Those with an interest in Australian patent law will have already seen many useful summaries of the High Court’s 17 August 2022 judgment in Aristocrat Technologies Australia Pty Ltd v Commissioner of Patents  HCA 29, and likely the read the judgment! Readers will know that there was a 3:3 split, with the result being that the appeal did not succeed in overturning the Full Court’s decision that the claims were not to a manner of manufacture.
This note sets out a few thoughts about the case and what its impact may be. In particular it considers the issue of the patentability of a game, and whether it was really that issue that split the High Court. The reader is assumed to have a general familiarity with the facts of the case.
Is Aristocrat binding on future trial judges and appeal courts?
Where an appeal is dismissed by reason of an even split between judges, there is no majority view of the Court in respect of the issue on which the court split, and so no binding precedent results. This was considered by a recent Full Court of the Federal Court in Harvard Nominees Pty Ltd v Tiller (2020) 282 FCR 530 at , citing earlier High Court authority.
For such an issue, a question would then arise as to whether a future single judge would instead be bound to follow any Full Court precedent, if a relevant ratio was established, including in Aristocrat itself. This is considered further below.
Some areas of agreement
Both pluralities in Aristocrat treated previous Full Court decisions considering computer-implemented inventions as being correctly decided on their own facts (even if some aspects of their reasoning were criticised), including CCOM Pty Ltd v Jiejing Pty Ltd, Grant v Commissioner of Patents, Research Affiliates LLC v Commissioner of Patents, Commissioner of Patents v RPL Central Pty Ltd, Encompass Corporation Pty Ltd v InfoTrack Pty Ltd, and Commissioner of Patents v Rokt Pte Ltd.
The two pluralities agreed that a plan, a scheme, or a game that is merely implemented on a computer does not involve a manner of manufacture. However, if what is claimed involves a contribution to the operation of a computer, that may involve patentable subject matter.
Both pluralities also agreed that it was important to properly characterise the invention that is described and claimed, but they disagreed as to what that appropriate characterisation was for the patent in suit. They agreed that if the proper characterisation of the invention was that it was a mere scheme, plan or game, then there would not be patentable subject matter.
The need to first characterise the invention has arisen in computer-related cases for the following reason. If plans and schemes are not per se patentable, no matter how ingenious, then it does not seem right that a plan or scheme becomes patentable merely because it is implemented using an ordinary computer. Patentees in previous cases have pointed to artificial effects that take place in the computer, including rewriting memory stores and changes in the computer’s display, and in response courts have adopted the approach of requiring that the invention first be ‘characterised’. For cases of mere implementation of a plan or scheme on a computer, characterising the invention as being the scheme itself has allowed such artificial effects to be sidelined and ignored for the purposes of considering manner of manufacture.
More recently, the High Court in D'Arcy v Myriad Genetics Inc emphasised the need to properly characterise an invention before assessing manner of manufacture; although in that case the Court found that the artificial state of affairs resulting from the isolation of a genetic sequence did not give rise to patentable subject matter because the claimed invention, when properly characterised, was the information encoded by that sequence.
The departure between the two pluralities – the perceived patentability of games
The three judges dismissing the appeal in Aristocrat (Kiefel CJ, Gageler and Keane JJ) considered that an electronic gaming machine (EGM) was effectively a computer. They concluded that the claimed invention was:
no more than an unpatentable game operated by a wholly conventional computer, using technology which has not been adapted in any way to accommodate the exigencies of the game or in any other way.
That plurality had previously stated their view that ‘the devising of a scheme or game for entertainment is [not] the invention of a manner of manufacture.’
The Keifel CJ plurality’s characterisation of the invention involved first identifying the differences between the claimed EGM and common-general-knowledge embodiments of EGMs. The plurality concluded that no technical contribution was involved because the elements of the scheme that distinguished the claimed invention from the prior art were elements of a scheme or game.
The three judges upholding the appeal (Gordon, Edelman and Steward JJ) were instead heavily influenced by their assumption that a game may be patentable if it involves the use of physical components that are new. Their Honours cited a single authority; a 1953 decision of Lloyd-Jacobs J, Re Cobianchi’s Application, which held that a claim to a ‘pack of Canasta cards’ having certain specified markings was capable of being a manner of manufacture. While the plurality sought to elevate the status of that decision by noting that Lloyd-Jacob J was ‘the first specialist British patent judge’, the judgment in that case ran to a single page, cited no authority, and involved only limited reasoning. Moreover, it involved a claim to a pack of cards per se, and not a game involving the use of that new pack of cards.
The Gordon J plurality next cited the Commissioner’s acceptance in argument that the game of Monopoly was patentable subject matter being ‘embodied in a physical form such as a designed cardboard board, dice, and playing characters’. The transcript does not disclose clear acceptance by the Commissioner of that proposition, and I am not aware of any clear case law that requires such a conclusion.
It is this assumption as to the patentability of a game that is played using new ‘physical materials’ that appears to underpin the Gordon J plurality’s view as to the proper characterisation of the invention claimed by Aristocrat, leading to their conclusion that the claim was to a manner of manufacture.
The Gordon J plurality raised at the outset of their judgment that in the 21st century, the fact that digital images could be generated by complex software and hardware should not change the analysis of patentability, and later, addressing the issue of games more directly, said:
In the 21st century, it would be absurd if the application of this principle were any different where the idea of the game is combined with a digital representation rather than a cardboard representation of the game.
So just as courts have rejected the suggestion that a scheme or plan can become patentable merely because it is implemented in a computer, the Gordon J plurality reasons that if a game involving new physical materials is patentable subject matter, that conclusion should not change merely because it is implemented in an equivalent manner using a computer. In their concluding paragraph, the Gordon J plurality explained that it was the interdependence of the game controller, which controlled the specific visual elements for the game, and the player interface, which then displayed those elements, that was critical to the invention as claimed, and resulted in a manner of manufacture.
The patentability of games is far from straight-forward, and there is only very limited case law. Dr Mark Summerfield has written about the evolving approach of the Australian Patent Office in examining such applications, which has moved away from acceptance. In my view it is this complexity that lies at the heart of the split in Aristocrat, although there was unfortunately no real analysis of this issue. This is unfortunate because I consider that there is real doubt about whether games, including a game such as Monopoly, involve patentable subject matter.
A note about the counterfactual of a wholly mechanical gaming machine
A counterfactual raised by the Gordon J plurality in support of their conclusion was the proposition that: (1) a mechanical gaming machine (using cogs, reels and motors) must be a manner of manufacture; and (2) the position should not logically be any different merely because the same functionality can now be achieved using an EGM, rather than mechanically.
This logical leap disregards that a computer itself is no less a manner of manufacture than a mechanical gaming machine. It is the fact that a computer can be used to process a series of logical statements, so as to implement a plan or a scheme, that has resulted in courts adopting the approach of analysing manner of manufacture only once the proper characterisation of the invention has been determined.
If, hypothetically, it were routine to vary a mechanical gaming machine to implement a different game, then the analogy with a computer may then apply. The Kiefel CJ plurality addressed this as follows:
It may readily be accepted that on the first occasion a mechanical poker machine was invented, it was patentable. But a later model of the same machine would not have been patentable simply because it allowed a new variant of poker to be played. That difference would not have relevantly differentiated it from the prototype.
The Gordon J plurality appeared to recognise this same logic when they later said that if a computer or ‘any other machine’ merely manipulated an abstract idea without creating an artificial state of affairs (and a useful result) then there would not be patentable subject matter.
The mechanical gaming machine analogy illustrates the difficulties of analogous reasoning. It may be that achieving a new outcome with a mechanical gaming machine requires ingenuity in the mechanical makeup of the machine – that would involve a manner of manufacture. But if instead, in the above hypothetical scenario, such a machine can be readily manipulated in a manner that has an equivalence to the reprogramming of an EGM, then that should give rise to similar issues (and difficulties) in assessing whether there is a manner of manufacture, rather than somehow provide the answer to that issue.
A few words about Philips v Mirabella
Both pluralities spent some time discussing the Court’s 1995 decision in Philips v Mirabella. But it was the Kiefel CJ plurality that appeared to place greater reliance on it, as part of their approach in characterising an invention.
Philips was an odd case, and considered the meaning of the word ‘invention’ in the introductory words of section 18(1). The majority found, in obiter, that there was a ‘threshold’ requirement that was not satisfied if it was apparent, on the face of the specification, that the necessary quality of inventiveness was not present. In more than 25 years since Philips was decided, I am not aware of any other court applying Philips to find that the threshold requirement of inventiveness was not present. Five years after Philips, the Full Court in Bristol-Myers Squibb Co v FH Faulding & Co Ltd grappled with Philips (as well as Ramset and Microcell) and noted that it was unclear whether the majority in Philips contemplated that their threshold assessment be made solely by reference to what was disclosed on the face of the specification, or whether common general knowledge could also be considered. They noted that if it was the latter, that would leave section 18(1)(b)(ii) redundant, and emphasised the desirability of considering such issues as part of a conventional inventive step analysis.
The Kiefel CJ plurality apparently considered that common general knowledge could be considered in deciding whether what was claimed involved a relevant difference in the machine itself (be that a computer, EGM or otherwise). Their approach was to consider whether what was disclosed ‘outside the common general knowledge’ gave rise to a ‘technical contribution’ (as distinct from considering whether it gave rise to an inventive step). They said there would be no ‘technical contribution’ if those differences merely involved elements of a scheme, plan or game.
It is difficult to see what Philips added to the analysis, or indeed how it directly translates to considerations of the characterisation of an invention for the purposes of assessing manner of manufature. Applied strictly, it is a potentially limiting approach if, rather than identifying what the invention is as a matter of substance (following Myriad, and Full Court authority), it elevates the significance of any physical difference between what is claimed and common-general-knowledge embodiments.
Does the two-step test of Middleton and Perram JJ have to be followed?
If there is no binding precedent established by the High Court’s decision in Aristocrat as to the proper analysis of an EGM, then a question arises as to whether the Full Court’s two-step test must be followed by a single judge.
In the Full Court, Middleton and Perram JJ’s test first asked whether the invention was a computer-implemented invention, and if so, then asked whether it could be broadly described as an advance in computer technology.
The Kiefel CJ plurality considered that this ‘unnecessarily complicates’ the analysis, which should instead focus only on the proper characterisation of the invention. The Gordon J plurality did not seek to support Middleton and Perram JJ’s test. In these circumstances, it seems unlikely that a future single judge will consider themselves bound to apply that two-step approach, which was not one that was proposed by previous Full Courts. Instead, a future court will likely seek to characterise the invention, including by reference to common general knowledge, to determine whether any plan, scheme or game is being implemented by a machine so as to give rise to a relevant artificial effect or outcome, or whether there is otherwise a contribution being made to the operation of that machine (be it computer, EGM or other analogous machine). It may be that in the case of machine-based games that the Full Court’s approach in Aristocrat to characterising such an invention influences that court’s approach in favour of concluding that gaming integers involve a mere scheme, but no doubt future cases will throw up new and interesting factual variations.
 By reason of s 23(2) Judiciary Act 1903 (Cth).
 At , -.
 In IBM v Commissioner of Patents the application of methods to better enable a computer to display a curve was patentable, and in CCOM Pty Ltd v Jiejing Pty Ltd the method for dealing with the retrieval of key strokes to enable a computer to display Chinese language characters was similarly patentable.
 The Gordon J plurality’s comment at  that none of the cases involved an artificial state of affairs does not reference this issue.
 See for example Research Affiliates LLC v Commissioner of Patents (2014) 227 FCR 378 at , .
 (2015) 258 CLR 334.
 At .
 At .
 At , Re Cobianchi’s Application (1953) 70 RPC 199.
 At . There was also no elaboration as to whether any patent claims would needed to have been drafted to as to encompass a set of associated rules of the game.
 It may be noted that the Kiefel CJ plurality also referenced Re Cobianchi’s Application, at , footnote 69, but did not directly address it.
 At .
 At .
 At .
 At -.
 At , footnote omitted.
 At 
 N V Philips Gloeilampenfabrieken v Mirabella International Pty Ltd (1995) 183 CLR 655.
 That introductory wording was changed in 2000, although it relevantly retains reference to the word ‘invention’.
 (2000) 97 FCR 524 at
 See at - and  per Black CJ and Lehane J, Finkelstein J agreeing.
 At , see also at .
 Commissioner of Patents v Aristocrat Technologies Australia Pty Ltd (2021) 286 FCR 572 at -.
 At .