Introduction
In his 1964 book The Morality of Law, the American legal philosopher Lon Fuller set out to answer a question that had occupied jurisprudence since antiquity: what separates a genuine legal system from a mere apparatus of coercion dressed in legal language? Fuller’s answer was not framed in terms of the content of laws – whether they were just, fair, or morally good in a substantive sense – but in terms of their form and the manner of their administration. He proposed what has become known as the “internal morality of law”: eight procedural conditions that any system of rules must satisfy if it is to be recognised as law at all, rather than the arbitrary exercise of power by those who happen to hold it.
Fuller’s eight desiderata are these: laws must exist and bind everyone, including those who govern; laws must be published; laws must operate prospectively rather than retroactively; laws must be reasonably clear; laws must not contradict one another; laws must not demand the impossible; laws must possess sufficient constancy to allow people to plan their affairs, while still permitting timely revision; and, finally, the actual conduct of officials must align with the laws as declared. Taken together, these conditions form less a checklist than a diagnostic instrument – a way of testing whether a society that proclaims its commitment to the rule of law is, in practice, living up to that commitment.
Applying Fuller’s framework to contemporary Australia is an uncomfortable but necessary exercise. Australia likes to think of itself as a mature liberal democracy, a nation governed by laws rather than the arbitrary will of office-holders, and on the broad comparative measures used by bodies such as the World Justice Project, that self-image is not unfounded: Australia was ranked 11th of 142 countries in the 2024 Rule of Law Index, ahead of most peer nations on paper.[1] Yet aggregate rankings of this kind measure formal institutional strength rather than the uneven, granular experience of law described by Fuller’s eight conditions, and a country can score well in the former while still falling short, in specific and recurring ways, on the latter. When the eight principles are applied with any rigour to the actual conduct of Australian governments, police forces, and administrative bodies, the picture that emerges is considerably less flattering than either the national self-image or the headline ranking suggests. This essay examines each of Fuller’s principles in turn, asking not whether Australia has laws – it plainly does, and by international comparison does reasonably well – but whether those laws, and the institutions that administer them, satisfy the deeper morality that Fuller believed was constitutive of legality itself.
Principle One: Laws Must Exist and Bind Everyone, Including Officials
The first and most fundamental of Fuller’s principles is also the one most often taken for granted: that there must be rules, and that those rules must apply equally to the governed and the governing. A legal system in which officials are effectively exempt from the laws they enforce is not a legal system in the Fullerian sense; it is a system of privilege masquerading as law.
Australia’s difficulty here is not the absence of laws binding officials – there are, on paper, robust frameworks of administrative law, judicial review, and statutory accountability mechanisms. The difficulty lies in the gap between formal applicability and practical enforcement. Royal commissions, coronial inquests, and parliamentary inquiries into police conduct, the treatment of people in custody, and the administration of welfare and immigration law have repeatedly found that the rules which bind ordinary citizens are applied with markedly less rigour to the agencies of the state itself. When misconduct by police, prison officers, or government departments is identified, the consequences are frequently internal, administrative, and opaque, rather than the kind of public, consequential accountability that the law imposes on private citizens for comparable conduct.
This asymmetry matters because it strikes at the heart of what Fuller meant by the rule of law binding officials. A speed limit that police themselves routinely ignore without consequence is not, in any meaningful sense, the same law that applies to other drivers. The same logic extends to far graver matters: the use of force in custody, the conduct of investigations, and the handling of complaints against government bodies. To be clear, the issue is not that Australian officials are formally above the law – no statute exempts a police officer or public servant from the criminal code – but that the practical mechanisms for enforcing the law against officials are structurally weaker, slower, and more self-referential than those available against ordinary citizens, producing a de facto rather than de jure inequality before the law. Where these mechanisms are governed by processes that shield officials from the ordinary consequences of wrongdoing, the first Fullerian condition begins to fray, even where the statute books remain formally intact.
Principle Two: Laws Must Be Published
Fuller’s second condition is comparatively straightforward to satisfy in a modern administrative state: Australian statutes, regulations, and case law are publicly accessible, and the architecture of legislative publication is, by global standards, reasonably sophisticated. Acts of parliament are gazetted, consolidated, and made available through government databases; court judgments are published, with some exceptions for matters involving children or national security.
Yet publication in the narrow sense of textual availability is not the same as the kind of practical accessibility that gives the principle its moral force. A law that exists in a database accessible only to those with the resources to engage specialist legal counsel is published in form but not in substance. The proliferation of delegated legislation – regulations, ministerial directions, and departmental guidelines that carry the force of law but are drafted and amended with far less scrutiny than primary legislation – creates a layer of governance that is technically public but practically obscure to the citizens it affects. Migration law, social security law, and the administrative guidelines governing police discretion are particularly dense with this kind of quasi-legislative material, much of it unfamiliar even to experienced practitioners. The result is a system in which the formal requirement of publication is met while the underlying purpose of the principle – that citizens should be able to know, in advance, what the law requires of them – is only partially achieved.
Principle Three: Laws Must Be Prospective, Not Retroactive
The prohibition on retroactive lawmaking is among the oldest and most widely accepted of legal principles, reflected in Australia’s constitutional architecture through section 51(xxxvi) and the general common law presumption against retrospectivity, as well as the more specific constitutional protection against retroactive criminal punishment found in cases interpreting Chapter III of the Constitution. In the ordinary criminal context, Australian courts have been consistently vigilant in policing this boundary.
The more troubling pattern emerges not in the criminal law proper but at its margins – in matters of citizenship, immigration status, and social security entitlement, where parliament has on several occasions enacted legislation with retrospective effect, validating past administrative conduct that would otherwise have been unlawful at the time it occurred. The most prominent recent illustration is the Robodebt scheme, in which an automated income-averaging system issued unlawful debt notices to several hundred thousand welfare recipients between 2015 and 2019, a practice the 2023 Royal Commission found had been pursued despite internal legal advice warning of its unlawfulness.[2] The retrospectivity problem here is not statutory but practical, and in some respects more troubling for being so: rather than the executive seeking parliamentary authority before acting, debts were raised and enforced against citizens under a method later confirmed never to have had lawful authority in the first place, with the legal correction arriving only after years of harm had already been inflicted. Citizens were, in effect, held liable under a version of the law that did not yet – and as it turned out, never did – exist.
Citizenship-stripping legislation directed at dual nationals raises the more conventional form of the same concern, with provisions in some instances purporting to apply to conduct that predated the legislative scheme itself.
These are not isolated technical anomalies. They represent a recurring temptation for the executive and legislative arms of government: when administrative conduct is found to have exceeded its legal authority, the path of least resistance is not to accept liability under the law as it stood, but to rewrite the law after the fact so that the conduct becomes retroactively lawful. Each time this occurs, the prospective character of law – its function as a guide for future conduct rather than a post-hoc justification for past power – is weakened.
Principle Four: Laws Must Be Reasonably Clear
Clarity is not merely a stylistic virtue; for Fuller, it is a precondition of fairness, because a law that cannot be understood cannot be obeyed, and a law that cannot be obeyed becomes a tool of arbitrary enforcement rather than a guide to conduct. Here Australian law presents a mixed record. Core criminal offences are, for the most part, drafted with reasonable precision. But significant areas of public order, protest, and emergency-powers legislation have moved in the opposite direction.
The proliferation of state-based “move on” powers, anti-protest legislation, and emergency public-health and public-order statutes enacted across multiple jurisdictions in recent years has drawn sustained criticism – including from the judiciary itself – for vagueness that grants police and other officials extraordinarily broad discretion. When a statute criminalises conduct that “disrupts” or “obstructs” public order without precise definition, the practical effect is to transfer the genuine content of the law from the legislature, where it is at least nominally subject to democratic deliberation, to the individual officer exercising discretion on the street. The High Court’s 2017 decision in Brown v Tasmania is instructive here: a majority struck down key provisions of Tasmania’s anti-protest legislation partly because the boundaries of the areas it regulated were so imprecisely defined that neither police nor protesters could reliably determine where the law applied, with the result that lawful protest was being deterred even where no valid direction to leave had actually been given.[3] Where legislation survives only because it has not yet been tested, or because litigation is prohibitively expensive for those most affected by it, the Fullerian requirement of clarity is satisfied in theory but compromised in the lived experience of citizens subject to its discretionary application.
Principle Five: Law Must Avoid Contradiction
Internal coherence is one of the less glamorous but more practically important of Fuller’s conditions. A federal system such as Australia’s, with concurrent state and Commonwealth jurisdiction across many areas of law, is structurally prone to a particular kind of contradiction. This is not the logical impossibility of obeying two rules at once. It is the practical incoherence of being subject to inconsistent legal regimes depending on jurisdiction, agency, or even the individual officer exercising discretion. Fuller himself treated contradiction as a failure that could exist within a single statute, between statutes, or – most insidiously – between a rule and the unwritten practice that has grown up around it. It is this last, quieter form of contradiction that does the most damage in the Australian context, because it is the hardest to detect, litigate, or correct through ordinary legislative amendment.
This manifests in the differential application of bail laws, custodial protocols, and police conduct standards between states. It manifests more troublingly within a single jurisdiction too, where formal policy and operational practice frequently diverge. Custody guidelines that prescribe particular welfare checks or notification requirements exist in every Australian jurisdiction; coronial findings into deaths in custody have, with grim regularity, identified failures to follow those very protocols. A further layer of contradiction arises from the interaction between Commonwealth and state law in areas such as bail, sentencing, and police powers, where a person’s substantive legal protections can depend less on the nature of their conduct than on which side of a state border they happen to be standing. The contradiction here is not between one statute and another, but between the law as written and the law as practised. Fuller’s framework treats this gap as a serious failure of legality: a rule that is systematically not followed by those charged with following it ceases, in any functional sense, to be a rule at all.[4]
Principle Six: Law Must Not Command The Impossible
Fuller’s sixth condition prohibits laws that demand of citizens what they cannot reasonably perform. Australia’s social security and migration systems provide the clearest contemporary illustrations of this principle under strain. The Robodebt scheme again serves as the paradigm case: an automated income-averaging system that issued debt notices to welfare recipients based on a calculation method that bore no necessary relationship to their actual earnings, effectively demanding that recipients disprove debts calculated by a method the government’s own lawyers had been advised was unlawful.[5] Many recipients, lacking the income records being assumed of them or the resources to contest the debts, were placed in a position where compliance with the law as administratively applied was not merely difficult but, for practical purposes, impossible.
A subtler version of the same problem recurs in bureaucratic systems that impose documentation or procedural requirements on people without the literacy, language access, or institutional knowledge to meet them – requirements that function, in practice, as mechanisms of exclusion rather than genuine preconditions for entitlement. When the practical demands of a legal or administrative process exceed the capacity of the people subject to it, the formal existence of a right or process becomes close to meaningless, and the principle that law must not command the impossible is honoured only in the letter, not the spirit.
Principle Seven: Law Must Have Constancy, Yet Allow Revision
Fuller’s seventh principle captures a genuine tension at the heart of legal systems: law must be stable enough to allow people to plan their lives around it, yet flexible enough to be revised when social and political circumstances change. Australia’s record here is genuinely mixed, and in some respects creditable – the capacity of the system to produce reform in response to royal commissions and public inquiries, however slow, demonstrates a functioning revision mechanism.
The difficulty lies less in the absence of revision than in its unevenness. Areas of law affecting the politically powerful tend to be revised relatively promptly in response to identified problems; areas of law affecting groups with less political capital tend to languish for decades despite repeated, well-documented findings of failure. The recommendations of the 1991 Royal Commission into Aboriginal Deaths in Custody remain, more than three decades later, only partially implemented, even as the underlying problems they identified have persisted and, in custody rates, worsened.[6] This is not a failure of the seventh principle in the abstract – the legal system does revise itself – but a failure of equal application: constancy is too often preserved precisely where it should be revised, and revision is too often withheld from those who need it most.
Principle Eight: Official Action Must Be Consistent With Declared Rules
The eighth and final principle is, in many respects, the culmination of the previous seven, because it asks the most practical question of all: regardless of what the law says, does official conduct actually conform to it? This is the principle most directly engaged by the recurring pattern of misconduct findings against police forces, government departments, and other state agencies across Australian jurisdictions – findings that establish, often after years of litigation or inquiry, that the conduct of officials departed substantially from the standards the law itself prescribed.
The pattern is a familiar one: an individual or group raises a complaint about official conduct; the relevant agency investigates itself, or is investigated by a body with limited independence and weaker powers than the agency under scrutiny; the process takes years; and even where wrongdoing is eventually established, the consequences are administrative rather than the kind of accountability the law would impose on a private citizen for equivalent conduct. Acquittals, exonerations, and successful judicial review applications against government decisions are, in this sense, a vital part of demonstrating that the legal system retains some capacity for the eighth principle to operate – but the years of delay, financial cost, and personal toll required to reach that point illustrate how far short of Fuller’s standard the underlying administrative culture frequently falls. A right that exists on paper but requires extraordinary personal sacrifice to vindicate is not, in the fullest sense Fuller intended, functioning as law.
Conclusion
Fuller’s eight principles were never intended as an abstract philosophical exercise divorced from practical consequence. They were meant as a working test – a way of asking, of any system that calls itself a legal order, whether it actually behaves like one. Applied to Australia, the test yields a sobering verdict: not a wholesale failure of the rule of law, but a pattern of selective and uneven compliance, in which the formal architecture of legality is substantially intact while its practical application varies enormously depending on who is asking, and who is being asked to comply.
This is, in some ways, a more troubling diagnosis than outright lawlessness would be. A society with no rule of law at all is at least honest about the nature of power within it. A society that maintains the institutions, language, and self-image of the rule of law, while permitting systematic and well-documented departures from its own standards in the treatment of those least able to demand accountability, runs a different risk: that the rule of law becomes, for a significant portion of its population, less a lived reality than a promise repeatedly deferred. Fuller’s framework does not tell us how to close that gap. But it gives us, at minimum, a precise vocabulary for naming it – and naming it precisely is the necessary first step toward holding any government to the standard it claims, on paper, to have already met.
[1] World Justice Project, Rule of Law Index 2024 (Report, World Justice Project, October 2024) 12, 28, ranking Australia 11th of 142 countries and jurisdictions overall, and 11th of 47 high-income countries and jurisdictions, on the basis of household and expert surveys assessing eight factors: constraints on government powers, absence of corruption, open government, fundamental rights, order and security, regulatory enforcement, civil justice, and criminal justice.
[2] Royal Commission into the Robodebt Scheme, Report (Final Report, 7 July 2023, corrected 11 July 2023) vol 1, 9-11, vol 2, ch 5. The Commission, chaired by the Honourable Catherine Holmes AC SC, found that the scheme was “devised without regard to the social security law”, that the use of income averaging “was essentially unfair, treating many people as though they had received income at a time when they had not”, and that senior officials, including the then Secretary of the Department of Human Services, had been made aware of substantial legal doubts about the scheme’s validity well before it was discontinued in 2019. See also Prygodicz v Commonwealth [2021] FCA 634 (the Federal Court class action settlement approval).
[3] Brown v Tasmania (2017) 261 CLR 328; [2017] HCA 43 (Kiefel CJ, Bell, Keane, Gageler and Nettle JJ; Gordon J partially dissenting; Edelman J dissenting), holding invalid ss 6(1)-(4), 8(1), 11(1)-(2) and (6)-(8), 13, and pt 4 of the Workplaces (Protection from Protesters) Act 2014 (Tas) on the ground that they impermissibly burdened the implied freedom of political communication. The plurality (at [73]) found the provisions vague enough that “lawful protests will be prevented or discontinued and protesters will be deterred from further protesting” because of police directions given under the Act, even where those directions had no proper legal basis. See also Lange v Australian Broadcasting Corporation (1997) 189 CLR 520 and Coleman v Power (2004) 220 CLR 1 for the underlying test the Court applied.
[4] Lon L Fuller, The Morality of Law (Yale University Press, revised ed, 1969) 65-70, 81-91, discussing the requirement that law avoid contradiction, both as a matter of formal logical consistency between rules and as a matter of the divergence between a declared rule and the administrative practice that grows up around it – what Fuller elsewhere in the same work calls the gap between “law in books” and “law in action”.
[5] Royal Commission into the Robodebt Scheme, Report (Final Report, 7 July 2023, corrected 11 July 2023) vol 2, ch 5, concluding that the use of income averaging as the sole basis for raising a debt under the Social Security (Administration) Act 1999 (Cth) was not authorised by that Act, and that the scheme required recipients to disprove debts calculated by a method the Commission found had no proper legal foundation.
[6] Commonwealth, Royal Commission into Aboriginal Deaths in Custody, National Report (Report, 1991), chaired by Commissioner Elliott Johnston QC, comprising five volumes and 339 recommendations directed at reducing the number of Aboriginal and Torres Strait Islander deaths in custody and addressing the underlying causes of overrepresentation in the criminal justice system. For subsequent implementation tracking, see Australian Institute of Health and Welfare and Productivity Commission reporting on Closing the Gap and deaths-in-custody data published in the years following the report’s release.
– – –
Bakchos is the founder of Blak and Black, an Australian media and advocacy platform established in 2010. Bakchos writes from the intersecting perspectives of Wiradjuri heritage, Jewish identity, and humanism.
© Bakchos, July 2026
