AustLII Home | Databases | WorldLII | Search | Feedback

Melbourne University Law Review

Melbourne Law School
You are here:  AustLII >> Databases >> Melbourne University Law Review >> 2018 >> [2018] MelbULawRw 6

Database Search | Name Search | Recent Articles | Noteup | LawCite | Author Info | Download | Help

Fella, Andrew --- "The Concept of Coherence in Australian Private Law" [2018] MelbULawRw 6; (2018) 41(3) Melbourne University Law Review 1160


THE CONCEPT OF COHERENCE IN
AUSTRALIAN PRIVATE LAW

Andrew Fell[1]*

The concept of ‘coherence’ has become a fundamental principle in Australian law. On numerous occasions, the High Court has emphasised the importance of coherence within the common law, and between statute and common law. But it has not fully explained what ‘coherence’ is. In most decisions in which coherence is mentioned, it refers to some kind of ‘consistency’ in the law, but this is quite unspecific. The Court has also not explained why it considers coherence to be important. This article attempts to explicate the concept of ‘coherence’, as it appears in decisions of the High Court. It argues that the High Court is attempting to achieve consistency in the law’s underlying normative reasons, and explains in more detail what this requires. It also identifies several arguments that potentially support the reliance on coherence in common law reasoning.

Contents

I Introduction

The concept of ‘coherence’ is a fundamental principle in Australian law.[1] It has been relied on by the High Court in statutory and constitutional interpretation,[2] but is relied on most often in the development and application of the common law. In that context, which will be the focus of this article, there are many cases in which the importance of ‘coherence’ within the common law, and also between the common law and statute law, has been acknowledged.[3] In those cases, it is generally regarded as a conclusive reason against a proposed ruling or decision that it would give rise to incoherence.[4] ‘Coherence’ has even been described as ‘the central policy consideration’ in determining the consequences of a plaintiff acting illegally in the circumstances giving rise to their claim.[5]

Despite its prominence, this ‘aspiration’ towards ‘[c]oherence within and between the common law and the statute law’ is ‘under-defined and under-studied’, as Chief Justice Robert French suggested in 2015.[6] The High Court itself has not fully articulated what it means by ‘coherence’,[7] and has said little about why it is important. Although some commentators have sought to elucidate the Court’s approach,[8] it is generally not well understood. At a general level, the Court indicated in Miller v Miller that a ‘lack of coherence’ involves some kind of ‘incongruity’, ‘contrariety’ or ‘inconsistency’ between legal rules.[9] But it did not explain the nature of this inconsistency in great detail, although it is clear that when statutory rules are concerned, their purpose is relevant in identifying it.[10] Subsequently, in Equuscorp v Haxton, ‘the objective of maintaining coherence in the law’ was equated with ‘the negative goal of avoiding self-stultification in the law’.[11] This was a reference to the notion of ‘stultification’ relied on by the late Professor Peter Birks, who suggested that it provided the basis of the illegality defence in the law of unjust enrichment.[12] Beyond this, the concept of ‘coherence’ or ‘consistency’ has not been explained. Complicating all this is that there are references to ‘coherence’ that clearly do not relate to the notion of consistency in the law.

If the High Court is going to rely on the ‘coherence of the law’ (understood as referring to some kind of consistency) in its decision-making, it is necessary to identify at least what the Court means, and the possible justificatory arguments that might support its approach. This is especially given the likelihood that coherence will become increasingly relevant in Australian legal adjudication. In contrast to past legislative practices, Grantham and Jensen note that modern legislation, when entering into common law fields, tends to establish comprehensive legal regimes that are intended to replace substantial parts of the common law but without codifying private law as a whole. In some areas, statutes therefore ‘exist side-by-side with remnants of the common law’.[13] As Barker points out, the resulting increase in overlap between common law and statutory norms is attended by an increased risk of contradiction between them.[14] Given that ‘[m]essy patterns of gradual, legislative accretion seem more likely than grand codification covering large fields’, he suggests that a proper understanding of the norms relating to ‘incompatibility’ between common law and statute is all the more important.[15]

But ‘coherence’ is not only relevant to the interaction between statute and common law. As Barker also observes, the novel applications of the (potentially expansive) principles of negligence and unjust enrichment, among other examples, highlight the potential for increasing overlap and contradiction purely within the common law. Australian courts rely on various ‘second-order’ principles, often in the name of ‘coherence’,[16] in order to prevent such conflicts between norms of different categories. There is, in his view, ‘an increasing need to research, review and rationalise the operation of these second-order principles as the complexity of the system continues to increase’.[17] A useful starting point in this process is to identify and examine the kind of inconsistency that these more specific principles have been tailored

to avoid.

The bulk of this article will be concerned with elucidating the concept of ‘coherence’, as used by the High Court in the development and application of the common law. In order to facilitate this discussion, I will provide a brief overview of the possible meanings of ‘coherence’ in Part II. Part III will consider its meaning in decisions of the High Court. There is a variety of such meanings, but on most occasions the High Court is referring to some kind of ‘consistency’ in the law. I will argue in Part IV that when used in this way, the High Court can be understood as referring to consistency in the law’s underlying normative reasons. This kind of consistency will be explained

in more detail in Part V, drawing on philosophical literature on the theory

of reasons and Ronald Dworkin’s writing on ‘political integrity’. Returning

to a more general level, Part VI then identifies some possible arguments

that might support the reliance on ‘coherence in the law’ in common

law reasoning.

II Coherence Generally

The word ‘coherent’ is part of ordinary language. It is also a concept that is used in specialised academic disciplines other than law.[18] But mostly, it is left undefined or vague.[19] For the purpose of clarity, it is worth pausing to provide an overview of its possible meanings before turning to the decisions of the High Court. Raz has suggested that ‘[w]hat is incoherent is unintelligible, because it is self-contradictory, fragmented, disjointed. What is coherent is intelligible, makes sense, is well-expressed, with all its bits hanging together.’[20] However, these different phrases are not perfectly synonymous with one another; the notion of ‘unintelligibility’ is conceptually distinct from that of ‘self-contradiction’, which is again different from ‘fragmentation’ and ‘disjointedness’. These differences are often overlooked. At least three distinct meanings of the word ‘coherent’ can be discerned in Raz’s statement. The first is ‘intelligible’, which means something like ‘capable of being understood or comprehended’. Secondly, similarly to the phrase ‘makes sense’, it can also mean something like ‘sound’ or ‘based on good reasons’.[21] Thirdly, as a relation between things or parts of a thing, it can mean ‘without contradiction or inconsistency’. Later, Raz also equated coherence and ‘unity’,[22] and this meaning can be added to the first three. This final meaning is also reflected in Raz’s explaining of ‘incoherence’ in terms of fragmentation and disjointedness. It conveys the notion of being joined together (not necessarily in a physical sense), and of singularity.

Although these meanings refer to conceptually distinct properties, ‘coherence’ of one kind can contribute to coherence of another. A set of random words is not capable of being comprehended because it lacks unity. A novel that is riddled with inconsistencies might be difficult to understand or follow. The operation and interaction of the various parts of a legislative scheme might be difficult to identify and comprehend if it does not reflect a sound and unified normative policy. There are no doubt other examples. The main point to note for present purposes is that this multiplicity of closely interrelated meanings undoubtedly contributes to a lack of precision in using and interpreting the concept. Bearing this range of possible meanings in mind will assist in understanding the decisions of the High Court.

III Coherence in Australian Law

The context in which the High Court most frequently refers to ‘coherence’ is in describing, developing and applying the common law. In that context, a variety of meanings can be discerned, which generally map to those identified in the previous section. Consider the recent decision in Prince Alfred College v ADC.[23] In that case, the majority observed that ‘[c]ommon law courts have struggled to identify a coherent basis for identifying the circumstances in which an employer should be held vicariously liable for negligent acts of an employee’.[24] There are several indications of what the word ‘coherent’ might mean here. Firstly, the heading immediately preceding the paragraph in which the statement appears questions whether there is a ‘general basis’ for vicarious liability. The majority also noted that a ‘general principle’ has ‘eluded the common law for a long time’.[25] Secondly, the majority suggested that a general principle might provide certainty, but that one partly based an exercise of judgment of what is ‘fair and just’ (like that utilised in the UK) would involve policy choices, on which minds may differ.[26] Such an approach would also ‘not proceed on any principled basis or by reference to previous decisions’.[27] Finally, repeating comments in an earlier decision, the majority observed that ‘a “fully satisfactory rationale for the imposition of vicarious liability” has been “slow to appear in the case law”’.[28]

Considering this, the majority, in referring to a ‘coherent’ basis for imposing vicarious liability, could variously be understood as referring to one which is general, certain in its application, non-discretionary, or based on a satisfactory normative rationale. The reference to generality can be seen as connected with the concept of unity. A ‘general’ basis for vicarious liability would be one that takes the form of a single principle that is capable of being stated and that, when applied, identifies when vicarious liability should be imposed and when it should not be. The references to certainty and non-discretion, in contrast, are appealing to the notion of intelligibility. One of the main features of a legal rule or concept is the way it applies in particular circumstances. Understanding how it applies is, therefore, one important aspect of understanding the rule or concept itself. If a legal concept or rule is vague, rendering its application uncertain, then it cannot be understood or comprehended to that extent. If its application depends on an exercise of discretion, then this also affects whether its application can be ‘understood’ in this sense.

Finally, the mention of difficulty in identifying a ‘coherent’ basis for vicarious liability can also be understood as connected with the suggestion that a satisfactory rationale — or, in other words, a ‘good reason’ for the imposition of vicarious liability — has been slow to appear. As mentioned earlier, ‘coherent’ can mean ‘sound’ or ‘based on good reasons’. The lack of a satisfactory rationale, in this sense, goes to whether vicarious liability is ‘coherent’. The Court could also be understood here as questioning the ‘intelligibility’ of vicarious liability, although in a different sense to that mentioned in the previous paragraph. A rule or principle that is not evidently based on any good normative reason, such as vicarious liability, is unintelligible in that it is difficult to understand why it exists or operates as it does. The majority’s use of the word ‘coherent’ could be understood as involving a reference to any or all of the abovementioned ideas in combination.

Each of these meanings can also be discerned in other decisions.[29] However, the High Court’s references to ‘coherence’ generally mean something else. Most can be understood as referring to some kind of ‘consistency’ in the law (although this in itself is still rather vague). As I mentioned at the beginning, there is a significant body of High Court jurisprudence suggesting that common law rules should be developed, altered and applied in a way that is conducive to ‘coherence’ (understood to mean ‘consistency’) within and between the common law and statute law. This is arguably the aspiration to which Chief Justice French was referring.[30] In saying this, I am not suggesting that the High Court is unconcerned about whether the law is intelligible, based on sound reasons and unified where this would be desirable. I am only pointing out that, when explicitly talking about ‘coherence’, most of the time the High Court is referring to some kind of ‘consistency’. I will discuss this kind of ‘coherence’ in the next part.

IV Coherence as Consistency

The High Court has relied extensively on this kind of ‘coherence’ in its reasoning. But it has not explained the notion in more detail. It is important that the concept is well understood if it is used in the Court’s decision-making, whether for the purpose of evaluating decisions in which it has been relied on, or to assist other judges or practitioners in applying coherence-based arguments. So in this and the next part, I will attempt to explicate it in more detail. In this part, I will consider some examples of ‘consistency-based’ reasoning. For now, the sole purpose is to understand, although in a general way, the Court’s references to ‘coherence’ in this context.

A Examples of Consistency-Based Reasoning

1 Sullivan v Moody

Although it had been relied on previously by Deane J in Hawkins v Clayton[31] and Gummow J in Hill v Van Erp,[32] the importance of consistency-based ‘coherence’ was first accepted by the Court as a whole in Sullivan v Moody.[33] Sullivan involved two joint appeals that had substantially similar facts. In each case, a child had been examined by a medical practitioner for evidence of possible sexual abuse. The practitioner concluded that the child had been abused, and reported this opinion to the relevant authorities in accordance with mandatory reporting legislation. A government social worker then assessed the child and reached the same conclusion. In each case, the father, who was the appellant, was initially regarded as the most likely suspect, but was never convicted of any offence. The appellants alleged that they suffered psychiatric harm and other loss as a consequence of the reports made by the medical practitioners and social workers, and argued that the examinations, diagnosis and reporting had been carried out negligently. The question was whether the fathers were owed a duty of care.

The Court referred to the need, in determining the existence of duties of care, ‘to preserve the coherence of other legal principles, or of a statutory scheme which governs certain conduct or relationships’.[34] There were two coherence-related issues that the Court identified in that particular case. Firstly, noticing that the appellants were alleging that they had been injured by the communication of information about them to others, it was said that the appellants’ complaint intersected with the law of defamation, which ‘resolves the competing interests of the parties through well-developed principles about privilege and the like’.[35] Applying the law of negligence so as to give rise to liability would resolve this competition ‘on an altogether different basis’, and ‘would allow recovery of damages for publishing statements to the discredit of a person where the law of defamation would not’.[36]

The second issue that the Court perceived as ‘present[ing] a question about coherence of the law’ was that the respondents had other responsibilities with which a duty of care would conflict.[37] The Court stated that ‘if a suggested duty of care would give rise to inconsistent obligations, that would ordinarily be a reason for denying that the duty exists’.[38] Relevantly, the professional and statutory functions of those examining the children required them to investigate into and report the facts ‘without apprehension as to possible adverse consequences’ to the potential perpetrators of sexual abuse, or legal liability to them.[39] A duty of care, in contrast, would tend to require those consequences to be avoided. Further, the statutory scheme also contained a general provision requiring the respondents ‘to treat the interests of the children as paramount’.[40] Their interests were said to be irreconcilable with the interests of those suspected of causing them harm. But, again, a duty of care would require the interests of the suspects to be furthered, sometimes at the expense of the children’s. As a result, the Court suggested that a duty of care owed to the fathers, as suspects, would be ‘inconsistent with the proper and effective discharge of those responsibilities’.[41] Given that ‘to find a duty of care would so cut across other legal principles as to impair their proper application’, it concluded that no such duty was owed.[42]

2 Miller v Miller

Alongside Sullivan, Miller v Miller is generally regarded as one of the paradigm instances of coherence-based reasoning in the High Court’s decision-making.[43] In that case, the plaintiff and defendant stole a car. The plaintiff, who was the passenger, was severely injured when the car crashed. The defendant was driving dangerously at the time. The question was whether the defendant owed the plaintiff a duty of care, despite the fact that the plaintiff was acting illegally. In deciding this issue, the majority suggested that ‘the central policy consideration at stake is the coherence of the law’.[44] This consideration was said to be important at two levels. Firstly, ‘the principles applied in relation to the tort of negligence [in determining the consequences of illegality] must be congruent with those applied in other areas of the civil law (most notably contract and trusts)’.[45] Secondly, ‘and more fundamentally’, the Court was also concerned as to whether any ‘incongruity’ between the statute proscribing the plaintiff’s conduct and the common law would arise if the defendant was held to have a cause of action for negligence.[46]

On this first issue, the majority noted that that illegality can sometimes render a contract or trust unenforceable, even when the statute does not explicitly or implicitly provide for this consequence. As Jacobs J noted in Yango Pastoral v First Chicago Australia, the refusal to enforce a contract in such a case is based on the ‘policy of the law’, the relevant policies being those contained in the applicable statute.[47] In determining the consequences of illegality, regard is had primarily ‘to the scope and purpose of the statutory provision, to consider whether the legislative purpose will be fulfilled without the courts regarding the contract as void and unenforceable’.[48] Mason J endorsed substantially the same approach.[49] This approach had also been adopted in subsequent cases relating to both trusts and contracts,[50] and the majority in Miller suggested that ‘the same path should be taken’ in cases of negligence.[51] Interestingly, the approach based on the ‘policy of the law’ was redescribed in terms of the subsequently developed principle of ‘coherence’. On this understanding, the basis of the Court’s refusal to enforce an obligation is the need to preserve the coherence of the law.[52] Like in the previous decisions, the Court stated: ‘It will be by reference to the relevant statute, and identification of its purposes, that any incongruity, contrariety or lack of coherence denying the existence of a duty of care will be found.’[53]

Turning to the present case, the fact that the purposes of the applicable statutory provision included the protection of property rights and the advancement of road safety did not ‘speak to’ the question of liability.[54] Rather, an incongruity stemmed ‘immediately’ from the fact that the plaintiff would herself be criminally responsible for the defendant’s dangerous driving, because that driving was a ‘probable consequence’ of the plaintiff and defendant’s ‘common intention to prosecute an unlawful purpose’.[55] For this reason, to conclude that the defendant owed a duty of care to the plaintiff would ‘not be consistent with the purpose of the statute proscribing dangerous driving’.[56] But even in cases in which dangerous or reckless driving does not eventuate, and the only offence committed by the plaintiff is illegal use, an incongruity between a duty of care and the statute would nevertheless arise ‘from the recognition that the purpose of the statute is to deter and punish using a vehicle in circumstances that often lead to reckless and dangerous driving’.[57] The argument seems to be based on the ‘association’ between the offences of dangerous driving and illegal use.[58] If the former would be inconsistent with a duty of care, the latter must also be, given its ‘association’ with the former. However, on the facts of the case, the plaintiff, by asking to be let out of the car on two occasions, had withdrawn from her illegal use of the vehicle. For this reason, there would be no incongruity in recognising that a duty of care was owed to her.[59]

3 Equuscorp v Haxton

This coherence-based approach to illegality was also adopted in Equuscorp v Haxton.[60] In that case, loans had been made to investors to facilitate their investment in a particular scheme. That scheme, however, was illegal, because its promoters had failed to register a prospectus. The loan contracts were thus held to be unenforceable, on the basis that they furthered an illegal purpose.[61] The question was whether the lender could obtain restitution of the loans despite the illegality.

In determining that issue, French CJ, Crennan and Kiefel JJ repeated the comment in Miller that ‘[t]he central policy consideration at stake ... is the coherence of the law’.[62] The question of whether the claims to recover the loaned money would give rise to incoherence depended ‘upon whether vindication of those claims would have frustrated or defeated, or have been inconsistent with, the statutory purpose’ of the prospectus provisions.[63] They noted Birks’s argument that the appropriateness of restitutionary relief in the context of an illegal contract should depend on whether the law would be self-stultifying if it allowed relief,[64] and suggested that ‘the negative goal of avoiding self-stultification in the law may be expressed positively as the objective of maintaining coherence in the law’.[65] In their view, allowing the claim would give rise to incoherence for several reasons.[66] Firstly, the lender was involved in the promotion of the scheme, and was not an arm’s-length financier. Secondly, the loans were an integral part of the scheme and furthered the illegal purpose. Thirdly, the purpose of the relevant statutory provisions was to protect those from whom recovery was sought (namely, potential investors who are otherwise incapable of obtaining sufficient information about the scheme). Arguably, their overall point is that the purpose of the statute is to protect the investors from people like the claimant. In their judgment, coherence required the claim to be rejected. Gummow and Bell JJ agreed with this conclusion, suggesting that a restitutionary action, if allowed, would ‘stultify’ the relevant statutory policy.[67]

4 Apotex v Sanofi-Aventis

The issue in Apotex was whether a method of medical treatment is a patentable invention.[68] In that case, the invention consisted in the use of an existing chemical compound for the purpose of medical treatment, for which it had previously not been used or known to be useful. Section 18(1)(a) of the Patents Act 1990 (Cth) specified that for an invention to be patentable, it must be ‘a manner of manufacture within the meaning of section 6 of the Statute of Monopolies’. French CJ observed that ‘the ascertainment, application and development of the principles determining whether a claimed invention is a “manner of manufacture”’ involve common law processes, rather than statutory interpretation.[69] One constraint in the development of common law principles is that an issue ‘involving large questions of public policy and reconciliation of interests in tension is, for the most part, best left to the legislature’.[70] In contrast, ‘a qualification or exception to a general principle may have become anomalous to such an extent that its removal would enhance the logical and/or normative coherence of the law’, and its removal, as part of the ‘endeavour to achieve coherence ... falls more readily within the institutional competence of the courts’.[71]

After considering the previous applications of the phrase ‘manner of manufacture’ to methods of medical treatment, French CJ suggested that the exception was indeed ‘an anomalous qualification on the principles governing patentability’.[72] He noted that ‘[t]he history of the exclusion of medical treatments from patentability does not disclose a stable, logical or normative foundation’ and that its exclusion is in logical and normative tension with the patentability of pharmaceutical products.[73] French CJ therefore concluded that methods of medical treatment are patentable, which ensured ‘the application of the rubric “manner of new manufacture” in a logically

and normatively coherent way’.[74] Interestingly, in D’Arcy v Myriad Genetics,

it was implied that a majority of the Court in Apotex had engaged in this

kind of reasoning, despite the fact that only French CJ referred explicitly

to ‘coherence’.[75]

B What Conception of ‘Consistency’?

It is clear that in these cases, the High Court is concerned to avoid some kind of inconsistency between different rules of law (or the consequences for which they provide). But the Court’s discussion in these and other decisions does not completely specify what it is that would make them ‘inconsistent’ in the relevant way. In this section, I will attempt to identify the kind of consistency to which the High Court is appealing. It will be necessary, firstly, to consider previous efforts to do so.

1 Consistency in the Operation of Rules?

Grantham and Jensen also regard the High Court, when talking about ‘coherence’, as referring to some kind of ‘consistency’. They note that ‘[i]n its most basic sense’, coherence requires that ‘[r]ules which belong to the same legal system must not prescribe different outcomes in relation to the same set of facts’, and suggest that ‘[i]t is in this most basic sense, as mere consistency in the operation of rules, that the High Court should be understood as having referred to coherence’.[76]

However, this understanding does not fit all of the abovementioned cases. Consider Apotex. French CJ suggested that there was a lack of ‘coherence’ between the rule that pharmaceuticals are patentable and the rule that methods of medical treatment are not. The inconsistency could not, however, have been that those rules provide different results, because there is no set of facts in relation to which they actually differ in the result they provide. In one unimportant sense, however, this last sentence is false. In a case involving a pharmaceutical, for example, the pharmaceutical rule would provide that the pharmaceutical is patentable, but the other would not, because it does not ‘apply’ to such a case. But I do not think that this kind of difference is what Grantham and Jensen are talking about. Arguably, they are suggesting that incoherence arises when, on a particular set of facts, two rules both apply and prescribe precisely opposite results. It is in that sense that they refer to ‘different’ outcomes. And on this understanding, the rules in Apotex did not provide for different results. The same can be said of the rules in Equuscorp. The actual text of the relevant statutory provision in that case did not (explicitly or implicitly) provide that the contractual or restitutionary claims were unenforceable. It said nothing about that issue. This being so, the problem could not have been that the common law of unjust enrichment provided a result different than that provided by the statute. If there was an inconsistency, it must have been of some other kind.

It is true that some instances of incoherence that are identified by the High Court do involve rules which prescribe opposite outcomes on the same set of facts. In Sullivan, the law of negligence (if it was held to give rise to liability) would have provided the opposite result to the law of defamation. But this is arguably not, by itself, what incoherence consists of. After all, the defence of qualified privilege could be understood as providing the opposite result to the ordinary rule against defamatory publications, and the High Court does not appear to regard this as involving any incoherence. It therefore seems necessary to point to something other than the difference in results provided by the rules in order to demonstrate their incoherence, although it is not immediately clear what this ‘something’ is.

2 Stultification

Birks’s notion of ‘stultification’ provides another possible lead. After all, the High Court suggested that ‘stultification’ and ‘incoherence’ are the same thing. On this assumption, a better understanding of ‘stultification’ facilitates a better understanding of ‘incoherence’. As will appear, this lead is a promising one.

According to Birks, ‘[t]o stultify the law elsewhere is to contradict it for no good reason’.[77] In discussing the various defences to unjust enrichment claims, he identified a category of ‘stultification’ defences which ‘aim to prevent the law of unjust enrichment from making nonsense of the law’s considered positions in other areas’.[78] This mention of the law’s ‘considered positions’ indicates that Birks is not referring solely to the content or consequences of the relevant legal rules, but to the considerations or reasons on which they are based. For example, Birks suggests that the defence of ‘bona fide purchase from a third party’ operates so as to not contradict the reason underlying the money exception to the nemo dat doctrine: namely, that it ensures confidence in the currency of money.[79] Allowing a personal claim for restitution of money against a bona fide purchaser would stultify the law’s reliance on this reason, and ‘would do nothing to ensure confidence in the currency of money’.[80] A personal claim for restitution would simply contradict the purportedly good reasons underlying the exception.

As the decision in Equuscorp indicates, Birks also observed that illegality can be a defence to an unjust enrichment claim. He argued that, in most cases, the basis of the defence is to avoid ‘the danger of stultifying the law’s refusal to enforce an illegal contract made between the parties’.[81] He suggested that the possibility of bringing an unjust enrichment claim to recover anything provided by one party under the contract is capable of providing ‘a lever to compel performance [of the illegal contract by the other party] and a safety-net in case that indirect compulsion fails’.[82] As an example, Birks referred to the decision in Boissevain v Weil,[83] in which the defendant had failed to repay a loan made in violation of exchange control regulations.[84] The possibility of bringing an unjust enrichment claim can be used as a kind of threat by the lender to compel the other party to repay the loan. And even if the threat is ineffective, the ability to obtain restitution provides the lender with a safety net, protecting them from the consequences of the other party’s non-repayment. Without the possibility of this safety net, the lender would be less likely to make any transfer under the illegal contract in the first place.

Again, this argument can only be understood as appealing to the reasons that underlie the refusal to enforce the contract, whatever they might be. A rule providing that contracts of a certain kind are illegal says nothing about restitutionary claims that might arise out of those contracts. The notions of levers and safety nets certainly go well beyond the content of or results provided by the rules themselves, and only become relevant once it is observed that certain legal rules or prohibitions are designed to influence people to act in particular ways. That the underlying normative reasons are Birks’s concern is clearly demonstrated by his subsequent examples. He observed that the possibility of an unjust enrichment claim will not stultify the relevant prohibition if such a claim ‘would compliment [sic] and fulfil the policy underlying the illegality’.[85] For example, a prohibition on contracts which require the lessee of residential premises to pay key money is undoubtedly intended to protect the lessee.[86] In other words, the purpose of such a statute would be to prevent those in the position of the landlord from exacting key money from the tenant. The possibility of allowing a tenant to recover key money if paid would actually facilitate rather than contradict the purpose of such a statute. This is why, in Birks’s view, allowing such a claim would not give rise to contradiction.

In that kind of case, there is no possibility of ‘stultification’ because the statutory purpose does not disfavour the claim. But even when it does, allowing the claim does not necessarily give rise to any inconsistency. Birks explicitly acknowledges that there can be good reasons for ‘contradicting’ the illegality by providing restitution to avoid some ‘greater evil’ that would follow from a denial of it.[87] According to Birks, any contradiction of this kind is only apparent or prima facie. His example is of a case in which an irregular migrant worker seeks to recover the value of their work despite the illegality of their employment contract. Even if the reasons underlying the illegality favour the rejection of the claim, ‘to refuse the non-contractual action would leave the immigrant with no remedy at all and open the way to slave-labour’.[88] So, even when the reason underlying the particular statute would favour the denial of liability, Birks’s conception of ‘stultification’ allows for the common law to give effect to a stronger yet competing reason without giving rise

to contradiction.

These provide some good examples of what the avoidance of stultification does and does not require. But the main point for present purposes is to observe what ‘stultification’ is. I think it is clear that, for Birks, ‘stultification’ consists of inconsistency in the normative reasons or considerations on which different legal rules are based. Conversely, the avoidance of stultification requires consistency in these reasons. I think that this understanding of ‘stultification’ is capable of illuminating the High Court’s conception of ‘coherence’ (as consistency). In all of the abovementioned examples (and many other decisions), the High Court can be understood as using the concept of ‘coherence’ to refer to the consistency of the law’s underlying reasons.

Consider Sullivan. Recall that there were two coherence-related issues in that case, the first concerning the intersection between the law of negligence and defamation, and the second the incompatibility of a duty of care with the professional and statutory duties of the medical practitioners and social workers. As to the first issue, I suggested above that there is nothing in the outcomes or consequences of the different rules that explains what the potential inconsistency consists of. On the present understanding, the explanation is obvious. The content of the law of defamation is based on considerations about the appropriate scope of liability for statements ‘to the discredit of a person’.[89] The reasons underlying the law of defamation, and, in particular, the reasons for having the defence of qualified privilege, would (potentially) be contradicted by the law of negligence if it allowed for liability in circumstances in which the defence precludes it.[90]

But the second issue, relating to the incompatibility of the proposed duty of care with other duties, appears quite different from the first. Nevertheless, it can still be understood as involving an inconsistency between reasons. At the most basic level, the imposition of any duty is based on a normative judgment that the conduct required by the duty should be performed instead of other conduct. If one duty requires a person to act in a way that breaches another of their duties, the reasons underlying each of the duties are at least prima facie inconsistent with each other. Each suggests that the conduct that each requires should be performed instead of the conduct required by the other. Further, even if one duty does not necessarily require a person to breach another duty, it might still require them to act in a way that is disfavoured by the reasons underlying the second duty. For example, the performance of one duty might make it practically much more difficult to perform another, even if it does not make it impossible. On this argument, the ‘incoherence’ given rise to in Sullivan by the imposition of inconsistent duties would, like in relation to the defamation issue, also have consisted of contradiction in the reasons underlying the different rules. The professional and statutory duties would have been imposed for particular reasons that would favour the performance of conduct required by the duties. To impose a duty of care, which required the person to act in a way that made it impossible or practically difficult to perform their other duties, would be inconsistent with those reasons.[91]

The decisions in Miller, Equuscorp and Apotex can be more easily understood as appealing to the notion of consistency in the law’s underlying reasons. The judges in Miller and Equuscorp in particular referred specifically to coherence as requiring consistency with the purpose of the applicable statute, which is simply a more specific way of referring to its underlying reason. In fact, the Court could not be understood in any other way, given that the statutory rules themselves did not expressly or impliedly affect civil liability. The same can be said of Apotex. More clearly in that case, there is no contradiction at all between the content of the rule relating to pharmaceuticals and the exception for methods of treatment. French CJ’s reference to the ‘normative tension’ between those rules indicates that he is instead referring to the normative reason or principle on which the pharmaceutical rule should be understood as based. Even though the forms of inconsistency that arose in the various cases were different, they can all be understood as involving an inconsistency in the law’s underlying reasons.[92] Although I will not discuss them here, I also suggest that this conception of coherence is capable of explaining every other decision of the High Court in which ‘coherence’ refers to some kind of consistency within or between the common law and statute law.

3 Alignment between Rules and Their Reasons?

Before moving on, it is worth discussing one other possible explanation. In a recent article, Hudson argued that the distinction between common law and equitable estoppels should be maintained in the interests of ‘coherence’, understood as ‘the alignment between an individual doctrine of law and its underlying rationale(s)’.[93] If this is an intended explanation of the High Court’s conception of coherence, I suggest that the explanation I offered in the previous section is preferable. More often than not, the decisions of the High Court in which the issue of coherence arises involve the intersection of different legal categories, for example, negligence and defamation. The issue in such a case is whether the reasons underlying each of those ‘doctrines’ are consistent with one another, not whether each of the individual doctrines aligns with its own underlying reason(s). Similarly, cases in which the alleged inconsistency is between the purpose of a statute and the common law cannot be explained in terms of Hudson’s conception. In those cases as well, the question is whether the common law provides a result that is inconsistent with the reasons underlying the statute, not whether the relevant rules align with their own underlying reasons. Again, I am not suggesting that the alignment between rules and their underlying reasons is not important. It is just not what the High Court is referring to in many of its decisions.

V A Closer Look at Reasons and Consistency

The reference to ‘stultification’ in Equuscorp provided the initial key in unlocking the Court’s approach, highlighting the relevance of the law’s underlying normative reasons. But even Birks’s conception was itself quite underdeveloped. What is needed is a more detailed understanding of normative ‘reasons’, and what is required for them to be ‘consistent’ when they are relied on in the law. It is then possible to consider the extent to which the decisions of the High Court properly apply these requirements.

A Normative Reasons

The account of moral or normative reasons that I will rely on here is Jonathan Dancy’s.[94] According to Dancy, reasons are ‘contributory’.[95] They ‘favour’ particular actions, but do not require them. In other words, if an action is favoured by a reason (or more than one reason), it does not necessarily follow that it ought to be done. There might be other, competing reasons that more strongly disfavour it. An action ought to be done only if it is most favoured on the balance of all the applicable contributory reasons. It is possible to speak of there being an ‘overall’ reason to do the action most favoured in this way. But an ‘overall’ reason is not a different or additional kind of reason. It is merely a way of describing ‘where the contributory reasons come down’.[96] In short, ‘there are no overall reasons’.[97] Normative reasons can only ever contribute to an overall ‘ought’.[98]

However, a particular descriptive ‘feature’ (for example, the fact that you made a promise to someone) does not necessarily give rise to a moral reason of equal strength whenever it obtains. Rather, the difference it makes depends on the context in which it appears. Dancy refers to this as the ‘holism’ of reasons.[99] In some cases, a promise gives rise to a moral reason to fulfil it. But in others, it would not. If the promise was given under duress, this, it might be argued, ‘disables’ it from giving rise to such a reason. The duress does not act as a reason against fulfilling the promise. It simply negates any moral reason that might otherwise have existed to fulfil it.[100] Similarly, some features can ‘intensify’ (or attenuate) the strength of the reason given rise to by another feature. The fact that someone nearby is in need of help provides a reason for you to help them, but that reason is stronger if you are the only person in a position to do so.[101] Finally, Dancy also argues that a particular feature might even give rise to a favouring reason in some circumstances, and a disfavouring reason in others. It might be good on some occasions to be helpful, but not if you are helping someone break into a car. Of course, Dancy acknowledges that there are some reasons that are close to ‘invariant’ in their ‘polarity’.[102] It is almost always a reason against a particular action that it would consist of the infliction of gratuitous pain on unwilling victims. But that is so because

of the reason’s specific content rather than some necessary truth about

reasons generally.

B Consistency between Normative Reasons

The points to take from all of this are that normative reasons are ‘contributory’ rather than ‘overall’, that there can be many applicable reasons in a particular case both for and against particular actions, and that normative reasons operate ‘holistically’: that is, with their strength and polarity potentially altered or negated by the context. This, I think, is sufficient to enable a useful discussion of what ‘consistency’ between reasons is or requires. But it is not necessary to start this discussion completely from scratch. Ronald Dworkin has considered in detail the notion of ‘coherence’ or ‘consistency’ in the law’s underlying reasons or principles. His account, in my view, conforms to the conception of normative reasons discussed in the previous section. It therefore provides a useful foundation. I will begin by outlining Dworkin’s account, and then identify in more detail the kind of consistency that Dworkin is referring to.[103]

Dworkin’s theory of law and legal adjudication is based on the notion of ‘political integrity’, which is his conception of ‘coherence’. Throughout Law’s Empire, he uses those phrases interchangeably, along with the phrase ‘consistency in principle’. He began his discussion by noting ‘the catch phrase that we must treat like cases alike’.[104] According to Dworkin, that idea ‘requires government to speak with one voice, to act in a principled and coherent manner toward all its citizens, to extend to everyone the substantive standards of justice or fairness it uses for some’.[105] Dworkin suggested, however, that this demand of political morality is not well described in the abovementioned ‘catch phrase’.[106] He instead referred to it as ‘political integrity’, in order ‘to show its connection to a parallel ideal of personal morality’.[107] ‘Personal’ integrity, according to Dworkin, requires that people act ‘according to convictions that inform and shape their lives as a whole, rather than capriciously or whimsically’.[108] When understood as a political notion, ‘it is inconsistency in principle among the acts of the state personified that integrity condemns’.[109]

More concretely, Dworkin suggests that ‘[c]heckerboard statutes are the most dramatic violations of the ideal of integrity’.[110] Elsewhere, he describes them as ‘an obvious way of infringing a general principle of coherence’.[111] According to Dworkin, a ‘checkerboard’ law is one which treats similar cases differently ‘on arbitrary grounds’.[112] A good example is a statute that permits abortion for women born in odd years but prohibits it for those born in even years. But what is the ‘inconsistency in principle’ in such a statute? The problem is not that the statute is inconsistent with correct moral principles. Dworkin demonstrates this by reference to a person who regards abortion as always wrong, even when the pregnancy is the result of rape. To such a person, a statute that prohibited abortion except in cases of rape would be morally incorrect (in part). But they would probably not regard it as involving the same kind of inconsistency as the checkerboard rule.[113] There is something more blatantly inconsistent about it.

The problem is more that, in enacting such a statute, the state ‘must endorse principles to justify part of what it has done that it must reject to justify the rest’.[114] But this somewhat bare statement needs clarifying, particularly as to the words ‘endorse’ and ‘reject’. Dworkin is not suggesting that, if the state ‘endorses’ a principle to justify one decision, it then ‘rejects’ that principle if it makes a decision on another issue that is disfavoured by that principle. This is clear from Dworkin’s suggestion that the rape-abortion statute involves no inconsistency. Dworkin understands that statute as ‘giv[ing] effect to two recognizable principles of justice, ordered in a certain way’.[115] Presumably the two principles are that abortion is wrong and that a person should be able to abort a pregnancy that is the result of rape. Clearly, the former principle is not given complete effect. In some circumstances, the statute provides a result that is not favoured by that principle. But Dworkin does not perceive this as involving any inconsistency. It seems to be regarded as permissible to give effect to the second principle at the expense of the first when the two conflict. In other words, consistency does not require a principle to be ‘relied on’ whenever it applies.

Indeed, on two other occasions in Law’s Empire, Dworkin emphasises that integrity does not rule out ‘competition’ as opposed to ‘contradiction’ between principles. Consider one of his examples, drawn from the law of ‘accidents’. He posits two moral principles, one which suggests that liability should be attributed to those responsible for harm, and the other which holds that people must be protected from being ruined by massive liability.[116] Sometimes, those principles will conflict in the result they favour in a particular case. It would be impossible to give complete effect to both. But this does not mean that recognising both gives rise to ‘incoherence’. He suggests, rather, that ‘any moral vision would be defective if it wholly disowned either impulse’, and that ‘it would be a serious misunderstanding of the logic of principle to consider them contradictory’.[117]

This reference to the ‘logic’ of principle directs attention to Dworkin’s well-known ‘logical distinction’ between rules and principles, according to which rules are applicable in an ‘all-or-nothing fashion’ and principles have a ‘dimension of weight’.[118] The fundamental point is that a principle ‘states a reason that argues in one direction, but does not necessitate a particular decision’.[119] As Dworkin observes, ‘[t]here may be other principles or policies arguing in the other direction’.[120] In a particular case, the applicable ‘abstract’ principles must be weighed or compromised against each other in some (‘nonarbitrary’)[121] way to determine the ‘concrete’ outcome.[122] This, I think, is saying nothing more and nothing less than that principles and reasons operate in a ‘contributory’ rather than ‘overall’ way, as Dancy emphasised.

Given that reasons are contributory, there is nothing necessarily inconsistent in the law relying on reasons that conflict. A principle against abortion might provide a conclusive reason in some cases, but it might be outranked in others. There is nothing inconsistent in this. And as Dworkin’s mention of ‘compromise’ indicates, there is also nothing inconsistent in compromises between two or more principles or reasons, neither of which is given complete effect in a particular case. Dworkin posits an example involving a decision about inheritance tax.[123] He notes that there are two applicable principles, one which suggests that people can do what they want with their property, and another suggesting that people should begin life on equal terms. On the question of whether rich people should be allowed to leave their wealth to their children in their will, these principles compete. A scheme of inheritance tax might give effect to both principles in a limited way ‘by setting rates of tax that are less than confiscatory’.[124] This would not involve a violation of integrity. But when a particular legal solution on some occasion strikes a balance between the two principles, integrity then requires that ‘whatever relative weighting of the two principles the solution assumes must flow throughout the scheme’, if the particular solution is part of a broader scheme, ‘and that other decisions, on other matters that involve the same two principles, respect that weighting as well’.[125]

This notion of ‘relative weight’ can be used, I argue, to explain the requirements of consistency more generally. When a decision is made about what ought to be done, a judgment is made about the relative ‘weight or importance’ of all the reasons that apply to it.[126] This will require either that one or more be ranked above others, or that a compromise be found. A decision cannot be understood solely in terms of the reason or reasons that favour it, because there are no overall reasons that are capable of justifying a decision by their mere application. Obviously, when looking at some decisions, it will be possible to identify the reason that was regarded as conclusive, but this should not be allowed to conceal what is actually going on. The judgment is that the reasons favouring the decision are stronger than those disfavouring it.

For sure, on this understanding, a single decision involves a judgment about many different reasons. Even a statute prohibiting murder, for example, contains a judgment about the preferences of murderers, and the strength of the reason they provide against the statute. However, even though the judgment is about many different reasons, it is only about the reasons that actually favour or disfavour the decision. In making a decision, no judgment is made about reasons that do not apply to it.

Considering all of this, it is possible to explain why Dworkin’s checkerboard statute involves an inconsistency, and the rape-abortion statute does not. If the state decides that it will permit abortions for women born in odd years, it judges that that action is more favoured on the balance of contributory reasons than not. It judges that the reasons against permitting abortion were not a strong as those in favour of it. But when the state then prohibits abortions for women born in even years, it acts in a way that is inconsistent with that judgment. The same set of reasons, no more and no less, applies to that decision. It can only be justified, if at all, by relying on a judgment about the relative weight of the applicable reasons that is different than the judgment in the former cases. It judges one reason to be stronger than another in one case, and then judges the very same reasons to have a different relative strength in another. That, in my view, is the relevant inconsistency. It is not that a decision was reached that is disfavoured by a reason ‘relied on’ elsewhere. It is inconsistency in the judgments about the weight of the same reasons relative to each other that is relevant.

But must the statute be understood as involving this kind of inconsistency? Could it not be understood as reflecting a judgment that ‘the oddness or evenness of the year of one’s birth carries ... moral significance’?[127] If it did, it would not involve the kind of inconsistency between judgments that is relevant here. In fact, any apparently inconsistent statute could be understood as only involving a moral error, rather than some other special defect. The problem with that argument, however, is that it would be wrong to attribute the principle about oddness and evenness of years to the legislature. The principles that lawmakers can be understood as giving effect to must be principles ‘that would be recognized as a moral principle [in our community], so that it could make sense to attribute that principle to its law-makers’.[128] So the abortion statute cannot be understood as giving effect to the year principle, because it is not regarded as even a possibly correct moral principle in our community. That is why the statute can be understood as involving an inconsistency rather than the reliance on an incorrect principle.

It is also clear that the rape-abortion statute, in contrast, does not contain an inconsistency. In ordinary cases, in which a pregnancy is not the result of rape, the rape principle simply does not apply. There are reasons favouring the prohibition, and reasons against it, but the rape principle falls into neither category. The decision to prohibit abortion in ordinary cases therefore makes no judgment about it. In then deciding to permit abortion in cases of rape, the rape principle of course applies. The making of that decision requires the rape principle to be regarded as stronger than the reasons against that decision, including the principle that abortion is wrong. But in doing so, the rape principle need not be regarded as stronger than any reason that the former decision (about ordinary cases) judges it to be weaker than, because it simply does not apply to the former decision.

Of course, even when the same set of reasons applies to different decisions, it is necessary to bear in mind the fact of ‘reasons holism’ that I mentioned earlier (that a particular reason can have different weight and polarity depending on the circumstances).[129] Even if a reason is judged to be stronger than another in one case, but weaker in a different case, it is possible that this can be justified on the basis of some other difference in the features of the case that affects their relative weight. The crucial point, however, is that there is more to the question of consistency than identifying the ‘main’ reason that is relied on to justify a decision or action. It is necessary to identify what other reasons were judged to be outweighed by that reason. What consistency requires is that the law not give effect to inconsistent judgments about the relative weight of the same reasons.

There are several other points worth noting. Firstly, Dworkin was only discussing the notion of consistency between principles, but I think that what I have said here applies equally to all kinds of normative reasons, including reasons of ‘policy’. In the passage in which he suggests that principles are reasons arguing in one direction, he also appears to contemplate that reasons of policy operate in the same way.[130] That is what matters for present purposes. The only difference is that holism is more likely to be relevant in relation to reasons of policy. The fact that a particular policy has been pursued in the

past is likely to affect the weight of that policy in future cases, whereas it is

the nature of reasons of principle that their strength does not diminish in

this way.

Secondly, the discussion here should not be understood as suggesting that only blatant checkerboard solutions are capable of giving rise to inconsistency. As Dworkin suggests,

[i]ntegrity is flouted not only in specific compromises of that character ...
but whenever a community enacts and enforces different laws each of which
is coherent in itself, but which cannot be defended together as expressing a
coherent ranking of different principles of justice or fairness or procedural
due process.[131]

He suggests that ‘our own legal structure constantly violates integrity in this less dramatic way’.[132] Given that true checkerboard solutions are likely to be rare, most instances of incoherence, then, will probably be of this subtle kind. Indeed, most of the decisions of the High Court have been concerned with these more subtle inconsistencies.

C High Court Examples

1 Sullivan v Moody

These considerations can be applied in examining and explaining some of the High Court’s decisions. Consider the defamation issue in Sullivan. The High Court suggested that the tort of negligence would, if it was held to provide for liability, operate inconsistently with the law of defamation. If the claim was for defamation, the defendants could have relied on the defence of qualified privilege. That defence is only negated if the defendant acted with malice (which there was no evidence of in Sullivan). Negligence, which is generally regarded as a lower level of fault, is not enough. However, negligence is (obviously) the only kind of fault that is needed in the tort of negligence. If the defendant was held liable for negligence, would this be inconsistent with the law of defamation?

Not necessarily. The defence of qualified privilege is based on the ‘public interest in permitting [people] to communicate frankly and freely with one another about matters in respect of which the law recognises that they have a duty to perform or an interest to protect in doing so’.[133] The law judges the abovementioned public interest to be stronger than any liability-favouring reason in cases of defamation, even those in which the defendant is negligent. But the facts that give rise to a claim for the tort of negligence are different than those that are relevant to a claim for defamation. In particular, the law of negligence focuses on the foreseeability of the harm and the relationship between the parties. It also responds to different kinds of harm. It is possible that those facts, when combined with the defendant’s negligence, give rise to a different kind of liability-favouring reason that is not present when the elements of defamation obtain, even if the defendant is negligent.[134] If they do, it is not sufficient to point out that the reasons underlying the defence favour the rejection of liability, or have been judged to outweigh the liability-favouring reason in defamation cases. Inconsistency only arises when the same reasons are weighted inconsistently. If the reason arising in cases of negligence is different, the defence of qualified privilege does not reflect a judgment about its weight relative to the abovementioned public interest. Providing for liability through the law of negligence, therefore, would not necessarily be inconsistent with the defence of qualified privilege. Further, even if the relevant liability-favouring reason is the same as in ordinary defamation cases, the additional facts might intensify that reason. There would be no inconsistency in allowing negligence claims if the intensified reason is stronger than that underlying the qualified privilege.

This is not necessarily to suggest that the decision was wrong. It is possible that the additional facts obtaining in cases of negligence do not either give rise to any different kind of reason or intensify the ordinary reason, in which case it would be inconsistent to allow negligence but not defamation claims. And even if they do give rise to a different reason, that reason might be no stronger than the liability-favouring reason in cases of defamation. One of these propositions must be true if the High Court’s conclusion is to be sustained. But neither appears to have been put forward in the decision.

2 Miller v Miller

In Miller, the Court concluded that a duty of care would give rise to incoherence with both the plaintiff’s criminal liability for the defendant’s dangerous driving and for her own illegal use of the vehicle. As to the cases in which the plaintiff is criminally responsible for dangerous driving, the majority did not explain why there is an ‘incongruity’ between the purpose of the statute and a duty of care. It only suggested that this conclusion was ‘evident’.[135] The Court might be suggesting that the plaintiff, given her criminal responsibility for dangerous driving, was responsible for her own harm in some sense. But even if this is right, it is hard to see how it leads to the conclusion that a duty of care would be inconsistent with the purpose of the statute. Presumably the purpose of the statute relates in some way to the deterrence and punishment of dangerous driving. Maybe it is possible to argue that the recognition of a duty of care would hinder these goals, but I do not think that this is the Court’s argument on this occasion. The Court is pointing more to the fact that the plaintiff was criminally responsible for the very conduct that caused her own harm. Again, it is not clear how this relates to the purpose of the statute, or to some inconsistency with it. In my view, it does not.[136] If the Court’s argument to the contrary is doubted, this would also have implications for its conclusion about the inconsistency of a duty of care with the illegal use provision. Given that it was regarded as following from the conclusion about reckless and dangerous driving, it must also be doubted.

What if the Court is instead interpreted as appealing to the purposes of deterrence and punishment, and the inconsistency of a duty of care with these purposes? After all, it did suggest that the inconsistency of a duty of care with the illegal use provision ‘arises from the recognition that the purpose of the statute is to deter and punish using a vehicle in circumstances that often lead to reckless and dangerous driving’ (although I think that the Court in saying this was pointing more to the association between illegal use and dangerous driving than to the purposes of punishment and deterrence).[137] It can be accepted that those purposes favour the rejection of a duty of care. Recognising a duty would, if anything, induce rather than deter people from illegally using vehicles. It would also lessen the punishment suffered by an offender like the plaintiff. Rejecting the duty would therefore seem to promote the purposes of the statute.

But these purposes should not be treated in an ‘overall’ way. As I have emphasised, normative reasons are contributory rather than overall. They only ‘favour’ particular outcomes without necessarily requiring them. A decision about what ought to be done involves a normative judgment about the balance of contributory reasons in the particular circumstances, rather than the selection of a single reason that must be relied on whenever it applies. Consistency only requires that these normative judgments are respected. In Miller, the legislature had judged the reasons that favour the illegal use provision as outranking the reasons against it. But what actually were those reasons? Whatever they might be, it is unlikely that they include the reason that favours liability for negligence. It would seem very odd to argue against either the prohibition itself, or the penalty for its breach, by appealing to that reason. If so, the statute contains no judgment about it. There is therefore no inconsistency in reaching the result that is favoured by that reason (ie recognising a duty of care), even if the ‘purposes’ of the statute favour against that result.

Further, attempting to reinforce the purposes of punishment and deterrence by rejecting the duty might even give rise to incoherence.[138] The statutory penalty itself reflects a normative judgment on the balance of contributory reasons. There is a whole range of possible penalties, from very small to very large. There are reasons that favour a smaller penalty and others that favour a larger one. That the penalty would punish or deter the offender are reasons in the latter category. The amount of the penalty reflects a judgment about how those reasons are best balanced against the competing reasons that favour a smaller penalty, analogously to Dworkin’s inheritance tax example. That weighting of the competing reasons must be respected by the common law. For the common law to provide an additional mechanism of punishment or deterrence would be to rely on a different and therefore inconsistent relative weighting of the reasons applicable in deciding the magnitude of the penalty. After all, the particular weighting reflected in the statute, unless some mistake has been made in implementing it, favours the penalty in fact provided by the statute, no more and no less.

This kind of argument does not apply only to penalties, but to any statutory consequence that is intended to give legal effect to a prohibition.[139] It suggests that ‘coherence [in] the law is best served by a court respecting and enforcing [the] legislative choice’ of the appropriate consequences, not ‘supplementing’ them.[140] There are some cases in which additional consequences might be needed, such as those in which civil liability would effectively shift a criminal penalty from the plaintiff to the defendant,[141] or require the performance of a crime.[142] Nevertheless, when the statute does not explicitly negate civil claims, courts should be slow to deny them.[143]

The general point is that cases of illegality should not be approached by identifying the statutory purpose and determining whether that purpose would be furthered by rejecting the claim. Grantham and Jensen suggest that such an approach is ‘not particularly helpful’ and ‘may be positively harmful’, given that ‘[t]he form which a statute takes may be the result of a series of compromises among several competing goals’.[144] Further, even when there is no compromise, a statute only contains a judgment about the weight of its purpose relative to the reasons that actually disfavour it. Consistency only requires that these judgments, whether involving a compromise or ranking, are respected. It does not require any purpose to be treated as conclusive whenever it applies.

3 Equuscorp v Haxton

In making this argument, Grantham and Jensen referred to the dissenting judgment of Heydon J in Equuscorp.[145] There, Heydon J emphasised the ‘primacy of legislative language’ in determining the effect of illegality on the enforceability of the claim.[146] In his view, ‘[t]he scope and purpose of the statute depend solely on the meaning of its language’.[147] He denied that the claims were negated by the statute, given the lack of express or clear words to that effect,[148] the unjustness of the result if they were denied,[149] the provision of onerous sanctions in the statute itself[150] and the contingent relation of the loans to the illegal scheme.[151] His approach contrasts with that of the majority, who relied heavily on the broadly stated purpose of ‘protecting investors’. It was probably correct to regard that as part of the statutory purpose. It favours the prohibition on the issuing of interests without a prospectus, and the imposition of penalties for the breach of that prohibition. But, again, consistency does not necessarily require the result that is favoured by that purpose, only that judgments about the weight of that purpose relative to other reasons are respected.

In cases involving a failure of basis, the reason that favours restitution is based on the imperfection of the plaintiff’s intention or consent to enrich the defendant.[152] Like in Miller, it is unlikely that the statute in Equuscorp contains a judgment that ranks that reason below the purpose of protecting investors. The restitution-favouring reason does not disfavour either the relevant prohibition or the penalties for its breach. So even if the restitution-favouring reason competes with the purpose of protecting investors,[153] the statute does not contain a judgment that ranks it below that purpose. There is therefore no inconsistency in permitting restitution.[154] This highlights why Heydon J was right to emphasise the importance of the statutory text in determining the consequences of illegality. As well as indicating the positive purpose of the statute, it will also reveal the reasons about which a judgment of weight relative to the positive purpose has been made. It is only those reasons to which the common law cannot give effect at the expense of the statutory purpose.

4 Apotex v Sanofi-Aventis

In light of the earlier discussion, it is possible to explain in more detail the argument that French CJ made in Apotex. When considering whether pharmaceutical products can be the subject of a patent, a particular set of contributory reasons applies. There is judged to be an overall reason in favour of their patentability. What French CJ can be understood as suggesting is that there is nothing in the differences between pharmaceuticals and methods of treatment that changes the set of applicable reasons, or their strength. If methods of medical treatment are excluded from patentability and pharmaceuticals are not, the law can therefore be understood as relying on an inconsistent normative judgment about the relative weight of the applicable contributory reasons.

VI Why Is Consistency (between Reasons) Important?

All of this, however, raises the important question of whether the High Court should aspire towards this kind of coherence in the law. As I mentioned at the beginning, the High Court itself has said little in support of this aspiration. But if it is to remain as a fundamental principle of Australian law, having at least an idea of the potential justificatory arguments in support of it is a necessity. Fortunately, there is a significant body of academic literature relating to the notion of ‘coherence’ in the law, although not specifically relating to the High Court’s approach. This literature contains many arguments that, at least prima facie, can be applied in favour of the High Court’s conception. In the scope of this article I will not be able to evaluate whether these arguments justify the Court’s approach, all things considered. Further, it would be impossible to even mention all the possible arguments here. I have chosen those that I consider to be of most justificatory weight, if sound.

Before turning to these arguments, it is worth clarifying that some of them appeal to the importance of the law being coherent, whereas others appeal to the importance of judicial decisions being (as far as possible) ‘coherent’ with the existing law. This distinction is not often acknowledged. In practice, however, both kinds of argument can potentially require the same thing. An argument that appeals to the importance of the law being coherent might, if premised on the view that coherence is more important than all other values, claim that coherence should be achieved by deciding cases coherently with the law that we already have. But an aspiration towards coherence in the law does not necessarily require this. If coherence is considered to be one of several important values, sometimes the best decision in a particular case might be one that is incoherent with the existing law, even if complete coherence is still regarded as desirable in general.

A Democracy

The only argument that can be found in the decisions of the High Court is in the judgment of French CJ in Apotex. He suggested that ‘[t]he endeavour to achieve coherence ... falls more readily within the institutional competence of the courts than an endeavour to strike some balance between competing public and private interests’, the latter being a matter largely ‘for the legislature’.[155] A similar argument appears throughout the coherence literature. It has been suggested that judicial decisions must be coherent with the reasons underlying the existing law so as to ensure that judges are separated from politics,[156] or in other words, do not engage in ‘unacceptable versions of judicial law-making’.[157] French CJ’s argument does not go this far. He is not suggesting that coherence-based reasoning is necessary to avoid a political or legislative kind of decision-making, only that it does avoid it. Notwithstanding this difference, the underlying point of all these arguments is that policy or lawmaking functions ought to be left to the democratically legitimate and accountable legislature, and that coherence-based reasoning ensures that they are.[158] Further, all appear to suggest that this is true regardless of whether the ‘existing law’ (with which a new decision is coherent) is sourced in judicial decisions or legislation.

This argument seems to have particular force when the question is whether the common law should be coherent with statute law specifically. Indeed, the High Court, in cases like Miller and Equuscorp, seems to have regarded it as necessary and not merely desirable on some occasions that the common law is coherent with statute law. Legislation is, of course, the product of decision-making by a democratically accountable and representative legislature. When the common law operates ‘incoherently’ with legislation, it prefers its own (or rather, the judiciary’s own) judgment about the values and purposes that the law ought to reflect and on which society ought to be based. It is plausible, at least prima facie, to suggest that this involves an assumption of the legislature’s proper function by the judiciary.

B Arbitrariness

Reflecting the close connection between ‘coherence’ and ‘reasons’, many have also suggested that incoherence of this kind simply is one form of arbitrariness. In a sense, all the arguments that I will mention here link coherence and justification, in that each identifies a different reason why coherence in the law is important. But the present argument posits a more conceptual connection between the two notions: inconsistent judgments, the argument goes, cannot both be true, and therefore cannot both be justified. To that extent, inconsistency is arbitrary. Dworkin, for example, suggests that when a person fails to act with personal integrity (that is, ‘according to convictions that inform and shape their life as a whole’) they act ‘capriciously or whimsically’.[159] He also describes ‘checkerboard’ legislation as treating ‘similar accidents or occasions ... differently on arbitrary grounds’.[160] More recently in Justice for Hedgehogs, he also suggests that inconsistency across ‘departments’ of personal conviction involves ‘arbitrary difference rather than principled distinctions’.[161] This point is also apparent throughout Birks’s discussion of ‘stultification’ (even though he does not refer explicitly to ‘arbitrariness’),

and is reflected in the writing of other scholars.[162] In requiring the law’s underlying reasons to be consistent, the High Court’s approach avoids this form of arbitrariness.

It is worth noting, however, that others have regarded the conceptual connection between coherence and justification as entailing a different conception of coherence. For example, Weinrib has argued that coherence in private law requires more than mere consistency. Rather, it requires the justificatory considerations underlying private law to be ‘unified’, or in other words, to ‘interlock into a single integrated justification’.[163] This claim was based on the following conception of justification:

A justification justifies: it has normative authority with respect to the material to which it applies. The point of adducing a justification is to allow that authority to govern whatever falls within its scope. Thus if a justification is to function as a justification, it must be permitted, as it were, to expand into the space that it naturally fills. Consequently, a justification sets its own limit. For an extrinsic factor to cut the justification short is normatively arbitrary.[164]

Weinrib thought that this conception of justification entailed his conception of coherence. He claimed that if private law attempts to give effect to more than one justificatory consideration, thereby giving rise to ‘incoherence’, each consideration will inevitably cut the other(s) short. As an example, he pointed to the understanding of the law of torts on which liability is justified by considerations of compensation and deterrence. On this understanding, the defendant is held liable when they engaged in conduct that ought to be deterred. The plaintiff has the benefit of this liability because, as a result of the defendant’s actions, they are in need of compensation. However, both considerations are ‘cut short’ on this understanding. Plaintiffs that are injured without anyone being at fault will not be compensated, despite being no less deserving of it. Defendants that happen to avoid injuring anyone will have no deterrent enforced against them. This is despite the justificatory strength of neither consideration depending on the application of the other. From this, Weinrib concludes that, if arbitrariness is to be avoided, it is necessary to identify a single consideration that is capable of justifying the connection of the particular plaintiff and particular defendant.[165]

It is not possible in the scope of this article to assess the soundness of this argument. For now, it is sufficient to say that many authors have put forward a conception of coherence as consistency on the basis that it avoids one form of arbitrariness. If this argument is sound, then it supports the High Court’s approach. But it is important to keep Weinrib’s argument in mind. At least at first glance, it appears to require a different conception of coherence. If Weinrib is right, this would imply that the High Court’s approach is incorrect.

C Inner Morality of Law

Several arguments in the academic literature appeal to the values that Fuller associated with the ‘inner morality of law’,[166] and which others have associated with more ‘formal’ conceptions of the rule of law.[167] Some of these arguments appeal to the importance of the law being coherent, whereas others appeal to the importance of deciding cases coherently with the existing law. Consider Amaya’s argument that the law is more ‘certain’ when it is coherent, which is an argument in the former category. She suggests that coherence promotes legal certainty by, among other things, ‘facilitating knowledge of the law, for a coherent body of norms is more easily remembered and understood’.[168] She also suggests that a coherent body of norms ‘is also easier to be applied and followed’.[169] It is noteworthy that these arguments are based on empirical hypotheses for which Amaya herself provides no support.[170]

In contrast, most other arguments that posit a link between coherence and the inner morality of law appeal to the importance of coherence-based reasoning itself. For example, MacCormick suggests that decisions not derivable from the reasons underlying the existing law, and thus not ‘coherent’ with it, ‘impos[e] new liabilities in a purely retrospective way’.[171] Here MacCormick is appealing to one of the well-known aspects of the inner morality of law: that legal standards should not be applied retrospectively to circumstances that occurred before those standards were available to those subject to them. Birks has made a different point. He argued that if judges are permitted to contradict the existing law ‘without attending to the need to explain’, they are simply exercising ‘pure power behind a semantic smoke-screen’, and ‘[p]ower goes hand in hand with uncertainty’.[172] Finally, Amaya suggests that ‘social stability’ is secured by the reliance on coherence-based reasoning.[173] All of these arguments can be possibly understood as suggesting that coherence-based reasoning is consistent with the inner morality of law.

D Integrity and Community

So far I have neglected to mention what might, given the extensive references to the notion of ‘integrity’, seem the most obvious argument in favour of the High Court’s conception of coherence. Dworkin himself argued at length that state coercion is only morally justified when the law on which it is based exhibits the requirements of integrity, or ‘consistency in principle’.[174] According to Dworkin, it is ‘close’ to a necessary condition of state coercion being (generally) morally justified that citizens have a (general) moral obligation to obey the law.[175] He suggests that arguments commonly made in support of such a moral obligation fail, given that they do not explain how someone can be ‘morally affected by being given what he does not ask for or choose to have’.[176] Dworkin’s own explanation relied on the concept of ‘associative or communal obligations’, the special responsibilities social practice attaches to membership in some group, ‘like the responsibilities of family or friends or neighbors’.[177] Importantly, these obligations do not necessarily arise by ‘choice or consent’,[178] although ‘the members of a group must by and large hold certain attitudes about the responsibilities they owe one another if these responsibilities are to count as genuine fraternal obligations’.[179] He identifies four such attitudes:

First, they must regard the group’s obligations as special, holding distinctly within the group, rather than as general duties its members owe equally to persons outside it. Second, they must accept that these responsibilities are personal: that they run directly from each member to each other member, not just to the group as a whole in some collective sense. ... Third, members must see these responsibilities as flowing from a more general responsibility each has of concern for the well-being of others in the group ... Fourth, members must suppose that the group’s practices show not only concern but an equal concern for all members.[180]

Given the abovementioned lack of choice or consent by the people, Dworkin’s argument is that ‘political obligation’, and in particular, the ‘obligation to obey the law’, must be understood as a special kind of ‘associative obligation’, owed by each member of the political community to all other members.[181] He suggests that these four conditions are only satisfied in a community committed to political integrity. In ensuring that the same principles are extended and applied to all, the law of such a community ‘assumes that each person is as worthy as any other, that each must be treated with equal concern according to some coherent conception of what that means’.[182] So, on this argument,

the High Court’s aspiration towards coherence contributes to the morality of state coercion.

VII Conclusion

The concept of ‘coherence’ is undoubtedly a difficult one. It has several meanings which significantly overlap with one another and differ only in subtle respects. It is not surprising, then, that multiple conceptions of coherence appear in the High Court’s decisions. The one that it relies on most often and aspires towards in a concerted way requires ‘consistency’ in the law, but it has not fully explained the kind of ‘consistency’ to which it is appealing. I suggest that the High Court is best understood as requiring consistency in the law’s underlying normative reasons or justifications. I have argued that this requires consistency in the weight or importance attributed to contributory reasons relative to each other. And although the High Court itself has provided only minor justification for its reliance on coherence, I have mentioned several possible arguments that might be at least capable of supporting it, relating to the value of democracy, the avoidance of arbitrariness and the inner morality of law. Dworkin also argued that ‘integrity’ in the law is a necessary condition of the morality of state coercion. As mentioned at the beginning, this understanding of the High Court’s principle of coherence (as consistency) and its possible normative basis will certainly allow it to be utilised more effectively. But all of this is only the start. More is needed before any firm conclusion can be reached about whether and why the High Court should continue to rely on ‘coherence’ in its decision-making.


* LLB (UQ); PhD Candidate, TC Beirne School of Law, The University of Queensland. I

thank Kit Barker, Ross Grantham and the anonymous referees for their helpful comments. This research is supported by an Australian Government Research Training Program

(RTP) Scholarship.

[1] It has also been ‘a topic of sustained and deep thought’ in academic literature for decades: Ross Grantham and Darryn Jensen, ‘Coherence in the Age of Statutes’ [2016] MonashULawRw 12; (2016) 42 Monash University Law Review 360, 362–3. See also Neil MacCormick, Legal Reasoning and Legal Theory (Clarendon Press, 1978); Ronald Dworkin, Law’s Empire (Fontana Press, 1986).

[2] See, eg, Williams v Commonwealth [2012] HCA 23; (2012) 248 CLR 156, 238 [157]; Plaintiff S4/2014 v Minister for Immigration and Border Protection (2014) 253 CLR 219, 236 [42].

[3] For some of the most recent examples, see Tabet v Gett [2010] HCA 12; (2010) 240 CLR 537, 551 [18];

Miller v Miller [2011] HCA 9; (2011) 242 CLR 446, 454–5 [15]–[16], 479–80 [93]–[94], 481–2 [101]–[102]; Equuscorp Pty Ltd v Haxton (2012) 246 CLR 498, 518 [34], 522–3 [45]; Barclay v Penberthy (2012) 246 CLR 258, 317 [164]; Apotex Pty Ltd v Sanofi-Aventis Australia Pty Ltd [2013] HCA 50; (2013) 253 CLR 284, 316 [44], 319 [50]; Hunter and New England Local Health District v McKenna [2014] HCA 44; (2014) 253 CLR 270, 279 [19]; Brookfield Multiplex Ltd v Owners Corporation Strata Plan 61288 [2014] HCA 36; (2014) 254 CLR 185, 201–2 [25] (French CJ), 214 [69] (Crennan, Bell and Keane JJ); Australian Financial Services and Leasing Pty Ltd v Hills Industries Ltd [2014] HCA 14; (2014) 253 CLR 560, 625 [156]; Gnych v Polish Club Ltd [2015] HCA 23; (2015) 255 CLR 414, 434–5 [72]–[73]; D’Arcy v Myriad Genetics Inc [2015] HCA 35; (2015) 258 CLR 334, 351 [28], 352 [30]; Crown Melbourne Ltd v Cosmopolitan Hotel (Vic) Pty Ltd [2016] HCA 26; (2016) 333 ALR 384, 416–18 [143]–[149]; Badenach v Calvert [2016] HCA 18; (2016) 257 CLR 440, 451 [23] (French CJ, Kiefel and Keane JJ), 458 [59] (Gageler J); Prince Alfred College Inc v ADC [2016] HCA 37; (2016) 258 CLR 134, 169 [116].

[4] See, eg, Sullivan v Moody (2001) 207 CLR 562, 580–1 [53]–[55].

[5] Miller (n 3) 454 [15]; Equuscorp (n 3) 513 [23], 518 [34].

[6] Chief Justice Robert French, ‘Trust and Statutes’ [2015] MelbULawRw 31; (2015) 39 Melbourne University Law Review 629, 631.

[7] Elise Bant, ‘Statute and Common Law: Interaction and Influence in Light of the Principle of Coherence’ [2015] UNSWLawJl 13; (2015) 38 University of New South Wales Law Journal 367, 367.

[8] See, eg, Grantham and Jensen (n 1) 362–7.

[9] Miller (n 3) 473 [74], 481–2 [101]–[102].

[10] Ibid 467 [56], 473 [74].

[11] Equuscorp (n 3) 520 [38].

[12] Ibid 519–20 [37], quoting Peter Birks, ‘Recovering Value Transferred under an Illegal Contract’ (2000) 1 Theoretical Inquiries in Law 155, 203.

[13] Grantham and Jensen (n 1) 362.

[14] Kit Barker, ‘Private Law as a Complex System: Agendas for the Twenty-First Century’ in Kit Barker, Karen Fairweather and Ross Grantham (eds), Private Law in the 21st Century (Hart Publishing, 2017) 3, 3.

[15] Ibid 17.

[16] See, eg, Sullivan (n 4) 579–80 [50], 580–1 [53]–[55]; Lumbers v W Cook Builders Pty Ltd

(in liq) [2008] HCA 27; (2008) 232 CLR 635, 662–3 [78].

[17] Barker (n 14) 19.

[18] For an account of the relevance of coherence in other disciplines, see Amalia Amaya, The Tapestry of Reason: An Inquiry into the Nature of Coherence and Its Role in Legal Argument (Hart Publishing, 2015) chs 3–9.

[19] Ibid 472, 477.

[20] Joseph Raz, ‘The Relevance of Coherence’ (1992) 72 Boston University Law Review 273, 276.

[21] See, eg, Ronald Dworkin, Justice for Hedgehogs (Belknap Press, 2011) 43: ‘People often say that some proposition makes no sense when they mean only that it is silly or obviously wrong.’

[22] Raz (n 20) 286.

[23] Prince Alfred College (n 3).

[24] Ibid 148 [39] (emphasis added).

[25] Ibid 149 [44].

[26] Ibid 149–50 [45].

[27] Ibid 156 [68].

[28] Ibid 148 [39], quoting Hollis v Vabu Pty Ltd [2001] HCA 44; (2001) 207 CLR 21, 37 [35].

[29] See, eg, D’Orta-Ekenaike v Victoria Legal Aid [2005] HCA 12; (2005) 223 CLR 1, 102 [325]; Neilson v Overseas Project Corporation of Victoria Ltd [2005] HCA 54; (2005) 223 CLR 331, 369 [111]; A-G (SA) v City of Adelaide Corporation [2013] HCA 3; (2013) 249 CLR 1, 28 [36].

[30] French (n 6) 631.

[31] (1988) 164 CLR 539, 584.

[32] (1997) 188 CLR 159, 223–34.

[33] Sullivan (n 4) 579–80 [50], 580–1 [53]–[55].

[34] Ibid 580 [50].

[35] Ibid 581 [54].

[36] Ibid. Although the Court did not explicitly refer to the problem arising from this intersection as ‘incoherence’, the Court subsequently implied in Koehler v Cerebos (Australia) Ltd [2005] HCA 15; (2005) 222 CLR 44, 56 [31] that this issue, as well as the second issue to be discussed below, is one which involved questions of legal coherence. The way in which the Court introduced the issue of ‘coherence’ in Sullivan (n 4) 579–80 [50] can also be understood as implying that the defamation issue involves a question of coherence.

[37] Sullivan (n 4) 581 [55]; see also at 582 [62].

[38] Ibid 582 [60].

[39] Ibid 582 [62].

[40] Ibid.

[41] Ibid.

[42] Ibid 580 [53].

[43] Miller (n 3).

[44] Ibid 454 [15].

[45] Ibid.

[46] Ibid 454–5 [16].

[47] Yango Pastoral Co Pty Ltd v First Chicago Australia Ltd [1978] HCA 42; (1978) 139 CLR 410, 432.

[48] Ibid 434.

[49] Ibid 428–30.

[50] See Nelson v Nelson [1995] HCA 25; (1995) 184 CLR 538; Fitzgerald v FJ Leonhardt Pty Ltd [1997] HCA 17; (1997) 189 CLR 215, 227.

[51] Miller (n 3) 473 [74].

[52] ‘Coherence’ has also been identified as the basis of the illegality doctrine in other common law jurisdictions: see, eg, Hall v Hebert [1993] 2 SCR 159; Patel v Mirza [2016] UKSC 42; [2017] AC 467. It is true that, in Patel, Lord Toulson SCJ referred favourably to Burrows’s ‘range of factors’ approach to illegality, in which the coherence of the law is only one factor among many in determining the ‘appropriate response’ to illegality (at 498 [93]–[94], quoting Andrew Burrows, A Restatement of the English Law of Contract (Oxford University Press, 2016) 229–30) and appeared to accept the relevance of a similar range of factors: at 499–501 [101]–[109]. There is, however, an important difference between the approaches of Lord Toulson SCJ and Burrows. On Lord Toulson SCJ’s approach, coherence or consistency appears to be the underlying basis of the defence, and the various factors are treated as relevant to the extent that they bear upon what coherence requires in the particular case: at 499–500 [99]–[101]. This is different to Burrows’s approach, which treats the factors as relevant to the extent that they bear upon what is the ‘appropriate response’ to the illegality.

[53] Miller (n 3) 473 [74].

[54] Ibid 478 [90].

[55] Ibid 480 [94].

[56] Ibid; see also at 479–80 [92]–[93].

[57] Ibid 482 [101].

[58] Ibid 481–2 [99]–[101].

[59] Ibid 482–3 [103]–[106].

[60] Equuscorp (n 3). It was also relied on in Gnych (n 3), although only Gageler J used the language of ‘coherence’: at 434–5 [72]–[73].

[61] Equuscorp (n 3) 513–14 [24].

[62] Ibid 518 [34].

[63] Ibid 514 [25].

[64] Ibid 519–20 [37], quoting Birks, ‘Recovering Value’ (n 12) 169, 203.

[65] Equuscorp (n 3) 520 [38].

[66] Ibid 522–3 [45].

[67] Ibid 543–4 [111].

[68] Apotex (n 3).

[69] Ibid 301 [17].

[70] Ibid 316 [44].

[71] Ibid; see also at 317–18 [46].

[72] Ibid 317 [46]; see also at 304–16 [23]–[43].

[73] Ibid 316 [44]. However, contrary to French CJ’s suggestion, the problem is not one of logic: see Lon L Fuller, The Morality of Law (Yale University Press, rev ed, 1969) 65–6.

[74] Apotex (n 3) 319 [50].

[75] D’Arcy (n 3) 352 [30] n 85.

[76] Grantham and Jensen (n 1) 363, 364.

[77] Peter Birks, Unjust Enrichment (Oxford University Press, 2nd ed, 2005) 224.

[78] Ibid.

[79] Ibid 242. This exception provides that ownership of money, but not other chattels, ‘is extinguished every time it is received honestly from, and for value given to, a non-owner’.

[80] Ibid.

[81] Ibid 247.

[82] Ibid.

[83] [1950] AC 327.

[84] Birks, Unjust Enrichment (n 77) 248.

[85] Ibid 249 (emphasis added).

[86] Ibid.

[87] Ibid 249–53.

[88] Ibid 250.

[89] Sullivan (n 4) 581 [54].

[90] Cf Spring v Guardian Assurance plc [1996] 2 AC 296.

[91] See also Cherie Booth and Dan Squires, The Negligence Liability of Public Authorities (Oxford University Press, 2006) 209.

[92] Michael Gillooly has also made the point that ‘incoherence’ can take different forms: Michael Gillooly, ‘Legal Coherence in the High Court: String Theory for Lawyers’ (2013) 87 Australian Law Journal 33, 38.

[93] Jessica Hudson, ‘The Price of Coherence in Estoppels’ [2017] SydLawRw 1; (2017) 39 Sydney Law Review 1, 1.

[94] See Jonathan Dancy, Ethics without Principles (Oxford University Press, 2004). In some aspects, Dancy’s account is controversial. My conclusions, however, do not depend on all of it. On the parts that are most important for my argument, others hold similar views to Dancy’s. Joseph Raz, for example, takes issue with some of the points Dancy makes: see, eg, Joseph Raz, ‘The Trouble with Particularism (Dancy’s Version)’ (2006) 115 Mind 99. Nevertheless, like Dancy, he seems to accept that reasons only ‘count in favour’ of particular actions, and that reasons can conflict and outweigh each other in particular cases: Joseph Raz, Practical Reason and Norms (Princeton University Press, 1990) 186–7. See also Derek Parfit, On What Matters (Oxford University Press, 2011) vol 1, 31–3; John Broome, Reasoning through Rationality (Wiley Blackwell, 2013) 51–60; TM Scanlon, Being Realistic about Reasons (Oxford University Press, 2014) ch 5. As I will argue below, Dworkin also appears to share this conception of reasons.

[95] Dancy (n 94) 16. I do not deny the evaluative character of this and the following ‘metaethical’ claims about the role of reasons in normative justification: see generally Dworkin, Justice for Hedgehogs (n 21) chs 2–5.

[96] Dancy (n 94) 16.

[97] Ibid.

[98] See generally ibid 15–17, 29.

[99] Ibid 7, 73.

[100] Ibid 38–9.

[101] Ibid 41–2.

[102] Ibid 73–8.

[103] I acknowledge the useful discussion in Dale Smith, ‘The Many Faces of Political Integrity’ in Scott Hershovitz (ed), Exploring Law’s Empire: The Jurisprudence of Ronald Dworkin (Oxford University Press, 2006) 119, 125–42, which assisted in structuring my thoughts on this issue. Several of the points made here were also emphasised by Smith.

[104] Dworkin, Law’s Empire (n 1) 165.

[105] Ibid.

[106] Ibid 165–6; see also at 219–20.

[107] Ibid 166.

[108] Ibid.

[109] Ibid 184.

[110] Ibid.

[111] Ronald Dworkin, ‘Response’ in Scott Hershovitz (ed), Exploring Law’s Empire: The Jurisprudence of Ronald Dworkin (Oxford University Press, 2006) 291, 296.

[112] Dworkin, Law’s Empire (n 1) 179.

[113] Ibid 183.

[114] Ibid 184.

[115] Ibid 183.

[116] Ibid 268–9.

[117] Ibid 269.

[118] Ronald Dworkin, Taking Rights Seriously (Duckworth, rev ed, 1978) 24–7.

[119] Ibid 26.

[120] Ibid.

[121] Dworkin, Law’s Empire (n 1) 269.

[122] Dworkin, Taking Rights Seriously (n 118) 93–4.

[123] Dworkin, Law’s Empire (n 1) 435–6.

[124] Ibid 436.

[125] Ibid.

[126] Dworkin, Taking Rights Seriously (n 118) 26.

[127] Dworkin, ‘Response’ (n 111) 296.

[128] Ibid.

[129] See Part V(A).

[130] Dworkin, Taking Rights Seriously (n 118) 26.

[131] Ibid 184.

[132] Ibid.

[133] Horrocks v Lowe [1975] AC 135, 149.

[134] See Spring (n 90) 324–5.

[135] Miller (n 3) 479 [93].

[136] This claim seems to be endorsed in Gray v Thames Trains Ltd [2009] 1 AC 1339. In that case, Lord Hoffmann, who gave the leading judgment, held that a causal connection between the claimant’s illegal conduct and their harm precludes recovery for that harm: at 1376–77

[50]–[55]. He observed that there are ‘wider’ and ‘narrower’ versions of this rule: at 1370 [29]. The narrower rule is that ‘you cannot recover for damage which flows from loss of liberty, a fine or other punishment lawfully imposed upon you in consequence of your own unlawful act’. The wider rule is that ‘you cannot recover compensation for loss which you have suffered in consequence of your own criminal act’. Of course, the High Court in Miller (n 3) emphatically rejected a causation-based approach: at 455–7 [17]–[22], 467 [56]. However, some of Lord Hoffmann’s remarks are nevertheless applicable. He explicitly indicated that, in contrast to the narrow rule, the wide rule is not (ie could not be) based on the need to ensure consistency between the civil and criminal law: Gray (n 136) 1376 [51]. The harm in Miller, however, is only covered by the wide rule. Therefore, on Lord Hoffmann’s argument, recovery for that harm would not be inconsistent with the criminal law.

[137] Miller (n 3) 482 [101].

[138] See Ernest J Weinrib, ‘Illegality as a Tort Defence’ (1976) 26 University of Toronto Law Journal 28, 45–6.

[139] The regulatory powers of the Independent Liquor and Gaming Authority in Gnych (n 3) are an example of a consequence that is not a penalty: at 428–30 [52]–[55] (French CJ, Kiefel, Keane and Nettle JJ).

[140] Ibid 435 [73] (Gageler J). See also Yango Pastoral (n 47) 429.

[141] See Weinrib, ‘Illegality as a Tort Defence’ (n 138) 50–4. The basis for such a claim might be that a criminal statute, in providing a specific and limited list of defences, exhaustively identifies the circumstances in which a person ought not to bear the consequences of its breach. Therefore, even if the plaintiff only suffered a criminal penalty because of the defendant’s negligence, this reason for liability would be judged in the statute as outweighed by the statutory purpose.

[142] See Patel (n 52) 514 [159]–[160] (Lord Neuberger P).

[143] See James Edelman and Elise Bant, Unjust Enrichment (Hart Publishing, 2nd ed, 2016) 161.

[144] Grantham and Jensen (n 1) 366.

[145] Ibid, quoting Equuscorp (n 3) 549 [124], 551–2 [133].

[146] Equuscorp (n 3) 548 [122].

[147] Ibid 549 [124].

[148] Ibid 550 [127]–[128].

[149] Ibid 550 [129].

[150] Ibid 550 [130]. Heydon J also relied on a provision that preserved liabilities owed to the investors. It provided that such liabilities are not negated by reason of a contravention of the prospectus provisions. This would preserve the liability of the lender under the loan contract: at 546 [119]. He regarded this as implying that any ‘all other liabilities and all corresponding rights under the loan contracts’, including the lender’s right to repayment, are extinguished: at 548 [123]. The magnitude of this consequence was thought to partly ‘indicate’ that the statutory consequences were regarded as sufficient, and that claims for money had and received were therefore not unenforceable: at 550 [130]. However, arguably, the provision to which Heydon J referred does not extinguish the lender’s contractual right to repayment. The legislature might be understood as generally deferring to the common law, only intervening in matters to which it applies if this is necessary to achieve the purpose of the statute. In the present case, that purpose was the protection of investors. A provision that preserves their rights is necessary for this purpose, but the same is not true of the rights of those such as the lender. The legislature, therefore, might simply have decided to leave the lender to the common law result, whatever its content might be, for reasons of deference. It should not be understood as positively extinguishing their contractual rights.

[151] Ibid 551 [131].

[152] See, eg, Edelman and Bant (n 143) 126–7; Kit Barker and Ross Grantham, Unjust Enrichment (LexisNexis Butterworths, 2nd ed, 2018) 222.

[153] It is even arguable that the purpose of ‘protecting investors’ did not favour the rejection of the claims. The change of position defence would ensure that the investors were not placed in a worse position than they were in before entering the loan agreements.

[154] This conclusion seems consistent with the decision in Patel (n 52), in which it was suggested that restitution for failure of basis ‘will not prima facie be debarred’ in cases of illegality:

at 504 [116] (Lord Toulson SCJ). Only in ‘rare’ cases, in which there is a particular reason why the claim would lead to inconsistency, will it be denied.

[155] Apotex (n 3) 316 [44], 317 [46].

[156] See Ernest J Weinrib, The Idea of Private Law (Oxford University Press, rev ed, 2012) 45.

[157] See Neil MacCormick, Rhetoric and the Rule of Law: A Theory of Legal Reasoning (Oxford University Press, 2005) 203. See also MacCormick, Legal Reasoning and Legal Theory

(n 1) 187–8.

[158] See also Peter Birks, ‘Equity in the Modern Law: An Exercise in Taxonomy’ [1996] UWALawRw 1; (1996) 26 University of Western Australia Law Review 1, 97–9.

[159] Dworkin, Law’s Empire (n 1) 166.

[160] Ibid 179.

[161] Dworkin, Justice for Hedgehogs (n 21) 108.

[162] See, eg, Allan Beever, Rediscovering the Law of Negligence (Hart Publishing, 2007) 22–3.

[163] Weinrib, The Idea of Private Law (n 156) 33.

[164] Ibid 39.

[165] Ibid 39–42.

[166] Fuller (n 73) ch 2.

[167] See, eg, Jeremy Waldron, ‘The Concept and the Rule of Law’ (2008) 43 Georgia Law Review

1, 6–7.

[168] Amaya (n 18) 541.

[169] Ibid.

[170] Hedley has made this point, although in a slightly different context: Steve Hedley, ‘Rival Taxonomies within Obligations: Is There a Problem?’ in Simone Degeling and James Edelman (eds), Equity in Commercial Law (Lawbook, 2005) 77, 84–8.

[171] MacCormick, Rhetoric and the Rule of Law (n 157) 203.

[172] Birks, ‘Equity in the Modern Law’ (n 158) 52, 97.

[173] Amaya (n 18) 541.

[174] See generally Dworkin, Law’s Empire (n 1) 186–216.

[175] Ibid 191.

[176] Ibid 195.

[177] Ibid 196.

[178] Ibid.

[179] Ibid 199.

[180] Ibid 199–200 (emphasis in original).

[181] Ibid 206.

[182] Ibid 213.


AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback
URL: http://www.austlii.edu.au/au/journals/MelbULawRw/2018/6.html