Aristocrat Technologies Australia Pty Ltd v Commissioner of Patents [2022] HCA 29
Patent validity – manner of manufacture – claim to electronic gaming machine
The High Court heard an appeal by Aristocrat from the decision of the Full Court that Aristocrat’s claim to an electronic gaming machine that could run a particular game was not a manner of manufacture as required by section 18 of the Patents Act. The six Justices of the High Court who sat on the appeal were evenly divided on the question. The result is that the appeal is dismissed, but there is no binding authority. It seems likely that the High Court will be called upon to revisit the issue, at which time the court will very likely be differently constituted. In our article, we critically analyse each judgment and seek to give some practical guidance to practitioners as to what to make of the decision.
Background facts
Aristocrat owns four innovation patents for “A system and method for providing a feature game”, which relate to a gaming system and a method of gaming. The specification described the architecture of the claimed invention as a computerised electronic gaming machine, or EGM.
Claim 1 of one of the patents, which was accepted as an exemplar by both parties, was to a gaming machine comprising various physical features common to EGMs including a game controller (comprising a processor and memory storing game program code) which controller allowed the triggering of free games during the course of the base game, with any credits won during the course of the free games adding to the player’s total in the base game.
The Commissioner, having been asked to examine the patents, revoked each on the basis that the claims were not to a manner of manufacture as required by section 18 of the Patents Act 1990 (Cth). Justice Burley allowed Aristocrat’s appeal from the Commissioner’s decision, on the basis that the claim to an EGM which included a combination of physical parts and computer software to produce a particular outcome in the form of gameplay was not a mere scheme or plan.
The Full Court of the Federal Court overturned Justice Burley’s decision in Commissioner of Patents v Aristocrat Technologies Australia Pty Ltd 286 FCR 572 (Aristocrat FCAFC). Justices Middleton and Perram did so on the basis that the invention claimed a computer-implemented invention which could not be described as an advance in computer technology, and Nicholas J did so because he did not consider that the claimed invention created an artificially created state of affairs, because it did not solve a technical problem or make some other technical contribution to the field of the invention.
Aristocrat obtained special leave to appeal to the High Court.
In Aristocrat Technologies Pty Ltd v Commissioner of Patents [2022] HCA 29 (Aristocrat HCA), six members of the High Court delivered two judgments. Kiefel CJ, Keane J and Gageler J concluded that the claims were not to a manner of manufacture and that the appeal should be dismissed. Gordon, Edelman and Steward JJ concluded that the claims were to a manner of manufacture and that the appeal should be allowed.
What is the effect of Aristocrat HCA?
Section 23(2)(a) of the Judiciary Act 1903 (Cth) provides that, when the Justices sitting as a Full Court of the High Court are equally divided in opinion as to the decision to be given on any question in a case where a decision of the Federal Court of Australia is called into question, the decision appealed from shall be affirmed. That is, where the case is in the High Court’s appellate jurisdiction, the decision appealed from stands. Accordingly, the result was that Aristocrat’s appeal was dismissed. However, that says nothing about the precedent to be followed in future cases and in lower courts.
In Federal Commissioner of Taxation v St Helens Farm (ACT) Pty Ltd [1981] HCA 4; (1981) 146 CLR 336, Barwick CJ observed that:
It is quite clear, in my opinion, and indeed settled doctrine in this Court that a case decided in this Court on an even division of opinion … does not constitute a binding precedent and that, notwithstanding it, the Court is at liberty, indeed, in my opinion, bound, to approach the question with which that case dealt, de novo, each Justice participating in the subsequent case to proceed upon his own view of the relevant law.
Aickin J held at page 430 that (emphasis added):
It is well settled that a decision of an equally divided Full Court is not a binding authority in subsequent cases in this Court.
And at page 432:
The ultimate court of appeal in Australia for all federal and many State matters is the Full Court of the High Court of Australia, and it is only unanimous or majority decisions of the Full Court which have binding authority.
Accordingly, at least the High Court must consider the question again de novo if it came up. However, the High Court’s observations above do not expressly address what lower courts should do with an evenly split High Court decision. As to that, French J (as he then was) in Long v Minister for Immigration & Multicultural & Indigenous Affairs [2002] FCA 1422 at [35]-[40] held that where there is no “clear common ratio” of a majority of the Justices of the High Court on a question, there is no binding principle on that question, and the courts below are not bound to follow one set of reasons over the other. Consistently with that approach, Lee, Anastassiou and Stewart JJ observed in Harvard Nominees Pty Ltd v Tiller [2020] FCAFC 229 at [55] that the seriously considered dicta of three members of a six-judge bench need not be followed, because it is not the majority of the High Court.
So then, what was the reasoning in each judgement in Aristocrat HCA; what can practitioners take away from the decision; and what should practitioners do in cases which involve the patentability of computer implemented inventions?
The reasoning of Kiefel CJ, Gageler and Keane JJ (the Kiefel CJ judgment)
The Kiefel CJ judgment reiterated the long-standing view that a mere scheme, plan or discovery, or mere abstract ideas or information are not patentable subject matter, and that neither the discovery of a natural phenomenon or law of nature, nor a scheme or plan devised for the accomplishing of a task, nor a set of rules whether devised for the conduct of a business or the playing game, is the proper subject of letters patent. The Kiefel CJ judgment then held that the practical implementation of a discovery of an abstract truth about nature, or a strategy devised for the conduct of business, or a set of rules devised for a game is not patentable if the mode of implementation is not patentable.
The Kiefel CJ judgment referred with approval to the principles that a claimed invention will be a proper subject of letters of patent if it has some “concrete, tangible, physical or observable effect” as distinct from “an abstract, intangible situation”, or “a mere scheme in which the invention is an abstract idea or mere intellectual information”, and stated that an artificial state of affairs may be created where the invention can broadly be described as an improvement in computer technology, where the computer is integral to the invention, rather than a mere tool.
Their Honours went on to affirm the correctness of CCOM Pty Ltd v Jiejing Pty Ltd (1994) 51 FCR 260 (except insofar as it dealt with the correctness of the Full Court’s decision in N V Philips v Mirabella ); Research Affiliates LLC v Commissioner of Patents (2014) 227 FCR 378; Encompass Corporation Pty Ltd v InfoTrack Pty Ltd (2019) 372 ALR 646; Commissioner of Patents v Rokt Pte Ltd (2020) 277 FCR 767, and referred with apparent approval to Commissioner of Patents v RPL Central Pty Ltd (2015) 238 FCR 27.
Revealingly, the Kiefel CJ judgement stated that “no one would suggest that a new variation of the rules of a game such as poker is patentable subject matter. In such a case, the well-known pack of cards is put to a different use for the purpose of the new game, but no-one would suggest that there is an invention because the only difference from the common general knowledge lies in the idea of the new game.” It is apparent that their Honours’ conclusion was that the facts of Aristocrat were analogous – that is, that the claimed invention did no more than use a well-known electronic gaming machine to play a new game.
That is also apparent from their Honours’ later observation 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” and their observation that “NRDC does not support the view that new, but unpatentable subject matter presented or operated through generic technology is patentable.”
The Kiefel CJ judgment relied heavily on the statement by Brennan, Deane and Toohey JJ in N V Philips Gloeilampenfabrieken v Mirabella International Pty Ltd (1995) 183 CLR 655 that section 18(1)(a) of the Patents Act contains a threshold requirement to the effect that if it were apparent on the face of the specification, when properly construed and understood, that a subject process was nothing more than a new use of an old product, the specification would itself disclose the absence of an alleged invention, and Brennan, Deane and Toohey JJ’s rejection of the contention that section 18(1)(b)’s requirements of novelty and inventive step are exclusive and exhaustive.
Chief Justice Kiefel, Gageler and Keane JJ stated that focussing on the features of the claimed invention which differentiate it from the common general knowledge in characterising the “invention” did not conflate the issue of manner of manufacture with issues of novelty and inventive or innovative step.
These authors note that it is unclear from the Kiefel CJ judgment precisely how to determine whether the threshold requirement of whether there is an invention is met. It appears that the judgment goes further than N V Philips in suggesting there can be reference, perhaps extensive reference, to the common general knowledge beyond what is described as such in the specification. The authors consider that, inevitably, such a threshold assessment will blur with considerations of inventive step.
Their Honours then characterised the claimed invention as a claim for a new system or method of gaming. They said: “The claimed invention takes its character, as an invention, from those elements of the claim which are not common general knowledge. If that were not so, every EGM conforming to the generic physical and hardware components, including computer components, described in the claim would be patentable simply because it, like every other EGM, allowed a new game to be played. And the only thing differentiating each new claimed invention, as an invention, would be that unpatentable game.” The authors note that this approach to characterising the invention is very difficult to reconcile with longstanding authorities, such as Lockwood Security Products Pty Ltd v Doric Products Pty Ltd (No 2) (2007) 235 CLR 173, which make it clear that inventiveness and novelty are to be characterised by reference to the combination as a whole, not by reference to individual integers.
The Kiefel CJ judgment continued: “Unlike CCOM, the present cannot be said to fall within a category of case in which, as an element of the invention, ‘there [is] a component that [is] physically affected or a change in state or information in part of a machine’.” The authors consider that this reasoning is difficult to understand. It appears that the process of playing the game would have involved a change in the state or information in the EGM, what was displayed on the machine, and its ultimate (monetary) output. The basis on which a distinction is drawn between Aristocrat’s claimed invention and the claim in CCOM (apparently because the change in state or information is not “an element of the invention”) is not clear, at least to the authors.
The Kiefel CJ judgment nevertheless criticised the judgment of Middleton and Perram JJ on the basis that it was not apparent that asking whether the claimed invention is an advance in computer technology or indeed is any advance in technology was helpful. Their Honours observed “the issue is whether the implementation of what is otherwise an unpatentable idea or plan or game involves some adaptation to or alteration of or addition to, technology otherwise well known in the common general knowledge to accommodate the exigencies of the new idea or plan or game”. Their Honours also observed that there was no reason to conclude that the claim was claiming an advance in gaming technology other than the use of a generic computer to play a new game.
Their Honours said that “a new idea implemented using old technology is simply not patentable subject matter”. This explains their Honour’s apparent rejection of the observation in CCOM that “whilst a claim for the ball point pen now would fail for anticipation [want of novelty] and obviousness [want of an inventive step], it would still be a claim for a manner of manufacture”. The Honours explained that the “ball point pen would not, now or at the time CCOM was decided, have met the threshold requirement of s 18 that it be an invention because it is not, and was not then, new”. The authors understand their Honours to be saying that the ball point pen could not meet the threshold test of “invention” if it is not novel or involves an inventive step.
The Kiefel CJ judgment concluded that the claim in Aristocrat’s innovation did not disclose any technical contribution to either computer or gaming technology outside common general knowledge: at best it disclosed a new game, which is not patentable subject matter.
The Kiefel CJ judgment stated that its decision involved no reliance on the decision in Alice Corporation Pty Ltd v CLS Bank International (2014) 573 US 208, noting the relevant US statutory provision is different from the relevant terms of the Act. The authors note that this conclusion is difficult to reconcile with their Honours’ apparent endorsement of Research Affiliates, which refers to Alice at [51]-[59].
The authors consider that the Kiefel CJ judgment, if followed, breathes new life into the threshold test for invention in section 18 of the Patents Act by allowing an opponent or revoker to reduce a claimed invention to only those integers that are not common general knowledge (and perhaps simply not novel), and then ask whether the remainder is not patentable subject matter or is not simply a new use of an old thing. This raises interesting questions with respect to innovation patents, given it is arguable that the new use of an old thing may be both novel and involves an innovative step, but may not meet such a threshold test.
The reasoning of Gordon J, Edelman and Steward JJ (the Gordon J judgment)
At the outset, the Gordon J judgment noted that the Commissioner of Patents accepted that, if the relevant claim had involved a mechanical implementation using cogs, physical reels and motors to create the gameplay, there would have been no doubt that the relevant claim was a manner of manufacture, and continued “In the 21st century, a law such as s 18(1A) of the Patents Act that is designed to encourage invention and innovation should not lead to a different conclusion where physical cogs, reels and motors are replaced by complex software and hardware that generate digital images”.
The Gordon J judgment rejected the Commissioner’s characterisation of the claim as a mere scheme or abstract idea, observing that such a characterisation could only be achieved by filleting from the claim essential and interdependent integers providing for the implementation of the game on the EGM. Their Honours observed that that characterisation must be undertaken as a matter of substance, not mere form, and requires consideration of all integers of the claim in light of the facts and matters in the specification (a view with which these authors agree). Their Honours observed that “the characterisation of the claim at the appropriate level of generality should include all of the matters that properly form part of the idea, as well as its implementation.”
While the Gordon J judgment acknowledged that there is a threshold question as to whether the subject matter can meet the description of an alleged invention, their Honours emphasised the observation in D’Arcy v Myriad Genetics Inc (2015) 258 CLR 334 that this exclusion arises “based on an admission, on the face of the specification, which makes clear that the invention claimed is not novel or does not involve an inventive step”. Their Honours observed that the threshold is low and necessarily imposes a less stringent requirement than the independent requirements of novelty and inventiveness. With respect, these authors agree with this analysis, and prefer this characterisation of the threshold requirement to the characterisation put forward in the Kiefel CJ judgment, which seems to leave the question of the threshold requirement more at large.
The Gordon J judgment then observed that a manner of manufacture thus requires only the existence of some material and artificial advantage, and that any product that improves, restores or preserves a vendible product and thereby creates some new and useful effect will be a manner of manufacture.
Their Honours observed that, although a mere scheme or plan, or the rules of a game, are not patentable, one way in which such a method will become a manner of manufacture is when it is practiced or used in a way that is embodied in a physical form. Their Honours referred to Lockwood in support of the proposition that when an idea is incorporated into a means for carrying out an idea, the idea can be taken into account when considering validity, and inventiveness may repose largely in the idea. Their Honours observed that this analysis applied where the idea consists of rules of a game, when combined with physical materials for playing the game (such as cards which are marked differently).
Their Honours stated: “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. For instance, as senior counsel for the Commissioner properly accepted on this appeal, the game of Monopoly is proper subject matter where it is embodied in a physical form such as a designed cardboard board, dice, and playing characters. It could not possibly be the case that the game of Monopoly ceases to be patentable subject matter if the graphics are displayed on a machine rather than on cardboard”. For that reason, they concluded that the requirement of a physical effect includes a concrete effect or phenomenon and extends to a change in state or information in part of a machine.
The Gordon J judgment said that it is not enough that the scheme involves the use of a machine to manipulate abstract ideas. Where the manner of manufacture relies upon some change in state or information in a machine, then that change must produce an artificial state of affairs or a useful result. Their Honours went on to cite numerous examples “where the proper characterisation of the claim is one that merely involves the use of a machine to manipulate an abstract idea rather than involving the implementation of the idea on a machine to produce an artificial state of affairs and a useful result” and observed that an “idea that uses a computer but does not generate some artificial state of affairs, remains no more than idea”. Among the examples cited by their Honours were Research Affiliates, RPL Central, Encompass and Rokt.
Their Honours continued: “Although there was no artificial state of affairs created in any of these cases, and the results in all of these cases are plainly correct, some of the statements explaining the results in these and other cases must be read in the context of what was being decided”. Their Honours gave one example of loose language – the expression of the characterisation question as whether the implementation of the scheme could be described as an improvement in computer technology in Rokt – but did not exhaustively identify what other aspects of the language in those cases are called into question or ought to be treated with caution.
The Gordon J judgment then stated: “A better way of expressing the point … would be to ask whether, properly characterised, the subject matter that is alleged to be patentable is (i) an abstract idea that is manipulated on a computer; or (ii) an abstract idea which is implemented on a computer to produce an artificial state of affairs and a useful result. The artificial state of affairs may be an improvement in computer technology, but it need not be. It is enough that the artificial state of affairs and useful result are carried out “by the way in which the method is carried out on the computer”.
Their Honours then went on to warn against relying on UK and US authority.
The Gordon J judgment observed that claim 1 should not be characterised either by focusing on the integers which were common to generic EGMs or by reference only to those elements that are deduced from the specification to involve alleged novelty and inventiveness. They concluded “the best characterisation of Claim 1 is an EGM incorporating an interdependent player interface and a game controller which includes feature games and configurable symbols.”
Their Honours then observed that it is not of assistance to consider whether the subject matter of a claim is computer implemented. The Gordon J judgment concluded that the claim was to a manner of manufacture, because on its proper characterisation, the claim was not merely the idea of a game. Their Honours considered that the operation of the game controller could not be severed from the interdependent player interface, and that the claimed operation of the game controller, displayed through the player interface, is an altered EGM involving an artificial state of affairs.
With respect, the authors do not find the Gordon J judgment’s re-characterisation of the test helpful. The basis for distinguishing between an abstract idea which is “manipulated on a computer” and an abstract idea which is “implemented on a computer to produce an artificial state of affairs and a useful result” is unclear. Their Honours observe that “the requirement of a “physical effect” includes “a concrete effect or phenomenon and extends to “a change in state or information in a part of a machine”. They then explain that “[w]here the manner of manufacture relies upon some change in state or information in a machine, then that change must produce an artificial state of affairs and a useful result”.
Their Honours thus appear to contemplate that, in such circumstances, the requisite “artificial state of affairs” which gives rise to a manner of manufacture is separate from the “physical effect” or “change in state or information in a part of a machine” which inevitably results from the “manipulation” of an abstract idea on a computer.
However, the authors’ view is that the identification of the requisite artificial state of affairs and useful result which was ultimately applied by their Honours is unclear. The authors find it difficult to understand how their Honours concluded that the claim under consideration did produce an artificial state of affairs and a useful result, but the claims in each of Research Affiliates, RPL Central, Encompass and Rokt did not. Nor does their Honours’ reference to the patentable subject matter in CCOM and Burroughs make it any clearer as to why the earlier Full Court authorities were plainly correctly decided but this case was not by the Full Court below.
Are there any clear principles arising from Aristocrat?
These authors consider that three matters are clear from Aristocrat HCA (in that all justices seemed to agree on them), although they do not sit easily together and they may not be of much assistance to practitioners.
First, the majority judgment of the Full Court in Aristocrat FCAFC incorrectly states the law and should not be followed. It appears that the judgment is incorrect because it is not of assistance to ask whether an invention is “computer implemented” or involves an improvement in computer technology (although the judgments differ on the nature of enquiry which should be made).
Second, each of the earlier Full Court decisions (Research Affiliates, RPL Central, Encompass and Rokt) was correctly decided (albeit that Gordon, Edelman and Steward JJ cavil with the wording of those decisions, and this concession is a little hard to reconcile with the Gordon J judgment, because it is not clear why those cases did not involve an artificial state of affairs and a useful result on the reasoning of Gordon, Edelman and Steward JJ).
Third, the law of the United States is not relevant to the question of manner of manufacture in Australia (although this is difficult to reconcile with the conclusion that each of the earlier Full Courts was correctly decided, given that Research Affiliates drew heavily on US decisions, and the later decisions rely heavily on Research Affiliates).
What should practitioners do?
It seems almost inevitable that the question of manner of manufacture in the context of computer implemented inventions will return to the High Court sooner rather than later. We hope that when it does, more clarity results.
Pending that result, and in the absence of a binding judgment or legislative change, we consider it likely that courts will now apply the earlier Full Court authority (not the majority’s judgment in Aristocrat FCAFC). The earlier Full Court authority is conveniently (and helpfully) summarised in Nicholas J’s decision in Aristocrat FCAFC. Having said that, it would be open to any litigant to argue that a claim to a computer implemented invention is patentable because it is implemented on a computer to produce an artificial state of affairs and a useful result and for the matter to return to the High Court for a hearing de novo (where none of the justices would be required to adopt the reasoning in Aristocrat HCA).
That is, patent applicants could prosecute claims on the basis that the Gordon J judgment is correct, but they would need to do so on the understanding that the matter would likely need resolution by the High Court. While it would also be open to argue that Research Affiliates, RPL Central, Encompass and Rokt were wrongly decided, the authors consider the likelihood of a new High Court accepting that would be very low, given that both judgments indicate those cases were correctly decided.