Sydney Law Review
Legislative Constitutional Baselines
Lael K Weis[∗]
‘Constitutional baselines’ are interpretive tools that are widely used in constitutional reasoning, although often implicit and unarticulated. They provide standards for measure that enable courts to evaluate the adequacy of the state’s provision of constitutionally guaranteed goods. This article identifies constitutional baselines as a distinctive issue in constitutional interpretation, and it examines an important but under-theorised way that the High Court of Australia defines constitutional baselines: namely, by adopting legislatively-defined norms or standards. The best-known example of this is the electoral franchise line of cases: in determining what the constitutional guarantee of representative government requires, the High Court frequently consults Commonwealth electoral legislation. However, while other commentators have observed and criticised this interpretive practice, it has not been properly understood or evaluated. This article clarifies how legislative constitutional baselines function, refines objections to their use, and develops an analytical framework for their evaluation. It ultimately argues that, at least under some circumstances, legislative constitutional baselines are justified because they provide a more plausible and more defensible method of defining constitutional baselines than methods that rely on other sources of constitutional meaning.
‘Baselines’ are starting points or points of departure that are used in a variety of different types of reasoning and analysis. They provide a standard for — although not necessarily of — measure. In the context of moral reasoning and analysis, baselines establish a frame of reference for what might otherwise be an unstructured, all-things-considered evaluation of possible ends or states of affairs. For example, philosophical theories of distributive justice rely on baselines to analyse whether existing patterns of wealth distribution are justified, and under what circumstances redistribution is justified. A baseline is what allows such theories to make sense of the notion of ‘redistribution’. Although baselines are often associated with the status quo — and where they are misapplied, with forms of status quo bias — this is not necessarily the case. Thus a distributive baseline might identify a distribution that in fact obtained prior to the enactment of some policy, or it might identify a distribution that is argued would obtain under an idealised set of circumstances, such as a ‘state of nature’ or a ‘veil of ignorance’. At the same time, however, baselines do not, in general, purport to identify an optimal standard. While this is a possible feature of a baseline, it is not a necessary feature. For example, non-ideal theories of justice use baselines to specify a set of minimum conditions of justice, such as the satisfaction of basic needs.
Baselines also play an important role in constitutional reasoning. One such role lies in analysing state compliance with constitutional requirements in relation to ‘constitutionally guaranteed goods’, broadly defined. How do we know whether state actions taken in relation to a constitutionally guaranteed good or that affect the provision of that good are consistent with constitutional requirements? Baselines help provide answers to interpretive problems of this kind because they help make sense of the constitutional requirement. They define standards against which state actions can be evaluated for adequacy. At the same time, they avoid the need to determine an optimal measure or method of provision.
This article examines a specific type of constitutional baseline that has been used in Australian constitutional jurisprudence: namely, legislative constitutional baselines. A ‘legislative constitutional baseline’, in the precise sense used here, is:
1. a legislatively-defined norm or standard
2. that a court deems to have constitutional significance
3. as a measure for the adequacy of the provision of a particular constitutionally guaranteed good.
Understanding legislative constitutional baselines has great importance for both the theory and practice of constitutional adjudication. When they are used, courts define constitutional requirements by reference to legislation. This appears to invert the foundational hierarchy of constitutional law over ordinary legislation, and to violate the fundamental principle that ordinary legislative enactments cannot legally bind future parliaments. In this respect, legislative constitutional baselines complicate conventional understandings of legal norm hierarchies within legal constitutionalism, as traditionally understood. They complicate the distinction that is standardly drawn between ordinary legislation and constitutional law within the domain of legal constitutionalism. Moreover, they complicate the distinction that is standardly drawn between legal and political constitutionalism.
This article does not attempt to address these theoretical puzzles. Before they can be addressed, legislative constitutional baselines must be properly identified and understood. Accordingly, the objectives of this article are to provide greater clarity about the nature of this interpretive practice, and to develop a set of conceptual tools for analysing and evaluating legislative constitutional baselines.
Part II of the article sets up this analysis by clarifying the use of constitutional baselines in general and contextualising them in Australian constitutional jurisprudence. Although constitutional baselines are a pervasive feature of constitutional reasoning, they are often implicit and unarticulated. As a result, scholarship to date has not articulated them as presenting a specific problem of constitutional interpretation, and they are largely under-theorised and under-unexamined. This Part identifies constitutional baselines as an interpretive tool that raises a distinct set of interpretive issues.
Part III of the article then shifts to the examination of legislative constitutional baselines in particular. The article focuses on a central example that has attracted significant commentary and criticism. This is the High Court’s constitutional jurisprudence on elections and the electoral franchise, where the Court has frequently used Commonwealth electoral legislation to give content to the constitutional requirement that Parliament be ‘chosen by the people’. Here the article describes the electoral franchise line of cases, situating them within the constitutional and interpretive context that informs the Court’s jurisprudence in this area. Drawing on this line of cases, the article then goes on to examine how constitutional baselines function, identifying objections to using legislation to define them, and ultimately arguing that this interpretive practice may nevertheless be justified.
Part IV considers and refines objections to legislative constitutional baselines. It distinguishes two different ways that constitutional baselines can function: as ‘strict’ constitutional baselines, whereby departures are determinative of invalidity, and as ‘threshold’ constitutional baselines, whereby departures are only prima facie invalid, and can be justified. The article accepts that using legislation to define a strict constitutional baseline is clearly objectionable because it is inconsistent with the principles of parliamentary and constitutional supremacy. Identifying the strict approach is helpful for crystallising objections to legislative constitutional baselines. However, strict constitutional baselines are rarely used in practice: most courts, including the High Court, use the threshold approach. Moreover, objections to using legislation to define a threshold constitutional baseline are not as severe. Accordingly, once the usual function of constitutional baselines in judicial reasoning is taken into account, judicial use of legislation to define constitutional baselines is at least potentially capable of justification.
Part V of the article proposes an approach for evaluating legislative constitutional baselines. The analytical framework that it develops rests on two core contentions. First, it is contended that some interpretive problems call for using a constitutional baseline, and that the task of defining a constitutional baseline often, if not typically, requires consulting extrinsic sources. Second, it is contended that legislative constitutional baselines must be evaluated by assessing the relative strengths and weaknesses of the non-statutory extrinsic sources that would otherwise be needed to define the constitutional baseline. Applying this approach, the article ultimately argues that in the context of the specific interpretive problem presented in the electoral franchise cases, legislation provides the High Court with a more plausible and more defensible way of defining the relevant constitutional baseline than the alternative sources of constitutional meaning that are available for this purpose. This demonstrates that there are at least some circumstances where legislative constitutional baselines are justified, and that this interpretive practice merits further examination and study.
Constitutional baselines are interpretive tools that permit a meaningful evaluation of state compliance with otherwise indeterminate constitutional requirements in relation to constitutionally guaranteed goods. They provide a standard of measure that answers the ‘compared to what?’ question. At the same time, constitutional baselines help courts avoid the need to articulate an ideal standard — something that is often thought to be undesirable for reasons having to do with the relative capacity and competency of the judiciary versus the legislature. Constitutional baselines are focused on adequacy, not optimality.
The concept of a ‘minimum core’, associated with debates about social and economic rights, is a familiar example from the global constitutional context. Although there are a variety of ways of understanding ‘minimum core’, the basic function of the concept is to help courts identify a standard for evaluating the adequacy of legislative and executive actions in relation to the provision of a good guaranteed by a constitutional right. In this way, ‘minimum core’ provides a standard for evaluating whether the state has met its constitutional obligations, but does not tell us what beyond this is feasible, desirable, or optimal.
Beyond this example, however, it is fair to say that constitutional baselines have not been a focal point in debates about constitutional interpretation. Indeed, although baselines feature pervasively in judicial reasoning about constitutionally guaranteed goods, they are often implicit and unarticulated. Courts may simply rely on analogies to findings across a line of decided cases without averring to the existence of a baseline, or rely on an established doctrinal test without explaining that test in terms of a baseline.
For example, even the most basic interpretive questions concerning constitutional guarantees of ‘liberty’ or ‘equality’, which are widely found in constitutions throughout the world, cannot be meaningfully assessed without a baseline. ‘Free’ or ‘equal’ compared to what? Yet, the focal point of jurisprudential analysis and critique typically concerns the kinds of constitutionally suspect ‘burdens’, ‘classifications’ and the like that trigger judicial scrutiny, and the manner in which judicial scrutiny is exercised — for example, whether to apply a strict prohibition or to use some form of balancing that accounts for competing interests, and if so how that balancing ought to be structured. The relevant baseline — which may simply be the status quo ante (that is, the state of affairs that obtained prior to the relevant state action under review) — is not often identified as a distinctive site of contention.
Similarly, baselines have not been identified as a central controversy or topic of debate in Australian constitutional jurisprudence. In the Australian context, however, inattention to constitutional baselines is also a function of the fact that the Australian Constitution has few express ‘rights’ or ‘rights-like’ guarantees. Whereas rights typically do raise questions about baselines, baselines are less frequently at issue in the central interpretive questions that Australian courts grapple with, which more commonly concern powers and structure. Issues of powers and structure do not, in general, present interpretive questions about the adequacy of state action that require reference to a baseline. For example, whether a Commonwealth law falls within a head of legislative power is a matter of characterisation: the law either bears a sufficient connection to the subject matter of a head of power (and is intra vires), or it does not (and is ultra vires). There is no further question concerning the adequacy of Commonwealth action taken in relation to that subject matter.
Even so, it is evident that constitutional baselines do play a significant, if underappreciated, role in Australian constitutional jurisprudence. For example, even in the legislative power context, there are exceptional cases that may present a baseline problem. The race power, Australian Constitution s 51(xxvi), which authorises the enactment of ‘special laws’ for ‘the people of any race’ is one possible example. Although the precise scope of the power is unsettled, on at least one approach — that advocated by Gaudron J in Kartinyeri v Commonwealth — analysing whether a Commonwealth law falls within the head of power has a ‘discrimination’ element that requires a standard of measure. More specifically, to enliven the power ‘there must be some difference’ that warrants singling out persons of a particular race for special treatment, or a ‘real and relevant difference necessitating the making of a special law’. Defining what counts as ‘real and relevant’ in this context requires a baseline. Is it current community standards? A standard based on principles of morality? Or something else?
More broadly, as Simpson has pointed out in her work on the constitutional concept of discrimination, there are various provisions in the Australian Constitution that require similar ‘assessments of who is relevantly “like” whom’, and therefore present ‘considerable room for debate about appropriate baselines’. As Simpson argues, a baseline issue arises whenever ‘constitutional rules invoke comparative concepts such as equality, discrimination, uniformity and preference’ — whether expressly by the terms of their text (Australian Constitution ss 51(ii), 99, 102, 117), or implicitly through the development of doctrinal tests interpreting their text (Australian Constitution ss 51(xxvi), 92). However, as Simpson’s discussion demonstrates, the issue of how to identify and define the relevant baseline is not treated as presenting a distinctive interpretive question and is often suppressed and unexamined in judicial reasoning.
Finally, the baseline problem is often present in interpretive questions raised by constitutional guarantees, whether express or implied. This includes ss 92, 99, and 117 of the Australian Constitution — all examples discussed by Simpson. It also includes implications from representative government, discussed below. Another possible example is s 51(xxxi), the constitutional requirement that the Commonwealth acquisition of property be ‘on just terms’. Determining whether a regulatory law (as opposed to a law that directly expropriates property) effects an ‘acquisition’ requires a point of departure for assessing which variations in rights, liberties, powers, privileges and immunities enliven the ‘on just terms’ guarantee. Practically all regulatory laws have this effect. However, it is accepted that not all such legislative variation amounts to an ‘acquisition’ of property. Here, too, the baseline problem is evident, and yet suppressed and unexamined.
In summary, the question of how to identify and define a constitutional baseline poses a distinctive interpretive problem for courts, even if it is not often identified as such. As the foregoing examples suggest, there are many different ways that a constitutional baseline could be identified and defined. Moreover, the interpretive problem cannot be avoided by simply relying on the status quo ante. For, even if treating the status quo ante as the relevant baseline is unproblematic, there is a further question concerning what features of the status quo ante should count toward defining the baseline. This article focuses on one way that courts sometimes define constitutional baselines, which singles out one particular feature of the status quo ante: namely, norms and standards defined by legislation.
Having identified and situated the interpretive problem presented by constitutional baselines, I now turn to the consideration of legislative constitutional baselines. A central example of the High Court of Australia’s use of legislative constitutional baselines comes from the body of jurisprudence concerning implications from the constitutionally prescribed form of government. This includes, in particular, the constitutional requirement that Parliament be ‘directly chosen by the people’. These implications limit the Commonwealth Parliament’s power to define the electoral franchise, including who votes and the manner of voting. In this sense, they function as a constitutional guarantee of a representative and democratic form of government. I refer to this line of cases as the ‘electoral franchise cases’, and use ‘the implied guarantees’ to refer to the relevant set of constitutional guarantees, which includes the implied freedom of political communication as well as implications concerning elections and the electoral franchise.
Before examining how legislation has been used to define a constitutional baseline in the electoral franchise cases, it is imperative to say a few words about interpretive context in order to situate the specific issues raised by this method of defining constitutional baselines. A key methodological assumption underlying this article’s analysis is that understanding and evaluating practices of constitutional interpretation are tasks that require taking account of contextual features, including constitutional and interpretive tradition. Although these will no doubt be familiar to most readers, it is important to be clear about the features of Australian constitutionalism that inform the analysis.
When compared with other, similarly situated, common law jurisdictions, the Australian tradition of constitutional interpretation is best described as highly ‘legalistic’. This is marked by the High Court’s tendency to approach constitutional interpretation in the same manner as ordinary statutory interpretation, albeit with due regard for the distinctive subject matter and drafting considerations that set the Australian Constitution apart from ordinary statutes. Emphasis is on text and structure, read in light of enactment history. By contrast, approaches that treat constitutions as ‘living’ instruments, the meaning of which evolves to reflect society’s changing needs and values, are generally regarded with scepticism.
This legalistic approach to constitutional interpretation bears a close relationship to the distinctive character of the Australian Constitution, which is best described as a practical charter of government. It establishes the basic framework for the exercise of powers by the Federal Government (the Commonwealth of Australia), and the division of powers between the Commonwealth and the states. The Australian Constitution famously does not include a bill of rights, and despite high levels of public involvement in the constitutional drafting process (especially for that time), Australian constitutionalism is generally thought to lack a central ‘founding moment’ that can be said to have ‘constituted’ the Australian people. The absence of these features makes it more difficult to contend that the Constitution has broader normative significance as a founding document or source of popular aspirations.
The foregoing description is not, of course, uncontested. Nevertheless, it is fair to say that it represents an orthodox view of Australian constitutionalism. It is this view that informs conceptions of judicial role, and the High Court’s interpretive practices ought to be understood and evaluated from this perspective. Importantly for present purposes, these contextual features help explain why constitutional guarantees implied from representative democracy are controversial. Understanding these features therefore helps situate the High Court’s approach to interpreting and applying the implied guarantees.
For example, although now well established, early cases recognising the implied freedom of political communication were heavily criticised for lacking a firm basis in the text and structure of the Australian Constitution. Although the Constitution clearly establishes a form of representative government, by design it does not contain a bill of rights. Moreover, interpreting and applying the freedom of political communication appears to require the Court to draw upon the kinds of interpretive resources and techniques used in jurisdictions with express constitutional rights of free speech, which sits uneasily with legalism and modest conceptions of judicial role. The contextual features just described help make sense of the development of the High Court’s jurisprudence in this area, which can be understood as an ongoing effort to align the freedom of political communication with constitutional and interpretive tradition, by tethering the implied freedom to its textual basis and to the prevailing understanding of the Constitution as a practical charter of government. Thus, ever since the decision in Lange v Australian Broadcasting Corporation, the Court has steadfastly maintained that the implied freedom is not a ‘constitutional right’, but rather a structural feature of the constitutional system, the content of which is to be determined in relation to the Constitution’s text and structure.
The electoral franchise cases have been criticised on similar grounds. However, in contrast to the political communication line of cases, what is particularly striking about the Court’s jurisprudence in this area is the fact that much of the implied guarantee’s content appears to be derived from legislation. The High Court has relied upon both current electoral legislation and legislative developments over time to interpret the constitutional requirement that Parliament be ‘chosen by the people’, a criterion which is then used to evaluate whether particular ways of regulating elections or restricting the electoral franchise are constitutionally valid. In the same way that jurisprudential developments within the political communication line of cases reflect contextual features of Australian constitutionalism and interpretive tradition, I ultimately suggest that this jurisprudential development, too, ought to be understood and evaluated in light of contextual features that inform the specific interpretive problem.
Turning to the electoral franchise cases, the case that provides the most compelling illustration of the High Court of Australia’s use of legislation as a source of constitutional meaning in this area is Rowe v Electoral Commissioner. Rowe concerned an amendment to the Commonwealth Electoral Act 1981 (Cth) that shortened the window of time for new voter enrolments and transfers in existing voter registrations in the period leading up to an election. Prior to 1983, the Act stated that the electoral roll closes on the day that the writs issue for an election. However, by longstanding convention, elections were announced well in advance of issuing writs — thereby giving eligible voters a pre-election window to enrol or transfer. Then in 1983, the Fraser Government departed from this convention, leaving only one day between announcing the election and issuing the writ. This event ultimately led to an amendment to the Act giving voters seven days to enrol or transfer. At dispute in Rowe was a 2006 amendment that shortened that seven-day window, requiring enrolment by 8pm on the day that the writs issue, and transfer by 8pm on the third working day after the writs issue — thereby effectively reducing the pre-election enrolment period by seven days, and the pre-election transfer period by four days. A 4:3 High Court majority held that these changes offended the constitutional guarantee of representative government because it would result in a Parliament that was not ‘chosen by the people’.
A critical aspect of the majority reasoning in Rowe was the notion that, in determining whether or not a law regulating elections or the electoral franchise is compatible with the constitutional requirement of choice ‘by the people’, the relevant constitutional baseline is universal adult suffrage (more precisely, universal adult citizen suffrage). While acknowledging that universal adult suffrage is not expressly mandated by the Australian Constitution, the majority emphasised that legislative developments since the framing have established this constitutional baseline. Choice ‘by the people’ is thus a constitutional concept that appears to derive its content from legislation. As French CJ explained,
[t]he content of ... ‘chosen by the people’ ... is now informed by the universal adult-citizen franchise which is prescribed by Commonwealth law. The development of the franchise was authorised by ... the Constitution ... [and] [i]mplicit in that authority was the possibility that the constitutional concept would acquire, as it did, a more democratic content than existed at Federation. That content, being constitutional in character ... cannot now be diminished.
Importantly, however, the proposition that universal adult suffrage is the relevant constitutional baseline, and the Court’s use of legislation to define that baseline, were not without precedent.
In Attorney-General (Cth) ex rel McKinlay v Commonwealth, an electoral districting case decided prior to cases establishing the implied guarantees, McTiernan and Jacobs JJ had suggested that ‘the long established universal adult suffrage may now be recognized as a fact and as a result it is doubtful whether ... anything less than this could now be described as a choice by the people’, referring to provisions of the Commonwealth Electoral Act. This suggestion received some initial support from three justices in McGinty v Western Australia, another electoral districting case, and was ultimately adopted in Roach v Electoral Commissioner.
In Roach, a law that disqualified all prisoners from voting, irrespective of the nature of the offence or the length of sentence, was found to be constitutionally invalid. In reaching this conclusion, the majority held that universal adult suffrage is the constitutional baseline that is used to determine whether a disqualification is consistent with the constitutional requirement of choice ‘by the people’. In their joint reasons, Gummow, Kirby and Crennan JJ asserted that universal adult suffrage is a standard that reflects contemporary understandings and values. Their Honours emphasised that representative government is not a ‘static concept’ and acknowledged legislative evolution, but they were otherwise not explicit about exactly how the relevant baseline is defined. However, a key point of emphasis in the reasons of Gleeson CJ is the idea that legislation provides the relevant set of ‘facts’ — as averred to by McTiernan and Jacobs JJ in McKinlay — that are in turn used to define the constitutional baseline. This was the approach adopted in Rowe.
Most recently, this understanding of Rowe as adopting and applying a legislative constitutional baseline was reinforced by Murphy v Electoral Commissioner. Murphy involved a challenge to the seven-day window for new enrolments and transfers that had been reinstated after Rowe invalidated the shortening of that period. The plaintiffs argued that a longer period was required to give effect to the constitutional requirement of choice ‘by the people’. Accepting this argument would have required the High Court to adopt an ideal measure (that is, an optimal standard) — namely, a criterion of maximal participation — rather than a baseline (that is, a standard of adequacy) for the purpose of evaluating the constitutional validity of laws regulating elections and the electoral franchise.
The High Court in Murphy unanimously rejected the adoption of an ideal measure to determine constitutional validity in favour of using a constitutional baseline. Moreover, Rowe was distinguished on the basis that it had involved a departure from the constitutional baseline. All of the justices emphasised that Commonwealth electoral legislation has always provided for a suspension of new enrolments and transfers prior to the closure of the electoral roll, and that for over three decades (since 1983) that suspension period has been seven days. At the same time, however, at least some members of the Court were equivocal about the use of legislation to define the constitutional baseline. Four members of the Court appeared to endorse the use of legislation for this purpose. Two appeared to be agnostic, simply accepting Rowe as authority for the legislative constitutional baseline. Finally, Gordon J expressed some reservations, noting that ‘[a]s a matter of constitutional interpretation, treating legislative diminution as a criterion raises complicated issues.’ At the same time, however, her Honour appeared to accept that this is an area where legislation is relevant to discerning constitutional meaning. Indeed, this was a recurrent theme, if not a basic premise, underlying all of the judgments.
The High Court of Australia’s use of electoral legislation to interpret the constitutional requirement that Parliament be ‘chosen by the people’ has not escaped attention. Within the electoral franchise line of cases, Rowe in particular has attracted controversy. One reason for the focus on Rowe is the contentious nature of the disqualification argument in that case. Unlike the total ban on prisoner voting at issue in Roach, it is not obvious that shortening the window of opportunity for voter enrolment and transfer in the period leading up to an election diminishes choice ‘by the people’. In this respect, Rowe drew attention to the legislatively-defined character of the constitutional baseline in a way that previous cases had not.
Indeed, this aspect of the Rowe majority’s reasoning was singled out for criticism by the dissenting judges, and has been criticised by commentators since. Most of these criticisms do not appear to object to the proposition that discerning what choice ‘by the people’ requires is an interpretive task where it makes sense to utilise a constitutional baseline. Or, at the very least, critics do not endorse defining an ideal measure or optimal standard. Nor do critics necessarily object to the proposition that universal adult suffrage is an appropriate constitutional baseline. Rather, the principal objection is to the manner in which the High Court has relied upon legislation to define this point of departure. For instance, Twomey has written that:
To suggest that the enactment of ordinary legislation ... amounts to changed facts which require the interpretation of the Constitution to change ... is to take a radical new approach to constitutional interpretation that is not justified by the text or structure of the Constitution.
It is imperative to be clear about the precise nature of the criticism. The objection is not against the use of extra-textual or ‘extrinsic’ sources to interpret the constitutional requirement. Nor is the objection against the use of a constitutional baseline per se. The text and structure of ss 7 and 24 — even when read in the context of other constitutional provisions and in light of enactment history — do not provide definitive answers to the interpretive question. The High Court has variously described the constitutional requirement of choice by the people embedded in the guarantee of representative democracy as: ‘a question of degree’ that ‘cannot be determined in the abstract’; a concept that ‘is not fixed and precise’;  and ‘a very large constitutional idea’. Even McHugh J, a moderate but committed textualist, once described the phrase ‘chosen by the people’ as ‘words of inexact application, dependent upon matters of fact and degree and always involving a value judgment’, and the term ‘the people’ to which that phrase refers as a ‘vague but emotionally powerful abstraction ... whose content will change from time to time.’
Facing interpretive challenges of this kind, Australian judges frequently do consult extrinsic sources of constitutional meaning, notwithstanding legalist orthodoxy. Moreover, it is precisely interpretive challenges of this kind that lend themselves to utilising a constitutional baseline. A constitutional baseline, howsoever defined, provides a needed frame of reference for an interpretive question that otherwise invites a wide array of moral, political, philosophical and sociological inquiries about the meaning of representative democracy. Importantly, it also avoids positing an ideal criterion, such as a requirement of maximal participation — an approach that the High Court and most commentators appear to agree would be undesirable in this context.
What we need to do, then, is to try to isolate the objection to using legislation to define a constitutional baseline. Assuming that a constitutional baseline is an appropriate doctrinal tool in this context and that extrinsic sources are needed to define it, what makes putting legislation to this interpretive use so objectionable?
Focusing on Rowe, the central objection, at least as articulated by leading commentators, has to do with the apparent inflexibility or rigidity of the legislative constitutional baseline. The concern is that, by treating a legislatively-defined norm or standard as a constitutional baseline, this interpretive practice effectively ‘entrenches’ the legislative status quo ante and sets in motion what is sometimes referred to as a ‘one-way ratchet’. In other words, although it permits subsequent parliaments to make the relevant legislative norm or standard more demanding, it appears to prohibit them from making that norm or standard less demanding (that is, Parliament can ‘ratchet up’, but not down).
For example, since 1973 the electoral franchise has extended to ‘all persons who have attained 18 years of age’. The rationale for this extension appears to have been based upon: a recognition that 18 is the legal age of adulthood for a variety of other purposes; a belief that 18 year olds are no less adults than 21 year olds in terms of their maturity and independence; and the notion that with adulthood comes the right and obligation to vote. Let us suppose, however, that new sociological studies indicate that young persons in their late teens to early 20s today typically exhibit a degree of immaturity and dependency that is inconsistent with actual (if not legal) adulthood due to a variety of life circumstances that have become increasingly common for persons of this age (for example, undertaking university study, living at home, relying on parents for financial support, etc). Relying on this evidence, could Parliament now amend the Electoral Act to restrict the franchise to persons 21 years of age and older, as was the case prior to 1973?
The concern is that by treating current legislation as defining the constitutional baseline, a decision to deny the franchise to 18 to 20 year olds would be invalid despite Parliament having (at least arguably) good reasons for the change. On the other hand, Parliament would be free to extend the franchise to 16 year olds. Once extended, however, it would appear that the electoral franchise could not then subsequently be restricted to return to 18 years of age consistently with the constitutional requirement of choice ‘by the people’.
This method of evaluating legislative changes for constitutional validity is contentious for reasons that are aptly captured by Heydon J’s dissenting judgment in Rowe. His Honour objected that:
The constitutional validity of legislation depends on compliance with the Constitution, not on compliance with ‘higher’ standards established by the course of legislation ... . The question is not whether an impugned legislative provision ‘regresses’ from some ‘higher’ standard established by the status quo. It is only whether it fails to meet a constitutional criterion. Legislative development, durable or otherwise, does not create constitutional validity or invalidity which would not otherwise exist. Otherwise the legislature could enact itself into validity.
Here we can disaggregate two distinct objections to legislative constitutional baselines, implicit in this quoted passage. Each appeals to a well-established principle of constitutional law.
The first objection, which I will call the ‘parliamentary supremacy objection’, is that this approach to constitutional interpretation results in a state of affairs where ordinary legislation is legally binding on future parliaments. A legislatively-defined norm or standard is used to define a constitutional requirement against which the validity of other legislation — or, indeed, the validity of any variation of the legislation used to define the baseline — will be determined. In this respect, legislative constitutional baselines appear to violate the principle of parliamentary supremacy that a parliament cannot bind its successors. This is highly controversial. Absent a compelling reason to think that legislatively-defined norms and standards represent points on a teleological trajectory toward a maximally optimal end-state, we want parliaments to be able to revise those norms and standards in response to new and emerging developments in information, resources, technology, social needs and values, and so on.
The second objection, which I will call the ‘constitutional supremacy objection’, is that the interpretive use to which legislation is being put violates the principle that constitutional law is ‘fundamental’ law, in the sense that it governs the validity of ordinary legislation. Treating ordinary legislation as having constitutional significance collapses the hierarchy of constitutional law over ordinary legislation that is foundational to legal constitutionalism. This, too, is highly controversial. If ordinary legislation is a source of constitutional meaning, then there is a sense in which legislative enactments establish the criteria for their own validity, violating the well-known maxim that ‘the stream cannot rise above its source’.
These are serious objections. The question, then, is whether they can be overcome. Is using legislation to define a constitutional baseline ever permissible or justified? In order to evaluate this interpretive practice, it is necessary to consider how constitutional baselines, howsoever defined, function in judicial reasoning. My suggestion is that once this is properly understood, it is apparent that the objections to courts using legislation to define constitutional baselines are not as strong as they first appear.
One possibility, implicit in concerns about ‘entrenchment’ and the ‘one-way ratchet’ effect, is that constitutional baselines function as an absolute minimum standard for the adequacy of the state’s actions in relation to constitutionally guaranteed goods, such that any ‘legislative departure’ (that is, legislation giving effect to something less than that standard) is constitutionally invalid. I will refer to this as a ‘strict’ constitutional baseline. The ‘strict’ approach clearly raises the objections to legislative constitutional baselines outlined above. Using legislation to define a strict constitutional baseline effectively ‘entrenches’ the legislative status quo ante: it turns an ordinary legislatively-defined norm or standard into an inflexible or rigid norm or standard that is incapable of downgrade.
But this is not the only, or indeed even the usual, way that constitutional baselines function in judicial reasoning. It is far more common for courts to use constitutional baselines to establish a prima facie case of constitutional invalidity.
I will refer to this as a ‘threshold’ constitutional baseline. A threshold constitutional baseline functions as a ‘rule of thumb’: it indicates the minimum adequate provision of a constitutionally guaranteed good that can ordinarily be expected, other things being equal. Legislative departures from a threshold constitutional baseline indicate that judicial scrutiny is required, but are not determinative of invalidity. Which is to say, legislative departures can be justified and therefore constitutionally valid. The method that courts standardly use to undertake this evaluation involves some form of ‘limitations analysis’. This includes proportionality testing and, in the Australian context, the ‘reasonably appropriate and adapted’ criterion and related methods.
This description of threshold constitutional baseline reasoning is consistent with the High Court’s approach in the electoral franchise cases, including Rowe. As even the strongest critics of Rowe acknowledge, the majority’s conclusion of invalidity did not rest exclusively upon legislative departure from the constitutional baseline, but also upon the lack of a compelling justification by applying a form of limitations analysis that the franchise cases refer to as the ‘substantial reason’ test. That test requires examining whether the legislative end is consistent with the constitutional guarantee of representative democracy and evaluating the ‘fit’ between that end and the legislative means adopted (including whether there is a rational connection between means and end, and whether there are less restrictive means available).
One of the difficulties with relying on Rowe to evaluate legislative constitutional baselines, however, is that it obscures the usual function of constitutional baselines. Due to the contentious nature of the disqualification issue, the majority’s analysis of the justification issue is relatively thin. Rowe required far more argument and analysis than Roach to make the threshold determination that there had been a legislative departure from the constitutional baseline. In Roach, this threshold determination was straightforward: the Act diminished the adult citizen franchise by expressly disqualifying previously qualified electors. Accordingly, the focal point of the majority judgments in Roach concerned the application of the ‘substantial reason’ test. In Rowe, by contrast, the threshold determination was complicated: it was not obvious that the Act’s procedural changes to enrolment and transfer diminished the adult citizen franchise, and therefore whether the ‘substantial reason’ test needed to be applied at all.
The difficulty with the relative ‘thinness’ of the Rowe majority’s reasoning on the justification issue is that it creates the impression that the constitutional baseline is much stricter than it is. In particular, the contentious nature of the majority’s finding on the disqualification issue makes it appear as if all legislative changes that result in a less inclusive electoral regime are constitutionally invalid. In this sense, the ‘entrenchment’ and ‘one-way ratchet’-type concerns raised about Rowe seem better understood as an effort to crystallise why the High Court’s use of legislation to define the constitutional baseline is objectionable, and not as an effort to describe how the baseline in fact functions. However, while this characterisation may be helpful in sharpening objections to legislative constitutional baselines by presenting them in their strongest possible form, it is not obvious that these objections apply with the same force once we account for how constitutional baselines more commonly function in judicial reasoning: namely, as threshold rather than strict baselines.
To begin with, while it is true that legislative constitutional baselines privilege the legislative status quo ante by imposing justificatory demands on legislative departures, they do not prevent legislative change. Constitutional baselines function as a ‘trigger’ for the judicial scrutiny of constitutional validity. Assuming that it is possible in both principle and practice for legislation to survive judicial scrutiny — a point that I will return to in a moment — it therefore remains possible for Parliament to enact legislation that departs from the baseline.
Moreover, once the function of constitutional baselines in judicial reasoning is accounted for, it is evident that there is also greater flexibility in the manner in which legislative constitutional baselines are defined than is assumed by ‘entrenchment’ and ‘one-way ratchet’-type concerns. It is true that once a court defines a constitutional baseline, the doctrine of precedent dictates that it will not be easily overturned. But it does not follow from this that it is unrevisable or irreversible. Whether a constitutional baseline is revised or reversed is always a matter of judicial reasoning, and the weight that courts give to the interpretive sources used to define the baseline — legislative or otherwise.
Accordingly, it is not necessarily the case that all legislative developments that result in a more demanding standard will move the baseline. For example, even if the Commonwealth Parliament expands the electoral franchise to include 16 year olds, the baseline might remain universal adult suffrage, with the recognition that some non-adults are now part of the franchise. Nor is it necessarily the case that only legislative developments that result in a more demanding standard are capable of altering the baseline. In the same way that a court might revise an established constitutional baseline by reconsidering other extrinsic sources — say, framing understandings, or new social attitudes — that reflect a more restrictive standard, there is nothing that rules out the possibility that a court might determine that legislative developments which reflect a more restrictive standard are not only justified departures from the baseline (and therefore constitutionally valid), but that they provide sound reasons for revising the constitutional baseline itself.
Now, as noted above, an important assumption in the foregoing analysis is that it is possible both in principle and in practice for Parliament to discharge its justificatory burden. If the justification required for departing from a constitutional baseline is extremely demanding, then it may operate in effect as a strict baseline. In these circumstances, using legislation to define the constitutional baseline will likely be inconsistent with the principles of parliamentary and constitutional supremacy, and should be rejected on that basis. On the other hand, in order to count as a constitutional baseline at all, judicial scrutiny must impose real demands of justification upon Parliament to provide reasons for legislative departure. For these reasons, analysing the applicable method of limitations analysis will be a necessary preliminary step in evaluating legislative constitutional baselines. It is important to emphasise, however, that these issues turn on the method of judicial scrutiny applied once ‘triggered’ by legislative departure. There is therefore a sense in which objections to legislative constitutional baselines that always result in the invalidity of legislative departures are better directed, at least in the first instance, to the form of limitations analysis deployed and not to the use of legislation to define the baseline per se.
In summary, then, legislative constitutional baselines impose burdens of justification on Parliament to provide reasons for legislative departures from established legislatively-defined norms and standards that a court has deemed to have constitutional significance. This may ultimately result in the constitutional invalidity of legislative departures from the baseline. Unless the baseline is strict, however, the legislatively-defined content of the baseline is not rigid or inflexible in the sense of imposing a standard that is conclusive of invalidity or incapable of becoming less demanding. It is therefore inaccurate to say that when courts use legislation to define a constitutional baseline it results in a state of affairs where future parliaments are legally ‘bound’ by prior parliaments. Future parliaments are burdened by requirements of justification when they enact legislation that departs from the baseline, but they are not strictly bound. Similarly, it is overly simplistic to think of legislative constitutional baselines as ‘entrenching’ legislatively-defined norms and standards. As with any other constitutional rule that requires interpretation, the process of defining a baseline is always mediated by judicial reason and the weight given to interpretive sources.
This is not to suggest that using legislation to define constitutional baselines is unobjectionable. Rather, it is to suggest that a different strategy for evaluation is required. Legislative constitutional baselines evidently do function as constraints on legislative power, and they do so by treating legislatively-defined norms and standards as having constitutional significance. And, like other judicially-defined constitutional rules, they are not directly revisable through the enactment of ordinary legislation. However, these kinds of concerns arise any time that a court relies on extrinsic sources to define a constitutional baseline. Accepting that there are legitimate sources of constitutional meaning outside the ‘four corners’ of the constitutional text, why should certain kinds of norms and standards, but not others, be treated as having constitutional relevance?
The objection to legislative constitutional baselines therefore becomes an objection to legislation as an interpretive source as against other available extrinsic sources. Pressed any further, the objection becomes an argument against using extrinsic sources in constitutional interpretation. Or — pressed even further — the objection becomes an argument against legal constitutionalism altogether: that is, it becomes an argument against constitutional supremacy and in favour of parliamentary sovereignty, as opposed to the weaker parliamentary supremacy.
The foregoing considerations thus permit us to refine the objections to legislative constitutional baselines as follows:
The status quo bias objection: legislative constitutional baselines privilege the legislative status quo ante. Absent good reasons for thinking that the legislative status quo has constitutional significance, courts should rely on other sources to define the baseline.
The norm-hierarchy objection: legislative constitutional baselines give ordinary legislation constitutional standing. Absent good reasons for thinking that legislatively-defined norms and standards are needed to interpret the constitutional requirement, courts should rely on other sources to define the baseline.
Thus refined, these objections focus attention on the decision of courts to draw upon legislative as opposed to other sources of constitutional meaning in defining a constitutional baseline. Evaluating legislative constitutional baselines in light of these objections is the task I turn to next.
So far, I have drawn upon the electoral franchise cases to illustrate a characteristic way that the High Court uses legislation to define constitutional baselines, and I have clarified and refined the objection to using legislation to define constitutional baselines. In this Part, I propose a framework for evaluation. I will ultimately argue that, at least under some circumstances, using legislation to define constitutional baselines can be justified.
As discussed in Part IIIA above, which set out the relevant interpretive context, the default starting point for constitutional interpretation in Australian constitutional practice is the Constitution’s text and structure, read in light of its enactment history. But they are not, of course, the ending point. When text and structure, read in light of enactment history, do not yield a determinate answer to the interpretive question, then courts must consult extrinsic sources, purely as a matter of practical necessity. As the electoral franchise cases demonstrate, legislation is among the possible extrinsic sources available
The principal claim I advance is this: evaluating legislative constitutional baselines requires a comparison of the merits of using legislation to define the baseline with the merits of using other available extrinsic sources to define the baseline.
Evaluating those alternative sources requires, in turn, examining the parameters of the specific interpretive problem posed by the constitutionally guaranteed good being interpreted and applied, having regard for contextual features of Australian constitutionalism and interpretive tradition that inform the interpretive problem. Substantively, I argue that the constitutional guarantee of representative democracy, and the specific requirement of choice ‘by the people’, provides an example of circumstances where a legislative constitutional baseline is both a plausible and defensible method of evaluating constitutional validity.
In developing a framework for evaluating legislative constitutional baselines, it is helpful to recall the discussion above in Part II. There we saw that the utility of constitutional baselines as an interpretive tool lies in providing a way for courts to meaningfully evaluate state compliance with otherwise indeterminate constitutional requirements in relation to constitutionally guaranteed goods. Constitutional baselines provide courts with a frame of reference for what would otherwise be an unstructured, all-things-considered evaluation of any number and variety of factors that bear on the nature of the good and the reasons for affording it constitutional status. At the same time, they help courts avoid making contentious determinations about the optimal measure or method of provision: they define a standard for adequacy, not optimality. Moreover, drawing on the discussion above in Part IV, we have seen that the usual way that constitutional baselines function is to indicate what is ordinarily required by way of adequacy and not what is necessarily required in all circumstances. They ‘trigger’ judicial scrutiny of legislative departures but are not conclusive of invalidity.
Bearing in mind the role of constitutional baselines in judicial reasoning, the evaluation of legislative constitutional baselines therefore turns on how well legislation fares in defining a workable constitutional baseline for courts to use in this manner when compared with other extrinsic sources. If courts do not use legislation to define the constitutional baseline, then what are the alternatives? And what are their relative strengths and weaknesses? What objections are courts likely to face when using them? This Part maps out these alternative non-statutory extrinsic sources and considers their status as interpretive sources within the Australian context. Table 1 below summarises these sources and the objections that courts are likely to face when using them, in light of key assumptions about judicial role specific to Australian constitutional and interpretive practice.
Without purporting to be definitive or exhaustive, the main possible alternative non-statutory extrinsic sources appear to fall within two broad categories: (1) social understandings and values, and (2) substantive views. Within the first category, social understandings and values, we can include:
(a) Founding understandings and values: that is, evidence of what the founding generation thought was adequate; and
(b) Contemporary understandings and values: that is, evidence of what the Australian people today think is adequate.
Notice that both (a) and (b) refer to the beliefs and attitudes actually held by groups of persons whose understandings and values are sometimes thought to bear on constitutional meaning, for one reason or another. They are said to bear on constitutional meaning because they were (in the case of the founding generation) or because they are (in the case of the Australian people today) in fact held by those persons. In this respect, these sources can be distinguished from the second category, substantive views, which includes:
(c) Substantive moral views: that is, views about the good or the right that are grounded in notions of constitutionalism, in philosophical doctrines, and so on; and
(d) Substantive non-moral views (that is, views about what is the case or what exists that are grounded in natural science, in human psychology, and so on).
Sources (c) and (d) refer to views that, while held by some persons, may not be widely held (if held at all) by members of the founding generation or even by members of the Australian public today. They are said to bear on constitutional meaning not because they reflect the attitudes and beliefs of persons whose understandings and values bear on constitutional meaning, but because they provide the best account (that is, the most plausible or correct account) of a relevant topic. Deploying these sources thus requires identifying a particular ‘epistemic community’ or the group of experts which is well-placed to provide the best account of the topic in question.
Having identified these basic categories, we can turn to the question of their relative merits. The extent to which Australian courts are inclined to regard these as plausible and permissible sources of constitutional meaning, I want to suggest, turns on both practical considerations and normative considerations. Clearly, these are interrelated: for example, practical difficulties with using a particular source may bear on whether it is regarded as normatively defensible. Nevertheless, it will be helpful to disaggregate these considerations. For, whereas practical considerations present general operational challenges for courts wishing to use a particular interpretive source, normative considerations draw attention to specific features of Australian constitutionalism that inform established interpretive practice.
Starting with practical considerations, each source presents courts with operational challenges. The challenge lies in providing evidence to prove certain facts upon which key assumptions that establish the source’s relevance to constitutional meaning rely. Starting with social understandings and values, the key operational challenge here is attribution-driven and lies in providing evidence to demonstrate that certain beliefs and attitudes are, or were, in fact held by the relevant group of persons. Thus, other things being equal, these sources of meaning are more difficult to justify where there is a lack of reliable evidence that persons within that group actually held certain beliefs and attitudes, or where the available evidence is insufficient to demonstrate that those beliefs and attitudes were sufficiently widely held that they can be attributed to the entire group.
The key operational challenge for substantive views, by contrast, is soundness-driven. The challenge here lies in identifying the relevant epistemic community and articulating the view that is defended within that community as the best account of the relevant topic, in the sense of being the account that is most plausible or most widely regarded as correct. Other things being equal, then, the more contentious the substantive view is within the relevant epistemic community, the more difficult it will be for a court to justify its use as a source of constitutional meaning.
Bearing these operational challenges in mind, it is a further question whether using these sources is normatively defensible. For instance, even if establishing the beliefs and attitudes held by members of the Australian public on a particular topic is empirically uncontentious, it is a further question whether a court ought to use those beliefs and attitudes as an interpretive resource. Normative defensibility is informed by the parameters of the specific interpretive problem: it is a function of both the source’s relevance to the subject matter of the provision being interpreted, and how the source is regarded within the interpretive community. It is therefore an issue that is not possible to analyse in the abstract.
For one thing, different constitutional guarantees pose different interpretive challenges simply by virtue of their subject matter. For example, interpreting the freedom of interstate trade, guaranteed by s 92 of the Australian Constitution, requires making assumptions about economic theory, including how markets function, and how regulation affects production and exchange. Moreover, the provision is drafted in a way — as an unqualified imperative that interstate trade ‘shall be absolutely free’ — that makes its prohibition opaque unless consideration is given to its purpose and place within the project of Federation. Accordingly, in the s 92 context substantive non-moral views have greater interpretive relevance than moral views (that is, views concerning the right or the good), and the framers’ beliefs and attitudes have greater interpretive relevance than those of the public today.
But subject matter does not provide a complete picture. How courts understand a particular constitutional guarantee, and therefore how the interpretive problem is framed, is highly sensitive to constitutional and interpretive context. For instance, courts in other jurisdictions often interpret and apply constitutional guarantees of free speech by drawing upon popular beliefs about the value of speech and expression, as well as moral views about the significance of speech and expression for democracy and human flourishing. However, the High Court of Australia has held that such an approach to the implied freedom of political communication misapprehends the nature of the constitutional guarantee. In particular, the Court has insisted that extrinsic sources that courts in other jurisdictions use to interpret express guarantees of free speech are inapplicable in the Australian context because the implied freedom is not a ‘right’, but rather a structural feature of the constitutional system of representative government. How far this legalistic, ‘structural’ understanding of the implied freedom goes to distinguish it from express guarantees found in other constitutional systems as a conceptual or normative matter is subject to debate. However in terms of constitutional practice, as discussed above in Part IIIA, it manifestly has influenced how the High Court approaches the implied freedom, which, in turn, influences the interpretive resources that the Court is willing to draw upon.
What the foregoing discussion underscores is that evaluating the judicial use of a particular extrinsic source to interpret a constitutionally guaranteed good cannot be viewed narrowly in terms of subject matter. Even if subject matter establishes the prima facie relevance of particular extrinsic sources, constitutional and interpretive context — which inform a court’s understanding of the good and the interpretive problem that it presents — are the more fundamental considerations at play.
Focusing specifically on Australian interpretive practice, a few generalisations about the judicial reception of the non-statutory extrinsic sources identified in this Part can be offered. In offering these generalisations, it is important to bear in mind the discussion of interpretive context in Part IIIA and, in particular, the High Court of Australia’s consistent and long-standing emphasis on the primacy of text and structure — a core tenet of Australian legalism. It is also important to bear in mind that the following discussion is meant to be descriptive: the aim is to understand the defensibility of relying on legislative sources from within the Australian constitutional and interpretive tradition. The article does not offer a view about whether the High Court’s approach to extrinsic sources is normatively desirable.
To begin with, the strength of legalism as an interpretive tradition means that the High Court of Australia is generally far more reluctant than its counterparts elsewhere in the common law world to draw upon contemporary understandings and values. For example, Australian interpretive practice stands in sharp contrast to the Supreme Court of Canada’s ‘living tree’ approach, which not only permits but requires drawing on contemporary values and understandings. On the other hand, founding understandings and values are now a well-established and relatively uncontroversial source of constitutional meaning. Unlike contemporary understandings and values, founding understandings and values are regarded as consistent with legalism insofar as they bear upon the Australian Constitution’s enactment meaning, so long as they do not amount to a quest to identify the framers’ subjective intentions.
Similarly, the High Court is often willing to draw upon substantive non-moral views, as the s 92 example demonstrates. By contrast, the Court generally resists drawing upon substantive moral views (that is, views about the right or the good) unless there is evidence that the framers held a moral view that bears upon a provision’s enactment meaning. Even then, such views are often characterised as non-moral in the sense that they fall within the special lawyerly expertise held by the framers as an epistemic community. An example of this is the debate among members of the High Court throughout the 1980s and 90s about the purpose of s 90 of the Australian Constitution in giving the Commonwealth the exclusive power to levy duties of excise. Although the central debate turned on competing normative views of fiscal relations within a federation, it proceeded as a debate about which of those two views better described the framers’ design.
In summary, then, evaluating the use of a given extrinsic source to define a constitutional baseline will always be case-specific. Evaluation must have regard to: the character of the constitutional good being interpreted and applied; features of the constitutional and interpretive context that inform that interpretive problem; and operational challenges involved in deploying the particular interpretive source. Likewise, I want to suggest, evaluating legislative constitutional baselines requires examining how statutory sources hold up against the available alternatives.
Table 1: Summary of Non-Statutory Extrinsic Sources
Nature of source
Social understandings and values
Beliefs and attitudes actually held by a group of persons.
Framing understandings and values
Contemporary understandings and values
Operational objections: evidence of beliefs/attitudes is unreliable or insufficient.
Normative objections: insufficiently related to subject matter or inconsistent with interpretive commitments and understanding of judicial role.
Contemporary understandings and values generally regarded as more objectionable.
The best account of a topic, as defended within an epistemic community.
Substantive moral views
Substantive non-moral views
Operational objections: insufficient consensus within epistemic community to establish plausibility/correctness.
Normative objections: insufficiently related to subject matter or inconsistent with interpretive commitments and understanding of judicial role.
Substantive moral views generally regarded as more objectionable.
In this final section, I outline an analytical framework for evaluating legislative constitutional baselines. Here I return to the electoral franchise cases, where we have seen that the High Court predominantly relies upon legislation to define the constitutional baseline. Developing an evaluative approach by working through a specific example will help clarify the steps that are required by making the discussion more concrete: the same framework can be applied in other cases where legislative constitutional baselines feature in judicial reasoning.
In developing this evaluative framework, the objective is to identify reasons why legislation may provide courts with a more attractive, and indeed more defensible, interpretive resource for defining constitutional baselines than non-statutory extrinsic sources. My central claim is that this interpretive practice is both plausible and permissible insofar as it helps courts overcome objections to using non-statutory extrinsic sources in circumstances where the specific interpretive problem otherwise seems to require relying on those sources. The electoral franchise cases, I suggest, present precisely such a scenario. These cases therefore demonstrate that there are at least some circumstances where legislative constitutional baselines are defensible.
Let us return, then, to the interpretive problem presented by the requirement that Parliament be chosen ‘by the people’. As argued above, this phrase presents precisely the kind of interpretive problem that calls for defining a constitutional baseline. Moreover, the text and structure of the Australian Constitution do not provide ready answers. As Gageler J has recently remarked, it is a concept that ‘defies being diced or squashed to fit within a judicially constructed box’. Extrinsic sources are required. So, which sources should the High Court use? And is the Court’s current practice of using legislative sources legitimate?
Step 1: Identify Relevant Non-Statutory Extrinsic Sources
Our first step is to identify the non-statutory extrinsic sources that seem relevant to the task of defining the constitutional baseline, in light of the subject matter of the constitutional guarantee. The concept of choice ‘by the people’ is a normative concept grounded in the theory and practice of representative government. It reflects moral views about the body politic, including membership within the polity and participation in self-governance. As observed above, even mainstream, moderate Australian judges have described it as a ‘vague but emotionally powerfully abstraction’, and as embodying ‘a very large constitutional idea’. Substantive non-moral views thus appear to have limited relevance. However, all three of the other sources are at least potentially in play.
To begin with, the constitutional requirement could be understood in terms of founding understandings and values. Those understandings and values were evidently quite different at the time of the Australian Constitution’s framing, nearly 120 years ago, than they are today. For instance, women and Indigenous peoples were excluded from the electoral franchise in a majority of the colonies and, thus, it is arguable that they were not among ‘the people’ contemplated by the constitutional requirement. Although it is uncommon to rely on founding understandings and values in this context, there are at least some critics of Rowe who adopt this approach. This would evidently result in a much different baseline than the one that the High Court has adopted, however. Accordingly, for present purposes this source can be left aside.
By contrast, contemporary understandings and values (which appeal to beliefs and attitudes held by members of the Australian public today), and substantive moral views (which here encompasses positions found in normative political theory), both provide support for the established constitutional baseline. Without attempting to offer an argument, it is plausible to think that the proposition that representative democracy ordinarily requires universal adult suffrage is reflected in widely held popular beliefs and attitudes about representative democracy, and in the most widely accepted views about representative democracy that are defended by contemporary political theorists.
Step 2: Assess the Relative Merits of Non-Statutory Extrinsic Sources
Having established the relevance of these non-statutory extrinsic sources to the interpretive problem, the next step is to consider their relative merits as sources of constitutional meaning. Both appear to converge upon the same constitutional baseline. But so, too, does Commonwealth electoral legislation. Why, then, might legislation provide a more attractive, and potentially more defensible, source of constitutional meaning? Answering this question requires considering contextual factors that inform the interpretive problem. In particular, it requires appreciating why contemporary understandings and values and substantive moral views are regarded as suspect sources of constitutional meaning within the Australian constitutional and interpretive tradition, and why they are particularly fraught in this context.
To begin with, both interpretive sources are at least prima facie in tension with the basic commitments of legalism. Treating them as legitimate sources of constitutional meaning suggests that constitutional meaning turns on considerations that are independent of text and structure, and that constitutional meaning evolves over time to reflect new social understandings and values. From this perspective alone, they seem objectionable and potentially difficult to justify when applied to the interpretive problem.
Moreover, both interpretive sources involve inquiries that go beyond the extremely modest understanding of judicial role and the strict separation between judicial and legislative power that colour Australian constitutionalism. Relying on contemporary understandings and values appears to require judges to make freestanding inferences about beliefs and attitudes held by society at large. Relying on substantive moral views appears to invite, if not require, judges to make determinations about the right or the good — or, at the very least, to take sides in contentious moral debates. These are the kinds of tasks that invoke accusations that judges are simply asserting their own preferences and opinions rather than engaging in constitutional interpretation. They are also the kinds of tasks that the High Court of Australia has consistently held to belong to the domain of legislative power and not judicial power. The subject matter at issue in the electoral franchise cases is particularly apt to attract these kinds of objections: representative democracy is a topic that is extremely complicated, highly contested, and widely debated.
Step 3: Assess the Relative Merits of Statutory Sources
The last step in the analysis is to consider the relative merits of statutory sources. My suggestion here is that appreciating the objections to using contemporary understandings and values and substantive moral views to define the constitutional baseline that Australian judges are likely to encounter in this context — as I have just outlined — helps us see the attraction of a legislative constitutional baseline. I also want to suggest that understanding how statutory sources provide an alternative to these non-statutory extrinsic sources, which otherwise seem to be required to define the constitutional baseline, demonstrates that there are circumstances where legislative constitutional baselines may well be justified.
The High Court of Australia has not explicitly defended its use of legislation to define the constitutional baseline. Nevertheless, the Court has made several observations about the relevance of legislation in interpreting and applying the implied constitutional guarantee of representative democracy. These observations are also helpful in making sense of this interpretive practice. Generalising from the reasoning in the Court’s two most recent electoral franchise cases, Rowe and Murphy, legislation has been used in three different ways:
1. Evolution of the concept in response to social change: The High Court has used legislative developments to show how the concept of representative democracy has evolved over time, at least in part in response to changing social values and understandings about membership in the body politic. This includes the expansion of the electoral franchise through amendments to the Electoral Act.
2. Stability of the concept in the contemporary era: The High Court has presented the inclusiveness of the current electoral franchise, as defined by the Electoral Act, as a durable and readily identifiable social fact. It is significant in this regard that the electoral franchise has remained relatively unchanged for at least three decades, with the notable exception of the amendments challenged in Roach and Rowe.
3. Constitutional division of labour in defining the concept: The High Court has emphasised that representative government is an abstract concept, the concrete conception of which is largely defined by legislation. In doing so, the Court has suggested that this is a division of labour that is not only practically unavoidable, but that the Australian Constitution itself expressly contemplates, by conferring wide powers upon the Commonwealth Parliament to define the incidents of the electoral system.
Drawing on this set of observations about the relevance of legislation to the central interpretive problem posed by the constitutional requirement of choice ‘by the people’, I now want to advance two theses about why it is both plausible and defensible to use legislation for the purpose of defining a constitutional baseline in this context.
My first and primary thesis is that statutory sources can provide an alternative and less contentious means of relying on contemporary understandings and values and substantive views (and substantive moral views in particular). In effect, legislation functions as a ‘proxy’ for these otherwise problematic sources of constitutional meaning and helps overcome objections that beset invoking them directly. I will refer to this as the ‘proxy source thesis’.
My second and more tentative thesis is that, in some circumstances, there is an independent rationale for using statutory sources to define a constitutional baseline, which I suggest is analogous to the idea of ‘legitimate expectations’. This rationale is based on the premise that, for at least some constitutionally guaranteed goods:
(i) legislation more or less comprehensively defines the set of norms that structure human interaction concerning that good, and
(ii) the Australian Constitution expressly contemplates that legislation play this role, by allocating powers and responsibilities to the Commonwealth Parliament to give effect to that good.
I will refer to this as the ‘legitimate expectations thesis’. Although this article will not attempt to provide a full analysis or defence of this rationale for legislative constitutional baselines, it will suggest that the electoral franchise cases provide a fruitful starting point for interrogating this possibility.
(a) The Proxy Source Thesis: Legislation as a Proxy for Non-Statutory Extrinsic Sources
To begin with, legislation can potentially provide a more reliable and, therefore, less objectionable source of evidence about contemporary understandings and values than freestanding judicial determinations about popular beliefs and attitudes. Legislation is the product of the representative branches of government. It thus arguably reflects the understandings and values of the body politic — particularly when it is regularly amended and updated, as is the case with electoral legislation. Indeed, the relative ease of revising legislation in response to changing social needs and public opinion is one reason for allocating the institutional responsibility for certain goods, as a matter of constitutional design, to Parliament in the first place (a point that I will return to below in discussing the legitimate expectations thesis).
To be sure, these points also count against legislative constitutional baselines insofar as they hold Parliament to standards defined by the legislative status quo ante. However, Parliament is not strictly bound: legislative departures can be justified. Moreover, judicial scrutiny of legislative departures, conducted through limitations analysis, can, and typically does, account for the need for legislation to respond to changes in social needs and public opinion. For instance, this is a central consideration in judicial analysis of whether legislation pursues a ‘legitimate end’: a core component of both the ‘reasonably appropriate and adapted’ criterion and its variants, and of proportionality testing. This includes the ‘substantial reason’ test, where restrictions on the electoral franchise are examined in light of whether they are responsive to widely held notions of community membership, the rights and obligations of citizenship, and the capacity to exercise choice.
Therefore, where contemporary understandings and values are relevant to the meaning of a constitutionally guaranteed good — as they are in the context of the electoral franchise cases — consulting legislatively-defined norms and standards is arguably defensible because it helps courts overcome evidentiary hurdles that they would otherwise confront in attributing attitudes and beliefs to the body politic. In other words, courts do better by relying on legislation as a proxy for contemporary understandings and values than they do by trying to directly ascertain what beliefs and attitudes the public holds.
It is unclear, however, whether relying on legislation as a proxy for contemporary understandings and values renders this source of constitutional meaning more normatively defensible absent independent reasons for thinking that it is relevant to the interpretive problem. This is reflected in the dissenting judgment of Hayne J in Rowe, where his Honour complained that:
The content of the constitutional expression ‘directly chosen by the people’ neither depends upon, nor is informed by ... ‘common understanding’ or ‘generally accepted Australian standards’ ... . The ambit of the relevant constitutional powers is not set by the political mood of the time, or by what legislation may have been enacted in exercise of the powers. Political acceptance and political acceptability have no footing in established doctrines of constitutional interpretation.
The essence of this criticism is that, while legislation may well provide more reliable evidence of contemporary understandings and values, more reliable evidence of an illegitimate and therefore irrelevant source of constitutional meaning does not transform it into a legitimate source.
This is not the only way that legislation can serve as a proxy for non-statutory extrinsic sources, however. Legislation concerning the provision of a constitutionally guaranteed good can also be understood as giving effect to a substantive view on a topic related to that good. In the context of the electoral franchise cases, we are focused on substantive moral views. The same considerations arguably also apply to substantive non-moral views. As discussed above, however, these are generally not regarded with the same degree of suspicion in Australian interpretive practice. Using legislation as a proxy for substantive non-moral views thus may not provide as strong a justification for a legislative constitutional baseline, the idea being that there is no need to rely on a legislative ‘proxy’ where there is no real objection to relying on the source directly. Nevertheless, this possibility is worth bearing in mind.
Relying directly on a substantive view, whether moral or non-moral, to define a constitutional baseline requires judges to resolve several matters. They must identify the relevant epistemic community for the topic. They also must articulate the view and defend the attribution of the view to that epistemic community, explaining why the view is thought to be the most plausible or correct account of the topic within the epistemic community. As suggested above in Part IV, this presents the greatest difficulties for judges where the topic is contentious and subject to disagreement and debate within the epistemic community. Here, too, we should add that there may also be debate about what the relevant epistemic community is — for example, where there are multiple sources of expertise with different methodological commitments, assumptions, and objectives.
Legislation can help overcome these difficulties because it permits judges to rely on substantive views in an indirect way. By relying on statutory sources, judges can defer to parliamentary determinations about these matters. Moreover, given the deliberative capacities and informational resources that importantly distinguish legislatures from courts, Parliament is arguably better positioned to make these determinations. In the case of the implied guarantee, this means that the High Court can avoid taking sides in debates about the best account of representative democracy and why it requires universal adult suffrage — a task that sits uncomfortably with the modest conception of judicial role in Australian constitutionalism. Instead, the Court only needs to demonstrate that electoral legislation instantiates or gives effect to a particular substantive view — a task that uses ordinary methods of statutory interpretation that belong to the core judicial function. By giving the Court a way to defer to parliamentary judgment, relying on legislation as a ‘proxy’ makes invoking a substantive moral view to interpret a constitutional requirement more like invoking a substantive non-moral view.
(b) The Legitimate Expectations Thesis: Legislation as an Independent Source
When legislation serves as a proxy for interpretive sources that would otherwise be needed to define a constitutional baseline, it effectively ‘borrows’ the rationale for its relevance as a source of constitutional meaning from those other sources. Thus, in the example that we have been considering, one reason why electoral legislation is relevant to what choice ‘by the people’ requires is because it provides evidence both of popular beliefs and attitudes and of the best contemporary accounts of representative democracy. I now want to suggest, somewhat more tentatively, that there may be an independent rationale for legislative constitutional baselines that does not rely on legislation serving as a proxy for other interpretive sources.
The independent rationale for legislative constitutional baselines that I want to put forward here has to do with the concept of ‘legitimate expectations’. The basic idea is this: there are some subjects where legislation can be said to define a relatively comprehensive set of legal norms coordinating human conduct and, in doing so, to establish a network of mutually-reinforcing social expectations regarding that subject. Where that subject concerns a constitutionally guaranteed good — I want to suggest — legislation of that kind may be said to establish a set of ‘legitimate expectations’ with respect to the state’s provision of, and actions taken in relation to, that good, such that departures from the legislative status quo ante require justification. If this suggestion is sound, it would provide a distinctive rationale for legislative constitutional baselines (as opposed to other interpretive uses of legislation that might be imagined).
A few points of clarification about this proposal are in order. In the first place, the legitimate expectations rationale would only apply where there is legislation that can be said to perform the kind of social coordination function just described. In general terms, this plausibly describes the type of legislation sometimes referred to as ‘super-statutes’. The term ‘super-statues’ refers to legislation that establishes a normative or institutional framework with respect to some subject, such that it defines a set of ‘default’ or ‘background’ assumptions against which government conduct in relation to that subject is evaluated (both by members of the public and legal officials) as a matter of sociological fact. A second point is that the legitimate expectations rationale seems most compelling where it is the case not only that legislation in fact plays this role, but also that a constitution expressly contemplates that legislation play this role. This might be said to occur, for example, where a constitution allocates specific powers or even imposes duties on Parliament to enact legislation that defines the norms and standards, creates the institutions, and so on, which are necessary to realise or otherwise give effect to a particular constitutionally guaranteed good.
This situation plausibly describes Commonwealth electoral legislation in relation to the constitutional guarantee of representative democracy and the requirement of choice ‘by the people’. Indeed, these considerations help make sense of an observation that the High Court of Australia has often made about the relevance of legislation in the electoral franchise cases. As noted above, the High Court has emphasised the idea that the Australian Constitution, by design, allocates the primary institutional responsibility for defining the incidents of representative democracy — an otherwise abstract and indeterminate constitutional concept — to Parliament. This idea, I suggest, can be understood to go beyond the observation that, as a purely descriptive matter, most features of representative democracy are legislatively-defined. Consider, for example, some of Gordon J’s comments in Murphy. Despite sounding a note of caution about legislative constitutional baselines, her Honour emphasised the idea of a constitutionally-mandated division of labour. In particular, her Honour suggested that there is a sense in which the legislatively-defined electoral scheme is constitutive of choice ‘by the people’. Moreover, her Honour indicated that Commonwealth electoral legislation has a special status — or at least a different constitutional standing — that sets it apart from legislation regulating the incidents of political speech in the context of the implied freedom, because Parliament is under a constitutional obligation to enact and maintain electoral legislation.
This is not yet to defend the view that the High Court’s use of a legislative constitutional baseline in this context can be justified on a legitimate expectations basis. Nevertheless, I do want to suggest that it provides at least a partial and supporting justification. More importantly, I want to suggest that it provides fertile ground for exploring this rationale for legislative constitutional baselines in greater depth.
This article has drawn attention to constitutional baselines as an interpretive tool, and it has examined how the High Court of Australia uses legislation to define constitutional baselines. In doing so, it has deepened understanding of this interpretive practice, clarifying how constitutional baselines function and the challenge that using legislation to define them poses for traditional principles of constitutional law. Finally, the article has proposed an approach for evaluating legislative constitutional baselines, which requires identification of:
1. The categories of non-statutory extrinsic sources that are relevant to the interpretive problem;
2. The possible objections to the use of those sources; and
3. The extent to which legislation can either:
a. overcome those objections (the proxy source thesis); or
b. provide an independent basis for a constitutional baseline (the legitimate expectations thesis).
Applying this evaluative framework to the electoral franchise cases, I have demonstrated why legislation appears to provide the High Court with a plausible alternative to the non-statutory extrinsic sources that would otherwise be required to define the constitutional baseline, and why it may be justified on independent grounds as well.
This is not to suggest that legislative constitutional baselines will always (or even often) be defensible, but rather that they are capable of justification in at least some circumstances and therefore merit greater attention. I also do not mean to suggest that using legislation as a source of constitutional meaning, whether as a proxy or otherwise, is without operational challenges of its own. It is worth briefly mentioning some of these here, as they present issues that a more complete study of legislative constitutional baselines — one that surveys their use in other common law jurisdictions besides Australia — would need to address.
One difficulty has to do with commensurability. Legislation that purports to concern the same topic as a constitutionally guaranteed good may in fact concern a qualitatively distinct set of issues. Thus, to import its content into the constitutional domain may distort the distinctive constitutional content of the good. Another difficulty lies in determining which elements of a legislative scheme constitute the baseline. For instance, the dissenting judgments in Rowe suggested that the Act’s procedural changes to voter enrolment and transfer did not amount to a departure from the baseline because the affected individuals were in breach of their statutory obligations to enrol or transfer in a timely manner. This can be understood as a criticism about cherry-picking the statutory provisions that are relevant to defining the constitutional baseline. It is unclear whether these concerns pose more serious issues for statutory sources than for non-statutory extrinsic sources, which also give rise to incommensurability and cherry-picking problems. Nevertheless, both concerns have merit and require further consideration.
Finally, a more complete study would also need to consider the extent to which there are specific features of constitutional and interpretive practice that vary between different common law jurisdictions and that lend themselves to utilising legislative constitutional baselines. This article’s analysis suggests that within the Australian context, legislative constitutional baselines are responsive to a set of competing pressures that courts face when interpreting constitutional guarantees. On the one hand, the Australian Constitution is very old and has proven difficult to amend. This puts pressure on judges to update constitutional meaning to reflect changed circumstances. On the other hand, Australia’s legalistic interpretive tradition puts pressure on judges to resist this, and to be wary of interpretive sources that go beyond constitutional text and structure. The tensions that arise from these competing pressures are particularly acute in the case of the article’s central example: the concept of representative democracy and what choice ‘by the people’ requires are necessarily informed by social understandings and values, substantive views, and other facts that have changed over time.
A more comprehensive study thus might examine whether legislative constitutional baselines are utilised less frequently in jurisdictions where courts do not face these competing pressures. If so, then perhaps the Australian context provides the most fertile ground for evaluating this interpretive practice. Alternatively, if legislative constitutional baselines are used with the same frequency in other jurisdictions, then perhaps the practice is better understood as responsive to a species of a generic interpretive problem encountered by common law courts.
This article therefore provides an important foundation and starting point for future research that examines legislative constitutional baselines across different common law jurisdictions. Its analysis, while confined to Australian constitutional practice, holds broad interest for constitutional law and theory.
[∗] Senior Lecturer, Melbourne Law School, The University of Melbourne, Victoria, Australia. Email: email@example.com.
 It is perhaps worth noting that baselines are often conspicuously absent when it would be helpful to have them. Consider, for example, the recent report on levels of sexual harassment in universities in Australia: Australian Human Rights Commission, Change The Course: National Report on Sexual Assault and Sexual Harassment at Australian Universities (2017) <https://www.humanrights.gov.au/
node/14533>. Levels of harassment were mainly presented as ‘very high’, but without much by way of reference to a baseline. In evaluating the result, it would be helpful to know, for instance, whether levels of sexual harassment are higher in universities than in other kinds of institutions.
 I use this term in a capacious sense to refer to any constitutionally guaranteed interest, including but not limited to rights and freedoms.
 The phenomenon examined here is therefore distinct from the legislative baselines described by Eskridge and Ferejohn in their work on ‘super-statutes’: William N Eskridge Jr and John Ferejohn, A Republic of Statutes: The New American Constitution (Yale University Press, 2010). For Eskridge and Ferejohn, a legislative baseline is a social fact that exists independently of the interpretive practices of courts: it exists where a statutory standard ‘sticks’ in public culture, meaning that it is difficult to change (and is in this respect, they claim, ‘entrenched’), forming a widely assumed point of departure against which subsequent policies are evaluated and adopted: at 6–9.
 See Katharine G Young, ‘The Minimum Core of Economic and Social Rights: A Concept in Search of Content’ (2008) 33(1) Yale International Law Journal 113.
  HCA 22; (1998) 195 CLR 337.
 Ibid 366 .
 Ibid 367  (emphasis added).
 Amelia Simpson, ‘The High Court’s Conception of Discrimination: Origins, Applications, and Implications’  SydLawRw 10; (2007) 29(2) Sydney Law Review 263, 274, discussing Permanent Trustee Australia Ltd v Commissioner of State Revenue (Vic) (2004) 220 CLR 388.
 Simpson 274. See also 264–5.
 Conferring the power to make laws ‘with respect to ... taxation; but so as not to discriminate between States or parts of States’ (emphasis added).
 ‘The Commonwealth shall not, by any law or regulation of trade, commerce, or revenue, give preference to one State or any part thereof over another State or any part thereof’ (emphasis added).
 Conferring the power to make laws that ‘forbid, as to railways, any preference or discrimination by any State, or by any authority constituted under a State, if such preference or discrimination is undue and unreasonable, or unjust to any State’ (emphasis added).
 ‘A subject of the Queen, resident in any State, shall not be subject in any other State to any disability or discrimination which would not be equally applicable to him if he were a subject of the Queen resident in such other State.’ (emphasis added).
 As discussed in this article above at 485.
 Through the application of the test adopted in Cole v Whitfield, where the High Court unanimously held that the s 92 requirement that interstate trade be ‘absolutely free’ prohibits laws that discriminate against interstate trade and that are protectionist in effect, in the sense of conferring a competitive advantage:  HCA 18; (1988) 165 CLR 360, 393, 407–8. The analysis of both elements requires a baseline.
 Simpson (n 8) 265–6, 268–9, 285–6 (discussing s 92 and leading cases); 265, 273–7 (discussing s 99 and leading cases); 265, 267–8, 292–4 (discussing s 117 and leading cases).
 Although the High Court has acknowledged the baseline problem in this context, rather than attempting to define a standard for measure, the Court has instead focused on defining categories of exemption: see Lael K Weis, ‘Property’ in Cheryl Saunders and Adrienne Stone (eds), The Oxford Handbook of the Australian Constitution (Oxford University Press, 2018) 1013, 1018–20; Lael K Weis, ‘“On Just Terms,” Revisited’ (2017) 45(2) Federal Law Review 223, 240–5.
 Australian Constitution ss 7, 24.
 To be clear: in adopting this methodological stance, the article takes no position on whether this approach is desirable or correct.
 See Jeffrey Goldsworthy, ‘Australia: Devotion to Legalism’ in Jeffrey Goldsworthy (ed), Interpreting Constitutions: A Comparative Study (Oxford University Press, 2006) 106; Adrienne Stone, ‘Judicial Reasoning’ in Cheryl Saunders and Adrienne Stone (eds), The Oxford Handbook of the Australian Constitution (Oxford University Press, 2018) 472, 474–5.
 See Elisa Arcioni and Adrienne Stone, ‘The Small Brown Bird: Values and Aspirations in the Australian Constitution’ (2016) 14(1) International Journal of Constitutional Law 60, 75–6.
 See Patrick Emerton, ‘Ideas’ in Cheryl Saunders and Adrienne Stone (eds), The Oxford Handbook of the Australian Constitution (Oxford University Press, 2018) 141.
 See Jeffrey Goldsworthy, ‘Constitutional Implications and Freedom of Political Speech: A Reply to Stephen Donaghue’ (1997) 23(2) Monash University Law Review 367; Tom Campbell, ‘Democracy, Human Rights and Positive Law’  SydLawRw 16; (1994) 16(2) Sydney Law Review 195.
 Adrienne Stone, ‘Expression’ in Cheryl Saunders and Adrienne Stone (eds), The Oxford Handbook of the Australian Constitution (Oxford University Press, 2018) 952, 958–60.
  HCA 25; (1997) 189 CLR 520.
 (2010) 243 CLR 1 (‘Rowe’).
 Ibid 12  (French CJ), 79  (Hayne J).
 Commonwealth Electoral Legislation Amendment Act 1983 (Cth) ss 29, 45.
 Citizenship is an important qualification. As the case law and commentary generally refer to the baseline without this qualification, however, I follow that practice.
 Rowe (n 27) 18–19 – (French CJ), 48–9 – (Gummow and Bell JJ), 116–17 – (Crennan J).
 Rowe (n 27) 18  (emphasis altered).
 A-G (Cth) ex rel McKinlay v Commonwealth  HCA 53; (1975) 135 CLR 1, 36 (emphasis added), 37 (‘McKinlay’).
  HCA 48; (1996) 186 CLR 140, 166–7 (Brennan CJ), 201 (Toohey J), 287 (Gummow J).
  HCA 43; (2007) 233 CLR 162 (‘Roach’).
 Ibid 174  (Gleeson CJ); 198 , 199  (Gummow, Kirby and Crennan JJ).
 Ibid 186–7 , 197 , 198 – (Gummow, Kirby and Crennan JJ).
 Ibid 173–4 –.
 (2016) 261 CLR 28 (‘Murphy’).
 Ibid 39 , 53  (French CJ and Bell J); 58  (Kiefel J); 73–4 – (Gageler J); 75 , 82–3 – (Keane J); 102  105–6  (Nettle J); 111–2 , 126–7  (Gordon J).
 Ibid 54–5  (French CJ and Bell J); 44 , 60  (Kiefel J); 73–4 – (Gageler J); 75 , 86–7 –, 87 , 91–2  (Keane J); 105 –, 111  (Nettle J); 125 –, 126–7  (Gordon J).
 Ibid 39 , 46–7 –, 53–5 – (French CJ and Bell J); 68–70 –, 72–3 – (Gageler J); 74 –, 76 , 90–2 – (Keane J). It should be noted that, while endorsing the legislative constitutional baseline, Keane J expressed disagreement with how it was applied in Rowe: Murphy (n 40) 99–100 –. At the same time, his Honour suggested that longstanding legislative enactments, at least in the area of electoral law, cannot become constitutionally invalid due to ‘changes in the world’— a proposition that appears to be even more radical than the notion of a legislative constitutional baseline: Murphy (n 40) 92–3 –.
 Murphy (n 40) 60  (Kiefel J); 105 –, 111  (Nettle J).
 Ibid 125  (Gordon J).
 This aspect of her Honour’s judgment is discussed further in Part V, below. See ibid 114  (emphasising that fundamental features of representative democracy are legislatively–defined), 120  (suggesting that the electoral scheme is constitutive of ‘choice by the people’), 123 – (suggesting that electoral legislation has special status as ‘constitutionally obligatory’ legislation).
 See, eg, Rowe (n 27) 89  (Hayne J); 102  (Heydon J); 130–31 – (Kiefel J).
 See especially Anne Twomey, ‘Rowe v Electoral Commissioner — Evolution or Creationism?’  UQLawJl 9; (2012) 31(2) University of Queensland Law Journal 181.
 Most commentators expressly reject this, wanting to resist the notion that there is a constitutional imperative to maximise participation in voting or opportunities for participation: see, eg, Graeme Orr, ‘The Voting Rights Ratchet: Rowe v Electoral Commissioner’ (2011) 22(2) Public Law Review 83, 87–9; Twomey, ibid 184. As discussed above, the High Court unanimously rejected this reading of Rowe in Murphy in favour of the baseline reading.
 Although some, of course, do. I would count Hayne and Heydon JJ in this camp. It is also the view of James Allan: ‘The Three ‘R’s of Recent Australian Judicial Activism: Roach, Rowe and (No)‘Riginalism’  MelbULawRw 19; (2012) 36(2) Melbourne University Law Review 743, 745–6, 750–77.
 Twomey (n 48) 185.
 McKinlay (n 33) 36 (McTiernan and Jacobs JJ).
 Ibid 56 (Stephen J).
 Murphy (n 40) 69  (Gageler J).
 Langer v Commonwealth  HCA 43; (1996) 186 CLR 302, 342 (‘Langer’).
 Twomey (n 48) 184–5, 187 (describing the Court’s approach as a ‘one–way evolutionary track’). See also Orr (n 49) 87–8 (acknowledging this concern, but suggesting that the Court’s approach falls well short of this).
 Commonwealth Electoral Act 1973 (Cth); Commonwealth Electoral Act 1918 (Cth) s 93(1)(a).
 Commonwealth, Parliamentary Debates, House, 28 February 1973, 40–3 (Fred Daly).
 A proposal that has recently been given some consideration. A Bill to this effect was introduced by Greens Senator Jordan Steele–John on 18 June 2018: Commonwealth Electoral Amendment (Lowering Voting Age and Increasing Voter Participation) Bill 2018.
 Twomey alludes to this possibility: (n 48) 187–8.
 Rowe (n 27) 102 .
 Another variation on this complaint is that legislative constitutional baselines circumvent the Constitution’s formal amendment procedure, and thereby ‘usurp’ the power of the people to amend the Constitution: see Twomey (n 48) 181. However, this is a complaint about any interpretive practice that relies on extrinsic sources: see, eg, Jeffrey Goldsworthy, ‘The Case for Originalism’ in Grant Huscroft and Bradley W Miller (eds) The Challenge of Originalism: Theories of Constitutional Interpretation (Cambridge University Press, 2011) 42, 57–60. As discussed below, taken any further, this objection becomes an argument against legal constitutionalism.
 James Stellios, Zines’s The High Court and the Constitution (Federation Press, 6th ed, 2015) 332.
 See Roach (n 35) 174 – (Gleeson CJ), 199  (Gummow, Kirby and Crennan JJ); Rowe (n 27) 19–20 – (French CJ), 58 , 59 , 61 – (Gummow and Bell JJ), 119 , 120–1  (Crennan J).
 Contra Twomey (n 48) 184, 202. Indeed, this is true even on a strict approach.
 This is a complaint, for example, that is often made about the application of ‘strict scrutiny’ in American constitutional jurisprudence: namely, that it is ‘“strict” in theory, but fatal in fact’: Gerald Gunther, ‘The Supreme Court, 1971 Term — Foreword: In Search of Evolving Doctrine on a Changing Court: A Model for a Newer Equal Protection’ (1972) 86(1) Harvard Law Review 1, 8.
 This complaint is often made about the application of the first stage of the Oakes test in certain areas of Canadian constitutional jurisprudence: namely, that it effectively treats all legislative activity as a prima facie burden on a constitutionally protected interest. See, eg, Grégoire Webber, ‘Rights and Persons’ in Grégoire Webber et al, Legislated Rights: Securing Rights through Legislation (Cambridge University Press, 2018) 27, 36–9.
 Twomey acknowledges this, conceding that while ‘it is true that the form of entrenchment is not absolute ... it is still a significant limitation on Commonwealth legislative power’: Twomey (n 48) 189.
 See n 62.
 To these groups, we could also add any number of various other groups across time whose understandings and values are thought to bear on constitutional meaning: for instance, to show the progression or evolution of social understandings and values.
 See, eg, Betfair Pty Ltd v Western Australia  HCA 11; (2008) 234 CLR 418, 480–2 – (Gleeson CJ, Gummow, Kirby, Hayne, Crennan and Kiefel JJ).
 Cole v Whitfield (n 15) 392.
 See generally Eric Barendt, ‘Freedom of Expression’ in Michel Rosenfeld and András Sajó (eds)
The Oxford Handbook of Comparative Constitutional Law (2012) 891, 895–900.
 See McCloy v New South Wales  HCA 34; (2015) 257 CLR 178, 202–3 – (French CJ, Kiefel, Bell, and Keane JJ); 228–9 – (Gageler J); 258  (Nettle J); 283–4 – (Gordon J).
 See Adrienne Stone, ‘Rights, Personal Rights and Freedoms: The Nature of the Freedom of Political Communication’ (2001) 25(2) Melbourne University Law Review 37.
 Consider, for instance, the following statement in the Supreme Court of Canada by Dickson J:
A constitution is drafted with an eye to the future ... . It must, therefore, be capable of growth and development over time to meet new social, political and historical realities often unimagined by its framers. The judiciary is the guardian of the constitution and must, in interpreting its provisions, bear these considerations in mind.
Hunter v Southam  2 SCR 145, 155.
 See Stone, ‘Judicial Reasoning’ (n 20) 477–8. This is not, of course, to suggest that the High Court ought to be relatively more confident in relying on founding understandings and values. As Irving has persuasively argued, there is an important difference between consulting historical sources and doing history, and there are good reasons for being skeptical about how good judges are at the latter: Helen Irving, ‘Outsourcing the Law: History and the Disciplinary Limits of Constitutional Reasoning’ (2015) 84(3) Fordham Law Review 957.
 Again, this is not to suggest that the High Court ought to be confident in relying on this source.
 Compare the majority and dissenting judgments in Ha v New South Wales  HCA 34; (1997) 189 CLR 465, 491–5 (Brennan CJ, McHugh, Gummow and Kirby JJ), 505–7 (Dawson, Toohey and Gaudron JJ).
 Murphy (n 40) 69 .
 Langer (n 55) 342 (McHugh J).
 Murphy (n 40) 69  (Gageler J).
 See Allan (n 50); Rowe (n 27) 71–2 –, 76 –, 89  (Hayne J), 97–100 – (Heydon J). It is somewhat unclear, however, whether these critics are proposing a different constitutional baseline or simply refusing to define one.
 See above Part IIIA in this article.
 This is not, of course, to suggest that they cannot be justified within such a perspective: see, eg, Patrick Emerton, ‘Political Freedoms and Entitlements in the Australian Constitution — An Example of Referential Intentions Yielding Unintended Legal Consequences’  FedLawRw 6; (2010) 38(2) Federal Law Review 169.
 See, eg, Rowe (n 27) 18 – (French CJ), 105–6 – (Crennan J); Murphy (n 40) 69–70 – (Gageler J).
 See, eg, Rowe (n 27) 18–19 – (French CJ); Murphy (n 40) 39 , 53 , 54  (French CJ and Bell J); 69–70 –, 72–3 – (Gageler J); 74 – (Keane J). See also McKinlay (n 33) 36 (McTiernan and Jacobs JJ), Roach (n 35) 173–4 – (Gleeson CJ).
 See, eg, Rowe (n 27) 48–50 – (Gummow and Bell JJ); Murphy (n 40) 69–70 –, 70–1  (Gageler J); 76 , 81 , 82 , 86–7 – (Keane J); 106  (Nettle J); 113–4 –, 120 , 123 – (Gordon J).
 To be clear: I am invoking this concept in a general way to explain why legislation might be a plausible interpretive source for defining a constitutional baseline. This is not to be confused with the idea of ‘legitimate expectations’ as a ground for judicial review of administrative action, which the High Court of Australia has rejected: see Matthew Groves, ‘Substantive Legitimate Expectations in Australian Administrative Law’  MelbULawRw 15; (2008) 32(2) Melbourne University Law Review 470.
 For a detailed account of how proportionality reasoning and its constituent elements figure in Australian constitutional jurisprudence, see Anne Carter, ‘Proportionality in Australian Constitutional Law’, Proportionality and the Proof of Facts in Australian Constitutional Adjudication, ch 5 (PhD Thesis, The University of Melbourne, 2018).
 See above n 64 and accompanying text.
 Rowe (n 27) 89 .
 Again, to be clear: I am invoking this concept in a general way. The argument presented here is unrelated to issues concerning ‘legitimate expectations’ as ground of judicial review of administrative action: see n 89.
 Eskridge and Ferejohn (n 3). An important qualification in drawing this connection is that Eskridge and Ferejohn’s work on this topic has a different objective: whereas this article’s focus is on how courts use legislation to define constitutional baselines, their aim is to demonstrate that this particular genre of statutes should be regarded as ‘constitutional’ in their own right, irrespective of whether they play this kind of interpretive role: at 1–28.
 Murphy (n 40) 120 .
 Ibid 123 –. The suggestion appears to be that electoral legislation therefore warrants greater deference when exercising judicial scrutiny for compliance with the implied guarantee.
 See, eg, Rowe (n 27) 77  (Hayne J), 93 , 95–6  (Heydon J). Justice Gordon raises the same concern in a more general way in Murphy: (n 40) 123–4 –.