Structure over Text: the High Court splits on constitutional implications
16.08.23
Many foundational principles of Australian public law arise from constitutional implications, to the extent that that within the Constitution ‘much of the greatest importance is implied’ (McGraw-Hinds (Aust) Pty Ltd v Smith at 668 (Murphy J)). The High Court has identified implied principles protecting states as bodies politic (Melbourne Corporation v Commonwealth), affirming the separation between federal judicial power and non-judicial power (R v Kirby; Ex parte Boilermakers ’Society of Australia) and guaranteeing freedom of political communication (Nationwide News Pty Ltd v Wills).
In Zurich Insurance Company Ltd & Anor v Koper & Anor [2023] HCA 25 (Zurich), the High Court unanimously rejected an argument identifying a novel constitutional implication. The Court was divided, however, on the proper approach to drawing implications within the Constitution. That division maps onto longstanding debates regarding the proper approach to constitutional interpretation.
This post considers the different approach taken by each judgment on the point of implications in light of the dominant methodology of textualism, and questions the distinction between text and structure which informs their reasoning.
The decision
The dispute arose over a residential unit in Auckland owned by Mr Koper, and built by BMX NZ. Mr Koper obtained judgment against BMX NZ in the High Court of New Zealand in respect of defective design and construction of the building. He then brought proceedings in the Supreme Court of New South Wales against the appellants, BMX NZ’s insurers, seeking leave to bring substantive proceedings against them under the Civil Liability (Third Party Claims Against Insurers) Act 2017 (NSW).
Mr Koper's entitlement to bring proceedings in the Supreme Court of New South Wales depended on the proper interpretation of the Trans-Tasman Proceedings Act 2010 (Cth) (TTPA), a bilateral agreement relating to trans-Tasman court proceedings. Sections 9 and 10 of the TPPA provide that defendants in New Zealand can be served initiating documents issued by Australian courts, without the need for the court to give leave or be satisfied of any connection between the proceedings and Australia (at [21]).
Zurich argued that ss 9 and 10 of the TPPA could not validly apply to an initiating document issued by the NSW Supreme Court if it related to a matter within State jurisdiction. This argument was rejected both at first instance and by the NSW Court of Appeal. Before the High Court, the appellants argued for the recognition of an unrecognised constitutional implication, which limited the capacity of the Commonwealth Parliament. Specifically, it argued that the Commonwealth Parliament lacks legislative power to ‘alter the scope and reach of State judicial power’, except to the extent that it is expressly empowered to do so by s 51(xxiv), s 77(ii) or s 77(iii) of the Constitution.
All seven High Court Justices rejected that argument, writing two separate judgments. Kiefel CJ, Gageler, Gleeson and Jagot JJ wrote for the majority, with Gordon, Edelman and Steward JJ in the minority. The two judgments agreed that the appeal should be dismissed, but were split over the proper approach to constitutional implications. The significance of Zurich is therefore not to the proposed limitation on Commonwealth power, or any other specific constitutional implication, but rather the nature and derivation of constitutional implications generally.
The differing approach to implications
The majority judgment began by citing familiar pronouncements on implications from decisions such as McGinty v Western Australia (McGinty). In particular, the majority judgment emphasised Mason CJ’s observations in Australian Capital Television Pty Ltd v The Commonwealth (ACTV) at 135:
It may not be right to say that no implication will be made unless it is necessary. In cases where the implication is sought to be derived from the actual terms of the Constitution it may be sufficient that the relevant intention is manifested according to the accepted principles of interpretation. However, where the implication is structural rather than textual it is no doubt correct to say that the term sought to be implied must be logically or practically necessary for the preservation of the integrity of that structure.
The majority read this passage as distinguishing between two types of implication: textual and structural. For the latter, Mason CJ was interpreted as enforcing a test of logical and practical necessity for the preservation of the integrity of the constitutional structure (at [28]). Applying this to the dispute before them, the majority held: ‘The implied constitutional limitation on the capacity of the Commonwealth Parliament for which the Insurers argued fails to meet that threshold requirement for recognition’ (at [29]).
The minority judgment of Gordon, Edelman and Steward JJ began by quoting the same passage from Mason CJ in ACTV. They took issue with the majority’s reading of the passage, writing (at [42]):
Contrary to how that statement is applied in the Joint Reasons, Mason CJ was saying little more than that the process of constitutional interpretation is not mechanistic or literalist. The meaning conveyed by the words of the Constitution may be deeper and more nuanced than that which is revealed by the attribution of singular dictionary meanings to particular words.
The minority judgment further took issue with the sub-categorisation of constitutional implications advanced by the majority: ‘Contrary to what is said in the Joint Reasons, this Court in Lange v Australian Broadcasting Corporation and Gerner v Victoria did not endorse an approach of categorising constitutional implications as either ‘textual’ or ‘structural’ ... [T]he notion of necessity referred to by Mason CJ in ACTV is not a threshold requirement, it simply “reflects the need for any implication to be ‘securely based’ in the text and structure of the Constitution”’ (at [43], quoting Burns v Corbett at [175]).
Implications and textualism
Constitutional implications are a longstanding point of tension in Australian law. They were the subject of vehement disagreement in McGinty, where there was talk of the illegitimacy of deriving constitutional implications from representative democracy as though it was a ‘free-standing principle’ (234 (McHugh J), 169 (Brennan CJ), 188 (Dawson J)). Much of that unease can be drawn down to debates about construction, and fidelity with which implications adhere to the text; in LibertyWorks Inc v Commonwealth of Australia, Steward J doubted the existence of the implied freedom of political communication, suggesting that it ‘may not be sufficiently supported by the text, structure and context of the Constitution’ (at [249]).
The restrictive approach to implications is informed by concerns over the legitimacy of the judicial exercise, and the need to confine judicial choice. Dixon J’s pronouncement that implications must be ‘compelling’ (Melbourne Corporation at 82) and Mason CJ’s statement that they must be ‘securely based’ (ACTV at 135) reflect a desire to limit indeterminate exercises in drawing implications. This reluctance is identifiable in the language used to describe the process itself: whereas Dixon J acknowledged the element of judicial choice by talking of ‘making’ implications (Australian National Airways Pty Ltd v Commonwealth at 85), Windeyer J stated ‘I would prefer not to say “making implications”, because our avowed task is simply the revealing or uncovering of implications that are already there’ (Victoria v Commonwealth at 402).
As ever, this dovetails with broader debates as to the basic theory or method of constitutional interpretation. The narrowest doctrines of literalism leave little room for judges to ‘discover’ new implications from text on the page. As one moves down the interpretive scale, that scope of legitimate choice increases. Disregarding the determinacy of the original text and intent, there is far less constraint as to which constructions can be reasonably adopted. Implications with no direct textual basis—or those based on the structure of the Constitution—tend towards this end of the spectrum.
Neither judgment in Zurich encouraged a freewheeling methodology of interpretation in order to identify new constitutional implications. The majority judgment may have edged towards a more open perspective in recognising a sub-category of purely structural implications, which suggests implications without immediate textual referents. Yet the majority were eager to cut down any potential for open-textured decision-making by affirming that any implication must be ‘logically or practically necessary for the preservation of the integrity of the constitutional structure’ (at [28]). (The disjunctive here is curious, leaving open the possibility of implications which are logical but impractical, or illogical but practical.)
The minority judgment, on the other hand, may be read as adhering to a more restrictive tradition of constitutional implication, which rejects the notion of implications not anchored in the text. They said that the process of recognising implications could not be divorced from both the text and wider context of the instrument, echoing the modern approach to statutory interpretation. The minority judgment contains a pointed affirmation of the necessity of all these elements (at [44]):
[T]his Court was not confused in its unanimous decisions in Lange and Gerner by insisting upon the presence of both text and structure in the process of recognising implications.
Viewed in distinction to the majority judgment, this is a more stringent test, militating against flexible implications drawn on the basis of structure.
Insofar as this was a response to the majority’s approach, it may not be justified. As outlined above, the majority’s reading of Mason CJ’s judgment in ACTV was not itself supportive of an unbound approach to constitutional interpretation. Rather, the majority’s own stringent test of practical necessity indicates the continuation of an established practice. It is uncontroversial that doctrines such as parliamentary supremacy and aspects of the rule of law lack specific statement in the Constitution, yet have shaped implications in the past: see Theophanous at 196 (McHugh J), Kartinyeri v Commonwealth at 381 (Gummow and Hayne JJ) and Cheng v The Queen at 356-7 (Gaudron J).
The orthodoxy of the majority’s approach is emphasised by their reference to the Engineers’ Case in rejecting Zurich’s argument (at [30]). That decision entrenched an objective legalism, and inveighed against vague or ‘individual’ implications being drawn from the text (at 145). As Rosalind Dixon and Gabrielle Appleby have recognised, the Court’s judgment in the Engineers’ Case is presented in terms of ‘a methodological critique of earlier cases, as taking too broad and flexible an approach to constitutional interpretation generally and particularly to the drawing of constitutional implications’ (at 363). This injuncts against any reading of the majority judgment as urging for a departure towards more a more radical or subjective approach.
Conclusion: The myth of structural implications
One questionable assumption underlying both judgments is the idea of a separation between text and structure. The majority enforces the idea of a neat bifurcation by suggesting a different test for ‘wholly structural’ implications (at [28]). The minority, while rejecting that sub-categorisation, nonetheless maintains a conceptual separation between the two, stating that ‘[t]he structure of the Constitution is part of the context that is always considered together with the text’ and acknowledging that ‘the recognition of some implications will rely more heavily upon the text of the Constitution than upon its structure or other context’ (at [44]).
Each judgment continues the long-standing tendency of the High Court to refer to structural and textual implications independently. But that binary does not bear scrutiny. As Jeremy Kirk recognised almost a quarter-century ago, the two are inseparable: ‘In so far as judges are referring to implications arising from the general internal structuring (meaning the ordering) of the Constitution itself, they speak of multi-provisional or inductive implications’ (at 664). By the same token, references to external structures such as representative democracy or federalism are ‘established by a series of provisions’ (at 664). Referencing the Constitution’s structure as distinct from its text is like referencing painted figures independent of their brushstrokes: any neat distinction between the two is confused.
Once this is realised, the minority’s emphasis on text and structure appears tautological: unless one abandons the document entirely, it is impossible to draw structural implications distinct from the words that make them up. By the same token, the majority’s focus on ‘wholly structural’ implications raises the question of where, absent the wording of the Constitution, those implications arise.
It should be emphasised this was a case in which all Justices rejected the assertion of a constitutional implication. It is therefore not certain how the majority’s test will be used positively in future cases, or whether it will survive at all given the division in the Court on this point. It may be some time before a new implication is designed, drawn, or discovered.
Harry Sanderson is a law tutor at Corpus Christi College, Cambridge.
Suggested citation: Harry Sanderson, ‘Structure over Text: the High Court splits on constitutional implications’ on AUSPUBLAW (16 August 2023) <https://www.auspublaw.org/blog/2023/8/structure-over-text-the-high-court-splits-on-constitutional-implications/>