision alone does not advance our inquiry one way or the other. Yet the learned Justice suggests at the same time that only someone who has already been convicted is an offender, as appears, for instance, from the following passage in his judgment (par. 25):
Even an admission by the applicant for a pardon that he committed an offence, is of no consequence, for he is presumed innocent until convicted by the court.
I must confess that I find the emphasis given to circumscription of the term "offender" in the context of our instant inquiry, somewhat perplexing. It is common cause that the main (some hold, the only) purpose of a full pardon, is to make amends for a serious miscarriage of justice which has resulted in the conviction of an innocent person. If that be the main (or sole) purpose, then the recipient of the pardon is no "offender" at all, but the victim of an error. If we adopt the arguments of the petitioners, and of certain jurists, that only the court is competent to stamp a person as an "offender" for the purposes of pardon, we shall find that it is precisely that kind of error which the President is unable to repair - a situation that is contrary to all logic. Of what avail is it for the court to find, upon hearing evidence and argument, that the accused indeed committed the crime - and thus branding him an "offender" - if the essence, and main purpose, of a pardon be to proclaim that he is not such? This reasoning alone would warrant the conclusion that an "offender" includes someone to whom the commission of an offence is attributed.
It is pertinent, moreover, to recall that Art. 16 of the Order in Council empowered the High Commissioner to pardon an offender (an accomplice) before conviction, if he was prepared to give information and evidence concerning the principal offender or any such offender. We must bear in mind the proximity in time between the repeal of the said Article 16 and the enactment of sec. 6 of the Transition Law, a proximity which provides further indication that the term "offender," as already pointed out in the judgment of Shamgar P., was intended to refer to someone to whom the commission of an offence "is attributed." The learned President cited many convincing examples of the lawgiver's use of the term "offender," in a variety of contexts, from which too one may learn that this term does not necessarily mean someone who has already been convicted. In other Laws the term may indeed import otherwise, depending upon the legislative context and intent, but the abovementioned examples all relate to the same or closely the same kind of material as our present matter (for instance, general amnesty), and convincingly show that the term "offender" should not be understood only as someone who has been tried and convicted. In addition to the above illustrations, among many other possible ones, I might also mention sec. 6 of the Secret Monitoring Law of 1979, which provides a framework for secret monitoring, inter alia, if necessary "to prevent offences or detect offenders." It is clear from the context that the Law envisages the monitoring and exposure of the conversations of a person involved in a criminal act (whether not yet committed, in the process of commission or after its commission) and all, of course, in the stage preliminary to the trial and, certainly, before conviction of the suspect.
In essence, my learned colleagues and I all agree that linguistically speaking sec. 11(b) of the Basic Law suffices, as it stands, to encompass also the power of pardon before conviction. The requisite interpretative nexus for this purpose is there, and the statutory provision cannot be said to lack a linguistic foundation for such a construction.
(b) We must now, after disposal of the linguistic aspect, deal with the main criterion, namely, the legislative purpose of the pardoning directive.
Justice Barak holds in his judgment that a construction according to which the President of the State may pardon someone before his trial and conviction, is inconsistent with the purpose of the pardoning directive. He states that in order to choose between the possible linguistic options we must turn to the legislative purpose, and he holds the true objects of the pertinent statutory provision to be those enunciated by Justice Agranat in A. v. The Law Council and in Matana, and none other, namely:
The primary purpose... is to redress the wrong done to a person who was convicted while innocent, and the second purpose - the value of which should also not be underestimated - is to reduce the sentence of the offender in circumstances which justify this. It is clear that the exercise of such a power by one of the highest State authorities is essential for the effectiveness of any governmental regime, since in no country whatever has there yet been created a system of justice capable of perfect and unerring operation, and of dispensing justice in every case without fail ( A. v. The Law Council, at p.751).
Justice Barak then goes on to make this comment:
This reasoning naturally only holds true in relation to a convicted offender. It is not at all applicable to someone who has yet to be convicted. How, then, is this reasoning of Justice Agranat to be reconciled with his view that the President has power to pardon before conviction? Such power would necessitate a different rationalisation, of the kind that is not to be found either in A. v. The Law Council or in Matana (my italics - M.B.P.).
This seems a cogent argument in support of the conclusion that, measured by the test of legislative purpose, the presidential power is restricted in its exercise to a pardon after conviction only. It is not so, however. Thus, for instance, in A. v. the Law Council Justice Agranat mentioned additional objects of a pardon, remarking inter alia as follows (at p. 755):
Third, I have not overlooked the possibility that a pardon may also be granted for reasons which do not stem, necessarily, from the innocence of the convicted person.
These remarks link up with what Justice Agranat said later in the Matana decision (at p. 451):
It is quite easy to think of a case in which the need to use this system would arise when the public interest alone, and not that of the prisoner, requires his release from custody. It may, for example, be proper to liberate a prisoner who is a national of an enemy state on condition that he leave Israel territory immediately and permanently, in order to facilitate an international arrangement which will ensure, in return for such a pardon, the immediate release of a "Zionist prisoner" in custody in that state.
The above example happens to relate to a convicted prisoner, yet this underlying purpose is not to reverse an injustice but to prefer the public interest, to which the rule of equality before the law must bow. That is to say, we have here a conflict between two very important interests: one - equality before the law, which requires that every offender against the law should answer for his conduct; the other - the safeguarding of a vital public interest. The proper balance between the two is the determining factor. The President of the State was in the same predicament in relation to the matters raised in the petitions before us. Thus he declared his conclusion that despite finding merit in the opposing viewpoint, he was satisfied at the time that vital security interests of the State were at stake and also that it was necessary to put an end to the "devils' dance," as he described it, and therefore he decided to accede to the requests for a pardon.
The primary purpose of a pardon, at least until the retrial procedure was instituted, has indeed been to correct an injustice resulting from an error in judicial proceedings. But that has not been the only purpose of a full pardon. Thus, as already mentioned, it was possible under Article 16 of the Order in Council to pardon an accomplice (before trial) in order to induce him to give information and evidence against the principal offender or any such offenders. The purpose of such a pardon was not to reward its recipient, but to achieve an object considered by the pardoning authority more important than trying the person pardoned. (A similar approach is also to be discerned in civil law - see Mistry Amar Singh v. Kulubya [47], where the plaintiffs claim, though tainted with illegality, was sustained in order that the purpose of the law should not be defeated.) For the attainment of the same purpose a pardon or clemency may conceivably also be granted to a convicted offender, by way of a reduction of sentence, if the latter, only at that late stage, is prepared to disclose important information against other offenders who committed serious crimes.
Furthermore, such rationalisation - that the public interest sometimes prevails over the interest of bringing the offender to trial or of having the trial run its full course - is to the best of my understanding, contrary to the opinion of Barak J., also to be found in the cases of A. v. The Law Council, and Matana. I base this conclusion in the first instance on the abovementioned remark of Agranat J. in A. v. The Law Council (at p. 755), that a pardon may also be granted for reasons "which do not stem, necessarily, from the innocence of the convicted person," and also on his following statement in the same case (at p. 747) :
lts main purpose - and I do not overlook its other purposes - was and remains to declare before all that the person tried and convicted, and now receiving a pardon, is free of guilt and that his offence has been wiped out (my italics-M.B.P.).
Thus, we seem to find in the two abovementioned precedents precisely such "different rationalisation," according to which a full pardon may also be granted to someone other than a wrongly convicted person who is serving his sentence. This pardoning consideration, provided it is applied correctly and carefully in the proper cases, operates with the same validity and force both before and after the conviction of the offender. In fact, even Justice Barak countenances the possibility that it may be better, in rare cases, to pardon a person before conviction rather than to stay the proceedings against him - for instance when the person is suffering from a malignant disease - save that my learned colleague does not consider such exceptional cases to warrant a wide interpretation of the statutory directive. By the same token I would hold it desirable, in a rare case, for a person to be pardoned before trial and conviction for the sake of protecting a vital public interest. The Attorney-General is indeed the competent party to decide whether a person shall stand trial or not, to which end he may, even must, weigh considerations of a social or security nature. Like Justice Barak, I too find support for my view in the report of the Agranat Commission on the Powers of the Attorney-General (1962), where it was stated, inter alia, that
In certain circumstances a matter of security, political or public interest may dictate that no criminal charge be preferred (p. 6).
The Commission dealt also with the Attorney-General's need to consult with the political authorities when making his decision, reporting thus (p. 13):
The stated duty to consult arises particularly when criminal proceedings are being instituted in relation to a matter of security, political or public interest. In such event it is always incumbent on the Attorney-General to consider whether the act of instituting criminal proceedings (or halting the same) is not more likely to prejudice the interests of the State than refraining from taking such action. This the Attorney-General will only be able to do after having sought information and guidance from those who carry the primary responsibility for safeguarding the State from the security, political and public aspects - that is to say, from those who, so we must presume, are more experienced and knowledgeable in those fields than we are. As already indicated, he will generally need to refer to the Minister of Justice for the required direction and advice; but sometimes, that is in cases which give rise to questions of "high policy," there will be no alternative but to obtain guidance from the Government as a body (my italics - M.B.P.).
I shall come back to these statements later. First, however, I wish to consider the situation that arises when pursuant to sec. 59 of the Criminal Procedure Law (Consolidated Version) of 1982, a police investigation has to be opened upon a complaint of the commission of a felony, and it transpires that the very conduct of the investigation (including the taking of statements from witnesses) may seriously impair the security of the State. Who will then be empowered to decide whether the investigation shall be completed or discontinued?
Mr. Harish, the Attorney-General, has submitted that he lacks the authority to order that a police investigation be discontinued (or not opened after the police learn of the commission of a felony). It is arguable, perhaps, that the police, being an arm of the Attorney-General for criminal investigation purposes ( Reiner v. Prime Minister of Israel [20], pp. 485, 486), is obliged to act as directed by him, thus rendering the Attorney-General competent to order discontinuance of the investigation. The matter, however, is far from clear. For instance, there is the express directive in sec. 60 of the above Law, that upon completion of the investigation of a felony, the police shall transmit the material to the District Attorney; the police, however, may refrain from investigating lesser offences, though only for the reasons set forth in sec. 59 of that Law. A measure of support for this separation of powers is even to be found in the following passage from the judgment of Justice Barak (in paragraph 25, though his remarks were intended for a different purpose):
This conclusion as to a "separation of powers", between the presidential pardon and the powers of other State authorities, is reinforced when regard is had to Israel's general constitutional framework. The other State authorities (the police, the prosecution and the courts) have the means to establish the facts. The police has its investigating facilities and seeks to reach an assessment of the factual situation. The prosecution, to whom the police must transmit the material, will handle and process the same until delivery of the judgment. The courts possess the institutional and normative facilities for elucidating the question of innocence or guilt.
That is to say, each authority has its own field of responsibility. The police, upon receiving a complaint or otherwise learning of the commission of a felony, is obliged to conduct an investigation, and upon its completion to transmit the evidence to the appointed authority, the prosecution, which only then comes into the picture. It is even doubtful whether the police may halt its investigation in cases of felony, even where the police considers the Attorney-General to have good reasons for wishing to do so. On a plain reading of the conclusion to sec. 59 of above Law, the police has such a discretion (on the grounds of there being no public interest involved) only in relation to misdemeanours, and not felonies. And most important, even the Attorney-General's power to intervene in the investigation by directing that it be discontinued is, as already mentioned, a matter of doubt. Thus, there is the express provision (sec. 61 of the Law) that the Attorney-General may direct the police to continue investigating if, after receiving the material, he "considers it necessary for a decision as to prosecution or for the efficient conduct of the trial." This provision would seem to be superfluous if the Attorney-General is indeed competent to decree at will the completion or halting of police investigations.
On the assumption that the police is legally obliged to complete its investigation of a felony, and then to refer the material to the District Attorney, the grant of a pardon would indeed be the only way of halting an investigation if it endangered a vital public interest such as the security of the State. The abovementioned view of the Agranat Commission that the security, political or public interests of the State may in certain circumstances require that no criminal charge be preferred, holds good also as regards the halting of an investigation for similar reasons. It is quite likely that in this situation (in contrast with the stage when the police refer the material of the investigation to the District Attorney) a presidential pardon will be the only way of halting the process.
Again, even assuming the Attorney-General to have power to halt a police investigation, a difference of opinion may yet arise, in a particular case, between the Attorney-General and the State authorities with whom, as the Agranat Commission required, he must consult. While it has to be presumed, according to the Commission, that the security authorities are more experienced and better informed than the Attorney-General, and though they bear primary responsibility for safeguarding the security and other vital interests of the State, the latter nevertheless has to make his own decision on matters within his sphere of responsibility. In this situation, with each party insisting upon its own viewpoint, how will the conflict be resolved? It seems to me that the situation bears comparison with the conflict that arises between the need to withhold privileged evidence and the right of the accused to defend himself against a criminal charge. It is a hallowed principle of penal law, embedded in the structure of a democratic regime, that the accused shall be given every opportunity to avail himself of any evidence in the hands of the prosecution. Yet this right has been qualified in the Evidence Ordinance (New Version) of 1971, sec. 44(a) whereof provides:
A person is not bound to give, and the court shall not admit, evidence regarding which the Prime Minister or the Minister of Defence --- has expressed the opinion that its giving is likely to impair the security of the State --- unless a Judge of the Supreme Court on the petition of a party who desires the disclosure of the evidence finds that the necessity to disclose it for the purpose of doing justice outweighs the interest in its non-disclosure.
If a Supreme Court Justice (not involved in a particular criminal case) should come to the conclusion that the need to disclose certain evidence, in the interests of justice, has to defer to the State interest in keeping such evidence privileged, he will uphold the latter even if the accused's ability to defend himself is affected thereby. If, on the other hand, it is decided that the evidence should be disclosed in the interests of justice, there would still be the possibility that the security authorities, taking a different view as to the measure of harm that may result, would prefer not to disclose the evidence, even if this should lead to the accused's acquittal. The acquittal may represent a more valuable prize to the accused than even a pardon, since he may be guilty and undeserving thereof; yet, in the view of the authorities concerned, the one interest (equality before the law) will have to yield to the other (safeguarding the security or existence of the State).
The above situation was deliberated in the case of Livni v. State of Israel [21], where Justice Barak commented as follows (at p. 736):
Once the court has decided that the evidence should be disclosed, the prosecution is faced with the dilemma of deciding whether or not to continue with the criminal proceedings. If the trial continues, the prosecution will have to disclose the evidence; if the prosecution believes that disclosure of the evidence will endanger the security of the State, it may have to stay the proceedings and sometimes even cause the accused to be acquitted. Thus, whereas initially the conflict was between the need to disclose the evidence in the interest of doing justice, and the need to keep it privileged in the interest of State security, we now find - upon the decision of the court - that the conflict is between the need to proceed with the trial by way of disclosure of the evidence, and the need to keep the evidence privileged by way of the discontinuance of the trial. The former conflict is resolved by the Judge in adoption of the procedure prescribed in sec. 44(a) of the Evidence Ordinance; the latter conflict is resolved by the prosecution within the framework of its general discretion in the conduct and stay of criminal prosecutions.
Equally in point are these remarks of Barak J. in continuation (at p. 735):
On the other hand, there is the consideration that it is sometimes in the public interest to keep the material of the investigation privileged, if its disclosure may prejudice the security of the State. It is an important public interest to protect the security of the State against all harmful subversive acts, which are mostly the product of underground planning and organisation. The struggle against such harm calls for the gathering of intelligence information without its sources becoming known.... This war is being waged by the security services, whose struggle would be gravely prejudiced by the uncovering or identification and public exposure of these sources (Miscellaneous Applications 52/82)............................................ This consideration asserts itself in every country, but does so with particular sharpness in the State of Israel, whose security has been threatened ever since its establishment. We are a "democracy on the defensive"... which has to fight for its survival, not only in large-scale wars but also in the day-to-day campaigns thrust upon it by its enemies. We must not close our eyes to this bitter reality.
It cannot be overlooked that those who discharge a clear security function find it especially difficult to act always within the law. The measure of departure may vary from country to country, but it exists as a fact, also in democratic regimes whose fidelity to civil rights is beyond question. In this regard Prof. B. Akzin has commented as follows ( Elements of International Politics, Akademon, 1984, in Hebrew, p. 332):
It should be added that while the police strives (or, at least, should strive) to act within the frame of the existing law, the intelligence and espionage services, including counter-espionage, are less punctilious about observing the law, and it sometimes happens that they knowingly and seriously violate it. Even in times of peace, let alone in times of war, they engage in acts of violence and sabotage, both in foreign countries and in their own. This reality leads to situations of embarrassment for countries which adhere to the rule of law, and places them in the dilemma of the comparative priority between the principle of legality and intelligence interests. That is no easy dilemma. If we compare the practice of some established democracies in this regard, we shall find that in the United States, for instance, the scope of intelligence operations is often (though not invariably) curtailed by the need to keep within the law, whereas in Britain and France the principle of legality does not restrict intelligence operations to the same extent (my italics - M. B. P.).
Naturally, the smaller the deviation from the legal norm, the easier it would be to reach the optimal degree of harmony between the law and the protection of the State's security. But we, as judges who "dwell among our people," should not harbour any illusions, as the events of the instant case well illustrate. There simply are cases in which those who are at the helm of the State, and bear responsibility for its survival and security, regard certain deviations from the law for the sake of protecting the security of the State, as an unavoidable necessity.
Barak J. has correctly pointed out in the Livni case [21] that when the two interests of a fair trial and the security of the State are competing for primacy, the conflict must be resolved. Both interests are of concern to the public, and the resolution of the conflict takes different forms in different countries, the form sometimes changing within the same country. He added that
this struggle between the conflicting interests is particularly sharp in Israel, since on the one hand we are insistent upon fairness in criminal proceedings and maintaining public confidence in them.... while on the other hand we are subject to considerable security risks, which have beset us for a long period (ibid., p. 735).
It is true that when the issue of privileged information arises in the course of a trial, the conflict between the two interests is resolved by a Supreme Court Justice (who is not hearing that particular case). But even when he rules that the evidence must be disclosed, that is not the end of the matter, for such disclosure may be avoided by a discontinuance of the trial, leading even to the acquittal of the accused.
By analogy it seems to me that a decision by the Attorney-General to refer a complaint to the police - despite the objections of the Executive authorities (in our present context, the inner Cabinet) that so to do would harm security interests - is also not necessarily final. There would be nothing improper in the Attorney-General's attitude that an investigation should be conducted notwithstanding the advice given him to the contrary, but equally there is no ground for questioning a resolution of the dilemma by way of its referral to the President as the Head of State - the symbol of the people whom he represents. I do not hold that the only course open to the Executive in the above circumstances, is for the Government to dismiss the Attorney-General, for his attitude is a legitimate one, and he may properly adhere thereto. Nor, by the same token, is any injury done to the standing of a Supreme Court Judge who has ruled that the evidence in question should be disclosed. The same may also be said of the authorities entrusted with the security and survival of the State, and who bear primary responsibility for this onerous task. In the present case it may be presumed of the President that he properly considered all aspects of the dilemma, and so it indeed appears from his public statement quoted in the judgment of Shamgar P. He mentioned his understanding of the opposing viewpoint, but was convinced that the interests of security should prevail. My respected colleagues and I all agree that if a presidential power to pardon before conviction exists, the considerations weighed by the President at the time of granting the pardons are valid.
At the same time, however, it is necessary to stress the gravity of the offences disclosed before us, the nature and quality of which should alert us to the need for a thorough review of the security establishment, with a view to the determination of just norms and directives as far as this is possible.
Justice Barak concedes the possibility of a valid presidential power of pardon before conviction, for exercise on rare occasions alone, but holds this to be undesirable as likely to increase in frequency and become the norm. He is accordingly deterred from building constitutional norms on what he regards as hope alone. With all due respect, I find no adequate basis for this apprehension. On the contrary, it was shown to us that the pre-conviction power of pardon has been exercised most rarely during the past thirty-five years, since the decision in A. v. the Law Council. That is no small guarantee that this will continue in the future as well, as indeed it should. Moreover, already in the Matana case, the fear of an excessive exercise of this power was allayed by Agranat D. P., in these terms (at p. 454):
Nor have I overlooked the fact that to endow the power in question with its "full" content may lead to its excessive use, which in turn involves the danger that the authority of the law in the eyes of the public will be weakened. My reply to this point, however, is that every instrument of pardon by the President requires the countersignature of the Prime Minister or one other Minister (sec. 7 of the Transition Law, 1949) [now sec. 12 of the Basic Law: The President of the State - M.B.P.]. This means that even if the decision to pardon or to reduce a sentence must be the personal decision of the President, it is also conditional upon the recommendation of the Minister concerned. This Minister will ordinarily be the Minister of Justice who has the means of conducting a precise investigation into the circumstances of the case before submitting his recommendation to the President. It is clear that this recommendation, and therefore the decision to pardon as well, are subject to review by the Knesset and it is this possibility which must be regarded as the guarantee laid down by law against the danger referred to.
There is therefore someone who is answerable to the Knesset (the Prime Minister or some other Minister) and this safeguard is now fortified by the possibility of challenging the pardoning decision indirectly. Another important factor is the special status of the President as representing the people and standing above political or public controversy. The State President presumably weighs all necessary considerations before deciding to exercise his power to grant a full pardon, whether before or after conviction. This is a power which has to be most rarely exercised. The pre-conviction pardon was not designed for the purpose of redressing an injustice done to the person pardoned, for the fact of his guilt is taken for granted and he is assumed to have committed the offence attributed to him (by the police or the prosecution). What has to be weighed, therefore, is the seriousness of the offence against some other interest - humanitarian, security, and the like. In other words, the pre-conviction pardon always entails a conflict between the interest of equality before the law and some other, vital, extraneous interest. This fact acts greatly to restrict the range of cases in which the exercise of this power will be justified.
A constitutional directive gives expression to the will of the people, to its "credo." If under a directive of this kind the power to pardon offenders has been conferred on the President, the latter must be seen as the proper authority for the discharge of this difficult task (with the countersignature of the Minister concerned, who is also answerable to the Knesset, and subject further to indirect judicial review of the President's decision). In those cases where the offender benefits from a pardon, though not for the reason of his innocence of the charge but for the protection of a higher interest - whether before or after conviction - the principle of equality before the law will well be breached, but this will happen also when, for example, an acquittal results from the ruling of a Supreme Court Judge that privileged information be disclosed, in the circumstances outlined above.
My abovementioned remarks as to the President being the ideal authority to grant a pardon, find support in the following statement of an American authority quoted in the judgment of Shamgar P.:
...Crime is an offense against the people, prosecuted in the name of the people, and the people alone can bestow mercy by pardon... the people may confer the pardoning power upon any officer or board that they see fit ( Am. Jur., at p. 10).
Like Justice Shamgar, I believe that the decisive factor is not the rank of the State President within the Executive hierarchy, but the fact that he symbolises the State and represents the people in holding and exercising the power of pardon .
An undesirable exercise of the pardoning power must be avoided at all times, whether it takes the form of an uncontrolled or unjustifiable remission of sentence, or the grant of a full pardon after conviction. A reduction of punishment granted one offender but not another in comparable circumstances (so we assume), amounts naturally to a discrimination between equals. An ill-timed pardon, or one granted on grounds already deliberated by a judicial tribunal, is tantamount to an intervention in the domain of the judicial authority. It has to be borne in mind that the facilities available to the court - rules of procedure and evidence for the greater part determined by legislative means and partly by judicial means - offer the most effective may of establishing the truth and ensuring a fair trial. The reversal by non-judicial means of a judicial decision, particularly in an age when a retrial is possible, is a process the retention of which may well be questioned. But that is a matter for the lawgiver. What concerns us here is that the grant of a pardon for reasons other than the correction of an injustice, and involving a conflict between the principle of equality before the law and some other vital interest, invokes a power which should rarely be exercised, and only after much careful consideration.
(c) From the aspect of the separation of powers, the President must be seen as holding a power termed "residuary" (by Justice Barak) or "reserve" (by Justice Agranat in the cases of A. v. the Law Council and Matana ). It is right that the power remain of such a nature, and that the President refrain from exercising it as long as some other authority is still competent to act in the desired direction. In his judgment (par. 25), Justice Barak cites the example of a pardon granted after conviction but before sentence is passed. I hold, unlike my learned colleague, that in this case the power itself is there, but its exercise (the grant of a full pardon alone is possible at this stage) would entail a gross interference with the judicial function and a possible lowering of its prestige. The same applies when a pardon is granted where the possibility of a retrial exists. Like the President, Justice Shamgar, I would not discount the need for a change in the existing constitutional arrangement on the subject, perhaps along the lines proposed by legal scholars such as Professor Feller.
(d) Justice Barak states that a pardon is given without publicity, whereas a stay of criminal proceedings is manifest and publicly known. I believe this picture should be put into its proper perspective.
(1) The Attorney-General's decision to stay proceedings must be reasoned, and conveyed to the complainant (pursuant to sec. 63 of the Criminal Procedure Law, Consolidated Version). Information concerning the decision must not, however, be disseminated among the general public, but may only be furnished to certain specified authorities consistently with the provisions of the Crime Register and Rehabilitation of Offenders Law, 1981 (see sec. 11(a) thereof). It may be noted that these authorities are mostly public organs (the Attorney-General, the police, the General Security Service, and others).
(2) As regards a pardon, the initiative is usually taken by the Ministry of Justice and the warrant requires to be countersigned by the Prime Minister or another Minister, normally the Minister of Justice. In case of complaint originally lodged with the police, it is clear that the fact of the pardon will also be brought to its notice, if it has to discontinue the investigation. Such discontinuance would also obligate the police to notify the complainant accordingly (sec. 63 of the Criminal Procedure Ordinance, Consolidated Version). There may be some difference in the measure of disclosure occasioned in each of the two cases, but the gap should not be exaggerated.
6. With reference to the petitioners' prayer concerning investigation of the complaint lodged with the police, I concur in the opinion of my learned colleagues that at this stage the court should rest content with the Attorney-General's intimation that the whole incident will be fully investigated. The petition, therefore, should be dismissed in this regard.
It has occurred to me that this result - dismissal of the prayer regarding the investigation - might have the effect of converting an indirect challenge of the pardons into a direct one, which would not be permissible under the existing law. I have not, however, delved too deeply into the question and, having regard to the attitude of my colleagues, have likewise preferred to deal with the question of the inherent validity of the pardons.
7. Having affirmatively answered the question as to the President's power to grant the pardons here deliberated, I must now deal with the second question confronting us, namely: were the requirements for the grant of such pardons satisfied?
I should state that I have found the Warrants of Pardon to provide only a general description and not to be sufficiently particularised, though less so in relation to the Head of the General Security Service. In the latter case it is recorded that the pardon was to extend to all the offences "connected with the incident called 'bus no. 300' and occurring on the night between 12 April and 13 April 1984, whether committed on the day of the incident or subsequently in connection therewith until the date of this Warrant." In the remaining Warrants it was stated that the pardon extended to all the offences "connected with the so called 'bus no. 300 incident,' and committed from the time of the incident on the night between 12 April and 13 April until the date of this Warrant." On its own this would be an inadequate particularisation, but with the declarations we have of the pardoned persons, together with the contents of the pardon applications as well as the subsequent Warrants, we now have sufficient particulars to identify the offences concerned. I need hardly add that the pardon extends solely to those offences and none other.
As to the State President's grounds for granting the pardons, we ruled earlier (on 1 July 1986) that there was no need for any declaration to be lodged concerning his reasons for deciding as he did.
From the material before us it may be learned that the negotiations with the President were commenced some considerable time before the pardons were granted, and only the final, formal stage was completed in haste, on account of the pressure arising from the surrounding circumstances. It has been declared that all the particulars relating to the incident were explained to the President, and I have no reason to doubt the truth thereof.
8. In conclusion, I must emphasise that I, like my respected colleagues, have endeavoured to deal with the central issue - the President's power of pardon before conviction - in isolation from the stormy public controversy aroused by this unfortunate incident. Such detachment is enjoined by our judicial task, which we must fulfil to the best of our understanding. We are obliged to adopt an attitude, even with regard to matters of public controversy and even though part of the public may not approve of that attitude. What is conclusive is the court's decision, as distinct from its views (by way of analogy, see Shalit v. Minister of the Interior [22], at p. 520, per Witkon J.).
9. For the reasons given above I concur in the judgment of the respected President, Justice Shamgar, and in his conclusion.
BARAK J .
I am of the opinion that the pardon granted by the State President to the Head of the General Security Service and three of his assistants is null and void, for lack of a presidential power so to act. It would follow that, as intimated by the Attorney-General, the investigation is to continue.
A. Our Judicial Function
1. After carefully perusing the judgment of my respected colleague, President Shamgar, I find myself agreeing with some of his opinions and not with others. The whole issue is important, lying as it does at the heart of our constitutional life. Interwoven with the immediate problem of the presidential power of pardon and the manner of its exercise, are questions of the rule of law and its supremacy. All these we shall examine from the legal standpoint. The issue has provoked a stormy public reaction, but we have not allowed that to influence our decision. We function in accordance with constitutional criteria and fundamental legal principles which reflect the "credo" of our national life. It is not passing moods that guide our approach, but fundamental national perceptions as to our existence in a democratic state. This guideline was succinctly stated by Shamgar P. , in Neiman v. Chairman of 1lth Knesset Elections Central Committee [19], in these terms (at p. 259):
Judicial decisions in constitutional matters, even in difficult cases, should properly be founded and shaped according to principles rather than considerations of policy structured according to what is viewed as desirable and responsive to the need of the hour or the feeling of the majority.
We are aware of the public controversy that is raging around this matter, and in the dynamics of political life our judgment here may well come to be used as a lever in the struggle between the opposing political forces. That we regret, but we have to fulfil "our function and our duty as Judges," as was pointed out by Landau D.P. in Duikat v. Government of Israel [23], at p.4:
It is still much to be feared that the court may come to be seen as having abandoned its proper place and descended into the arena of public discussion, and that its decision will be enthusiastically welcomed by a section of the public while another section loudly and utterly rejects it. In this sense I see myself - as one whose duty it is to decide in accordance with law any matter brought before the court according to law - constrained to proceed undeterred in the discharge of my task. Even so, I know full well that the wider public will look not to the legal reasoning but to the final conclusion alone, with the attendant risk of damage to the rightful standing of the court as an institution beyond the divisions of public controversy. But what shall we do, if that be our function and our duty as Judges.
We are an arm of government, whose task it is to review the functioning of the other authorities, so as to ensure the government's adherence to the rule of law. These arms of government are of high status, but the law stands above them all. We should be failing in our judicial duty, were we not to pass under review, within the framework of petitions properly lodged, the activities of other State authorities in the circumstances disclosed in the petitions before us. I propose first to examine some of the questions on which I share the opinion of Shamgar P., and then to deal with the State President's power to grant a pardon before conviction. Following that, I shall endeavour to clarify my reasons for dissenting from the judgment of my learned colleague, Shamgar P., and shall conclude with some general remarks pertaining to the functioning of the State authorities in the present case. I regret the length of this judgment, but I did not have enough time to write a shorter one.
B. Locus Standi
2. Like Shamgar P., I hold the petitioners to have due standing to approach the court in the present matter. I do so for various reasons. In the first place, a number of persons lodged complaints with the police relating to offences committed in the "bus no. 300 incident." Under sec. 58 of the Criminal Procedure Ordinance (Consolidated Version), it is open to "any person" to lodge a complaint with the police, and the complainant must be informed of a decision not to investigate the same (sec. 63). He may then lodge an objection with the Attorney-General, whose decision is subject to judicial review and the complainant certainly would have standing in such a petition (Ashkenazy v. Minister of Defence [24], at 371). It is true that not all of the petitioners lodged complaints, but their standing before the court may be recognized on a second ground, which I shall now state.
When there arises before the Supreme Court a legal problem of constitutional import, the court will take a liberal view in matters of legal standing:
in such cases it is desirable to grant access to the court, without examining too carefully the interest at stake, provided this is in furtherance of the rule of law ( Segal v. Minister of the Interior [17], at p. 443).
The rule of law would be so served in the present case, having regard to the allegation that the Head of the General Security Service, and a number of his assistants, committed very serious offences involving loss of life and interference with the processes of investigation and the administration of justice. According to the material before us, these allegations - raised by the Attorney-General, Prof. Zamir - were not being investigated, though such investigation was said to be called for. The petitions accordingly involve basic questions of the rule of law, of equality before the law and of the susbservience of the principal centres of power in the State to the law as it stands. In these circumstances it is fitting that the petitioners be recognized as having sufficient standing to approach the court as they have done.
C. The Petition Concerning the Investigation
3. Some of the petitioners have concerned themselves with instigating a police investigation into the "incident." In a written communication received by us from the Attorney-General (on 15 July 1986), it was intimated as follows:
The attitude of the Attorney-General, communicated here with the confirmation of the Inspector General of Police (the remaining respondents have no standing whatever as regards the investigation), is that the police will investigate the said complaints pursuant to its duty under sec. 59 of the Criminal Procedure Law (Consolidated Version) 1982.
In his oral argument before us, the Attorney-General repeated his above intimation in these terms:
There will be a police investigation. The investigation will be conducted without qualification or reservation, until its conclusion, and will encompass all levels from top to bottom, including the political hierarchy. It is not intended to leave any matter uninvestigated, nor to exclude any person from the investigation.
The Attorney-General further emphasized that the investigation had already commenced, and in that situation, he argued, there was no room for confirmation of the order nisi - as prayed for by some of the petitioners - but the petitions, so far as they related to the investigation, should be dismissed. I agree with Shamgar P. that the Attorney-General's view should be sustained.
D. The State President as a Respondent
4. A number of petitioners joined the State President as a respondent. We ordered that his name as a respondent be deleted. As was pointed out by Shamgar P., this ruling was dictated by sec. 13(a) of the Basic Law: The President of the State, under which "the President of the State shall not be amenable to any court or tribunal... in respect of anything connected with his functions or powers." The President's act of pardon although, in my opinion, ultra vires, was nevertheless "connected with" his functions or powers, so that this court has no jurisdiction to entertain any direct challenge against his conduct. In the criminal appeal in Matana v. Attorney-General [25], Berinson J. noted (at p. 979) that when the President purported to act within the scope of his functions and powers, he would, if he exceeded these, be subject, like everyone else, to the laws of the State, and "amenable to the jurisdiction and authority of the courts." It seems to me, however, that even when the President exceeds his powers, but does so in a matter connected with his functions and powers, in good faith and in furtherance of what he considers to be the discharge of his duties - this court will have no jurisdiction over him. This limitation falls away where it is not sought to render the President answerable directly, but only to challenge indirectly the legal competence of a presidential act. The question arose in Bar Yosef v. Minister of Police [12], where the Supreme Court held as follows:
We accept that the State President has a discretion in the exercise of his power under sec. 11(b) of the Basic Law: The President of the State, and that this discretion - as distinct from the President himself - is, in proper circumstances, subject to indirect judicial review.
This perspective emerged from the approach of the Supreme Court in the abovementioned criminal appeal in Matana v. Attorney-General [25], where Berinson J. commented thus (at p. 786):
If indeed the President lacked authority to act as he did, there would be no need in the present case to disqualify the act itself. It would suffice for us to refrain from granting it validity and from aiding in its implementation, so far as this depends on us.
Elsewhere in the judgment, he added:
this does not mean that the legality of his official conduct and acts which may be prejudicial to the individual, cannot be indirectly reviewed without the President himself appearing as a party.
In the Attorney-General's original reply to the petitions (dated 30 June 1986), he noted that "once a pardon has been granted to all the members of the General Security Service who are mentioned as suspected of having committed the offences attributed to them, there is no longer any ground for investigating this complaint." This approach is challenged by the petitioners, and incidental to this main line of attack (against the Attorney-General), they are also challenging the President's pardoning decision. That they are entitled to do.
Such indirect judicial review is essential, for in its absence the power becomes unlimited in practice. Unlimited powers wielded by government authorities are alien to a democratic regime. Absolute powers, as Justice Douglas has rightly pointed out, are "the beginning of the end of liberty" (see New York v. United States [60], at p. 884, which statement was cited by this court in Kahana v. Speaker of the Knesset [26], at p. 92). The same is true of the pardoning power, which is not publicly exercised, the exercise of which need not be reasoned and which is little known to the public (see M. Gottesman, "Arbitrariness & Sympathy: The Criteria for Granting a Pardon," Mishpatim 1 [1968], 211; R. Gabison, "Arbitrariness & Sympathy: A Further Note," ibid., p. 218; D. Friedman, "Amnesty: Disclosure of Reasons," Hapraklit 25/1 [1969], 118; M. Ben-Ze'ev, "Matters of Amnesty," Hapraklit 25/2 [1970], 368). Such a power, if not subject to judicial review - even if only indirect - poses, upon its improper exercise, danger of the kind most destructive to all orderly government. Bentham has clearly outlined this danger:
From pardon-power unrestricted, comes impunity to delinquency in all shapes; from inpunity to delinquency in all shapes, impunity to maleficence in all shapes; from impunity to. maleficence in all shapes, dissolution of Government; from dissolution of Government, dissolution of political society ( The Works of Jeremy Bentham, ed. Bowring, New York, 1962, vol. 1, p. 530).
These are strong words, perhaps too strong, but they speak eloquently of the need for judicial review. Since the lawgiver left no opening for challenging directly the President's exercise of this power, it is well that we do what is next best, and exercise indirect judicial review.
E. The Pardoning Power
(1) The Matter in Issue
5. I now come to the central issue in the petitions before us. This issue has a twofold aspect: first, does the State President have the power to pardon someone before he has been convicted? second, assuming this power to exist, were the conditions for the grant of a pardon to an unconvicted suspect satisfied in the instant case? I am of the opinion that the State President lacks the power to pardon before conviction, and it is therefore unnecessary for me to deal with the latter question concerning the propriety of the President's exercise of his pardoning power.
As regards the first aspect, i.e. the existence of a presidential power of pre-conviction pardoning, the question is by no means an easy one, and has been the subject of keen controversy. In the circumstances, the proper path to have followed seems to be that appointed by the lawgiver, in see. 32(a) of the Consolidated Version of the Courts Law of 1984, namely:
Where a petition for a pardon or for the reduction of a penalty has been submitted to the President of the State, and a question arises which in the opinion of the Minister of Justice deserves to be dealt with by the Supreme Court, but which cannot provide a ground for a retrial under section 9, the Minister of Justice may refer such question to the Supreme Court.
The circumstances were pressing, however, and the opportunity was missed. We have no option, therefore, but to examine the validity of the power within the framework of a petition to the High Court of Justice.
(2) "To Pardon Offenders"
6. In principle, the starting point for our inquiry is sec. 11(b) of the Basic Law: The President of the State (the "Basic Law"), which reads:
The President shall have power to pardon offenders and to lighten penalties by the reduction or commutation thereof.
To ascertain the meaning of the expression "to pardon offenders," we must, as with any other act of interpretation, start with a linguistic inquiry. I believe the question whether the terms "to pardon" and "offenders" in themselves provide an answer to our inquiry, must be answered in the negative. In Israel legislation the term for pardon* does not have one single meaning only. Thus besides its use in sec. 11(b) of the Basic Law, it is also used in Knesset enactments to designate amnesty (thus the General Amnesty Ordinance of 1949, the Amnesty Law of 1967). There is no disputing that the two kinds of pardon differ from each other. The presidential pardon is an individual act, whereas the Knesset amnesty is a general, all-embracing act. The two pardons differ also in their consequences. Despite these differences between the two forms of pardon, the lawgiver has used the same term to describe both. In fact, the term haninah has not acquired any scientific precision or conceptual clarity in Israel, and the term on its own does not enable us to define its meaning. The reasons for this uncertainty - which is not unique to Israel - are hinted at by Dr. Sebba, in these terms (On Pardon and Amnesty, at p. 140):
The lack of clarity on this matter stems from a number of factors, but mainly from a confusion in the choice of terminology, historical changes in the development of these institutions, and a lack of definition of the functions of pardon in its different forms - both as regards their objectives and their legal consequences.
It is quite clear that the term "to pardon" in the Basic Law, relates only to individual pardon. On the other hand, the "pardon" mentioned in sec. 149(9) of the Consolidated Version of the Criminal Procedure Law of 1982, would seem to embrace both individual pardon and general amnesty, but apparently refers mainly to the latter since diffferent situations of individual pardon (in the context of preliminary pleas in a criminal trial) are already covered in sec. 149(5) of the Law, which mentions "a former acquittal or former conviction."
The term "pardon" (haninah) seems, therefore, to have no uniform meaning in Israel law. We have not yet evolved for ourselves an operative jurisprudence the reflective processes of which would generate "jurisprudential" expressions such as "pardon," having a recognized meaning for the entire legal community. Other countries - among them France, Italy and Germany - are more fortunate in this respect, since their own terms for the concept of a pardon granted by the authority at the head of and symbolizing the State (grace, grazia, Begnadigungsrecht), are all self-understood as relating to (individual) pardon after conviction. We have yet to reach such unanimity in Israel, and here, as already indicated, the term haninah encompasses both pardon and amnesty. As regards the question whether an individual pardon - with which alone sec. 11(b) of the Basic Law deals - has any reference to an unconvicted suspect, our own operative jurisprudence offers no answers. That leaves us no alternative but recourse to judicial interpretation, from which there shall evolve, in the course of time, the kind of operative jurisprudence that is responsive to the existing conceptual need.
(3) "Offenders"
7. We have next to examine whether the term "offender" throws any light on our inquiry. Can an unconvicted suspect be deemed an "offender"? This question was discussed by Prof. Klinghoffer, who wrote as follows" ("Lectures on Amnesty," at p. 5):
The Law mentions the power to "pardon offenders." Now it is a cardinal rule in the constitutional law of Israel that a person suspected or accused of a criminal offence is presumed innocent until duly and finally convicted. That means no person is an "offender" until a final convicting judgment has been given against him.
The same approach was adopted by the then Attorney-General, M. Ben Ze'ev, when the Constitution, Law and Justice Committee of the Knesset was considering the proposed Basic Law: The President of the State. He said:
The designated meaning of the word [offender], in my opinion, is someone who has been convicted in a court of law. For if not so, we shall come into conflict with the cardinal rule in our system that a person is presumed innnocent until duly convicted according to law, and anyone might come to the President and say: "I am under suspicion, grant me a pardon" (quoted in the opinion of the Attorney-General, Prof. Zamir, dated 15 June 1985 and appearing in directive no. 21.333 of the Attorney-General's Directives).
I naturally accept that every convicted person is an offender, but it does not follow that someone who has not yet been convicted cannot for the purpose of some particular enactment likewise be deemed an offender (cf. Gold v. Minister of the Interior [27]). Thus, for example, when sec. 3 of the Police Ordinance (New Version) speaks of the employment of the police in "the apprehension and prosecution of offenders...," it is clear from the context that the term "offenders" specifically excludes convicted persons; someone who has already been convicted of a particular offence may not be "apprehended" by the police or "prosecuted" for that same offence. Yet a convicted person is certainly an "offender" for purposes of the Basic Law. In fact, the lawgiver has made a far from precise use of the term, and has not always distinguished clearly between persons suspected, accused, or convicted of a criminal offence - having sometimes included all three possibilities within the purview of this term.
8. The term "offenders" raises further questions about its meaning. It will be found amenable to more than one meaning in the context of sec. 11(b) of the Basic Law. Besides certainly embracing someone who has been duly tried and convicted, in a final judgment (as distinct from the meaning of the same term in the new version of the Police Ordinance), does it also include someone who has been convicted in a judgment that is not yet final? And what is the situation of a person who has not been convicted but in respect of whom the court has held "the charge proved" and issued a probation order under sec. 1 of the Probation Ordinance (New Version) 1969? And in particular, what is the situation of someone who has not yet been charged at all, or who has been charged but whose trial has not yet reached completion? "Offender" is therefore a vague term, ambiguous and open to different interpretations in different contexts.
(4) The Legislative Purpose
9. It is now clear that a linguistic examination of the term "offender" does not suffice to dispose of our interpretative problem - as indeed it rarely should be expected to do (Kibbutz Hatzor v. Rehovot Assessment Officer [28], at p. 74). Among the different possible meanings we should select that which ensures attainment of the legislative purpose - "the Law is an instrument for the achievement of a legislative purpose, and therefore needs to be construed according to its inherent purpose" (per. Sussman J. in Estate Late E. Bergman v. Stossel [29], at p. 516). This purpose can be ascertained, first and foremost, from the intention of the lawgiver. The legislative history of an enactment is a source from which one may ascertain the legislative purpose.
(5) The Intention o1the Legislature
10. In order to ascertain the intention of the Legislature when investing the State President with the power "to pardon offenders," we must return to the Transition Law. It represented the first Israel Law to deal with the presidential powers. In sec. 6 of the Transition Law the Presidential office had been established, inter alia, with the "power to pardon offenders." The objects of this directive were elucidated by Agranat D.P. (as he then was) in the rehearing in Attorney-General v. Matana [3] (at p. 441). He pointed out that as the basis for its debates at the time, the Constitution Committee of the Provisional Council of State relied on the draft constitution of Dr. L. Kohn and a memorandum submitted by E. Vitta. I have carefully considered all this material, from which it clearly transpires that it was not the pardoning powers of the English Monarch, nor those of the American President, the High Commissioner for Palestine, or the Head of any other State, that were envisioned by the draftsmen of the Transition Law as the model for the powers of our own President. Dr. Kohn did not elaborate on the presidential pardoning power, beyond a bold statement (in sec. 59 of his proposal) that the President be reserved the right to grant a pardon. Vitta changed the wording slightly, proposing that the presidential functions include the grant of pardon and the reduction of punishments. Commenting upon Dr. Kohn's proposal, Vitta opined that the presidential power be restricted to individual cases, with a power of general or even partial amnesty entrusted to the Legislature alone, for implementation by way of a formal statute. In a comprehensive debate on the President's proposed status conducted by the abovementioned Constitution Committee, the presidential powers in France, Czechoslovakia and Switzerland were mentioned, slight reference was made to the King of England, while the American President was only hinted at. With regard to the power of pardon, there is recorded only Z. Warhaftig's opinion that the directive be phrased to empower the President "to pardon and reduce punishments" (Proceedings of the Constitution Committee of the Provisional Council of State, Debate on the Executive Authority). The proposal was adopted. In introducing the Bill for the Transition Law, 1949, before the Knesset, Y. Idelson made only a brief statement, and the subsequent debate on the presidential powers was also short. Neither the English King nor the High Commissioner was mentioned in the context of pardon, while the office of the American President was mentioned only as differing from our own form of presidential office. Our survey accordingly leads to a twofold conclusion: first, we lack full information concerning the extent of the pardoning power which the Knesset sought to confer on the President at that time; second, it is clear that the Knesset did not consider imitating any particular model of the power, and certainly not the power of the English King, the High Commissioner or the American President.
11. The provisions of sec. 6 of the Transition Law were repealed with the enactment of the Basic Law. We have no access to the debates of the Knesset Constitution, Law and Justice Committee, which are closed, but I am prepared to accept the following account thereof given by the Attorney-General, Prof. Zamir, who apparently had the opportunity to peruse the minutes of the relevant proceedings (see his abovementioned opinion):
The Legislature's intention may also be gathered from the preparatory stages of the Law. The question before us was not discussed when the Knesset plenum debated the Bill for the Basic Law: The President of the State, but it did arise in a discussion of the Bill at a meeting of the Constitution, Law and Justice Committee (on 5 February, 1964). It appears from the discussion that all the speakers considered the President empowered to grant a pardon to convicted offenders only. The then Attorney-General, Mr. M. Ben Ze'ev, said at that meeting: "The designated meaning of the word [offender], in my opinion, is someone who has been convicted in a court of law. For if not so, we shall come into conflict with the cardinal rule in our system that a person is presumed innocent until duly convicted according to law, and anyone might come to the President and say: 'I am under suspicion, grant me a pardon.' " And Knesset Member, H. Zadok, remarked at the end of the discussion on this point: "It seems to me we have no difference of opinion on the substance of the matter. We intend to empower the President to pardon persons who have been criminally tried and convicted."
This is further evidence that it was not the pardoning powers of the English Monarch, the High Commissioner, or the American President that served as a basis for the above Committee's discussions. On the contrary, the subjective thought of those who dealt with the question was -"We intend to empower the President to pardon persons who have been criminally tried and convicted".
12. Speaking for myself, I would not attribute too much weight to the factor of the legislator's intention in the instant case. The legislative history of the Transition Law offers us scant details and hardly advances our inquiry. As for the Basic Law, we know the opinions of members of the Knesset Committee who dealt with the Bill, but not what the Knesset itself thought. Actually, as faithful interpreters of the law, it is our task to act by way of "analysis of the law and not psychoanalysis of the lawgiver" (Agudat Derekh Eretz v. Broadcast Authority [30], at p. 17). We must not seek to establish a Knesset Member's attitude towards a particular problem confronting us from the legislative history of an enactment. The solution of such problems is our responsibility, and ours alone (FIatto Sharon v. Knesset Committee [31], at p. 41;"Kach"Faction v. Chairman of the Knesset [1], at p. 141). Elsewhere, I have had occasion to comment thus:
The Judge does not seek a concrete answer to the practical problem he has to decide in the history of a legislative enactment. The court is not interested in the specific pictures and concrete likenesses contemplated by the Legislature. In the legislative history of an enactment we seek its purpose; we seek the interests and objectives from which, after compromise and balance between them, there was distilled the policy underlying the norm which is being construed. What we seek is the fundamental perception rather than the individual application - the abstraction, the principle, the policy and purpose. We are interested in the Legislature's concept as to the purpose of the Law, and not in its conception as to the resolution of the specific dispute before the court ("Of Ha-Emek"v. Ramat Yishai Local Council [32], at pp. 143-144) .
We must accordingly continue our search for the legislative purpose behind the statutory provision concerned.
(6) The Legislative Purpose: a "Spacious View"
13. The proper path to follow was indicated by Agranat D.P. (as he then was) in the Matana rehearing [3]. Referring to the Transition Law, which was then in force, Justice Agranat observed (at p. 444):
The "omission" in which my learned colleague found the expression of the desire of the Israel Legislature to cut down the provisions of Art. 16 of the Order in Council and therefore to restrict the President's power of pardon, is in no sense proof of any such intention. It is more correct to say, as was said by Smoira P. in another context... that the Israel Legislature "neither copied nor omitted, but built its law as an independent structure."
The Transition Law was indeed an independent Israel Law, as is the Basic Law which followed it, and the presidential powers conferred thereunder are autonomous and original. The Israel legislator, far from "copying or omitting" anything, fashioned by its own means the constitutional framework for our national life, producing an "independent structure" which must also, therefore, be construed in the same way.
14. We are in fact dealing with an independent Israel Law of constitutional content. This element is of basic importance in the construction of the Law, as was pointed out by Agranat D.P. ([3], p. 442) with reference to the statement of Justice Frankfurter (in Youngstown Sheet and Tube Co. v. Sawyer [51]), that when a matter touched a document which laid down the framework of the government of the State, the court was to take a "spacious view of the powers herein prescribed." I myself followed this approach in the Neiman case [19], where I made these observations (at p. 306):
Basic provisions must be construed according to a "spacious view"- to use an expression of Justice Frankfurter in Youngstown Sheet and Tube Co. v. Sawyer, quoted by Agranat D.P. in the Matana case - and upon the understanding that we are dealing with a directive which determines the national pattern of life. A basic constitutional directive is not to be construed in the same way as an ordinary legislative provision. It was Chief Justice Marshall of America who, in the early stages of the shaping of the American constitutional perspective, stated that in interpreting the Constitution it had to be remembered that it was no ordinary document -"it is a constitution we are expounding" (M'Culloch v. Maryland). We are concerned here with a human endeavour which has to adapt itself to the changing realities of life. If we have said of an ordinary Law that it is not a fortress to be conquered with the aid of a dictionary, but a frame for a living legislative idea (Cr. A. 881, 787/79, at p. 427), how much more should we be so guided when engaging in the interpretation of directives of a constitutional nature.
Constitutional enactments must indeed be interpreted with the structure of the whole system in mind. A Law is "a creature living within its environment" (per Sussman J. in Shalit v. Minister of the Interior [22], at p. 513), and the "environment" of a constitutional Law is, inter alia, the other constitutional enactments which determine the essential character of the regime. Every constitutional enactment is but a building block in the overall structure, which is erected upon given foundations of government and law. Hence, when construing a constitutional enactment, it is the judge-interpreter's function to bring the same "into harmony with the foundations of the existing constitutional regime in the State" (Justice M. Landau, "Rule and Discretion in the Administration of Justice," Mishpatim 1 (1969), 292). That expresses the real importance of Justice Agranat's perception that a "spacious view" must be taken of a constitutional enactment.
15. To take such a "spacious view" when construing the presidential power "to pardon offenders," means to view the presidential powers as part of the general distribution of powers among the State authorities. The presidential power of pardon must be seen as a component in the complex of governmental powers comprising the "constitutional scheme," as was stated by Justice Holmes in Biddle v. Perovich E56] (at p. 486):
A pardon in our days is not a private act of grace from an individual happening to possess power. It is a part of the constitutional scheme.
The pardoning power forms part of the fabric of our democratic life: it flows from the regime's republican system of power allocation. This aspect was elucidated in the American case of Schick v. Reed [58], where it was held per Marshall J., at p. 276):
The references to English statutes and cases are no more than dictum: as the Court itself admonishes, "the power [of pardon] flows from the Constitution alone"... .Accordingly, the primary resource for analyzing the scope of Art. II is our own republic system of government.
16. What conclusions are to be derived from the above mode of interpretation, according to which the presidential pardoning power must be "spaciously" viewed against the background of our own constitutional structure? Two main conclusions seem to be warranted. First, for the purpose of construing the President's pardoning power, we cannot be guided by the powers of pardon conferred on officeholders in other countries whose status, in the devolution of powers in their respective countries, differs materially from that of our own President. We may, however, by the same token, learn about his pardoning power by comparing it with such powers conferred on like officeholders in other countries of similar constitutional structure. The second conclusion is that whatever the scope of similar powers in other countries, we must in the final analysis construe the State President's own power against the domestic constitutional background, and in the end we can gain but limited interpretative guidance from the situation in other countries.
(7) The Legislative Purpose: Guidance from England?
17. Let us examine the power of pardon of the English Monarch. This power had its origin in the seventh century during the reign of the Anglo-Saxon kings (see L. Radzinowicz, A History of English Criminal Law, London, 1948, vol. 1, pp. 107-137). These kings had the power of life and death:
the power to take life included the power to save it... and the pardon power was identical in scope with the power to punish (Boudin, "The Presidential Pardons", at p. 9).
This basic standpoint prevailed for a long period of time. Thus in 1686, an English court held that the Kings of England were absolute sovereigns, that the laws were the Kings' laws and that the King had the power "to dispense with any of laws of the Government as he saw necessity for it" (Godden v. Hales [48], at p. 1051). A number of attempts to curtail the royal power of pardon were made over the years, but in essence it remained as wide as before. Its ideological foundation was the notion that the King was the "fountain of justice." He was the defender of the public and dispenser of justice; he established courts and executed the law, he prosecuted offenders and granted pardons. As Blackstone has commented (Book 1, at pp. 268-269):
As the public, which is the invisible body, has delegated all its power and rights, with regard to the execution of the laws, to one visible magistrate, all affronts to that power, and breaches of those rights, are immediately offences against him, to whom they are so delegated by the public. He is therefore the proper person to prosecute for all public offences and breaches of the peace, being the person injured in the eye of the law... and hence also arises another branch of the prerogative, that of pardoning offences; for it is reasonable that he only who is injured should have the power of forgiving... of prosecutions and pardons.
With the conversion of the English Crown to a constitutional monarchy, the power of pardon itself underwent no real change, although a change did assert itself as regards the exercise of the power. Thus the royal power of general pardon was recognized, but never came to be exercised in practice. The royal power to pardon before conviction has likewise remained recognized, but has not been exercised since the middle of the nineteenth century. At that time this possibility was limited to an immunity from prosecution given to someone who turned "King's evidence."
Nowadays, however, with the development of police powers as well as the Attorney-General's power to stay criminal proceedings, and with the establishment of the office of the Director of Public Prosecutions, the power of pardon is no longer exercised even in the above exceptional circumstances. An English authority has described the situation thus (see J.L. Edwards, The Attorney-General, Politics and the Public Interest, London, 1984, at p. 414):
During the nineteenth century it was common practice to grant a pardon to an accomplice who was to turn Queen's evidence, but this resort to the machinery of pardons, prior to the registering of a conviction, has long since become obsolete. Where the reluctance of a witness to testify on behalf of the Crown did not stem from his being an accomplice but arose on the ground that he would incriminate himself, it was also known for the Crown to prepare a free pardon in advance, ready to be produced by prosecuting counsel. The last occasion when a free pardon was granted to a witness in these circumstances was in 1891. There is now a general understanding among British constitutional law authorities that the practice of conferring a pardon upon a principal offender before conviction has fallen into disuse.
The English King's historical power of pardon is rooted in the royal prerogative, with the King perceived as the source of justice. That perception provides no guidance so far as concerns the President of Israel, as was pointed out by Berinson J. in Matana v. Attorney-General [25] (at pp. 976-977):
I cannot say that the Israel Legislature in conferring the power of pardon upon the President of the State in the Hebrew language, intended to include therein the full content which the concept of pardon has acquired over the ages in English law, pardon which is wholly in the hands of the Crown without reserve or limit by virtue of the ancient royal prerogative. There is no point of comparison between the status of the President in our country and that of the Crown in England. The President is a creature of statute and his powers are defined by law. Like everyone else in this country, he enjoys no rights or privileges which are not accorded to him by the laws of the State and every official act of his which exceeds the limits of the law is null and void.
Unlike the English Monarch, the President of Israel is not "the fountain of justice," he does not execute the law or prosecute public offences. Accordingly, as regards the presidential power "to pardon offenders," no interpretative guidance is to be derived from the pardoning power enjoyed in principle by the English Monarch. But we could certainly be guided as to the scope of the presidential pardoning power by the practice followed by the English Monarch today. And as we have seen, this practice does not extend to a pardon before conviction, since
such practice is out of harmony with modern views as to the propriety of granting dispensation before the normal process of the criminal law has run its course (Edwards, The Attorney-General, p. 475).
(8) The Legislative Purpose: Guidance from the U.S.A.?
18. In the American case of Schick v. Reed [58] Justice Burger relates that when the American Constitution was under preparation, a short discussion took place on the scope of the pardoning power to be entrusted to the President. The view that it be confined to exercise after conviction only, was rejected for the reason that this would preclude the possibility of using accomplices as prosecution witnesses in conformity with the English practice at the time. The American courts have since then interpreted the President's pardoning power as being similar to that of the English Monarch. This power embraces not only individual pardon but also general amnesty, and results from the perception that the President of the U.S.A. is charged with the execution of the laws:
A pardon is an act of grace, proceeding from the power entrusted with the execution of the laws (United States v. Wilson [55], at p. 160, per Marshall C.J.).
The customary view, which is based on dicta in a number of old cases, is that the existing pardoning power extends also to the grant of a pardon before conviction. There seem to be several reasons, however, why even the American model cannot guide us with regard to the presidential power of pardon in Israel. First, in the U.S.A. the President heads the Executive authority. Like the English King in the past, so the American President today is responsible for execution of the law, so that there is a certain logic in entrusting him with a power not to execute the law in certain cases by way of granting a pardon. That is not the situation of the President of Israel, who holds no powers so far as execution of the law is concerned. Second, the U.S. President is empowered to grant a general amnesty, also to unconvicted suspects. There is a certain logic in the contention that the authority competent to grant an amnesty to unconvicted suspects should also be competent to grant an individual pardon before conviction (see P.B. Kurland, Watergate and The Constitution, Chicago, 1978, p. 145). This argument doesn't hold good in Israel, where the President is not empowered to grant an amnesty, and from this viewpoint there is no logical basis for empowering him to grant a pardon before conviction. A third reason for distinguishing the American situation from our own is that the framers of the American Constitution were mindful of the English experience, which they themselves had shared in the colonial period. This was pointed out by Judge Wayne in Ex Parte Wells (1856) [61]:
At the time of the adoption of the Constitution, American statesmen were conversant with the laws of England, and familiar with the prerogatives exercised by the Crown. Hence, when the words "to grant pardons" were used in the Constitution, they convey to the mind the authority as exercised by the English Crown, or by its representatives in the colonies. At that time both Englishmen and Americans attached the same meaning to the word "pardon."
A different situation has existed in Israel. As we have seen, the draftsmen of the Transition Law did not seek to emulate the experience of the English monarchy, and those who legislated the Basic Law into existence did not consider the presidential pardoning power to be exercisable before conviction.
19. The result is that we cannot be guided by the Anglo-American experience when seeking to construe the scope of the President of Israel's power "to pardon offenders." This does not mean that the constitutional situation in England and the U.S.A. cannot ever provide any interpretative guidance for us. On the contrary: our own interpretative processes, ever since the establishment of the State, have drawn extensively on the Anglo-American constitutional experience, and the outlook in these countries on many subjects, among them human rights, have often inspired our own approach. Yet such nourishment has to be controlled, and the inspiration can only flow from a comparison between institutions, processes and perspectives which have a common basis. Thus we too can learn from the American recognition of the fundamental human rights, since both our countries have democratic regimes committed to the rule of law and the separation of powers. That, however, does not apply to the power of pardon, which in England and the U.S.A. is based on an approach that differs entirely from our own.
(9) The Legislative Purpose: Proper Guidance
20. The absolute French monarchy also wielded a wide power of pardon, both before and after conviction, individual as well as general. Like powers were enjoyed by other absolute monarchies in Europe. A drastic change came with the French Revolution, when the existing form of pardon was abolished in France and replaced by a more restricted form. Since then it has been customary in most of the Continental countries for a limited power of pardon - not exercisable before conviction - to be conferred on the titular, and not executive Head of State, that is to say, the person holding the powers which symbolise the State. This is the situation in modern France (see e.g. sec. 35 of the Constitution of the Fourth Republic, and see also Monteil, La grace); in Italy (see art. 87 of the new Constitution, and see also Manzini, Trattato Di Dirrito Penale Italiano, 1981, p. 510; Bortolloti, "Il principio Constituzionale Della Clemenza," Rivista Trim. Di Dir. Civ. [1978], 1681): in Germany (see sec. 60(2) of the new Constitution) and in many other countries (including Holland and the Phillipines). It is interesting to discern the same trend in the new democracies which became a part of the British Commonwealth of Nations. Thus the King of England and the Governor-General of Canada cannot grant a pardon in that country except after conviction (see. 683 of the Criminal Code of 1970). The same holds true in Australia, except in the context of persons who turn Queen's evidence. Section 72 of the Indian Constitution empowers the President to grant a pardon after conviction only (see Balkrishna, "Presidential Power of Pardon," J. of Indian Law Institute 13 [1971], 103). It might also be noted that in a number of countries (among them France, Italy and Germany) the pardoning power is constitutionally defined in general terms, while their equivalents of the expression "to pardon offenders" have been construed, in judicial decisions and by commentators, as relating solely to pardon after conviction.
21. I may now conveniently summarize my observations on the interpretative guidance to be derived from a review of the pardoning methods in other countries. I have sought to show that in countries where the law of pardon is not laden with historical memories from the era of the absolute monarchy or coloured by other similar influences of a bygone era, and at the head of which stands a King or President who symbolises the State, this authority holds a restricted power of pardon. The most important restriction is the limitation of the power to the stage after conviction. This conclusion does not, however, put an end to our interpretative search, for we have seen that constitutional directives require a "spacious view" in their construction. In our present context, that means we have to construe the pardoning power in the light of the general governmental structure in Israel. This I now proceed to do.
(10) The Legislative Purpose: General Governmental Structure
22. During the era of absolute rule, when the power of pardon was wielded by the sovereign himself, there would have been little point in examining the division of authority among the different governmental organs. The ruler held supreme authority, and was therefore entitled to grant a pardon (individual or general) when so disposed, before or after conviction or the conduct of an investigation. It is different in a democratic constitutional regime. The sovereignty there lies with the people, the ruler is no longer omnipotent, and the rule itself is divided among the different authorities. Each has to function within its own sphere, though in general synchronization with the others and subject to mutual checks and balances. It is not in keeping with the democratic character of the regime that any authority, be it the President himself, should hold a paramount power which enables it to change a decision of any of the other authorities which have acted within their responsibility in the framework of criminal proceedings. Such a power may be fitting for an absolute ruler who wishes to show grace to his subjects, but is alien to a holder of high office who wants to serve his subjects. This contrast is well-illustrated in the American precedents. At first it was held by the Supreme Court that a pardon granted by the President, like one granted by the English King, was an act of grace (see U.S. v. Wilson [55], at p. 160) per Marshall C.J.):
A pardon is an act of grace, proceeding from the power entrusted with the execution of the laws....It is the private, though official, act of the executive magistrate, delivered to the individual for whose benefit it is intended, and not communicated officially to the court.
This approach, however, fell into disfavour (see Buchanan, "The Nature of a Pardon under the U. S. Constitution,") and was later expressly rejected in Biddle v. Perovich [56], where Justice Holmes observed as follows (at p. 486):
We will not go into history, but we will say a word about the principles of pardons in the law of the United States. A pardon in our days is not a private act of grace from an individual happening to possess power. It is a part of the constitutional scheme.
The power of pardon of the President of Israel is, indeed, a part of the constitutional scheme, within which it has to find its rightful place.
23. What then are the implications for the pardoning power of the need for its coordination with the other State organs and authorities? In the first place, it seems to me that the President's power of pardon must not be construed as placing him in paramount authority over all the other authorities involved in the administration of criminal justice. The pursuit of criminal justice involves different authorities in different stages, from the commission of the offence until the delivery of a final judgment: the police, the prosecution, the courts, and the prison services (for holding suspects in custody). It would be contradictory to this constitutional arrangement to enable the President to intervene in the normal process by exercising his power of pardon concurrently with the powers exercised by the other State authorities. Only an unworthy constitutional arrangement would permit the President a power to halt a police investigation or the prosecution of a criminal charge, or to intervene at any stage in the course of the adjudicatory process. Such a situation was decried by Landau J. in the Matana rehearing in these terms ([3] at p. 461):
I am unable to see any purpose which can justify such confusion in methods of punishment and the division of powers between the authorities of the State.
The proper interpretative approach indeed requires us to focus attention on the division of powers between the different State authorities, the ramifications of which are to be gathered from the "constitutional scheme" underlying our legislation. The proper construction of the pardoning power against this background is that it should be exercised by the President only after the other authorities have discharged their own functions. If in that situation there be need for a pardon, the President will be empowered to grant it. This was the fundamental philosophical approach to the question of pardon in the U.S.A., as was pointed out by Hamilton (Federalist, no. 74):
The Criminal Act of every country partakes so much of necessary severity, that without an easy access to exception in favor of unfortunate guilt, justice would wear a countenance too sanguinary and cruel.
The same idea was expressed by American Chief Justice Taft, in Ex Parte Grossman [49] (at pp. 120-121):
Executive clemency exists to afford relief from undue harshness or evident mistake in the operation or enforcement of the criminal law. The administration of justice by the courts is not necessarily always wise or certainly considerate of circumstances which may properly mitigate guilt. To afford a remedy, it has always been thought essential in popular governments, as well as in monarchies, to vest in some other authority than the courts power to ameliorate or avoid particular criminal judgments. It is a check entrusted to the Executive for special cases.
That was also the approach of Justice Agranat in A. v. the Law Council [2], where he commented thus (at p. 751):
Here - as in England - the primary purpose... is to redress the wrong done to a person who was convicted while innocent, and the second purpose - the value of which should also not be underestimated -is to reduce the sentence of the offender in circumstances which justify this. It is clear that the exercise of such a power by one of the highest State authorities is essential for the effectiveness of any governmental regime, since in no country whatever has there yet been created a system of justice capable of perfect and unerring operation, and of dispensing justice in every case without fail. I need hardly state that not all the material which may throw light on the matter before the court is invariably produced at the trial, and even the judges, who are but human, may err from time to time. It is essential, therefore, that there be available such a reserve power in order to prevent the harmful consequences of an injustice, and also to enable the reduction of a person's punishment - even when properly convicted - should the circumstances so dictate or warrant .
It accordingly transpires that the power of (individual) pardon is exercisable only upon conviction of the offender in a final judgment. Only then will the different State organs have exhausted their own powers, and only then can there arise the need to correct an injustice. Until that stage, the fate of the suspect is to be determined by the appointed authorities in the ordinary course of the administration of criminal justice. A presidential invasion of the province of any of these authorities is an inconceivable possibility in a democratic-constitutional country.
24. The perception that the presidential power of pardon must not be construed as being in rivalry with the powers of the other State authorities, underlies the interpretative conclusion that the power does not extend to the grant of a general amnesty. The expression "to pardon offenders" - according to its plain meaning, as I have already explained - would seem to embrace also the grant of a general amnesty, since it too effects a pardon for offenders. Yet there is no disputing the view that the President lacks the power to grant a general amnesty, as was in fact held by Agranat D.P. in the Matana rehearing (at p. 455):
It must be understood that when the First Knesset conferred the power of pardon upon the President, its intention was that he should not be empowered to declare a general amnesty, the result of which would be to pardon all offenders, for the intention was to reserve the power of general amnesty to the legislative authority itself.
I believe the rational explanation for the President's lack of the power of amnesty to be the perception that amnesty is a legislative act which is properly the function of the Knesset rather than that of the President. The latter must exercise no powers that impinge on those of the legislative authority or, by analogy, on those of other authorities. The powers of the police to conduct investigations, and of the Attorney-General to prosecute offenders, including their respective powers to discontinue the investigation, or the prosecution, must not be subject to encroachment by the President through exercise of his pardoning power. Neither, indeed, should this power be construed as warranting presidential intervention in the authority of the court to acquit or convict and impose whatever punishment it sees fit. It has to be recognized that the grant of a pardon in the course of the investigation of an offence, or a trial, is an intervention in the exercise of these executive powers just as unwarranted as an intervention by the President in the Knesset's exercise of its legislative powers. The undesirability of construing the pardoning power in a manner permitting such presidential intervention, was stressed by the Attorney-General, Prof. Zamir, in his abovementioned directive, in these terms:
A Presidential power to intervene in criminal proceedings pending before the court, in a manner permitting termination of such proceedings at any time, is undesirable in principle. The pardoning power of the President bears no comparison with the Attorney-General's power to intervene in criminal proceedings by way of staying the same. The Attorney-General functions from the start as an integral factor in criminal court proceedings, for he is empowered by law to prefer the charge on behalf of the State. This power naturally entails also the power to withdraw the charge as well as to stay the criminal proceedings at any stage before final judgment, such decision being founded on a close involvement in and familiarity with the proceedings. The President, on the other hand, is an extraneous factor in criminal proceedings. In this situation, his grant of a pardon in the course of the trial might be seen as an unwarranted intrusion into the domain of the court.
Such an unwarranted intervention would occur if, for instance, the court were to convict the accused and he be pardoned by the President before sentence.
25. This conclusion as to a "separation of powers" between the presidential pardon and the powers of other State authorities, is reinforced when regard is had to Israel's general constitutional framework. The other State authorities (the police, the prosecution and the courts) have the means to establish the facts. The police has its investigating facilities and seeks to reach an assessment of the factual situation. The prosecution, to whom the police must transmit the material, will handle and process it until judgment. The courts possess the institutional and normative facilities for elucidating the question of innocence or guilt. That, however, is not the situation of the President, who has no facilities for ascertaining the truth and testing the facts. It is therefore only natural that in matters of pardon the President be guided by the court's rulings. If it finds the accused not guilty, that is the end of the matter; if it convicts the accused and sentences him, that will be the President's starting point. Before conviction of the accused the President has no factual basis whatever for weighing the justifiability of a pardon. Even an admission by the applicant for a pardon that he committed an offence is of no consequence, for he is presumed innocent until convicted by the court.
26. My approach to the construction of the presidential power of pardon is also dictated by the reality of Israel's own constitutional structure. The President is the "Head of State," and not the head of the executive authority. He is a kind of additional authority to those four already existing (the legislative, executive, judicial and supervisory authorities). In the Israel constitutional context, the President is perceived as symbolizing the State. He is not party to the power struggles in the country, and stands above the day-to-day political strife. It is unreasonable to assume that a President so constituted should be endowed by the lawgiver with a power of intervention in the daily functioning of the remaining governmental authorities. That would be like descending into the "arena," and not fitting for the President. Here, indeed, is a material difference between the President of Israel and the American President. The latter heads the Executive and has to do battle every day. That is why his power of pardon may be construed as extending also to a general amnesty as well as a pardon before conviction of the offender.
27. Moreover, the exercise of power must be subject to judicial review, without which arbitrariness will result, for without the judge there is no law. Hence our constitutional perspective that the activities of each of the governmental authorities are subject to judicial review, consistently with the scope of the powers of that authority. This court has held even the functioning of the Knesset to be subject to limited judicial review (see Bergman v. Minister of Finance [33]; Sarid v. Knesset Chairman [15]; "Kach" Faction v. Knesset Chairman [1]; Kahana v. Knesset Chairman [26]). However, it was seen fit in the Basic Law, see. 13(a), to free the discharge of the presidential function from direct judicial review. There does remain the possibility of indirect review of his decisions, but only in a restricted way. For example, if he pardons a particular person, but refuses to pardon someone else in a similar position, the latter person would have no remedy. It is inconceivable that that same Legislature which freed the presidential functions from judicial review, should have granted him pardoning powers in place of those of other authorities amenable to the jurisdiction of the courts. We should, on the contrary, construe the presidential pardoning power as a residual, or a "reserve" power - as Justice Agranat called it - for use when the powers of the other authorities have reached the limits of their exercise.
28. I have so far assumed that our own "constitutional scheme" requires the pardoning power to be construed as not competing with the powers of other authorities. Hence my conclusion that the President lacks the power to pardon before conviction. It might be argued that the desired result could be achieved by recognizing the President's power to pardon before conviction in the expectation that he would make only a limited use of that power, as is the case in England. There the law has left the constitutional monarch with the pardoning power of the absolute monarch, but ensured that he does not in practice exercise that power except as consistent with the democratic character of the regime. This alternative, attractive as it may seem, is unacceptable to me for a number of reasons .
In the first place, constitutional norms cannot be built on hopes. Basic principles of government are not shaped on the assumption that all will proceed as planned. Quite the contrary. The entire constitutional edifice is testimony to the realization that checks and balances must be provided to prevent, or cope with, situations that are likely to go wrong. If under our "constitutional scheme" the presidential powers must not rival those of other authorities, it would be most undesirable to rest the attainment of this objective on the expectation of presidential restraint, and his refraining from the exercise of his available powers. What if the presidential conduct doesn't come up to expectations? And - should we pursue this approach - why not say that the President has a general power of pardon and amnesty? The fact is that we are dealing here with a matter of constitutional import, impinging as such on our lives within the national framework. When it comes to the shaping of basic principles of government, we have to adopt a clear stand one way or the other. The matter should not be left for resolution on a casuistic basis of distinction between case and case, exceptional or otherwise, that would leave everything exposed to the vagaries of the passing political rivalries. We have been so instructed by my respected colleague, President Shamgar, in Neiman v. Central Knesset Elections Committee [19], where he held as follows (at p. 260):
When constitutional matters are under review, their import and implications must be considered in the long term, and proper weight given to their influences on the political and social frameworks within which they operate. If these be subjugated to the needs of the hour and we adopt a casuistic approach in matters of constitutional content, particularly concerning the freedom and rights of the individual, we shall miss the mark and deal less than justly with the subject.
Were we to resolve the problem by a casuistic determination that the State President has the power, in principle, to pardon before conviction, with everyone left hoping that he will only rarely exercise that power, we should be guilty of doing exactly as admonished not to do.
Secondly, the very existence of a power, albeit slumbering, invariably arouses expectations of its use. The President would be subject to constant individual and public pressure to exercise his power of pardon before conviction, and thrust himself into the centre of public controversy whether he accede to or refuse the request for a pardon. It is precisely the need to regard the President as the symbol of the State and isolate him from political rivalries, that demands a clear and unequivocal determination as to the scope of his pardoning power, and its negation before conviction of the offender in a final judgment.
Thirdly, the indirect - in Israel the only possible - judicial review of the President's discharge of his functions, would be ineffective if he intervened in the activities of the other authorities, for if he acted within his lawful powers, there would be little opportunity for intervention by the court. If we are bent upon a "separation of powers," it is necessary that we keep the powers duly separated.
29. It is accordingly my conclusion that our constitutional framework precludes a construction of the presidential pardoning power as being concurrent with the powers of other State authorities. It demands, in fact, that the power only be exercised after the other authorities have exhausted their own powers. It might be asked whether this approach is not unduly rigid, and whether it may not result in injustice in certain, perhaps exceptional, cases. Can it be said that the other authorities will weigh the same considerations as does the President, and that in a rare case the presidential pardon will not offer the most effective remedy if granted before conviction? These questions are important and must be answered.
Our starting point is that in the overwhelming majority of cases, the different State authorities are sufficiently equipped to deal with the problems with which the power of pre-conviction pardoning is likely to be confronted. True, the mechanisms are not the same, nor is the legal consequence identical, but the main problems are solved. If someone's personal plight (for example, a malignant disease) indicates that he should not be interrogated or prosecuted, a presidential pardon is not the only satisfactory solution. The police and the prosecution may take the suspect's personal circumstances into account and, for example, the Attorney-General may stay the proceedings for lack of public interest in the continuation of the trial on grounds of personal considerations affecting the applicant. Legally speaking, pardon is of course a "stronger" remedy than a stay of proceedings, yet the latter "milder" remedy suffices to solve the problem of the incurably ill suspect, or other problematic cases, by achieving the generally desired effect: The Attorney-General may likewise take into account general social considerations (embracing also matters of security and foreign policy). It was pointed out in the report of the Agranat Jurists' Commission on the Powers of the Attorney-General (1962) that in certain circumstances a situation involving a security, political or public interest may demand that no criminal charge be preferred. In this connection the Commission reported as follows on the Attorney-General's need to consult with the political authorities (at p. 13):
The stated duty to consult arises particularly when criminal proceedings are being instituted in relation to a matter of security, political or public interest. In such event it is always incumbent on the Attorney-General to consider whether the act of instituting criminal proceedings (or halting the same) is not more likely to prejudice the interests of the State than refraining from taking such action. This the Attorney-General will only be able to do after having sought information and guidance from those who carry the primary responsibility for safeguarding the State from the security, political and public aspects - that is to say, from those who, so we must presume, are more experienced and knowledgeable in those fields than we are. As already indicated, he will generally need to refer to the Minister of Justice for the required direction and advice; but sometimes, that is in cases which give rise to questions of "high policy," there will be no alternative but to obtain guidance from the Government as a body.
It will be found that most of the problematical cases for which a pre-conviction pardon is sought, can and should properly be handled through the existing mechanisms, which have been structured in advance to deal with that very kind of case. Every person is presumed innocent until convicted, and a suspect's legal status cannot change except upon conviction by the court. The particular problems, for the suspect himself and for the general public, can normally be satisfactorily handled through the authorities charged with the administration of criminal justice. The former situation of the comparatively poor facilities available for the "extinction" of the offender's criminal past, has now been significantly improved with the enactment of the Crime Register and Rehabilitation of Offenders Law of 1981, according to which a presidential pardon - and clearly only a pardon after conviction is envisioned there - is an extinction of the conviction for all intents and purposes (sec. 16(c).
30. I am ready to acknowledge the possibility of very exceptional circumstances in which the power of pardon before conviction would offer the most practical and effective means of dealing with the problem. Even that, however, would not be a decisive consideration, for the reason that the "spacious view" we have to take when construing a provision of constitutional content, which is the case here, otherwise dictates. We must take into account not only the individual's plight, but also the interests of the general public, and the possibility of the abuse of the power. We must also remember the dynamics of a progression in which the rarest exceptions become less rare, and then become the general rule. The Attorney-General's power to stay proceedings offers an illustration of such a progression. We accordingly have to strike a balance between the different interests - between the hypothetical special exception and the need for the determination of clear and sharply defined limits for the exercise of executive powers at the highest level. Since there is no ideal solution, we can only strive at one that promises the least evil by balancing between the clashing interests. In so doing, in these circumstances, there is no need to give any priority to anomalies and exceptions. The best way to deal with the special cases is through the powers of the authorities who deal regularly with the situations concerned, and not through the conferment of exceptional powers on the State President. That has been the experience of many countries with political regimes resembling our own. Thus the Heads of State in France, Italy, Germany, India, Australia, Canada and in numerous other countries, do not enjoy a power of pardon before conviction. This lack has not, so far as I am aware, led to injustice grave enough in exceptional cases to prompt any move towards amendment of the existing law of pardon. The modern trend seems rather to indicate the contrary, and countries which were formerly subject to the English King's power of pardon have acted to restrict exercise of the pardoning power in their own countries to the post-conviction stage. Edwards, for example, states the situation in these terms (The Attorney-General, at 474):
A review of the independence constitutions within the Commonwealth, negotiated with the United Kingdom Government prior to the transfer of sovereignty, provides substantial support for a pre-conviction limitation of the pardoning power.
And the trend in our own country has been towards refinement of the functioning and facilities of the existing authorities, as witness the provisions of the Crime Register and Rehabilitation of Offenders Law of 1981. Even the most anomalous and exceptional circumstances should not, therefore, be allowed to controvert our fundamental constitutional doctrines.
31. Before concluding this part of my opinion, I wish to refer to a problem connected with the President's post-conviction pardoning power. Does the pardoning power avail, after conviction, in situations where other State authorities have their own powers to deal with the problem? For instance, can a presidential pardon be granted someone whose case is under examination in a retrial, or be granted for reasons of "permanent ill-health" when this question is under examination by the Minister of Police in the framework of his powers under sec. 49(d) of the Penal Law of 1977? These questions do not arise in the matter now before us, and must await elucidation at the opportune time. However, I may point out that the situations described raise a question different from that occupying us in the present petitions. Our concern so far has been whether the term "offender" includes also an unconvicted suspect, the term itself being "open" and amenable to different possibilities, so that we are assigned the task of selection in accordance with the legislative purpose. In the above problematical situations (such as retrial) the accused, who has already been convicted, is by any linguistic test an "offender" and the question is whether such a person can be said to fall outside the ambit of sec. 11(b) of the Basic Law. These situations raise interpretative questions of the greatest complexity. Hence, it may happen that the legislative policy, though its trend be clear, will not be given to implementation in respect of a certain class of "offender" concerning whom the language of the law is insufficiently flexible. The answer, whatever it may be, cannot however affect the construction of the term "offender" in those cases where the language of the Law is flexible enough to serve the legislative policy. It would be unreasonable to hold that since the Legislature failed to attain its objective in some of the possible cases, it then becomes desirable to construe its directives generally in a manner thwarting achievement of the legislative purpose. I believe we should take the opposite interpretative approach. As faithful interpreters, we are committed to attainment of the legislative purpose as far as possible, bearing in mind always that while the interpretation is not bound to the words used, the words do limit the interpretation and so restrict our interpretative freedom. We may implement a legislative purpose within the bounds of a maximum-minimum semantic gradation, but we may not implement a legislative purpose which has no foundation at all in the language of the statute. The point was elucidated by this court in Haddad v. Paz [34], in the following terms (at p. 670):
The legislative purpose constitutes an interpretative aid when it serves as a guide in choosing between different, linguistically permissible, interpretations. It is essential, therefore, that the interpretative option which would implement the legislative purpose, find a receptacle in the language of the Law. There has to be a verbal connection, even a minimal one, between the language and the purpose of a Law. It is necessary to find an Archimedean hold for the legislative purpose in the language of the Law. The interpreter may not implement a purpose that finds no linguistic anchorage in the Law.
We have dealt with the legislative purpose. Linguistically speaking it may be achieved in relation to an unconvicted person. He is not in the category of an "offender." It is doubtful whether this purpose is attainable in relation to a convicted person whose case is under retrial, or under examination by the Minister of Police in the context of his powers in situations of permanent ill-health. These are difficult questions awaiting clarification at the proper time.
32. It may possibly be contended that such linguistic "rigidity" is indicative of a basic misconception as to the legislative purpose. It should perhaps be said that because the President has a power of pardon where there is a retrial, he has the like power where no trial has been held at all. My answer to this legitimate question is that the available evidence - factual (the views of the Knesset members concerned) and legal ("the constitutional scheme") - does not support such a conclusion.
(11) Interim Summary
33. The empowerment of the President "to pardon offenders" is couched in "open", equivocal language, offering in itself no answer to the question whether the power of individual pardon is exercisable also before conviction, or the term "offender" includes also an unconvicted suspect. For the purpose of choosing between the possible linguistic options, we must have recourse to the legislative purpose. That purpose was not the equation of the State President's pardoning power with that of the English King, or of the President of the U.S.A., so we cannot be guided by those models. The Basic Law: The President of the State, is indeed an original Israel Law - the Israel Legislature "neither copied nor omitted, but built its law as an independent structure." It must accordingly be construed against the background of our own national experience, with interpretative guidance sought from the countries which have a similar constitutional arrangement. In discharging the interpretative function we must take a "spacious view," having regard to our "constitutional scheme." Against this background, my own approach is not to presume that the lawgiver sought to confer on the State President - who does not head the executive authority but symbolizes the State, and whose functioning is not subject to direct judicial review - executive powers concurrent with those of other State authorities (the police, the prosecution, the courts). The lawgiver cannot be presumed to have favoured presidential intervention in criminal proceedings before these have run their full course. Therefore, I interpret the expression "to pardon offenders" as extending only to persons against whom a final convicting judgment has been given. This approach finds support in the scholarly treatises of two of Israel's most distinguished jurists, Professors Feller and Klinghoffer, in the criminal law and constitutional law fields respectively. In the view of both scholars, each from the angle of his own specialized field, the presidential power of individual pardon extends only to duly convicted persons (see Prof. Feller's article, "Rehabilitation," p. 5). The same approach was also adopted by Landau J. in the Matana rehearing, where he held as follows (at p. 461):
My main ground in opposing the wide interpretation proposed by the Deputy President is that matters of punishment in criminal cases fall within the jurisdiction of the courts. It is clear that side by side with this jurisdiction the special power of pardon is required in order to correct any serious error of the court which cannot otherwise be corrected, and as an act of grace after the offender has served part of his sentence.
The matters of correcting "any serious error of the court," and "an act of grace after the offender has served part of his sentence," have relevance after conviction only. I am conscious of the fact that Deputy President Agranat (as he then was) and Justice Berinson, two of Israel's most distinguished Judges, expressed a different opinion. I shall seek to explain how this occurred when I examine the approach of my respected colleague, President Shamgar, which I now proceed to do.
F. The Approach of Shamgar P.
(1) The Gist of his Approach
34. I shall seek to set out the main points in the judgment of Shamgar P. on which our approaches diverge. My colleague's starting point seems to be that the Anglo-American model was envisaged by the Israel Legislature as the prototype for the presidential pardoning power in Israel at the time when sec. 6 of the Transition Law was enacted. Further, that it was also so held in the case of A. v. the Law Council and in the Matana rehearing. This "historical-interpretative" approach would dictate the conclusion that the presidential pardoning power in Israel is the same as that of the English King or the American President - and different from other models which may be disregarded - and that it embraces also the grant of a pardon before conviction. In the opinion of Shamgar P., this same situation was envisaged by the Israel Legislature when it later enacted the Basic Law: The President of the State. Hence, so far as pertaining to the issue now before the court, the text of the pardoning provision in the Transition Law was re-enacted without change in the form in which it had been interpreted in the precedents - an indication that no need was seen to change the then existing legal situation. Accordingly, it could not be contended that parallel powers held by other authorities (such as the Attorney-General's power to stay criminal proceedings) might affect the presidential pardoning power, without it first being proved that the existence of such parallel or overlapping powers have implicitly repealed the presidential power. In the opinion of Shamgar P., no such implied repeal could be established in the present matter since the existence of the parallel powers created no conflict. The State President was accordingly competent to grant a pre-conviction pardon and, in the view of Shamgar P., this conclusion was also consistent with the interpretative perspective that constitutional powers must be given an expansive interpretation.
35. Shamgar P. also made reference to other matters in his important judgment, but I believe I have sufficiently stated the essence of his approach. With all due respect, 1 am unable to agree with that approach. Lack of time prevents me from elaborating upon many of the points on which 1 am in disagreement with my respected colleague, and I shall confine my remarks to certain matters which seem to me important for the resolution of the problem before us.
(2) The Anglo-American "Mould and Prototype"
36. As I have already indicated, Justice Shamgar took as the starting point for his construction of the presidential pardoning power, the view that it was structured according to the powers of pardon of the English Monarch and the U.S. President. I have sought to show that this view is not in accord with the facts. It was not the latter powers that the lawgiver had in mind when the power of pardon was enacted under the Transition Law. I have examined the relevant legislative history, without finding any hint of factual support for this theory. The powers of pardon of the English Monarch did not serve as the "prototype" for the pardoning power of the President of Israel. Even Justice Agranat did not find as a fact, in A. v. the Law Council [2], that in enacting sec. 6 of the Transition Law the Israel Legislature had in mind the English Monarch's pardoning power as a matter historically established. Justice Agranat's approach was legal and not historical and he assumed, as a matter of law, that the pardoning power of the President of Israel was the same as that of the High Commissioner in the Mandatory period, which in turn he assumed was the same as that of the English Monarch. On this basis he drew the logical conclusion that the pardoning power of the State President was the same as that of the English Monarch, the learned Justice holding as follows (at pp. 750-751):
I am of the opinion that the power of pardon of the President of Israel is the same, generally speaking, as the power of pardon of the King of England, in its nature and in respect of the consequences which flow from its exercise. Before the enactment of the Transition Law of 1949, the High Commissioner was empowered under Art. 16 of the Palestine Order in Council (inter alia) to "grant to any offender convicted of any crime... within Palestine... a pardon either free or subject to lawful conditions, or any remission of the sentence passed on such offender.... " The power of pardon held by the High Commissioner was accordingly the same as the power of pardon held by the King of England.... If the provisions of sec. 6 of the Transition Law of 1949 were enacted in place of Art. 16 of the Order of the King-in-Council, as I believe happened, then the power of pardon held by the President must be deemed the same as the power formerly possessed by the High Commissioner, and later by the Provisional Government. That is to say, this power is parallel, in its nature and in the consequences which flow from its exercise, to the power of pardon exercised by the King of England.
It is generally accepted today that the above parallel drawn by Justice Agranat in A. v. the Law Council contained two errors, as was indeed pointed out in the dissenting opinion of Landau J. in the Matana rehearing. In the first place, it is clear that the State President's pardoning power is not the same as was the power possessed by the High Commissioner. If these were the same, the question now before us might never have arisen, since except with regard to offenders who turned "King's evidence," the High Commissioner had no power of pre-conviction pardoning. This error was later acknowledged and corrected by Justice Agranat in the Matana rehearing, in the following terms ([3] at pp. 443-444) :
resort to a system of comparison between the language of Art. 16 of the Order in Council and that of sec. 6 of the Transition Law, 1949, in order to ascertain the intention of the Israel Legislature in the latter section which it enacted - resort to this mode of interpretation is out of place... The language of Art. 16 of the Order in Council need not prevent the giving of a wide interpretation to the President's power of pardon.
The second error was Justice Agranat's perception of the High Commissioner's pardoning power as equal to that of the King of England. This aspect too was later dealt with by him in the Matana rehearing, where he pointed out that the High Commissioner held a delegated power which was not the full power of the English King:
the power of pardon granted to the High Commissioner under Art. 16 of the Order in Council... is none other than the power delegated to him by the King from that accorded to the latter by virtue of the prerogative... In view of the rule of construction mentioned above which demands a restricted interpretation of the "delegated" powers of one who has the status of the governor of a British colony, it was imperative to define clearly and precisely the power which was delegated in this respect to the High Commissioner, and it is to this that the detailed and exact language of art. 16 must be attributed. The truth of the matter is that the power of pardon of the King of England was never delegated, in its entire scope, to the High Commissioner. This is proved by the fact that whereas we learn from the passage from Halsbury's Laws quoted above that the King is empowered to grant a pardon also " before conviction," it is provided by Art.16 of the Order in Council that the High Commissioner may only exercise this power in respect of "any offender convicted" (ibid., pp. 439-440).
There accordingly appears to be neither a factual nor any legal basis for the "historical-interpretative" premise that at the root of the State President's pardoning power lay the power of pardon of the English Monarch. Nor, I need hardly add, is there any factual-legal basis for attributing such a role to the pardoning power of the U.S. President.
37. I shall now proceed to analyse the judgment of Agranat D.P. in the Matana rehearing. I have endeavoured above to show that he did not found his decision on any "historical-legal" basis of an Anglo-American "prototype" of the pardoning power. The starting point of his approach was the perception of the Transition Law of 1949 as an original Israel enactment. Agranat D.P. relied in this regard on a dictum of Smoira P., that the Israel Legislature "neither copied nor omitted, but built its law as an independent structure", and went on to add as follows ([3] at p. 444):
I have so far tried to show that the language of Art. 16 of the Order in Council need not prevent the giving of a wide interpretation to the President's power of pardon and that the formulation of this power in sec. 6 of the Transition law "suffers" such a construction. Is there any positive justification for this and how far should the line be stretched? To answer this question we must first consider the nature of the various powers of pardon.
The approach of Agranat D. P. is accordingly to be seen as interpretative rather than historical. Reading the text of sec. 6 of the Transition Law, he examines whether the language "suffers" the construction concerned, and among the possibilities "suffered", chooses the meaning for which there is "positive justification" but taking care not to "stretch the line" too far. He takes into account the fact that he is dealing with a constitutional provision which, he holds, need not be given a restrictive interpretation but calls for the taking of a "spacious view" (ibid. p. 442). It was against this background that Agranat D.P. examined the substance of the pardoning power. Making a thorough examination of the pardoning powers held by the King of England and the American President, he was confronted by the judgment of Berinson J. in the criminal appeal in Matana v. Attorney-General [3], where the latter dwelt on the difference between the English King and the President of Israel. Countering this argument, Agranat D.P. held as follows:
My reply to these words of dissent is twofold. Firstly even if the fact that the local provision is drafted in the Hebrew language must not be lightly disregarded, it would also not be right to give it undue weight. Not only do the expressions "pardon" and "reduction of punishments" have a universal meaning, but the power of pardon, in its scope under the common law, is the power which passed to the Provisional Government by virtue of sec. 14 of the Law and Administration Ordinance, 1948, and was known to local jurists at the time when that provision was framed.
Secondly, the importance of the lesson from American precedent arises from the fact that although it was the clear aim of the draftsmen of the Constitution of the United States (and it was this very object which they wished to achieve) to ensure - by means of the provision imposing upon the President the duty of supervising the faithful implementation of all the laws of the State - that no trace should remain of those prerogative powers which served the King as a means of relaxing the bonds of various laws and statutes, nevertheless they left the power of pardon within the realm of criminal law just as it was, and introduced a specific provision conferring such power upon the President of the State.
And in reply to the contention that the status of the President of Israel, as "symbolizing" the State, differed from that of the American President as "conducting its affairs," Agranat D.P. had this to say (ibid., pp. 453-454):
The reply to this argument is that also in France, where at least until 1958 the status of the President was basically similar to that of the President of Israel, it was found necessary to confer upon the President of the Republic the right to grant pardons.... The result is that the ground of the absence of any similarity or comparison between the status of the President of our country and that of the British Crown (or of the President of the United States) is erroneous.
Justice Agranat accordingly did not construe the Transition Law on the basis that its legislative purpose "was fashioned in the Anglo-American mould, which served as its prototype." His approach was to take a pervasive constitutional perspective, to take a "spacious view" in construing the relevant statutory provision. In ascertaining for himself the meaning of "pardon," he availed himself of the English experience as well as the American, the French and the German. He did not distinguish between the legal situations in the different countries, and he was apparently unaware of the fact that in France the term grace mentioned by him, was not interpreted in the same way as the term "pardon" in the U.S. Constitution. He sought to uncover the essence of the matter, seeing the term "pardon" as harbouring a concept of "universal significance." I wish to adopt that same approach in the matter now before us.
.38. It cannot be overlooked that in his judgment in the Matana rehearing, Agranat D.P. also held, specifically, that the State President has the power to pardon before conviction, and so did Berinson J. As for the import of this determination, I believe it is generally recognized as carrying the weight of an obiter dictum. That would entail a twofold consequence. First, from the viewpoint of our legal system, the abovementioned conclusion constitutes no authoritative declaration that Israel law empowers the President to pardon suspects also before their conviction. Hence a District Court Judge, for instance, would still be free to hold that the presidential power of pardon avails after conviction only. For the same formal reason M. Ben-Ze'ev, a former Attorney-General, felt himself free to declare before the Knesset Constitution, Law and Justice Committee - after the decisions in A. v. The Law Council and in the Matana rehearing - that the presidential pardoning power was exercisable after conviction alone. For this reason too legal scholars have expressed the same opinion. In so doing they were not merely reflecting the desirable state of affairs, but what seemed to them to be the existing legal situation. It is true that "the final form in which the Law is shaped, is the form given it by the Judge" (Justice Sussman, "The Courts and the Legislating Authority," Mishpatim 3 [1971], 213). Also, "Once the Supreme Court has construed a legislative enactment as it did, in a dispute before it, this construction becomes part of that enactment" (Justice S. Agranat, "The Contribution of the Judiciary to the Legislative Endeavour," Iyunei Mishpat 10 (1984), 244). But these statements are true only with regard to the ratio decidendi, and do not apply to obiter dicta. The rational explanation for this is that in his passing remarks the judge does not sense the same responsibility as he does when setting forth the reasons for his decision. Knowing that his remarks in passing have no binding force, he may feel greater freedom in expressing them. It seems to me that our instant matter well illustrates the point. Thus Berinson J.'s statement concerning the presidential power to pardon before conviction does not, I believe, accord with his general line of thinking. Seeking to restrict, he in fact widened the interpretation. His own perspective provided little explanation for that result. As for Justice Agranat, he referred not only to Anglo-American law but also to Continental law, without, however, carefully examining the latter. He was therefore unaware of the fact that the Continental countries generally did not recognize the possibility of individual pardon without prior conviction. He would certainly have made a thorough study of the matter had he considered it central to his decision. Moreover, in substantiating the need for the pardoning power, Justice Agranat set forth the following two reasons alone (in A. v. The Law Council and in the Matana rehearing):
The primary purpose... is to redress the wrong done to a person who was convicted while innocent, and the second purpose - the value of which should also not be underestimated -is to reduce the sentence of the offender in circumstances which justify this. It is clear that the exercise of such a power by one of the highest State authorities is essential for the effectiveness of any governmental regime, since in no country whatever has there yet been created a system of justice capable of perfect and unerring operation, and of dispensing justice in every case without fail (A. v. the Law Council [12], at p. 751).
This reasoning naturally only holds true in relation to a convicted offender. It is not at all applicable to someone who has yet to be convicted. How, then, is this reasoning of Justice Agranat to be reconciled with his view that the President has power to pardon before conviction? Such power would necessitate a different rationalization, of the kind that is not to be found either in A. v. The Law Council or in Matana.
The second implication (of holding a judicial statement to be obiter) is that much significance may nevertheless attach to obiter dicta. If these flow directly from a coherent basic perception, they are capable of heralding accurately the reasons for a decision in the future. As a result these dicta create public expectations which are frequently acted upon. The resulting practice may in turn contribute in the course of time to the adoption of a construction that is in keeping with the original dictum, the expectations thus fulfilling themselves. All of this, of course, will fail to be decisive if a later court holds the dictum to be wrong. But it will all be of great importance if the later court should hesitate between two possible constructions. It is true that as between truth and stability, we should prefer truth, yet sometimes when truth and truth vie with each other - stability is to be preferred (see Of Ha-Emek Cooperative Society v. Ramat Yishai Local Council [32]). These considerations do not apply in the case before us for several reasons. First, because the dicta of Justices Agranat and Berinson were not a necessary concomitant of their basic perspective, and with regard to Justice Agranat I have sought to show that he did not perceive the Anglo-American method of pardon as the "prototype" for our own, but that his underlying approach was to give the expression "pardon" its universal meaning. By such universal standards, the dictum that the State President has the power to pardon before conviction certainly cannot be said to have any compelling foundation. A second reason for not following this dictum is that no constitutional practice actually evolved in its wake. In fact, the contrary appears to be the case, for, by internal directive, requests for a pardon have generally not been acceded to before conviction. In argument before us only a very small number of cases of pardon before conviction could be cited. It seems that the dictum created no expectations which could influence our interpretation.
It accordingly transpires that the question of the presidential power to pardon before conviction has remained essentially unanswered, and we are now called upon to resolve it for the first time. So far the question has been the subject of passing judicial statements, legal articles and jurists' opinions. It is now the time for this court to have its say.
39. My colleague, President Shamgar, relying on the statements of Agranat D.P. in the Matana rehearing, holds that constitutional provisions should be given an expansive interpretation. This is an important determination, since Shamgar P. seeks to give the presidential power of pardon a wide construction. I have two comments in this regard.
First, Justice Agranat's starting point (in the Matana rehearing) was that constitutional directives did not necessarily require a restrictive interpretation, but called for a "spacious view" to be taken. I agree with that approach and it also forms the basis of my own judgment here. In my opinion, however, it does not follow that every directive of constitutional content should be expansively construed. That is unfeasible, since constitutional directives deal in the nature of things with the reciprocal relations between the State authorities, and the occasional expansive construction of a particular authority's powers necessarily entails a narrowing of those of some other authority. Furthermore, an expansive interpretation of a governmental power may often entail a narrowing of basic rights, which too is an inconceivable result. In fact, the question whether the construction should be expansive or restrictive does not determine the mode of interpretation, but is itself the interpretative outcome. Thus Justice Agranat himself held that the presidential power of pardon did not extend to a general amnesty, this conclusion being the result of a narrow construction of the pardoning power. The constitutional proposition, in my opinion, is that constitutional directives must be construed in a manner fitting their preferred standing, and in consonance with their capacity to determine the national pattern of life. A basic provision is not intended to perpetuate an existing situation, but to give direction to human experience. Its construction accordingly calls for a pervasive perception, and not a technical approach.
My second comment is that if called upon to choose between an expansive and a restrictive construction of the presidential pardoning power, I should prefer the latter for several reasons: in the first place, in order to avoid the kind of unwelcome rivalry between the different authorities that I have already described; secondly - and this is the main consideration here - because pardon creates an inequality between "offenders," and a statutory provision relating to pardon must accordingly be given a narrow interpretation. The matter was discussed by Landau J. in Bergman v. Minister of Finance [33], in the following terms (at p. 698):
It is accordingly proper, especially in borderline cases, that where a statutory provision is open to two constructions, we should prefer the construction which upholds the equality of all before the law and does not set it at naught.
This principle has been reiterated by the court on a number of occasions (see Abu Hatzeira v. Attorney-General [35]; Raundanaf (Korn) v. Hakim [36]).
40. Before concluding my remarks on the instant problem, I should state that President Shamgar's basic standpoint that Knesset Members contemplated a particular model of the pardoning power when our own version was enacted, raises many questions in my own mind. Supposing it were to be established that the English or the American model indeed served as the "prototype" for the wording of sec. 6 of the Transition Law, would that require us to construe the provision in accordance with the American tradition? I believe not. A Law, as I have already mentioned, is a creature living within its awn "environment," and the environment of an Israel statute differs from that of an English or American statute, even if they be similarly worded. "The law of a people must be studied in the light of its national way of life" per Agranat J. in "Kol Haam" Ltd. v. Minister of the Interior [37] at p. 884). The judicial discretion in the interpretation of a statute, so Justice Landau has guided us, must be exercised "in order to bring it into harmony with the constitutional regime in existence in the State" (see his abovementioned article in Mishpatim, 1[1969], at p. 306). All does not depend, therefore, on the model or prototype contemplated by the lawgiver when the pardoning provision was enacted and, important though this may be, it is not decisive. We must interpret the law in consonance with our national way of life, and this may change with the passage of time. If so, the interpretation of a Law will undergo a corresponding change. "If times have changed," Justice Sussman wrote, "the Law suffers a sufficiently flexible construction to enable its adaptation to the changes" (see his abovementioned article in Mishpatim 3 [1971], at p. 215). In this regard Agranat J. has commented thus (Kaufman v. Margines [38], at p. 1034):
When the Judge is confronted by a factual situation stemming from new conditions of life rather than those which called forth the existing ruling, it will be the Judge's task to re-examine the logical premise on which the ruling formulated against a different background was based, with a view to adaptation of the same to the new conditions.
This adaptative need applies not only when the facts change, but also when the legal context or "environment" changes. The enactment of new Laws creates a new legal context, and these have the capacity to influence the construction of an earlier statute. It is to be observed that the question is not one of a repeal, expressed or implied, of an earlier Law, but of the effect of the very existence of the new and different Laws on the interpretation of the earlier Law. The point was discussed in State of Israel v Pahima [39], where it was held as follows (at p. 828) :
Sometimes a Law, upon its enactment, presents a number of interpretative options, but with the passage of time arid the enactment of additional Laws on the same subject, some of these options fall away, while others take their place. Apposite here is Justice Sussman's statement that "a term in an enactment is a creature living within its environment" (H.C. 58/58, at 513). This environment includes, besides other directives in that enactment, other statutory enactments which throw light on the interpretation of the Law concerned. It must be observed that here the additional enactments bring about no "legislative" change in the Law, only an "interpretative" change. The new enactments have created a new "environment," which by its very existence influences the manner of interpretation of the Law.
Hence the "prototype" contemplated by the lawgiver when the State was established, for all its importance, cannot in itself determine the contemporary interpretation of the Law. A Law is a dynamic creation, adaptable to changing exigencies. This quality was thus elucidated by Justice Agranat in his abovementioned article (Iyunei Mishpat 10 [1984], at p. 239):
Experience teaches that words have a dynamic life of their own. That is to say, with the changes wrought by time in the conditions of life and the concomitant changes in the different social outlooks, words gradually "shed" their original meaning and "assume" a new significance, or come to harbour additional shades of meaning. This factor may well - though it need not always - bring about a construction of the Law the result of which, although falling within the purview of the Law's general purpose, is not the interpretation contemplated by the lawgiver.
It follows that new legislation (such as, for instance, the Crime Register and Rehabilitation of Offenders Law) enacted after the passing of sec. 6 of the Transition Law and sec. 11(b) of the Basic Law, is able to affect the mode of interpretation of the latter provisions.
(3) The Legislative Authority and its "Acquiescence"
41. The expression "to pardon offenders" was repeated in the Basic Law without change, just as it stood in the Transition Law. From this Shamgar P. infers that the Knesset adopted for itself the construction of the majority as well as the minority opinion of the court, on the question of the pre-conviction pardoning power, in the Matana rehearing. This approach is neither factually nor legally acceptable to me.
42. Factually, the above thesis seems in conflict with the views held by members of the Knesset Constitution, Law and Justice Committee when they discussed sec. 11(b) of the Basic Law prior to its enactment. It appears from the views they expressed, as cited by me above (see par. 11), that they considered the presidential pardoning power to be exercisable after conviction alone. They so understood the words "to pardon offenders," and that was the result they desired. I need only repeat the following statement made by Knesset Member H. Zadok towards the end of the Knesset Committee's discussion of the matter:
It seems to me we have no difference of opinion on the substance of the matter. We intend to empower the President to pardon persons who have been criminally tried and convicted.
Against this background, I do not see how it can be said that the Knesset "rested content" with the pre-conviction pardoning situation as interpreted by the court in the Matana rehearing. The Knesset focused its attention on the problems which inspired the ratio decidendi, and clarified the matter by way of a subsequent amendment to the wording of the Law on the troublesome point then in issue. The Knesset did not address itself at all to the obiter dictum on the matter of pardon before conviction.
43. Legally speaking I am equally unable to accept the aforementioned thesis of Shamgar P. My own approach is that the Knesset legislates only when it actually enacts, and not when it refrains from so doing (see A. Shapira, "The Silence of the Legislature: A Canon of Statutory Construction?," Hapraklit, 21, 293; G. Tedeschi, "Recent Trends in the Theory of 'Stare Decisis'," Hapraklit, 22, 320). The proposition was succinctly stated by Berinson J. as follows (in the Matana rehearing [3], at p. 470):
When have we found that the Legislature is able by silence or inaction to put its seal on a particular course of action of one of the State authorities?
Accordingly, the Knesset's mere repetition in the Basic Law of the wording used in the Transition Law, cannot be said to have put the seal of a binding norm on the above dicta in the Matana rehearing.
(4) Implied Repeal
44. My colleague, President Shamgar; has devoted a considerable part of his judgment to the question of a repeal by implication. In my perception, however, this question fails to arise at all. It is not my view that the powers conferred on the different State authorities (the police, the prosecution, the courts) have repealed by implication the presidential power to pardon before conviction. To have thought so, would necessarily have entailed a recognition of the presidential pardoning power also before conviction. In fact, my approach is that the presidential power of pardon does not avail at all before conviction, so that no question of an implied repeal arises here. In my view, the various Laws dealing with the powers of the different authorities form part of the legal context or "environment," within the framework of which the pardoning power must be construed. These Laws have not implicitly repealed the provisions of sec. 11(b) of the Basic Law, but they do constitute a factor in the interpretation of the Basic Law (see State of Israel v. Pahima [39], at p. 828).
G. On the Rule of Law
45. Before concluding my judgment, I might observe that the petitions before us harbour in the background formal, as well as substantive, questions of the rule of law. In its formal sense the rule of law requires that all persons and bodies in the State - individuals, associations and governmental agencies - act in accordance with the law, and that any act in conflict with the law must be confronted by society's organized sanction. In this sense the rule of law has a twofold meaning: lawful rule and supremacy of the law. This embodies a formal principle, since it is not the content of the law that concerns us here, only the need for it. In this sense the rule of law is unconnected with the nature of the regime, but only with the principle of public order. As far as the executive authority is concerned, the rule of law concerns itself with the legality of the administration. The Executive is subject to the law, and its agencies have no rights, powers or immunities, unless conferred by law. It follows that a State functionary as such holds no greater rights, powers or immunities than does any other person in the State, and is therefore equally answerable for his actions. In this connection I may quote the well-known words of A.V. Dicey:
With us every official, from the Prime Minister down to a constable, or a collector of taxes, is under the same responsibility for every act done without legal justification as any other citizen (The Law of the Constitution, 10th ed., p. 193).
Consequently, if the Attorney-General be of the opinion that the available material provides prima facie justification for the opening of an investigation concerning very serious offences allegedly committed by members of the executive authority, the rule of law will require that the matter be examined and investigated. That is how we should treat anyone else, and State officials should be treated no differently. Security considerations dictate no contrary result, for there is no security without law, and the rule of law is a component of national security. Security needs dictate that the proper investigative machinery be found, or else the General Security Service will be unable to fulfill its task. The strength of the Service lies in the public confidence it enjoys, in the trust placed in it by the court. If security interests become the paramount consideration, the public as well as the court will lose their trust in the Security Service and in the legality of its operations. Without trust, the State authorities cannot function. That is the case with the public trust in the courts (see Tzaban v. Minister of Religious Affairs [40]), and so it is with the public trust in the other governmental organs .
46. The rule of law carries, in addition to its formal attributes, also a substantive significance, namely: rule of the appropriate law, law which displays a balance between individual and the public needs. The primary implication thereof is the equality principle, equality in the application of the law and its use. The rule of law is negated where there is discrimination between equals. The matter was discussed by this court in Neiman v. Knesset Central Elections Committee [19], where Shamgar P. made these observations (at pp. 261-262):
The rule of law finds its main expression in the fact that it is not the rule of persons - according to their own unfettered decisions, considerations and desires - but is founded upon stable normative directives which are equal for all and bind everyone in equal measure. The manner of definition of a right and even its recital in the Law do not in themselves constitute an effective safeguard, for these do not secure full realization of the right. Rights are practically realized when they are respected by applying them equally in practice, without unjust discrimination. The value and potency of a Law which confers rights lie in the facts that the rights thus conferred do not remain in the realm of an abstract idea, however lofty in spirit and trend, that also the letter of the Law comes down to what is concrete and available, that it is applied according to standards of an equality among equals, from which there be no deviation for improper reasons.
The subjection of one person to an investigation, but not another who is in an equal situation, is an impairment of the rule of law, just as it is to grant one person a pardon but not another in equal circumstances, or to afford one person every opportunity of defending himself and stating his version of events whilst withholding the same benefits from someone else with an equal claim thereto.
47. Historians tells us that Chief Justice Coke, when he was unable to dissuade King James I from asserting authority in the judicial sphere, addressed these memorable words to the King:
Quod rex non debet sub homine, sed sub deo et lege (the King is subject not to men, but to God and the law).
So be it.
The petitions concerning the investigation dismissed by unanimous opinion; the petitions concerning the pardons dismissed by majority opinion. The orders discharged.
Judgment given on 6 August 1986.
* In Hebrew - haninah, חנינה-Translator's note