Earlier this morning, the High Court of Hong Kong handed down an important judgment on the “balance” between personal liberty and national security. Readers will be aware that for the last few months, there have been mass public protests in Hong Kong. In response, the Hong Kong government passed the Prohibition on Face Covering Regulation [“PFCR”] which, as the name suggests, prohibited protesters in public spaces from wearing face masks to hide their identities. The PFCR was passed under the authority of the Emergency Regulations Ordinance (ERO), a colonial-era law that allowed sweeping powers to the Executive in an “Emergency” or during times of “public danger.”
The PFCR was passed on 4th October. It was promptly challenged (along with the ERO). The High Court heard arguments at the end of October, and handed down its judgment today. The Court struck down the ERO to the extent of its application during times of “public danger”, while leaving open the question of whether it was valid for “Emergencies.” The Court went on to hold that the PFCR was an unconstitutional and disproportionate violation of the freedom of expression of the citizens of Hong Kong.
In this post, I will discuss both holdings. The striking thing about this judgment is that despite conceding a high margin of discretion to the Executive, and despite accepting the Executive justification of maintaining law and order, the Court still found that the indiscriminate and non-targeted nature of the measure, which failed to distinguish between violent protesters and ordinary citizens, was disproportionate. As we shall see, this is by no means the judgment of an activist Court, which placed the claims of personal liberty beyond all question. On the contrary, this was a judgment by a cautious and deferential Court, which still found the ERO and the PFCR to violate Hong Kong’s Basic Law (the Constitution). And at the heart of its judgment, as I shall show, was a very simple logic: Constitutions allow the government to declare states of Emergency, and suspend certain civil rights. If, however, the government has elected not to declare an Emergency, it is not for the Court to presume their exists one. In terms of law and constitutionalism, there is no halfway house between Emergency and normalcy, where – in the absence of an Emergency proclamation – the Court nonetheless adopts a hands-off approach towards civil rights violations. Rather, if there is no Emergency, then the judicial approach towards civil rights violations must be one that applies constitutional principles with their full rigour.
The ERO was a 1922 law, passed by the colonial British regime. Effectively, it authorised the Chief Executive in Council [“CEIC”, or “Executive”] to make “regulations” in times of Emergencies or public danger. These regulations were extremely wide in scope, including powers of censorship, seizure of property, amendment of laws, trial and punishment, and so on.
The High Court struck down the ERO on seven substantive grounds. Under Hong Kong’s basic law, it found that the Legislative Council [“LegCo”] was the primary legislative organ. The CEIC’s powers were limited to accepting or vetoing bills, and passing subordinate legislation. This is, of course, a familiar arrangement in parliamentary democracies. The Court then made the familiar point that “this constitutional scheme does not permit the LegCo to grant and the CEIC (or, for that matter, any other body) to receive and be vested with what is essentially the LegCo’s own constitutional power and function as the legislature of the Hong Kong SAR to enact, amend or repeal laws, except for an authorisation of subordinate legislation.” (paragraph 52)
This, of course, is the “excessive delegation” test known to students of constitutional and administrative law everywhere. Applying this test, the Court found that the “ERO confers general legislative powers on the CEIC.” (paragraph 55) This was because:
… the ERO is not a statute that legislates on a subject matter in principle leaving another body to devise the detailed legal norms that elaborate or put flesh on the broad matters laid down in the primary legislation. The long title of the ERO specifies that its object is to confer on the CEIC power to make regulations on occasions of emergency or public danger. But it gives no shape or direction of what the regulations that may be made are to be about. For example, the PFCR was enacted under the ERO not to work out and fill in the details for certain broad norms established by primary legislation, but as the very first piece of legislation in Hong Kong that has anything to do about face covering. This is fundamentally different from one’s ordinary conception of subordinate legislation. (paragraph 56)
Next, the Court found that the scope of the power delegated to the CEIC was extremely broad – to make “any regulations whatsoever” that it considered to be in the public interest. Thirdly, the powers could be invoked “on any occasion” when the CEIC was satisfied that there existed an emergency or public danger – neither of which were defined in the statute. In other words, such wide power was accorded to the Executive, that it was virtually unconfined – effectively (as the Court noted) it could never be argued that the Executive was going beyond the authority conferred by the Legislature, as the authority itself had no boundaries. Not only that, but the ERO actually authorised the CEIC to amend existing legislation – i.e., it conferred – in so many terms – legislative power upon the executive (the Court’s fourth point). Furthermore, the powers of punishment conferred upon the Executive went beyond what was authorised in primary legislation (fifth); and there was no time limit upon the “validity and force of the regulations made under the ERO, nor any mechanism for constant review” (paragraph 68) (sixth).The power of “negative vetting” was held not to be a substantial check on the executive (seventh).
Drawing upon insights from comparative law, the High Court therefore concluded that:
“…the ERO, once invoked, seems to us to create in Hong Kong a separate source of laws that are primary legislation in all but name, but which are not made by the legislature in accordance with legal procedures (Art 73(1)) or reported to NPCSC (Art 17), and are not subjected to the scrutiny concomitant with the normal legislative process. Whenever the CEIC considers an occasion falling within the ERO has arisen, the CEIC becomes a legislature.” (paragraph 80)
This was evidently unconstitutional. And in response to the government’s argument that there were times that necessitated “swift and decisive action”, the High Court made the crucial observation that “the need for an urgent response is no justification for departing from or impugning the constitutional scheme.” (paragraph 95)
Let us now come to the prohibition of face coverings in public spaces. The PFCR prohibited the use of “facial covering that is likely to prevent identification” while a person was at an unlawful assembly, an unauthorised assembly, or even an authorised or lawful assembly (see paragraphs 25 – 29 for an explanation of these terms in Hong Kong law). It was common cause between the parties that the PFCR restricted the freedom of speech, assembly, and privacy. It was also common cause that the constitutional validity of the restrictions was to be determined according to the proportionality standard:
“…(1) does the measure pursue a legitimate aim; (2) if so, is it rationally connected with advancing that aim; (3) whether the measure is no more than reasonably necessary for that purpose; and (4) whether a reasonable balance has been struck between the societal benefits promoted and the inroads made into the protected rights, asking in particular whether pursuit of the societal interest results in an unacceptably harsh burden on the individual.”
The government argued that the goal of the prohibition was “(i) deterrence and elimination of the emboldening effect for those who may otherwise, with the advantage of facial covering, break the law, and (ii) facilitation of law enforcement, investigation and prosecution.” (paragraph 130) The question then arose: why was the measure blanket in nature, targeting both potential law-breakers as well as legitimate public protesters? To this, the government argued – on the first count (deterrence) – that “those protesters who are not prepared to break the law may comply with the PFCR and this would generally result in lessening the support for the more radical and violent protesters”; that “masked protesters mix themselves into larger groups and instigate violence and vandalism“; that “non‑radical protesters will be less likely to be influenced by or emulate their violent peers and will think twice before emulating them when they know their identity is not concealed“; that “the PFCR can act as an effective deterrent against at least some students from wearing masks when joining a protest (lawful or unlawful), which thereby substantially reduces the chance that they will be induced to break the law.” (paragraph 133) On the second count (law enforcement), the government argued that protesters were using “black bloc” tactics (i.e., appearing in indistinguishable groups and wearing similar clothing), which made specific and targeted identification difficult. A prohibition on face covering would facilitate the police in being able to identify which of the protesters were acting unlawfully.
The government’s arguments will sound eerily familiar to those following the litigation around the communications lockdown in Kashmir. In both cases, the State’s primary justifications for blanket restrictions is (i) targeting is impossible, and (ii) the bad guys will mingle with and influence the innocent guys, and so we have to restrict everyone’s freedom. The only difference is that the Hong Kong government’s arguments before the Court at least sounded more sophisticated and plausible than the Indian government’s ham-fisted “terrorists use mobile phones” justification.
How did the Court engage with this argument? The first thing to note is that it did not deny that there existed an ongoing law and order situation in Hong Kong. In paragraph 132, it observed that:
“… there is evidence before us of the enormity of the damage and danger created by some of the protesters. In the few months leading to 4 October 2019, Hong Kong has witnessed numerous instances where certain protesters charged police cordon lines with weapons, blocked public roads and tunnels with a variety of large and heavy objects, attacked drivers who voiced complaints at such blockades, vandalised public facilities and buildings, burned public property, hurled inflammable liquid bombs at the police and at and inside Mass Transit Railway stations, damaged shopping malls, shops, banks and restaurants (with reports of looting and theft in some of the damaged shops), damaged residential quarters of the disciplined forces, crippled the operations of transport infrastructure, and harassed and attacked ordinary citizens holding different political views. These acts of violence and vandalism had increased in intensity and frequency, with the incidents on 1 October 2019 being especially serious. The more violent protesters were often all suited up and masked by facial covering such as surgical masks, balaclavas and gas masks which concealed their identity.”
Along with these facts, the Court also conceded that the government had to be given a “wide margin of discretion”, and that the necessity prong within the proportionality standard required only that the government’s action was “reasonably necessary.” Now within this framework, how did the Court apply the proportionality standard? It began by noting that “some participants in demonstrations may wish to wear facial covering for legitimate reasons, such as to avoid retribution.” (paragraph 148) Consequently, the restriction on freedom was not minor or trivial, but a serious one. By contrast:
… the effect of s 3(1)(b), (c) or (d) is to impose a near‑blanket prohibition against the wearing of facial covering by the participants, without any mechanism for a case‑by‑case evaluation or assessment of the risk of any specific gathering developing or turning into a violent one such as would make it desirable or necessary to impose the prohibition in relation to that gathering only. (paragraph 155)
It is not clearly stated whether, to be caught by the prohibition, the person must be a participant in the relevant gathering, or whether it suffices for that person to be merely present at the gathering, eg a person who goes to the scene for the purpose of taking photographs, or giving first-aid to persons in need of help, or even a mere passer‑by who has stopped to observe the gathering. (paragraph 156)
This was, thus, prima facie evidence of over-breadth and disproportionality. What of the government’s argument that it was the only way to prevent violence? The Court noted in response that “the evidence before us is far from clear that the PFCR has achieved to any substantial degree the intended aims of deterrence and elimination of the emboldening effect for those who may otherwise, with the advantage of facial covering, break the law, or facilitation of law enforcement, investigation and prosecution.” (paragraph 164) In other words, under the proportionality standard, the burden was upon the government to justify its rights-infringing measures on the touchstone of necessity and proportionality, with evidence (and not in a sealed cover). And the Court was unimpressed by the government’s exhortations of public danger, noting that “even in these challenging times, and particularly in these challenging times, the court must continue to adhere to and decide cases strictly in accordance with established legal principles.” (paragraph 165) Thus:
… having regard to the reach of the impugned restrictions to perfectly lawful and peaceful public gatherings, the width of the restrictions affecting public gatherings for whatever causes, the lack of clarity as regards the application of the restrictions to persons present at the public gathering other than as participants, the breadth of the prohibition against the use of facial covering of any type and worn for whatever reasons, the absence of any mechanism for a case‑by‑case evaluation or assessment of the risk of violence or crimes such as would justify the application of the restrictions, the lack of robust evidence on the effectiveness of the measure, and lastly the importance that the law attaches to the freedom of expression, freedom of assembly, procession and demonstration, and the right to privacy, we do not consider the restrictions of rights imposed by s 3(1)(b), (c) and (d) to be proportionate to the legitimate aims sought to be achieved by the imposition of those restrictions. (paragraph 166)
Thus, except insofar as it applied to unlawful gatherings, the prohibition was struck down. The Court used similar analysis to strike down Section 5, which empowered “a police officer to stop any person in any public place who is using a facial covering and to require that person to remove it so that his or her identity may be verified, if the officer reasonably believes the facial covering is likely to prevent identification.” The Court held that its indiscriminate character (“no limitations as to circumstances or period”) violated the proportionality standard.
The Hong Kong High Court’s judgment is a shot in the arm for civil rights. It demonstrates that even in the time of the proverbial “clash of arms”, courts can ensure that the laws are not silent. A few salient features of the analysis stand out. First, the High Court took seriously the indiscriminate and blanket nature of the prohibition, which failed to distinguish between criminals, and those who were lawfully exercising their constitutional rights to demonstrate and protest. This was perhaps the most damning feature of the government’s measure. Secondly, the government’s efforts to justify this fell flat. In particular, given that the restriction was blanket and indiscriminate, the onus was on the government to show that there was no other way to achieve the goals of law and order – and to show this with evidence. Unsurprisingly, the government failed, because there was no evidence. And lastly, the Court thoroughly rebuffed the government’s efforts to immunise its actions by making claims about the law and order situation. The Court’s approach to this issue can be summed up in paragraph 108:
- In times of a public emergency officially proclaimed and in accordance with the other requirements of s 5 of the HKBORO, measures may be adopted under the ERO which derogate from the Bill of Rights (even so, excepting the specified non‑derogable provisions and discrimination on the prohibited grounds). Subject to the conditions of s 5 (including that the derogations are limited to those strictly required by the exigencies of the situation), this may have the effect of temporarily suspending the relevant human rights norms.
- In other situations, measures adopted under the ERO may not derogate from the Bill of Rights, which means that if any such measure has the effect of restricting fundamental rights, then like any other restriction in normal times, it has to satisfy the twin requirements that the restriction is prescribed by law and meets the proportionality test.
The important of this observation cannot be understated. What the Hong Kong government was trying to do in this case – and indeed, what the Indian government is trying to do in the Kashmir litigation – is to create a third, midway category of (what is effectively) a “permanent, undeclared Emergency.” For political reasons, the government is wary of formally declaring an Emergency and suspending civil rights. But by repeated invocations of “law and order” and “security”, it is attempting to persuade the Court to act as if the situation is one of Emergency, and thereby, adopt the “judicial hands-off” approach” that it would be compelled to do if there was an Emergency. Paragraph 108 of the Hong Kong High Court’s judgment refutes this disingenuous and dangerous argument: it makes it clear that if there is no Emergency, then the restriction on civil rights must be judicially examined as “in normal times”, and the usual doctrines of reasonableness and proportionality applied. And as we have seen, proportionality takes a particularly dim view of blanket and indiscriminate restrictions (which are in effect, if not in form, Emergency-style suspension of rights – if everyone is prohibited from exercising a certain right, then it is hardly deniable that that right has been suspended for the relevant territory altogether). Such measures, therefore, must almost always be struck down as disproportionate.
It remains to be seen whether the excellent judgment of the Hong Kong High Court will be followed elsewhere, where governments make similar arguments to deny civil rights to their citizens.