A whimsy inspired by Dworkin
Everyone knows that Hercules has been revered by generation after generation as a great hero, indeed a demigod, of ancient Greece. Criminal lawyers regard him with a special kind of awe, seeing him as a particularly interesting case study. He was found to have killed his wife and children, but got off with a community service order, known to us all as the twelve labours of Hercules. His cousin Theseus, who was apparently acting for him, pointed out that Hercules killed them in a fit of madness sent to him by his arch enemy the goddess Hera – so arguably one of the diminished responsibility defences would have applied.
But students of Jurisprudence know a different Hercules. Only students of Jurisprudence, who have read Dworkin, know that while Hercules did have superhuman abilities,[1] he was not an ancient hero at all, but a common law Judge, and therefore not even Greek – in fact Dworkin “supposes” that Hercules is a judge “in some representative American jurisdiction”.[2] One might be surprised to find him employed in any such modest position, and fancy that with abilities like those Dworkin attributes to him, he would surely have been elevated to the US Supreme Court. Or perhaps not.
According to Dworkin, Judges do not make the law, they find it. Even in hard cases.[3] The positivist view, that when a case cannot be decided by reference to an established rule of law the judge has “a ‘discretion’ to decide the case either way”[4] is, according to Dworkin, wrong. Dworkin’s ideal judge, who he names “Judge Hercules”, finds rather than makes the law, and does so in the following way. He formulates a political theory that would justify the constitution, statutes, case law, conventions, and standards of the jurisdiction in which he is working. This theory is nothing less than “the political morality presupposed by the laws and institutions of the community”.[5] Having identified that theory, hard cases are no longer matters for discretion – he simply extrapolates from that theory to get the correct decision in the case before him. Obviously there will always be only one right answer. While mortal judges can only do their best, they should follow the same process as far as mortal intelligence will allow.
Dworkin’s views are widely accepted, but is it even logically possible to do what Dworkin envisages Hercules as doing? Once, in a moment of whimsy during an otherwise ordinary day, I was seized by the determination to do the experiment. Have you ever tried to find the general proposition that encapsulates all the behests of any body of law, let alone all of the law? Is there in fact a first principle, (which of course might be a complex one) that logically implies all the lawful determinations made any field of law?
In some fields of law, it does actually seem possible to perform the labour that Dworkin assigns to Judge Hercules. For example, Administrative Law. As far as that field of law is concerned, I respectfully submit that Dworkin is right. Judge Hercules could do the task Dworkin sets him, indeed so could mere mortals.
Let’s start at its beginning. Once you have conceded that the King will not “proceed with force” against a person “except by the lawful judgement of his equals or by the law of the land”,[6] you have predicated your governing on the proposition that government must be according to law. And once you have done that, I suggest that you don’t have to import any other normative propositions to get to (for example) the philosophy that informs the Judicial Review Act[7] and the cases based on it. The whole canon of Administrative Law does indeed seem to be entirely consistent with and derivable from this single jurisprudential proposition: namely that government must be conducted according to law.
It goes like this. If executive government must be conducted according to law, then as the law changes, and the process of making the law changes, the principle that government must be conducted according to law adopts the new rules and new institutions and operates within them. So, when the principle that government must be according to law was conceded (in 1215), there was no legislature. But the first parliament was established 50 years later.[8] From that moment[9] the principle that government must be according to law then logically implied that government must be conducted in obedience to parliamentary legislation.
The proposition that government must be conducted in accordance with the determinations of the Judiciary also follows from the first principle. One of the functions of the Judiciary is to determine the meaning of legislation. In other words, articulating the consequences of the legislation in accordance with which government must be conducted. In other circumstances the pronouncements of the Judiciary are also posited as pronouncements of law. Ergo, Executive government must be conducted in accordance with them.
Looking at the particular behests of the canon of Administrative Law, I suggest that they all follow logically from the principle that government must be conducted according to law. By way of illustration let’s just run through a few of the major behests in the Judicial Review Act. I suggest that no other normative proposition is needed to entail those behests. Indeed, I suggest that the provisions of that and similar Acts merely identify and describe some of the circumstances in which it would make no sense to maintain that a decision was made in accordance with law.
The list in section 20 of the Act begins with requiring that decisions must be made with regard to the principles of Natural Justice.[10] Can anyone reasonably believe that a decision was being made according to law if the decision maker were biased? How could anyone be sure that it was the law and not the bias that led to the decision? Nemo debet judex in propria sua causa is not a separate normative principle, but a logical consequence of the first principle, that government must be according to law. Likewise, can anyone reasonably believe that a decision is according to law if the decision maker does not hear both sides? If she or he does not hear both sides, how could he or she even know whether the decision they were making was according to law? Audi alterem partem is not another normative principle, but a logical consequence of a commitment to the proposition that government must be according to law.
Then please consider the provision that allows a decision to be reviewed if procedures required by law are not observed.[11] If the omitted procedure is sufficiently important,[12] the decision maker can hardly be seen as making the decision that the law requires. And if the decision maker did not have the jurisdiction to make the decision,[13] or if the decision he or she made is not authorised under the enactment according to which it was supposed to be made,[14] the decision maker cannot rationally be seen as governing according to law: they are manifestly guided by something else.
The same considerations apply to all the improper exercise of power provisions in section 23 of the Act. If a decision is taken having regard to an irrelevant consideration or without regard to a relevant consideration,[15] how can it be a decision according to law? It is not even a decision about the case before the decision maker – just a decision about a case very like the one before the decision maker. And decision makers who act under dictation,[16] or who make decisions for an ulterior motive,[17] and so on, are not making their decision according to law, but according to other considerations entirely. Likewise, if a decision is Wednesbury unreasonable,[18] so unreasonable that no reasonable person would make it,[19] then it cannot be a decision according to law. Because such decisions are arbitrary, and law is meant to replace arbitrary government and is logically antithetical to it.
For these reasons I suggest that the entire canon of Administrative Law contains no normative provision except such as are derivable from the fundamental principle that government must be conducted according to law. The statutory provisions and the propositions of law in the casebooks are best seen as particularised articulations of the general proposition. They are but binding guidelines derived from one single principle, namely that Executive government must be conducted according to law.
If I am right so far, then Dworkin’s jurisprudence is right as far as it applies to Administrative Law, and the ambitious project that Dworkin has assigned to Judge Hercules is on the right track. The problem is that Hercules needs to clean out all the stables of Augeius, not just one of them.[20] It is not so arguable that other bodies of law are explicable by reference any coherently stateable general principle.
For instance, consider Criminal Law. For the sake of brevity, I will use the term colloquially to include regulatory offences as well. Now this body of law, seen in one aspect, is just a list of all the things thou shalt not do. But a mere list is of no use to Dworkin, or to Justice Hercules, when he is struggling with a hard case, and has to decide what to do when the conduct of the accused falls within a lacuna in respect of which the previously written or articulated law is silent. Given that what Hercules J proposes to do is to find a political theory that justifies all the decisions of a jurisdiction, and which that is going to imply how he should decide the case before him, the list alone is not going to help him to decide whether the prisoner before him has done something that falls within the description of what is on the list. He needs a normative proposition to do that. Ever since David Hume pointed out that you cannot derive a moral imperative from a description of reality,[21] moral philosophers have been wrestling with what they commonly refer to as the “is/ought gap.” Bottom line is that Judge Hercules needs something action guiding, and you can’t get that from a list that doesn’t cover the circumstances before you. One needs a grundnorm[22] that includes a theory of culpability and punishment, and which implies the criminal law of the jurisdiction.
So, according to Dworkin, Judge Hercules must identify “the political morality presupposed by”[23] the list of things that thou shalt not do. And, I am going to respectfully suggest, he will be unable to identify any such thing, because no normative theory will recommend punishment of all the conduct which the criminal law prohibits, and any normative theory will prohibit some conduct that the criminal law will not punish.
But let me go one step at a time. Let’s start at the beginning of this field of law too. King Hammurabi of Babylon, who around 1750 BC was as far as we know the first to publish a set of criminal laws, also spelled out the “political morality”[24] on which he based his laws. You might say he did a bit of a Dworkin. But his philosophical position was not necessarily internally consistent. First, he was, he said, the agent of Anu, “the lord of Heaven and earth” (as his Divine Majesty the sky god then was). In this way Hammurabi cast himself as a religious reformer aiming to “bring about the rule of righteousness”.[25] But he was also a retributivist, literally an eye for an eye retributivist,[26] his project being to “destroy the wicked and evil doers”.[27] At the same time, he was a utilitarian seeking to ensure that “the strong should not harm the weak” and that “the wellbeing of humankind” should be advanced.[28]
But none of these strands of thought are going to help Judge Hercules. First because every jurisdiction has victimless offences, and second because every jurisdiction has offences that are committed merely by being on the wrong side of an arbitrary line in the sand. Whether you are a retributivist or a utilitarian, (or both, as Hammurabi and many since him have tried to be) your philosophy is not going to imply the entirety of the canon of criminal and regulatory law.
Think of any victimless offence you like. The example that occurs to me is the person who drives their critically ill partner to the emergency department of the local hospital even though the only vehicle available to them is roadworthy but unregistered. Neither a retributivist who believes that wrongdoers should be punished just because they are culpable, nor a utilitarian who believes that the only point in punishment is to protect society, will find in their philosophy any normative basis for punishing someone who has committed a victimless offence such as this. Courts punish some offenders, at least to the extent of finding them guilty whether they are further punished or not, for no other reason than that they have broken the law.
Even some offences that are arguably not victimless wear the aspect of offences that are punished for no other reason than that they constitute a breach of the law. Consider the fine lines that, of necessity the law has to draw. The law draws lines for distance time and age.[29] Any of these will make the point. We could turn our thoughts to time, and ponder the old common law year and a day rule for murder that still exists in some jurisdictions. Or to age, and spare a thought for the fine young Australians who did not get a vote in the last election because they were 17 years and 364 days old. But let us consider distance. Two diners leave their table for a smoke break. One of them is the requisite 5 metres from the doors of all non-residential premises. The other is 4.9999 metres away. If one of their “victims” had reported them, that individual would be surprised to discover that the legal system ended up convicting only one and found the other to be an upright citizen.
No doubt one of the two is guilty and the other innocent – but any candidate for “the political morality presupposed by”[30] the jurisdiction would be hard put to find the guilty one more culpable than the other. Yet the law has to draw a line in the sand, and when it is randomly drawn – the mandated distance could have been set at ten metres, or four metres – no political morality can imply it. The normative cannot imply the random.
For good measure let us imagine that the two persons of interest are facing each other and the one that is standing the legal distance away is facing the al fresco diners, and as a result is exhaling in their direction. The illegal one is facing the other way and is not doing so. If the al fresco diners were to smell something, it would likely to be caused by the legal one, not the illegal one. Anyone who was prepared to make the effort to apply their political morality to this detail would find that it would attribute more culpability to the lawful conduct than the unlawful conduct. But the law, being the law, has to draw a line in the sand, and this is the consequence.
And where do victimless offences and arbitrary lines in the sand leave Judge Hercules? They leave him without the ability to induce a “political morality presupposed by the laws and institutions of the community”.[31] The reality is that some conduct is punished, not because it violates some political morality or other, but just because the law has prohibited it.
Let us dwell on this a moment longer. In any organised society the list of things that thou shalt not do will have to include not only conduct that is mala in se but also conduct that is mala prohibita. And every legal system is going to have to draw fine lines.
No doubt obedience to the law is a virtue. And it might be argued that the political morality that Judge Hercules will posit as that which is presupposed by the laws of the jurisdiction, will have as one of its tenets that obedience to law is a duty. It might be churlish to argue that this is merely assuming what Judge Hercules is seeking to justify. Instead, I will argue that even allowing obedience to law as part of the political morality, of the jurisdiction, Judge Hercules will still be unable to find a basic normthat presupposes the punishment of victimless offences or implies the culpability of persons who are on the wrong side of an arbitrarily chosen line. Because normative standards, being normative, cannot inure themselves to moral differences or prescribe arbitrary outcomes.
If he cannot find the political morality presupposed by all the laws of the jurisdiction, could Judge Hercules find the political morality presupposed by most of the laws of the jurisdiction – that is all except the victimless offences and the offences defined by fine lines? Well, if he could, he still would not be able to work with it. His normative theory would criminalise too much. Not only is any normative theory of punishment going to fall short of implying the whole canon of any jurisdiction’s criminal law, but it is also going to imply that people the law elects not to punish ought to be punished.
Consider the concept of moral luck. Two friends go out drinking and have far too much. They both recklessly choose to drive themselves home. One gets home safely and without being breathalysed: the other kills a pedestrian and is jailed for Dangerous Driving Causing Death. Every “political morality” Judge Hercules can think of will identify the party who got home safely as culpable, but the legal system has no penalty for them.
Consider also the concept of limited duty. A couple of strong swimmers are taking a stroll along the beach and notice a stranger, to whom they have no obligation to provide the necessaries of life,[32] drowning not far from the shore. They walk past muttering “Thou shalt not kill but need’st not strive officiously to keep alive”[33], and the law does not criminalise them. But any normative theory that Judge Hercules might conceive of as “the political morality presupposed by the laws and institutions of the community”[34] would condemn them – and praise the Good Samaritan who risked her or his life to save the stranger.
In sum judge Hercules is not going to find a normative theory that entails all the criminal laws of a jurisdiction, nor will he find one that entails only the criminal laws of a jurisdiction.
Should you ever, gentle reader, find yourself with an idle moment, you might like to do the same test with other fields of law that are familiar to you. It may be that Dworkin’s distinction between principles and rules is of assistance.[35] My guess would be that fields of law that are largely based on principles are more likely to be receptive to Herculean efforts that those that are based on rules. And you will probably find that of all the fields of law that he has to compass, finding the political morality underlying the Criminal Law is the hardest of the labours of Judge Hercules.
Indeed, if either of them existed, Hercules the Judge might well envy the labours of Hercules the Hero. Because getting girdles from Amazon Queens,[36] and triple headed dogs from beyond the gates of Hell,[37] physically impossible though those labours may have been for a mortal, Hercules the Hero at least had the advantage that he was not required to do anything that was logically impossible.
[1] Ronald Dworkin, Taking Rights Seriously, Harvard University Press 1977, page 105
[2] Ronald Dworkin, Taking Rights Seriously, Harvard University Press 1977, page 105
[3] Ronald Dworkin, Taking Rights Seriously, Harvard University Press 1977, Chapter 4, passim
[4] Ronald Dworkin, Taking Rights Seriously, Harvard University Press 1977, page 81
[5] Ronald Dworkin, Taking Rights Seriously, Harvard University Press 1977, page 126
[6] Magna Carta, Article 39
[7] Judicial Review Act 1991 (Qld)
[8] As a result of the Baronial revolt led by Simon de Montfort
[9] When I say “from that moment” I mean only as a matter of logic. In terms of history, the doctrine of parliamentary supremacy was not fully recognised until around the middle of the 17th century after the civil war. Courts and kings continued to act as though the decisions of parliament were reviewable until then. Henry VIII clauses, which enable the Executive to effectively amend an Act, continue to be referred to as such in the Queensland Parliament today, and are specifically prohibited by the Legislative Standards Act 1992 section 4(4)(c)
[10] Judicial Review Act 1991 (Qld) section 20(2)(a)
[11] Judicial Review Act 1991 (Qld) section 20(2)(b)
[12] Project Blue Skies v Australian Broadcasting Authority 194 CLR 355
[13] Judicial Review Act 1991 (Qld) section 20(2)(c)
[14] Judicial Review Act 1991 (Qld) section 20(2)(d)
[15] Judicial Review Act 1991 (Qld) section 23(3)(a) and (b)
[16] Judicial Review Act 1991 (Qld) section 23(2)(e)
[17] Judicial Review Act 1991 (Qld) section 23(2)(d)
[18] Wednesbury Corporation v Minister for Housing and Local Government [1965] 1 All ER 186
[19] Judicial Review Act 1991 (Qld) section 23(2)(g)
[20] The cleaning of the Augeian stables was traditionally the fifth labour of Hercules, the fourth being the capturing of the Erymanthian boar, and the sixth being the elimination of the Stymphalian birds.
[21] David Hume, A Treatise of Human Nature, 1740
[22] Hans Kelsen, General Theory of Law and State, 1945: Kelsen uses this term to describe a putative basic norm that justifies the legal system – a concept interestingly similar to Dworkin’s idea of a “political morality” underlying the law.
[23] Ronald Dworkin, Taking Rights Seriously, Harvard University Press 1977, page 126
[24] Ronald Dworkin, Taking Rights Seriously, Harvard University Press 1977, page 126
[25] The Code of Hammurabi, circa 1750 BC, preamble
[26] The Code of Hammurabi, circa 1750 BC, article 196
[27] The Code of Hammurabi, circa 1750 BC, preamble
[28] The Code of Hammurabi, circa 1750 BC, preamble
[29] Acts Interpretation Act 1954 Part 9
[30] Ronald Dworkin, Taking Rights Seriously, Harvard University Press 1977, page 126
[31] Ronald Dworkin, Taking Rights Seriously, Harvard University Press 1977, page 126
[32] Criminal Code section 285
[33] From “The latest Decalogue” by Arthur Clough
[34] Ronald Dworkin, Taking Rights Seriously, Harvard University Press 1977, page 126
[35] Dworkin says that rules require a particular decision if the facts they adumbrate are proved. A legal principle “states a reason that argues in one direction, but does not necessitate particular decision.” Ronald Dworkin, “The Model of Rules” 1977 page 26. One of his examples is the principle that no person may profit from their own wrong.
[36] Traditionally the ninth labour of Hercules
[37] Traditionally the twelfth labour of Hercules
The Honourable Dean Wells was Attorney General of Queensland between 1989 and 1995, and is currently in practice at the Queensland Bar. The Honourable Rod Welford was Attorney General of Queensland between 2001 and 2005, and in 2003 introduced into the Queensland Parliament the Bill which was enacted as the Civil Liability Act 2003 (Qld). Hearsay thanks them for this contribution apropos of the provision of such Act pertaining to proof of causation in breach of duty of care causes of action.
But for a seminar supported by the Bar Association we would not have thought to write this note. In September last year Professor Stapleton gave a presentation entitled “Unnecessary and Insufficient Factual Causes”,[1] in which she examined the jurisprudence underlying the role of causation in tort law. The commentator at the seminar was Pat Keane KC, then very recently retired from a distinguished term on the High Court Bench.
Professor Stapleton’s discussion focussed on the traditional role of necessary and sufficient conditions in tort law. Drawing examples from many jurisdictions, she advanced the thesis that a large number of cases in modern tort law could not really be explained in terms of the “but for” test of causation. Although the main thrust of her argument was of universal application, and her examples were international, she did mention one specifically Queensland matter: in particular she suggested that Queensland’s Civil Liability Act[2]opened the door to finding liability in circumstances in which the “but for” test would not permit it to be found.
This note is about only the small part of Professor Stapleton’s thesis that related to the Civil Liability Act.
The Minister’s Second Reading Speech introducing the Bill for the Act, made on 11 March 2003, did not address whether the Parliament intended to any extent do away with the “but for” test. It did note that the Bill was not intended to create any new cause of action, and that it was not meant to be a codification of the law. The authors of this note are very conscious of the fact that in the absence of an explicit statement relating to the “but for” test, the only useful information as to the intention of the legislature is obtained by judicial statutory interpretation. So we are not purporting to give evidence on what the parliament intended.
What we can do is to draw attention to some propositions of legal philosophy that appealed to us at the time and would justify the prescriptions in the Act in question. In summary, we suggest that the jurisprudence that would justify the legislature’s language in the Act is entirely consistent with a world in which the courts continue to apply a necessary and sufficient test. Indeed we suggest that there is nothing in the Act with which Hart and Honore would have a problem.[3]
For ease of reference, we set out section 11 of the Act so far as is relevant to this discussion:
(1) A decision that a breach of duty caused particular harm comprises the following elements—
(a) the breach of duty was a necessary condition of the occurrence of the harm (factual causation);
(b) it is appropriate for the scope of the liability of the person in breach to extend to the harm so caused (scope of liability).
(2) In deciding in an exceptional case, in accordance with established principles, whether a breach of duty—being a breach of duty that is established but which cannot be established as satisfying subsection (1)(a)—should be accepted as satisfying subsection (1)(a), the court is to consider (among other relevant things) whether or not and why responsibility for the harm should be imposed on the party in breach.
Rather obviously the Act grants the courts a very wide discretion. Tort law is an area in which a sensible legislature would want to allow courts the flexibility to have regard to all of the particular circumstances of a case. We note however, that this wide discretion was not given in a vacuum: the provision specifically refers to the retention of “established principles”. The legislature thus accorded the courts that wide discretion in circumstances in which the same test of causation had the place it did under common law.
Deploying the concept of “responsibility for the harm”, the Act allows that a person, whose contribution to the harm in question is not of itself necessary or sufficient to cause the harm, may be held liable if their conduct nevertheless contributed to the cause of the harm (to an extent that the court considers such conduct entails “responsibility”). When seen in its full legal context, this does not do away with the necessary and sufficient test: it makes it possible for the courts, if they choose to do so in a particular case, to refocus it. Reading the legislative provision together with the common law at the time, if one looks for the philosophical position a lawgiver would have to hold to prescribe it, it can be seen to be a provision based on the jurisprudence that a court should be able to hold a person liable in tort if their act was part of a set of conditions which, taken as a whole, were necessary and sufficient to cause that civil wrong.
The Queensland Parliament has often used such a device. For example, in the party provisions of sections 7 and 8 of the Criminal Code. The guilt of a person who is a party to a crime does not depend on whether his or her contribution was independently sufficient to ensure the commission of the crime. Culpability is established by their participation, with others, in a set of actions which did in fact culminate in the commission of a crime. Whether their participation was effective or not is not the question. The question under the Code is whether a crime was committed, and if so whether the accused was a party. A legal philosopher might justify such a provision in the Code by noting that what the accused is alleged to have done was a malum in se.
The jurisprudence that would explain section 11(2) of the Civil Liability Act is, we suggest, that it should be open to a court to find liability if someone has done something that made them a party to the commission of tortious harm. As we noted above, Pat Keane KC was the commentator on Professor Stapleton’s paper. He has made the point that historically responsibility for causing harm in tort law was resolved as a question for the jury. To use his phrase, there was thus always a moral squint necessarily involved in the determination of this question. He also noted that nothing in the legislation suggested an intention to alter the nature of the question.
The public policy that would justify treating a party to a tort as a tortfeasor is simply a public policy to discourage the infliction of public harm. Which, after all, is the core business of any legislature. A legal philosopher might argue that the legislature was extending the concept of “malum in se” by analogy from the criminal sphere to the sphere of civil wrongs.
The point is that the provision provides the courts with the discretion to find that even when the actions of a defendant are not of themselves a necessary or primary causal trigger for the resulting harm, the actions may nevertheless give rise to a level of culpability that constitutes sufficient grounds for holding the defendant accountable for the harm. The application of this principle is most evident in the context of two or more defendants contributing to the resulting harm.
However, it is not unreasonable to expect that it could apply equally to a single participant in the civil wrong. The Act makes it clear that liability can be found if, for example, the act was performed in circumstances where the defendant’s action was likely to have consequences, which would not have occurred in the absence of those circumstances. The thinking here has obvious analogies to the jurisprudence that courts rely upon in “egg shell skull” cases. This, we respectfully suggest, can also be explained as viewing the act in question in the way we suggest above: as part of a set of conditions which, taken as a whole, were necessary and sufficient to cause that civil wrong.
Accordingly, we submit that the underlying jurisprudence of the Civil Liability Act is not such as to set aside or overturn the precedents based on the existing “necessary and sufficient” test in Queensland law. It simply expands its ambit of application. Thus, for example, if a number of persons recklessly or negligently pursue a course of action that is apt to cause harm, or is apt to contribute to causing harm, to another person, then to ask whether the actions of a particular agent alone were both necessary and sufficient to cause that harm, is to ask only one of the questions permitted by the Act. The other question is whether the action of that agent was part of a set of conditions which were, collectively, necessary and sufficient to cause the harm.
[1] Jane Stapleton KC (Hon) FBA, Life Fellow, Christ’s College Cambridge, Hon Prof. U Sydney Law School: her seminar “Unnecessary and Insufficient Factual Causes” was held in the Banco Court on 14 September 2023.
[2] Civil Liability Act (Qld) 2003.
[3] Hart and Honore, Causation in the Law, Oxford University Press 1959.
