Gtedeschi - Prospective Revision Of Precedent.pdf

+(,121/,1( Citation: G. Tedeschi, Prospective Revision of Precedent, 8 Isr. L. Rev. 173 (1973) Content downloaded/printed from HeinOnline Fri Nov 23 15:10:25 2018 -- Your use of this HeinOnline PDF indicates your acceptance of HeinOnline's Terms and Conditions of the license agreement available at https://heinonline.org/HOL/License -- The search text of this PDF is generated from uncorrected OCR text. -- To obtain permission to use this article beyond the scope of your HeinOnline license, please use: Copyright Information Use QR Code reader to send PDF to your smartphone or tablet device

ISRAEL LAW REVIEW Vol. 8

April, 1973

No. 2

PROSPECTIVE REVISION OF PRECEDENT By G. Tedeschi*

The Problem 1. Innovations in case law have traditionally always had retroactive effect also, first on the case in question, and therefore on past conduct, and then, to the extent that authority attaches to the precedent or it is at least followed in practice, on other cases subsequently heard, although the conduct involved here also preceded the declaration of the new rule. Justice Holmes, in a dissenting opinion of 1910, wrote: "I know of no authority in this case to say that in general state decisions shall make law only for the future. Judicial decisions have had retrospective operation for near a thousand years".: It is fairly certain that Justice Holmes had the Common Law world in mind, but the picture would not be different if we looked at the situation elsewhere. In France as well, for instance, the ritroactivitd des nouvelles 2 jurisprudenceshas been considered as une infirmiti du syst~me jurisprudentiel. Nevertheless, it is a fact that the situation excites some consternation among lawyers and the public, which is particularly strong in those matters where retroactivity is singularly severe in its effects and the legislature normally refrains from imposing it, although not forbidden to do so under constitutional principles (laws being interpreted and applied non-retroactively in the absence of any express provision in this regard) and in those instances where there was reliance, explicit or presumed, by the parties on the rule which is subsequently abandoned. Because of this disturbing feeling, case law, it would appear, endeavours to abstain from new departures which it might otherwise embark upon, at times to the extent of forbidding innovation by adopting a rule of unchangeable precedent-which is one way, even if not the most desirable, of radically solving the problem. Indeed it cannot be said that the reliance placed by the parties on the old rule is properly the sole ground * Professor of Civil Law, Hebrew University of Jerusalem. 1 2

Kuhn v. Fairmont Coal Co. 215 US 349, 372 (1910). Roubier, Le droit transitoire (Conflits des lois dans le temps) (2nd ed., 1960) no. 56, p. 248. Cf. G6ny, Methode d'interpritation (1899) 430 et seq.; Lambert, Etudes de droit commun ligislatif, vol. I, p. 20.

ISRAEL LAW REVIEW

[Is.L.R. Vol. 8

for a correct solution of the matter. The new rule may possibly be so much more just than the old that it warrants ignoring all reliance upon the previous rule.3 But the reverse is also conceivable, and that is the assumption which the legislature regards as prevailing when it does not order its innovations to be retroactive. Again it is a fact that, as against the "thousand years" of Justice Holmes, it is more than one hundred years since a movement began in the United States to correct or confine the drawbacks which may arise from implementing changes in case law by limiting as far as can be the sacrifice of public anticipations. First there were isolated judgments which proceeded upon the special circumstances of the cases concerned, without the judges directing their minds to the general problem. Then came a maturer awareness of the new orientation in the emergence of which outstanding lawyers, among them Wigmore and Cardozo, participated. This movement manifested itself in particular in the matters already suggested above 4 and where there was occasion for assuming that the parties had indeed relied upon the old rule.5 In other countries lawyers did not remain uninfluenced by this American trend. As late as 1966, W. Friedmann wrote that prospective overruling On the consideration whether the innovation will achieve its deserved result without retroactive application (when, for instance, deterrence is the main object), see inter alios S.R. Shapiro, "Prospective or Retroactive Operation of Overruling Decision" 10 A.L.R. 3rd, 1371. 4 These matters are primarily title to property, questions of personal status involving the family and commercial transactions. See the analysis of the cases in C.M.D., "Limited Retroactivity of Overruling Decisions" (1938) 25 Va.L.R. 210; Note, (1947) 60 Harv. L. R. 437; Note (1965) 51 Va. L. R. 201. There is also agreement that an adverse change with regard to criminal responsibility should not be applied retrospectively. Taxation matters have in this regard attracted special attention in W. Germany. See Grunsky, Grenzen der Riickwirkung bei einer Aenderung der Rechtsprechung (Karlsruhe, 1970) 6. (Abridged English translation: "Limitations to Retroaction in Cases of Amendments to Precedent" 3 Law and State (Tilbingen, 1971) 73). In America on the other hand the trend with regard to taxation matters is not uniform: Note (1947) 60 Harv. L. R. 437, 444 et seq.; Shapiro, op. cit., 1437 et seq. Compare the list of matters given in the Practice Statement (Judicial Precedent) of the House of Lords in 1966 (Hansard Report 677 (July 26, 1966); [1966] 1 Weekly L.R. 1234 (H.L.)): "[their Lordships] will bear in mind the danger of disturbing retrospectively the basis on which contracts, settlements of property and fiscal arrangements have been entered into and also the special need for certainty as to the criminal law". In American law the problem of non-retroactivity arises also in connection with the abolition of immunity in torts (in order to enable the person responsible to insure himself). 5 As to whether the question of "reliance" is to be dealt with on the basis of proof or assumption, see inter alia Note (1947) 60 Harv. L.R. 437, 440; C.M.D., op. cit., pp. 213 et seq.; Note (1961) 71 Yale L.J. 907, 945-948; Note (1966) 51 Minn. L. R. 79, 80, 82. Cf. Grunsky, op. cit., pp. 26, 27 (English abridgement, pp. 88-89). 3

No. 2, 1973]

PROSPECTIVE REVISION OF PRECEDENT

175

"appears to be virtually unknown outside the United States",' but in recent years there also, lawyers have begun to deal with the question which they had previously ignored, regarding it as an irreparable defect and consoling themselves by such varying considerations as the infrequency of the cases in which the parties truly rely on earlier case law or the infrequency of situations in which the changes could not have been foreseen. Thus, in contrast to the eleventh edition of Salmond on Jurisprudence 7 where it is said that "it is true that [case law] is invariably retrospective", the twelfth edition expresses a different view: "it is by no means true that it must be retrospective".' Likewise Diplock L. J. echoes: "the rule that a new precedent applies to acts done before it was laid down is not an essential feature of the judicial process"? So also with regard to the Practice Statement (Judicial Precedent) issued by Lord Gardener L. C. in 1966 and the reservation therein with regard to certain matters, it has been asked whether the intention was not that "the Law Lords were prepared to overrule with prospective application only"? ° In Western Germany interest in the American trend to prospective revision has been considerable, and there also the tendency to give prior notice of the innovations which the judges have in mind to introduce in the case law has been marked.'1 In France, Roubier advanced the idea "s'il ne serait pas possible de privoir, et de rigler, les conflits entre des jurisprudences successives",12 an idea which raises doubts whether indeed that irremediable "infirmitg du systme jurisprudentiel" exists, about which, as we have seen, he speaks elsewhere. 6 W. Friedmann, "Limits of Judicial Lawmaking and Prospective Overruling"

(1966)

29 Mod. L. R. 593. 7 Salmond, Jurisprudence (11th ed., 1957) 148. Cf. Cross, Precedent in English Law (2nd ed., 1968) 29: "retrospective judicial legislation must surely be a necessary evil". 8 Salmond, Jurisprudence (12th ed., 1966) 127-28. It may be observed that G. Williams, who edited the l1th edition of Salmond, addressed himself later to the prospective innovations in American case law in his treatise, The Criminal Law (2nd ed., 1961) para. 106. 9 Diplock, The Courts as Legislators (1965) 17. Cf. Dias, "Precedents in the House of Lords" (1955) Camb. L.J. 153, 155. And see infra text and n. 26. 10 Leach, "Revisionism in the House of Lords" (1967) 80 Harv. L. R. 797, 802. 11

Cf. Grunsky, op. cit.; Knittel, Zum Problem der Riickwirkung bei einer Aenderung der Rechtsprechung-Untersuchung zum deutschen und US-amerikanischen Recht (1965); Kisker, Die Riickwirkung von Gesetzen (1963) 120-134; Friedrich, "Die

Riickwirkung im Steuerrecht" (1951) Stu. W. Sp. 81, 116; Vogel, "Zur Frage der Riickwirkung von Steuergesetzen" (1960) NIW 1182; Weisensee, Riickwirkende Anwendung von Entscheidungen des Bundesfinanzhof?" (1961) Stu. W. Sp. 17; Miiller, "Zur Bindung des Bundesfinanzhofs an seine Rechtsprechung" (1961) Stu. W.

12

Sp. 683; Zweigert/Kbtz, "Die Bedeutung des Riickwirkungsverbot fUr die Auslegung von Steuergesetzen" (1969) BB 453. Op. cit., supra n. 2, no. 7, p. 24 et seq.

ISRAEL LAW REVIEW

[Is.LR. Vol. 8

Subsequently, however, he confronts the suggestion with the existence of "un obstacle insurmontable dans notre droit: c'est l'impossibilitj oit Von est de diterminer, avec une precision suffisante, le moment oii la jurisprudence est fixie"; since in France what is determinative is not the isolated precedent but a series of uniform precedents. This objection does not appear unanswerable because in any event every series of precedents must commence with an initial precedent (although that does not attain decisive authority) with regard to which the question may be asked whether it is proper to give effect to the innovation by applying it retrospectively or to limit its retroactivity or even avoid it by prior public announcement of the new trend which the judges have in mind to introduce. At all events it is clear that in many respects the question varies with different countries. The intention here is to consider the ideas that have been expressed about this tendency in as far as these are of interest for Israel law and to suggest the position that may be taken up in this area of our law. 2. Theoretically indeed the problem of retroactivity finds no place when the very idea of new departures in case law is denied. If the legal system is assumed to be complete, it embraces all answers, at least by implication. It follows that when the solution of a question is given in a judgment, it is not deemed to have retrospective application since it is regarded as having already existed in the legal system before its actual formulation and revelation. However, the premiss of such reasoning now appears to many to be arbitrary and fictitious. Even were it allowed that the apparent innovation need only be regarded as corrective of errors previously made by case law or by prevailing professional opinion, that would not affect the state of mind of the public, at least where the change in question was not foreseeable. Theoretically, again, the problem is narrowed down when in a particular legal system and in respect of particular instances the precedent is regarded as being not open to correction. The question of retrospective application does not arise within those areas in which the very possibility of innovation is denied. But whatever the accepted theory may be, there is no doubt that in practice precedents are on occasion corrected; and the surreptitious manner in which under these circumstances the change is made cannot reduce the difficulties created for those who have acted until then in reliance on the apparent legal position. On the contrary the theory itself that precedent cannot be corrected will indeed strengthen the reliance which people place upon it and thus aggravate their disappointment when a precedent is in fact set aside or modified. Finally, in theory the problem is narrowed down in another aspect when the authority of precedents is in general denied. To the extent that a precedent -in contrast to previous case law or communis opinio-is innovative it will constitute a retrospective application of the new rule in respect of the case concerned, but that does not mean that the innovation need also apply to other

No. 2, 1973]

PROSPECTIVE REVISION OF PRECEDENT

cases, either past (those involving dealings that have taken place previously) or future. This assumption, however, is likely to hold more in theory than in practice. Even when a precedent is not binding, it normally awakens in practice some expectations that it will be followed in the future. That is especially so in those legal systems where rules intended to assure uniformity of decision exist and therefore a judgment cannot be said to be exhaustive of all its aspects material to the actual litigants alone, and all other persons concerned with its ratio decidendi are simply "overhearing" things not intended for them. At all events, Israel law generally recognizes the authority of precedents either as binding or as "guiding". Even if the provisions of sec. 3313 of the Courts Law, 1957 are formally directed to the judges, the result is that the precedents dealt with there come to be norms from which the individual may reasonably derive future expectations. On the other hand there are no precedents in Israel law from which one may not deviate. It follows that in the Israel system the problem presents itself more fully and more clearly than it does in other legal systems where one of these two conditions are absent. It is true that the fact that every precedent is open to change may moderate or lessen the legal value of the expectation resting upon it. The private individual must know that any given precedent, even of the Supreme Court, does not constitute a basis upon which he can rely as though it were enacted law. Although both precedent and statute may change, a new precedent which diverges from the old operates retroactively (within the limits of res judicata)-- either on its own alone or alongside the old precedent, as the case may be-whereas the amendment of a statute is normally not retrospective. That, however, is no reason for not supporting the aspiration that decisional innovations also should as far as possible be effected without sacrificing any victims, without disappointing private expectations. "Prospective Overruling" 3. American jurists who are protagonists of the new trend have repeated again and again that it could not have been contemplated so long as there prevailed the declaratory theory of which Blackstone was the best known and most widely cited spokesman. They add, however, that if we abandon the fiction that the judges are confined to finding the law and acknowledge that frequently they need to create it, no difficulty arises in permitting "prospective overruling"" 4 as it is commonly termed. English lawyers and others would appear to agree in putting the question in this form.15 13 Sec. 33 reads as follows: a) A court shall be guided by a precedent established by a higher court. b) A precedent established by the Supreme Court binds every Court except the Supreme Court. 14 Cf., e.g., B.H. Levy, "Realist Jurisprudence and Prospective Overruling" (1960) 109 U.Pa.L.R. 1; Carpenter, "Court Decisions and Common Law" (1917) 17 Col. L.R. 593, 604; Note (1939) 24 Corn. L.R. 611, 612. 15 Cf. e.g., Diplock, loc. cit.; Dias, loc. cit.; A. et S. Tunc, Le droit des Etats-Unis

ISRAEL LAW REVIEW

[IsL.R. Vol. 8

Yet to speak of "prospective overruling" is like speaking of squaring the circle. "Overruling"-that is, the adoption of a new ratio decidendi in place of another that that was adopted in an earlier judgment-cannot be prospective only; it must form the foundation of a judgment which necessarily concerns certain past conduct. And if it is merely prospective-a "prophecy", in the words of Cardozo, of what the courts will decide for the future-it can only be prospective to a future overruling and not an actual one, according to what was said above regarding the nature of the latter. Even when we entirely reject the "declaratory theory", the difficulty inherent in the rule of precedent remains."' The supporters of this movement, or some of them, endeavour to distinguish between the adjudicatory and the rule-making functions of the judge." But this distinction cannot be established within the framework of the rule of precedent, simply because, despite their declared aspiration, rule-making in the theory of precedent is a by-product of adjudication. That is the difference between law (or custom) and precedent: the former obtains even without application in judical proceedings, the latter does not exist unless it operates as the ratio decidendi of a judgment. The rule-making of the judge author of the precedent is paradigmatic: what is authoritative is the example given by his decision as such and not any possible formulation of the ratio of his decision. The distinction can only be drawn if judicial rule-making occurs outside the framework of judicial decision and therefore of precedent. That was the case, for instance, with the arrits de r~glement of the French parlements in the past, which only bore a judicial form, and were normally not issued in the course of judicial proceedings, and of the regulae praxis issued by the English courts, as well as by the courts of Palestine during the Mandatory period. Many more examples may be given of instances in which the courts have fulfilled and continue to fulfil general normative tasks which are unconnected with their judicial function or connected with it but always external thereto. 8 Kocourek in 1931 proposed that the Supreme Court should be allowed to issue declarations of rules for future application even when such rules are not embodied in the decision of the court in an actual case (and that such declarations should be regarded as evidence for future cases of the existence d'Amirique, Sources et Techniques (Paris, 1955) 220. W. Bigiavi, Appunti sul dirittogiudiziario, (Urbino, 1933) (Reprint from (1933) VII Studi Urbinati) 89 et passim. 17 See in particular Levy, op. cit., at p. 3 and Carpenter, op. cit. at p. 604, as well as the observations of the Hon. George F. Campbell who was the first to formulate this new trend in his address of 1917: see Rep. S. C. Bar Ass'n, 1917, p. 17, cited also in R.H. Freeman, "The Protection Afforded Against the Retroactive Operation of Overruling Decisions" (1918) 18 Col. L. R. 230, 248. Is We have dealt with a number of these examples in Studies in Israel Law (Jerusalem, 1960) 116 et seq. 16

No. 2, 1973]

PROSPECTIVE REVISION OF PRECEDENT

179

of reasonable reliance for the purpose of the application of these rules in subsequent matters)."9 The declaration would only be made in connection with a question raised on the facts presented to and found by the court. Thus, although we still remain near the area of precedent, on the margins thereof, in truth we are already outside it since the declaration is not embodied in an actual decision. In any case, Kocourek's proposal was merely de lege ferenda and in fact was impliedly abandoned by him later on. 20 Among the supporters of prospective overruling generally, and in American practice which is influenced by it, no such revolutionary ideas are to be found and general declarations which are 21 not applied in a concrete case are regarded simply as obiter dicta. Indeed, ordinarily the appellation obiter dicta is given to those rules which are pronounced in a judgment but do not affect the case in hand on the facts as found by the judge-rules which perhaps were not even discussed in the argument of counsel-that is, an abstract hypothesis differing from that to which regard is had in the decision itself. Here, on the other hand, the situation is otherwise. If the court does not limit itself in its declaration that it intends in the future to depart from the course which it has up to then pursued but goes on to formulate at once the new rule which it is going to adopt in future cases, it is certainly possible that the new rule was discussed in the instant proceedings.22 Moreover, since the new rule has been considered as an alternative solution to the matter which has been determined, it cannot be said to be unconnected with the concrete facts as found by the judges. Yet the new rule is not the basis for any actual decision for which the judge assumes judicial responsibility and it may therefore not be regarded as a precedent. If obiter dicta is a negative concept (everything that does not form part of the ratio decidendi), it is rightly so called. 4. Thus we find that the "prospective overruling" of American theory and practice is not what it says it is. One of its two components is taken in a special sense, quite different from the ordinary meaning of the word. Either it is truly overruling-and then it is retrospective with regard to the case affected by the new ratio decidendi (even if it is stated in the judgment that this ratio shall not apply to other conduct that has already occurred)-or it is simply prospective and then it is not truly overruling. "Retrospective Decisions and Stare Decisis: a Proposal"

19

Kocourek,

20

A.B.A.J. 180. Kocourek and Koven, "Renovation of the Common Law through Stare Decisis" (1935)

21

22

(1931)

17

29 II.L.R. 971.

See as against this Lupoi, Valore dichiarativo della sentenza ed irretroattiviti del mutamento giurisprudenziale nel diritto statunitense (estratto da Quaderni del "Foro Italiano" Roma, 1969) 4-5, who deems such a declaration as true overruling to which lower courts would be subject. Cf. on this point Mishkin and Morris, On Law in Courts (1965) 308.

ISRAEL LAW REVIEW

[Is.L.R. Vol. 8

The first of these possibilities can be termed retroactivity of limited scope. It will arise, as has been said, when the new rule is applied to the matter in hand whilst the judgment contains a declaration that it shall not apply to any matter or to a particular group of like matters that preceded the decision.2 " The judges here may be said to have a mind to accord a sort of immunity to certain matters against the new precedent so that they are not subject to it. The value to be attached to such a declaration will not be discussed here. It has already been observed that it is no more than obiter dictum, that is, it does not achieve what is sought. In fact it happened that the judges who themselves have made such a declaration have not abided by it when other cases have come before them. In any event, even apart from the possibility of the authors not observing their declaration, a similar result-wholly or partially-occurs when, although the new precedent is unaccompanied by any such restriction, judges subsequently dealing with "old" cases of a similar kind are regarded as free not to follow, and in fact do not follow, in its footsteps or do not do so when in the case before them there was no reliance (proved or presumed) of the parties upon the old rule ("selective retroactivity"). The alternative to retroactivity of limited scope has been called in the United States "prophecy"; we would prefer to define it "a declaration of future departure from precedent" or briefly "a disjunctive declaration". That occurs when, in the course of their judgments according to old precedents, judges state or make it understood that they will not continue along the old path. Rather than a precise design, there is here a series of possibilities which has been suggested and in part realized in the United States and Germany. This technique is somewhat similar to that adopted by the Canonists with regard to custom. When custom conflicts, not with Divine law but solely with the law of the Church, its invalidation is not per modum declarationis, by virtue of which it is considered void ab initio, but per modum dispositionis, by virtue of which the particular custom, although acknowledged as having been operative in the past, is declared to be without effect as regards the future: 24 Ipsam [consuetudinem] mandamus de cetero non servari. The most moderate form of a "disjunctive declaration" is the expression of doubt in a judgment whether the decision actually taken is rightly to be followed in the future. The significance of this expression of doubt is in the nature of a caveat to those who may thereafter wish to rely on the decision. 23

24

See e.g. in American case law what was said in Meyers v. Drozda 34 U.S.L. Week 2594 (Neb. Sup. Ct. April 22, 1966): "the new rule applies to all causes of action arising after April 22, 1966, the filing date of this opinion. In respect to other causes of action the new rule applies if, but only if, the nonprofit charitable hospital was insured against liability on the claim of the patient, and then only to the extent of the maximum applicable amount of its insurance coverage". Trummer, Die Gewohnheit als kirchliche Rechtsqueile, Ein Beitrag zur Erkliirung des Codex Juris Canonici (Wien, 1932) 56 et seq.; Bigiavi, op. cit., pp. 88-89.

No. 2, 1973]

PROSPECTIVE REVISION OF PRECEDENT

181

Possibly also it constitutes an invitation to jurists to go into the problem for the purpose of working out a different solution which it might be proper to adopt in the future. 2 At a slightly higher level, we find the court (or one or other of its judges), instead of expressing doubt, declaring that in its view the line which the judgment in fact takes should be abandoned for the future; thus the caveat is rendered clearer and more forceful, even though the new line which the judges have a mind to adopt is not indicated or not precisely indicated. A still higher level is the unambiguous declaration of the new direction which it is henceforth intended to follow (something indeed which is not required when the problem is such that only one solution presents itself with the rejection of a given alternative). Similarly in England, the judges sometimes also adopt measures to express their view as to the future. W. Friedmann observes that in Hadley Byrne 26 "the House of Lords could have been content to dismiss the action on the ground that the defendants had excluded any legal responsibility for their statement. It chose instead to enunciate [ ...] a future principle for responsibility [... ].The House of Lords overruled Candler v. Crane, Christmas & Co. 27 with respect to future situations. [... ]. The decision [... ] operates in effect as a 'prospective overruling' ",.2s Finally the declaration may take a more manifestly normative form, when a particular date or period is laid down (such as the end of the next session of the legislature) for the commence29 ment of the new rule. 5. There is another reason for preferring the expression "prospective revision" to "prospective overruling". Overruling imports the abrogation of the ratio decidendi of an earlier judgment. But the question before us is posed even when the court adopting the new line is of such a rank that it may not abrogate the old rule (a precedent of an equally ranking court) or when it is doubtful whether it may do so (a precedent of its own-see infra section 12). The question may arise, in these cases also, both as a question of the most suitable technique to be adopted by the judge initiating the new line and as a question of the choice by subsequent judges between the precedent which it is to be assumed the parties relied upon and some other precedent 25

For Germany, see Grunsky, op. cit., p. 24 et seq. (English abridgement, p. 87

et seq.) 26 27 28 29

[1966] A.C. 465. [1951] 2 K.B. 164. W. Friedman, op. cit., p. 605. In theory the determining date-the date when judgment was issued or some other date mentioned therein--can relate to understands the American decisions and is retroactive as regards cases occurring was entered thereafter: see Roubier, op. courts, however, to relate to the entry the case.

the entry of the summons. Roubier thus he observes that in this way the new rule before the judgment, even if the summons cit., p. 29. It is not the habit of American of the summons but to the occurrence of

ISRAEL LAW REVIEW

[Is.L.R. Vol. 8

which emerged after they had acted as they did and was innovatory, although the latter did not confine its application to future cases alone. The second situation may appear to be not comprehended in the expression "prospective revision", but it can be, if we understand by that not only a revision which is expressly stated to be such, wholly or partially, but also one which can be so regarded because of the use made of it afterwards. 6. The expression "prospective overruling" has met with other criticism. Professor Mishkin takes the view that "[a] power of prospective limitation cannot be confined to cases involving overruling of precedent ... .]. For the logic of the position extends to any situation in which the new decision produces a substantial change from what had previously been considered to be the lawwhether or not specific authority could be cited for that previous view". 30 From this point of view the term "prospective revision" should also be ruled out, but it is not easy to find a short appellation that fits the subject. Mishkin's "prospective limitation" is incomplete and if we are to understand it as "prospective limitation of precedent", it does not appear apt to embrace "prophecy" as well, since the latter is not a precedent, not even a limited precedent. We should say "prospective innovations of case law". Whether the cases mentioned by Mishkin should be taken into account is another question. Although cases such as these have been adjudicated in America, 31 it is not very reasonable to avoid giving retroactive effect to a decision when the reliance of the litigants upon a different legal situation did not rest upon any authoritative foundations. It may be said with Cardozo that these are risks inherent in the game of life.3 2 In any event what is involved here are marginal cases as opposed to the essential problem affecting the revision of precedents. Correcting Erroneous Precedent and Reforming Undesirable Precedent 7. We have not found that the advocates of prospective revision confine their ideas to revision of precedent when not marred by gross error, as being contra legem. Cicero said: cujusvis hominis est errare; nullius, nisi insipientis, in errore perseverare (Philippica XII, 2). Hence the well-known scholastic saying: errare humanum est, perseverare autem diabolicum. The question remains whether a judge can behave like the fool mentioned by Cicero or like the devil mentioned by the Scholastics. What we mean is whether a judge should declare: "This precedent is indeed erroneous-since it overlooks the law or is founded on an imprecise reference to the law or the 30

31 32

Mishkin, "Foreword, The Supreme Court 1964 Term" (1965) 79 tHarv. L.R. 56, 65. Cf. ibid.; Note (1947) 51 Va. L. R. 201, 250. Cardozo, Nature of Judicial Process (Yale University Press, 1925) 148. Cf. Note (1947) Va.L.R. 201, 250.

No. 2, 1973]

PROSPECTIVE REVISION OF PRECEDENT

183

like. Nevertheless it is as well that it continue to apply to the matter now before me (or alternatively not to this matter but to every other matter that has already occurred but has not yet been legally decided) ". It may perhaps be urged that it is right to establish whether in the particular case before him, the judge is confronted by a precedent binding upon him, being set by a superior court; and if that is so, then it may be urged that sec. 33 of the Courts Law, 1957 does not permit him to depart from it. The argument is not, however, wholly tenable. Even if the language of sec. 33 is not restrictive, it must be given a reasonable interpretation having regard to the English theory of precedent which before the enactment of the Israeli Law was held in principle to be operative in this country and which therefore served as the background to that Law. So it should be deemed, for instance, with regard to denying obiter dicta any authority, although such denial is not explicit in the terms of the Law themselves. 33 So also with regard to a decision 33

Incidentally, I may be permitted a reply to Dr. Witkon ("Some Reflections on Judicial Law-making" (1967) 2 Is. L. R. 475, 479) who writes that I "criticized the court for expressing disapproval of Lamm J.'s refusal in the District Court to follow a previous decision of the Supreme Court which, in Professor Tedeschi's view, merely contained an obiter dictum and was therefore not binding on the District Court". In my note on the case in question (in (1964) 20 HaPraklit 99) I did not intend to express such a view as being my own but to refer to the view of Witkon J. himself which I cited verbatim: "The course which the judge took was not correct even on the assumption that the question did not arise in the first case except by way of an example with which the judge sought to buttress his opinion". See Israel Permanent Investment Co. v. Silver (1963) 17 P.D. 1655, 1660. Dr. Witkon adds the following in his aforementioned article: "As regards obiter dicta, I would suggest that they are of two different kinds, those which are gratuitously thrown in and have, at best, persuasive influence, and those which are stated as a necessary corrolary [scil., corollary] to the ratio decidendi of the case. Disregard of the latter kind of obiter dictum is tantamount to a denial of the precedent itself". One may observe as to this distinction between two kinds of obiter dicta, of which there is no suggestion in the judgment, that what is a necessary consequence of the ratio, whether or not stated in the judgment, must in logic be considered as belonging to the ratio itself. Against this conclusion one may cite from English law the well-known remarks of Lord Halsbury: "I entirely deny that it [scil., a case] can be quoted for a proposition that may seem to flow logically from it" (Quinn v. Leath [1901] A.C. 495, 506)-provided we do not understand these remarks as Cross does in his Precedent in English Law (2nd ed., 1968) 189 et seq. In any case, according to the first solution (the consequence of the ratio regarded as part of the ratio itself) one cannot raise the question which Dr. Witkon seeks to do. And according to the second solution, the conclusion will be that the corollary is not binding. The fact of the matter is that either the corollary is binding as ratio decidendi or it is not binding at all. Why therefore term the corollary expressed in the decision-as Dr. Witkon does--obiter dictum rather than an interpretation--correct on his assumption-of the ratio decidendi or of some aspect thereof? And if what was involved in the above-mentioned judgment was a necessary corollary of the ratio decidendi, why did

ISRAEL LAW REVIEW

[Is.L.R. Vol, 8

per incuriam in the instances mentioned above. Such a precedent has never been considered binding in English law, 34 and a fortiori it has no authority under Israel law, since in Israel the entire authority of precedent derives from clear and express statutory dispensation, and it cannot be imagined that the legislature thought to attribute to the judges the power of waiving its own laws. At all events, our conclusion is still more certain where it is not a matter of an inferior court having to decide whether it must follow an erroneous precedent of a superior court. It is inconceivable then that the court, in order to honour the expectations based upon it, would decide-even if only for the time being-according to the precedent erroneous in the above sense. Yet, Professor Grunsky has expressed a contrary view in regard to West German law, not indeed in respect of the specific case of a judgment per incuriam but without expressly excluding that instance from his general statement in favour of the operation of an incorrect precedent. He adverts to the criticism that may levelled against his view ("The court [... ] could not be compelled knowingly to pass a judgment which is considered wrong") but he counters: "This objection can be invalidated first of all by pointing out the corresponding dilemma of the legislator. If the latter wishes to create a new law because the law as it exists is no longer satisfactory to him, he is nevertheless prevented by the principle of constitutionalism from subjecting the already existing facts of a case to the new regulations. Then it would no longer be inconcievable that this also applied to the judge. Above all, however, the question as to whether the decision in an individual case is "right" or "wrong" cannot be answered without taking current practice into consideration. The "intrinsically right" solution can be unjust, and therefore wrong, if it comes as a surprise to the parties concerned and thus could not be included in their calculation from the outset".3 5 (In Grunsky's reasoning, we must replace "principle of constitutionalism" by "the normal practice of the legislator regarding the retrospective operation of a law" in order to adapt Witkon J. say that "the fidelity required of a judge to a rule decided by a superior

court (sec. 33 of the Courts Law, 1957 [supra n. 13]) is not a formal matter. Its source lies in public order and the courtesy which one court expects of another as well as in the will to avoid barren deliberations", as if sec. 33 sees precedent even beyond the ratio decidendi? 34 Even Reem v. Minister of Finance (1954) 8 P.D. 494, which dealt with the question of "self-deference" before the enactment of the Courts Law, 1957, excluded a precedent per incuriam therefrom-but the question was not raised whether in the opinion of the judges such erroneous precedents could bind inferior courts. 35 Grunsky, op cit., p. 10 (English abridgement, p. 78). Cf. C.S. Rogers, "Perspective on Prospective Overruling" (Winter, 1968) 36 U.M.K.C.L. Rev., 35, 73: "[...] it is more just for the courts to correct their own errors, while not retroactively penalizing those who had relied upon the erroneous decision, than it is either to leave the error uncorrected or to apply the overruling decisions retroactively [...]"

No. 2, 1973]

PROSPECTIVE REVISION OF PRECEDENT

185

it to Israel law or any other legal system which possesses no constitutional restrictions like the West German regarding the retroactivity of law). Clearly the analogy does not hold. There is no similarity between the freedom of the legislator (which is complete in the absence of any constitutional restraint as mentioned above) and the subordination of the judge to the law. In point of law, the legislator does not amend an error when he effects the reform of an unjust or undesirable law, in contrast to a judge when he corrects an error into which a judgment, given per incuriam and contra legem, may have fallen. There is no similarity in the public expectation and its legitimacy between these two instances. Where the legislator amends a statute, it is proper to have regard to legitimate expectations based upon the rule repealed, be the reason for repeal what it may. But when the law says "white" and the precedent "black" it would be arbitrary and anti-legal to assume that "black" alone was in the public mind or at least was more in its mind or so understood than the "white" of the law. If we had to assume that the citizen is incapable of understanding the laws, he would also be incapable of understanding the precedents (moreover, the latter are in a number of respects more difficult to interpret than the law); and the entire theory of prospective revision would be vacuous. It may also be assumed that the jurists will make it easier for the public to understand the law and will be watchful in pointing out any inconsistency between it and per incuriam judgments. Our conclusion is that the correcting of a precedent contra legem must be retroactive. In this manner alone is the hierarchy of the sources of law maintained and an attitude of respect for the law ensured. 8. It is clear that the instances with which we have dealt do not exhaust all the circumstances in which a revision of precedent arises. The abovementioned instances (of a precedent suffering from gross error), even if not very infrequent, are exceptional. At the other end of the scale, an old precedent may have been laid down when a lacuna existed in the legal system and thus may have led to the creation of a rule in the absence of any relevant legal regulation. If another judge should subsequently depart from the precedent, that does not mean that he does so because he considers it to be contrary to the ruling principles of the legal system but more perhaps because he thinks another ordering of the matter is more just and more desirable or expedient, whether that appears to him to have been the situation from the very outset or whether new social conditions have brought that about. Although in the past many have thought that every legal system is free of lacunae and provides a single proper solution for every question that may arise, this view has been abandoned by the majority, but this is not the occasion to dwell upon that doctrine and its fictitious nature. In instances such as that last illustratedthe divergence of a new precedent from one which filled an existing gap in the legal system-it is not a matter of rectifying an error that has occurred in an earlier precedent in view of the state of the law but of reforming an undesirable

ISRAEL LAW REVIEW

[Is.L.R. Vol. 8

regulation. Clearly no reasons exist here which call for retroactivity in the other instance. It should also be noted that on the last assumption a revision looking to the future only would not be open to the objection raised against prospective revision in general, that is to say, that the introduction of this practice may encourage the courts to "legislate" and so avoid "the often unpleasant consequences of retroactivity",. 6 Since, on the assumption we have posited, a lacuna in the legislative rules existed, there is no reason for setting up obstacles to the normative activity of the courts. A precedent concerning a lacuna may be compared with a precedent that emerges when a legislative rule is vague, insufficient or ambiguous. Here also the view has been, and still is, taken that in every system all solutions but one are wrong, so that no distinction is to be drawn between this instance and the previous one of the precedent made per incuriam, contra legem. Nevertheless, if the legal system does not demand such a fiction, the foregoing argument can be dismissed as being artificial. The existence of legislative rules because of their vagueness cannot go so far as to admit of a single solution appropriate for every problem-even if the honest interpreter must regard these rules as disqualifying certain solutions. As for statutory interpretation pure and simple, 7 where a judge does not follow some precedent because he prefers another interpretation of the law, we must for our present purpose ascertain in which sense he regards his own interpretation as preferable. If he considers the old precedent to be negatived by the law, then the situation is similar to that of a precedent per incuriam, on which we have already dwelt. If, on the contrary, the interpretation given in the old precedent seems to the judge a possible one-although another is regarded by him to be more desirable-we have a situation very close to the other field, i.e. or reform (and not of amending a mistake). Here the work of the judge is objectively similar to that of a legislator changing the earlier law. The principle of the judge's subjection to the law does not require retroactivity, if we must understand by law that which the law expresses and not that which one may wish to attribute to it. On the other hand, in this instance there are reasons which generally counter the retrospective operation of enacted law. And these reasons indeed are perhaps of greater weight here, since reform of the law is obviously more

36 3

Cf., e.g., Note (1961)

71 Yale L.J. 907, 925, 926, 928. It is to be noted that American case law, even when it normally followed the path

of retroactivity, recognized an exception with regard to the interpretation of statutory law relating to contract. It was thought that the construction becomes as much a part of the statute as the text itself, and a change of decision is to all intents and purposes the same in its effect on contracts as an amendment of the law by means of a legislative enactment. See, e.g., von Moschzisker, "Stare Decisis in Courts of Last Resort" (1924) 37 Harv. L. R. 409, 422-24; C.M.D., op .cit., 212.

No. 2, 1973]

PROSPECTIVE REVISION OF PRECEDENT

187

expressive of the democratic process and often is largely heralded by public debate and therefore less likely to take the public by surprise than changes in case law. That does not mean that every reform of an undesirable precedent (as distinct from the amendment of a mistaken precedent) must actually be prospective only. The legislator also does not always effect a reform for the future alone. One must distinguish between different matters, between different factual situations, between different kinds of reform, and in this light certain expectations based on the previous legal situation may or may not be regarded as justified; and even when the expectation is legitimate, other considerations may well enter into the picture as a decisive counterbalance. When a reform effected by the judiciary is involved it is possible to take the concrete circumstances into account and adapt the proper process to them, in much larger measures than with reform by a legislator. Apart from all the foregoing, and whatever the desirable balance of opportunities, we may not ignore one important restriction that flows from the limited power of the judges under the given system-in our case the Israel system. The Power of the Courts 9. A question different from the previous one (when will the denial of retroactivity adversely affect the principle of the judiciary's subjection to the law and when will it not) is the question whether, at least in Israel law, the judges are empowered to adopt the techniques of prospective revision as suggested by American (and German) lawyers and as employed, in large part by American law (and to some extent also by West German law). More particularly, may the court declare "a break with the past in anticipation of the future" in its various forms or the non-application of the new rule to "old" cases? Further, may a "disjunctive declaration" in its most positive form attain to "self-shackling" by the court? In any event, can it alter the legal position deriving from precedents which precede it in respect of inferior courts? And with regard to the non-applicability of the new rule to old cases, can the declaration release the judges from the authority of the new precedent so far as these old cases are concerned? Or can it possibly become a prohibition on inferior judges to decide according to the new precedent? In America, only some aspects of these questions have been discussed, i.e., those associated with the Constitution, which are not of interest here. In general terms, American lawyers do not greatly concern themselves with these questions. Cardozo, in dealing with "disjunctive declaration" which he views more in a negative manner-a departure from an earlier trend-than the laying down of a new line-says "I am not persuaded altogether that competence to proceed along these lines does not belong to the judges even now without the aid of statute", but adds immediately: "If the competence

ISRAEL LAW REVIEW

[Is.L.R. Vol. 8 3

does not exist, it should be conferred by legislation". Kocourek initially suggested statutory regulation of the matter 39 but subsequently took the view that that was unnecessary for the following reasons: "It needs to be emphasized that both the declaratory principle with its supplement, the retrospective rule and, likewise, the stare decisis rule, were created by the courts. No court can limit its future competence any more than a present legislature can limit the competence of a future legislature. The same power that has made can also unmake. The argument presented leads to the simple conclusion that [...] in no case of justifiable reliance should an overruling decision disturb transactions reasonably entered into on the faith of existing declarations to the law. 4 The courts already inherently have the power to accomplish this program". 0 As against this, Justice von Moschzisker criticizes the new trend and sees "prophecy" as tantamount to "legislation". It is not, however, clear whether for this reason he thinks that it (the prophecy) is forbidden to the courts as exceeding the limits of their competence, since he simultaneously calls it "legislation" and obiter dicta (two things which are prima facie contrary to one another).41 Prof. Mishkin on his part deals with the decision in Molitor42 where the judges preferred to apply a new rule (although they announced that it would not apply to other "old" cases) with the declared purpose, inter alia, that it should not remain merely obiter dictum. Mishkin observes that this concern of the court excites irony since its announcement that the new rule should not apply to other cases was itself obiter dictum (and also unreliable because it was not followed in connection with Molitor itself nor in other instances) .43 10. The rules regulating precedent are enigmatic in their different aspects in almost every legal system, both those which recognize in principle the binding authority of precedent in a generally accepted form and those where a hiatus exists between theory and practice in this regard. What is particularly unclear is the "dynamic" in the order of things, the question whether from the legal point of view there is the possibility of varying the rules which regulate precedent, and in this case what is the extent of the changes which are permitted and which are the organs competent to effect them; or whether under the law all possibility of change is barred, at least change by the courts themselves, so that if change does in fact occur it must be regarded as a "revolution". 38 Cardozo, N.Y. State Bar Address (1932) N.Y.S.B.A. Report 263, 297. 39 Kocourek, in (1931) 17 A.B.A.J., cit. 40 Kocourek and Koven, op. cit., 971, 999. 41 von Moschzisker, op. cit. 42 Molitor v. Kaneland Community Unit School District 18 Ill. 2d 11, 163 N.E. 2d 89 (1959). 43 Mishkin and Morris, op. cit., p. 309 et seq. See also Mishkin (1965) 79 Harv. L.R. 56 et seq.; Note (1965) 51 Va. L. R. 201, 216-17 and the further references there.

No. 2, 1973]

PROSPECTIVE REVISION OF PRECEDENT

189

It is only possible to assert with certainty that from the analytical point of view the principle of the authority of precedent does not flow from precedents themselves. As Salmond puts it, "you must possess authority before you can confer it".44 The fact that historically it was the courts which imposed the rule of precedent in the English common law system does not involve going on to say that-from the dogmatic point of view (namely, from the point of view of the legal system)-the power of the courts themselves to change these rules is to be recognized; and even if it be so, whether the competence rests in the generality of the judges as a group (their position being revealed in long usage) or is decentralized and each judge may, even by a single decision, provide for his own precedents, or whether the power is altogether concentrated in the hands of some superior court. Thus, for instance, the question arises in English law whether the rule according to which it holds itself bound by its own decisions, which the Court of Appeal once adopted, derived its force from the view of the matter taken by the Court itself or from the subsequent affirmation of the House of Lords.45 The last paragraph of the latter's Practice Statement (Judicial Precedent) of 19664G may possibly be understood as showing that the Lords think that they have this power, but some have taken the view that to release 47 the Court of Appeal from its own precedents requires legislative intervention, 48 whilst some have held that the Court of Appeal may do so itself. There is also the question of how to reconcile the 1966 Statement under which the Lords are no longer bound by their own precedents with the view previously held which found expression in the well-known remarks of Lord 49 Campbell in Beamish v. Beamish. 11. Whatever the situation in other legal systems, a number of matters emerge clearly from Israel law under sec. 33 of the Courts Law. Here, the principle does not exist that the force of precedent is, in the terminology of Kelsen, a Grundnorm, as is suggested in other systems. Nor is it an outcome of judicial practice-whether a regular practice or one characterized by decentralization, that is a practice attributing to individual courts or to one of them alone the autonomy to prescribe regulations and to vary them, even by

Salmond, Jurisprudence (11th. ed.) 137. Cf. Cross, Precedent in English Law (1961) 246 et seq. 45 Ibid., p. 247. 46 "This announcement is not intended to affect the use of precedent elsewhere than in this House". 47 See the Law Society's letter to the London Times of July 27, 1966. 48 Dias, loc. cit.; Goodhart, "The Abolition of Absolute Precedents in the House of Lords" (1966) 82 L.Q.R. 441, 443. 49 (1861) 9 H.L. Cas. 274, 338-39. 44

ISRAEL LAW REVIEW

[Is.L.R. Vol. 8

means of a solitary decision, as indicated above. In Israel it is a normal legislative rule made by the legislature and open to any alteration by it. Prior to the Courts Law of 1957 the Supreme Court manifested a certain tendency to regard precedent as a matter of its own autonomy. The crop harvested was the judgment of the High Court of Justice in Reem v. 5 Minister of Finance " which purported to regulate the question of the Supreme Court being bound by its own decisions. But it is because of this very background that the statutory regulation of the matter under sec. 33 acquires all its value. The statutory provision is intended to negative not only the "self-shackling" of the Supreme Court but also autonomy of the courts in general as regards precedent. The authority of precedent derives from the legislature and the legislature alone. Professor Yadin takes the view that a matter such as the question relating to the "self-shackling" of the Supreme Court is not capable of being legislated (just as certain questions are not considered to be justiciable)." In the present context indeed an American court has held that a statute which restricts overruling decisions to prospective effect constituted a legislative attempt to curtail the power of the judiciary and was "an invasion of the powers and authority of the judiciary and violating the doctrine of the separation of powers". 52 We think on the contrary that all matters are capable of being legislated (in our law, at least) and that all legislation presupposes and requires that the judge be loyal to the law and therefore ready-here as in every other matter-to yield to the view of the legislature. In the light of these principles one must deal with the question of the force to be attached to judicial pronouncements that purport to restrict the authority of a given precedent (retroactivity of limited scope) or are directed to bind for the future, the court making them by means of its own "prophecy", as it were.

It will be urged vainly on behalf of the first kind of declaration that just as, for instance, the Supreme Court can by its own precedent in the normal manner bind courts inferior to it, it may also bind them in respect only of certain instances (the "new" cases): the greater includes the lesser. In truth, in Israel law the "deference" never proceeds from the court; it is the legislature that binds, assuming the existence of a precedent. The declaration of the Supreme Court prohibiting other courts from applying a new rule to "old" cases, or releasing them from the force of the new precedent in respect of such cases, does not only actually constitute obiter dictum (as Mishkin says with reference to American law: supra section 9) but it would be dictum forbidden to the judge making it. 5 50 51

52

53

(1954) 8 P.D. 494. U. Yadin, "Truth v. Stability" (1972) 28 HaPraklit 152, 154. Davis v. G.M.A.C. 176 Neb. 865, 871, 127 N.W. 2d 907, 912 (1964). Rogers (op. cit. supra n. 35 at p. 62) is critical of these considerations of the court. This conclusion under current law is not to be regretted, and accordingly there

No. 2, 1973]

PROSPECTIVE REVISION OF PRECEDENT

191

A similar conclusion suggests itself with regard to the "disjunctive declaration" by the Supreme Court, if deemed as a release5 4 of the lower courts from an old precedent or as a prohibition upon them-or upon itself-to follow an old precedent in the future. As against this, a caveat that the public should no longer rely upon a rule hitherto applied cannot be regarded as in any way affecting the principles of our law. On the contrary, the use by the Supreme Court of this method may encourage it to respect the law. At present the Supreme Court does not tend to depart from its precedents, although sec. 33 of the Courts Law negatives self-shackling. One may assume that it would be more prepared to change its case law, were there no room for fearing that the expectations of the public would be disappointed. Kocourek indeed thought that after making such a declaration a court will regard itself morally bound to follow the path it has indicated and that it is therefore possible to rely upon the declaration. 5 In fact American practice has not always been so and cases have occurred where declarations of this kind by the courts have not been abided by.5 6 In any event a declaration importing the moral obligation of the Supreme Court--or any other court-

is no occasion for initiating a legislative innovation permitting "limited retroactivity". This affects the principle of equality in discriminating among people who acted before the judgment: the litigants in the particular case are discriminated (favourably as "chance beneficiaries" and unfavourably as "victims" as the case may be) as against all other potential litigants in connection with "old" cases. Those who favour this technique regard the discrimination as the necessary price to be paid for being able to make the desired change in the case law, since the alternative technique ("prophecy" or "disjunctive declaration") discourages the initiative of a litigant who cannot hope to benefit from a change of trend brought about through himself. Nevertheless this criticism is answered by the fact that (a) the institutional litigant will always be interested in a change in case law as regards the future; (b) sometimes a potential litigant may hope that the court will apply the new rule retroactively for him by reason of the special circumstances; (c) many occasions present themselves in which the court may in its judgments hint that a particular rule, affecting only incidentally a question then before it, is proper to be changed in the future. See Note (1961) 71 Yale L.J. 907, 945; Kocourek and Koven, op. cit., p. 945; Note (1947) 60 Harv. L.R. 437, 440. Cf. also Black J. in Williams case, 364 Mich. 111, N.W. 2d 1, 16 (1961). 54 See as against this the suggestion made by Dias, loc. cit., regarding English law, that the court will declare: "let the present case be governed by the precedent to be overruled for the last time [...] but let the precedent be deprived of effect for the future". Cf. Aigler, "Law Reform by Rejection of Stare Decisis" (1964) 5 Ariz. L.R. 155, 157: "One may find interest in speculation as to what would be the effect if [...] judges were to incorporate in their opinion a statement in substance: 'This decision shall not constitute a precedent' ". 55 Kocourek and Koven, op. cit., p. 995. 56 Schaefer, "The Control of 'Sunburst': Techniques of Prospective Overruling" (1967) 22 Record of N.Y.C.B.A. 394, 402.

ISRAEL LAW REVIEW

[Is.L.R. Vol. 8

with regard to its future action is not in tune with the spirit of sec. 33 which is opposed to self-shackling. It is inconceivable that an Israeli judge would follow the practice of the Roman praetor who declared in advance in his edictum in which cases judicium dabo; such a declaration bound him (according to the lex Cornelia of 67 BCE) and restricted his powers, although some scholars dispute this consequence, at least in the sense that in their opinion the litigant had no remedy if the praetor did not abide by his edict. Certain American writers assert that there is no real difference between precedents and "prophecies", because in their opinion the substance of a precedent is no greater than that of a "prophecy".5 " Whether or not that is correct according to American law, it seems to us that under Israel law the two sides of this presumed equation need changing so that they are not equal to one another but vastly different. On the one side, the authority of a precedent is not merely a moral authority in our law but a precise legal authority of two levels according to sec. 33 of the Courts Law. On the other side, the provisions of sec. 33 itself serve as an argumentum a contrario for denying any authority to "prophecy" and for prohibiting it from meaning "self-shackling" or for imposing any duty of "deference" upon the courts inferior to the court making the declaration. For this reason one cannot agree, so far as our law is concerned, with the view denying that "prophecy" be regarded merely as an obiter dictum simply because "anything in a court's opinion beyond the judgment order is legislative to the extent to lay down a general rule",5 s if by that it is meant to attach any authority to what is said in a judgment beyond the ratio decidendi. Nevertheless, the negation of all authority from "disjunctive declaration" is not a denial of its practical importance. Even without any obligation upon a judge to render judgment in the future according to such a declaration or to attach greater value to it than it warrants on an independent appraisal that is persuasive for him at the time, it is important for the public to know what is the present opinion of the judges with regard to their future judicial activity; and it is also important that the judge who will adjudicate in the future should then know that the public was already prepared for the possibility of a change in precedent. As in other social, political and legal questions, communication is a wise thing and productive of good and harmonious relations.

Note (1947) 60 Harv. L. R. 437, 440: "a holding itself is but a prophecy"; R.H. Freeman, "The Protection Afforded Against the Retroactive Operation of Overruling Decisions" (1918) 18 Col. L. R. 230, 232: "the obligations imposed upon the courts by the doctrine of stare decisis is a moral obligation only". Cf. Chamberlain, Stare Decisis (1885) 19. 5s8 Schaefer, loc. cit. 57

No. 2, 1973]

PROSPECTIVE REVISION OF PRECEDENT

Sell-overruling? 13. The situation in which for the present purpose an Israel District Court finds itself is dependent inter alia upon an answer to the following question: can a judgment of the Supreme Court overrule a previous precedent of that court? The situation in a Magistrates' Court is similar in face of different precedents of any particular District Court (besides those of the Supreme Court). This is one of the questions to which no answer is given in the laconic wording of sec. 33. Prior to its enactment there does not appear to have existed any particular local practice which would enable one to gather the intention of the legislature. The English background of the Israel theory of precedent is also not very clear in this respect. On the one hand Salmond says "since overruling is the act of a superior authority, a case is not overruled merely because there exists some later opposing precedent of the same court or a court of co-ordinate jurisdiction. In such circumstances a court is free to follow either precedent".5 9 Cross suggests another course. Although he admits "that no-one has even suggested that a High Court judge of first instance can overrule, as opposed to differ from, the decision of another High Court judge" and agrees that a court which is "self-shackling" cannot impliedly overrule itself, nevertheless as regards express overruling he thinks, contrary to Salmond, that certain judicial observations may be cited which appear to admit such a possibility, for instance those of Channel J. in Logsdon v. Trotter.'" In the United States the ability of a court to overrule a previous precedent of its own is recognized but the American rules relating to precedents would not appear to serve as a basis for interpreting the Israel provisions on the subject. The question in Israel law can perhaps be sub-divided into a number of secondary questions: (a) Can the ratio decidendi adopted in a Further Hearing overrule the different ratio decidendi embodied in the judgment which is the subject matter of that Further Hearing? (b) Can the decision in a Further Hearing which is contrary to the ratio decidendi of a previous ordinary precedent of the Supreme Court in other matters overrule such a precedent? (c) Can an ordinary precedent of the Supreme Court overrule (at least expressly) a previous precedent of its own; and similarly a District Court precedent as against a previous precedent of that court? The answer to the first of these questions seems to be in the affirmative. 59 Salmond, Jurisprudence (12th ed., 1966) 148. 60 [1900] 1 Q.B. 616, 623. For other examples, see Cross, Precedent, cit. (1961) et seq. ((2nd ed., 1968) 123 et seq.).

125

ISRAEL LAW REVIEW

[Is.L.R. Vol. 8

In the whole theory of precedent, a decision which is not authoritative with regard to the case in hand cannot be considered as authoritative as a precedent. As against this, the answer to the second question is less certain. On the one hand, in the institution of Further Hearing is manifest the consideration of the Further Hearing decision as a precedent (witness the fact that the initiation of a Further Hearing can lie with the Supreme Court itself); and that may be a ground for preferring the decision over any prior ordinary precedent. Yet, on the other hand, the very fact that the previous case could have been brought to Further Hearing upon official initiative but that was not done may perhaps be regarded as a reason for comparing the two precedents and therefore to equate the second and third question. It should be mentioned that sec. 33 speaks of "a precedent established" and not of "the precedent established". Had the legislator said the second, that would have been evidence in favour of "self overruling", since the legislator himself would have directed that the rule taken into consideration is a single rule and therefore by implication the one lastly established. This may lead us to a negative conclusion with regard to the third question. Another reason can be found in the fact that in our law the whole theory of precedent is statutory. The legislator himself-and not the judge-determines when a precedent is authoritative, as regards whom and on what level of authority. The words in which the Supreme Court purports to abrogate an earlier precedent of its own are not a precedent and are without authority, since under the Law the other courts are only bound by precedents (of the Supreme Court). Just as the judge cannot confer any precedental authority, so logic requires that he should be unable to abrogate such authority. The legislator himself impliedly provides when overruling is automatic by virtue of the higher status of the judge who creates the new precedent as opposed to the judge who created the earlier one. On these lines-not contemplating self-overruling except within the context of the first question and at most of the second-we think that District and Magistrates' Courts are free to choose between various precedents of the Supreme Court"' even if the last of them has purported to overrule its predecessors. (This conclusion must of course be understood with two reservations, one which was postulated with regard to the first question above and the other that in the meantime no new legislative rules have been introduced which remove the foundations of some precedent or another). A similar freedom exists for the Magistrates' Courts in respect of the various precedents of the District Courts. But let us examine the questions in greater detail. If a Supreme Court judgment makes a disjunctive declaration about the 61

As against this, the Ontario Supreme Court has held that although an inferior court must follow the rules laid down by the Court of Appeal, where the latter itself has deviated from its own rules the inferior court must adopt the most recent of such inconsistent rules and leave it to the Court of Appeal to decide which rule is to be preferred: Chliwniak v. Chliwniak (1972) 24 D.L.R. 3d 646.

No. 2,1973]

PROSPECTIVE REVISION OF PRECEDENT

line which it took previously regarding a particular question, such a declaration would not alter the obligation of another court to follow the old precedents. One can indeed contemplate that after a "disjunctive declaration" the Supreme Court will rectify accordingly every judgment of the District Courts inspired by the old rule and touching a "new" case. After one or more of such rectifications (and perhaps in anticipation that these will occur, even if they have not in any concrete instance) one can also contemplate that the other courts will deem it futile to continue to follow the old precedent of the Supreme Court which it has already criticized and thus deprived of its moral force, and accordingly they will in fact proceed in the new direction which the Supreme Court has indicated although the latter has not yet adopted it in practice in its judgments. 62 Nevertheless that would, in strict law, constitute an infraction of sec. 33;63 and if such a practice should spread, it would have to be regarded as a judicial "revolution". Where the Supreme Court decides contrary to a past precedent and declares that with regard to other "old" cases the old rule should still apply, other courts ought not to have regard to the latter since it is contrary to the law. If the Supreme Court is denied the capacity of abrogating its precedents (apart from the situation in a Further Hearing with regard to the judgment in question), the other courts would be at liberty to proceed along one or other direction, since they would be presented with self-contradictory authorities. This freedom of choice would also avail them if the above restriction was not contained in a judgment of the Supreme Court; and they might always be at liberty to prefer the first (previous) solution for "old" cases (or those of such "old" cases in which it can be shown or implied that the litigants placed reliance upon the old rule) and the second (new) solution for "new" cases. If the Supreme Court is deprived of the power to annul its previous precedents, such freedom of choice will rest in the other courts even when, after a disjunctive declaration is made in respect of the old rule for the future, the Supreme Court has the opportunity to decide according to the new rule in respect of a "new" case. The other courts would be permitted 62

With regard to the possibility that a court bound by the precedent of a superior court will jump the gun, that is to say, will disregard that precedent, in the hope that the superior court will diverge from it, see for the trend in America to tolerate such practice Auerbach et al., The Legal Process (1961) 183. For the opposite trend see the observations of the Chief Justice of Australia in Jacobs v. Utah Contruction

63

& Engineering Pty. Ltd. (1966) 40 A.L.J.R. 306, 308; Cross, "Recent Developments in the Practice of Precedent-The Triumph of Common Sense" (1969) 43 Austr. L.J. 3, 5. By contrast, there would not occur an infraction of sec. 33 if the District Court would say in its judgment: "We must decide according to the old rule; the losing side in this case should appeal and the Supreme Court may vary this judgment in accordance with the new rule which it has declared."

ISRAEL LAW REVIEW

[Is.L.R. Vol. 8

to act in one way or another and would do so according to whether a "new" or "old" case is before them. If, on the other hand, the Supreme Court is deemed to be free to annul its precedents-and if the last judgment, given under the influence of the new trend, constitutes a true annulment-then in strict law the other courts would have to follow the new direction even with regard to "old" cases. The Supreme Court which is never bound by its own precedents may always amend such a decision by itself applying the old rule to an old case. Similar conditions apply to the precedents of District Courts in relation to the decisions of Magistrates' Courts, but subject to two observations. Even if a District Court is regarded as being free to set aside its own precedents and if in fact that happens when a new rule is adopted-whilst there exist precedents of another District Court in line with the old rule-the Magistrates' Courts would always be at liberty to choose between the two rules and might therefore exercise this choice according to the circumstances, that is to say, according to whether they have to decide an "old" or "new" case. One always assumes that no authority of the Supreme Court is involved. Moreover, the authority of District Courts precedents is only "guiding" and therefore the Magistrates' Courts would always preserve some freedom with regard to such precedents, in contrast to the situation created in the case of a Supreme Court precedent, when every other court is denied freedom of choice.