Canterbury Law Review
Suppose, for example, that the law were to command the permanent establishment of a previously decreed state religion. Could this be viewed as expressing the lawmaker's real will and intent? One should first ask himself whether a people may enact a law to the effect that certain tenets of faith and outward religious forms, once adopted, should remain forever; that is, may a people prevent itself (in future generations) from progressing in religious insight or from correcting what may be old errors?... [S]uch a law would be null and void in itself, because it runs counter to the destiny and ends of mankind.
In 1793, Kant demanded that permanently unalterable laws should not be allowed to hold back human progress. Today, the danger posed by such laws is minimal, but the problem lingers in a more subtle form. Legislation normally continues in force until it is amended or repealed, regardless of changes in social, technological or other conditions.  Over time, the meaning that it bore in the context in which it was enacted may come to produce anachronistic results, or fail to fulfill its purposes. If our legal system is to serve the values and aspirations of the community, then such Acts must be updated. Parliament has a clear mandate to undertake this task, but systematic legislative updating has proved impossible, because of its institutional limitations and the sheer number of statutes. Therefore, the courts have assumed an ambulatory role, updating statutes to reflect changed conditions. This raises a difficult issue: whether and how the judiciary can update legislation by changing its meaning. The idea of the courts altering Parliament's dictates may seem at odds with our constitutional structure, but it is increasingly advocated both in academic literature and case law. A complex theoretical debate has developed between 'dynamism' and 'ambulatory originalism'. Dynamists permit the meaning of a statutory text to evolve over time. Ambulatory originalists hold that any judicial updating must be consistent with the original meaning of the statute, but usually accept that what are effectively changes in meaning may sometimes be reconciled with it. This essay develops a model for allowing changes in meaning that is consistent with the political theory on which our constitution is grounded. It allows true changes in meaning to occur where the original meaning of a statute has become inconsistent with fundamental values. It endorses dynamism to this extent, but otherwise requires that changes in meaning are brought within an ambulatory originalist framework.
Three techniques for achieving this, the 'preexisting innovations', 'mobile words' and 'dynamic intention' doctrines, are discussed. While the courts' use of these techniques has sometimes been problematic, a principled reformulation of them suggests that they are powerful tools for judicial updating. Nonetheless, the limits of the model must be recognised. Ultimately, if law is to foster rather than restrain the aspirations of the community, Parliament must shoulder the greatest part of the burden of updating legislation.
Statutes are creatures of the context in which they are enacted. The draftsperson and legislator make law in response to the facts and circumstances of which they are aware. The future is murky: technological, social and other changes are difficult to predict and therefore likely to be considered only in the most general way. As time passes, their original intentions often become increasingly inappropriate. Expectations regarding the specific objects to which a statute applies may become outdated as analogous but unforeseen developments occur. More problematically, changed social circumstances may render the policy underlying an Act anachronistic. Yet the Act remains law. If laws are to remain effective and appropriate over time, then such legislation must be made to reflect modern conditions. The preferable solution would be for Parliament to revise legislation systematically, but its response to elderly statutes has proved haphazard and partial. The resource and time constraints upon the legislature and the executive agencies which support it operate to prevent continuous scrutiny of the statute book, and political legislation, rather than law reform, dominates the legislative agenda. Hence, an ambulatory role has fallen upon the courts. The legitimacy of ambulatory interpretation is now beyond question. The occasional dictum or decision suggests that the court must construe provisions exactly 'as they would have been the day after the statute was passed', so that changed conditions can never affect its application, but the modern view is that 'an enactment applies to circumstances as they arise'. There is a presumption that an Act should receive an ambulatory interpretation. Only legislation clearly intended to retain its precise original meaning is outside the ambit of the rule. This category is largely limited to legislation involving particular grievances or problems, such as Acts in the nature of contracts between the government and business interests. Section 6 of the Interpretation Act 1999 confirms this approach, though it is likely that the courts would have rejected the static view independently. It is highly impractical, as it necessitates amendment whenever circumstances change. Furthermore, it rests upon the questionable assumption that statutes refer to specific, existing things, rather than general concepts capable of encompassing future developments. The latter view seems more realistic, given that Parliament can be expected to anticipate the possibility, if not the content, of subsequent developments. Though the ambulatory approach is now orthodox, uncertainty remains regarding its scope. Unforeseen innovations, such as developments in technology, or novel social or commercial practices, pose little difficulty. These can be accommodated within existing legislation if its text and purpose allow. For example, in Victor Chandler International Ltd, it was held that information sent from computers to teletext screens was a 'document', despite the fact that teletext had been introduced subsequently to the enactment of the legislation. Its regulation fell within the purposes of the Act, and the word 'document' was capable of covering projections. Often, however, the problem is that the construction which was or would have been placed on the text in its original context no longer seems appropriate because of changes in social or other conditions. Questions have arisen as to whether the courts may alter the meaning of legislation to reflect such changes. Obviously, this will be of great importance to practitioners faced with archaic Acts. If the meaning of legislation can change in accordance with altered social conditions, then established interpretations cannot necessarily be relied upon before the courts. Unfortunately, there is substantial judicial disagreement over whether and how such changes can occur, as two recent decisions of the House of Lords illustrate. In Fitzpatrick v Sterling Housing Association, the House divided by three to two over whether the homosexual partner of a tenant could be a 'member of the original tenant's family' for the purposes of Schedule 1 of the Rent Act 1977. Subject to a residency requirement, such a person takes an assured tenancy on the death of the tenant, which protects them against eviction and allows them to pay fair, rather than market, rent. Fitzpatrick had shared a 'longstanding, close, loving and faithful, monogamous, homosexual relationship' with Thompson, a deceased tenant. In the contemporary world, such a relationship would probably be thought familial. However, the Schedule was drawn from the Rent and Mortgage Interest (Restrictions) Act 1920. Undoubtedly, at the time the provision was enacted, a homosexual partner would not have been considered to fall within it. Two of the judges considered that the intentions of the enacting legislature were determinative, so homosexual couples could not fall within the ambit of the Act. However, a bare majority took the opposite view. Since it was now recognised that committed homosexual relationships shared the essential characteristics of a family, they had come to be covered by the provision.
In Birmingham City Council v Oakley, the Lords considered whether s 79 (1) (a) of the Environmental Protection Act 1990, which empowered local authorities to issue abatement notices for statutory nuisances, enabled an order to be made requiring the shifting of toilet facilities so that the tenants of a flat would have access to a washbasin without passing through the kitchen. The section allowed an order to be made if premises were 'in such a state as to be prejudicial to health or a nuisance'. The issue was whether the word 'state' covered the layout of a house, where this created a risk that people would endanger their health, or was limited to the presence of particular things which were in themselves unhealthy or noxious. While the word 'state' was held to be capable of covering matters of layout, the majority adopted the narrower meaning. This followed from the context in which the provision was first enacted: in 1850, it was unlikely that Parliament intended to regulate layout. Lord Hoffman accepted that modern standards of hygiene supported the broader view, but claimed that the ambulatory approach did not allow the court to 'construe the language of an old statute to mean something conceptually different from what the contemporary evidence shows that Parliament must have intended'. The minority, by contrast, argued that the Act should be given a modern interpretation regardless of its legislative history, and that the original purposes of the Act extended to regulating layout, albeit the wide scope for such regulation had been recognised only in the light of modern developments in standards of hygiene.
These cases raise a difficult question: whether the words of an Act can change their meaning to accord with altered conditions. There are two theoretical approaches. The orthodox account, which might be called 'ambulatory originalism', sanctions updating only where this is consistent with the original meaning of the legislation, thus curtailing the scope for changes of meaning. On this view, the time of enactment is the 'canonical moment' at which a statute acquires its meaning. The range of meanings attributable to the text at that time and the purposes of the enacting legislature therefore govern statutory construction. The court must operate within these limits, though where there is genuine ambiguity, it should adopt that result which is more appropriate in modern conditions. Thus, neither linguistic nor contextual change can alter an Act's meaning. The courts could not, for example, revise the meaning of an old statute to give effect to the changes in the meaning of the word 'seduce'. Equally, an Act cannot be reinterpreted to serve new goals, however desirable these may be. An example is Wahrlich v Bate. In that case, McGechan J held that the finitude of resources was not a relevant consideration in exercising the discretion to grant legal aid under the Offenders Legal Aid Act 1954. While the Judge considered that allowing a balancing of the interests of the accused against the probable needs of others was an appropriate goal in modern conditions, the 1954 Act had been passed at a time when New Zealand was in a strong economic position and legal aid had not been perceived to require significant public expenditure. Since the legislature had regarded resource allocation concerns as irrelevant, the judge was constrained to ignore them. New policies could not be introduced into the Act without legislative sanction.
However, the ambulatory strain of originalism does not require legislation to be applied exactly as it would have been in its original context. Rather, it is assumed that the legislature enacts general concepts capable of application to future circumstances in a sensible way. This allows what are effectively changes in meaning to occur, but requires their reconciliation with the original meaning of the statute. Nonetheless, because this will not always be possible, it operates as an important and restrictive limit on the scope for judicial updating. Goodes v East Sussex County Council is a recent example. In that case, all of the Lords refused to extend the duty on Highway Authorities to 'maintain' highways to cover gritting or salting to prevent ice formation. Historically, the duty had not extended so far, and it was not for the courts to alter the law to achieve this result, even if it were desirable in the modern era.
Dynamic theories, by contrast, allow a more expansive role for ambulatory interpretation. The common thread underlying all such views is that the meaning of a provision may evolve overtime, in response either to linguistic changes or changes in the context in which it operates. There are many different models. Some theorists permit an altered interpretation where this is necessary to achieve the broad goals of an Act in the new circumstances, but most accept that an Act may be adapted to serve new purposes so that it continues to operate in an appropriate way over time. Eskridge, perhaps the leading contemporary theorist, adopts a pragmatic stance, in which interpretation is guided by multiple perspectives. The interpreter is enjoined to consider the text, the legislative history, the evolution of the statute (practice and precedents) and contemporary norms. These factors reciprocally interact, so each must be weighed and evaluated in the light of the others. He accepts that the text, being the most concrete factor, should generally be followed where its meaning is clear. Equally, legislative history is significant for the period immediately after an Act's passage. However, where the words themselves do not control the meaning of the Act, and the legislative history has lost its relevance because the assumptions upon which the original intentions of the legislature were based have been negated, a dynamic interpretation giving weight to contemporary values becomes appropriate. In such circumstances, the court is enjoined to seek the best substantive result allowed by a liberal reading of the text, guided but not limited by the history and evolution of the statute.
A clear rule defining the scope for changes of meaning has not yet emerged. The Interpretation Act does not resolve the issue. Section 6 might be thought to provide some support for ambulatory originalism, as its focus on the 'application' of statutes to new circumstances arguably suggests that a fixed original meaning operates upon arising circumstances. However, the significance of this should not be overstated. First, it can be plausibly argued that the application of an Act to new circumstances sometimes requires that it be given a different interpretation. The legislative history is ambiguous on this point. Furthermore, even if the section does support ambulatory originalism, it is doubtful whether it would prevent the courts from adopting a broader view if this were theoretically compelling. The focus of the 1999 Act seems more enabling than regulatory: the courts are unlikely to construe it as limiting the development of principles of interpretation which go beyond, but are not inconsistent with, those it espouses. The case law is also ambiguous. Courts throughout the common law world have traditionally adhered to ambulatory originalism. In Attorney-General for Ontario & Viking Houses v Regional Municipality of Peel, Laskin CJ considered that changes of meaning were impermissible as 'the courts cannot turn their role of construction into one of naked legislating, however well disposed they may be to solutions proposed for problems which arise under deficient legislation'. Equally, Lord Wilberforce's speech in the Royal College of Nursing Case suggests that the role of the courts is simply to apply the text and purposes of an Act to new situations. These sentiments are echoed in several New Zealand cases. For example, in McClenaghan, Chilwell J affirmed that 'the only way in which the law can be changed is by Parliament rethinking and changing its speech' . In Wahrlich, McGechan J accepted that the policy decisions involved in any change of meaning were best left to the legislature. However, while many of their decisions are still couched in the language of ambulatory originalism, the English courts are increasingly receptive to dynamic theories. Recent dicta in the House of Lords may signal the beginning of a radical change in approach. Some of the Law Lords appear to endorse a form of dynamic interpretation, which would at least allow the meaning of statutes to be revised in the light of changes in linguistic meaning or contemporary conditions, to ensure that their original purposes were fulfilled. Lord Steyn is perhaps the strongest advocate of such a model. Delivering the leading judgment in a recent case, he held that:
Although out of considerations of piety we frequently refer to the actual intention of the draftsman, the correct approach is simply to consider whether the words of the 1861 Act, considered in the light of contemporary knowledge, cover a recognisable psychiatric injury.
The lack of constraint upon the relevance of contemporary knowledge suggests that Lord Steyn views interpretation as a dynamic exercise. He has affirmed this view in other cases, and extra-judicially. In Birmingham, for example, he noted briefly that an 'appeal to Victorian social history, and legislative history going back 150 years, is in my view not appropriate to the context. The 1990 Act must be given a sensible interpretation in the modern world' .
Lord Steyn's is not a lone voice. In Fitzpatrick, obiter statements by two of the majority judges appeared cautiously to endorse dynamic interpretation. Though he clearly preferred more orthodox justifications for the decision, Lord Slynn noted that, 'an alternative question is whether the word 'family' in the 1920 Act has to be updated so as to be capable of including persons who today would be regarded as being a member of each other's family' . Updating could take the form of a change in the meaning of the word. Similarly, Lord Clyde suggested that while there was 'a relatively rare category of cases where Parliament intended the language to be fixed at the time when the original Act was passed', in most cases the courts may update statutes to accord with changes in meaning as a matter of language. As yet, there has been no express discussion of this approach in New Zealand. However, the comments of Potter and Robertson JJ in a recent adoption decision perhaps indicate support for it. They held that:
We have had some regard to what was said in Parliament at the time the statute was enacted, but must recognise the substantial move in community attitudes towards adoption since 1955. We do not consider that comments made at that time, which were directed to rather different considerations, are of particular value to us.
Their decision to give little weight to statements in the House suggests that the original meaning of the enactment is not determinative. It is not just that the Act is capable of meeting changes in social attitudes, but that the incorporation of such changes need not be closely tied to the original meaning of the Act.
Unfortunately, these dicta have not been accompanied by any substantial discussion of the issues of policy and principle raised by dynamic interpretation. It appears that the rejection of the static approach to interpretation has led to the wholesale rejection of originalism, rather than a principled reformulation of it. The courts' reasoning seems to be that if Acts can develop to apply in ways unforeseen by the drafters, then their task is simply to apply legislation sensibly to new circumstances. Before this conclusion can be reached, it is necessary to evaluate the relative merits of dynamism and ambulatory originalism.
Fundamentally, the debate between ambulatory originalism and dynamism concerns the proper role of the courts. The view that the courts have only a subsidiary legislative role and must defer to Parliament underlies ambulatory originalism. Since this means that they must give effect to legislative policy choices, it is arguable that ambiguities must be resolved in the light of a reconstruction of the policy that the legislature meant to enact. Against this, dynamists claim either that the court's subordinate role is consistent with dynamic interpretation, or that overall responsibility for the proper development of the law is shared between the courts and Parliament. The two theories are often seen as diametrically opposed, but it is submitted that in reality they reflect different facets of the judicial function. Thus, ambulatory originalism will ordinarily be the proper framework for interpretation, though a dynamic approach is warranted in cases where fundamental values are at stake.
The dispute ranges over several issues. The first is whether the idea of original meaning is tenable, given the nature of interpretation itself. Proponents of dynamic models claim that originalism rests upon an outdated 'speaker's meaning' theory of interpretation, which holds that texts should be interpreted by hypothesizing what the author intended, on the basis of the text and other conventional evidence. According to perspectival theories of interpretation, such as hermeneutics, this picture is radically incomplete. Interpretation is said to involve the creative fusion of the perspectives of the author and interpreter. The interpreter draws from the text that meaning which is thrown up by the frame of reference within which he or she views it. A frame of reference is a web of beliefs, norms and understandings, which shape the way texts are perceived. These influences mean that the interpreter cannot simply reconstruct the author's perspective. Rather, he or she must recreate the meanings of a text. Since this creative role is inherent in interpretation, legitimate interpretations of a text may alter overtime. Eskridge claims that this insight is applicable to statutory interpretation. Parliament does not establish the meaning of a text in enacting it: rather, it is constructed and reconstructed as the text is interpreted. This arguably necessitates a realignment of the relationship between author and interpreter. Since interpretation is viewed as a collaborative process, in which the interpreter has an equal role, the legislative author loses authority over the text. Therefore, the judiciary, as part of constructing its meaning, may refer to contemporary concerns and values.
However, even if this is the right picture of interpretation, it does not lead inexorably to dynamism. Originalism is not inconsistent with hermeneutics. Though a judge may have an equal role in constructing the meaning of legislation, the crucial issue is how that role should be exercised. One possibility would be for the judge to defer to his or her best reconstruction of the original intentions of Parliament. Such a reconstruction would necessarily be imperfect, but might nonetheless serve as a useful orienting principle. Consequently, while hermeneutics provides a philosophy of interpretation which allows dynamism, whether it is appropriate is ultimately a question of political theory.
Therefore, the basic issue is consistency with fundamental constitutional values. There are two general accounts of the content of these values. One holds that majoritarianism is the fundamental basis of the democratic polity; the other that substantive values may outweigh it in some circumstances. It is submitted that wholesale dynamism is justifiable on neither account, but the latter provides persuasive grounds for interpreting statutes dynamically where fundamental values are at stake. The traditional account is strictly majoritarian. It claims that, due to the prima facie equality and freedom of all citizens, the only way in which law can acquire legitimacy is by popular consent. Consent is achieved by incorporating the people
into the legislative process. Repeated elections ensure that the legislature is representative of and accountable to them, and that all citizens have an indirect role in lawmaking. Of course, consent also legitimates the wider political process, as it is maintained by the will of the people. This explains the quasi-legislative functions of the courts. However, the democratic accountability of the legislature means that legislation can be seen to express the popular will more directly. Consequently, it must override any tacit consentto lawmaking by appointed and unaccountable officials. Individual judges cannot substitute their personal views for that of the majority of representatives. Since the basis of our polity is majoritarian, the decisions of a representative majority must carry greater weight than those of an appointed judiciary.
As this entails that the judiciary must follow the directives of the legislature, it seems to rule out dynamic interpretation. Deference to legislative supremacy requires that the court attempt to produce an interpretation reflecting what Parliament wanted. Since lawmaking is a purposive endeavour, it is constrained by the original purposes of the legislation. The words of an Act cannot simply be twisted to further a policy which Parliament never approved. This means that although the courts necessarily exercise a creative, interstitial role in interpretation, they operate within the limits of plausible constructions of what the legislature meant. It follows that the courts cannot alter the meaning of legislation in response to changed circumstances. Except in the rare case where the common law is capable of developing to fill agap left by legislation, the effects of its original meaning will continue until it is amended or repealed. Proponents of dynamic interpretation have attempted to assail this conclusion, but their efforts are largely unpersuasive. One strategy is to claim that dynamic interpretation can be legitimated by reference to current legislative majorities, rather than the enacting legislature. For Eskridge, dynamic interpretation is consonant with majoritarian values because it gives effect to the preferences of the majority of legislators, even though these have not been encapsulated in legislation. Others develop the argument further: the acquiescence of the majority of legislators to the continued existence of an Act is treated as essentially canonical. It is said that since legislation is retained because it is thought to serve contemporary needs, the courts are empowered to adapt the law to reflect these needs. However, neither view is tenable. Both ignore the procedural safeguards inherent in our constitutional structure. To avoid uncertainty and ensure that measures receive proper debate and consultation, the legislature makes law only by following specific procedures. These important values should not give way to the preferences of the majority, particularly where it has chosen not to exercise its power to give them legal status. Furthermore, the latter approach rests upon a problematic inference from acquiescence to the acceptance of dynamic interpretation. It is more likely that restraint reflects acceptance of the original meaning of the legislation, given that conscious acquiescence will usually rest upon advice about the past effects of a law and that unconscious acquiescence does not indicate positive acceptance of a change in the law.
A second approach holds that dynamic interpretation is not inconsistent with majoritarianism because it operates beyond the limits of the specific policy choices made by the enacting legislature. The original meaning of legislation can be said to be linked to its original context, because legislative intentions rest upon assumptions about the context in which an Act will operate. The specific intentions of the enacting legislature regarding the working of legislation are shaped by the need to further its general goals without creating a disproportionate counter-mischief. These intentions are necessarily directed at the original context, since everything will depend upon the conditions which the Act regulates. Consequently, it is said that once the context alters their relevance diminishes. The courts must act as 'honest agents' of the legislature, exercising a subordinate policy-making role. They must adapt the meaning of the Act so that its general aims are achieved in the new context. For example, in enacting the Adoption Act 1955, Parliament may have intended to allow only married couples to adopt children jointly, because of an assumption that only marital relationships have a high degree of permanence and stability. As de facto relationships have become more prevalent, this assumption has been undermined. Arguably, the legislature lacked any intention that the law would require marriage as a precondition of joint adoption in modern conditions, as it never considered that they might arise. Therefore, it might be thought that the courts must determine whether the general purposes of the Act are now best served by recognising joint adoption by de facto couples.
This argument has a certain appeal, but it is submitted that it is flawed. The real complaint here is simply that the original meaning of an Act has become inappropriate. This is a matter for the legislature. The argument assumes, problematically, that a broad purpose requiring different meanings in different contexts can be abstracted from the actual intentions of the legislature. However, it can be inferred from the relatively determinate character of most legislation that Parliament normally intends to enact a specific rule. This specific purpose qualifies the general aims of the Act. It is this qualified policy which Parliament licenses. In the absence of a clear delegation of the power to alter the meaning of legislation, it cannot realistically be said that Parliament has endorsed the adaptation of the law to meet changed conditions. It has simply provided a fixed rule, reserving the power to alter it if amendment becomes desirable. Thus, the 'honest agent' metaphor seems flawed. The court does not work within the policy choices made by Parliament in interpreting legislation dynamically. Rather, it legislates to fill a gap caused by changed circumstances. This is an impermissible usurpation of the legislative function, which cannot be countenanced in a strictly majoritarian system.
Nonetheless, this argument encapsulates an important truth: carrying out the purposes of an Act may sometimes require changes in its interpretation. The problem is that this premise is applied to an over-general conception of legislative purpose. The question must be whether the specific purposes of the Act require a rule capable of development overtime. If so, changes in meaning may be justified, though not dynamic interpretation. Consequently, it seems that a strictly majoritarian conception of democracy is inconsistent with dynamism. However, an alternative conception of our constitutional structure is often thought to be more supportive of it. Its basis is that majoritarianism is not the sole basic value in our polity. Rather, fundamental human rights and other values may have equal or greater importance in a democracy. Indeed, such notions are implicitly connected with the ideas of freedom and equality which motivate democracy itself. As Lord Cooke has said, 'some rights are inherent and fundamental to democratic civilized society' . An ideal legislature would always respect these rights, but the legislature is not ideal. Democratic principles are imperfectly translated into practice by fallible human legislators. The role of the courts, then, should not be conceived as if Parliament lived up to its normative aspirations. Rather, democracy must be reconceptualised as a 'collaborative enterprise' between the courts and Parliament. The judiciary must act as guardians of fundamental values, sharing with Parliament the responsibility for developing our legal framework in accordance with them. This means that the courts should not be viewed as subordinate to a supreme legislature, but as their partner in a joint endeavour. This may lead the courts to, inter alia, disapply legislation infringing upon fundamental values where it results from legislative mistake or ignorance. It seems that mistake will be presumed in the absence of very clear evidence of an intention to bring about the infringement. This means that these values are effectively protected, since the political cost of enacting such legislation is likely to be unacceptably high. While the collaborative enterprise model is not yet orthodox, it is at the heart of a sea-change in the relationship between Parliament and the courts, the beginnings of which have been felt both in England and New Zealand. It is appealing because it promotes the substantive promises of democratic theory while preserving scope for majoritarian decision-making. Thus, it partially resolves the paradox that rigid adherence to majoritarianism may produce results inimical to the values that motivate democracy, without sacrificing its demotic character. Hence, it is submitted that it is the better view of the democratic polity Dynamists often appeal to similar theories to justify dynamic interpretation. Eskridge suggests that 'Countermajoritarian statutory interpretation may be normatively desirable if it contributes to the overall legitimacy of the political system'. The idea seems to be that the legislature and judiciary share equal responsibility for producing the best possible body of substantive law. This enterprise is hindered by the inability of the legislature to update legislation, so the courts must intervene by interpreting statutes dynamically. The fundamental democratic norm that laws should reflect the will of the current citizenry, and other important values such as equality before the law would otherwise be compromised, thus undermining the integrity of the legal system itself.
It is questionable whether the notion of a collaborative enterprise can stretch so far. Although the doctrine supports an extension of the judicial role, that role must still be defined in a principled fashion. This means that the values which are to be protected must be weighed against the interest in majoritarianism, which is itself a fundamental norm. The less serious the value infringed and the more the decision raises policy issues which are appropriately resolved democratically, the less willing the courts should be to intervene. This restricts the scope for dynamic interpretation greatly. Originalist interpretation will only rarely encroach upon fundamental values. In particular cases, outdated interpretations may come to undermine human rights or other values, but the most that can be said of the typical situation is that it works injustice on one of the parties. This provides a very weak ground for intervention, as a degree of injustice is tolerated in our legal system, in the interests of certainty and the rule of law. It might nonetheless be said that systemic values are endangered by originalism. The law may fail to reflect contemporary norms, and the consequent injustice may diminish the legitimacy ofthe system itself. However, while these are important interests, they lack force in this context. Originalist interpretation will only rarely cause unequivocally unjust results, and Parliament is able to ameliorate the effects of highly undesirable decisions. Furthermore, since systemic values are primarily aspirational, and the failure to achieve them does not directly harm citizens, it is arguable that their protection should be left to more traditional mechanisms. The interest in majoritarian decision-making must be balanced against these factors. It is particularly strong in cases where important policy issues are raised by a proposed change in the meaning of an Act. There are both theoretical and practical reasons for this. At a theoretical level, where there is legitimate scope for disagreement over an issue, the prima facie equality of all citizens in a democracy requires that it be settled in a democratic way, rather than imposed by appointed and unaccountable office-holders. Persuasion, rather than authority, must be the basis of policy-making. Of course, a degree of policy-making discretion falls upon the judiciary in interpreting statutes and developing the common law. Rigid adherence to formalist constitutional theory would produce an unconscionably inefficient system. However, outside the traditional domains of the common law, the courts should take a restrained approach, as our system confers responsibility for such laws upon Parliament. Ultimately, the question is one of degree: the larger the policy-content of a decision, the less appropriate is its judicial resolution.
The judicial updating of statutes will usually raise policy issues appropriate for legislative resolution, as there will be scope for different views on the desirability of, or at least the proper form of, updating. This flows in part from disagreement over means, and in part from the normative pluralism which holds in modern society: deep disagreements exist over values and what they require. While pluralists perhaps overstate the extent of disagreement on certain fundamental values, the lack of consensus is otherwise apparent. The decision in Oakley exemplifies the problem. The minority were concerned to uphold modern standards of hygiene by extending the scope of the provision to include matters of layout. There seems general agreement that this is an important value. However, itmustbe balanced against the interests of property-owners, who might be put to a great deal of time and expense in renovating their houses by the change. For this reason, it is arguable that layout should be regulated only as new houses are constructed. Because there is legitimate scope for differing views on the issue, it was appropriate to leave the matter for Parliament. Practical problems would also result from the negative systemic effects of the courts resolving policy questions. First, the value-laden nature of policy decisions exposes the judiciary to suggestions of partiality and politicization. Because of the non-majoritarian nature of judicial decision-making, this may bring into question the legitimacy and authority of their decisions, thus undermining the systemic values that dynamism purports to uphold. Secondly, dynamic interpretation may diminish the law reform role of Parliament. Because of the time and trouble involved in legislative change, interest groups may attempt to change the law through litigation and legislators may leave technical law reform to the judiciary. This is undesirable, as the parliamentary process, while slow-moving, is undoubtedly the preferable method for amending legislation. It allows wider consultation and debate, and is likely to produce better law because the legislature is not constrained by the existing statutory text. This goal is not unrealistic. While the systematic updating of statutes has proved unachievable, Parliament is capable of responding to anachronistic laws which are brought to its attention by interest groups or judicial decisions. The practice of the courts should support rather than undermine this process.
These factors indicate that dynamism cannot be elevated into a general rule. It is likely to raise significant policy issues and the grounds for ignoring the principle that such issues should be settled democratically are very weak. However, in particular cases where fundamental values are at stake, it is submitted that the courts are justified in taking a dynamic approach. The policy-making supremacy of the legislature is less important where the original meaning of a statute can simply no longer be countenanced. Hence, an old statute purporting to exclude the right to a trial in certain criminal cases should be reinterpreted if the words could be strained to bear a more modern meaning. Equally, this may explain the practice of construing criminal statutes narrowly when the rationale for the criminalisation of particular conduct has disappeared. The liberty of the individual should not be endangered by vagaries of the legislative process. Thus, the crime ofblasphemous libel, for example, might be narrowed almost to non-existence if brought before the contemporary courts. As this technique becomes more established, its ambit is likely to expand. For instance, a dynamic interpretation of the Rent Act 1977, at issue in Fitzpatrick, could arguably be justified on the basis that freedom from discrimination on the grounds of sexual orientation has become a fundamental value. However, the scrutiny of the courts is likely to be less rigorous in such areas, as the precise scope of the rights involved is still contested.
The above discussion indicates that ambulatory originalism will be the proper approach to interpretation in most cases. However, this does not preclude what are effectively changes in meaning from occurring overtime. Several techniques for reconciling effective changes in meaning with originalism have been developed. These require a difficult balancing exercise of the judges, who must permit changes in meaning where they are required to give effect to the purposes of the Act, without usurping the policy-making functions of the legislature. Unfortunately, the courts have not always succeeded in this task. Several of the techniques have been used with undue liberality or conservatism, and others fail to provide a coherent justification for changes in meaning. Consequently, this section attempts to provide a principled reformulation of the various doctrines in this area.
The common foundation of these techniques is a conceptual approach to the meaning of statutory texts. This is the view that, because legislation is intended to last over time, Parliament typically enacts general concepts under which existing objects happen to fall, rather than referring to those objects specifically. Dickerson has explained this approach by drawing upon the difference between connotative and denotative linguistic meaning. The connotative meaning of a word is its broad meaning, as distinct from its precise denotations, or those things which fall within its rubric. For example, in R v Misic, Anderson J, construing the meaning of the word 'document' in s 229A of the Crimes Act 1961, held that its connotative meaning was simply 'a material record of information'. The particular medium upon which the information was recorded was merely a denotation of that concept, so this feature could vary overtime. This conceptual approach is the rationale for extending legislation to cover unforeseen developments, but can be adapted to permit what are effectively changes in meaning. The conceptual approach can be applied in two ways. Courts sometimes establish the conceptual meaning of a word or phrase by providing criteria, or necessary conditions, for an object to fall within its scope. These conditions are defined in the light of the purposes of the Act and the language used. This approach has been applied in several recent cases, including two decisions of the House of Lords and one in the New Zealand Court of Appeal. The alternative method is more impressionistic. The courts simply apply the law to the object, considering whether it falls within a liberal construal of the text and purpose of the Act. Nothing turns upon the method chosen: a conceptual definition should produce the same outcome as simply applying the text and purpose of the Act, since these guide its creation. The first method will generally be preferable because it provides guidance for later courts, but in some cases such certainty may be thought undesirable, so a more intuitive approach will be appropriate.
There are three ways in which these criteria might be defined. Each supports a technique for recognising what are effectively changes in the meaning of statutory texts. The first method is to define the necessary conditions as objective criteria, or legal and actual characteristics of objects. This view underlies what might be called the 'preexisting innovations' technique. Secondly, they might be defined by both objective criteria and subjective criteria, which refer to variable attitudes, values and beliefs. This is the foundation of the 'mobile words' technique. Finally, according to the 'dynamic intention' doctrine, these subjective and objective criteria may be subject to a meta-criterion: the court's view of what an Act ought to mean in current circumstances. The scope of each of these techniques will be considered below.
When confronted with an unforeseen development, such as the introduction of a new technology, the courts typically employ objective criteria to determine whether it falls within the scope of existing legislation. Occasionally, this technique has been used to adapt an Act to cover something which already existed when the Act was passed, but would not then have been thought to fall within its terms. The judgment of Lord Slynn in Fitzpatrick stands out. He considered that:
The first question is what were the characteristics of the family in the 1920 Act and the second whether two same-sex partners can satisfy those characteristics so as today to fall within the word family.
According to Lord Slynn, these characteristics were:
That there should be a degree of mutual interdependence, of the sharing of lives, of caring and love, of commitment and support.
Since Fitzpatrick's relationship with the Thompson fell within these criteria, it was held to be familial. A striking feature of these criteria is that they are objective: they hold independently of beliefs or values about such relationships.
There is an obvious difficulty with this approach. If the original meaning of legislation is to be preserved, the conceptual meaning adopted must explain the way it operated in its original context. Putative objective criteria are prima facie erroneous if they would have produced a different result at the time of enactment to that which was or would actually have been applied then. This means that as long as the objective character of a pre-existing object does not alter, so that it is really a new development, it cannot straightforwardly come to fall within the ambit of an Act under this technique. The reason is that the objective criteria must be defined so as to have excluded it at the time of enactment. Fitzpatrick exemplifies the problem. When the Rent and Mortgage Interest (Restrictions) Act was passed in 1920, the English Parliament was aware that homosexuals sometimes cohabited in stable, loving and monogamous relationships. However, there seems no doubt that the Act would have been interpreted to exclude homosexual partners from being members of each other's family, as the goal of extending rights to homosexual couples cannot reasonably be ascribed to the 1920 legislature. Therefore, the conceptual meaning of the word 'family' must have excluded homosexual couples in its original context. If this meaning is defined using solely objective criteria, then a change in meaning is virtually impossible. Nothing has altered in the character of such relationships: they have always involved commitment, love and interdependence. This exposes the flaw in Lord Slynn's reasoning. His objective criteria are plainly inconsistent with the original meaning of the Act, as homosexual couples would always have fallen within them. Lord Hutton emphasised this in his dissent, holding that:
In 1920 the fact of homosexuals living together in permanent relationships was known to Parliament, and if a homosexual couple was not intended by Parliament to come within the term "family" at that date I do not consider that changed public attitudes towards homosexuality mean that a new state of affairs has come into existence which extends the meaning of the term.
Nonetheless, the conceptual reasoning which underlies the innovations technique may support a changed interpretation in some circumstances. The conceptual nature of legislative language means that the specific objects that the legislature expected to fall within a provision cannot always be determinative. Where the character of an object was virtually unknown when an Act was passed, the courts may conclude that it always fell within the scope of the concept established by the legislation, even though this was not recognised by the enacting Parliament or earlier courts. For example, if homosexuals had been thought incapable of having loving, monogamous and stable relationships in 1920, then it might be said that the enacting legislature simply laboured under a mistake about the scope of the concept of a 'family'. The purpose of the Act would always have included homosexual couples, even though the legislature and the courts at the time would not have recognised this. Unfortunately, on the facts in Fitzpatrick, this assumption appears incorrect. However, it is arguable that this principle explains the much-criticized decision of the House of Lords in Ireland. In that case, the Lords held unanimously that 'bodily harm' included mental injury for the purposes of acriminal statute enacted in 1861. It was acknowledged that mental injury was not thought to have a bodily component when the provision was enacted. Consequently, the enacting legislature would not have expected that the Act would create criminal liability for mental injury. However, it is arguable that the purpose of the Act was simply to render criminal all forms of harm to the body of another. The reason that the infliction of psychiatric illness would not have been thought to fall within that category at the time was that psychiatry was in its infancy. Mental illness was not thought to relate to the body at all, and therefore not to be straightforwardly linked to the actions of others. In the light of modern scientific knowledge, it can be plausibly argued that there is no essential difference between mental injury and other examples of bodily harm for the purposes of the Act. This indicates that 'bodily harm' always covered mental injury, albeit the 1861 courts would never have recognised this.
A more difficult question is whether the wilful blindness of the legislature could ever justify a similar approach. This is perhaps the underlying basis of Lord Slynn's judgment in Fitzpatrick. It might be said that the purpose of the provision, granting security to those in certain close, loving relationships with tenants, logically encompassed homosexual couples, but that the legislators failed to recognise this because of irrational prejudices and so were mistaken about the precise scope of that purpose. Since the court must give effect to the purpose of a statute even where this is contrary to the expectations of legislators, it is arguable that the beliefs of legislators that they were not extending rights to homosexual couples should be ignored. However, the problem with this argument is that there is a link between expectations, intentions and purposes. It cannot really be said that the purpose of an Act was to extend rights to homosexual couples if the legislature was aware of the relevant facts but nonetheless refused to do so. The complaint is rather that the purposes of the Act have become irrational by modern standards. Thus, wilful blindness arguments conceal an illegitimate form of dynamic interpretation. The 'pre-existing innovations' technique is applicable only in cases of true ignorance.
The limitations on the 'pre-existing innovations' technique can be avoided if the conceptual meaning of a provision is held to refer to some variable subjective criterion, like contemporary scientific understandings or ethical standards. As the standard shifts, the application of the word will change, though its conceptual meaning remains the same. Such provisions have been described as 'mobile phrases'. One recent example is Moreton v Police, which concerned s 312A of the Crimes Act 1961. The section states that a communication is not private for the purposes of the regime governing the use of intercepted communications as evidence in criminal proceedings if the participants ought reasonably to have expected its interception. William Young J accepted that:
Because the concept of reasonable expectations is embedded in the definition of what constitutes a "private communication," the definition appears to have an ambulatory application.
Thus, increased public awareness of the likelihood that cellular phone communications would be intercepted was relevant to whether a communication was private, although this was ultimately insufficient to render it public. This technique is widely accepted, but difficult issues have arisen relating to its conceptual basis and scope.
There are two alternative explanations of the doctrine, which have contrasting implications for the scope and character of mobility. One view is that the meaning of mobile words actually alters overtime, in accordance with changes in linguistic meaning. Alternatively, they might be viewed as having a fixed conceptual meaning, which has as one of its criteria a subjective standard relating to contemporary values or beliefs. The application of the concept would vary as the standard shifted, but its meaning would remain constant. It is submitted that the latter view is preferable, as it ensures that the application of the phrase is linked to the meaning and purpose of the legislation as it was enacted. Simply applying the contemporary linguistic meaning of a term would mean that linguistic shifts might expand or contract the ambit of an Act, thus frustrating its purposes, or extending the law beyond them. For example, the word 'obscene', if enacted tomorrow, would refer to sexual conduct which society considered too corrupting and depraved to be publicly exposed, but its linguistic meaning might expand overtime to include hate speech. If the meaning of the word in law were simply its current linguistic meaning, then hate speech would come to fall within the provision. This is unlikely to have been a goal of the enacting legislature, as it raises different policy issues from the regulation of sexual impropriety. It seems absurd that a mere change in the meaning of a word could bring about such a fundamental change in the law. Equally, if the meaning of the word narrowed to cover only certain types of sexual impropriety, though other types were still seen as equally deserving of restriction and condemnation, then the latter acts would cease to be regulated by the statute. It would therefore fail to achieve its purposes fully, for no reason other than a change in linguistic meaning. These problems are avoided if the second approach is adopted, as it entails that the dimensions along which a word is mobile may be relatively narrow. A provision can be treated as establishing both objective and subjective criteria for its application to particular types of conduct. For example, the word 'obscene' would require as an objective criterion that the conduct be sexual in nature. This would be supplemented by a subjective criterion, social attitudes to the conduct. The question would not be what society currently describes by using the word 'obscene', but rather what conduct of a sexual nature is now regarded as deserving of restriction and condemnation.
This has important consequences for the practice of interpretation. If the contemporary linguistic meaning of a word cannot simply be applied to mobile phrases, then a careful analysis of the scope for variability will be required in every case. The court must determine, in the light of the text and purpose of the Act, what particular values and beliefs are relevant to the provision's meaning.
Difficult questions have also arisen regarding the kind of words and phrases to which the doctrine applies. It most naturally applies to words or phrases which inherently refer to an external standard. Typically, words referring to community values or expectations, such as 'indecent', and 'exceptional depravity', will be mobile. Equally, read in context, scientific or medical terms may appear relative to their respective knowledge communities. For example, the phrase 'psychopathic personality' in a provision establishing grounds for the refusal of residency permits, seems naturally to refer to the current medical consensus regarding the scope of the condition. The penumbra of the concept of a value-laden term extends quite far. For example, the phrase 'by his conduct precluded', in s 23 of the Sale of Goods Act 1908, while not necessarily referring to the ethical standards of the business community, might readily be interpreted in this way. However, some words and phrases, such as 'spouse', 'state' and 'paint', appear to refer to facts, rather than beliefs or values. Such words ordinarily refer to a determinate set of objects, rather than varying in accordance with a subjective criterion. While their meaning may alter over time, this is a contingent change driven by changes in linguistic practice. Though such changes may reflect altered values or beliefs, they need not do so.
The use of such a determinate word arguably indicates that the legislature does not intend its application to alter over time, but nonetheless such words have sometimes been recognised as being mobile. The English Rent Acts provide a useful case study. Many commentators, and several judges, have suggested that the word 'family' in the legislation varies by reference to either linguistic usage or social attitudes. For example, in Fitzpatrick, Lord Clyde defined the family in terms of certain objective characteristics, but nonetheless considered that these should be applied so that the test for whether a person was a 'member of the tenant's family' was 'the ordinary, popular understanding of the word at the date when it falls to be applied'. Yet the word 'family' has a determinate meaning. It is perfectly consistent with its meaning for it to retain its original scope despite radical changes in social attitudes. Prima facie, the ordinary, determinate meaning of the word should be applied. Hence, there is a suspicion that this practice disguises a limited form of dynamic interpretation. However, it is submitted that it will sometimes be tenable to treat a determinate word as mobile, though this raises difficult conceptual issues which the proponents of the practice have failed to consider fully.
Two arguments are offered for treating determinate words as mobile. The first holds that words used in a popular sense vary in accordance with current linguistic usage. A word is used in its popular sense where it has no technical legal meaning, but rather bears the meaning that is assigned to it in everyday speech. The principal authority for this approach is Dyson Holdings v Fox, in which the English Court of Appeal held that the meaning of the word 'family' in the Rent and Mortgage Interest (Restrictions) Act 1920 had shifted as the linguistic meaning of the term had changed. Ms Fox had lived in a childless de facto relationship with a statutory tenant for 21 years. She claimed that she was entitled to a statutory tenancy under the Act, which conferred protection upon a member of a deceased tenant's family who was residing with the tenant at the date of his or her death. The question was whether de facto partners could be members of each other's family for the purposes of the Act. It is probable that, at least in the absence of children, this would not have been accepted in 1920. Indeed, this was the effect of an earlier decision of the Court of Appeal, so the court was unable to avoid this conclusion. Nonetheless, Ms Fox succeeded, as the court applied the contemporary meaning of the word. For James and Bridge LJJ, the reason was that if a word in a statute was used in its popular sense, the meaning of the provision would alter as the ordinary usage of the word changed overtime. James LJ considered that as 'the popular meaning of the word "family" is not fixed once and for all time', the meaning of the word in the statute was capable of change and had in fact altered to include de facto relationships . Thus, neither the fact that the meaning of the word as enacted had been narrower, nor that an earlier court had expressly determined that this narrower meaning applied to the Act, was relevant. The Act effectively received a new interpretation to render it consonant with contemporary understandings of the word 'family'. Lord Denning MR concurred for other reasons, though these broadly complemented the views of the majority.
This approach has been strongly criticized. It is not obvious that the mere fact that a word is used in its ordinary or popular sense alters its determinate character. This perhaps explains the significant disquiet which Dyson has engendered both in England and New Zealand. In Helby v Rafferty, for example, the English Court of Appeal, construing the same statute, doubted the approach, though they considered themselves bound to follow it. Stamp LJ asserted that:
Apart from authority, I would have taken the view that the language of a statute by whatever process you apply to its construction - whether you construe it in its natural or ordinary meaning or whether you construe it in a popular way or whether you construe it in what has sometimes been called "a legal way" (and I am not sure that I understand what the difference is) - cannot alter its meaning from time to time and that, in order to find out what Parliament intended by the statute, you must ascertain what the words of the statute meant when Parliament used those words.
Similar dicta are found in other cases. Furthermore, the failure of the House of Lords to deal substantively with this reasoning in Fitzpatrick at least suggests a preference for more orthodox modes of reasoning. The reception in New Zealand has been similar. Though it is mentioned in numerous cases, Dyson has been doubted, and the principle has not been applied in any reported case.
However, few of these criticisms address the reasoning in Dyson itself, perhaps because it is so briefly and obliquely expressed. The apparent basis of the decision is that if a word is used in its ordinary, popular sense, it is for the tribunal of fact to apply its intuitive understanding of the word to the particular circumstances. One of the consequences of this position is that the word is left undefined: the courts eschew the use of synonyms to clarify the meaning of popular words. From this, the majority appear to have concluded that the current understanding of the word must be applied, since this is the only intuitive understanding available to the decision-maker, who cannot step back into the past. Thus, the use of a popular word can be seen as an instruction to apply its current meaning. From this perspective, not to do so would be contrary to the original meaning of the legislation, as it would require the courts to shift the meaning of the word from its popular sense to a technical one.
However, this reasoning gives the popular usage of a word an overwhelming significance which it simply cannot bear. The fact that a word is used in an ordinary sense does not entail that it must be applied solely according to the intuitive understanding of the tribunal of fact. Parliament has simply used ordinary linguistic conventions to express its message. Conveying that message in changed circumstances may require some explanation of these conventions. The judiciary is empowered to provide such explanations by creating guidelines for the application of popular words. Dyson itself is a good example. Lord Denning ruled that, at least in circumstances analogous to those in the case, a person was a member of his or her de facto partner's family. These principles can be used to maintain the original popular meaning of a word over time. This means that the inference from popular usage to variability is unjustifiable. In reality, some other factor must be present before a word will be mobile. The obvious candidate, which will be considered below, is a supportive legislative purpose. Consequently, it appears that the Dyson principle rests upon a conceptual mistake. The significance of the distinction between popular and technical words has been greatly overstated, perhaps because many of the leading cases involving 'popular' meanings, such as Brutus v Cozens, concerned value-laden terms which required the tribunal of fact to apply contemporary standards. The two concepts may have been conflated. Regardless, the Dyson approach is untenable. It is not a legitimate form of ambulatory interpretation.
The second approach holds that a purposive reading of a statute may sometimes require that the meaning of a relatively determinate word be strained to allow some mobility. This is now the more widely subscribed view. It is often used to explain both Dyson and Fitzpatrick. For example, Zuckerman states that:
The enactment of the provision arose... from the recognition that security for the tenant was indivisible from security for his family. It follows then that the legislature's object was to create a test capable of adaptation to changes in family patterns.
The most obvious way to achieve this result would be to hold that, in addition to the objective criteria cited by Lord Slynn, the particular type of relationship must be socially accepted. Dicta from the majority judgments in Fitzpatrick provide some support for this approach. The clearest statement was Lord Nicholls' view that:
The underlying legislative purpose was to provide a secure home for those who share their lives together with the original tenant in the manner which characterises a family unit. This purpose would be at risk of being stultified if the courts could not have regard to changes in the way people live together and changes in the perception of relationships.
Lord Clyde agreed that allowing variability in accordance with changing perceptions was 'at least consistent with the evident purpose of the legislation'.
This approach has wide-ranging implications. It could allow substantial variability in almost all statutory tests. The doctrine might apply wherever adeterminate phrase can plausibly be linked to subjective beliefs orvalues. This is likely to be tenable for most forms of social legislation and certain other statutes, including those regulating commercial behaviour and resource management. Its significance is enhanced by the fact that subj ective criteria may allow great changes in the application of a provision over time. The reason is that the technique allows the operation of a statute in its original context to be explained without closely tying its conceptual meaning to the objective features which characterised that situation. Birmingham exemplifies the possibilities. Even if it is accepted that the phrase 'in such a state as to be prejudicial to health or a nuisance' originally excluded matters of layout, the provision might expand to incorporate them if it is treated as mobile. It could be viewed as requiring only that the state be some feature of the premises and that it be contrary to current standards of hygiene. This would explain its original operation, as in 1850 it was not recognised that layout could cause a danger to health, but would also allow the provision to expand now that this assumption has been negated. The breadth of such a reading obviously poses certain dangers, so it is important to consider the scope and legitimacy of this doctrine. It is submitted that in some circumstances it is appropriate for the courts to interpret a relatively determinate word as being mobile. Everything will depend upon the text and purpose of the legislation. The starting point must be that the ordinary meaning of the provision has a determinate character. This indicates that a fixed rule is intended, since Parliament could always have chosen to express itself using mobile words. However, for two reasons, this should not always be decisive. First, while determinate words do not have an inherent evaluative dimension, sometimes their meaning is nonetheless intimately related to the values and beliefs of the community. The meaning of a word like 'family' is, on a realistic view, shaped by common views of the morality and appropriateness of certain close relationships. In a sense, the word contains an element of moral approval. Given the intimate connection between values and language, it is obvious that excluded groups may come to stand in an analogous position to included groups as circumstances change. Parliament can be expected to be aware of this and therefore, because statutes are typically intended to last indefinitely, to use such words in a special sense, contemplating value-or-belief-driven changes over time. Secondly, where a word or phrase has a loose and impressionistic character, rather than a precise conceptual meaning, its use can be interpreted as a delegation to the courts of responsibility for producing a meaning which accords with the purpose of the legislation. This suggests that a determinate meaning is not intended, as it shows that Parliament did not have a precisely defined class of objects in mind in enacting the legislation.
Secondly, the Act's purposes must require mobility. It will often be plausible to construe the general purposes of an Act to support this, but it is submitted that this is insufficient. As was noted above, legislation is typically aimed at achieving a specific result in furtherance of a general aim. If the policy-making supremacy of the legislature is to be respected, then a word must be treated as mobile only where the specific purposes of legislation are broad enough to allow a degree of mobility overtime. Usually, the evidence will be ambiguous either way. Hansard may contain categorical statements about the effect of legislation, but these cannot be decisive, as legislators are unlikely to consider the possibility of contextual change in making such assertions. Of course, the view of the legislature should prevail in the rare case where the issue is discussed. However, in the majority of cases, the sole guides will be considerations of workability and the policies underlying the particular legislation. Where a provision is motivated by a social purpose that is only contingently linked to particular facts, so that it would acquire an arbitrary character if it were unable to adapt to changed conditions, it is sensible to assume that the purposes of the Act are broad enough to permit responses to such changes. This provides a rationale for Fitzpatrick and Dyson. The Rent Acts were intended to provide security to a class of people, not because of any quality necessarily limited to the class originally covered by the Acts, but because the people within that class were deemed to be entitled to protection due to their close relationships with tenants, and the perceived social merit of their claims. Since others might come to meet these criteria over time, it is arguable that the word 'family' must be treated as mobile for the Acts to operate effectively. Whether the specific purposes of an Act should be viewed in this way is ultimately a matter of impression. Consequently, values and views of the judicial function will shape the practice of the courts. However, some broad guidelines can be given. Where any change in application would involve significant policy issues, the proper inference is that the Act has a determinate purpose. Contentious political legislation, such as that regulating genetic modification, should not vary, as the political salience of the issues it raises means that even small alterations should receive Parliamentary sanction. Equally, the expansion of legislation creating an exception to a compelling right or interest should be left to Parliament. In Pearce, for example, the court was faced with an argument that s 80 of the Police and Criminal Evidence Act 1984 (UK),which provided that a husband or wife was compellable as a witness only in certain cases, should be updated to cover daughters and de facto partners. Kennedy LJ rightly rejected this contention. The provision constituted a limited exception to the general principle that, in criminal matters, witnesses should be required to testify so that the best evidence is heard. Consequently, a liberal construction of the provision was inappropriate. It was for Parliament to extend the scope of the exception if this was desirable. This approach is supported by s 6 of the Interpretation Act. While the provision is literally concerned with the application of legislation to new circumstances, the corollary of this is that enactments should be construed so as to be capable of applying to them. This underlies the rule that legislation should normally be given a conceptual reading, so that subsequent developments will fall within its scope. It is equally applicable here. The ultimate concern of the section is that an enactment's purposes should not be frustrated by contextual change. Hence, where a determinate interpretation would prevent an Act from operating effectively overtime, it is arguable that a degree of subjective variability must be introduced so that it continues to apply effectively to circumstances as they arise. The legislative history of the provision provides general support for this view. The Law Commission, for example, contemplated that the section would allow the judiciary to respond to changed social attitudes.
Against this, it might be said that legislation should be given a determinate construction in the absence of clear evidence that variability is intended. Two reasons might be cited. First, it is arguable that permitting mobility in the absence of such evidence amounts to an usurpation of the legislative function, as it allows the court to implement social policies which have not received legislative sanction. For example, many view the decision to extend rights to homosexual couples taken in Fitzpatrick to have been a matter for Parliament, rather than the courts, yet it seems to fall squarely within the expanded conception of the mobile words technique. However, it is submitted that this view rests upon an unduly narrow conception of the judicial role. Recognising a word as mobile does not devolve matters of high policy to the courts. Rather, it allows them to apply an existing policy to new circumstances. Where an object has become functionally analogous to those things traditionally covered by a statute, it falls straightforwardly within the scope of a statute's purposes. A low-level policy choice is required to determine whether the analogy holds, but this is simply an incidence of the normal judicial function of determining whether particular things fall within the purposes of an Act. Thus, in Fitzpatrick, the court's decision to adopt a modern approach to homosexual couples was permissible. As a firm consensus that homosexual relationships are socially acceptable has emerged, there is no longer a principled basis for distinguishing them from other familial relationships. Therefore, it seems strongly arguable that they have come to fall within the purposes of the Rent Act. Secondly, it is arguable that the doctrine of mobile words should be tightly constrained because of the desirability of certainty in statutory interpretation. In reality, however, the dangers arising from a broad application of the doctrine have been overstated. First, certainty is most important in cases where there are important reliance interests. For example, people should be able to rely on the established meanings of criminal provisions to regulate their conduct. Equally, uncertainty is inappropriate where substantial financial interests are at stake, as in the charities legislation. The proper inference in such cases is that the provision is immobile, as the need for certainty suggests that a fixed rule will best serve the purposes of the Act. In other cases, the danger of uncertainty is likely to be minimal, as the courts are likely to require compelling evidence before accepting that a provision has varied. This means that there is little scope for subtle and complex arguments that social change has undermined an established meaning. Consequently, the possibility of change should readily be predicted by counsel. While this does not avoid all uncertainty, any resulting inefficiency is likely to be less costly than using Parliamentary processes to make minor amendments to statutes, and outweighed by the importance of achieving the goals of legislation. Therefore, it is submitted that it will often be plausible to read determinate words as mobile. However, a cautious approach should be taken to determining the dimensions along which a phrase is variable. The courts must balance rule of law values, respect for the policy-making role of the legislature, and the need to apply legislative policy choices to changed circumstances. First, the conceptual meaning of aphrase should not permit its application to evolve beyond the broad range of meanings attributable to the text. It is what Parliament has chosen to enact, and that choice must be respected. Furthermore, the rule of law requires that citizens should be able to determine at least the possible constructions of the content of their obligations from the text of the law: the text of an Act may be strained, but should not positively mislead. This requirement can be met by incorporating objective criteria which narrow the scope for mobility into a phrase's conceptual meaning. For example, it seems necessary to require a legal marriage where the words 'husband and wife' are used.
Secondly, the altered application must fall within the scope of the policy decisions made by the enacting legislature. Consequently, the objective and subjective criteria for the application of a provision must ensure that new objects are functionally analogous to those objects caught by the provision in its original context: variations raising new policy issues should be left for legislative consideration. For example, the provision at issue in Birmingham should be mobile only on narrow grounds. The particular kinds of unhealthy things for which an abatement notice may be issued must be determined in accordance with contemporary standards of hygiene. However, objective criteria should be used to limit the scope of the section to things in themselves, rather than matters of layout. The different policy issues raised by the costs and difficulties involved in ordering changes to the layout of existing houses mean that this is not functionally analogous to ordering the removal of an unhealthy thing. Equally, the immense burdens which a requirement to remove ice and snow would place on public authorities mean that the duty to 'maintain the highway' in the Highways Act 1980 should not be construed to be capable of altering over time to encompass this.
A final technique reconciles dynamic interpretation with originalism on the basis of a legislative intention permitting the courts to update the meaning of an Act as circumstances change. This goes beyond the mobile words doctrine, in that the conceptual meaning of a word is simply that meaning which the judge considers appropriate in current conditions. Hence, the objective and subjective criteria for the application of a word may actually change over time. One uncontroversial example is the granting of broad discretionary powers to courts and administrative agencies, which can be seen as a delegation of the authority to forge rules that are appropriate to arising circumstances. For instance, in Director of Proceedings v Nursing Council, Baragwanath J held that the requirements of a provision empowering the Nursing Council to determine whether to hold hearings in public were to be determined by reference to contemporary values, rather than those current when the Act passed. Entirely new factors could become relevant. One such was the great importance now accorded to public accountability. This seems a sensible approach on the facts, though it must always be a matter of interpretation whether Parliament intended to fix certain mandatory relevant considerations for the lifetime of an Act.
While this technique provides the broadest scope for updating, it is of narrow application. The reason is that it is prima facie inconsistent with the constitutional role of the courts, as it empowers them to decide policy questions normally reserved for legislative consideration. Consequently, a dynamic intention should be inferred only where there is clear evidence of a delegation by Parliament. This might take the form of a statement in Hansard or the express conferral of a broad discretionary power. In rare cases, it might also be inferred from the context. For example, the difficulties with amending doubly entrenched legislation may indicate that Parliament intends it to be interpreted dynamically over time. However, there is no warrant for applying the doctrine wherever legislation is couched in imprecise terms and a fixed meaning might frustrate its purposes. Some argue that since this indicates that Parliament has delegated to the courts the responsibility for shaping the provision, it implies that the courts are permitted to change the meaning of legislation so that its purposes are achieved overtime. However, while this may indicate that some variability is intended, it suggests only that Parliament intends that its policy choices be applied to new circumstances, not that the courts are empowered to decide major policy issues. Consequently, such cases are more appropriately dealt with under the mobile words doctrine, as this permits variability without sanctioning the introduction of new policies by the courts. This means that a dynamic intention will rarely be found, as there is usually little evidence regarding how the legislature intended an Act to be treated by the courts.
It is submitted that these doctrines provide the preferable approach to judicial updating in ordinary cases. Nonetheless, one plausible criticism of them is that they are poor cousins of dynamism, which achieve similar results but require complex and sometimes artificial analyses of legislative purpose. However, it is submitted that this criticism is flawed. The virtue of ambulatory originalism is that its principles direct the courts to operate within the existing policy decisions of the legislature. Thus, it enables them to avoid encroaching upon its policy-making function. Of course, judicial minds differ on the legitimate scope for judicial policy-making . An activist judiciary might employ ambulatory originalism to justify making important social policy decisions. However, it is surely better to begin from a constrained position and accept that constitutional purity will occasionally be smudged, than to endorse a practice which seems logically to entail substantial judicial policy making.
The effort to free society from the will of bygone Parliaments by judicial means has been only partially successful. The reach of the ambulatory approach is broad but finite. In the most important cases, where fundamental values are at stake, there is scope for a dynamic form of ambulatory interpretation. Otherwise, however, it is necessary to reconcile effective changes in meaning with originalism. While this restricts the scope for ambulatory interpretation, there is good cause for optimism. The three techniques discussed explain many of the controversial cases in the English courts which have extended the ambit of the ambulatory approach. Consequently, it appears that legislation will often be capable of adaptation to meet changed social and other conditions. Nonetheless, the constraints upon the model must be recognised. Each of the three ambulatory originalist techniques has a limited scope. The pre-existing innovations doctrine applies only where Parliament legislated in ignorance of the character of an object. The dynamic intention technique requires very clear evidence of a delegation of power to the courts. The mobile words doctrine has broader possibilities, but is tenable only where the language of a provision is expressly value-laden or capable of being read in this way in the light of the purposes of the Act. Their limitations share a common theme: significant policy decisions should normally be the domain of the legislature. While it is submitted that this strikes an appropriate constitutional balance, it nonetheless creates an unfortunate lacuna. The irony is that Parliament is likely to amend legislation which has come to infringe fundamental values because of the political cost of allowing such results to continue, but may ignore less consequential statutes which are equally anachronistic. However, it is submitted that the proper solution is not judicial action. The involvement of the courts in essentially legislative decisions is undesirable both in theory and practice. Parliament must make such choices. It is incumbent upon lawyers, interest groups, legislators, ministries and the Law Commission to monitorthe statute book for deficiencies. Ultimately, this reflects the reality of the collaborative enterprise: even if sovereignty is shared, the legislative branch must continue to pull its weight.
[*] The author is clerk to the Honourable Justice McGrath in the Court of Appeal, Wellington. This paper was awarded the Canterbury Law Review Prize for the best undergraduate Honours paper completed in 2002.
 I Kant, 'On the old saw: that may be right in theory but it won't work in practice' in S M Cahn (ed), Classics of Modern Political Theory (1997) 564.
 A Samuels, 'Current Developments' (2001) 22 Statute Law Review 154; S G Requadt, 'Worlds Apart on Words Apart: Reexamining the Doctrine of Shifting Purpose in Statutory Interpretation' (1993) 51 University of Toronto Faculty of Law Review 331.
 R Dickerson, The Interpretation and Application of Statutes (1975) 126; J F Burrows, Statute Law in New Zealand (2nd ed, 1999) 221; FAR Bennion, Statutory Interpretation: a Code (3rd ed, 1997) 687; J G Miller, 'Evolutionary Statutory Interpretation: Mr Justice Scalia meets Darwin' (2000) 20 Pace Law Review 409.
 W N Eskridge, Dynamic Statutory Interpretation (1994) 50; Bennion, Statutory Interpretation: a Code, above n 3, 686; Burrows, Statute Law in New Zealand, above n 3, 219; A Samuels, 'Current Developments' (2001) 22 Statute Law Review 154.
 In this essay, 'object' is used as a neutral term covering all those things which might fall within a statutory provision.
 Burrows, Statute Law in New Zealand, above n 3, 221-3; D J Hurst, 'The Problem of the Elderly Statute' (1983) 3 Legal Studies 21, 32.
 Hurst, 'The Problem of the Elderly Statute', above n 6, 21-2. Bennion, Statutory Interpretation: a Code, above n 3, 686-7. The recent fourth edition of Bennion has also been reviewed, but does not differ substantially from the third edition for the purposes of this essay. See FAR Bennion, Statutory Interpretation: a Code (4th ed, 2002).
 L C Marshall, '"Let Congress do it": the case for an absolute rule of statutory stare decisis' (1989) 88 Michigan Law Review 177, 201.
 See The Law Commission, NZLC R 17: A New Interpretation Act: To Avoid Prolixity and Tautology (1990) 37; M H Redish and T T Chung, 'Democratic Theory and the Legislative Process: Mourning the Death of Originalism in Statutory Interpretation' (1994) 68 Tulane Law Review 803, 836; Bennion, Statutory Interpretation: a Code, above n 3, 687.
 Bennion, Statutory Interpretation: a Code, above n 3, 687; The Law Commission, NZLC R 63: Annual Report 2000 (2000) 1.
 Sharpe v Wakefield  UKLawRpKQB 200; (1889) 22 QBD 239, 242 (Lord Esher MR). See also The Longford  UKLawRpPro 42; (1889) 14 PD 34. These dicta can be viewed as supporting an ambulatory, rather than a static, form of originalism: R Sullivan, Driedger on the Construction of Statutes (3rd ed, 1994) 144; Burrows, Statute Law in New Zealand, above n 3, 232. The cases are ambiguous as in neither case was an ambulatory interpretation tenable:  UKLawRpKQB 200; (1889) 22 QBD 239, 242;  UKLawRpPro 42; (1889) 14 PD 34, 36; noted in J F Burrows, 'The Problem of Time in Statutory Interpretation,'  New Zealand Law Journal, 253, 254.
 R Sullivan, above, n 11, 143-4; McCulloch v Anderson  NZLR 130, 132.
 The Interpretation Act 1999 (NZ) s 6. See also s 5 (d) Acts Interpretation Act 1924 (repealed.)
 While s 6 of the Interpretation Act is expressed as an absolute rule that 'an enactment applies to circumstances as they arise', the provisions of the Act give way where the context indicates that they are inapplicable (s 4 of the Act.)
 Rv Ireland  UKHL 34;  AC 147, 158 (Lord Steyn).
 See, eg, Bogoch Seed Co. Ltd. v Canadian Pacific Railway  SCR 247. In the case, an Act implementing an agreement between the Canadian government and the Railway with regard to commercially grown rape seed was held to be inapplicable to commercially grown grain, which was virtually unknown when the legislation was enacted. The scope of the Act was held to be limited to its initial subject matter: Sullivan, above n 11, 144-5; Bennion, Statutory Interpretation: a Code, above n 3, 700.
 See The Law Commission, NZLC R1 7: A New Interpretation Act: To Avoid Prolixity and Tautology (1990) for a discussion of the provision.
 Burrows, Statute Law in New Zealand, above n 3, 231.
 R G Glover, 'The Statutes Statute'  CanterLawRw 5; (1986) 3 Canterbury Law Review 61, 64.
 Bennion, Statutory Interpretation: a Code, above n 3, 686-7. For some judges, the rejection of the static approach also follows from the decision to limit the doctrine of contemporanea exposito to very old statutes. See, eg, Lord Clyde's speech in Fitzpatrick  1 AC 27, 49. However, this is not strictly relevant. Contemporanea exposito is an evidential doctrine, which requires the courts to apply the interpretation placed upon an Act by contemporary interpreters. The rule now applies only where it is impossible for the judge to interpret an Act properly because of its age. Logically, this restriction in the scope of the rule is irrelevant to the substantive static doctrine, as the court is capable of reconstructing the original meaning of statutes in cases where contemporary interpretations are not determinative. See Hurst, 'The Problem of the Elderly Statute', above n 6, 24, 30.
 For example, the development of the telephone (Attorney-General v Edison Telephone Co  UKLawRpKQB 112; (1880) 6 QBD 244), and of new methods of creating embryos (R (Quintaville) v Secretary of State for Health  EWHC 29;  QB 628).
 An example would be a new way of structuring a tax minimisation scheme.
 The orthodox view is that words cannot be read into a statute to allow updating.
 Burrows, Statute Law in New Zealand, above n 3, 221-3; R v Misic  NZCA 128;  3 NZLR 1;Royal College of Nursing v Department of Health and Social Security  UKHL 10;  AC 800. Subsequently to the writing of this paper, the House of Lords affirmed these principles in R v Secretary of State for Health, ex parte Quintaville  UKHL 13.
 Victor Chandler International Ltd v Customs and Excise Commissioners  1 WLR 1297. The case concerned s 9 (1) (b) of the Betting and Gaming Duties Act 1981 (UK). The question was whether advertising by teletext was covered by the prohibition on advertising by offshore bookmakers contained in the section.
 Ibid 1305 (Richard Scott V-C). See also Corkery v Carpenter  1 KB 102; Attorney-General v Edison Telephone Co  UKLawRpKQB 112; (1880) 6 QBD 244.
 No bright line can be drawn between unforeseen innovations and changes affecting the way existing objects are viewed. Whether a relevant feature of an object has changed so that it is an unforeseen innovation can be determined only in the light of the text and purpose of the relevant Act.
 Hurst, 'The Problem of the Elderly Statute', above n 6, 42; A Allott, The Limits of Law (1980) 108. Stare decisis will obviously play an attenuated role if the meaning of an enactment can change. Hence, though important, it is not the primary focus of this essay.
  1 AC 27.
 The Lords unanimously rejected a claim to the full protection of a statutory tenancy as the 'spouse' of the deceased, holding that, in context, the term implied a heterosexual relationship.
 See para 3 (1), schedule 1 of the Rent Act 1977.
  1 AC 27, 32 (Slynn L).
 On one view, the ordinary meaning of the word 'family' requires the presence of children. However, the particular statutory context suggested that a widower could claim as a family member, so this meaning was displaced: J Evans, Statutory Interpretation: Problems of Communication (1988) 144-5. Arguably, homosexuals would now fall within this broader meaning of the word family.
 There is a presumption that the reenactment of a phrase from an earlier Act is not intended to change the law. See P V Baker, 'Notes' (1979) 95 Law Quarterly Review 101, 102. Many of the recent examples of ambulatory interpretation, including both Fitzpatrick and Oakley, have concerned reenacted provisions. Baker suggests that it should not be assumed that the legislature intends to retain antiquated meanings in reenacting phrases. However, at least where the legislators have been advised as to the current law, it seems clear that this is the only possible result. Reference to legislative history can be used to determine whether advice was actually given. In the modern era, it is likely to be given in almost all cases. If legislators occasionally mistake the effect of that advice, the solution lies in education rather than judicial activism.
  1 AC 27, 35 (Slynn L); 64 (Hutton L).
  1 AC 27, 64 (Hutton L); 68 (Hobhouse L), albeit the latter took the starting point for the inquiry to be 1988, when the legislature reviewed the legislation but did not extend rights to homosexual couples.
 Lord Slynn of Hadley, Lord Nicholls of Birkenhead and Lord Clyde.
 See  1 AC 27, 38 (Slynn L); 51 Lord Clyde. Lord Nicholls broadly endorsed this view, but also referred to the 1988 amendment regarding de facto heterosexual couples as indicating an intention to remove the morality of the relationship from the inquiry: at 45-6.
  UKHL 59;  1 AC 617, 621 (Slynn L).
 Ibid 629 (Hoffman L); 627 (Slynn L).
 Ibid 628.
 Ibid 631 (Hoffman L).
 Ibid 628 (Steyn L).
 Ibid 634-5 (Clyde L).
 Sullivan, above n 11, 143-4; McClenaghan v BNZ  2 NZLR 528, 532 (Chilwell J); Birmingham  UKHL 59;  1 AC 617, 631 (Hoffman L).
 R Dworkin, Laws Empire (1986) 316.
 Allott, above n 28, 106; Dickerson, above n 3, 129; D J Farber, 'Statutory Interpretation and Legislative Supremacy,' (1989) 78 Georgetown Law Journal 281, 287.
 The justification for this is pragmatic: if a sensible interpretation is tenable, the courts should prefer it to an anachronistic one. Subsequent conditions therefore act as a kind of tie-breaker between equally plausible interpretations. See Redish and Chung, above n 9, 867.
 R Dworkin, Laws Empire (1986) 315-6; R Dickerson, above n 3, 126; P Mitchell, 'Just Do It! Eskridge's Critical Pragmatic Approach to Statutory Interpretation' (1996) 41 McGill Law Journal 713, 720.
 Bennion gives several examples of words used in statutes which now bear dramatically different meaning, including 'sad', 'seduce' and 'trespasser': Bennion, Statutory Interpretation: a Code, above n 3, 700.
 See R v Big M Drug Mart (1985) 18 DLR (4th) 321, 353 (Dickson J).
  NZHC 1015;  3 NZLR 97.
 Ibid 111.
 Ibid 109-10.
 Ibid 111.
 Sullivan, above n11, 144; Dickerson, above n 3, 129
 Sullivan, above n11, 143; Dickerson, above n 3, 128.
 Sullivan, above n 11, 142-3; Hurst, 'The Problem of the Elderly Statute', above n 6, 33.
 See Birmingham  UKHL 59;  1 AC 617, 631 (Hoffman L); Burrows, Statute Law in New Zealand, above n 3, 227; P A Côte, The Interpretation of Legislation in Canada (1984) 207.
  UKHL 34;  3 All ER 603.
 The duty is imposed by s 41 (1) of the Highways Act 1980.
 Ibid 612 (Hoffman L).
 Ibid 614.
 T A Aleinikoff, 'Updating Statutory Interpretation' (1988) 87 Michigan Law Review 20, 49. Eskridge, Dynamic Statutory Interpretation, above n 4, 49; T W Jones, 'Textualism and Legal Process Theory: Alternative Approaches to Statutory Interpretation' (2000) 26 Journal of Legislation 45, 49.
 See J Bell and G Engle, Cross: Statutory Interpretation (1995) 52; Editorial, 'How Does the Law Change' (1999) 20 Property Law Bulletin 41; Bennion, Statutory Interpretation: a Code, above n 3, 694. Bennion does not expressly address the point in either the third or fourth edition of his work. However, it seems clear that his insistence that the words of an Act be interpreted so as to give effect to the original intention of the framers in a changed context would allow changes in meaning to occur. An obvious example is his treatment of Williams and Glyn's Bank v Boland  UKHL 4;  2 All ER 408, which he characterises as involving a changed interpretation to give effect to original intentions: Bennion, Statutory Interpretation: a Code, above n 3, 694-5.
 Eskridge, Dynamic Statutory Interpretation, above n 4, 49; Aleinikoff, above n 64, 49; Requadt, above n 2, 333. This view seems to accord more with logic than the narrower views of the English theorists. The extent to which Bennion and the authors of Cross restrict dynamic interpretation is questionable: it is likely that the 'sensible' achievement of a purpose in a new context will mask the introduction of new purposes and social policies.
 See, for comparison, the weakly dynamic model of Bell and Engle, in which change in permissable only to serve the original purposes of the legislation and Aleinikoff's strongly dynamic view, on which all legislation would simply be interpreted as if it had been enacted yesterday, so that the original context and legislative history would indicate only the general area in which a statute should operate: Bell and Engle, above n 65, 52-4; Aleinikoff, above n 64, 49, 55.
 Eskridge, Dynamic Statutory Interpretation, above n 4, 55. A recent brief restatement of this view is found in W N Eskridge, P P Frickey and E Garrett, Legislation and Statutory Interpretation (2000).
 Eskridge, Dynamic Statutory Interpretation above n 4, 56.
 Ibid 55-6.
 W N Eskridge, 'Dynamic Statutory Interpretation' (1987) 135 University of Pennsylvania Law Review 1479, 1488.
 Eskridge, Dynamic Statutory Interpretation, above n 4, 52.
 S Fruehwald, 'Pragmatic Textualism and the Limits of Statutory Interpretation: Dale v Boy Scouts of America' (2000) 35 Wake Forest Law Review 973, 978. Eskridge, Dynamic Statutory Interpretation, above n 4; Eskridge, University of Pennsylvania Law Review, above n 71, 1550-1.
 Eskridge, Dynamic Statutory Interpretation, above n 4, 121, 126-8. See the discussion below, in the context of the mobile words doctrine.
 See The Law Commission, NZLC R 17: A New Interpretation Act: To Avoid Prolixity and Tautology (1990) 35-7; NZPD Vol 565, 1997: 5910. In the second reading debate, the Minister of Justice simply contrasted a 'dynamic' and a 'historical' approach to interpretation. There was no discussion of the ambit of this 'dynamic' approach, so it is unclear whether a true dynamic approach was endorsed.
  2 SCR 1134, 1139. See also Commonwealth v Welosky 276 Mass 398, 177 NE 656 (1931) 659 (Rugg CJ). While the judge failed to apply the ambulatory approach correctly, he nonetheless usefully stated the principles underlying ambulatory originalism.
 Royal College of Nursing v Department of Health and Social Security  UKHL 10;  AC 800, 822.
 McClenaghan v BNZ  2 NZLR 528, 532.
 Warlich v Bate  NZHC 1015;  3 NZLR 97, 110-11. See also Quilter v A-G (1996) 14 FRNZ 430 (Kerr J).
 See Birmingham  UKHL 59;  1 AC 617 (Hoffman L); Goodes  UKHL 34;  3 All ER 603; Bellinger v Bellinger  EWCA Civ 1140;  Fam 150 (Robert Walker LJ). See also the recent judgment of Lord Bingham in Quintaville  UKHL 13.
 Rv Ireland  UKHL 34;  AC 147, 158. This dictum is perhaps capable of a narrower interpretation. Lord Steyn may be saying that the concept of bodily harm is mobile by reference to contemporary scientific knowledge. However, this would be consistent with legislative intention, so the comment seems to endorse the broader view.
 Cf J F Burrows, ‘The Changing Approach to the Interpretation of Statutes' in D Carter and M Palmer (eds), Roles and Perspectives in the Law: Essays in Honour of Sir Ivor Richardson (2002) 750.
 RvK  3 All ER 897, 909; McCartan Turkington Breen (A Firm) v Times Newspapers Ltd  UKHL 57;  4 All ER 913, 926-7. See also his recent judgment in Quintaville  UKHL 13.
 See J Steyn, 'Pepper v Hart: A Reexamination'  Oxford Journal of Legal Studies 59, 68.
  UKHL 59;  1 AC 617, 628.
  1 AC 27, 35.
 Ibid 39.
 Ibid 49.
 Ibid 50. Furthermore, in the court below in Fitzpatrick  Ch 304, 336, Ward LJ (dissenting) noted Dworkin's dynamic model with apparent approval.
 T v J  2 NZLR 236.
 Ibid 237. Strangely, the judges cite Burrows' view that Acts may be applied to new facts as they arise for the proposition. His statement appears to be an endorsement only of ambulatory originalism.
 For example, Lord Steyn's comment in Ireland  UKHL 34;  AC 147, 158 that the court must determine whether it 'must search for the historical or original meaning of a statute or whether it is free to apply the current meaning of the statute to present day conditions', suggests that the court is confronted by a stark choice between static originalism and dynamism.
 Redish and Chung, above n 9, 805; Farber, above n 47, 281.
 Dickerson, above n 3, 13-5.
 Redish and Chung, above n 9, 853-4.
 Not all the arguments can be reviewed here. In particular, one alternative justification for dynamic interpretation should be noted. It is based upon the rule of law value that citizens should be able to discern the law by reading the statutory text. Since they are likely to read legislation in accordance with its current meaning, it is said that this should be adopted by the courts: Sullivan, above n11, 139. While there is something to be said for this view, it is out of step with the interpretive practices of the courts. Our system privileges the furtherance of legislative intent over this aspect of the rule of law: strained meanings may be adopted in appropriate cases.
 R Dworkin, Laws Empire (1986) 50, 315; Redish and Chung, above n 9, 805. At one point, Burrows appears to endorse the speaker's meaning view, noting that 'a statute is a form of communication and it is always relevant what the communicator, in this case Parliament, intended to say'. See Burrows, Statute Law in New Zealand, above n 3, 227.
 See Eskridge, Dynamic Statutory Interpretation, above n 4, 58; Eskridge, University of Pennsylvania Law Review, above n 71, 1482; Aleinikoff, above n 64, 57.
 Eskridge, Dynamic Statutory Interpretation, above n 4, 60.
 Ibid 58.
 Ibid 60.
 Eskridge, University of Pennsylvania Law Review, above n 71.
 Eskridge, Dynamic Statutory Interpretation, above n 4, 60-2; Aleinikoff, above n 64, 57.
 Aleinikoff, above n 64, 57.
 Cf Requadt, above n 2, 345, who appears to conclude that the originalist inquiry would be pointless if it were not objective. This seems beyond reason: even if a reconstruction is not entirely precise, the policy-making role of the courts could still be limited by reference to it.
 Eskridge, Dynamic Statutory Interpretation, above n 4. See also Aleinikoff, above n 64, 58. Nor, indeed, is a hermeneutic approach necessary to dynamic interpretation. One might maintain that dynamic interpretation is justified on policy grounds even though it is not a true incidence of interpretation but rather a quasi-legislative function.
 Requadt, above n 2, 344; Maltz, below n 119, 781; Redish and Chung, above n 9, 833.
 Fruehwald, above n 73, 983. See Kant, above n 1, for an excellent discussion of the relationship between freedom, equality and representative decision-making. See also J J Rousseau, 'Of the Social Contract or Principles of Political Right'; J Locke, 'Second Treatise of Government' in S M Cahn (ed), Classics of Modern Political Theory (1997).
 Requadt, above n 2, 344; Redish and Chung, above n 9, 841.
 Requadt, above n 2, 356.
 Joseph would probably object to this conception as being ahistorical. See P A Joseph, 'Demise of Ultra Vires - Judicial Review in the New Zealand Courts'  Public Law 354, 357. However, while it is an incorrect explanation as a matter of history, this view may nonetheless serve a useful role as a contemporary justification for the continued role of the courts.
 Hints at this view are found in S Fruehwald, 'Pragmatic Textualism and the Limits of Statutory Interpretation: Dale v Boy Scouts of America (2000) 35 Wake Forest Law Review 973, 983; Dickerson, above n 3, 7; Farber, above n 47, 292.
 Allott, above n 28, 106.
 Dickerson, above n 3, 13.
 Farber, above n 47, 290; Dickerson, above n 3, 7-8; N S Zeppos, 'Judicial Candor and Statutory Interpretation' (1989) 78 Georgetown Law Journal 353, 380; Mitchell, above n 49, 321.
 Therefore, the mere fact of judicial creativity does not imply that statutes may be interpreted dynamically. For examples of this argument, see Requadt, above n 2, 343-4; A C Spiropoulos, 'A Defence of Substantive Canons of Construction'  Utah Law Review 915, 931. Legislative intent still operates as a constraint on the possible meanings of a statutory text: Mitchell, above n 49, 721; Dickerson, above n 3, 13; Farber, above n 47. Only if it were a fictional judicial construct could it be argued that the meaning of a statutory provision is simply the product of judicial decision, and is therefore open to judicial revision. This, however, is contrary to all experience: Dickerson, above n 3, 79.
 Dickerson, above n 3, 15; Farber, above n 47. For a loyal affirmation of this principle, see Dupont Steels Ltd v Bellows  1 All ER 529, 545 (Diplock L).
 E A Maltz, 'Rhetoric and Reality in the Theory of Statutory Interpretation: Underenforcement, Overenforcement and the Problem of Legislative Supremacy' (1991) 71 Boston University Law Review 767, 786. This is likely to be rare, as the common law is largely constrained to its traditional areas by incrementalist assumptions and deference to the policy-making superiority of the legislature: 791. Some academics support a broader role for the development of the common law by analogy with statute. See J Beatson, 'The Role of Statute in the Development of Common Law Doctrine' (2001) 117 Law Quarterly Review 247. The argument is, however, beyond the scope of this essay.
 Eskridge, Dynamic Statutory Interpretation, above n 4, 154. See also Requadt, above n 2, 355.
 Requadt, above n 2, 350; Dworkin, above n 46, 349.
 Dickerson, above n 3, 9. Zeppos, above n 116, 382; Redish and Chung, above n 9, 876.
 See Redish and Chung, above n 9, 845.
 Requadt, above n 2, 345; Eskridge, Dynamic Statutory Interpretation, above n 4.
 Eskridge, Dynamic Statutory Interpretation, above n 4, 126-30.
 Eskridge, Dynamic Statutory Interpretation, above n 4, 126-30; Miller, above n 3, 409; Farber, above n 47, 409.
 In fact, the text of the Act probably precludes such an interpretation, although not all judges have taken that view. See Re R (adoption) (1997-1999) 17 FRNZ 498; cf Re TW (Adoption) (1998) 17 FRNZ 349.
 Parliament does not normally enact general statements of purpose, leaving the courts to establish rules that are appropriate to the circumstances.
 Dickerson, above n 3, 99.
 As Lord Wilberforce has said, 'There is one course which the courts cannot take, under the law of this country: they cannot fill gaps; they cannot begin by asking the question 'What would Parliament have done in this case - not being one in contemplation - if the facts had been before it?', attempt themselves to answer this question, if the answer is not to be found in the terms of the Act itself’: Royal College of Nursing v Department of Health and Social Security  UKHL 10;  AC 800, 822.
 P A Joseph, 'Constitutional Law'  New Zealand Law Review 449, 454; Eskridge, Dynamic Statutory Interpretation, above n 4, 156-7. Kant's notion of the protected sphere of freedom held by citizens in a republic can be seen as a precursor of this view. See Kant, above n 1.
 R (Daly) v Secretary of State for the Home Department  UKHL 26;  2 WLR 1622, 1636.
 This is, of course, subject to the question of any justified limitations upon rights.
 P A Joseph, Constitutional and Administrative Law in New Zealand (2nd ed, 2001) 506; 'Constitutional Law'  New Zealand Law Review 449, 456; adopted in R v Pora  NZCA 403;  2 NZLR 37, 75 (Thomas J).
 Joseph, Constitutional and Administrative Law in New Zealand above n 134, 510-11; 'Constitutional Law'  New Zealand Law Review 449, 452, 454, elaborating creatively upon R v Pora  NZCA 403;  2 NZLR 37.
 Joseph, 'Constitutional Law', above n 131, 449, 454.
 Joseph, Constitutional and Administrative Law in New Zealand, above n 134, 510-11.
 Fruehwald above n 73, 978, wrongly describes such views as antidemocratic, on the basis that they take a consequentialist approach which views procedural requirements as irrelevant. In New Zealand at least, the collaborative enterprise is concerned to balance consequentialist and procedural values. It does not wholly endorse consequentialist policy-making. Hence, the criticisms of the collaborative enterprise as resting upon an antidemocratic model miss their mark. See Redish and Chung, above n 9.
 For an early statement of this view, see A De Tocqueville, 'Democracy in America' S M Cahn (ed), Classics of Modern Political Theory (1997) 725-6.
 Space constraints prevent the point from being fully argued here.
 Eskridge, Dynamic Statutory Interpretation, above n 4, 157. In an earlier article, he makes a similar comment: 'A judiciary that always faithfully interprets ... statutes according to the original intent is not contributing to the legitimacy of our government as a whole. A model of dynamic statutory interpretation would contribute somewhat to the judiciary's positive role in our representative democracy, because it would help to ameliorate some of the biases attendant to the legislative process': above n 71, 1523-4.
 W N Eskridge, 'Spinning Legislative Supremacy' (1989) 78 Georgetown Law Journal 319, 331; Zeppos, above n 116, 382; Mitchell, above n 49, 730; D A Forkner and K M Kostka, 'Unanimously Weaving a Tangled Web: Walters, Robinson, Title VII and the Need for Holistic Statutory Interpretation,' (1999) 36 Harvard Journal on Legislation 161, 203.
 Eskridge, Dynamic Statutory Interpretation, above n 4; Redish and Chung, above n 9, 853-4.
 Requadt, above n 2, 335-6.
 Zeppos hints at this in 'Judicial Candor and Statutory Interpretation' (1989) 78 Georgetown Law Journal 353, 380
 Eskridge, Dynamic Statutory Interpretation, above n 4, 154-5; Eskridge, University of Pennsylvania Law Review, above n 71, 1523-4; Aleinikoff, above n 64, 57-9.
 cf R v Pora  NZCA 403;  2 NZLR 37, 75 (Thomas J).
 This formulation owes much to the law of judicial review as stated by Joseph, Constitutional and Administrative Law in New Zealand, above n 134.
 It might be thought that the application of the notion of a collaborative enterprise in these circumstances is wildly implausible. Joseph, one of the leading proponents of the doctrine, claims that it justifies the disapplication of statutes only where values of the greatest importance are at stake: Constitutional and Administrative Law in New Zealand, above n 134, 511. However, because ofthe less radical character of reinterpretation, it is at least arguable that a less compelling justification may suffice.
 In Birmingham, for example, the result was perhaps unfair to Oakley, but can hardly be said to have infringed a fundamental right.
 Marshall, above n 8, 235.
 Eskridge, Dynamic Statutory Interpretation, above n 4, 154-7; Aleinikoff, above n 64, 57-9.
 This ability is somewhat limited by the rule that the fruits of victory in litigation should not be disturbed by legislation, but the legitimacy of the rule is questionable and it is possible that Parliamentary practice will alter. See Joseph, 'Constitutional Law', above n 131, 449, 57.
 Obviously, such decisions may produce injustice in individual cases, but no direct harm is caused by the failure to achieve systemic values per se.
 Kant, above n 1, 560; Redish and Chung, above n 9, 833; Fruehwald, above n 73, 983.
 Zeppos, above n 116, 380; Farber, above n 47, 286.
 Cf Maltz, above n 119, 791.
 See Joseph, Constitutional and Administrative Law in New Zealand, n 134, 742, again in the context of judicial review. See also Sullivan, above n 11, 149.
 Redish and Chung, above n 9, 841.
 This was acknowledged by the majority:  UKHL 59;  1 AC 617, 623 (Slynn L).
 Ibid 632 (Hoffman L). See also I Loveland, 'Washbasins and Nuisances' (2001) 151 New Law Journal 701.
 Goodes  UKHL 34;  3 All ER 603, 614 (Hoffman L) is another example. The court held that the decision to impose an onerous duty to remove ice and snow on local bodies was for Parliament because of its far-ranging implications.
 Spiropolous, above n 117, 918; Zeppos, above n 116, 379; Redish and Chung, above n 9, 857.
 Marshall, above n 8, 210; Jones, above n 64.
 Burrows, Statute Law in New Zealand, above n 3.
 Jones, above n 64, 67.
 J C Nagle, 'Review Essay: Newt Gingrich, Dynamic Statutory Interpreter' (1995) 143 University of Pennsylvania Law Review 2209, 2245; Jones, above n 64, 59; Marshall, above n 8. An example closer to home is the amendment to the Commerce Act following the decision in Foodstuffs (Auckland) Ltd. v Commerce Commission  1 NZLR 533.
 Indeed, it is strongly arguable that they are so weak that even matters raising only low- level policy decisions should be left to Parliament, particularly given the interest in certainty. Regardless, such cases will often be resolved using the techniques for recognising effective changes in meaning, which are discussed below.
 Cf Joseph, Constitutional and Administrative Law in New Zealand, above n 134.
 The extent to which straining is permissible is debated. The Bill of Rights Act 1990 allows the adoption of a meaning consistent with the rights therein only if the words are capable of bearing it. By contrast, the Human Rights Act 1998 (UK) allows the courts to adopt highly strained meanings, for example by reading words into provisions. The latter view seems more consistent with the theory underlying the collaborative enterprise, though it may require legislation for the New Zealand courts to adopt it. That issue, however, is beyond the scope of this essay.
 Crimes Act 1961 (NZ) s 123.
 Since writing this passage, the decision of the English Court of Appeal in Ghaidan v Godin-Mendoza  EWCA Civ 1533 has been released. In the case, the court accepted that the Human Rights Act 1998 empowered the courts to reinterpret the phrase 'as husband and wife' in the schedule to the Rent Act 1977 to read 'as if husband and wife', thus extending its meaning to cover homosexual couples. The reason was that the statute infringed the freedom from discrimination provisions in the Human Rights Act. While the decision has been criticized (See Editorial, 'Same Sex Partners and Succession to Rent Act Tenancies' (2002) 9 Housing Law Monitor 12) it is submitted that, at least with regard to the flexible word 'family', a similar result would be tenable in the absence of legislation.
 Lord Hoffman employed this distinction in Birmingham City Council v Oakley  UKHL 59;  1 AC 617, 631. Anderson J adopted it in R v Misic  NZCA 128;  3 NZLR 1, 8. See also Dickerson, above n 3, 129-30; Burrows, Statute Law in New Zealand, above n 3, 221.
 See also Dickerson, above n 3, 128-30; Cf Commonwealth v Welosky 276 Mass 398, 177 NE 656, 659 (1931).
  NZCA 128;  3 NZLR 1, 8.
 See Dickerson, above n 3; Burrows, Statute Law in New Zealand, above n 3.
 Dickerson suggests that the connotative meaning of a word should be defined by reference to those characteristics of the things originally caught by the word which bear a functional relationship to the purposes of the Act: above n 3, 130. While this is undoubtedly correct, it is important not to lose sight of the statutory text itself.
 See Birmingham  UKHL 59;  1 AC 617, 631 (Hoffman L); Fitzpatrick  AC 27, 35 (Slynn L); 51 (Clyde L).
 R v Misic  NZCA 128;  3 NZLR 1, 8 (Anderson J).
 See, eg, R (Quintaville) v Secretary of State for Health  EWHC 29;  QB 628, 639 (Lord Phillips of Worth Metravers MR); Royal College of Nursing v Department of Health and Social Security  UKHL 10;  AC 800, 822 (Wilberforce L); Burrows, Statute Law in New Zealand, above n 3, 221.
 For example, where the ambit of a provision is very uncertain, the court may wish to give a result without prejudicing its ability to hear further argument on its precise scope.
 See, eg, R v Misic  NZCA 128;  3 NZLR 1; R (Quintaville) v Secretary of State for Health  EWHC 29;  QB 628; Royal College of Nursing v Department of Health and Social Security  UKHL 10;  AC 800.
  AC 27, 35.
 Ibid 38.
 Lord Clyde also focused upon stability and love as indicia of a familial bond: 51. However, he noted that these criteria should be applied so that the meaning of the word accorded with linguistic usage, which suggests that the case was seen as involving 'mobile words'. Lord Nicholls also seemed to endorse this approach at one point: 45. Ultimately, the judgments are not entirely consistent.
 To be fair, Lord Slynn later referred to the need to have regard to changes in attitudes: at 38, which may bring his view within the mobile words doctrine, but he failed to link these to the criteria which he had established.
 Normally, this will take the form of an actual change. For example, the sudden overabundance of a plant might cause it to become a noxious weed within the meaning of certain statutes. However, a change in legal status may sometimes alter the character of an object. This may explain Williams and Glyn's Bank v Boland  UKHL 4;  2 All ER 408. In that case, the House of Lords held that a wife with an equitable interest could be in 'actual occupation' of her matrimonial home, despite contrary authority indicating that her interest was merely the shadow of her husband's. The decision is trenchantly criticized by Hurst in 'Palm Trees in the House of Lords - Some Further Thoughts on Boland's Case' (1983) Statute Law Review 142. However, it is explicable on the basis that the conceptual meaning of 'actual occupation' simply requires that a person occupies land according to some lawful right. Over time, a wife's equitable right to occupy a home to which she had contributed came to be seen as independent in law from that of her husband. Consequently, the character of a wife's interest changed, so that it fell within the provision.
  1 AC 27, 64 (Hutton L); 68 (Hobhouse L).
  1 AC 27, 35 (Slynn L); 64 (Hutton L). See Fruehwald, above n 73, for a discussion of a comparable New York provision. For a discussion of the similar position of de facto heterosexual couples, see D C Bradley, 'Meaning of Family: Changing Morality and Changing Justice' (1976) 39 Modern Law Review 222; Hurst, 'The Problem of the Elderly Statute', above n 6, 37.
  1 AC 27, 64.
 See R Dickerson, above n 3, 23.
 The question of whether the scope of a provision can contract is more difficult. Ultimately, this will depend upon the impression formed by the court as to whether the purpose of an Act was to cover a specific range of objects, or to cover them only because they were considered to be within a particular genus. A cautious approach is warranted, particularly where there are important reliance interests, but it is submitted that Roch LJ's view in the Court of Appeal in Fitzpatrick  Ch 304, 321 that it is appropriate to extend but not restrict the meaning of legislation is too simplistic.
 Stare decisis should play an attenuated role here, as progress will have revealed that an earlier decision is simply wrong. However, this must be balanced against the rule that common error makes law. In cases where an established interpretation is frequently relied upon, intervention may be inappropriate for pragmatic reasons: Hurst, 'The Problem of the Elderly Statute', above n 6.
 For an analogous case, consider Lord Clyde's view in Birmingham that the purpose of the predecessor of the Environmental Protection Act had always extended to matters of layout, though because these were not seen as health hazards at the time this went unrecognised  UKHL 59;  1 AC 617, 634-5. Though it is submitted that this view rests upon erroneous factual premises, his reasoning appears valid.
 Critics include J Herring, 'The Criminalisation of Harrassment' (1998) 57 Cambridge Law Journal 10; S Gardiner, 'Stalking' (1998) 114 Law Quarterly Review 33, 38; A Clarke, 'Can Psychiatric Injury Amount to Bodily Harm' (1997) 141 Solicitors Journal 1020.
 The Offences against the Person Act 1861.
  UKHL 34;  AC 147, 158.
 See P M Churchland, Matter and Consciousness: A Contemporary Introduction to the Philosophy of Mind (Revised ed, 1988) for a brief discussion of the evolution of views of the nature of mind.
 Cf A Clarke, 'Can Psychiatric Injury Amount to Bodily Harm' (1997) 141 Solicitors Journal 1020, who claims that the different policy issues involved in criminalising psychiatric injury mean that it is outside the ambit of the statute. This view is also plausible, but contrary to the view taken of psychiatric injury in the cases discussed in Ireland.
 Redish and Chung, above n 9, 815; Mitchell, above n 49, 724. 204Dickerson, above n 3, 87-8.
 Burrows, Statute Law in New Zealand, above n 3, 224-5; Bell and Engle, above n 65, 52; Hurst, 'The Problem of the Elderly Statute', above n 6, 33.
 DPP v Jordan  AC 699, 719 (Wilberforce L).
  2 NZLR 234.
 Section 312M of the Crimes Act was in issue in Moreton itself.
  2 NZLR 234, 238 (William Young J).
 Ibid 239.
 Ibid 240-1.
 Dyson Holdings v Fox  EWCA Civ 8;  1 QB 503; R v Dunn  2 NZLR 481, 483. In Dunn, McCarthy J held that the word indecent should simply be given the meaning 'which it is accorded in general use'.
 Burrows, above n 3, notes the point briefly at 224. Of course, it will remain open in particular cases for the courts to hold that Parliament intended the meaning of the word to shift in accordance with its linguistic meaning.
 Indeed, if the linguistic meaning of a term were to change dramatically, bizarre consequences could ensue. Dickerson, for example, suggests that a shift in the meaning of the word 'President' to 'Commission' would alter the shape of the federal government: Dickerson, above n 3, 126. See also J Evans, Statutory Interpretation: Problems of Communication (1988) 145.
 See The Shorter Oxford English Dictionary (1973) 1428; DPP v Jordan  AC 699.
 The mirror image of this point is that sometimes the conduct may expand beyond the linguistic meaning of a word. Social change may run ahead of linguistic shifts. Of course, it is this, rather than mere linguistic changes, which founds the need for updating.
 R v Dunn  2 NZLR 481.
 C v C  Fam 23, in Bennion, Statutory Interpretation: a Code, above n 3, 696.
 It will always be a matter of interpretation whether a word is used in this sense or simply in the fixed sense in which a layperson might understand a word.
 The phrase was used in the Immigration and Nationality Act 1952 (US) under which aliens with a psychopathic personality were excluded from entry into the United States. Homosexuality, while originally a ground for exclusion, arguably ceased to be so when medical opinion ceased to regard it as a psychopathic condition. The Act is discussed by Eskridge: Dynamic Statutory Interpretation, above n 4, 51-7. However, this account differs from his, which suggests that an evolutive reading of the statute was justified: 54-5.
 See also DPP v Jordan  AC 699, 719 (Wilberforce L).
 One might question the general validity of this distinction, but it is submitted that it is useful here as a way of differentiating between different kinds of word.
 It might be thought that is a mere theoretical possibility, because of the high value attached to precedent in interpretation cases: Burrows, Statute Law in New Zealand, above n 3, 212; Cross v CIR  3 NZLR 6. In most cases, a phrase will have been judicially interpreted before any relevant change in circumstances occurs. Unfortunately, judges are unlikely to contemplate the possible mobility of a relatively determinate word until it is necessary to recognise a change. However, in reality this presents little obstacle. Early decisions can readily be interpreted as determining only how a provision was to be applied at the time of the decision.
 See Dyson Holdings v Fox  EWCA Civ 8;  1 QB 503; Fitzpatrick  AC 27, 50 (Clyde L) (albeit the dictum is somewhat ambiguous); Hurst, 'The Problem of the Elderly Statute', above n 6, 189.
 See Fitzpatrick  AC 27, 43 (Nicholls L); A A S Zuckerman, 'Formality and the Family - Reform and the Status Quo' (1980) 96 Law Quarterly Review 248, 264-5.
  AC 27, 50. See also Dyson Holdings v Fox  EWCA Civ 8;  1 QB 503.
 Cf Allott, above n 28, 106.
 Ibid 108; E A Kellaway, Principles of Legal Interpretation of Statutes, Contracts and Wills (1995) 69; Hurst, 'The Problem of the Elderly Statute', above, n 6, 38.
  EWCA Civ 8;  1 QB 503. The principle was also applied in LTSS Print and Supply Services Ltd v Hackney LBC  1 QB 663. The court held that the word 'warehouse' varied over time in accordance with the popular meaning of the term. However, it is arguable that this decision is explicable on a more orthodox ground: the objective criteria for an object being a warehouse were loose enough to allow premises used in a variety of different ways to be classified as warehouses by the state.
 This legislation was the predecessor of that at issue in Fitzpatrick.
  EWCA Civ 8;  1 QB 503, 511 (James LJ); 513 (Bridge LJ).
 Ibid 507 (Lord Denning MR).
 Section 12 (1) (g) of the 1920 Act.
  EWCA Civ 8;  1 QB 503, 508 (Lord Denning MR).
 Gammans v Ekins  2 KB 328.
  EWCA Civ 8;  1 QB 503, 511.
 Ibid 511 (James LJ); 513 (Bridge LJ).
 Ibid 511 (James LJ).
 Ibid 512 (James LJ); 513 (Bridge LJ).
 He held first that prior decisions should not bind the court in changed social circumstances. Since this ignores the importance of the original meaning of the statute, it may suggest a dynamic approach to interpretation. Insofar as it does, it is unpersuasive, for the reasons given above. Alternatively, however, he held that it was for the tribunal of fact to apply the popular meaning of a word to concrete circumstances, so an appellate court could interfere only if the decision was plainly unreasonable. On the facts, appellate intervention was said to be unwarranted: Ibid 509-510. This approach presumes the correctness of the majority position, as the decision of the trial judge could be seen to be reasonable only if the meaning of the word had changed to embrace de facto relationships.
 Hurst, 'The Problem of the Elderly Statute', above n 6, 37; Burrows, Statute Law in New Zealand, above n 3, 227-8; Allott, above n 28.
  1 WLR 13, 16. Roskill and Cumming-Bruce LJJ concurred at 23.
 See, eg, Watson v Lucas  1 WLR 1493, 1500 (Oliver LJ); 1498 (Stephenson LJ). The problems and difficulties with the decision were referred to obliquely in Carrega Properties Ltd v Sharratt  1 WLR 928, 930 (Diplock L), though the matter was left for later consideration. In the court below, Megaw and Lawton LJJ expressed similar views: Joram Developments Limited v Sharratt  1 WLR 3, 8, 10.
 In Quilter vA-G  NZCA 207;  1 NZLR 523, Keith J accepted that the interpretation of broad provisions might sometimes be flexible, and noted Dyson: 571. However, this brief mention does not imply a considered endorsement of the particular method of change. See also H v C  3 NZLR 503, 508 (Anderson and Potter JJ); R vH  1 NZLR 129 (Hardie-Boys J).
 In his careful judgment in Re R (adoption) (1997-1999) 17 FRNZ 498, 510, Judge Inglis QC noted that Dyson was of dubious authority and distinguished it on the facts.
 This approach is founded upon Lord Reid's statement in Brutus v Cozens  UKHL 6;  AC 854, 861, that where the statute uses a word in its ordinary sense, 'it is for the tribunal which decides the case to consider, not as law but as fact, whether in the whole circumstances the words of the statute do or do not as a matter of the ordinary usage of the English language cover or apply to the facts which have been proved'.
 In Brutus v Cozens  UKHL 6;  AC 854, 863, for example, Lord Morris stated that 'I find it unnecessary and indeed undesirable to compose a definition of a word which is of general use and which presents no difficulty of application or understanding'.
  EWCA Civ 8;  1 QB 503, 510.
 Brutus was concerned with the phrase 'insulting behaviour'.
 See, eg, A A S Zuckerman, 'Formality and the Family -Reform and the Status Quo' (1980) 96 Law Quarterly Review 248, 264-5; Fitzpatrick  Ch 304, 335 (Ward LJ);  1 AC 27, 52 (Clyde L). Samuels, above n 2; Cf S M Cretney & F M B Reynolds, 'Limits of the Judicial Function' (2000) 116 Law Quarterly Review 181, 183.
 Zuckerman, above n 250, 265-6.
  AC 27, 43 (emphasis added).
 Ibid 52.
  UKHL 59;  1 AC 617, 628, Lord Hoffman accepted this.
 Ibid 627, Cf Lord Slynn.
 The words 'male' and 'female' are also of this type, given modern understandings of the nature of gender: cf Bellinger  EWCA Civ 1140;  Fam 150. Arguably, even the concept of a 'state', used in Birmingham, is similar. What is considered to be a 'state' likely to be injurious to health is something that is likely to vary as circumstances, including standards of hygiene and beliefs about risks to health, change.
 This answers Lord Hobhouse's point in Fitzpatrick  1 AC 27, 68, that changes in social attitudes had little to do with the policies underlying the Rent Acts.
 A hint at this view can be seen in Samuels, above n 2, though the link between words and values is not expressly drawn. It is not clear how far this technique might extend. All words might be seen to be value-driven in this sense: see the discussion of postmodernism in M Davies, Asking the Law Question (1994). However, as a practical matter, it seems that a clear interrelationship between values and a word should be required for it to be given a mobile interpretation. The more remote the connection, the less plausible it is that the legislature would have used a word in a mobile way.
 This is the case for the word 'family'. See Joram Developments Limited v Sharratt  1 WLR 3, 9 (Lawton LJ).
 Eskridge, University of Pennsylvania Law Review, above n 71, 1517; Côte, above n 59, 208. See also the discussions of the significance of the strictness of the words used to the ambulatory approach by Lord Nicholls in Fitzpatrick  1 AC 27, 35 and Lord Wilberforce in Royal College of Nursing  UKHL 10;  AC 800, 822.
 See Dickerson, above n 3, 88-9, 98-9 for a discussion of the hierarchical nature of legislative purpose.
 This problem is seen, for example, in Lord Steyn's fears that increasing reference to Hansard would preclude the proper application of the ambulatory approach. See J Steyn, 'Pepper v Hart: A Reexamination'  Oxford Journal of Legal Studies 59, 68.
 One possible example is Bellinger v Bellinger  EWCA Civ 1140;  Fam 150, in which the English Court of Appeal considered whether a transsexual female who had undergone gender reassignment surgery was a 'female' for the purposes of s 11 (c) of the Matrimonial Causes Act 1973, which voids a marriage the parties to which are not respectively male and female. It was common ground that when the Act was passed, the state of scientific knowledge would have precluded the identification of a man who had undergone gender reassignment surgery as female. The majority refused to update the Act on the ground that the policy issues involved in such an extension of rights were best left to Parliament. However, Thorpe LJ, in dissent, held that changes in social attitudes to marriage and transsexuals, and advances in medical understanding justified the updating of the statute. One factor was a statement in the House by the promoter of the Bill that the position of transsexuals would be left for the courts to develop in accordance with modern psychiatric understandings of their situation. This arguably indicates an intention to allow the meaning of the phrase to vary by reference to the psychiatric consensus on the meaning of the concepts 'male' and 'female'.
 There is a complex interrelationship between considerations of workability and legislative purpose. See Burrows, Statute Law in New Zealand, above n 3, 136, 210. In this case, it is not useful to rigidly separate the two factors: on a realistic view, the characterisation of purposes must be influenced by common sense considerations.
 Zuckerman, above n 250, 265-6; Samuels, above n 2.
 Lord Nicholls hints at this view in Fitzpatrick  1 AC 27, 45-6. See also Lord Clyde 52.
 See Sullivan, above n11, 149. Albeit her comment is directed at the ambulatory approach generally.
 See Royal College of Nursing v Department of Health and Social Security  UKHL 10;  AC 800, 822, in which Lord Wilberforce noted that the courts ought not to take a liberal approach to interpreting politically contentious legislation.
  1 WLR 1553 (CA.)
 Ibid 1555.
 Ibid 1557. See also De Richaumont Investment Company v OTW Advertising Ltd.  NZHC 310;  2 NZLR 831. Priestley J held that a cautious approach to the updating of s 128 of the Property Law Act 1952 was justified because the section was an infringement upon private property rights. This perhaps evidences an overcautious approach, given that the protection of property rights is, at least in the modern world, often overborne by other interests.
 See Dickerson, above n 3.
 Law Commission, Report No. 17. A New Interpretation Act: To Avoid Prolixity and Tautology (1990) 34-7.
 cf Lord Hobhouse's statement in Fitzpatrick  AC 27, 67, that 'it is an improper usurpation of the legislative function for the courts to adopt social policies which have not yet been incorporated in the relevant legislation'.
 Ibid. See also Fruehwald, above n 73, 985, citing Braschi v Stahl Associates Co 544 NYS 2d 782 (1989), a near-identical case concerning New York Rent Control Legislation.
 See Editorial, 'Homosexuality and Succession to Tenancies' (1999) 6 Housing Law Monitor 12.
 That is, there is no salient difference between the two objects when viewed in the light of the purposes of the Act.
 See Farber, above n 47, 287; Dickerson, above n 3, for a discussion of this aspect of the judicial role.
 This seems to be assumed by the majority in Fitzpatrick, but is strongly arguable given the general acceptance of anti-discrimination provisions in legislation in England (and New Zealand). There is some danger that such a consensus may be confined to higher socio- economic groups: 'The Problem of the Elderly Statute', above n 6, 32, but it is submitted that it is at least tenable to hold that homosexual relationships are now generally accepted.
 This contrasts with the situation in Bellinger  EWCA Civ 1140;  Fam 150. There, scientific opinion had not crystallized as to the gender of a transsexual.
 Hurst, for example, is clearly perturbed by the possibility of fluctuations in the meaning of statutory tests under the mobile words doctrine: 'The Problem of the Elderly Statute', above n 6, 42.
 Eskridge, Dynamic Statutory Interpretation, above n 4, 138-9.
 Hence, Hurst's fears for Eton College's charitable status are unjustified, as long as a principled approach is adopted: 'The Problem of the Elderly Statute', above n 6, 42.
 Consequently, this model seems to meet the requirements of certainty suggested by Loveland in 'And now wash your hands - Statutory nuisance under the Environmental Protection Act 1990 s 79 (1)' (2001) Journal of Planning and Environmental Law 1144, 1151.
 The position of unadvised citizens differs. Here, the relevant value is not certainty but the rule of law. Ideally, citizens should be able to guide their conduct in accordance with the statutory text: Sullivan, above n11, 139. This value is promoted by the proposed technique, as unadvised citizens are likely to read legislation in accordance with its contemporary meaning: Ibid. Even though mobility should not be tied to linguistic meaning, changes in values tend to be accompanied by linguistic shifts, so the technique is likely to support citizens' readings of a provision.
 L C Marshall, ' 'Let Congress do it': the case for an absolute rule of statutory stare decisis' (1989) 88 Michigan Law Review 177; Requadt, above n 2, 346.
 For a review of the stages involved in legislative change, from which the costly nature of the process can be inferred, see Burrows, Statute Law in New Zealand above n 3; R Mulgan, Politics in New Zealand (2nd ed, 1997). It might be said that the public purse, rather than private interests, should bear the cost of any uncertainty. However, given that the choice to litigate to resolve contentious issues remains with the parties, who can always settle their differences, it does not seem unfair to place the burden of resolving uncertainty upon them.
 See Requadt, above n 2, 334.
 Bell and Engle, above n 65, 22; Burrows, Statute Law in New Zealand, n 3, 192.
 Sullivan, above n11.
 Re R (adoption) (1997-1999) 17 FRNZ 498; cf Re Adoption by Paul and Hauraki  NZFLR 266. A traditional Maori marriage was considered sufficient to allow joint adoption under the Adoption Act 1955, ss 2, 3 which spoke of applicants being 'spouses' and 'husband and wife'. This is an arguable view, but the point cannot be fully considered here.
 Goodes  UKHL 34;  3 All ER 603. For some readers, there may be something ad hoc about this, as whether a provision can apply to an object is determined in the light of conditions subsisting after the enactment of the statute. However, this is not inconsistent with originalism. The question must always be whether the object falls within the scope of the original policy choices made by the legislature. This question can only be answered by reference to future developments.
 Requadt, above n 2, 353; Redish and Chung, above n 9, 868.
 It could, however, be viewed as an extreme example of the doctrine, in which the conceptual meaning of a word is defined solely by reference to a subjective criterion: what the court considers reasonable in the circumstances in the light of the purposes of the Act and the current context. Nothing turns upon this, however.
 See M V Yorke Motors (a firm) v Edwards  1 WLR 444, 450 (Diplock L).
  3 NZLR 360, 373-4. The relevant provision was s 43 (7) of the Nurses Act 1977.
  3 NZLR 360, 381-2. This decision could be subsumed to the mobile words doctrine on the basis of a broad 'public interest' criterion, but it seems more accurate simply to recognise that the court had the power to remake the provision in the light of contemporary norms.
 See, eg, Wahrlich v Bate  NZHC 1015;  3 NZLR 97, in which McGechan J held that a discretion to be exercised 'in the interests of justice' did not permit consideration of limited fiscal resources. The distinction between these cases seems to be that in the former, the discretion was in terms unlimited.
 Cf Eskridge, Dynamic Statutory Interpretation, n 4, 9; Bennion, Statutory Interpretation: a Code, n 3, 701. See alsoAGfor Ontario v AG for Canada  AC 127, 154 (Viscount Jowitt LC).
 Redish and Chung, above n 9, 868-9; Eskridge, University of Pennsylvania Law Review, n 71, 1517; Côte, above n 59, 208. This is also a possible interpretation of Lord Nicholl's speech in Fitzpatrick  1 AC 27, 45.
 It is difficult to conceive of cases involving loose and impressionistic language which could not be brought within the mobile words doctrine, particularly as in some cases very broad standards, such as community expectations regarding reasonable behaviour, may be adopted. Where this cannot be achieved, it seems strongly arguable that the legislative intention shown by the text was simply to establish a determinate rule. In such cases, the proper inference is that Parliament has delegated only the power to establish the boundaries of the rule, rather than the power to alter it.
 Of course, a version of the dynamic intention doctrine that was limited to low-level policy decisions could be created. However, it is submitted that it is better to bring updating within the mobile words framework. This provides clearer direction for counsel and judges regarding the ambulatory approach and avoids challenges to the legitimacy of judicial updating by characterising judges not as positive agents of social progress, but simply as recognising changed conditions.
 Requadt, above n 2, 353. Parliament could, of course, alter the Interpretation Act to license dynamic interpretation except where the context otherwise requires, but, in the light of the above reflections, it is doubtful whether this is desirable.
 See Requadt, above n 2, 353; Allott, above, n 28, 108. Eskridge's, Dynamic Statutory Interpretation, above n 4, 107, view that the courts invariably interpret statutes dynamically suggests a similar view.
 Côte, above n 59, 207.
 See Wahrlich v Bate  NZHC 1015;  3 NZLR 97, 111.