Earlier in September 2017, the Singapore High Court (HC) ruled in Sun Electric v Sunseap Group that the HC does not have original jurisdiction to hear patent revocation proceedings or to grant orders for revocation (click here for commentary of the HC decision), even by way of a counterclaim in patent infringement proceedings. This meant that a defendant in infringement proceedings cannot counterclaim for revocation of the asserted patent, but instead must initiate revocation proceedings separately before the Registrar at the Intellectual Property Office of Singapore (IPOS). The HC decision was appealed and the Singapore Court of Appeal (CA) issued its judgement on 10 January 2019 partially reversing the HC decision. The CA ruled that there are two distinct categories of cases which must be dealt with separately in deciding whether the HC has original jurisdiction for patent revocation.
The first category relates to applications for revocation which are brought by way of counterclaim in infringement proceedings. In this first category, the CA reversed the HC decision and ruled that the HC does have original jurisdiction to revoke a patent at least when such a relief is pleaded in infringement proceedings. The CA accepts that Section 82(1)(a) of the Singapore Patents Act (SGPA) by itself does not confer original jurisdiction on the HC to determine the validity of a patent. However, the HC acquires the jurisdiction to determine the validity of the patent by virtue of infringement proceedings commenced under Section 67(1) SGPA. The CA noted that the exhaustive grounds to invalid a patent are the same as those to revoke a patent as listed in Section 80(1) SGPA. Thus, it follows that if an alleged infringer is successful in establishing invalidity, it would also succeed in revoking a patent. The HC would have the power to revoke the patent by virtue of Section 91(1) and Section 80 SGPA. Essentially, the CA disagreed with the HC and clarified that it is not necessary for the HC to have ‘in rem jurisdiction to revoke a patent. Only ‘in personam’ jurisdiction is necessary to which the HC would acquire once infringement proceedings is commenced against a defendant.
The second category relates to applications for revocation which are entirely pre-emptive in nature and brought independently of infringement proceedings. In this second category, the CA held that the HC does not have original jurisdiction to hear applications for revocation because its jurisdiction has been excluded by Sections 82(1) and 82(2) SGPA. Pursuant to Sections 82(1)(d) and 82(2) SGPA, exclusive jurisdiction to determine the validity of the patent in this second category of cases is reserved to the Registrar of IPOS. The CA felt that this was sensible since at the time the SGPA entered into force in 1995, the patent system operated on ‘self-assessment’, and exclusively reserving original jurisdiction to the Registrar was intended to alleviate concerns that the HC might be inundated with multiple proceedings for the revocation of patents of dubious quality.
In arriving at the above decisions, in particular on the first category, the CA also clarified that since a defendant may only put validity in issue by way of defence in infringement proceedings, it follows that only the validity of asserted claims (in respect of which allegations of infringement are made) may be put in issue. In other words, the validity of unasserted claims cannot be challenged in the defendant’s defence and counterclaim. The CA appreciates that if the patentee does not assert all the claims, this begs the question as to how the defendant may obtain relief in the form of revocation.
This led to the CA setting out different scenarios in which a defendant may plead to revoke a patent. The first scenario is one where the validity of the entire patent is challenged and this seems straight forward as if the HC finds in the defendant’s favour that the entire patent is invalid, the HC should exercise its power to revoke the patent.
The second scenario is one where all the independent claims of the patent are asserted by the patentee to be infringed, and the defendant challenges their validity. If the defendant is successful, the CA said that the HC should revoke the patent.
The third scenario is one where some, but not all, of the independent claims of the patent are asserted by the patentee and challenged by the defendant. As the defendant cannot challenge the validity of the unasserted claims, this means that there would still be at least one independent claim which has not been impugned However, the CA clarified that the defendant would still be entitled to a declaration of invalidity in relation to those asserted independent claims and have legal recourse to challenge the validity of the remaining independent claim(s) in revocation proceedings before the Registrar.
As the CA puts it, the appeal involves an important point of law regarding revocation of patents, and who has the original jurisdiction to revoke a patent in what circumstances, and the judgment does indeed help to clarify this aspect.
However, the CA’s comments that a patent should be revoked if all the independent claims were found to be invalid are indeed puzzling. This scenario is elaborated at [70] of the CA judgement appended in part below.
[70] If the HC finds in the defendant’s favour that the independent claims are invalid, it follows that the dependent claims must also fall. This is because, as the nomenclature suggests, dependent claims (or “subsidiary claims”) refer back to the independent claim and incorporate all its features (see Lee Tat Cheng v Maka GPS Technologies Pte Ltd [2017] 3 SLR 1334 at [104]). This would be so even if the defendant did not challenge specifically the validity of the dependent claims in its defence. Thus, in practical terms, once the defendant succeeds in establishing that all the independent claims in a patent are invalid, the dependent claims must necessarily fall away and the patent as a whole must be regarded as invalid. In such circumstances, it would also be proper for the HC to exercise its power to order that the patent be revoked.
It is submitted that the CA’s comments that once an independent claim is invalidated, its dependent claims are automatically invalidated simply by virtue of dependency cannot be right, and indeed contradict the well-established patent practice and examination guidelines that each claim (including dependent claims) stand on its own. If the CA’s comments are correct, there would hardly be any use of Section 70 SGPA which relates to relief for infringement for a partially valid patent.
It is also interesting that the CA referred to the ‘Lee Tat Cheng v Maka GPS Technologies Pte Ltd‘ case to arrive at its comments, as that case seems to suggest otherwise. Referring to paragraph [104] of that case:
[104] I accept that it may well be correct for a court, in assessing the validity of the first or what may loosely be called the "independent claim”, to obtain guidance from the "dependent claims” or "subsidiary claims” which refer back to the independent claim and incorporates all its features, as an aid to interpreting the proper scope of the independent claim. But it does not follow that the novelty of subsequent claims must ipso facto stand or fall on the fate of the independent claim. Much will depend on the scope of the invention as set out in the subsequent claims. The label "dependent claim” or "subsidiary claim” should not distract the court from the enquiry which it is tasked to undertake, which is whether the elements or features in the subsequent claim(s) taken together with the invention as set out in the preceding claim meet the requirements of novelty.
The judge in the Lee Tat Cheng case clearly understood that the validity of the dependent claims is not dependent on the independent claims, and the dependent claims stand on their own even if the independent claims fall.
Additionally, it can also be inferred, especially from the third scenario, that there appears to be an advantage to the patentee if he does not assert all the independent claims in infringement proceedings. Doing so would avoid the risk of the HC revoking the patent since the defendant cannot challenge the unasserted independent claims and the defendant would be placed with the responsibility to initiate revocation proceedings before the Registrar.
While there is now more clarity under what circumstances a defendant may revoke a patent, it will be interesting to see if the courts (and certainly the IPOS in subsequent proceedings) follow the CA’s legal principle that a patent should be revoked if all the independent claims are found to be invalid, regardless of the validity of the dependent claims.
