1. Introduction
A. European patents for computer programs and business methods
Since 1978, the patent systems of European states have been based on the European Patent Convention (EPC).Footnote 1 Signed in 1973 as a “special agreement” under the Paris Convention,Footnote 2 the purpose of the EPC is to establish “a system of law, common to the Contracting States, for the grant of patents for inventions”,Footnote 3 and a European Patent Office (EPO) to administer it. While patents granted under the EPC are referred to as “European patents”, they take effect as a bundle of national patents under the laws of the States in which protection is sought. Consistent with this, post-grant issues of validity and infringement remain the exclusive jurisdiction of national courts. On the other hand, post-grant oppositions to a patent can be brought before the EPO,Footnote 4 and a European patent may only be revoked on an Article 138 EPC ground.Footnote 5
Throughout Europe, the substantive provisions of the EPC are transposed into the legislation of each Contracting State. With only three official (German, French and English) language versions, most transpositions are necessarily imperfect; and indeed, even the U.K. Patents Act 1977 (PA) fails to follow the official English text.Footnote 6 In combination with the system of dual EPO/national jurisdiction, this makes conflicting applications of the law inevitable, both between national courts inter se, and between those courts and the EPO.
In the U.K., the risk of conflicting decisions with the EPO and other Contracting States is reduced by section 130(7) of the PA, transposed from a Resolution annexed to the Community Patent Convention (CPC).Footnote 7 Section 130(7) records the intent of the U.K. LegislatureFootnote 8 that certain provisions of the PA “are so framed as to have, as nearly as practicable, the same effects in the United Kingdom as the corresponding provisions of the European Patent Convention”. As expressed by Lord Hoffmann in Merrell Dow Pharmaceuticals Inc v. H.B. Norton & Co. Ltd, this means that the U.K. courts must construe each provision “so that, so far as possible, it has the same effect as” its EPC counterpart:
For this purpose, [the courts] must have regard to the decisions of the [EPO] on the construction of the EPC. These decisions are not strictly binding upon courts in the U.K. but they are of great persuasive authority; first, because they are decisions of expert courts (the Boards of Appeal and Enlarged Board of Appeal of the EPO) involved daily in the administration of the EPC and secondly, because it would be highly undesirable for the provisions of the EPC to be construed differently in the EPO from the way they are interpreted in the national courts of a Contracting State.Footnote 9
Of central importance among the section 130(7) provisions is section 1(1) & (2) PA. According to section 1(1), a patent may only be granted for an invention that is new, involves an inventive step, is capable of industrial application, and for which the grant of a patent is not excluded (inter alia) by section 1(2), as follows:
It is hereby declared that the following (among other things) are not inventions for the purposes of this Act, that is to say, anything which consists of –
(a) a discovery, scientific theory or mathematical method;
(b) a literary, dramatic, musical or artistic work or any other aesthetic creation whatsoever;
(c) a scheme, rule or method for performing a mental act, playing a game or doing business, or a program for a computer;
(d) the presentation of information;
but the foregoing provision shall prevent anything from being treated as an invention for the purposes of this Act only to the extent that a patent or application for a patent relates to that thing as such.
Corresponding to section 1(1) & (2) is Article 52(1) to (3) EPC. According to Article 52(1), “European patents shall be granted for any inventions, in all fields of technology, provided that they are new, involve an inventive step and are susceptible of industrial application”. The phrase “in all fields of technology” is a recent addition, having been inserted at the EPC 2000 Revision Conference to reflect the wording of Article 27.1 of the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS), and to date has no PA counterpart. By Article 52(2), “[t]he following in particular shall not be regarded as inventions within the meaning of paragraph 1: (a) discoveries, scientific theories and mathematical methods; (b) aesthetic creations; (c) schemes, rules and methods for performing mental acts, playing games or doing business, and programs for computers; and (d) presentations of information”, though with patentability only to be excluded – under Article 52(3) – “to the extent to which a European patent application or … patent relates to such subject-matter or activities as such”. While Article 52(2) & (3) lacks the inclusive language of section 1(2) PA, the travaux confirm that it was not intended to be exhaustive,Footnote 10 and this has been accepted by the EPO.Footnote 11 As for the provisions' other differences: their potential effect has been avoided by the U.K. courts' treatment of Article 52(2) & (3) as the relevant source of U.K. law.
Since the commencement of the EPC, the definition of the invention in Article 52(2) & (3) has given rise to considerable controversy and legal uncertainty. The reason is its interpretation by the EPO as resolving to a requirement for “technical character”. Thus, if a subject matter has technical character, it is an “invention” within the meaning of Article 52(1), and therefore capable of supporting a patent. For it to have technical character, a subject matter need only include some technical feature, regardless of whether that feature dominates.Footnote 12 One result has been a recognition of the patentability of computer programs, including the business and other methods which they implement, notwithstanding the terms of Article 52(2)(c) & (3).Footnote 13 For example, in the recent case of T_1616/08 (AMAZON/Gift Order), the Technical Board of Appeal (TBA) held that a method “for placing a purchase order via a communications network” is an “invention” within the meaning of Article 52(1) EPC.Footnote 14
On its face, the EPO interpretation of Article 52(2) & (3) is difficult to accept: computer programs would seem to have technical character, while automated methods of purchasing goods are indisputably methods of doing business. According to the TBA, however, the exclusion of both from threshold patentability is avoided by the Article 52(3) “as such” proviso, which restricts the application of Article 52(2) to the enumerated subject matter abstractly conceived.Footnote 15 As the TBA explained in T_0154/04 (DUNS LICENSING ASSOCIATES/Estimating Sales Activity), Article 52(2) & (3) – on the EPO's understanding – is the drafters' expression of the distinction between “practical scientific applications and intellectual achievements”.Footnote 16 The result is that the exclusion from European patentability of computer programs and business methods as such does not prevent the grant of a European patent for an automated method of commercial transaction, nor for any other computer-implemented method.Footnote 17
At the time it was decided, the effect of Duns and the “technical character” theory of Article 52(2) & (3) was to vindicate the view in T_0935/97 (IBM/Computer Programs) that understandings of the “invention” ought to reflect the practices of other jurisdictions, including of Japan and the United States. Particularly influential has been US practice which, until 2008, when In re Bilski was decided, supported the inherent patentability of any process that produced a “useful, concrete, and tangible result”.Footnote 18 In In re Bilski, however, this test was rejected. According to the US Court of Appeals, a process is only eligible for a patent if “(1) it is tied to a particular machine or apparatus, or (2) it transforms a particular article into a different state or thing”.Footnote 19 The effect of this “machine or transformation” test was to exclude from the scope of the US system certain methods regarded by the EPO as having technical character,Footnote 20 undermining the rapprochement of European and US law. On the other hand, in June 2010 In re Bilski was restricted by the US Supreme Court, which described its “machine or transformation” test as a non-exhaustive test of the existence of an invention.Footnote 21 In the absence of a statement by the Court of the other valid tests, the case has left the scope of US law substantially uncertain, and underlined the difficulties of attempting to interpret the EPC in a manner that tracks US jurisprudence.
B. U.K. responses to EPO decision making
With some important exceptions, the EPO's approach to Article 52(2) & (3) has been skeptically received by the U.K. courts. This has been particularly so since 2005, when the EU's attempt to entrench the approach in a Directive on the patentability of computer-implemented inventions failed in a climate of intense controversy.Footnote 22 An important event in the history of that attempt was the European Economic and Social Committee's publication of an Opinion on the Proposal in September 2002, as part of the EU's mandatory legislative process.Footnote 23 The Opinion was deeply critical of the Proposal, as well as of the manner in which it was prepared. According to the Committee, the Proposal was a partisan documentFootnote 24 that lacked a sound economic basis, showed insufficient regard for public interest and democracy,Footnote 25 and put Europe at risk of increased legal uncertainty, and perhaps even “legal chaos”.Footnote 26 The Directive's provisions themselves were criticized as offering a “de facto acceptance and justification of the a posteriori drift of EPO jurisprudence”,Footnote 27 and as opening the way to the abolition of Article 52(2)(c) & (3) by envisaging the patenting of all methods implemented through programs or on electronic networks.Footnote 28 The doctrinal premise of the Proposal – the EPO distinction between “computer software by itself” and “software producing technical results” – was described as “indefinable” and “the product of legal casuistry”.Footnote 29
Immediately after the Directive's defeat, in CFPH's Application Footnote 30 and Halliburton Energy Services Inc v. Smith International Inc,Footnote 31 Mr Prescott Q. C. and Pumfrey J. eschewed conventional acquiescence with the EPO BoardsFootnote 32 by rejecting their view of Article 52(2) & (3) in favour of one earlier proposed by Laddie J. According to his Lordship in Fujitsu Ltd's Application Footnote 33 – in a view rejected in that case by the Court of AppealFootnote 34 – Article 52(2) & (3) does not resolve to a requirement for technical character, but rather contains independent policy exclusions having potentially different spheres of application.Footnote 35 In the judgments of Mr Prescott Q.C. and Pumfrey J., this approach was justified by the nature of the EPC as an international agreement (appropriately interpreted teleologically),Footnote 36 and the impossibility of reconciling EPO authorities.Footnote 37
In both CFPH and Halliburton, the policy objectives underlying the Article 52(2) exclusions were described as supporting their division along a spectrum of soft and hard exclusions; as Mr Prescott Q.C. said in CFPH, “the harder the exclusion, the more it is the policy of the law to insist that the use of the information be not foreclosed under patent law”.Footnote 38 In his view, computer programs and business methods were on the “hard” side of this spectrum; the EPC drafters having decided that granting patents for programs would “do more harm than good”,Footnote 39 and that granting patents for business methods was unnecessary to encourage business innovation.Footnote 40 In Halliburton Pumfrey J. refrained from detailed analysis of the individual exclusions, suggesting that the real problems with Article 52(2) are “caused by (c) and (d), which, by reason of their exclusion ‘only to the extent that the patent relates to such subject matter … as such’ are remarkably difficult to assess in cases lying near the boundary, particularly as it is difficult to discern an underlying policy”:
For example, do we only exclude computer programs as such because computer programs as such are protected by copyright, like aesthetic creations which can likewise be used industrially? Or is there some other reason? Whatever the reason, surely it is not the same as the reason for excluding methods of doing business?Footnote 41
In the end his Lordship eschewed the need to decide the reason, being content instead to determine the case pursuant to a positive conception of the “invention”, as any subject matter “tethered” to a specific industrial activity, and directed to advancing the technical arts.Footnote 42
The following year, in 2006, the scope of Article 52(2)(c) & (3) came before the Court of Appeal in Aerotel Ltd v. Telco Holdings.Footnote 43 Writing the judgment for the Court, Jacob L.J. summarized the relevant EPO cases as supporting six “mutually contradictory”Footnote 44 approaches to the “invention”, including one which he described as “not intellectually honest”.Footnote 45 The six approaches were a “contribution approach,” focused on whether the inventive step resides only in the contribution of excluded matter; a “technical effect approach,” focused on whether the subject matter makes a technical contribution to the known art; an “any hardware approach,” focused on whether the claim involves the use of, or is, a piece of physical hardware, however mundane, such as a computer or pencil and paper; and three variants of the “any hardware approach”. According to Jacob L.J., in a judgment which he described as constrained by past U.K. authorities, the approach required by U.K. law is “the technical effect approach with the rider … that a contribution which consists solely of excluded matter will not count as a technical contribution”; ie, an approach different from each of those which he attributed to the EPO.
Jacob L.J.'s judgment in Aerotel reflects a deep frustration over EPO decision making. At the heart of that frustration is a body of sparsely reasoned cases that are on the one hand impossible to reconcile inter se, and on the other hand of equal authoritative weight.Footnote 46 It was in the context of such cases that his Lordship suggested – “in a spirit of cooperation between national courts and the EPO”Footnote 47 – that “the time has come for matters to be clarified by an Enlarged Board of Appeal” (EBA).Footnote 48 In his suggestion, and based on submissions by the parties in the Aerotel case, the following three questions could appropriately be asked:
(1) What is the correct approach to adopt in determining whether an invention relates to subject matter that is excluded under Article 52? (2) How should those elements of a claim that relate to excluded subject matter be treated when assessing whether an invention is novel and inventive under Articles 54 and 56? (3) And specifically, (a) Is an operative computer program loaded onto a medium such as a chip or hard drive of a computer excluded by Article 52(2) unless it produces a technical effect, if so what is meant by “technical effect”? (b) What are the key characteristics of the method of doing business exclusion?Footnote 49
C. Seeking a referral to the EBA
Under Article 112(1) EPC, both the Boards of Appeal and the EPO President have a power to refer a matter to the EBA “[i]n order to ensure uniform application of the law, or if a point of law of fundamental importance arises”. In the case of the Boards, the power extends to “any question” on which the Board “considers that a decision is required for the above purposes”.Footnote 50 In the case of the President, it extends to “any point of law … where two Boards of Appeal have given different decisions on that question”.Footnote 51
A month after Aerotel, in the Duns Licensing case, a request was made for a referral of Jacob L. J.'s questions by the TBA, and rejected on the ground that neither of the Article 112(1)(a) bases was made out:Footnote 52 a decision of the EBA was not required to ensure uniform application of the law, nor to resolve an important point of law arising in the case. On the substantive points raised by Jacob L.J., the Board defended its jurisprudence as “constant” and capable of “succinct summary” in the manner above.Footnote 53 It also returned Jacob L.J.'s specific criticism by describing the U.K. Court's own approach to Article 52(2) & (3) as not “consistent with a good-faith interpretation” of the EPC.Footnote 54
Following Duns, there remained the possibility of a referral by the EPO President, to whom Jacob L.J. had expressly directed his remarks. In February 2007, however, the President – Alain Pompidou from France – wrote to him as follows:
Leaving aside Board of Appeal case law the line of reasoning of which has been abandoned by later case law, I believe there are insufficient differences between current Board of Appeal decisions dealing with Article 52 EPC exclusions on important points of law that would justify a referral at this stage.Footnote 55
This did not mean, he emphasized, “that a clarification of certain issues by the Enlarged Board of Appeal would not be welcomed”; only that a referral would not be appropriate without a divergence of authority sufficient to support conflicting patent granting practices.Footnote 56
Five months later, in July 2007, Mr Pompidou was succeeded by a new President – Alison Brimelow from the U.K. In October 2008, nearly two years after Duns, Ms. Brimelow exercised her power under Article 112(1)(b) to refer the following questions to the EBA:
[1.] Can a computer program only be excluded as a computer program as such if it is explicitly claimed as a computer program? [2.] (a) Can a claim in the area of computer programs avoid exclusion under Art. 52(2)(c) and (3) merely by explicitly mentioning the use of a computer or a computer-readable storage medium? (b) If Question 2(a) is answered in the negative, is a further technical effect necessary to avoid exclusion, said effect going beyond those effects inherent in the use of a computer or data storage medium to respectively execute or store a computer program? [3.] (a) Must a claimed feature cause a technical effect on a physical entity in the real world in order to contribute to the technical character of the claim? (b) If Question 3(a) is answered in the positive, is it sufficient that the physical entity be an unspecified computer? (c) If Question 3(a) is answered in the negative, can features contribute to the technical character of the claim if the only effects to which they contribute are independent of any particular hardware that may be used? [4.] (a) Does the activity of programming a computer necessarily involve technical considerations? (b) If Question 4(a) is answered in the positive, do all features resulting from programming thus contribute to the technical character of a claim? (c) If Question 4(a) is answered in the negative, can features resulting from programming contribute to the technical character of a claim only when they contribute to a further technical effect when the program is executed?Footnote 57
It is fair to say that support for Ms. Brimelow's referral was ambivalent from the start, including within the U.K. Aside from a number of amicus curiae submissions, this is apparent from the public statements of such influential people as Lord Hoffmann, who described her referral as inadmissible, and defended the EPO's interpretation of Article 52(2) & (3).Footnote 58 Given this, there was little surprise in May 2010 when the EBA reached the same conclusion.Footnote 59 In one of its longest ever Opinions, and only the eighth involving Article 112(1)(b) to date,Footnote 60 it dismissed the President's referral as inadmissible, and supported the EPO approach as based on sound (“consistent”) logic,Footnote 61 and as having “created a practicable system for delimiting the innovations for which a patent may be granted”.Footnote 62
In the analysis below I suggest otherwise, by criticizing the EBA's decision on two central grounds. First, I suggest that its basis is an interpretation of Article 112(1)(b) that is inconsistent with the principles of Articles 31–33 of the Vienna Convention, and that offends the constitutional principles from which it was purportedly derived. And second, I suggest that the EBA's support for the TBA approach is in any case inappropriate. This is because that approach (a) suffers from the same faulty logic for which the EBA criticized the President's referral, (b) lacks doctrinal and theoretical coherence, and (c) is incapable of producing legal certainty, either within the EPC or national (European) patent systems. Hence my conclusion, that “judiciary-driven legal development”Footnote 63 within the EPO has met its limits with respect to the European invention, with the result – as the EBA itself suggested – that “it is time for the legislator to take over.”Footnote 64
Some of the wider implications of this thesis will be apparent, including for international judicial cooperation. As the Aerotel and Duns exchange reflects in the context of patent law and the EPO at least, understandings of the requirements of such cooperation diverge, and in ways that transcend jurisdictional bounds. For example, in 2008, in Symbian v. Comptroller-General of Patents,Footnote 65 Lord Neuberger implicitly rebuked the Aerotel Court for its failure sufficiently to “compromise” on the nature of the invention, and its refusal to “march[ ] together” with the EPO.Footnote 66 On the other hand, in the Herchel Smith Lecture of 1996, Floyd J. commented as follows on Lord Hoffmann's speech from Merrell Dow:Footnote 67
I disagree that the members of the Boards of Appeal are more expert than the courts of this country or of the courts of Holland or Germany in interpreting the EPC or applying it. The EPO has, furthermore, no experience in the enforcement of the patents which it grants. On points such as those raised in Merrell Dow and Mobil [novelty and subject matter] there is enormous value in the point being decided by courts with experience in deciding infringement, or at least on the basis of an interplay between those courts and the EPO. I also disagree that such decisions should be followed slavishly so as to make the EPC a European House of Lords on patent law.Footnote 68
By restricting the scope for EBA referrals, and supporting the approach of the TBA notwithstanding widespread concerns regarding its doctrinal and theoretical basis, the EBA has confirmed its view of the TBA as something of a “European House of Lords”, and sanctioned an approach to EPC interpretation described by the EBA itself as “praetorian”.Footnote 69 It is submitted that this is unfortunate, and that if it represents the reality of life under the EPC system as it stands, we would indeed fare better with a European patent court, notwithstanding the dilution of common law values and jurisprudence which such a court would entail.
II. The President's Power to Refer Points of Law to the EBA
By provision of Article 112(1)(b) of the EPC:
In order to ensure uniform application of the law, or if a point of law of fundamental importance arises: … the President of the European Patent Office may refer a point of law to the Enlarged Board of Appeal where two Boards of Appeal have given different decisions on that question.
On its face this Article seems to require two things, consistent with it having two distinct limbs: (a) “different decisions” given by two Boards of Appeal, (b) on a point of law of fundamental importance, or producing such divergence that an EBA decision is necessary “to ensure uniform application of the law”. Implicit in this view is an interpretation of Article 112(1)(b) as supporting two heads of presidential referral: one aimed at clarifying points of law of fundamental importance, as suggested by Jacob L. J. in Aerotel,Footnote 70 and the other at ensuring the uniform application of the EPC.Footnote 71
In G_0003/08, the EBA seemed initially to share this view. Thus, in determining the admissibility of the referral before it, it asked whether the computer programs exclusion to which the referral related is “a point of law of fundamental importance”, and whether the cases cited by the President in support of that referral disclose differences on that point by two EPO Boards.Footnote 72 It decided that while the exclusion is a point of law of fundamental importance,Footnote 73 the cases do not disclose “different decisions” within the meaning of Article 112(1)(b).Footnote 74 Its reasons for the second of these conclusions were as follows. First, a comparison of the official versions of the phrase “different decisions”/ “décisions divergentes”/“abweichende Entscheidungen” reveals an interpretive ambiguity in the first limb of Article 112(1)(b) which can only be resolved with reference to the Article's “object and purpose”,Footnote 75 consistent with Article 31 of the Vienna Convention.Footnote 76 And second, such object and purpose are revealed by “current constitutional thinking”Footnote 77 to support a restrictive interpretation of Article 112(1)(b), as requiring (in all cases)
… a divergence or, better, conflict in the case law making it difficult if not impossible for the Office to bring its patent granting practice into line with the case law of the Boards of Appeal.Footnote 78
The “constitutional thinking” which led to this conclusion concerns democracy and the notion of legal development. According to the EBA, such considerations are relevant to Article 112(1)(b) by virtue of the nature of the European Patent Organisation as
… an international, intergovernmental organisation, modelled on a modern state order and based on the separation of powers principle, which the sovereign contracting states have entrusted with the exercise of some of their national powers in the field of patents.Footnote 79
From this, and the democratic nature and obligations of EPC Contacting States,Footnote 80 the EBA derived a duty on the part of the Organisation to ensure “the predictability and verifiability” of its actions, and to protect “the separation of powers, the rule of law and respect for human rights including fundamental procedural rights.”Footnote 81 It is this duty which provided the essential context for its reading of Article 112(1)(b).
Of particular importance to that reading was the separation of powers doctrine. According to the EBA, that doctrine supports a view of the EPO Boards as having primary responsibility for interpreting the EPC,Footnote 82 as the only proper judicial bodies within the Organisation.Footnote 83 By enabling presidential referrals to the EBA, it suggested, Article 112(1)(b) involves a limited usurpation of those functions, which can only be justified in extremis:
The fact that the Enlarged Board takes action only on a referral from the Boards of Appeal or the President (with the exception of petitions for review under Article 112a EPC, which however concern procedural matters and have a very narrow scope) and thus does not constitute a further instance ranking above the Boards of Appeal within the EPC judicial system is a clear indication of the extent of its significance for legal uniformity. The exhaustive list of admissibility criteria for a referral under Articles 112(1)(a) and (b) EPC implies that the Enlarged Board takes decisions on specific legal questions and that neither the Boards of Appeal nor the President are authorised to consult it whenever they so wish in order to clarify abstract points of law. For that purpose the President can call upon a separate Legal Department within the Office.Footnote 84
The result was a view of Article 112(1)(b) as having been “only intended” to have the limited scope above.Footnote 85 Thus, a mere difference in two or more cases, including one that reveals the law to be proceeding in a non-linear fashion,Footnote 86 that abandons or modifies an earlier approach, or that effects “a radical shift in [the Boards'] jurisprudence”,Footnote 87 will never be a valid basis for a presidential referral.Footnote 88 This is particularly important, the EBA suggested, in the context of the EPC, due to “its numerous undefined legal terms”,Footnote 89 the challenges of “new technologies”,Footnote 90 and the fact-specific nature of patent law.Footnote 91 As expressed in its opinion,
… the Enlarged Board cannot develop the law in the same way as the Boards of Appeal, because it does not have to decide on facts of pending appeals, but only in specific instances and only in the aforementioned context of points of law referred to it under Article 112(1) EPC. This applies a fortiori to referrals by the President, which do not even arise from a specific appeal.Footnote 92
By its reasoning, the EBA affirmed the view of the TBA in Duns Licensing, and Mr Pompidou in his letter to Jacob L. J. Nonetheless, that reasoning is problematic in several respects. One is that it proceeds without regard to the opening clause of Article 112(1)(b) – being the second part of that Article's second limb – and its text “or if a point of law of fundamental importance arises.” As suggested above, and clearly understood by Jacob L. J. in Aerotel,Footnote 93 this text envisages presidential referrals aimed at clarifying fundamental points of law, regardless of whether they are needed “to ensure [the law's] uniform application”. In contrast, by interpreting the first limb of Article 112(1)(b) as limiting the Article's applicability to situations in which it is “difficult if not impossible” for the EPO to apply the law as stated by its Boards,Footnote 94 the EBA rendered this head of referral redundant. Hence my suggestion, that its reasoning exploits the ambiguity in the first limb of Article 112(1)(b) to import considerations of constitutional policy, which considerations are then used to delimit the literal meaning of the Article's other terms, and thereby remove one of the President's heads of power. The result is a reading of Article 112(1)(b) which is difficult to reconcile with its literal terms, and the “good faith” interpretation required by Article 31 of the Vienna Convention,Footnote 95 and that restricts substantially the President's power of referral, thereby increasing the judicial authority of the TBA.
A second, albeit related respect in which the EBA's reasoning is problematic concerns its failure to refer to the unpublished travaux préparatoires for Article 112(1)(b) in reaching its decision. Of importance in this regard is Article 33.4 of the Vienna Convention:
Except where a particular text prevails in accordance with paragraph 1, when a comparison of the authentic texts [of a treaty] discloses a difference of meaning which the application of articles 31 and 32 does not remove, the meaning which best reconciles the texts, having regard to the object and purpose of the treaty, shall be adopted.
In G_0003/08, it was the EBA's comparison of the three official versions of the first limb of Article 112(1)(b) which created the interpretive ambiguity above, and thereby enabled the EBA's reliance on constitutional factors to read down that Article's scope of application. However, and as Article 33.4 makes clear, this interpretive approach – which involved reconciling the official versions of Article 112(1)(b) having regard to the object and purpose of the EPCFootnote 96 – was not open to the EBA without first attempting to resolve the ambiguity by “the application of articles 31 and 32”. Article 32 states as follows:
Recourse may be had to supplementary means of interpretation, including the preparatory work of the treaty and the circumstances of its conclusion, in order to confirm the meaning resulting from the application of article 31, or to determine the meaning when the interpretation according to article 31: (a) leaves the meaning ambiguous or obscure; or (b) leads to a result which is manifestly absurd or unreasonable.
The effect of this provision, in combination with Article 33.4, was to make the travaux préparatoires for Article 112(1)(b) an essential “supplementary means of interpretation” in G_0003/08. Indeed, even without regard to Article 33.4, their relevance as such is apparent from the explicit basis of the EBA's interpretation in the “object and purpose” of Article 112(1)(b), of which the travaux are primary evidence – as the EBA itself has remarked.Footnote 97 This, in combination with the difficulty of reconciling the EBA's interpretation of Article 112(1)(b) with that Article's literal meaning, and the EBA's consistent practice of considering the travaux to resolve ambiguities in other cases – including in all but two of those involving Article 112(1)(b) referralsFootnote 98 – supports the conclusion that in failing to refer to them in G_0003/08 the EBA proceeded inconsistently with Article 33.4, and arguably Article 31, of the Vienna Convention.
The result is of more than academic or methodological significance. The reason is that the travaux – which are not readily accessible outside the EPO – would seem to contradict the EBA's interpretation of Article 112(1)(b), by suggesting that the central aim of that provision was indeed “to ensure the uniform application of the law or to clarify a fundamentally important point of law”,Footnote 99 as suggested above. In addition, the travaux show that the drafters considered the constitutional issues relied on by the EBA, and supported Article 112(1)(b) even after doing so. Indeed, constitutional issues were foremost in their minds, as evidenced by the following record of their earliest discussions of Article 112(1)(b) in 1969:
Serious objections were brought forward against the possibility, provided for in paragraph 1 b, of the President of the European Patent Office asking the Enlarged Board of Appeal for an opinion on questions not arising in proceedings on a case: It was said that such a possibility would seriously impair the judicial character of the Enlarged Board of Appeal; Courts could not give an opinion on abstract questions put to them by administrative authorities but had to decide specific individual cases; If abstract questions are put to them for a decision they are recognized as having a quasi-legislative power; There was the danger that when the Enlarged Board of Appeal had given an opinion on an abstract question it would not depart from this in subsequent specific cases; There was no need to give the Enlarged Board of Appeal powers of this type, since the President of the Patent Office could first ask the Legal Department of the European Patent Office for an opinion on any questions arising for the first time; The President could also ask a member of the Enlarged Board of Appeal to give an opinion in his own name.Footnote 100
Not all of the Working Party members agreed with these objections, producing a conflict of opinionFootnote 101 over the desirable scope of the provision.Footnote 102 The result was an initial version of Article 112(1)(b) that included, in addition to its current first limb, a sub-section which authorized the President “at any time [to] ask the Enlarged Board of Appeal for an opinion on any question, except where such question arises in proceedings on a case”.Footnote 103 Alongside the Working Party's record of this version, the following explanatory note appears:
The Working Party points out that the second sub-section of sub-paragraph (b) limits the President's power to refer a matter to the Enlarged Board of Appeal. The Working Party was unable to agree on the advisability of giving the President the power provided for in the first sub-section of sub-paragraph (b). This question must be re-examined in consultation with the national legal advisers.Footnote 104
When the participants at the first of the EPC's preparatory conferences came to consider the initial version of Article 112(1)(b) in January 1970, they agreed to postpone a decision on the first sub-section for the same constitutional reasons as considered by the Working Party in 1969. According to the record of the Conference discussions:
Several delegations expressed doubts on the provision in question, particularly since it would involve the risk of obliging the Enlarged Board of Appeal to take decisions on abstract questions, which might hinder it later when it came to consider concrete questions of a similar nature, and also because the Enlarged Board of Appeal, notwithstanding the rules governing its composition, cannot be totally likened to a judicial body.
On the other hand, other delegations believed the provision in question could be most useful, especially during the running in period of the European Patent Office, during which a number of questions of interpretation were bound to occur, on which it would be expedient for the President to be able to obtain the opinion of the Enlarged Board of Appeal.
Finally, the Conference considered that it would be particularly useful to receive the opinions of the national legal advisers before taking a final decision on this problem.Footnote 105
Eleven months later the matter was considered by the Conference again. In the record of its meeting, the participants affirmed their commitment to the second sub-section, and their ongoing uncertainty regarding the first. According to the travaux:
The Conference agreed that the President should in any event have the power to ask the Enlarged Board of Appeal for an opinion in the case referred to in the second sub-section of sub-paragraph (b). On the other hand, there was no agreement on the question whether the President should also be given such a power in the other cases referred to in the first sub-section, which are not covered by the second sub-section. If an affirmative answer is given to this question at a later date, the text of the second sub-section could be deleted.Footnote 106
Then, in October 1971, the responsible Working Party
… agreed by a majority that the President of the European Patent Office may only call upon the Enlarged Board of Appeal when two Boards of Appeal have given different decisions on the same question; the more widely drafted first sub-section of this sub-paragraph was therefore deleted.Footnote 107
The result was the version of Article 112(1)(b) which remains today, formulated and accepted by the drafters after full consideration of its constitutional implications, including – as the preceding discussion demonstrates – for the separation of powers and the development of the law, and having regard to the fact-specific nature of judicial patent law-making, and the inability of the EBA to “be totally likened” to a court.Footnote 108 In combination with the historical evidence regarding the aim of Article 112(1)(b), this raises a question regarding the appropriateness of the EBA's reliance on constitutional arguments to restrict the literal meaning and terms of Article 112(1)(b). Indeed, it suggests that the EBA's interpretation involved a subversion of the very democratic principles from which that interpretation was purportedly derived. This is because it involved the EBA's usurpation of the constitutionally-mandated executive authority of the President, and the legislative authority of the EPC Contracting States.Footnote 109 Further, while the EBA's analysis was premised on a view of the Boards of Appeal as the only proper judicial bodies within the Organisation, that view ignores the identity of the Boards as “a limb of the [EPO's] granting office itself”,Footnote 110 and contradicts their description on the EPO's own website:
The current legal position is that the Boards of Appeal, together with their associated administrative services, are integrated into the European Patent Office as Directorate-General 3 (DG 3). However, the boards' administrative and organisational attachment to the EPO which is an administrative authority obscures their judicial nature and is not fully commensurate with their function as a judicial body.Footnote 111
Indeed, it is precisely with a view to addressing the Boards' lack of judicial autonomy that a proposal is on foot for the EPC to be revised.Footnote 112
III. The Technical Character Theory of Article 52(2) & (3)
The question arises as to the implications of this analysis for the outcome of G_0003/08. In the EBA's opinion, the differences of relevance in the cases cited by the President did not create sufficient difficulties for the EPO's uniform application of the law to justify the referral of any of the four stated Questions under Article 112(1)(b).Footnote 113 It is clear, however, that had the EBA interpreted Article 112(1)(b) in the manner above, this conclusion would not have been sufficient to resolve the case. Rather, the EBA would have also needed to ask whether such differences satisfied the first limb of Article 112(1)(b), and concerned a matter of law of fundamental importance. If they did, the President's request for clarification ought (on its own analysis) to have been valid.
With respect to Question 1, the EBA found that they did. Specifically, it found that the Question related to a point of law of fundamental importance, and that the cases cited by the President in support of its referral disclosed differences on that point by two Boards of Appeal.Footnote 114 By contrast, the EBA rejected the President's referral of Questions 2 to 4, on the basis that the cases cited by the President in support thereof did not disclose a relevant difference of authority.Footnote 115 Assuming that the EBA correctly resolved the semantic ambiguity in the first limb of section 112(1)(b) as requiring diverging if not conflicting decisions,Footnote 116 it ought to follow from this rejection that the President's referral of Questions 2 to 4 would still have been inadmissible on the view of Article 112(1)(b) advanced above.
It is not my purpose to engage with the reasons for the EBA's finding that the cases cited with respect to Questions 2 to 4 did not reveal “different decisions” of relevance in the sense required. Rather, my concern is with the EBA's analysis of Article 52(2) & (3), and its support for the Boards' interpretation of that Article's provisions. It is submitted that even applying the EBA's own principles, such support was inappropriate on several grounds. One is that the Boards' interpretation of Article 52(2) & (3) suffers from the same kind of logical flaw which the EBA attributed to the EPO President. In its opinion, by treating computer program claims and computer implemented method claims “as having an identical scope,”Footnote 117 the President confused “a set of instructions for carrying out steps and the steps themselves”,Footnote 118 and in doing so made an erroneous argument. In the event, the EBA was prepared to forgive the President her allegedFootnote 119 error, and proceed to consider the admissibility of her Question nonetheless.Footnote 120 However, it did not do so without explaining where she went wrong:
There is a general distinction in logic to be made between an object and a name or description of the object. Consider the following argument:
Tigers eat meat. Meat is a word. Therefore tigers eat words.
Clearly there must be an error in this argument. It occurs because “meat” is being used differently in the two premises. In the second premise what is being discussed is not the substance meat, but the name of the substance. These are two different things, and the usual way of distinguishing them is to put the name in quotation marks.Footnote 121
This “tigers eat words” syllogism clearly fails. On the other hand, and as argued in respect of its constitutional analysis above, the EBA's lesson in syllogistic reasoning draws attention to a shortcoming in its own analysis. That shortcoming lies in its support for a theory of Article 52(2) & (3) that has the same “logical weakness” as it attributes to the President in the passage above. Among other things, this makes it difficult to accept the EBA's view of the technical character theory of Article 52(2) & (3) as based on sound (“consistent”) logic.Footnote 122
According to the theory, and as described above,Footnote 123 the list of “non-inventions” contained in Article 52(2) & (3) resolves to a positive requirement for technical character, with the result that if a subject matter has technical character it is an “invention” within the meaning of Article 52(1), and therefore capable of supporting a patent. For it to have technical character, a subject matter need only include some technical feature, regardless of whether that feature dominates.Footnote 124 Further, while the requirement for an invention is expressed as being “essentially separate and independent of” the other patentability requirements,Footnote 125 in determining novelty and inventive step it is only the “technical features” of a subject matter that count.Footnote 126 Implicit in this understanding is the following argument:
Computer programs as such are not inventions. Inventions have technical character. Therefore computer programs as such lack technical character.
However, this argument suffers from the same kind of broken syllogism which the EBA attributes to the President above.
Consistent with the thesis I have elsewhere advanced,Footnote 127 there are additional reasons for criticizing the EBA's support of the Boards' theory of Article 52(2) & (3), including of the computer programs exclusion. One is that that theory is insufficiently supported by the text or history of the EPC, and lacks doctrinal and theoretical coherence. For example, it is difficult to reconcile with the terms of Article 52(2) & (3), which include among their list of non-inventions such apparently technical subject matter as computer programs.Footnote 128 Further, while the drafters of the EPC assumed that inventions would be drawn from the technological arts, there is no evidence that they intended Article 52(2) & (3) to resolve to a requirement for “technical character” alone, and nor – I have argued – do the EPC's object and purpose support that resolution.Footnote 129 Hence my suggestion, that the logical flaws of the Boards' understanding of Article 52(2) & (3) are compounded by its theoretical and doctrinal shortcomings. In addition, that understanding produces an unworkable definition of the “invention”, due to the undefinability and obscurity of the concept of “technical character” itself.Footnote 130 As evidence of this, one need only refer to the extensive litigation involving Article 52(2) & (3), and the judicial decisions (and complaints) which that definition has elicited. As one U.K. Judge recently suggested, despite extensive judicial consideration, the nature of the invention remains depressingly uncertain;Footnote 131 a suggestion supported by the fact and terms of the G_0003/08 referral itself. Finally, and contrary to the implication of the EBA, the inadequacy of the Boards' interpretation of Article 52(2) & (3) is not resolved by their interpretation of the other Article 52(1) patentability criteria, including novelty and inventive step.Footnote 132 This is because the determination of those requirements depends on a prior conception of the subject matter under the description of the “invention”,Footnote 133 which conception depends in turn on an understanding of what constitutes an “invention” itself. It follows that such determination cannot resolve the flaws in that understanding; rather, and as the Boards' decisions reflect, it can only extend their effect into other areas of European patent jurisprudence. And indeed, this is apparent from the reasoning in the Amazon case, in which the TBA followed the EBA-sanctioned approach.Footnote 134 Consider for example the following central passage:
The only relevant criterion for inventive step is the one stated in Article 56 EPC 1973, namely whether the invention is obvious to a person skilled in the art. Since in accordance with established jurisprudence this “art” cannot be a field of business or administration, only elements of the solution falling within the competence of a technically skilled person (here: a programmer or computer scientist) can be taken into account.Footnote 135
Thus, while the Board's de minimis conception of the “invention” supported the treatment of the method as inherently patentable subject matter, it was nonetheless excluded from patentability for failing to advance an art not covered by Article 52(2) & (3). By this approach, the EBA implied, EPO Boards achieve the same result as would be achieved by applying a more robust conception of the “invention” itself: “[m]erely the EPC article applied is different.”Footnote 136 However, this defence of the Board's jurisprudence on Article 52(2) & (3) with reference to its jurisprudence on Article 56 is problematic for at least two reasons. First, it ignores the importance of judicial reasoning to coherent and transparent judicial decision-making.Footnote 137 And second, it ignores the incoherence of the Boards' reasoning on inventive step. That incoherence is apparent from the finding in the case, that while business is not a “technical” art due to the categorization of business methods as non-inventions, computer programming and science are such arts despite their categorization as non-inventions. In reasoning such as this one re-encounters the failed syllogism above.
In conclusion, and for the reasons advanced, it is submitted that the EBA's support of the EPO Boards' jurisprudence on Article 52(2) & (3) was unfortunate. This is because, contrary to the EBA's assertion, that view is not based on solid logic,Footnote 138 has not “created a practicable system for delimiting the innovations for which a patent may be granted”,Footnote 139 and however amenable to clear formulation, is insufficiently coherent to be capable of supporting predictable and verifiable judicial decision-making.Footnote 140 In this sense, it fails the EBA's own test of a constitutionally robust legal order.
IV. Europeanization and Judicial Cooperation
With respect, the EBA's decision fails that test in other ways as well. For example, by treating the European Patent Organisation as having the equivalent status of a national democratic state,Footnote 141 the EBA failed properly to consider its role within the wider European patent system, including the impact of its decisions on national patent laws, and the additional obligations which this imposes, or ought to be regarded as imposing.Footnote 142 That impact derives from the CPC requirement that national European courts interpret their patent legislation in a manner that is consistent with the Boards' interpretation of the EPC,Footnote 143 and is reflected in the Court of Appeal's decision in Symbian Footnote 144 above. In that case, Lord Neuberger referred to the Court's commitment to cooperation with the EPO in denying any substantive difference between U.K. and EPO jurisprudence regarding Article 52(2)(c) & (3), and postponed the development of U.K. law with respect to that Article pending a decision of the EBA.Footnote 145 Among other things, this underlines an important point about Europeanization, which is the unequal burden of national courts to ensure its realization through “cooperation” with the EPO.Footnote 146 Hence my suggestion, that the very fact of this inequality imposes an obligation on European decision makers to take particular care to facilitate the task of national courts by offering detailed reasons for their decisions, whether or not they are bound by common law rules of precedent to do so.Footnote 147 Among other things, this is necessary to ensure transparent judicial analysis and consistent decision making, both of which were recognized as essential in G_0003/08.Footnote 148 It also leads to my conclusion, that by failing to give sufficient weight to the place and role of the Boards in the European patent system, the EBA rendered a decision that is higher on democratic language than democratic content. Given this, and the other factors above, the only hope is for the European or national Legislatures to recognize that “judiciary-driven legal development”Footnote 149 within the EPO has indeed met its limits, with the result – as the EBA itself suggested – that “it is time for the legislator to take over” on the computer programs exclusionFootnote 150 or for a European patent court.
V. The Need for a European Patent Court
Since the introduction of the EPC, the creation of a European patent court has been something of a preoccupation for the European patent community, and the focus of several EU initiatives. In 1989, for example, the 89/695/EEC Agreement relating to Community Patents was introduced, aimed at bringing the CPC and certain supporting instruments into force. Among other things, this Agreement provides for the creation of a Community Patent Appeal Court to hear referrals on questions of infringement and validity from designated national courts, and to hear appeals from the Revocation and Patent Administration Divisions of the EPO. Having been ratified by only 12 States, however, it is accepted as unlikely to enter into force.
In addition to the EEC Agreement, there exists Commission Proposal (EC) 2000/412 for a Council Regulation on the Community Patent.Footnote 151 Under this Proposal, the EU would accede to the EPC, the EPO would grant Community patents, and a Community Intellectual Property Court would exercise exclusive jurisdiction with respect to their validity and infringement. In 2003, two proposals were made for the creation of such a Court. OneFootnote 152 was to confer jurisdiction in respect of certain Community patent disputes on the European Court of Justice (ECJ), and the otherFootnote 153 was to establish a Community Patent Court to exercise the ECJ's jurisdiction on its behalf, a specialized chamber within the Court of First Instance to hear appeals from that Court, and a mechanism for review by the ECJ. In 2004, however, both proposals faltered, following the failure of the EU Competitiveness Council to agree on certain details regarding the first,Footnote 154 and a decision by the ECJ criticizing the second on a range of technical, substantive and institutional grounds.Footnote 155
Despite this, in May 2008 the Community Court initiative was revived with the submission of a Draft Agreement on the European and Community Patents Court to the Working Party on Intellectual Property (Patents) – followed in January 2009 by an amended version.Footnote 156 This Agreement differs from the 2003 initiatives in proposing a Court of First Instance with central and local/regional divisions, a Court of Appeal, and a mechanism for limited review by the ECJ. In March 2009, the EU also sought negotiating directives from the Council to conclude an agreement creating a Unified Patent Litigation System, and on 6 July 2009 the Council asked the ECJ to provide an opinion on the legality of the System. On 2 July 2010, The Advocates General published an opinion that the system is incompatible with EU law. At the time of writing an opinion from the ECJ is still pending.
A European patent court is widely perceived as an essential element of a European patent system. And yet, its creation will come as something of a disappointment, because of the dilution of common law jurisprudence and values it will necessarily involve. It ought also to be possible to make the current system work, and for U.K. and other national courts to play a central role in developing a European patent jurisprudence. This was certainly what the EPC drafters intended, including with respect to Article 52(2)(c) & (3) itself.Footnote 157 Nonetheless, and as the history of that Article's treatment underlines, this has not occurred, due partly to the tendency of national courts to treat the Boards “as if they were a supranational court like the ECJ”;Footnote 158 a tendency which the decision in G_0003/08 seems likely to encourage.
VI. Conclusion
In the 1960s, when the question of patents for computer programs first arose in U.K. law, it caused some initial difficulty for decision makers by challenging the distinction between products and methods, and between inventions and subject matter distinguished by its informational and literary content. However, it was soon decided by the Patents Appeal Tribunal that provided the claim was for the apparatus itself, programmed to operate in a certain way, or for the automated method of achieving a certain end, it would be valid and capable of supporting a patent.Footnote 159 In adopting this approach the Tribunal supported something similar to the In re Bilski “machine or transformation” test,Footnote 160 and rejected the suggestion of the Superintending Examiner that programs were inherently unsuitable for patent protection.Footnote 161
According to the EPO, the computer programs exclusion in Article 52(2)(c) & (3) gives effect to this distinction. However, on this view that exclusion adds nothing to the other Article 52(2) & (3) exclusions, including those of presentations of information and aesthetic works.Footnote 162 Further, the technical character theory of the invention on which the view depends has implications beyond computer programs themselves, including for business methods and biotechnological subject matter.Footnote 163 For these reasons alone, it is submitted that the EPO's approach to Article 52(2) & (3) is an inappropriate one, and that if the computer programs exclusion is regarded as undesirable, the appropriate course is to seek its legislative amendment.
