Could the High Court employ EU/UK/Canadian structured proportionality analysis recently embraced in McCloy v NSW to achieve a viable constitutional resolution of the dilemma posed by the need to protect secret national security information in anti-terrorism matters while at the same time affording a fair hearing to terrorism suspects?
UNSW’s George Williams argues that “now is the worst time for making new laws”, given shock, anger and consequent possibilities of overreaction which could undermine “the very freedoms we are seeking to preserve from terrorism”.
Obviously Williams was referring to last Friday’s dreadful terrorist attacks in Paris, and to the citizenship-stripping bill currently being considered by Parliament. However he might just as well have been talking about the Counter-Terrorism Legislation Amendment Bill (No. 1) 2015 (Cth) which was also introduced into Parliament last week by Attorney-General George Brandis. It was clearly triggered at least in part by the murder of NSW police accountant Curtis Cheng by a 15 year old radicalised Muslim fanatic at Parramatta just seven weeks ago. As Tamara Tulich and Jessie Blackbourn explain in an excellent article at The Conversation:
This bill includes a host of new measures designed to address the evolving threat posed by terrorism. These include:
- a new offence of advocacy of genocide;
- amendments to the control order regime, so it applies to persons 14 years and older, and new measures to monitor controlees; and
- clarification of the basis for issuing a preventative detention order.
But the bill’s most concerning aspect is the proposal to expand the secrecy provisions available to courts in control order proceedings.
Keeping national security information secret in court
Since 2004, legislation has been in place to deal with information that is likely to prejudice national security in federal court proceedings.
This legislation created a special closed hearing procedure to determine whether national security information could be disclosed in court and, if so, in what form. This process regulates disclosure between the parties – that is, who gets to see what. …
The bill’s effect is to allow secret evidence into control order proceedings.
“Secret evidence” is that which is not disclosed to an affected party and their legal representative. It is not new.
A successful claim of public interest immunity, for example, results in secret material being excluded from the evidence presented in court. What is new in the anti-terror context is legislation that allows the courts to rely on secret evidence in control order proceedings.
The Criminal Code already allows the Australian Federal Police (AFP) to exclude sensitive national security information at each stage of the control order process. This bill expressly provides that when deciding whether to impose a control order, a judge can rely on evidence that has not been disclosed to the controlee or their legal representative or been challenged – for example, through cross examination.
The government considers that “the inherent capacity of the court to act fairly and impartially” will ensure fairness in the proceedings. However, in the UK, the simple involvement of a judge in closed material proceedings has been deemed insufficient to guarantee a fair hearing.
At state level the secrecy horse has bolted
Unfortunately a State Act with almost identical secret evidence provisions to the Counter-Terrorism Legislation Amendment Bill (No. 1) 2015 (Cth) was considered by the High Court only a few years ago and held constitutional unanimously. A number of States and Territories have enacted anti-bikie gang laws in recent years. Some aspects of a couple of them have been struck down under the so-called Kable doctrine. In Kable v Director of Public Prosecutions (NSW) the High Court held that a State (or Territory) Parliament could not constitutionally confer functions on a state court which undermine its institutional integrity, impartiality and independence, given that state courts may exercise Chapter III judicial power.
Queensland’s anti-bikie law is the Criminal Organisation Act 2009 (Qld). It goes significantly further than most other States’ models (although the NSW law is similar) by providing that a criminal intelligence application is to take place ex parte and without notice to the organisation or person to whom it relates, and in a special closed hearing from which both the subject of the application and his/her lawyer are required to be excluded. In other words, there is no capacity in the Supreme Court to ensure any meaningful compliance with the rules of procedural fairness. Although the Act creates an office of public interest/criminal intelligence monitor to provide some degree of independent oversight of police conduct, it is clear that the monitor’s role is not to represent the accused organisation or individual.
The Queensland Act was considered by the High Court in Assistant Commissioner Condon v Pompano Pty Ltd. Despite the fact that the mandated procedure seemingly did not permit the Supreme Court to require any meaningful compliance with the rules of natural justice, and that those rules have repeatedly been held by the High Court previously to be core aspects of judicial power, the Court in Pompano unanimously held that the Queensland anti-bikie legislation was constitutionally valid.
Although the Kable doctrine had been previously deployed by the High Court to strike down aspects of other States’ anti-bikie laws in Totani and Wainohu, in Pompano the Court eschewed reliance on Kable and instead followed its earlier reasoning in Gypsy Jokers and K-Generation and developed its own judicial “jurisprudence of secrecy”. The Court unanimously concluded that the Queensland Act’s mandating of complete secrecy of criminal intelligence information did not impair the essential and defining characteristics of the Supreme Court of Queensland so as to transgress the limitations on state legislative power derived from Chapter III of the Australian Constitution. Their Honours resolved the tension between reliance on criminal intelligence and fair trial procedures in favour of secrecy, so long as courts retain discretion to independently assess classified information. It has also been said that as a result of these cases, in criminal intelligence and national security matters the High Court has abolished the doctrine of procedural fairness and replaced it by a newly-fashioned notion of “curial fairness”. The common reasoning linking Gypsy Jokers, K-Generation and now Pompano is that:
evidence that formerly would not have been available to the affected party, pursuant to public interest immunity, on which basis it was not utilised by the court, may now still not be available to the affected party but can be used by the court.
This peculiar notion of curial fairness as a substitute for natural justice is most clearly explained by French CJ in Pompano:
Understandably the respondents placed great emphasis on the provision for special closed hearings and the secrecy attaching to criminal intelligence tendered to the Supreme Court. Those provisions undoubtedly represent incursions upon the open court principle and procedural fairness. The Supreme Court, however, retains its decisional independence and the powers necessary to mitigate the extent of the unfairness to the respondent in the circumstances of the particular case. It retains the responsibility to determine what weight, if any, to give to criminal intelligence and, in particular, hearsay evidence relating to information provided by informants. The power of the Supreme Court to control its own proceedings in order to avoid unfairness also suggests that it would have a discretion to refuse to act upon criminal intelligence where to do so would give rise to a degree of unfairness in the circumstances of the particular case which could not have been contemplated at the time that the criminal intelligence declaration was made.
Surely this is empty sophistry. How can a judge meaningfully mitigate unfairness when he/she can have no idea what the suspect would say about any given allegation or what evidence they might be able to adduce to prove it false if he had known what the allegation actually was? Tulich and Blackbourn quote UK Supreme Court justice Lord Kerr in relation to secret evidence:
The central fallacy of the argument, however, lies in the unspoken assumption that, because the judge sees everything, he is bound to be in a better position to reach a fair result. That assumption is misplaced. To be truly valuable, evidence must be capable of withstanding challenge. I go further. Evidence which has been insulated from challenge may positively mislead.
Brennan J said something similar in Pochi and Minister for Immigration and Ethnic Affairs, an early case before the Administrative Appeals Tribunal:
[I]f an applicant is not given a full opportunity to deal with confidential information adverse to his interests, the probative force of the information must be particularly cogent if that information is to be acted upon. There are notorious risks in failing to hear an opposing view – slender proofs may falsely seem irrefragable, and the scales of justice may falsely seem to be tipped by the weight of insubstantial factors.
Brennan J ultimately excluded Mr Pochi (a suspected drug trafficker) from hearing the secret evidence (by a police undercover agent) but permitted his barrister and solicitor to hear and challenge it. Having done so he concluded that the evidence should be given little weight. However, under Attorney-General Brandis’s Bill neither the suspect/controllee nor his lawyers will be permitted to be present and the Court will have no discretion in that regard.
In light of the High Court’s unanimous pro-secret evidence decision in Pompano, a pessimist might well conclude that any challenge to the validity of the almost identical provisions in the the Counter-Terrorism Legislation Amendment Bill (No. 1) 2015 (Cth) would have little chance of success. But there is at least one constitutional argument that is still worth a shot given that this is a Commonwealth law where constitutionally mandated separation of powers applies.
Security-cleared special advocates
In Pompano French CJ considered at some length the development of jurisprudence in the EU, UK and Canada relating to secret evidence provisions in national security and related matters. All these jurisdictions of course have some form of human rights charter constraining or influencing decision-making. All of them give some weight to the existence or otherwise of a suspect’s right to security-cleared special advocates who are given access to the secret evidence and can challenge it on the suspect/client’s behalf. One of the Canadian decisions Charkaoui even notes that “special advocates constitute one example of an approach that is a more proportionate response to reconciling the need to keep some information secret and the need to ensure as much fairness and adversarial challenge as possible.”
French CJ goes on to record the UK Supreme Court’s resolution of this issue:
However, in Tariq v Home Office the Supreme Court of the United Kingdom held that a closed material procedure with special advocates, used in employment tribunals to deal with evidence involving national security, was in principle compatible with Art 6 of the European Convention. Moreover, it was not necessary, under Art 6, for a person to be provided with sufficient information about the allegations against him or her to enable the person to give effective instructions to a legal representative or to a special advocate. A distinction was drawn between cases involving the liberty of the subject and those which do not. In the former case:
“If the special advocate is unable to perform his function in any useful way unless the detainee is provided with sufficient information about the allegations to enable him to give effective instructions to the special advocate, then there must be disclosure to the detainee of the gist of that information”.
As Tulich and Blackbourn note, the Council of Australian Governments’ (COAG) counter-terrorism legislation review similarly proposed that a requirement for security-cleared special advocates should be made part of the anti-terrorism control/preventative detention order regime, and that there should be a guaranteed minimum level of disclosure to the subject/suspect , namely that “the applicant must be given sufficient information about the allegations against him or her to enable effective instructions to be given in relation to those allegations.” Neither recommendation has been adopted in Brandis’s Bill.
Proportionality – a reasonable long shot argument
Nevertheless, despite the High Court’s discouraging decision in Pompano, there remains scope for the Court effectively to mandate security-cleared special advocates as a minimum standard for valid judicial proceedings at federal level where a suspect’s liberty is at stake (as is the case with both anti-terrorism preventative detention and control orders). That scope emerges from the Court’s very recent decision in McCloy v New South Wales in which a clear majority (French CJ, Kiefel, Bell and Keane JJ) decisively adopted a localised version of EU/UK/Canadian structured proportionality analysis as an elaboration of the test laid down in Lange v ABC of whether a law which burdens freedom of political communication can be regarded as “appropriate and adapted” to a legitimate purpose within power. McCloy built on a series of earlier cases which had progressively adopted elements of proportionality analysis.
Anne Twomey summarises the newly-minted Australian version of proportionality analysis:
The first question remains whether the law effectively burdens the freedom in its terms, operation or effect. The second question is described as one that involves ‘compatibility testing’. It requires the identification of the purpose of the law and the means adopted to achieve that purpose and asks whether they are compatible with the constitutionally prescribed system of representative government in the sense that ‘they do not adversely impinge’ upon it. This clarifies previous uncertainty as to what amounts to a ‘legitimate end’ under the Lange test.
The third question is described as requiring ‘proportionality testing’. It is broken up into three parts. It asks whether the restriction imposed by the law on the freedom is justified as
(b) necessary; and
(c) adequate in its balance.
A law is ‘suitable’ if it has a rational connection to its purported purpose. It is ‘necessary’ if there is ‘no obvious and compelling alternative, reasonably practicable means of achieving the same purpose which has a less restrictive effect on the freedom’. It is ‘adequate in its balance’ if the court makes the value judgment that the importance of the purpose served by the law outweighs the extent of the restriction that it imposes on the freedom.
Given that a requirement for security-cleared special advocates is discussed at length in Pompano and is regarded by European, British and Canadian courts as an important protection where a suspect’s liberty is endangered by secret evidence provisions, there is a respectable argument that Australia’s High Court should also regard such a requirement as a minimum standard. It is an “obvious and compelling alternative, reasonably practicable means of achieving the same purpose” (providing necessary protection of sensitive national security information) in a manner which has a less restrictive effect on a suspect’s right to natural justice than complete exclusion of the suspect and his legal representatives from access to secret national security evidence.
Of course, unlike the EU, UK and Canada, the Australian Constitution does not entrench natural justice or due process, although it forms a part of most formulations of rule of law, which in turn is said via A.V. Dicey to be at the very least an underpinning constitutional value, and has always been regarded as a core aspect of the constitutionalised notion of judicial power, at least until Pompano. Perhaps structured proportionality analysis can provide a path whereby the High Court can resolve a crucial values conundrum and apply an approach which protects both vital national security interests and the rights of Australian residents to a fair trial where liberty is at stake, while avoiding trampling impermissibly on evaluative functions that are rightly the province of Parliament and the Executive.
Structured proportionality analysis has to date only been applied to the implied freedom of political communication, but it is apparent from the majority’s reasoning that it is at the very least equally applicable to other constitutional rights and freedoms.
Adrienne Stone has recently raised the possibility that proportionality analysis might well be extended to the task of evaluating the constitutional limits of judicial power:
Moreover, since then a majority of the High Court has explicitly adopted proportionality analysis in a different context.
The position is moreover in tension with established practice of Australian courts. As I have argued elsewhere, the task of implementing the Australian Constitution inevitably requires the same kinds of judgments as are involved in proportionality analysis: judgments as to the meaning of the many morally contested ideas that the Constitution adopts. The need for this kind of reasoning is especially obvious where judges have developed unwritten structural principles that resemble constitutional rights. In Australian constitutional law these include a right of freedom of political communication and a ‘rule of law’ principle. Equally, it is required by s 92 of the Constitution, which guarantees freedom of trade among the states.
But even if we are to dismiss these aspects of Australian constitutional law as either exceptional or themselves instances of impermissible disregard for the limits of the judicial power (as Jeffrey Goldsworthy views the freedom of political communication), the judicial role under the Australian Constitution is necessarily highly evaluative. … Put simply, it is the High Court that has shaped Australian federalism, the separation of powers doctrine, the fundamental institutions of government and while it is far from unconstrained in its role, it nonetheless has inevitably faced choices of considerable moment. It has to decide what is expedient, which outcomes are to be preferred and in doing so it chooses between contested conceptions of constitutional ideas. …
This is not to say that adoption of proportionality is an insignificant step. Proportionality analysis does bring these matters of evaluation to the fore. The Court now explicitly acknowledges the balancing element of these inquiries that (as explained here) had previously been implicit in its use of the standard ‘reasonably appropriate and adapted’.
We do indeed live in interesting (and sometimes frightening) times.
 Qld Act s 66; NSW Act s 28I.
 Qld Act ss 70, 78; NSW Act s 28K.
 Greg Martin, ‘Jurisprudence of Secrecy: Wainohu and Beyond’ (2012) 14(2) Flinders Law Journal 189, 189.
 Andrew Lynch, Tamara Tulich and Rebecca Welsh, ‘Secrecy and Control Orders: The Role and Vulnerability of Constitutional Values in the United Kingdom and Australia’ in David Cole, Federico Fabbrini and Arianna Vedaschi (eds), Secrecy, National Security and the Vindication of Constitutional Law (Edward Elgar Publishing, 2013) 156, 168.
 Steven Churches, ‘Paradise Lost: But the Station is Always There’ (2010) 12(1) Flinders Law Journal 1, 20 (emphasis in original).
 (2013) 252 CLR 38, .
 (2013) 252 CLR 38,  – .
  1 SCR 350 at 398  citing Roach, “Ten Ways to Improve Canadian Anti-Terrorism Law”, (2006) 51 Criminal Law Quarterly 102 at 120.
 At least the associated implied freedom of association (to the extent necessary to underpin freedom of political communication); freedom of interstate trade (section 92); freedom of religion (section 116); and freedom from discrimination based on State of residence (section 117).