In the words of the noted American philosopher Yogi Berra, the recent Federal Court decision of Aristocrat Technologies Australia Pty Limited v Commissioner of Patents (No 3) [2024] FCA 212 (Aristocrat No 3) feels like déjà vu all over again.
This decision is the latest instalment in the long-running contest between the Commissioner of Patents (i.e. the Australian Patent Office) and the gaming company Aristocrat Technologies Australia Pty Ltd (Aristocrat) concerning the patentability of electronic gaming machines (EGMs) which are said to fall broadly within the class of computer-implemented inventions.
Litigation History
Very briefly, this saga kicked off with rejections by the Commissioner on patentable subject matter grounds of a number of Aristocrat EGM patent applications. Aristocrat successfully appealed these rejections to the Federal Court (Aristocrat Technologies Australia Pty Limited v Commissioner of Patents [2020] FCA 778 (Aristocrat FCA) but this decision was then subsequently reversed on appeal by the Commissioner to the Full Federal Court (Commissioner of Patents v Aristocrat Technologies Australia Pty Ltd [2021] FCAFC 202 (Aristocrat FCAFC)).
As this matter concerned a number of rejected patent applications, it was agreed early on between the parties that a representative claim be selected from the initially-rejected applications on the basis that if this representative claim was found to be directed to patentable subject matter, then this finding would apply to all the remaining patent claims being rejected.
The parties also agreed that if this representative claim was ultimately found not to be directed to patentable subject matter, then Aristocrat could go to back to the Court for an assessment of a selection of the remaining claims of the rejected application (referred to as the “residual claims”) to determine if these might meet the requirements for patentable subject matter. If Aristocrat failed with respect to the residual claims, then they accepted that all claims of the rejected EGM patents would fail for lack of patentable subject matter.
Following the Full Court’s unfavourable decision, Aristocrat sought and was granted special leave to appeal to the High Court. Unfortunately, the makeup of the High Court hearing the Aristocrat appeal was comprised of only six justices, who then split evenly in Aristocrat Technologies Australia Pty Ltd v Commissioner of Patents [2022] HCA 29 (Aristocrat HCA), with three justices issuing joint reasons dismissing the appeal (dismissing reasons), and three justices issuing joint reasons allowing the appeal (allowing reasons). Notably, there was a clear conceptual divide between the two sets of reasons as to the correct approach for assessing the patentability of computer-implemented inventions.
In these circumstances where the High Court is evenly split, Section 23(2)(a) of the Judiciary Act 1903 (Cth) applies. This section essentially states that where there is a “tie”, then the decision of the Court below will be affirmed. In this case the Full Court had found the representative claim (i.e. claim 1 of Australian Innovation Patent No 2016101967) was not directed to patentable subject matter, as a result resolving this dispute, insofar as it concerned the representative claim, in favour of the Commissioner.
The Full Court’s Decision – Aristocrat FCAFC
At this stage it is worth examining the Full Court’s decision in some detail. As above, the Full Court reversed the decision at first instance, with all three judges finding for the Commissioner and concluding that the representative claim under consideration was not directed to patentable subject matter. Further, the Full Court ordered that the matter be remitted below for consideration of the residual claims (among other issues). Two sets of reasons were provided by the Full Court, with Middleton and Perram JJ joining for a first set of reasons (majority decision) and Nicholas J providing separate reasons (minority decision).
In the majority decision, their Honours rejected the approach taken at first instance and proposed an alternative two-step analysis for assessing computer-implemented inventions. This two-step analysis involved asking two questions, the first being is the invention claimed a computer implemented invention; and if so, can the invention claimed broadly be described as an advance in computer technology (Aristocrat FCAFC at [74])?
Problems with the Two-Step Analysis
It would be fair to say that the two-step analysis approach proposed in the majority decision has been the subject of some criticism. First, Nicholas J, while still allowing the appeal, did not agree with the approach of majority decision but instead adopted a more conventional analysis requiring that for an abstract idea or game to become patentable then the idea must be given “practical effect and transformed into a new product or process which solves a technical problem, or makes some other technical contribution in the field of the invention” (Aristocrat FCAFC at [124]).
In respect of the High Court, the allowing reasons warned against an approach to assessing patentable subject matter that involved a question of whether the subject matter is “computer-implemented” (Aristocrat HCA at [152]) and as Burley J notes in Aristocrat No 3 at [100] this could be understood as an indirect criticism of this two-step analysis. The dismissing reasons also criticised the two-step analysis as unnecessarily complicating the analysis of the actual critical issue, which is the “characterisation of the invention by reference to the terms of the specification having regard to the claim and in light of the common general knowledge” (see Aristocrat HCA at [77] and also Burley J’s review in Aristocrat No 3 at [86]).
Outside of the judicial world, the two-step analysis was also the subject of considerable concern and criticism as the requirement for an invention to satisfy the first step of the analysis, i.e. is the invention claimed a computer-implemented invention, would arguably be satisfied by claims directed to many modern electronic-based technologies involving a data processing aspect. Following the second step of the two-step analysis, many of these electronic-based technologies would arguably fail to have the requisite “advance” in “computer technology” to satisfy the second-step requirement given that the technical application of the invention may be solving a technical problem outside of the computer, and be unrelated to advancing computer technology.
There are significant concerns that if the two-step analysis was adopted, then it would follow that many more computer-implemented inventions would be rejected as not being directed to patentable subject matter. These issues (among others) with the two-step analysis were raised before the High Court by the Institute of Patent and Trade Mark Attorneys of Australia and FICPI who both appeared as amicus curiae in the High Court Appeal.
Back to the Residual Claims
Returning to the Aristocrat No 3 decision, this is the result of the Federal Court now going back to consider the patentability of the residual claims. Burley J was the original trial judge who presided over the first Federal Court proceeding and it now fell to his Honour to determine whether the residual claims were directed to patentable subject matter.
Determining the correct approach to be taken when assessing whether a claim to a computer-implemented invention was directed to patentable subject matter required deliberation, especially in light of the High Court’s divided approach to this question. His Honour first concluded that the High Court’s decision did not create any binding precedent, because a binding authority can only be established where there is a unanimous or majority decision (Aristocrat No 3 at [105]).
This was not the case here, and the effect of the affirmation of the Full Federal Court’s decision was that Burley J, being a single judge of the Federal Court, was bound by the orders of the Full Court and their reasons (Aristocrat No 3 at [110]) regardless of whether he was able to draw some applicable principle of law from the High Court’s reasoning (Aristocrat No 3 at [112]). Burley J was not persuaded that there was any principle that could be drawn from the High Court reasoning (Aristocrat No 3 at [112]) finding that while both sets of Judges were critical of the Full Court’s approach, they were critical for different reasons.
His Honour then proceeded to apply the two-step analysis to the residual claims under consideration. Unsurprisingly, given the previous findings in the majority decision of the Full Court that the representative claim involving no advance in computing technology, the residual claims being directed to other related aspects of an EGM, were also found to involve no advance in computing technology.
Conclusion
Despite the almost universal lack of endorsement of the two-step analysis approach to assessing patentable subject matter proposed in the Full Court’s majority decision, it has once again been applied in Aristocrat No 3. This is perhaps a unique set of circumstances, where the first instance judge has felt compelled to adopt this analysis because of the lack of binding authority from the High Court’s split decision and the resulting affirmation of the Full Court’s decision including the reasoning of the majority decision.
It seems unlikely that the two-step analysis requiring an “advance in computer technology” will find favour more generally outside of the special circumstances of this litigation given there are alternative approaches that have been endorsed by the Full Court to assessing the patentability of computer-implemented inventions and that the High Court was critical of the approach. However, it remains a risk.
Aristocrat will more than likely appeal to the Full Court but again there is the question of whether within the confines of this particular litigation, a differently constituted Full Court or even the High Court will be able to substitute another test for patentability for the two-step analysis, even though this outcome would seem perverse.
