Home | Databases | WorldLII | Search | Feedback Otago Law Review

Wall, Jesse --- "Prospective overruling - it's about time" [2009] OtaLawRw 7; (2009) 12 Otago Law Review 131

Last Updated: 26 February 2012

^{Prospective overruling – it’s about time}

Jesse Wall*

“The past is a foreign country; they do things differently there”.¹

The past has different standards, different values and different expectations flowing from different rules. It is as if it were a foreign jurisdiction. Time is therefore an important factor concerning how we should view the legal consequences of conduct. The courts, as adjudicators, determine the legal consequences of events well after they occur. When applying the common law, they are sometimes forced to cross temporal borders into the past but apply contemporary rules. This creates a problem for the rule of law in the form of ‘retrospective common law’.

In this article I will explain how the common law operates to produce retrospective law and how this is in turn premised upon certain jurisprudential assumptions. I will then turn to assess the prospect that ‘prospective overruling’ may provide a solution to the problem of retrospective common law. The purpose of this article is to provide an argument in support of appellate courts prospectively overruling prior precedents. Accordingly, I shall set out the merits of, and refute any objections to, prospective overruling. However, it is necessary to first explore theoretical foundations of the problem that prospective overruling is purporting to solve. It is necessary to explore these theoretical foundations so as to avoid the disagreement about prospective overruling dissolving into a disagreement about the law itself.

I. Retrospective Common Law

When a court decides to depart from a prior precedent, and restate the law, a court is attaching new legal consequences to events that occurred prior to that restatement of the law. This outcome is the result of an inevitable tension between two conflicting aims of the common law. On one hand, in accordance with the rule of law tradition, it is expected to be consistent, predictable and certain. On the other hand, it should also be malleable so as to arrive at fair and just outcomes. This tension between certainty and fairness operates beneath the surface of all judicial reasoning but it surfaces most abruptly when a court is considering overturning a prior precedent.

To ensure consistency and certainty one of the basic principles in

^* This article is a revision of a dissertation completed in 2007 in fulfilment of the LLB(Hons) at the University of Otago. The dissertation gained the 2007 New Zealand Legal Research Foundation Award for best unpublished student paper. I would like to sincerely thank Professor John Dawson for his continued support with regards to this article.

^{1 L P Hartley, The Go Between (Hamish Hamilton, 1953) 1.}

the administration of justice is to treat like cases alike.² Hence a prior precedent represents a statement of law that can be relied upon if a sufficiently analogous case were to arise. However, in order to achieve a just and fair result the court may find it necessary to depart from the precedent and replace it with a new statement of law. The aims of certainty and fairness may therefore conflict in a particular case.

However, this conflict does not cause retrospective law. The further catalyst is the courts’ adjudicative function.³ That is, retrospectivity arises because the courts’ function is to decide the legal consequences of past acts or omissions. This means that when a court departs from a prior precedent, the new statement of law determines the legal consequences of past acts or omissions. Yet the old statement of law (embodied in the prior precedent) arguably was the law when the act or omission was performed. By overturning a prior precedent, courts announce and apply law that attaches new legal consequences to past events, so there is an element of retrospectivity whenever they overturn a prior precedent.

The problem is that any form of retrospective change is an afront to the rule of law. Whereas prospective legal change allows for rearrangement of individual affairs to align with the new state of the law, retrospective change undercuts our ability to plan our affairs in reliance on the law promulgated at the time. A major ‘harm’ of retrospective law is therefore the injury it inflicts on the rule of law. The law cannot be predictable, and people are unable to be guided by it, if it can be changed after the fact. Alternatively, in the words of Bentham:⁴

It is the judges (as we have seen) that make the common law. Do you know how they make it? Just as a man makes law for his dog. When your dog does anything you want to break him of, you wait till he does it, and then [discipline] him for it. This is the way you make laws for your dog: and this is the way judges make laws for you and me.

The problem of retrospective common law has recently come to the forefront in the decisions of the House of Lords in National Westminister Bank Plc v Spectrum Plus Ltd⁵ and the New Zealand Supreme Court in Chamberlains v Lai.⁶ Both these cases illustrate the problem of retrospective common law change. National Westminister concerned a debenture which Spectrum Plus had created over its debts in favour of National Westminster. In particular, it concerned whether National Westminster had a fixed or floating charge. If it was a floating charge, Spectrum Plus’ creditors would have priority over the bank; if not, the bank would be entitled to the whole of the proceeds. This question of

² R Cross & J W Harris, Precedent in English Law (Clarendon Press, 1991)

^228.

^{3 National Westminster Bank Plc v Spectrum Plus Ltd [2005] UKHL 41, Lord}

^{Nicholls, at paragraph 4.}

^{4 J Bentham, Truth versus Ashurst, or, Law as it is, contrasted with what is said}

^{to be (1823).}

^{5 National Westminster Bank Plc v Spectrum Plus Ltd [2005] UKHL 41.}

^{6 Chamberlains v Lai [2006] NZSC 70; [2007] 2 NZLR 7.}

^{law was governed by the rule established in Siebe Gorman v Barclays Bank7}

^{that such a debenture operates as a fixed charge over the debt.8}

The law prior to Spectrum Plus was therefore that such debentures created a fixed charge in accordance with the law set out in Siebe Gorman. However, the House of Lords unanimously held Siebe Gorman was wrong in law and should be overruled.⁹ Their Lordships held, contrary to Siebe Gorman, that the key distinction between a fixed and floating charge was whether or not the asset was finally appropriated as security for the debt until a future event.¹⁰ As a consequence, the debenture was held to be a floating charge.

The new statement of law concerning the characteristics of a floating charge in Spectrum Plus attached new consequences in terms of the priority of creditors. The debenture had been arranged prior to the annunciation of the new statement of law and in reliance on the old statement in Siebe Gorman. The new statement in Spectrum Plus thus operated retrospectively to change the status of the debenture and to upset reliance on law that was expressed at the time of its formation.

However, as Lord Devlin extra-judicially commented at an earlier time, ‘A judge-made change in the law rarely comes out of a blue sky. Rumblings from Olympus in the form of obiter dicta will give warning of unsettled weather ’.¹¹

Even before Spectrum Plus there were such ‘rumblings from Olympus’. The English Court of Appeal in Re New Bullas Trading¹² supported the Siebe Gorman decision, but the Privy Council in Agnew v Commissioners of Inland Revenue¹³ held that the decision in Bullas Trading was wrongly decided. This cast doubt on the validity of the reasoning in Siebe Gorman.

Such ‘unsettled weather ’ does not avoid the retrospective effect of the judicial change in law enacted by the overruling of Siebe Gorman. Despite the Privy Council questioning the validity of Siebe Gorman, the House of Lords nonetheless felt compelled to overturn explicitly the Court of Appeal’s decision in Siebe Gorman. This suggests Siebe Gorman was the law in England until it was retrospectively changed, and that it was reasonable to previously rely on Siebe Gorman as the law within that jurisdiction.

The decision of the New Zealand Supreme Court in Chamberlains v Lai is a similar instance of retrospective change. When the events occurred that gave rise to the proceeding in Chamberlains ‘everyone would have

⁷ Siebe Gorman v Barclays Bank [1979] 2 Lloyd’s Rep 142.

^{8 Lord Scott of Foscote, at paragraph 105, held that there was no material}

^{difference between the debenture in Siebe Gorman and in Spectrum Plus.}

^{9 Spectrum Plus, Lord Nicholls at paragraph 2.}

^{10 Spectrum Plus, Lord Scott at paragraph 111.}

^{11 Lord Devlin, ‘Judges and Lawmakers’ (1976) 39 Modern Law Review 1,}

^10.

^{12 Re New Bullas Trading [1994] 1 BCLC 485.}

^{13 Agnew v Commissioners of Inland Revenue [2001] UKPC 28.}

said that under New Zealand law as it stood then barristers were immune from suits for negligence in relation to protected conduct’.¹⁴ Yet, eleven years after the event the Court found that barristers’ immunity ought to be abolished, and, as a consequence, the immunity was removed retrospectively. The Supreme Court held that the House of Lord’s decision in Arthur J S Hall v Simons¹⁵ had rendered the position of barristers’ immunity in New Zealand law uncertain. So, as in Spectrum Plus, the Court used the ‘gathering of clouds’ to obscure the retrospective effect of this common law change.

1. Rules and Retrospectivity

However, this problem of retrospective common law is only perceived from certain jurisprudential standpoints. The way one views retrospectivity in the common law depends on how one conceives the law itself. First of all, it is only a problem if the law is viewed through the conceptual framework of Legal Positivism. In contrast, from a Dworkinian perspective, the common law does not change retrospectively if the courts are approaching their task in the proper way.

HLA Hart provides perhaps the best Positivist account of law. According to Hart, the law is essentially a union of primary rules of obligation and secondary rules of adjudication, recognition and change. The primary rules relate to the substance of the law, the secondary rules to the procedure for determining the law. The way the law can change, according to Hart, is specified through secondary rules. “The simplest form of such a rule is that which empowers an individual or body of persons to introduce new primary rules...and to eliminate old rules”.¹⁶

For example, the rule that legislation may introduce new primary rules that defeat earlier primary rules arising out of custom or precedent is an example of a secondary rule of both recognition and change. The rule that the New Zealand Supreme Court can restate the common law in a way that bound the New Zealand Court of Appeal is a further rule of the same kind.

There are still many sources of ambiguity in rule following. First, when we come to identify which rule applies, not all cases fall squarely under existing primary rules. Some cases may fall within the gaps in the rules. It may be said this leaves areas of legal uncertainty caused by our inability to foresee all “possible combinations of circumstances”.¹⁷

In such cases, the outcome cannot be determined by a pre-existing rule. Instead, the judge may be forced to choose between competing interests to enact “subordinate legislation”¹⁸ or “interstitial legislation”¹⁹ through

¹⁴ Chamberlains, Tipping J at paragraph 132.

^{15 Arthur J S Hall v Simons [2002] 1 AC 615.}

^{16 HLA Hart, The Concept of Law (Clarendon Press, 1961) 92-93.}

^{17 Ibid, 128.}

^{18 Ibid, 134.}

^{19 J Bell, Policy Arguments in Judicial Decisions (Oxford University Press, 1982)}

^17.

judicial decision making. Second, in terms of interpretation, some cases will fall into the penumbra of legal uncertainty concerning the meaning of a particular rule, so the court must choose between competing interpretations of the rule.

A set of events, to be governed by a prior precedent, must be seen as squarely covered by the nucleus of current legal rules. Then the precedent can clearly determine the outcome of the case. The jurisdiction of a later court to depart from that prior precedent must then derive from a secondary rule that enables legal change. For example, in its Practice Statement (Judicial Precedent),²⁰ the House of Lords declared that it was able to depart from its earlier decisions, recognising its ability to introduce new primary rules and eliminate old ones.

According to this conception of law, prior precedent (a) establishes the primary rule (x) which can be identified through various secondary rules as the law. However, when a court relies on a secondary rule of change to create new precedent (b), the court overturns prior precedent (a) to eliminate rule (x) and introduce new primary rule (y).

_{Secondary Rule of Change}

^{Precedent (a) Precedent (b)}

_{Change in Law}

^{Primary Rule (x) Primary Rule (y)}

The law that could be determined through the secondary rules of recognition prior to decision (b) was primary rule (x). However, decision (b), in overturning that prior precedent, attaches new legal consequences to past conduct by applying primary rule (y) to acts and omissions that were performed whilst the secondary rules of recognition would have picked out primary rule (x) as the law in force. Therefore, if we perceive the legal system as operating as Hart’s conception of law suggests, retrospective change occurs.

However, a competing conception of the law is offered by Ronald Dworkin. One of the main reasons for Dworkin’s rejection of Hart’s conception of law was that it authorised just this kind of judicial, retrospective law-making in “hard cases”.²¹ Under Dworkin’s conception of law the common law can operate without retrospectivity both in hard cases at the margins of the law and when clear prior precedents are overruled.

Dworkin offers a wider ontology of the law. According to Dworkin, the law applicable to a case goes beyond the clearly recognisable rules found in the governing precedents in the immediate field. Dworkin argues that the law includes principles and other standards and that Positivism “forces us to miss the important roles of these standards that

²⁰ Practice Statement (Judicial Precedent) [1962] 1 WLR 1234.

^{21 R Dworkin, ‘Hard Cases’, in Taking Rights Seriously (Duckworth, 1987)}

^81.

are not rules”.²²

There are three key characteristics of Dworkinian principles. The first is that principles may be derived from sources beyond the immediate applicable precedents, such as a collateral area of law. Second, these principles have legal pedigree and can be relied upon to set aside a rule when the rule does not ‘fit’ the fabric of the legal system. Lastly, and most importantly, Dworkinian principles are already embedded in the law. When rules and principles conflict, one is held to be more important than the other in a particular context, but if a principle is defeated it is not necessarily excluded from our legal system. In that particular case the principle is simply outweighed by another rule or principle. According to Dworkin, rules, in contrast, do not have this characteristic: “If two rules conflict, one of them cannot be a valid rule”.²³

Dworkin would therefore contend that a rule may be set aside in favour of an existing principle (as was the case in Dworkin’s example of Riggs v Palmer²⁴). But this change does not operate with retrospective effect, as that principle already existed in the fabric of the law.

Dworkin contends that Hart’s rules of recognition cannot identify all the relevant principles. Nor can they determine the appropiate balance and priority to be struck between principles, since this requires a normative assessment of the competing principles derived from moral or political theories they represent. Hence, Dworkinian principles fall outside the rule-based concept of law that Legal Positivists defend.

In terms of retrospective common law, the key difference between Hart and Dworkin emerges when a principle provides the impetus to overturn a rule: in essence, when rules and principles conflict. Hart would contend that, since rules have the right pedigree, the law is represented by the rule, and to overturn the rule is to (unexpectedly) change the law. For Dworkin, the law includes the principle. Overturning the rule is not therefore a change in law, since the existing principle (with legal pedigree) has merely been prioritised over the rule.

As Dworkin’s ontology of law suggests, to view overturning precedent as merely rule (x) in precedent (a) being changed for rule (y) in case (b) may be too simplistic. There is, according to Dworkin, more than just rule (x) in the legal fabric. There may be other principles that conflict with the existing rule. This conflict may cause the ‘rumblings from Olympus’. If we view changes in common law through this lens we can see more clearly the role that Lord Devlin’s ‘unsettled weather ’ plays in predicting judge-made changes in the law.

Moreover, these other principles may point toward a new rule

^{that would better fit and integrate within the legal fabric. When a}

²² R Dworkin, ‘Is law a system of rules?’ in The Philosophy of Law (Oxford

^{University Press, 1977) 39.}

^{23 Ibid, 48.}

^{24 Riggs v Palmer 115 NY 506, 22 NE 188 (1889); see R Dworkin, The Philosophy}

^{of Law 44.}

court overturns a prior precedent, it is removing a legal rule that was inconsistent with higher principles already present in the law. The law as a whole therefore does not change when the prior precedent is overruled. The law is simply re-declared as if the new decision had always been the law but previous courts had failed to adequately apply the right principles in that situation. The newly stated legal rule is not

‘new’ because it was based upon ‘law’, including principles, that existed prior to the decision and pointed toward the current result. Thus, the common law would change with retrospective effect only if we adopted the positivist account of the law as the union of primary and secondary rules, and not Dworkin’s account which supplements the rules with principles.

2. Realism and Retrospectivity

The claim that when a court overturns a prior precedent the court enacts retrospective law not only rests on Positivist assumptions about rules, it also assumes a certain conception of judicial decision making. It assumes that judges perform a quasi-legislative function, that judges make the law and they may do so (in appropriate cases) by reference to policy factors and values.²⁵ This assumption underlies my explanation of retrospective common law: that “in order to achieve a fair and just result the court may find it necessary to depart from the statement of law in the precedent case”.

However, such a ‘Realist’ conception of the function of judicial decision making is contestable. Moreover, alternative accounts of the function of judicial decisions may yield different accounts of retrospectivity in the common law. In particular, according to the declaratory theory of judgment, the common law does not change with retrospective effect.

The declaratory theory of judgment also seeks to describe the phenomenon of judicial decision making. According to this theory, the entire common law already exists awaiting judicial declaration. As a consequence, “if the judges change their mind about a particular common law rule or principle, they are not changing the law”.²⁶ Rather, judges are declaring the true state of the common law.

Any theory of judgment is ultimately premised upon a particular judicial method, or a particular analytical approach to legal problems. The declaratory theory reflects a method that might be called legalism. This method considers that the “law consists largely of posited precepts laid down in legislation or leading judicial reasoning”²⁷ and that legal questions can be answered through “logical reasoning based on the text of the relevant law, to the exclusion of social, political and economic

²⁵ A Mason, ‘Legislative and judicial law-making: can we locate an

^{identifiable boundary?’ [2003] AdelLawRw 4; (2003) 24 Adelaide Law Review 15, 21.}

^{26 Ibid, 18.}

^{27 J Smillie, ‘Formalism, fairness and efficiency: civil adjudication in New}

^{Zealand’ [1996] New Zealand Law Review 254, 255.}

considerations”.²⁸ On this view, by adopting a legalist approach to judicial decisions, a Judge is able to work within a body of legal materials to find the correct rule and doctrine which are already part of the common law.

Since, according to this declaratory theory of judgment the court does not change the law, the common law does not operate with retrospective effect. This is because, when a court departs from a prior precedent, it is not doing so in an effort to arrive at a fair and just result. Rather, the court is removing erroneous decisions from the law. The erroneous decision (the prior precedent) was never the law per se, which means reliance on the precedent was not reliance on the law. Instead, the law has always been the rule that is later declared by the court. As Tipping J noted:²⁹

The traditional (declaratory) theory involved the proposition that in deciding what the law was the courts were deciding what it had always been...hence the courts could not, or at least should not, state that at the time of the relevant events the law was X but from the date of judgment onwards it was to be Y.

^{However, to some members of the judiciary, the declaratory theory of}

^{judgment is a fiction. As Lord Reid stated:30}

There was a time when it was thought almost indecent to suggest that judges make law – they only declare it. Those with a taste for fairy tales seem to have thought that in some Aladdin’s cave is hidden the common law in all its splendour, and that on a judge’s appointment there descends on him knowledge of the magic words ‘open sesame’. Bad decisions are given when judges muddle their passwords and the wrong doors open. But we do not believe in fairy tales any more.

Furthermore, with particular regard to retrospectivity, Lord Browne- Wilkinson has stated:³¹

The theoretical position has been that judges do not make the law; they discover and declare the law which is throughout the same. According to this theory, when an earlier decision is overruled the law is not changed; its true nature is disclosed, having existed in that form all along. This theoretical position is... a fairy tale in which no one believes... . But while the underlying myth has been rejected, its progeny – the retrospective effect of change made by judicial decision – remains.

The declaratory theory of judgment seems unpopular. Rather than exclusively confining their method to existing legal materials, most members of the judiciary seem to have “accepted direct responsibility for adapting and developing the law to meet current needs”.³²

An alternative theory of judicial decision making is often called

²⁸ Mason, 18.

^{29 Chamberlains, Tipping J at paragraph 130.}

^{30 Lord Reid, ‘The judge as lawmaker ’ (1972) The Journal of Public Teachers}

^{of Law 22.}

^{31 Kleinwort Benson Ltd v Lincoln City Council (No. 8) [1998] UKHL 38; [1999] 2 AC 349.}

^{32 Smillie, 258.}

legal realism. This theory holds that judges make the law and do so by reference to policy factors and values. The court’s function is still only quasi-legislative; it is still constrained to an extent by the existing law. Nonetheless, it is still accepted that the courts engage in some moral or political judgment when determining the law.

Just as the declaratory theory of judgment needs to be premised upon a judicial method, the legal realist approach also needs a method of addressing legal questions. After all, if judges are to depart from strictly legal materials and shape the common law towards ‘just and fair ’ outcomes, there needs to be a ‘philosophical touchstone’ or ‘analytical framework’ to guide the law toward that end.³³ As Sir Ivor Richardson has noted, for judges to make value judgments they need “a frame of reference against which to probe and test the economic, social and political questions involved”.³⁴ This is because ‘just’ and ‘fair ’ are rather vacuous terms without an analytical framework for substantiating their meaning.

For example, the ‘philosophical touchstone’ that informed Lord Atkin’s decision in Donoghue v Stevenson³⁵ was arguably Christianity’s neighbour principle. For Lord Cooke, the ‘philosophical touchstone’ was a form of fairness based on societal expectations: “that once the facts of any given case have been fully elicited most people would agree on the fair result”.³⁶ In contrast, Sir Ivor Richardson has been sceptical of the

‘fairness’ framework. Instead, Sir Ivor purported to adopt an objective approach to legal change, first, in the form of functional utilitarianism, then, later, in the form of economic analysis.³⁷

Regardless of the particular touchstone that judges adopt to give content to their notion of ‘justice’ or ‘fairness’, if judges are to depart from legal materials and engage in political and moral judgment, they need a method or framework to use. Under this general conception of judgment

– Legal Realism – a judge performs a slight legislative function by shaping the law in accordance with a particular philosophical touchstone. When a court is overturning a prior precedent, the court is changing the law in order to achieve ‘justice’. The adjudicative function of the judiciary means that the new statement of law is then applied as if it had always been the law. Therefore, if we view the function of judicial decision making through a legal realist lens, we can perceive the retrospective effects of a court overturning a prior precedent.

To summarise, so far we have addressed three layers of theoretical questions: (1) what is the law; (2) what is the proper analytical method for solving legal problems; and (3) what is the appropriate judicial

³³ Ibid.

^{34 Sir Ivor Richardson, ‘The role of the appellate judge’ (1981) 5 Otago Law}

^{Review 1, 9.}

^{35 Donoghue v Stevenson [1932] AC 562.}

^{36 Sir Robin Cooke, ‘Fairness’ (1989) 19 Victoria University of Wellington Law}

^{Review 421, 422.}

^{37 Smillie, 269.}

function? At face value, the two assumptions that my claim rests upon (grounded in the philospophies of Legal Positivism and Legal Realism) seem irreconcilable: if the law is the union of primary and secondary rules there may be no need to engage in moral or political judgement. However, there are at least three instances in which judges seem to resort to moral and political judgement even when operating within the framework of Positivism. The first two instances are when ‘rules run out’: either, in terms of their identification or their interpretation. The third instance is when the court opts to use its jurisdiction to depart from a prior precedent.³⁸ In doing so the court employs a secondary rule of legal change. However, the decision to depart from a prior precedent is not made with sole reference to existing law. The court, in deciding to change the common law, shapes it in accordance with a particular philosophical touchstone or concept of fairness.

Hence, the claim that a court enacts retrospective law when overturning a prior precedent seems largely to rest on two jurisprudential claims: first, that the law is the union of primary and secondary rules; second, that judges resort to moral and political judgement in deciding to depart from a prior primary rule and establish a new one. Elements of both Legal Positivism and Legal Realism therefore seem to be involved. When we view these theoretical claims alongside the court’s inherent adjudicative function, the problem of retrospective common law surfaces quite abruptly: the courts occasionally change the law (akin to legislation) whilst in the same move determining the legal consequences of past conduct (adjudication) so that the law applied is distinct from that previously promulgated. On this view, there is a problem of retrospectivity in the common law and any retrospective law abrogates the core principle that “the law should be such that people will be able to be guided by it”.³⁹

II. The Prospect of Prospectivity

Now that we have identified the problem of retrospectivity in the common law, we can assess the judicial device of prospective overruling as a possible solution. First, however, what prospective overruling entails needs to be outlined.

Prospective overruling can take various forms. The most common form is “pure” prospective overruling. In this form the effect of the court’s ruling applies exclusively to events that occur after the date of the decision.⁴⁰ All events occurring before that date are governed by the prior precedent, including the events that gave rise to the current proceeding. This pure approach is the orthodox form of prospective overruling in

³⁸ R v Chilton [2005] NZCA 295; [2006] 2 NZLR 341; Practice Statement (Judicial Practice) [1962]

^{1 WLR 1234.}

^{39 J Raz, ‘The rule of law and its virtue’ (1977) 93 Law Quarterly Review 196,}

^198.

^{40 Spectrum Plus, Lord Nicholls at paragraph 9.}

appellate courts in the United States⁴¹ and India.⁴² As will be discussed, it has been criticised for removing the incentive for litigants to challenge the prevailing view of the law, as, even if the litigant is successful, the change in law will not apply to their claim.

There are permutations of the pure form. The court may find that the circumstances require ‘selective’ prospective overruling so that the ruling is generally prospective in its effect although retrospective with regards to the parties to the current proceeding. This was, in essence, the approach taken by the Supreme Court of Ireland in Murphy v Attorney General.⁴³

The Supreme Court held certain tax provisions to be unconstitutional and void but restricted the restitutionary right to recover payments made by way of taxes to plaintiffs in the proceeding. It has been contended that this form of overruling, as will be later discussed, fails to treat like cases alike since it discriminates between the parties to the proceeding and others who have similar claims even though the relevant facts giving rise to their respective claims may be identical.

Alternatively, a court may find it appropriate to overrule a precedent with ‘delayed’ prospective effect so that the ruling will not take effect until a future date. For instance, the European Court of Justice delayed the effect of its ruling in Banca Popolare di Cremona v Agenzia Entrate Ufficio Cremona⁴⁴ to allow the State a reasonable opportunity to introduce new legislation.

In contrast to the Supreme Courts of the United States and India, and the European Court of Justice, prospective overruling has not found favour in Australia or Canada. In Re: Edward and Edward⁴⁵ the Saskatchewan Court of Appeal held prospective overruling to be “a dramatic deviation from the norm in both Canada and England”. The High Court of Australia in Ha v New South Wales⁴⁶ went as far as declaring that it had no power to overrule cases with only prospective effect.

The position in New Zealand and the United Kingdom is uncertain. The House of Lords in Spectrum Plus kept the door open to the concept of prospective overruling. However, the House of Lords refused to recognise any legitimate forms or instances of it. Tipping and Thomas JJ in Chamberlains entertained, in the abstract, the possibility of prospective overruling and Tipping J strongly favoured the selective form.

⁴¹ See Chevron Oil Co v Hudson [1971] USSC 180; (1971) 404 US 97; Linkletter v Walker (1965)

^{[1965] USSC 130; 381 US 618; Bingham v Miller (1848) 17 Ohio 445.}

^{42 See Golak Nath v State of Punjab [1967] INSC 45; (1967) 2 SCR 762; IndiaCement Ltd v State}

^{of Tamil Nadu (1990) 1 SCC 12.}

^{43 Murphy v Attorney General [1982] IR 241.}

^{44 Banca Popolare di Cremona v Agenzia Entrate Ufficio Cremona (Case C-}

^{475/03), 17 March 2005.}

^{45 In Re: Edward and Edward (1987) 39 DLR (4th) 654, 660.}

^{46 Ha v New South Wales [1997] HCA 34; (1997) 189 CLR 465.}

^{III. Objections to Prospective Overruling}

So far I have suggested that, to avoid the harm of retrospective common law-making, the court might overrule the precedent with prospective effect only. However, it has been contended that prospective overruling would create more problems than it solves.

In particular, there are three salient objections worthy of consideration. Prospective overruling can be objected to on (1) constitutional grounds, contending that prospective overruling is outside the legitimate function of a judiciary. A further concern is that pure prospective overruling may (2) remove the incentive to challenge the prevailing view of the law. Alternatively, it has been argued that (3) selective prospective overruling discriminates as it fails to treat like cases alike.

1. The Constitutional Objection

Lord Devlin considered prospective overruling to be “the Rubicon that divides the judicial and the legislative powers”.⁴⁷ More recently, Dr Juratowitch has argued that prospective overruling is inconsistent with the “proper scope of the judicial function in common law systems”.⁴⁸

This objection represents a cluster of claims that the judiciary, because of its institutional features, should not prospectively overrule. These institutional features include: insulation from political pressure and accountability, the limited perspective of viewing only the facts before the court, and the provision of relief that “looks backward and compensates, punishes or invalidates an act which has already been committed”.⁴⁹

Because of these institutional characteristics, it is argued that the courts should not prospectively overrule as the court would be making new law without applying it to the case before the court, and in doing so, the court is looking beyond the facts before the court which usurps the role of the legislature.

The essence of the objection is that the courts’ primary function is adjudicative so the courts should not perform legislative tasks. In response, I would question whether, those instances in which an appellate court overturns a prior precedent, it is always acting primarily as an adjudicator.

I have described the cause of retrospective common law as the tension that arises between the aims of achieving certainty and fairness in the law, compounded by the courts’ adjudicative function. However, there is a further hidden cause of retrospective common law that lies behind these competing aims, namely, the court’s legislative function.

The twin demands of producing certainty and fairness are placed on the common law because the courts not only decide the legal status of

⁴⁷ Lord Devlin, The Judge (Oxford University Press, 1979) 12.

^{48 B Juratowitch, ‘Questioning prospective overruling’ [2007] 3 New Zeland}

^{Law Review 393.}

^{49 A G L Nichol, ‘Prospective overruling: a new device for English Courts?’}

^{(1976) 39 Modern Law Review 548.}

past events (adjudicate) but also develop and determine the law as it will apply on future occasions (legislate). It is a corollary of the doctrine of precedent that each pronouncement of law is “descriptive of what the judges believe the law to be, and prescriptive of what it should be in the future”.⁵⁰ It cannot be said therefore that the court’s perspective is solely directed at the particular dispute before the court; rather, common law courts simultaneously prescribe rules for the future whilst settling disputes. This dual function is performed most obviously when a court overturns an established precedent.

The root of my disagreement with opponents, such as Dr Juratowitch, stems from disagreement about the extent to which appellate courts perform a legislative function by prescribing new common law rules. I contend that when an appellate court overturns a prior precedent the court is acting just as much as a legislator as it is an adjudicator. Accordingly, it may be appropriate for the consequences of the courts’ decision to be more akin to prospective legislation than retrospective adjudication.

In contrast, Dr Juratowitch suggests that the constitutional role of common law courts is to be “incidentally creative”⁵¹ and only develop the law “interstitially to meet the needs of the case before it”.⁵² This is an accurate description of the judicial role in most cases. However, when appellate courts are asked to overturn established law, the courts do not have the luxury of being only incidentally creative.

When the House of Lords in Spectrum Plus held Siebe Gorman to be wrong in law their Lordships were prescribing new rules for all bank debentures by changing the characteristics of fixed and floating charges. The House of Lords was thus performing the dual functions of adjudicating and legislating. The effect of the decision of the New Zealand Supreme Court in Chamberlains v Lai was also legislative as much as adjudicative. By holding that the particular barrister in the case was generally not immune from suits of negligence in relation to protected conduct the court enacted a significant change in the law of negligence that would have future effect.

Instances of overruling can be viewed from two perspectives: either the courts are usurping the legislative function, or, as a consequence of the doctrine of precedent, the courts are inevitably required to perform a legislative function. If we adopt the latter perspective, Juratowitch’s objection dissolves, as prospective overruling is no longer inconsistent with the common law judicial function since the court’s function is, at times, inevitably legislative.

Alternatively, if we view the overturning of prior precedents as improperly legislative, we still confront the reality that such overturning does occur. In practice, appellate courts are being legislative by prescribing

⁵⁰ Ibid, 542.

^{51 Juratowitch, 407.}

^{52 Ibid.}

new rules that have retrospective effect. Prospective overruling cannot be rejected on the grounds that it will make the courts legislative. The horse has already left the stables. To argue that we should close the stable gates, in the face of this reality, is simply not persuasive.

Once we recognise that courts are performing a legislative function by bluntly overturning prior precedents, regardless of our normative assessment of that state of affairs, we ought to recognise that it creates a problem in the form of retrospective common law. We cannot then reject the solution to this problem that can mitigate the harm that retrospective law-making can cause. Dr Juratowitch describes prospective overruling as “blatantly legislative”.⁵³ If it is, it is a blatantly legislative solution to a blatantly legislative problem.

2. The Pragmatic Objection (to pure prospective overruling)

Perhaps the most convincing argument against pure prospective overruling is the twin objection that prospective overruling leaves a successful litigant without a remedy and removes any incentive to challenge the prevailing view of the law.⁵⁴ The concern is that, although a court may agree with a litigant’s contention that the law ought to be changed, were the court to enact the change with prospective effect only, the litigant would not benefit from the change. In the immediate case, this would mean that a court may agree that the old rule is unfair, and should be changed henceforth, yet would still apply the old rule to the parties to the proceeding. From a broader perspective, this would mean that there would be no incentive to iniate proceedings to ask the court to depart from a prior precedent. That may in turn stifle development of the common law.

It is worth noting that this objection applies only to pure prospective overruling and it is this criticism that directs Tipping J in Chamberlains to favour the selective form. That form of prospective overruling avoids the above objection as it does confer the benefit of the change in law on the immediate parties even if not to other litigants in similar cases whose claims have already crystallised. However, as we shall see below, there are also compelling objections to this selective form.

I have three responses to the ‘lack of incentives’ objection. First of all, the harm of retrospective overruling itself also deters a court from making a change in the law and so deprives a worthy litigant of a remedy. When a court is considering whether or not to overturn a prior precedent, one consideration that can constrain it from introducing a new rule is the justified reliance placed upon the old rule by one of the parties. It may be that the retrospective change of the law upsets justified reliance on the old rule to the extent that the court declines to overturn the prior precedent. Strict adherence to that approach would prohibit overruling, which would again mean that litigants would have no incentive to initiate proceedings designed to seek such a change in the law.

⁵³ Ibid, 407.

^{54 Spectrum Plus, Lord Nicholls at paragraph 27.}

This matter is illustrated in the case of Geelong Harbour Trust v Gibbs, Bright & Co⁵⁵ in which the High Court of Australia was asked to re- interpret s 110 of the Geelong Harbour Trust Act 1951. The Court held that although the applicable precedent case of Townsville Harbour Trust v Scottish Shire Line⁵⁶ incorrectly interpreted the statutory provision, since the precedent had “stood without question for over fifty years”⁵⁷ and commerce had been conducted on that basis, the prior precedent ought not to be overturned. In this case therefore, a worthy litigant was not awarded a remedy and the law was not developed, due to concerns about the retrospective effect of overturning a prior precedent.

Whether pure prospective overruling is possible or not, there will be instances where worthy litigants are deprived of a remedy because of the courts’ unwillingness to upset justified reliance on an established authority. Therefore, the fact that a litigant may not receive the fruits of their labour is not a convincing argment against pure prospective overruling because the denial of such a remedy may have the same effect.

In addition, the deterrence from challenging the prevailing view of the law because of the risk of only prospective change should not be overstated. Prospective overruling is a discretionary judicial tool which the courts may apply in appropriate circumstances. From the position of litigants, it is only one possible outcome. The court may agree that the law should change but, in the circumstances, only with prospective effect. Such discretion is required to enable courts to balance the detriment of undercutting one litigant’s justified reliance on established law against the detriment of refusing a remedy to a litigant who has successfully convinced the court to depart from established law. So, it is also possible, in every such case, that the court will find change in law is required, but due to the weighing of the different detriments, the change ought only to be applied with retrospective effect.

Pure prospective overruling adds an extra variable into arguments about the overruling of prior precedents, but, since it is only one possible outcome alongside the possibility of retrospective overruling, it should not significantly deter litigants from challenging established law.

Even if, despite these two responses, pure prospective overruling still has the effect of deterring litigation and depriving litigants of proper remedies, such a concern would only apply significantly less to ‘institutional litigants’. An institutional litigant is interested in the long-term effect of the outcome of a particular dispute.⁵⁸ These ‘repeat players’ may be interested in later enforcement of the law or be generally concerned about the precedent value of the decision and how it will apply in future cases.

⁵⁵ Geelong Harbour Trust v Gibbs, Bright & Co [1970] HCA 16; (1970) 122 CLR 504.

^{56 Townsville Harbour Trust v Scottish Shire Line [1914] HCA 33; (1914) 18 CLR 306.}

^{57 Geelong Harbour Trust, 505.}

^{58 Nichol, 498.}

The lack of incentive objection counts much less with regards to these institutional litigants. If an institutional litigant is successful in convincing the court to overturn a prior precedent, although it may not immediately benefit from the outcome of the particular dispute, it will nonetheless benefit from the change in the long term. It follows that there would still be an incentive for these litigants to challenge the prevailing view of the law, even when the court has the option of prospectively overruling, due to the institutional litigant’s interest in the precedent value of the decision. There seems particular scope for prospective overruling when institutional litigants seek to overturn prior precedents.

3. Discrimination Objection (to selective prospective overruling)

The final objection criticises only the selective form of prospective overruling. Lord Nicholls in Spectrum Plus held this approach to be discriminatory because it treats persons in like cases differently.⁵⁹ The discrimination occurs when a court decides to overrule a prior precedent prospectively, except as regards the events that are subject to the current proceeding before the court. The current litigants may obtain their remedy, but no other litigants whose claim has already crystallised. The difficulty in this approach is that the events giving rise to the later litigants claim may be factually and temporally identical. The event that gave rise to the proceedings that prospectively overruled the old rule would be governed by the new rule. However, the old rule is applied to identical events that are not subject to the instant proceeding. Identical events, occurring at the same time, may therefore attract different legal consequences by virtue of the selective prospective overruling, and it may be largely a matter of chance that one event, and not its identical counterpart, came first before the courts.

It might be contended that enacting any legislative or quasi-legislative change necessarily entails a degree of discrimination, since any substantive change in law may make something unlawful today that was lawful yesterday. So, every legislative change may fail to treat like cases alike. However, a more accurate way to look at legislative change is to say that two events that occur at different times are not ‘like cases’ because a characteristic of an event is the time at which it occurred. Hence, legislation would not discriminate, as different legal consequences would be attached to different events. This is where the comparison between selective prospective overruling and legislation uncovers a crucial distinction. Unlike legislation, selective prospective overruling fails to treat cases alike that are really alike since two identical events that occur at the same time can attract different legal consequences.

Although Tipping J prefers the selective form, he notes that “exempting the immediate parties from any general non-retrospective operation is itself somewhat arbitrary”.⁶⁰ However, Tipping J dismisses this concern on the grounds that a “small degree of arbitrariness must be accepted in

⁵⁹ Spectrum Plus, Lord Nicholls at paragraph 27.

^{60 Chamberlains, Tipping J at paragraph 140.}

the interests of the law as a whole. To do otherwise ... would stultify the common law method...”.⁶¹ However, where Tipping J errs in his analysis is in assuming that prospective overruling uniquely stifles development of the common law. As discussed above, the constraining principle of justified reliance also stifles development. Since prospective overruling is a mere substitute for the constraining principle of justified reliance on a prior precedent, Tipping J is wrong in assuming that pure prospective overruling would further stifle development, and hence, there are no “interests of the law as a whole” to justify the arbitrary nature of selective prospective overruling.

What the discrimination objection successfully identifies is that the selective form of prospective overruling undermines a basic tenet of the common law: that like cases ought to be treated alike. Therefore, if a court were to prospectively overrule, the usual aproach should be to adopt the pure prospective form.

By means of summation, let us look at Geelong Harbour Trust from the perspective as if the High Court of Australia decided to overrule Townsville Harbour Trust with pure prospective effect. Under such an approach, all of the events prior to the decision of the Court, including the event of the damaged beacon that gave rise to the dispute, would be governed by the (incorrect) Townsville interpretation of s 110. Any event that occurred after the date of judgment would be governed by the new (correct) Geelong interpretion. In such a case, the law is able to develop and improve without upsetting any justified reliance on the law that was clearly stated. Moreover, the Harbour Board, as an institutional litigant, would also benefit from the prospective re-interpretation of the legislation although not benefitting in the instant case.

Conclusion

The twin demands of upholding certainty and fairness that are placed on common law courts force judges to sail between Scylla and Charybdis: trying to uphold the doctrine of precedent, which gives the common law certainty, whilst also trying to reach a fair and just outcome in the particular case. When the court is asked to overturn a prior precedent this task is made even more difficult. Prospective overruling charts a course that carefully balances the impetus to modify the common law, on one hand, with the presumption against enacting retrospective law on the other. This enables the court to be flexible, and to provide new remedies, whilst disrupting the rule of the law to the least degree possible. After all, “He that will not apply new remedies must expect new evils; for time is the greatest innovator”.⁶²

⁶¹ Ibid.

^{62 Sir Francis Bacon, Of Innovations, in B H Levy, ‘Realist jurisprudence and}

^{prospective overruling’ (1960) 109 University of Pennsylvania Law Review}

^1.