Constitutional Principles and Coherence in Statutory Interpretation

La Trobe Law School Symposium on the coherence of statutory interpretation

RTF version 207 kb

This broad topic was given to me by Professor Keyzer and may I say, at once, that it is not only imposing but also, potentially, ambiguous. I say that with good reason. As frequent interpreters of Commonwealth legislation, Federal Court judges as a caste, have an eye keenly attuned to the detection of ambiguity.^[1] The topic could, for example, be asking what effect constitutional principle has upon the coherence of statutes as interpreted or, perhaps more tantalizingly,^[2] it could be referring to the impact of constitutional principle upon the field of law known as statutory interpretation. The former topic makes, I think, for a short meal. The law about coherence in the actual interpretation of statutes is clear. We have it on the authority of Project Blue Sky v Australian Broadcasting Authority^[3] ('Project Blue Sky') that a statute is to be construed on the prima facie basis that its provisions are intended to give effect to harmonious goals so that where the language in which it is expressed gives rise to conflict, the meaning of the conflicting provisions is to be 'adjusted'.^[4] What kind of adjustment is required? One that will best give effect to the purpose and language of the provisions whilst 'maintaining the unity of all the statutory provisions'. In pursuit of this lofty principle, the High Court counsels us to give meaning to every word of the provision so as to avoid, so far as is possible, superfluity in the language in which the statute has been cast. Later in this speech I am going to return to this principle which, in a practical sense, is perhaps the most significant principle in statutory interpretation.^[5]

So interpreted, a discussion of the topic suggested by Professor Keyzer might, quite possibly, stop right here with not much more to say than has just been said. Whilst this might be a mercy to the listener, and certainly for the speaker, this is not, I apprehend, what was intended. I propose, therefore, to focus upon the second interpretation of the topic and to take a brief sojourn through the field of statutory interpretation itself, to ponder its coherence as a topic of legal discourse and then to assess the impact of constitutional principle upon that coherence.

Some immediate difficulties arise which stand in our way. What do we mean by coherence? What is a statutory interpretation and how might it be coherent? And, what are constitutional principles? It will be obvious that none of these questions is small. Largely fortuitously, it seems best to answer them in the order just posed.

A few words about coherence

As a matter of ordinary contemplation, coherence denotes a state of affairs free from inconsistency or perhaps to put it another way, a state of affairs which is logically structured and does not suffer from internal contradiction. Of course, those who have studied logic either as an aspect of philosophy or of pure mathematics^[6] will readily tell you that there is no link between truth and coherence. Perfectly logical arguments can be quite without content because logic is concerned with the validity of inferences drawn from premises and not with their truth or usefulness. An excellent example, which I owe to the late Dr John Bacon of the University of Sydney, is that of Skippy and the Helicopter. The following argument is logical:

Skippy is a crocodile.

All crocodiles are helicopters.

Ergo, Skippy is a helicopter.

Whilst logical the argument is also absurd; indeed, it may be doubted that it has any content at all beyond its formal validity.

Many people in the law believe that formal validity – 'logic' or 'coherence' if you will – is a worthwhile goal. I am not sure, however, that all those who invoke coherence as an informing legal value necessarily have themselves the most complete grasp on the distinction between valid reasoning (with which logic and coherence are solely concerned) and arguments that are correct or right (which may or may not also be logical or coherent). For example, the fact that an argument which is illogical leads to a particular outcome does not mean that the outcome is itself incorrect or false. I may reason illogically that because all cats are animals and because Felix is an animal that Felix is a cat. But my argument's lack of logic tells one nothing about whether Felix actually is a cat; in particular, the fact that my reasoning is illogical does not establish that my conclusion is wrong and that Felix is not a cat. Erroneous reasoning of that kind has a name: it is the fallacy fallacy or for those who like Latin, the argumentum ad logicam. The fallacy fallacy has, I fear, become endemic within the legal profession, the academy and the judiciary.^[7] It is very common in legal discourse to claim the contrary view is illogical and hence wrong but, as I hope I have just explained, that last step is itself illogical.

This is not necessarily a cri de coeur on my part that we should all begin to reason irrationally; rather, it is just to sound a note of caution about the limits to which coherence as an informing concept in shaping principles of justice may be put.

I am by no means the first person to blow on this trumpet. Much more famously it was Oliver Wendell Holmes Jr who observed that '[t]he life of the law has not been logic; it has been experience'.^[8] That famous statement has often enough been called upon by judges seeking to resist logical arguments which, if acceded to, would lead to results regarded as unpalatable. For example, many people have thought that in personal injury cases the head of damages known as Griffith v Kirkemeyer^[9] damages are illogical. This head of loss allows a badly injured plaintiff to recover from a defendant the commercial cost of voluntary services provided to her. The overarching position in negligence is, of course, that a plaintiff can only recover loss suffered by them^[10] but, on one view, Griffith v Kirkemeyer damages appear to be an example of the plaintiff recovering for loss actually suffered by someone else, namely, the carer. In CSR v Eddy (2005) 80 ALJR 59 at [91] McHugh J explicitly invoked Holmes' dictum to justify the anomalous position of this head of damages. Indeed, according to his Honour, 'Griffith v Kirkemeyer illustrates the truth of Holmes' dictum'. McHugh J is not alone. In another area of tort law, that concerned with the difficult question of whether psychiatric injury should be regarded as different to personal injury, Spigelman CJ invoked Holmes to say that '[i]n this field, the law manifests the classic aphorism of Oliver Wendell Holmes Jnr'.^[11] Indeed, he went somewhat further by unearthing^[12] a statement by Fullager J of which I must confess I was quite ignorant until I prepared this paper. We are, according to Fullager J, to resist:

'….the temptation, which is so apt to assail us, to import a meretricious symmetry into the law.'^[13]

'Meretricious' is a pretty harsh word. It derives from the Latin word meretrix, which means prostitute. 'Meretricious' has come to mean the quality of being apparently attractive but of having no real value. I think maybe Fullager J went a little too far in this; indeed, I doubt he was intending to suggest that judges should give up reasoning logically altogether. It was more, I think, a reminder that we should not be blinded entirely by formality.

That is a proposition which, so it seems to me, is acceptable. Without some form of commitment to logical reasoning there is a systemic risk that cases that are essentially or relevantly identical will be decided differently. Whilst there is a good deal of debate about the nature of justice there is a degree of support for the idea that it includes, at least, the notion that similar cases should be decided the same way. And that principle is, in substance, no more and no less than the principle of equality before the law. That is why the High Court has always turned its face against 'palm tree justice',^[14] that is to say, cases where a court simply decides a case entirely by how it personally feels the case should be decided. How the propinquity of the judge to a palm tree is apt to encourage this style of reasoning has always eluded me.^[15]

Consistently with this view of the issue, there are many statements of the highest authority urging judges towards coherence. Miller v Miller[16] concerned the question of whether the drunk driver of a stolen car owed the drunk passenger who had stolen the car a duty of care such that losing control of the vehicle and crashing into a pole was actionable in negligence. Breach of duty was admitted; the only issue was whether the duty existed. Of central concern was the interaction between tort law and the unlawful conduct of the plaintiff. At [15]-[16] six justices said:

 

'These reasons will show that the central policy consideration at stake is the coherence of the law. The importance of that consideration has been remarked on in decisions of this Court. Its importance in this particular context was emphasised by the Supreme Court of Canada. It is a consideration that is important at two levels. First, the principles applied in relation to the tort of negligence must be congruent with those applied in other areas of the civil law (most notably contract and trusts).

Second, and more fundamentally, the issue that is presented by observing that a plaintiff was acting illegally when injured as a result of the defendant's negligence is whether there is some relevant intersection between the law that made the plaintiff's conduct unlawful and the legal principles that determine whether the plaintiff should have a cause of action for negligence against the defendant. Ultimately, the question is: would it be incongruous for the law to proscribe the plaintiff's conduct and yet allow recovery in negligence for damage suffered in the course, or as a result, of that unlawful conduct? Other questions, such as whether denial of liability will deter wrongdoers or advantage some at the expense of others, are neither helpful nor relevant. And likewise, resort to notions of moral outrage or judicial indignation serves only to mask the proper identification of what is said to produce the response and why the response could be warranted.'

I have omitted the footnotes to this passage but the second sentence is accompanied by footnote 8 which contains a number of authorities all emphasizing the important of coherence as a legal value. They are: Sullivan v Moody[17] ('More fundamentally, however, these cases present a question about coherence of the law'); Agricultural and Rural Finance Pty Ltd v Gardiner[18] ('the need for coherence of legal principle and the effects of overly broad interpretations of waiver and estoppel upon other doctrines must be borne in mind'); and, CAL No 14 Pty Ltd v Motor Accidents Insurance Board[19] ('Another difficulty obscured by the narrow formulation of the duty of care in the light of the particular eventuality which came to pass is that of legal incoherence'). Other more recent cases also emphasise the importance of coherence as a legal value.^[20] Perhaps these are all best summarized by the observation of Lord Devlin in Hedley Byrne & Co Ltd v Heller & Partners Ltd[21] that 'no system of law can be workable if it has not got logic at the root of it'.

There is then an obvious tension between the Lord Devlin's view about the root need for logic in a legal system and Holmes' observation about the life of the law not being logic but experience. The High Court is not itself unaware of this tension. It was explicitly referred to by five justices in Collector of Customs v Agfa Gavaert[22] and both McHugh J and Gleeson CJ have separately discussed the tension between the two positions.^[23]

One has the position then that the law's approach to the value of coherence as a legal principle is itself not altogether coherent. It seems that the prevailing value is logic but that perhaps, to borrow the words of Branson J in a different context, it is at best to be seen as a 'counsel of perfection'.^[24]

Statutory Interpretation and Coherence

There is no particular skill involved in locating coherence issues in general legal discourse. This is because most areas of law are made up of rules calling for obedience by the subject or application by the judges. A coherence issue will arise in such a system wherever two rules in it conflict. This can happen internally where two principles in the same field of law appear to contradict each other. It can happen externally where disparate fields of law are brought into contact and require apparently conflicting outcomes. A very straightforward example of an internal conflict might be afforded in constitutional law by a State law which conflicts with a paramount federal statute. In such a case, the resolution is provided by s 109 of the Constitution. An example of an external conflict might arise from the imposition of a duty of care upon state officials in the exercise of some statutory power (perhaps the power to decide to fill a pot hole in a road) and the refusal of administrative law to have a Court review such a decision on its merits. The former tort rule requires a Court to decide whether the official's actions were reasonable; the latter forbids any such consideration by a court because it would infringe upon the separation of powers. My present point is only that every legal system is riddled with such conflicts and it is the everyday task of courts, especially intermediate appellate courts, to sort them out.

But consider the law of statutory interpretation. Although there are rules of law in statutory interpretation, much of this field of law is not made up of rules of law at all but rather by non-binding principles of interpretation. There are long lists of these principles in the textbooks. To cite just six well-known ones:

• in interpreting a provision of an Act, the interpretation that would best achieve the purpose or object of the Act is to be preferred to each other interpretation;^[25]
 
• the meaning of a word is to be derived from its context.^[26] Thus, in the expression 'house, office, room or place' the word 'place' does not include a public lane. Consequently, a person accused of betting in a public laneway should not have been convicted of betting in 'a place';^[27]
 
• general matters are constrained by reference to specific matters.^[28] Thus 'a railway, road, pipeline or other facility' does not include a storage facility;^[29]
 
• express reference to one matter excludes reference to another matter;^[30]
 
• a statute is to be construed on the prima facie basis that its provisions are intended to give effect to harmonious goals;^[31] and
 
• constructions involving superfluity should be avoided.^[32]

None of these principles requires any particular outcome. Furthermore, there is no set of principles which tell one what to do when more than one principle of statutory interpretation might apply and generate dissonance. There is no rule, like s 109, that tells one that the ejusdem generis rule prevails over the exclusio alterius rule. At best, there is an informal and somewhat hazy understanding in some quarters that some rules are more or less important – perhaps fashionable – than others. Thus, at the moment, we are told that the exclusio alterius rule is 'a valuable servant, but a dangerous master'.^[33] So too, most judges would recognize that the principle of harmonious construction advocated in Project Blue Sky is somehow a more cardinal principle than many others. Why some principles are in vogue and others not must remain always something of a mystery.

The present point is that these principles do not, and cannot, really conflict because none of them actually requires any particular outcome. Once that is appreciated it may appear that statutory interpretation is, for the most part, quite unlike other fields of law in that many of its principles are not really rules of law but rather merely available approaches to resolving ambiguity.

For myself, I think that observation tends to decrease the utility of asking whether statutory interpretation, as a field, is coherent. The coherence inquiry presumes the presence of rules which have a sufficiently hard edge that they can be seen to be inconsistent with one another. There are, to my mind, serious difficulties in asking whether, for example, as a matter of general principle the approach in Project Blue Sky is consistent with s 15AA of the Acts Interpretation Act 1901. The question just does not really make much sense.

Coherence more generally

Nevertheless, we may yet be able to identify a coherence issue if we can locate principles of statutory interpretation which transcend the ordinary kinds of rules enumerated above and which apply more generally. I think that this can be done, but it necessitates a certain Cartesian skepticism to pare the topic back to its essentials so that its underpinnings are exposed. This requires one to ask what it is that a court is doing when it interprets a statute. In putting my emphasis on what courts do, I am not seeking to ignore that there are many other persons and entities which use statutes besides courts (a point sometimes made by Parliamentary Counsel to judges who complain about legislative drafting styles). However, the role of the courts in interpretation is unique because theirs is the interpretation which is authoritative and binding on everyone else. How other people might read legislation, whilst a legitimate concern for those involved in drafting, seems to me to have little relevance to the constitutional function which courts discharge when doing so themselves.

What then is happening when a court interprets the written text which is a statute? One matter is clear: what is involved is a speech act by the legislature and a listening act by the courts. What are the features of such a situation? There are a number of matters worth pointing out. First, largely this is not a dialogue. The principal role of the legislature is to speak and that of the courts to interpret. With limited exceptions, there has not historically been much of an indication that what is taking place between courts and legislatures is a dialogue in which both parties are, in some sense, participating. That observation is contrary to some more recent statements to which I am going to return concerning the principle of legality. But before it burst on to the scene, there was only limited evidence for such a dialogue. There might be a dialogue in which the legislature is the listener to the extent, I suppose, that Parliamentary Counsel draft legislation with full knowledge of the interpretative rules which the Courts are going to apply. In that sense, the form of legislation will often, although not inevitably, be drafted in a way which is not tone deaf to the way the courts are likely to read it. But this is a fairly emaciated form of dialogue. So too, the courts are in some circumstances inclined to treat various legislative actions as endorsing prior judicial determinations.^[34] Whether this is a true dialogue or rather an assumption on the judiciary's part that it has been listened to may be a different question.

But these kinds of matter are quite limited examples of dialogue. Even assuming there is a dialogue in relation to the principle of legality (a proposition which, on empirical grounds, I venture to doubt below), even its most doctrinaire adherents would not argue that there is a more general dialogue between courts and legislatures about the meaning of legislation outside the context of fundamental rights.

So that is the first feature: we are concerned with a speech act which either is a monologue or, at best, a very limited form of dialogue. The role of principal listener and interpreter rests with the courts.

Secondly, we are concerned with a very particular kind of speech act. It is at the extreme end of formality. Indeed, it is difficult to imagine more formal English than that used in a statute. The law is not in slang. It contains no jokes. It is not read for fun or even for betterment. It is the formal embodiment of the outcome of the community's political processes. Further, not only is it highly formal but it is also the product of legislative chambers which are deliberative in nature and which consist of many persons. These are important matters. In every human language the listener derives considerable insight into what the speaker is saying by the context in which it is being said. Reflection on ordinary speech shows this to be so. The English we read and write is a different dialect to the English we speak, a fact to which we are so inured that we tend not to notice it. Written English is generally expected to make sense and be grammatically correct. Its sentences are orderly. This very paragraph may, hopefully, serve as an example of this dialect. On the other hand, the spoken English which we carry on between us, every day, is quite different. Its chaotic sentences are usually not finished. Verbs are left hanging. The speakers cut across each other; fragments of sentences lie scattered all about. The meaning of this mish-mash emerges not from the grammar of what is being said (which barely exists at all) but most often by the listener instinctively understanding what the speaker is attempting to say. This happens even in the comparatively formal setting of a courtroom. Here is some transcript of oral submissions recently made to me by reputable senior counsel:

But even if they were concerned about that risk, the fact that both people have significant risks does not promise in any way what the answer will approximate what the tribunal would do, for the very reason of what you can only say about it, which Mr Thorpe agreed, there was a substantial risk that would be substantially different. So if you've got a deal which is premised on that an assessment by a tribunal of the real value could be substantially different in each direction from that which is achieved, and that's the premise upon – that's the analysis, it seems somewhat strange, with respect, to then say, "That's the value."

I was not struck by the thought at the time that the submission made to me was gibberish, but that is what it appears to be on the page, robbed of its surrounding context of body language and any general understanding at the time of what was being talked about. This is not an isolated phenomenon. If you listen carefully to almost any conversation between humans you will discover that, grammatically, it very rarely makes any sense at all. Here is a transcript of an ordinary conversation retrieved from google.com:

'Note: (.) indicates a short pause

SALLY: so how's your new job going 

JOHN: well (.) you know (.) all right 

SALLY: what you doing then 

JOHN: this week I've just been on the phones 

SALLY: what (.) selling 

JOHN: no they're doing like a survey (.) what people think of the the service and all that 

SALLY: that's a bit off (.) so you've got to deal with all the moaning and complaining all the flak 

JOHN: no (.) no it's just like a questionnaire were your goods delivered on time yes or no 

SALLY: d'you think you'll stick it 

JOHN: next week I'm with the regional manager out on the on the road (.) that'll be all right 

SALLY: I was on the phones at the Town Hall (.) dead boring 

JOHN: tell me about it 

SALLY: so what's your mate doing 

JOHN: who 

SALLY: you know whatsisface (.) had the Mohican 

JOHN: Gizmo 

SALLY: yeh 

JOHN: he's gone back to college 

SALLY: weren't he dead good at computers (.) I thought he passed everything 

JOHN: he got a good grade for computing but he's doing resits (.) he couldn't get in (.) where was it (.) I don't know (.) Manchester or somewhere'

It is interesting that spoken English is so very incoherent. Just as interesting is the fact that no-one appears to be troubled by this and understands perfectly well what is being said. If nothing else teaches this lesson, surely this phenomenon shows that context is not peripheral to understanding speech acts but dramatically central.

Experience bears this out. The sentence 'He fed her cat food' has two potential meanings. Only by knowing the context in which it was said (captive woman force-fed cat food versus kindly neighbour looking after cat) can we unravel it. Further, it is not just the context in which the speech act occurs that determines its meaning. It can also be the identity of the intended audience. Thus, when spoken, the sentence 'The koala eats shoots and leaves' has a different meaning for a room full of marsupial biologists than it has for a room full of comedians.^[35]

So what does this mean for statutory interpretation? The high degree of formality and written nature of statutory language guarantee that its interpretation is to be approached in an equally formal way. The listener will assume that the words used have been carefully chosen and that the speaker means precisely what it is saying. This is quite contrary to spoken English and less formal versions of written English. If you want to see how true that is, try applying Project Blue Sky and the ejusdem generis rule to, depending on one's age, an episode of My Kitchen Rules or Gogglebox.

This assumption about the careful use of language allows a series of deductions to be made about meaning from the linguistic choices which have been made. The use of one word here and another different word there permits an inference to be drawn that a difference was intended; the use of general words in one place and particular ones in another carries with it a legitimately inferred connotation. Because the language is formal and considered, the listener is entitled to assume that a coherent meaning was intended. Professor Pearce calls interpretative assumptions of this kind 'intrinsic principles' of interpretation and that is, with respect, a useful label. Each of the intrinsic principles of statutory interpretation can, I think, be traced to an inference arising from the formality of the language. It may be that the waxing and waning which influences the various principles of interpretation reflects changes in understanding amongst modern English speakers about what can be inferred from formal language. The comparative decline of the exclusio alterius principle may, perhaps, be an example of this.

It remains to note one final feature of these intrinsic principles. It is that linguistic deductions of the present kind tend to lack what might be called an overtly ideological content. In a vacuum, one's personal beliefs about how society should be organised are unlikely to throw very much light on whether, at a high level of generality, particular words should exclude general ones. Of course, at the coal face where cases are actually decided, one may well find that one's enthusiasm for one principle of statutory interpretation over another is influenced in a particular case by one's worldview. A good example, controversial perhaps, is afforded by the High Court's decision in Kirmani v Captain Cook Cruises Pty Ltd (No. 2) (1985) 159 CLR 461. The antecedent case (No.1) had involved an interesting question of whether a particular Commonwealth statute had validly repealed part of an earlier Imperial statute. The unsuccessful party then sought leave to appeal to the Privy Council from the High Court. Such an application was expressly provided for by s 74 of the Constitution:

'No appeal shall be permitted to the Queen in Council from a decision of the High Court upon any question, howsoever arising, as to the limits inter se of the Constitutional powers of the Commonwealth and those of any State or States, or as to the limits inter se of the Constitutional powers of any two or more States, unless the High Court shall certify that the question is one which ought to be determined by her Majesty in Council.'

The words appearing after the word 'unless' might have been thought suitable for the application of the principle espoused in Project Blue Sky at 382 [71] that a court construing a provision 'must strive to give meaning to every word of the provision'. Of course, Project Blue Sky is only a statement about statutes whereas as s 74 is part of the Constitution. But given the nature of a written constitution and the rigours, in particular, of the amendment requirements in s 128 it seems even less likely that a court of interpretation could simply ignore words which are not to its taste.

Not so, it seems. In a terse, but unanimous judgment, the Court rejected the application for a certificate (at 465):

'Although the jurisdiction to grant a certificate stands in the Constitution, such limited purpose as it had has long since been spent. The march of events and the legislative changes that have been effected – to say nothing of national sentiment – have made the jurisdiction obsolete. Indeed an exercise of the jurisdiction now would involve this Court in passing the responsibility for final decision to the Privy Council in that class of constitutional case which the Constitution marked out above all others as the class of case which should be reserved for the final decision of this Court – and this at a time when, as a result of the legislative changes already referred to, no other constitutional cases can be taken to the Privy Council.'

The present point is not to suggest that that conclusion is wrong but rather merely to illustrate that intrinsic principles of interpretation having little ideological content may nevertheless be utilised (or, in this case, not utilised) in ways which are connected to larger domains of discourse. The basic point remains sound nevertheless. The intrinsic principles of interpretation which may be inferred as corollaries of the formal nature of statutory language do not have an ideological edge to them, however they might be applied in practice. Nor, for completeness, can they be seen as saying much about legislative intent save that the legislature intended to express itself carefully.

But these intrinsic rules are not the only principles which apply. We are additionally commanded by s 15AA to prefer an interpretation of a statute which is more consistent with the legislature's purpose. Much ink has been spilled on the nature of legislative intent both as a concept and as a tool. You will forgive me if I do not rehearse the issues which are usually thought to arise. My point is that, subject to a significant qualification, s 15AA does not require, and the Courts have not generally approached, the ascertainment of legislative purpose with any preconceptions about what kind of legislator the legislature is thought to be. In fact, all that s 15AA tells us to do is to read the statute consistently with the perceived purpose of the legislation giving us no particular preconceptions about what those purposes might be. In that sense, it presents as politically agnostic. This apparent political indifference finds itself reflected, perhaps embodied, in the High Court's modern position of insisting on the primacy of legislative text. A well-known instance of that posture may be found in Federal Commissioner of Taxation v Consolidated Media Holdings Ltd (2012) 250 CLR 503 at 519 [39]:

'"This Court has stated on many occasions that the task of statutory construction must begin with a consideration of the [statutory] text". So must the task of statutory construction end. The statutory text must be considered in its context. That context includes legislative history and extrinsic materials. Understanding context has utility if, and in so far as, it assists in fixing the meaning of the statutory text. Legislative history and extrinsic materials cannot displace the meaning of the statutory text. Nor is their examination an end in itself.'

(footnotes omitted)

This is apt to suggest that one should not approach the language of a provision with any a priori notions of what it might have been intended to mean; the language alone is to be paramount, the words supreme. This noble vision is certainly consistent with what the High Court has on other occasions said. For example, in Minister for Employment and Workplace Relations v Gribbles Radiology Pty Ltd (2005) 222 CLR 194 the Court said this (at [21]):

'Section 149 must be read in a way that gives effect, so far as possible, to its legislative purpose. Plainly, the purpose of the section and its predecessor succession provisions is, and always has been, to extend the operation of awards beyond those who were parties to the dispute that the award determined. But identifying that purpose does not answer the question that arises in this matter – how far does the extension go? It is only if some a priori assumption is made about the intended reach of the provision that considering its purpose casts light on the question. To reason in that way begs the question. Rather, it is necessary to consider the words of the provision. It is there that the intended reach of the legislation is to be discerned.'

Here then, I think, there is apparent consistency and coherence. The intrinsic principles of statutory interpretation are, so viewed, a blank slate without political content and this, itself, is reflected in the purposive approach to interpretation which tells one to focus only on the purposes of the legislature. Under both views, the judiciary leaves its political luggage at the interpretive door. I will come back to this judicial luggage shortly to make the point that political luggage, unlike real luggage, can be surprisingly difficult to lose.

What then about constitutional principles?

Constitutional Principles

This is a large sounding heading but may I, at once, trim it down? In the context of a discussion about statutory construction it is trite to say that Courts will endeavour to interpret statutory material consistently with the requirements of the Constitution. Section 15A of the Acts Interpretation Act requires as much as well as authorising reading down to facilitate it. No doubt, even without s 15A it would be necessary to read statutes subject to the Constitution. I do not think that this primacy of the Constitution over statute law tells one very much about the coherence of statutory interpretation. In a sense, it is a domain of discourse which lies outside, or at least beyond, statutory interpretation.

Rather more useful in the present context are a set of interpretative principles applied by Courts which might be called 'small-c' constitutional principles. By 'small-c', what I intend is that they are principles relating to, or touching upon, systemic elements of government or, to put it another way, that relate to the organisation, or literally, the constitution of government. Many of these will be familiar to you. They include:

• the presumption in Potter v Minahan (1908) 7 CLR 277 that legislation is not to be interpreted as altering common law doctrines without expressing that intention with 'irresistible clearness';
 
• the presumption that legislation is not to be interpreted as having an extra-territorial effect: Jambunna Coal Mine NL v Victorian Coal Mines Association (1908) 6 CLR 309 at 363;
 
• the presumption that legislation does not interfere with vested proprietary interests: Clissold v Perry (1904) 1 CLR 363 at 373; and
 
• the presumption that legislation is not to curtail certain human rights and freedoms (including personal liberty) unless an intention to do so is clearly manifested in unambiguous language – the principle of legality : see, e.g., Al-Kateb v Godwin (2004) 219 CLR 562 at [19] per Gleeson CJ.

There are many others. They are usefully collected in Chapter 5 of Pearce and Geddes (supra). Now an interesting question to ask is from where do these derive? I have tried to argue above that the intrinsic rules of interpretation are, in effect, colourless implications flowing from the formal nature of the language used and the very context of a deliberative legislature pronouncing rules of law. That other extrinsic tool of interpretation – legislative purpose –is likewise said to be focussed on the purposes of the legislature and does not permit the taking into account of a priori notions of what a statute might mean, leastwise judicial a priori notions.

Until the recent efforts of those behind the principle of legality to identify its legitimacy as deriving from a shared dialogue between legislature and courts, there had not been very much of an attempt to explain why these various constitutional assumptions were made. Looking at the traditional ones I identify above it is possible, I think, to discern an assumed set of values accepted, I think its authors would have thought, by everyone at the time the presumptions were first fashioned.

The presumption that legislation is not to be interpreted as abolishing common law doctrines without expressing itself with irresistible clarity is, so it seems to me, a direct invitation not to read legislation in accordance with its ordinary language. In such a context, ordinary words will not do; only the very clear will suffice. But in the interpretation of ordinary statutory language we are content to contend with ambiguity. Indeed, we have a complete armoury of tools – the intrinsic aids to interpretation – which allow us to make sense of the ambiguous.

The effect of this presumption (and the others to which I have referred) is to make unavailable all of the intrinsic aids for resolving ambiguity. Where the abolition of a common law doctrine is concerned, any ambiguity means that the common law doctrine is not affected even if the ordinary intrinsic rules of interpretation would have resulted in the opposite outcome.

This seems, arguably, to involve the a priori assumption that common law doctrines are something which should be preserved, if possible, from the blandishments of a legislature. What value does this represent? And whose value is it? A cynic might answer this by observing that the judiciary's tenderness to protect the common law reflects a certain chauvinism for judge made law. The presumption has its origins well before the twentieth century so we should, perhaps, not be too judgmental. For modern eyes, it is easy to identify historic common law doctrines which seem shocking and not in any way worthy of protection from a legislature. For example, it is the common law which gave us both the principle that a wife could not be raped by her husband^[36] and trial by combat.^[37] Why do we need words of unmistakeable clarity to get rid of these achievements of the common law?

Other presumptions may be seen to rest on similarly shaky foundations. The presumption that legislation does not have an extra-territorial effect is a close cousin of the principle that legislation is to be construed on the assumption that it does not breach principles of international law. Both assumptions proceed on a premise that, in some ways, it is desirable for municipal law to operate harmoniously with international law, whether customary or public. But modern events cannot leave one with any confidence that this is a shared value at all. Australia often legislates extra-territorially. The examples are not obscure or difficult to find. Extra-territorial operation lies at the heart of the extended application provisions in s 5 of the Competition and Consumer Act 2010 (Cth). The laws prohibiting whaling inside Australia's exclusive economic zone but outside its territorial waters are another example.^[38] Further, Australia is not shy about assertions in the international sphere which are, or may be, legally dubious. One example may be Australia's claim to large swathes of the Antarctic which is contested by many States.^[39]

I suppose it might be said that the Commonwealth is, of course, free to break international law if it wants but the judiciary proposes to proceed on the basis that it does not so intend. However, that is likely only to involve a restatement of the problem. Even so expressed, the presumption contains an a priori notion that the international legal order is something worthwhile. Whilst that is an idea which is attractive to many including myself, it is difficult to avoid the conclusion that it is an opinion lying outside any given legislative text.

A similar point may be made about each of the other items on the list set out above and I hope I will be forgiven for passing over them. It would, however, be remiss, and incomplete, not to mention the principle of legality. With the two assumptions I have just dealt with it might be accurate, although perhaps a little glib, to say that they reflect a certain lawyerly view of how the world ought be arranged. But that world view is not explicitly articulated anywhere. By contrast, in the case of the modern acceptance of the principle of legality there is, for the first time, an explicit statement of the values involved. In a well-known passage, Gleeson CJ explained in Electrolux Home Products v Australia Workers Union (2004) 221 CLR 309 at 329 [21]:

'The joint judgment in Coco went on to identify as the rationale for the presumption against modification or abrogation of fundamental rights an assumption that it is highly improbable that Parliament would "overthrow fundamental principles, infringe rights, or depart from the general system of law" without expressing its intention with "irresistible clearness". In R v Home Secretary; Ex parte Pierson, Lord Steyn described the presumption as an aspect of the principle of legality which governs the relations between Parliament, the executive and the courts. The presumption is not merely a common sense guide to what a Parliament in a liberal democracy is likely to have intended; it is a working hypothesis, the existence of which is known both to Parliament and the courts, upon which statutory language will be interpreted. The hypothesis is an aspect of the rule of law.'

(Footnotes omitted)

This brings out into the open what appears to be a set of a priori values which are to influence how the intention of Parliament is discerned from the text it uses. In short, they are the values of a modern liberal democracy. The wielding of such a set of values by the judiciary would, in a vacuum, be undemocratic for it is not the role of the Courts to impose the values of the members of the bench on legislation. This democratic deficit is apparently, however, overcome in the case of the principle of legality because the values are not those of the Courts alone but are instead a 'working hypothesis the existence of which is known both to Parliament and the courts'.

While I have no particular difficulty endorsing those values myself, I must say I am sceptical of the claim that they constitute a working hypothesis known to both branches of government. If that claim meant only that Parliament (by its Parliamentary Counsel) is aware of the manner in which Courts will interpret legislation and it is that knowledge which constitutes the working hypothesis then it would, whilst probably true, also be fairly trivial. Certainly, I am reasonably sure that the proponents of the principle did not only have in mind the bland concept that Parliament knows how the Courts are going to interpret its legislation.

It is much more likely that the principle includes not only the idea that the values thus encapsulated are good or worthwhile but also the notion that they are shared by both the Parliament and the Courts. It is, so it seems, a hypothesis upon which both branches are working together. If that were not so, the difficulty of the democratic deficit implicit in judges imposing an a priori set of values onto legislation would be acute. It is only this conception of shared values which can avoid that difficulty or (on the trivial interpretation) save the principle from a meaning which would otherwise be trite.

I doubt whether the principle of legality thus understood is empirically sound. It would be nice to believe that Parliament shares a commitment to the values which inhere in a modern liberal democracy, but in practice this does not appear to be the case. It seems to me therefore arguably inaccurate to ascribe to Parliament a shared set of liberal democratic values. As the current turmoil in the Western world shows, liberal democratic values are by no means immutable or even shared.

If that be so, then the principle of legality hangs by a thin conceptual thread. It looks like a set of a priori notions being applied to legislation by the judicial branch. That is to say, it appears to be an example of an approach to legislative interpretation which otherwise the authorities tell us is to be avoided at all costs. When we find ourselves in an area over which the principle of legality holds sway, the words on the page are not to be read by reference to their usual meaning. They are instead to be read as if they did not intend to interfere with a posited set of values unless the language is unavoidably plain. If one rejects the shared values premise upon which the principle rests – and there are good factual reasons to think it dubious – then this seems to me to open up the law of statutory interpretation to a charge that the principle of legality reduces its coherence particularly on the role of a priori notions. One might make useful comparison with this irregular verb: 'I am applying a shared liberal democratic value', 'You are resorting to an a priori matter', 'She is a judicial activist'. As I explained at the outset, however, whether that is a bad thing is altogether a different question well beyond the scope of this paper.