SDCV v Director-General of Security: ‘Closed evidence' and ‘practical justice'
David Hume
28.10.2022
In SDCV v Director-General of Security [2022] HCA 22, the High Court held, 4:3, that it was constitutionally-permissible for the Federal Court to have regard to 'closed' information, which was known to the Court and the Government’s lawyers, and was the subject of 'closed' submissions involving the judges and the Government’s lawyers but was not disclosed to SDCV or his lawyers.
The key issue in the case was whether the statutory scheme established by the Administrative Appeals Tribunal Act 1975 (Cth) – which permitted the Federal Court to rely on closed information, while prohibiting it from being disclosed to SDCV – required or permitted the Federal Court to act in a procedurally unfair way.
The majority was comprised of a joint judgment of Kiefel CJ, Keane J and Gleeson J, with Steward J writing separately. The majority emphasised that procedural fairness requires no more than the avoidance of 'practical injustice'. The key point made by the majority was that SDCV had made individual choices which exposed him to the very scheme he now asserted was procedurally unfair. That, the majority said, meant that there was no practical injustice.
The minority emphasised two points: first, the inherent procedural unfairness which arises when courts rely on secret evidence; and, second, the fact that this scheme took the choice of what was procedurally fair out of the hands of the Court and instead imposed procedures by legislative fiat.
In my view, there are tensions between the approach taken by the majority and broader Chapter III and constitutional right and freedom jurisprudence. The majority applied a form of reasoning more appropriate for individual rights when, in other cases, the Court has repeatedly made the point that most constitutional guarantees are in the nature of systemic freedoms.
In this post, I first address the factual and procedural background (which are interesting in and of themselves); secondly, I address some key aspects of the majority’s reasoning; and, thirdly, I make some observations on the reasoning and outcome.
The factual and procedural background
SDCV was a Lebanese citizen, who was married to an Australian citizen. In 2012, Australia granted him a partner visa. Thereafter, he applied for Australian citizenship, with his application subsequently approved.
In order to become an Australian citizen, individuals must attend a citizenship ceremony and give the pledge of commitment. However, before SDCV could attend his citizenship ceremony, the Australian Security Intelligence Organisation (ASIO) commenced an investigation into him. Ultimately, ASIO formed the view that SDCV had some male relatives connected with a terrorist organisation, ISIL, as well as a relative who had been convicted of a terrorism offence in Australia.
SDCV was then told that his citizenship ceremony was to be delayed, because consideration was being given to the cancellation of his visa as a result of ASIO's investigation. In August 2018, ASIO concluded that SDCV was a risk to security, and issued a certificate to that effect. ASIO also recommended that his visa be cancelled.
This was the genesis of the issue which ultimately went to the High Court. That is, while ASIO told SDCV some of the reasons for the conclusion that he was a risk to security, they omitted key parts of the information, effectively on the basis that it was against the public interest and prejudicial to national security to disclose it.
What SDCV was told was that ASIO considered that he supported terrorism and a terrorist organisation, and that he had used tradecraft (such as a covert phone) when communicating with people who were of security concern. However, it is clear that this was only part of the story, and there was in fact quite a bit more information that was important and was relied on, but was not given to SDCV. In effect, SDCV had the conclusions, but not the detail and evidence on which those conclusions were based.
We don’t know exactly what the information relied on by ASIO was, what its sources were and the probativeness it had. What we do know is that SDCV thought that ASIO’s conclusions were not justified, albeit he couldn’t put his finger on exactly why because he didn’t know the information that had actually been relied on.
All of this information – both open and closed – was then provided to the then Minister for Immigration (the Minister), who decided to cancel SDCV’s visa under s 501(3) of the Migration Act 1958 (Cth) on the basis that he was not of good character and that cancellation was in the national interest. Like ASIO, the Minister for Immigration relied on information which was kept secret from SDCV.
SDCV appealed the Minister’s decision to the Administrative Appeals Tribunal (AAT), which affirmed the Minister’s decision to cancel SDCV's visa, again by reference to secret information.
Pausing here, nothing to this point gave rise to any constitutional complaint as to denial of procedural fairness. All steps to this point involved the exercise of executive power, and there is no general constitutional imperative for the executive to afford procedural fairness. While statutory (and possibly non-statutory) powers will generally, as a matter of strong presumption, be subject to procedural fairness, that is always subject to contrary intention. The steps to this point could possibly have given rise to constitutional complaint on the basis that the executive was given the power to make decisions which, by reason of the secrecy of the underlying information, were practically immune from review. But that complaint – if made, which it was not – would have had its own difficulties.
The appeal to the Federal Court
Following the AAT's affirmation of the Minister's decision, SDCV appealed to the Federal Court, invoking the right of appeal on questions of law given by the Administrative Appeals Tribunal Act 1975 (Cth).
This is ultimately where the constitutional issue which was agitated in the High Court arose. This is because, while the executive does not have a general constitutional duty to afford procedural fairness, Chapter III courts (including the Federal Court) do have such a duty. That duty is at least an incident of the proposition that Chapter III courts must be and remain 'courts' within the meaning of the Constitution, and it is an essential element of a 'court' that it afford procedural fairness.
Procedural fairness ordinarily requires a fair hearing which, in turn, requires that those affected by a decision know the material on which the decision-maker relies. However, in this case, the way in which the statutory scheme worked was that the secret material was delivered to the Federal Court, and the Federal Court was empowered to look at that material, but could not disclose the material to SDCV or his legal representatives.
One of SDCV’s grounds of appeal to the Federal Court was that there was no evidence to support the key conclusions reached by the Minister and the AAT. He candidly accepted that he could not make meaningful submissions in support of this ground – because he did not know what evidence actually existed, that having been kept secret. This left the Federal Court in an interesting position. It had before it an argument that the AAT’s decision was wrong because it was unsupported by evidence, but it had no litigant before it who was actually willing and able to make submissions in support of that argument.
The Federal Court ultimately opted to review the secret information for itself, before conducting a closed hearing, attended only by the Commonwealth's representatives, to discuss some of the information. During the closed hearing, the Commonwealth was represented but SDCV’s representatives were unable to appear. Having then considered the secret information, the Federal Court decided to reject SDCV’s appeal.
So at this point SDCV now had had three decisions go against him, all the while without knowing the precise basis on which he had been held to be a security risk. The following sections explore SDCV's appeal to the High Court and the Court's decision in this case.
The High Court: submissions and decision
SDCV's argument before the Court started from the uncontroversial proposition that courts are constitutionally required to afford procedural fairness. According to SDCV, the statutory scheme was procedurally unfair because it allowed evidence adverse to him to be considered by the Federal Court, in circumstances where that information was kept secret from him, and he had no opportunity to respond to it. The Commonwealth’s case was that there is no absolute or invariable requirement that courts not act on secret information, and that in this case there was no overall practical injustice.
The High Court, by majority, agreed with the Commonwealth. The Court was closely divided, with four judges in the majority, and three judges (Gageler, Gordon and Edelman JJ) writing separate judgments in the minority. In my analysis, I will focus on two key aspects of the majority’s reasoning.
The majority was of the view that courts are sometimes allowed to rely on secret information when making decisions, and that this was one of those cases. In their Honours’ view, SDCV had made choices, the result of which was that he was exposed to a decision based on secret information. SDCV, having made those choices, suffered no practical injustice when the risk to which he exposed himself eventuated.
The first 'choice' relied on by the majority was SDCV’s decision to invoke a right of appeal to the Federal Court on a question of law (rather than, for example, seeking judicial review under s 39B of the Judiciary Act 1903 (Cth)). This right of appeal came with a potential benefit for SDCV, namely that the Federal Court could look at the secret information, but also a detriment to SDCV, namely, that if the Federal Court did look at this information, its contents could not be disclosed to SDCV. The majority said that SDCV was required to take the forensic burden (non-disclosure) with the forensic benefit (the ability of the Court to look at the information), and that, therefore, he suffered no practical injustice (see paragraph [83]).
The second kind of 'choice' relied on by the majority was that he had elected to seek and obtain the privilege of a visa. Having done so, he exposed himself to the risk that his visa might be cancelled, on the basis of closed information, kept secret from him during executive process and, subsequently, in Court. By way of example, the majority said (at [12]) that ‘it is necessary to appreciate the limited statutory rights of the appellant to enter and remain in Australia’ and ‘[t]he rights of a visa holder were always qualified by the statutory process of the executive government to deny the visa holder disclosure of security-sensitive grounds for the making of an [adverse security assessment]’: see also at [22], [69], [71]-[74]. There was, again, no practical injustice in the information being kept secret. As the majority said at [74]: ‘[t]he appellant suffered no practical injustice … He suffered no loss of rights by reason of being denied full disclosure because his right to hold a visa unless he was the subject of a valid ASA was circumscribed by the requirement that he not be informed of security-sensitive information in relation to decision’. The majority’s view, in short, was that the constitutional guarantee of natural justice was congenitally infirm in its application to those who were in Australia by way of conditional privilege.
SDCV and its tension with broader constitutional guarantee jurisprudence
In my view, the majority’s reasoning is out of step with conventional constitutional jurisprudence.
With limited exceptions, conventional constitutional doctrine is that constitutional guarantees are systemic freedoms, not individual rights. This has been repeatedly emphasised by the High Court in its implied freedom jurisprudence, as well as in the context of s 92 and Ch III of the Constitution, amongst other areas.
The observation that constitutional guarantees are systemic freedoms has important consequences. Normally, it tends against a conclusion that a law is inconsistent with a constitutional guarantee. For example, it tends towards an analysis which focuses on the system-wide effect of a law on a protected freedom, rather than an analysis which focuses on the effect of a law on a particular individual.
However, there are also some instances where it tends in favour of a conclusion that a law is inconsistent with the constitutional guarantee. One example of this is where it is said that choices made by the individual challenging the law render valid that which would or might otherwise be invalid, such as was the case for SDCV. So, the High Court has said of s 80 (trial by jury) that litigants cannot waive trial by jury due to the Trial by jury serves systemic functions (eg community participation in the justice system) the broader public interest. It is not for an individual litigant to waive a 'right' which, in truth, is a right of the public.
That courts are, and must remain, 'courts' is a matter of public concern; it is not a matter left to the hands of individual litigants. A Ch III court must be, and remain, a 'court' so as to ensure that it remains a suitable repository for the exercise of federal jurisdiction. This ensures consistency of federal justice, and maintains public confidence in the judicial system. Where a court does something in one case which departs from the basic and essential features of a court, that is no less a departure from constitutional requirements, and no less of constitutional concern, because the litigants in that case sanctioned the departure.
The difficulty with the majority’s reasoning, in my view, is this: the gist of the majority’s reasoning is that SDCV, by making particular choices, lowered the standard of federal justice to which he might otherwise have been entitled. This is out of step with the basic principle that individuals cannot waive the systemic guarantee that the Federal Court will be and remain a court.
Further, the majority’s reasoning that because SDCV was an alien and therefore only in Australia by way of conditional privilege, is, in my view, particularly discordant with basic principle. That entry by aliens is a privilege does not have the consequence that Parliament has a plenary legislative power to affect their interests (on the theory that all entry is conditional, and Parliament can attach whatever conditions it wishes). Aliens, like citizens, are indirectly protected by constitutional freedoms. For example, the constitutionally-protected supervisory jurisdiction protects the right of aliens, as much as citizens, to invoke the aid of courts. By way of further, example, the freedom of political communication protects the freedom of aliens, as much as citizens, to engage in political communication.
Although each case will depend on the precise provision in issue, it is difficult to see the High Court holding valid a law permitting a visa to be cancelled on the express basis that the individual visa-holder expressed political views with which the ruling government disagreed. It is equally difficult to see the High Court holding valid a law depriving aliens of standing to seek judicial review of an exercise of Commonwealth or State public power. This difficulty is based on the reason I’ve suggested above: namely, constitutional freedoms are not individual rights which attach to individuals such that the status of the individual (eg as alien or citizen) is constitutionally-significant.
The tension between the majority’s approach and conventional freedom jurisprudence is perhaps best exposed by considering the following questions. So far as the majority rested its decision on the alienage of SDCV, was the law valid in its application to aliens, but invalid in its application to others? So far as the majority rested its decision on the forensic choice made by SDCV to pursue an appeal under the AAT Act, is the law valid in respect of individuals who have made a deliberate forensic 'choice', but invalid in respect of those who cannot be said to have made such a deliberate 'choice’? What if a deliberate forensic choice was in fact made, but it was made because of representations by the Commonwealth as to what information SDCV were given (eg that SDCV would receive some or all of the closed information), being representations which were later reneged on? Would the law be invalid then because there would be practical injustice by reason of the defeat of SDCV’s legitimate procedural expectations?
The point to which these questions are directed is a fundamental one. The observation, well-established in the administrative law context, that procedural fairness requires 'practical justice' is closely allied to the proposition that what is procedurally fair depends on all the circumstances of the case. But, at the constitutional level, at least where one is dealing with a constitutional freedom, the circumstances of the particular case generally do not bear on validity. Conventional doctrine is that a law is either valid or invalid, and that validity turns on systemic and generalised considerations, not the circumstances in which the law has been applied in a particular case. There are difficulties in reconciling 'practical justice' considerations with issues of statutory validity.
It will be interesting to see whether these aspects of the majority’s reasoning become entrenched in the constitutional jurisprudence, and provide a foothold for future developments.
—
David Hume is a Barrister at Six Selborne Wentworth Chambers. He is the co-author of Human Rights under the Australian Constitution, and has appeared in a number of cases concerning constitutional guarantees.
Suggested citation: David Hume, ‘SDCV v Director-General of Security: ‘Closed evidence' and ‘practical justice' on AUSPUBLAW (28 October 2022) <https://auspublaw.org/blog/2022/10/sdcv-v-director-general-of-security-closed-evidence-and-practical-justice/>