Aristocrat's second chance on remittal quashed by procedural bind 8 min read
In recent years, Australian courts have had several occasions to consider the requirements for a computer-implemented invention to be patentable under Australian patent law. In 2022, the High Court of Australia grappled with this question in the Aristocrat v Commissioner of Patents case. However, the bench was split evenly. This meant the earlier decision of the Full Court of the Federal Court in the same case was affirmed by a technicality. The lack of a binding ratio from the High Court left prospective patentees, lawyers and patent attorneys with uncertainty as to the correct approach to the patentability of computer-implemented inventions.
The affirmed Full Court decision made a remittal order for the first instance judge, Justice Burley, to decide residual issues in light of the reasoning in the Full Court decision. Although the Full Court's reasoning was critiqued by both camps of High Court judges, Justice Burley decided that no binding authority results from the High Court decision and instead he was bound by the reasoning of the majority in the Full Court decision in relation to the residual issues.
In this Insight, we unpack Justice Burley's reasoning for the approach taken and consider where this latest judgment (Aristocrat Technologies Australia Pty Limited v Commissioner of Patents (No 3) [2024] FCA 212) leaves us on this topical issue.
Key takeaways
- Justice Burley has applied the Full Court decision in Aristocrat. It appears the 'two-stage' test set out in the majority reasoning of the Full Court decision remains binding law in Australia for now.
- We expect the existing practice in the Australian Patent Office in relation to computer-implemented inventions to remain largely unchanged.
- Among the few things that all members of the High Court bench agreed on was a critique of the 'two-stage' test. It therefore remains to be seen whether the High Court or the Full Federal Court will clarify the law on patentability of computer-implemented inventions in a future case.
- This may present an opportunity for legislative intervention on the question of patentable subject matter.
The story so far
In Justice Burley's own words, 'the procedural history is tortuous'. The case concerns the patentability of Aristocrat's claims in innovation patents relating to an electronic gaming machine (EGM) functioning in a certain way, and the central issue that went up to the High Court is whether the claimed innovation falls within the meaning of a 'manner of manufacture' in section 6 of the English Statute of Monopolies of 1623. Although the Statute of Monopolies historically formed the basis of patent law in many jurisdictions that adopted English law, Australia (under the Patents Act 1990 (Cth)) and New Zealand are the only major jurisdictions that still define whether particular subject matter is patentable by reference to this section of the Statute of Monopolies.
'the procedural history is tortuous'
In 2016, Aristocrat applied for innovation patents relating to an EGM in which certain sequences of symbols in the base game can be programmed to trigger a 'feature game'. The Australian Patent Office determined that the claims were new and innovative, but not a manner of manufacture.
On appeal to the Federal Court of Australia, Justice Burley held that the claims were a manner of manufacture and not a mere scheme. His Honour reviewed the authorities and adopted a two-step process:
- Was the claim for a mere business scheme?
- If so, did the invention lie in the manner in which it had been implemented in the computer?
A mere business scheme is not a manner of manufacture, and if the answer to the second question is 'no', then the computer-implemented invention is not a manner of manufacture.
On further appeal to the Full Court of the Federal Court, Justice Burley's decision was reversed. The majority (Justices Middleton and Perram) adopted a new two-stage test:
- Is the invention claimed a computer-implemented invention?
- If so, can the invention broadly be described as an advance in computer technology? If not, then the invention is not patentable subject matter.
On further appeal to the High Court, the bench was evenly split. Three of the judges (Chief Justice Kiefel and Justices Gageler and Keane) (the senior plurality) agreed with the Full Court decision that the claims were not a manner of manufacture, but they disagreed with the Full Court's reasoning. Three other judges (Justices Gordon, Edelman and Steward) (the junior plurality) disagreed with the Full Court decision entirely, holding that the relevant claims met the low threshold of manner of manufacture. We have previously reported on the reasoning of each of the two decisions more fully here,
Section 23(2)(a) of the Judiciary Act 1903 (Cth) provides that, where the High Court is equally divided in relation to a Federal Court decision, the decision appealed from 'shall be affirmed'. Therefore, by a technicality, the Full Court decision was affirmed. As a result, Justice Burley was ordered to determine the residual issues 'in light of the Full Court's reasons'.
The Full Court and High Court had focused on one claim (claim 1) of one of Aristocrat's applications, as the parties agreed that if this claim was a manner of manufacture, then the same conclusion would follow for various other claims. By the parties' agreement, on remittal, Justice Burley considered a different common claim (claim 5) of three of Aristocrat's applications. The parties agreed that, if none of these claims is found to be patentable subject matter, then none of the claims of each of the applications is patentable.
What is the binding law?
Justice Burley first addressed whether either set of reasoning from the High Court, or the Full Court's reasoning, should be applied. One major point of uncertainty arising from the High Court's evenly split decision was whether the two-stage test of the Full Court decision was still the correct approach given that each of the High Court pluralities critiqued this approach. Aristocrat in particular argued that the senior plurality's reasoning should be adopted.
After considering section 23(2)(a) of the Judiciary Act and the remittal order, his Honour concluded that the majority reasoning of the Full Court should be applied.
Justice Burley held that there was no binding principle arising from the High Court's critique of the two-stage test. The two-stage test of the Full Court decision is therefore affirmed.
The Judiciary Act was clear that the Full Court decision 'shall be affirmed'. The evenly split High Court decision did not create any binding precedent because only unanimous or majority decisions of the High Court are binding. Although case law does refer to 'special circumstances' where authoritative principles may be established by a decision of an equally divided Court, his Honour did not consider the current circumstance to fall within that scenario. The remittal order was also unequivocal—the residual issues were to be determined 'in light of the Full Court's reasons'.
Further, Justice Burley held that, even if any relevant principle of law was considered to emerge from the High Court decision, his Honour was 'not permitted the latitude that might apply to other courts to discern from the High Court any relevant binding principle'. His Honour's choice of words here, and the emphasis placed on the remittal order, suggests the possibility that there may be room for another court, which is not subject to the remittal order, to distinguish from the approach taken by Justice Burley in this case.
In considering the correct test to apply, Justice Burley noted that the senior and junior pluralities differed as to the basis for their critique of the two-stage test adopted by the majority in the Full Court decision. While both pluralities criticised the two-stage test's focus on whether the claimed invention is an advance in computer technology, the senior plurality preferred to characterise the invention by reference to the terms of the specification having regard to the claim and in light of the common general knowledge, whilst the junior plurality's rejection was more fundamental and their Honours preferred the question of whether the computer-implemented invention creates 'an artificial state of affairs and a useful result'. Justice Burley held that there was no binding principle arising from the High Court's critique of the two-stage test. The two-stage test of the Full Court decision is therefore affirmed.
His Honour also noted that both sets of reasons in the High Court approved the previous Full Court and High Court authorities. His Honour did not consider whether the Full Court decision was inconsistent with existing authority, noting that expressing any such view 'would be to ignore the judicial hierarchy in which this matter is to be decided'.
Outcome
In applying the Full Court's reasoning to determining the patentability of claim 5, Justice Burley did not expressly apply the two-stage test of the Full Court decision to each integer of claim 5. Instead, his Honour started from the outcome reached by the Full Court that claim 1 was not a manner of manufacture and assessed whether the additional integers in claim 5 of each application represented an advancement to computer technology. His Honour swiftly concludes that that no additional features or limitations in claim 5 warrant a different outcome to that of the Full Court decision in respect of claim 1. As such, each of the claims of the patents was concluded to not be a manner of manufacture.
Implications of the decision
There remains some residual uncertainty arising from Justice Burley's decision. With the even split of the High Court bench, the much-anticipated decision of Justice Burley was expected to clarify which opinion should be adopted. However, the conflicting comments in the series of decisions in Aristocrat leaves the future direction of patenting computer-implemented inventions unclear.
As it stands, there is a constant struggle between patent attorneys and examiners at IP Australia applying differing manner of manufacture tests to computer-implemented inventions.
Currently, examination for manner of manufacture at the Patent Office involves balancing and considering a variety of principles set forth by earlier Full Court decisions (Research Affiliates LLC v Commissioner of Patents, Encompass Corporation Pty Ltd v InfoTrack Pty Ltd, Commissioner of Patents v Rokt Pte Ltd and Commissioner of Patents v RPL Central Pty Ltd). These include, for example, whether the invention solves a technical problem, results in an advancement to computer technology or provides a practical and useful result. Satisfying one of these principles alone is often insufficient to confer patentability, making it more difficult in Australia (compared to other major patenting jurisdictions) for patentees to obtain patent protection for computer-implemented inventions.
As it stands, there is a constant struggle between patent attorneys and examiners at IP Australia applying differing manner of manufacture tests to computer-implemented inventions. This unstructured approach often results in a stalemate between patent attorneys and examiners as there is no clear and binding precedent or legislative guidance to follow.
Accordingly, further clarity regarding the test for manner of manufacture and reform of IP Australia's examination practice is needed to provide greater clarity and certainty for the future of computer-implemented inventions, particularly with the rapid rise of AI technology, which often involves innovations that are characterised as computer-implemented inventions.
The emphasis on procedural requirements in Justice Burley's reasoning means it remains to be seen whether other courts might attempt to depart from this approach. In any case, the issue as to patentability of computer-implemented inventions generally is now likely to be primed for a further appeal to the High Court. If no clear judicial guidance is delivered, a timely legislative intervention may be warranted.