When I went and watched part of the hearing that led to R J E v Secretary to the Department of Justice  VSCA 265, Maxwell and Weinberg seemed set to throw the Charter at the government, while Justice Geoffrey Nettle seemed to be sitting on the sidelines and even throwing the government a bone or two. That just goes to show that my lousy predicting ability is quite robust. As previously posted, the chatty duo bizarrely didn’t apply the Charter, while Nettle, seeing more value in comity, felt that only the Charter could justify ditching Callaway’s interpretation of s11 of the SSOMA from just two years ago.
And Nettle promptly applied the Charter to do just that. This is, of course, quite exciting: perhaps only the second time the Charter has actually made a difference to the outcome of a case (albeit, on this occasion, only the difference between a majority and a unanimous new interpretation.) Some will see more excitement in Nettle’s discussion of the interpretation mandate:
I acknowledge that, if TSL, Tillman and Cornwall are regarded as having been correctly decided according to ordinary conceptions of statutory construction, it must also be accepted that Parliament’s intention at the time of enacting s 11 of the Act was that ‘likely’ need not mean more likely than not. To adopt now the construction which I prefer is to accept that the intention has changed. But that appears to be the way in which the Charter was intended to operate.
It’s great to see that he’s clearly read Charter s. 32, even giving a bit of lip service to the purposive limitation. And he’s even read some overseas cases on rights compatible interpretation. Glory be. In particular, he adopted the discussion of the UKHRA interpretation mandate by Lord Woolf in Poplar Housing:
(a) Unless the legislation would otherwise be in breach of the Convention section 3 can be ignored (so courts should always first ascertain whether, absent section 3, there would be any breach of the Convention).
(b) If the court has to rely on section 3 it should limit the extent of the modified meaning to that which is necessary to achieve compatibility.
(c) Section 3 does not entitle the court to legislate (its task is still one of interpretation, but interpretation in accordance with the direction contained in section 3).
(d) The views of the parties and of the Crown as to whether a ‘constructive’ interpretation should be adopted cannot modify the task of the court (if section 3 applies the court is required to adopt the section 3 approach to interpretation)
The last of these principles is the most exciting, giving some hope that the Victorian judiciary will break free of slavishly parroting of the most intelligible of the lawyers before it. The second-last, while question-begging, is a reasonable point (and Nettle later tantalises us with an almost-endorsement of Ghaidan, not that UK-style interpretative high-jinx are needed to read ‘likely’ as ‘probable.’)
The first two points, alas, are the Charter-marginalising ‘ordinary interpretation first, Charter interpretation last’ method. Nettle says that he prefers this to Elias CJ’s dissent in Hansen due to its ‘clarity and simplicity’, not to mention its endorsement by Anthony Mason in his post-constitutional-senility life in Honkers. But: (1) The dispute in Hansen was about the interaction between the interpretation mandate and the reasonable limits provision, which is a combination peculiar to Victoria, the ACT and NZ, and doesn’t arise in the UK or HK. (2) Any chance, Nettle, of doing some freaking interpretation of the Charter, i.e. reading its words and structure, looking into its history, etc? Since when have ‘simplicity’ and ‘clarity’ been the sole test of a major structural question in a statute? Believe it or not, fundamental rights laws tend to raise some other issues too!
Still, all up, this is a rare instance this year of someone treating the Charter as something other than a bunch of soft suggestions. As Nettle says:
I consider that the interpretation of s 11 of the Act which was adopted in TSL is now inconsistent with an offender’s right to move freely within and without Victoria and the offender’s right to privacy, if not his or her right to liberty. It follows from Lord Woolf’s second direction in Poplar that the Charter cannot be ignored. Consistently with his Lordship’s third direction, however, I consider that to construe ‘likely’ in s 11 as meaning ‘at least more likely than not’ is within the permissible ambit of interpretation, well short of the forbidden territory of legislation.
Are you listening, the rest of the Victorian judiciary? Stop ignoring it.
Alas, Nettle’s relatively solid command of the operative provisions was not matched by his analysis of the rights provisions and the central notion of compatibility. His central reasoning appears in this dross:
Evidently, the purpose of s 11 of the Act is to guard against the dire consequences of the commission of a relevant offence. In some circumstances, that might justify significant encroachments on an individuals rights of freedom of movement and privacy and even liberty. But if ‘likely’ in s 11 of the Act is construed as including a less than even chance, it is capable of rendering the requirement for satisfaction to a high degree of probability illusory. For example, one might, well be satisfied to a high degree of probability (say, 80 per cent) that there is a 45 per cent chance of the commission of a relevant offence, and yet, according to the laws of probability, the risk of the commission of the offence as so assessed would be only 36 per cent. That would mean that a relatively low risk of re-offending could provide a sufficient basis for making an order. Even giving full weight to the purpose of s 11, I cannot conceive of the potentially far reaching restrictions on rights provided for in the Act as being capable of demonstrable justification in the relevant sense unless the risk of an offender committing a relevant offence is at least more than even.
Frankly, I just cannot believe that this is all he could come up with after three days of hearing.
The weakness of Nettle’s reasoning is evident in the very numbers he comes up with. Who thinks that there’s no conceivable justification for restricting the liberty of someone who poses a better than one in three chance of raping a child? OK, I can see a couple of hands, but chances are you folks think that’s it’s always wrong to lock up someone once they’ve served their time, right? Nettle’s argument, however, assumes that just about every reasonable person would reject out-of-hand restricting the liberty of a person who has a one-in-three chance of raping a kid, but happily restrict the liberty of someone who has a one-in-two chance of raping a kid. Personally, I doubt that there are many people like that. Indeed, most people would say: there’s very little relevant difference between 36% and 51% when it comes to protecting kids.
Actually, that’s precisely what the UK House of Lords said when it construed the word ‘likely’ in a statutory provision governing when a child could be removed from its parents because of abuse was ‘likely to occur':
In this context Parliament cannot have been using likely in the sense of more likely than not. If the word likely were given this meaning, it would have the effect of leaving outside the scope of care and supervision orders cases where the court is satisfied there is a real possibility of significant harm to the child in the future but that possibility falls short of being more likely than not. Strictly, if this were the correct reading of the Act, a care or supervision order would not be available even in a case where the risk of significant harm is as likely as not. Nothing would suffice short of proof that the child will probably suffer significant harm. The difficulty with this interpretation of section 31(2)(a) is that it would draw the boundary line at an altogether inapposite point. What is in issue is the prospect, or risk, of the child suffering significant harm. When exposed to this risk a child may need protection just as much when the risk is considered to be less than fifty-fifty as when the risk is of a higher order. Conversely, so far as the parents are concerned, there is no particular magic in a threshold test based on a probability of significant harm as distinct from a real possibility. It is otherwise if there is no real possibility. It is eminently understandable that Parliament should provide that where there is no real possibility of significant harm, parental responsibility should remain solely with the parents. That makes sense as a threshold in the interests of the parents and the child in a way that a higher threshold, based on probability, would not.
This quote is very pertinent for several reasons. First, as noted, it involves state action to protect kids from risk, the same issue (albeit not the same action) as is raised by s11. Second, according to NZ’s Court of Appeal, the origins of the word ‘likely’ in NZ’s ESO statute (which seems to be the source of the same word in our statute) is in the above UK judgment. Third, just recently, the House of Lords affirmed that this interpretation remains correct under the UKHRA. I’m just flabbergasted that Nettle doesn’t address any of this. Surely it was discussed at length in the hearing?
So, how did Nettle come up with the magic ‘one in two’ threshold? He doesn’t say, but it seems clear that all he was really doing was interpreting ‘likely’ as ‘narrowly’ as possible, in the sense that he was choosing the interpretation that least restricts RJE’s Charter rights. In other words, he’s just treating the interpretation mandate as if it’s just a pushier form of the common law principle of construing penal/rights-reducing statutes ‘narrowly’. And that’s wrong for three reasons:
First, if Hansen is right, then the ‘narrowness’ approach is wrong because the narrowness that flows from the interpretation mandate is itself restricted by the ‘reasonable limits’ provision. If 40% is reasonable, then that’s all the deviation that Charter s. 32 requires from the ‘ordinary’ (Callaway) interpretation. That being said, Nettle does claim every now and then that anything less than 50% is simply at odds with democracy, etc. But how can he say that? What is magic about 50%?
Second, more importantly, it totally ignores the fact that other rights than RJE’s are at stake. Why won’t someone think of the children? The UK judges sure did, and they didn’t even have Charter s. 17(2) more or less requiring them to act in their best interests:
17(2) Every child has the right, without discrimination, to such protection as is in his or her best interests and is needed by him or her by reason of being a child.
What’s Nettle’s excuse?
Third, and most importantly, the ‘narrowness’ approach grossly oversimplifies the interpretation mandate, which is about (a) interpretation; (b) compatibility; and (c) rights (yes plural.) As I argued in my substantive criminal law paper a few months back, the essence of the interpretation mandate isn’t about ‘narrow’ readings, but about ‘sensible’ readings. The question posed by s11 of the SSOMA, Charter s. 32 and the myriad of Charter rights (RJE’s and children’s) engaged by it is: what’s the best interpretation you can come up with of the term ‘likely’ that strikes the most reasonable balance between them, a reasonableness that ought to be guided by Parliament’s purposes in enacting both the Charter and the SSOMA (just a year apart.)
Now, the above doesn’t mean that I think that I think that ‘likely’ ought to be a lower threshold than ‘more likely than not’. It’s true that there are some sound reasons to lower the threshold when you consider the rights of children. But I don’t think that it’s possible to read all that much into Parliament’s intentions. The EM to the SSOMA doesn’t make any attempt to define ‘likely’, nor does it make any reference to either the NZ or UK precedents. Alas, it seems like the Parliament simply left the hard yards of defining thresholds to the courts, presumably so that it wouldn’t have to bear the political risk if things went pear-shaped. It’s just that sort of cravenly legislating that the Charter exists for, in my opinion. So, I wouldn’t give any margin of appreciation to Parliament on this one.
However, it seems to me that there’s one key consideration that cuts through this morass of competing issues. It’s this one:
21(3) A person must not be deprived of his or her liberty except on grounds, and in accordance with procedures, established by law.
Now, Tate did try to convince the Court of Appeal that nothing in the SSOMA affects anyone’s liberty. (In part, that was based on an appalling take on interpretation, that I’ll get to next post.) The Court of Appeal, alas, didn’t put the kibosh on that, with Maxwell & Weinberg dodging the Charter and Nettle noting the issue (somewhat skeptically, it seemed to me, but others may see it differently) but deciding that even RJE’s movement rights couldn’t be infringed unless there was an even bet that he was going to rape a kid. But what I think is so significant about the liberty right – and, I fear, is why Tate wants it read as narrowly as possible – is that it, more than any other right, is intimately associated, not with substantive justice, but with procedural justice. That’s clearest in the detention-specific rights in Charter s. 21 (like habeas corpus), but it also comes through in Charter s. 21(3). (It’s true that Charter s. 7(2)’s ‘subject under law’ requirement is probably to similar effect, but explicit protections are surely more powerful than vague ones.)
As I argued in my substantive criminal law paper, provisions like Charter s. 21(3) pick up the wealth of public law protections that are standard fodder in comparative human rights law: protections against vagueness, arbitrariness, overbreadth, over-delegation, capriciousness and, above all, lack of accountability. And that’s the issue that, in my mind, points most in favour of a ‘more likely than not’ standard’. While there’s nothing magic in numbers, there is also a lot of danger in trying to set standards that are either qualitiative – such as the UK’s ‘unacceptable risk’ version of ‘likely’ – or too quantitative. On the latter, think how impossibly subjective it would be to rate someone as either more or less than a 33% chance of raping someone. On the other hand, making judgments about whether something is more likely or not is something that everyone, but especially judges, are well familiar with. It’s a standard that is less likely than all the others to produce wildly and widely varying application amongst different judges. And that makes it the appropriate standard for an infringement on liberty like an ESO. (See here for my earlier post about the need for non-subjective standards when it comes to risk assessment.)
So, after all that, does that mean that the Court of Appeal got the result right after all? Well, they got the definition right, I reckon. But the details of the majority’s approach to its application to RJE, cheerily endorsed by Nettle, are quite disturbing. RJE turns out to be not-your-average child sex offender. He was convicted of three sexual offences in the space of what appeared to be a few months in 1997: he had sex with his newly-discovered sixteen year-old daughter, then raped and bashed his domestic partner; and then, after she moved out, raped the fifteen year-old friend of another daughter. The state’s experts both took the view that RJE’s conduct was an inappropriate approach to sex and the particular circumstances of the time (his unhappy relationship), rather than an intrinsically peadophilc tendency. Basically, he’s a nasty version of David Duchovny’s character in Calofornication. The experts stuck their necks out to say that RJE’s later willingness to lie to a new partner by saying he was in jail for defensive manslaughter, writing a dodgy letter to her daughter, and doing a runnner during his ESO hearing, were nothing worrying. But RJE’s County Court judge disagreed.
Now, there’s nothing wrong with the Court of Appeal disagreeing, in turn, with the County Court judge. But here’s their reasoning:
Predicting whether a particular person will commit a criminal offence in the future is notoriously difficult. The Monitoring Act recognises that the prediction of future dangerousness, if it is to be attempted at all, is a matter for expert opinion. As the report in the present case illustrates, the making of such a prediction in a particular case requires expertise in observation and assessment of those who commit offences of the particular type, and a detailed knowledge of the types of factors, both personal and environmental, which increase or reduce the risk of further offending. The necessary expertise combines the ability to make a qualitative assessment of an individual and the ability to utilise the available quantitative risk assessment instruments.
One thing is clear. Judges, including experienced criminal judges, have no such expertise… Judges are, of course, expert in assessing matters of fact, and an expert report ‘is no better than the facts on which it is based.’ But where – as here – the facts are not in contest, and the expert opinion is both cogent and unchallenged, a judge should ordinarily be slow to depart from the risk assessment which the expert has made.
Now, as someone who teaches a course on expert evidence law, where the dangers of deference to expert evidence is so well known, I find this just shocking. It’s true that, before this passage, Maxwell and Weinberg noted that the final decision is the judge’s. But this statement is as clear a message as there can be to lower court judges that what the expert says on ESOs is what should go. And, don’t think that their statement is limited to RJE’s situation, when the experts are opposed to an ESO. Here’s what Maxwell and Weinberg said next:
This point may be simply illustrated. If the undisputed opinion of the expert assessor was that the offender was likely to commit a relevant sex offence if released unsupervised, it is difficult to imagine on what other grounds a judge might fail to be satisfied that the likelihood existed. Conversely, if the assessor’s unchallenged opinion was that the offender was not likely to offend again if released unsupervised, it is again difficult to imagine what material – being material which (ex hypothesi) the assessor must not have considered – could be sufficiently demonstrative of risk to justify the court nevertheless being satisfied that the relevant likelihood existed. If such other material existed, it would seem to us to be far preferable for the expert to be asked whether it affected the risk assessment previously made. In the present case, as will appear, the assessor was asked to comment on particular matters in just this way.
This is a total debacle and one, I’d predict, that the High Court would instantly and angrily reverse if the case got to them. There’s no way that experts should ever be given such power and there’s no way that the language of s11 supports that. (Unfortunately, the drafters of s11 left out the ‘clear and compelling evidence’ standard that the Queenslanders used.) I imagine that many experts will blanch at being given such responsibility (and will inevitably err on the side of caution by recommending ESOs all the time; no-one wants a raped child on their conscience); some experts, more worringly, will relish the power to decide who is supervised and who isn’t. Indeed, the Court of Appeal’s take on the centrality of state experts is so extreme that it may well raise a Kable argument (i.e. that the court is now a rubber stamp for experts, thus damaging their precious role as a potential repository for federal power.)
More importantly, I think that the Court of Appeal’s approach is not only incompatible with Charter s. 21(3), but also with this Charter right:
24(1) A person charged with a criminal offence or a party to a civil proceeding has the right to have the charge or proceeding decided by a competent, independent and impartial court or tribunal after a fair and public hearing.
In short, in their unwillingness (the majority) and inability (Nettle) to properly apply the Charter, the Court of Appeal has replaced Callaway J’s hopeless interpretation of s11 of the SSOMA with an equally hopeless approach of their own. There’s no way it will last (though I suspect that, rather than a High Court appeal, or a five-judge bench, the whole thing will be put out of its misery by new Victorian detention legislation sometime in the new year.)