The Unfinished Sham Tsz Kit Case: Can Remedial Effectiveness Be the Way Forward? [1]
About one year ago, the LegCo voted down the Government’s Registration of Same-sex Partnerships Bill, and the Court of Final Appeal’s decisions in Sham Tsz Kit (No 1) and Sham Tsz Kit (No 2) took effect to declare the government in violation of art.14 of the Hong Kong Bill of Rights (“BOR”) for not providing an alternative framework that recognises same-sex partnerships and confers appropriate rights and obligations thereupon (together, “Same-sex Partnership Framework”). Nothing more has happened on this front since then.
This article tries to facilitate the discussion on the way forward by offering preliminary thoughts on certain matters regarding remedial effectiveness. I argue that it is instrumental to distinguish the breaches to BOR 14(1) and 14(2), and the present situation of remedial ineffectiveness vis-à-vis the breach to BOR 14(1) should trigger the courts’ wide and flexible remedial discretion under s.6(1) of the Hong Kong Bill of Rights Ordinance (“BORO”). I will also offer tentative suggestions on how the discretion should be exercised.
I. What is (are) Really Breached?
It is crucial that
Sham Tsz Kit (No 1) actually disclosed not one but
two breaches of BOR. They corresponded to the two subsections of
BOR 14:
(1) No one shall be subjected to arbitrary or unlawful interference with his privacy, family […].
(2) Everyone has the right to the protection of the law against such interference or attacks.
The first breach accordingly was the existing “state of affairs”
([257]-[258]) — i.e., the difficulties faced by same-sex partners
as such ([128]-[135], [142]-[145]) — which constituted the arbitrary interference with their right to private life. The second breach was the government’s failure to provide legal protection against such interference, i.e., failing the positive obligation to establish a Same-sex Partnership Framework.
With the two breaches separated, the majority and minority opinion in
Sham Tsz Kit (No 1) was reconcilable.
Arguendo the dissenting judges
may be right that the omission to establish a Same-sex Partnership Framework
as such could not constitute an interference under BOR 14(1). But on my reading the majority need not mean that the omission itself was the interference. Rather, the interference triggering BOR 14(1) was the “state of affair”
resulting from the omission. It was on top of this that BOR 14(2) was engaged to positively oblige the government to offer “protection against such interference”, namely a Same-sex Partnership Framework. The government’s failure to do so was the second breach — of BOR 14(2).
Admittedly, at times the majority judges themselves were equivocal about the distinction between the privacy-interfering “state of affair” and the source omission (see e.g., the first sentence of
[144] and the second sentence of
[258]). But it is a doctrinally sound distinction.
It is possible for official omission to result in a situation that amounts to impermissible interference under BOR 14(1) alone. In
Tornel v Spain, a prison failed to inform an inmate’s family of his deteriorating health. The Human Rights Committee (
“HRC”) found that “the passive attitude of the prison deprived the [inmate’s family] of information which undoubtedly had a significant impact on their family life, and which may be characterized as arbitrary interference with the family and as a violation of [ICCPR 17(1), equivalent to BOR 14(1)]” ([7.4]). At the same time there was no finding on ICCPR 17(2) (equivalent to BOR 14(2)).
Further, the distinction is neither artificial nor circular. Theoretically, having a right does not always entail imposing a positive obligation (
Stoyanova 2023, 9), so it is sensible for the former to be dealt with by one provision (BOR 14(1)), and the latter by another (BOR 14(2)). It is true that when BOR 14(1) is engaged, BOR 14(2) is almost invariably engaged as well, but that is just what BOR 14 as a whole has always intended (
Schabas 2019, 464).
But more pertinently, if the interference under BOR 14(1) had been the government’s omission to enact a Same-sex Partnership Framework as such, BOR 14(1) and (2) would fold into one to prescribe such framework as the only effective remedy. In contrast, if the interference had been simply a “state of affairs”, any remedy fixing the status quo may be deemed effective, be it from the government or, indeed, the court.
Viewed in this way, the remedies granted in
Sham Tsz Kit (No 2) — a supervisory jurisdiction and a declaration — may be necessary to redress the BOR 14(2) breach, but they are not necessary for the BOR 14(1) breach. Alternative remedies are possible to redress the latter, especially when the remedies granted turn out to be ineffective, as I shall now argue.
II. Remedial Ineffectiveness
A. ICCPR 2(3) and Remedial Effectiveness in Hong Kong
ICCPR 2(3) provides for the right to effective remedy, consisting of the provision (paragraph (a)), determination (paragraph (b)) and enforcement (paragraph (c)) of effective remedies. While Hong Kong implements ICCPR 2(3), it does not do so by providing for a
right to effective remedy, but by a judicial discretion to grant one. Still, ICCPR 2(3) provides the standard for when and how the discretion should be exercised.
BORO 6(1) empowers a court to grant remedy “as it has power to grant or make […] and as it considers appropriate and just in the circumstances”. Legislative materials state explicitly (LegCo brief,
pp 8-9) that s.6(1) is meant to implement ICCPR 2(3)(a) and (b). By necessary implication then, s.6(1) is also meant to implement paragraph (c): it is inconceivable for a law to provide for effective remedy but does not guarantee its enforcement (cf
Kenedi v Hungary (ECtHR) [47]).
Admittedly, s.6(1) does not create a right to effective remedy, but a judicial discretion. Yet, if implementing ICCPR 2(3) is the purpose of s.6(1), the latter’s interpretation must still be grounded in the former, at least in the following sense: the lack of effective remedy by ICCPR 2(3) standard would frustrate the purpose of s.6(1), which in turn is an overwhelming consideration suggesting that the courts should use s.6(1) to provide for a remedy effective within the meaning of ICCPR 2(3).
Apart from statutory interpretation, it is also for
constitutional reason that the compliance with ICCPR 2(3) weighs heavily in favour of exercising s.6(1) discretion: the courts
must (Ng Ka Ling (No 1) [61]) uphold the ICCPR-based BOR, an obligation to which ensuring effective remedy is integral — so much so that ICCPR 2(3) accepts no reservation (
HRC GC24 [11]).
B. No Effective Remedy
In
Sham Tsz Kit (No 2), the supervisory jurisdiction was a procedural remedy, but it is ineffective for anyone except Mr Sham.
On the other hand, assume that the declaration does require the government to do something. Independent of what is required and whether the government has complied with the declaration, insofar as little has improved for same-sex couples
in effect, the declaration is ineffective
vis-à-vis the BOR 14(1) breach. For a remedy to be effective, it must be effective in practice
(L.P. v Czech (HRC) [7.4]; cf
HRC GC31 [20]; cf
Iatridis v Greece (ECtHR[GC]) [66]) and be “full” (
Shelton 2015, 199-201;
HRC 2016, [1]-[3])
vis-à-vis the violation. These conditions are not satisfied, because the “state of affairs” faced by same-sex couples has barely improved since
Sham Tsz Kit (No 1) was decided.
It does not matter that the government may have been “duly diligent” — this at best counts towards procedural but not substantive effectiveness, which are independent matters. Thus in
Murne v Sweden [10.8], [12], the HRC ordered substantive remedy even when the state’s procedural remedy (the duty to investigate. See
Prutina v Bosnia and Herzegovina (HRC) [9.5]) was not ineffective.
If I am right that there has not been an effective remedy, at least vis-à-vis the BOR 14(1) breach, it is almost certain that the remedial discretion under BORO 6(1) should be exercised. The question is how. Here I can only offer some preliminary thoughts on the relevant factors.
III. BORO 6(1) How?
First, the scope of available remedies under BORO 6(1) is wide. The purpose of s.6(1) is to implement ICCPR 2(3). As the HRC emphasised, that article requires states to “use all appropriate means and organize their legal systems” to guarantee remedial effectiveness, and — most pertinent here — “where the execution of a final judgement becomes impossible […] other legal avenues should be available [for] adequate redress […]” (
Horvath v Australia [8.6]-[8.7]). The legislative materials of BORO 6(1) also say it permits “some flexibility in dealing with breaches of the BOR in order that existing remedies may be used in new ways or developed” to ensure an effective remedy is granted (LegCo brief,
p 9).
Admittedly, BORO includes the term “as it has power to grant”, and the legislative materials state that BORO 6(1) does not create new remedies (Legco brief,
p 9). The way around is to treat “as it has power to grant” not as the ceiling, but the starting point, of what remedies are available under s.6(1), such that the section is read as conferring a “meta-power” for the court to develop new remedies from existing remedies. In other words, in “may grant such remedy or relief […] as it has power to grant”, “may grant” is the meta-power for a court to develop new remedies from what “it has power to grant”.
Such is the effect of
Koon Wing Yee v Insider Dealing Tribunal [110]-[113] where Mason NPJ said s.6(1) “confer[s] a power which will enable the courts to” resolve remedial conundrums. The idea of s.6(1) as conferring an enabling power is consistent with my interpretation of a “meta-power”. I should further note that Mason NPJ in that case did not treat “as it has power to grant” as a limitation on the court’s remedial power.
Second, the exercise of s.6(1) discretion must not unjustifiably interfere with remedial effectiveness. The basis of this claim is jurisdiction rather than the right to effective remedy, for no such right exists in Hong Kong. Under ICCPR 2(3), restriction to remedial effectiveness must be justified (
Sechremelis v Greece (HRC) [10.4]). In turn, given its purpose, s.6(1) does not authorise the court to make orders that unjustifiably interfer with remedial effectiveness.
Third, even a judgment favourable to the victim of right-interference can be rendered an ineffective remedy if the victim had to go through an arduous process to obtain the judgment (
V.D.A. v Argentina (HRC) [9.4]). Potentially, this factor weighs in favour of a court retaining supervisory jurisdiction and even recognising class action in Hong Kong. A prototypical example for both can be found in
Ng Siu Tung v Director of Immigration, whereby the CFA worked with the Court of First Instance to retain jurisdiction for almost 6 years.[4] Nor is
Sham Tsz Kit (No 2) the sole precedent for Hong Kong courts to retain supervisory jurisdiction in LGBTQ+ issues: see the Court of First Instance judgment in
Ng Hon Lam Edgar v Hong Kong Housing Authority [80(3)], predating
Sham Tsz Kit (No 2).
Fourth, at minimum, the court should grant “declarations of legal union”, similar to the “declaration of parent at common law” in
NF v R [161], to committed and stable same-sex couples (
Sham Tsz Kit (No 1) [145]) who come before it.
IV. Postscript
There is no space for me to discuss whether the government has discharged its BOR 14(2) obligation merely by proposing the Registration of Same-sex Partnerships Bill. But I should record my view that it has not. The positive obligation under BOR 14(2) has minimum requirements. It must be an “effective” protection
vis-à-vis the identified impermissible interference of privacy (
L.P. v Czech (HRC) [7.4]). In my view the Bill plainly fell short of that standard.
This only reinforces my view that BORO 6(1) should be triggered, as the delay in implementing the remedy granted can itself be a violation of ICCPR 2(3) (
Czernin v Czech [6.7], [7.5]).
-----------------------------------------------------------------------------------------------------------------------------
[1] I thank Professors Cora Chan and Stefano Osella for their helpful comments on an earlier version of this article. All errors mine.
[2] The standard of “effectiveness” is broadly the same under ICCPR 2(3) and OP 1 Article 5 (rule of exhaustion of domestic remedy before application to HRC admissible):
Schabas 2019, 63-64 & fn 194 at 64. Therefore, case law under either article can supply interpretation for effectiveness.
[3] The cited paragraph concerns the admissibility requirement under ECHR 35(1). However, the standard of effectiveness is the broadly the same under ECHR 35(1) and 13, save for narrow exceptions irrelevant here (
Reiertsen 2022, 145). NB ECHR 13 concerns only procedural remedy:
Mugemangango v Belgium (ECtHR[GC]) [130]-[131].
[4] I thank Dr P Y Lo for mentioning this case to me.
[5] I thank Professor Po Jen Yap for raising this point at the “Court-Legislature “Dialogue”? Sham Tsz Kit and its Aftermath” Panel Discussion on 5 Nov 2025 held by HKU CCPL.
.jpg)
.jpg)
.jpeg)
.jpg)
.jpg)
.jpg)
.jpg)
.jpg)
.jpg)
