patentability of computer software
作者:何放 (伦敦政治经济学院)       本站发布时间:2003-5-17 10:19:59

Whether Europe needs to follow the US approach in allowing for the patentability of computer software

By He Fang 何放,  London School of Economics and Political Science

Licensed PRC attorney-at-law and trademark agent

As shown in the recent Proposal for a Directive on the Patentability of Computer-implemented Inventions of 20 February 20, 2002 (the “Proposal” hereafter), question about whether Europe should follow America’s approach in allowing for patentability of computer software gives rise to rather hot debates. In the above Proposal, the European Commission aims to harmonize the national laws relating to “Computer-implemented” inventions. [1] However, people too often are mistaking:

- Computer-implemented inventions (computer related inventions)

- Computer programs (software)

The two above issues do not correspond to two different expressions of a single idea. It should first be noted that the first one does not raise any problems any more, neither in the US, since the Diamond decision of the Supreme Court of 1981 [2] , nor with EPO, since the VICOM decision T 208/84 (OJ 1/1987) and the KOCH & STERZEL decision T 26/86 (OJ 1/1988 ) of the Boards of Appeal 3.5.1 and 3.4.1, respectively. Nevertheless, the Proposal only aims to harmonize the computer-implemented inventions at the commission level while still leaves “ Computer programs as such” untouched. It is the argued that the protection of computer software inventions is far behind the US.

The commonly accepted rationale of the justification of the patent system is that it creates incentives for innovation. [3] As Lord Oliver said in Asahi the “underlying purpose of the patent system is the encouragement to the public the inventor receives the benefit of a period of monopoly during which he becomes entitle to prevent others from performing his invention except by his licence.” [4] It is generally believed that that the world’s current software industry is dominated by most US-based companies while their small and media-sized European rivals are lobbying for the limit of “pure” computer programs in the EU jurisdiction. Under US jurisdiction, patents may be granted for any new and useful process, machine, manufacture or composition of matter. This kind of broad definition of subject matter has made the US take a lead in widening the scope of subject matter for which patent protection may be obtained. It is not a coincidence that the US is also in a leading position in the software industry over the past twenty years over its European counterparts. It is worth mentioning that the US’s attitude to the patentablility of computer programs has changed with the development of its software industry in the past decades. When the first computer was invented in 1964, computer programs only served as the ancillaries to computer hardware. However, the world soon saw the computer programs were so sophisticatedly developed that the hardware now began to serve as their ancillaries to solve specific problems. Before the anti-trust investigation on IBM in 1970, computer software was often sold with computer hardware without separate prices. Then the IBM decided to separately sold its hardwares and softwares. Afterwards, the 1970s saw a tendency to separate investment and selling in computer hardwares and softwares while the industry began to invest more and more in the development of computer softwares. It is quite natural for the software industry to demand improved intellectual property protection for their huge investment.

Although the US Supreme Court rejected Benson’s patent application on the processing of programmed conversion of numerical information [5] , they did not deny the possibility of the patentability of computer software. In the landmark case Diamond v. Diehr, ways were firstly open to include the computer softwares within the US patent system. After the Diamond v. Diehr, the commonly believed unpatentable “algorithm” was required to be interpreted restrictively so as to encompass only procedures intend to solve a mathematical problem [6] .  The USPTO’s Guideline also stated that as long as it is novel, a computer program that causes a computer to function in a particular fashion would be protected.

However, this is not the case in Europe. As it was shown in earlier cases of computer related patent applications [7] , many of them were rejected because the only novelty and inventiveness reside in the software elements, which consist of “algorithm” or “mental act”. Computer program “as such” is still prohibited as a patentable subject matter by the Article 52 of the EPC although the door to computer related inventions seemed to have been opened through the leading EPO decision of Vicom [8] . The computer program “as such” exclusion was de facto obsolete provided it can produce a “technical contribution”. As reiterated in the recent draft of the Proposal, most European legislations and case laws still insist on a “ technical character” or “technical contribution” in the patentable subject matters.

It is deeply doubted that whether the “technical contribution” requirement is really necessary for European patentable subject matters. This requirement is probably the biggest obstacle to exclude the world’s largest industry from the most important and relevant form of intellectual property law.

First,  “the technical contribution” necessity lacks legislative foundation. As a matter of fact, the introduction of technical character as de facto non-statutory requirement for patentablility owes its origin to an imaginative interpretation of Article 52 of the EPC [9] because all the categories in Article 52(2) have in common is that they are non-technical. There is a short inductive leap to conclude from those exclusions that the term “invention” relates to inventions of a technical nature. As a result, this conclusion was reinforced by the Rules and Guidelines of the EPC that clearly state that in order for an invention to be patentable it must be technical. [10] It is argued that such requirement does not conform to the Article 27of TRIPs which states that “patents shall be available for any inventions, whether products or processes, in all fields of technology” and that “patents shall be … enjoyable without discrimination as to … field of technology”. Moreover, TRIPs Article 9(2) states that copyright protection shall not apply to “methods of operations or mathematical concepts as such”.

Actually the Board of Appeal of EPC has noted this point and stated “as the EPO was not a signatory to TRIPs, they were not bound by it. However, on the basis that TRIPs aimed at ‘setting common standards and it acted as an indicator of modern trends, the Board noted that it was the clear intention of TRIPs not to exclude from patentablility any inventions, in particular, not to exclude programs for computers” [11] . It is worth mentioning that the 2002 Proposal has also realized this and stated that “Article 3, in the context of Recital 6, reflects Article 27(1) of the TRIPS Agreement, according to which patents shall be available for any inventions, whether products or processes, in all fields of technology, provided that they are new, involve an inventive step and are susceptible of industrial application.” According to the Proposal, a computer-implemented invention is defined as belonging to a field of technology. Meanwhile, “Technical contribution” is defined to mean a contribution to the state of the art in a technical field which is not obvious to a person skilled in the art in the Proposal. However, an algorithm which is defined without reference to a physical environment does not meet the definition of “computer-implemented invention” and does not fall within a field of technology. Nevertheless, the 2002 Proposal still have not clearly enough defined what “Technical contribution” is and it aims to “codify” the requirement of a technical contribution.

Second, the “technical contribution” requirement excludes many “pure” software applications and adds uncertainty of the outcome of many such applications. As discussed the above, because this requirement is only an extensive interpretation of the EPC in the matter of computer programs, it lacks judicial clarity. Thus, to answer when does a computer program have a technical character, the EPC and many national patent authorities developed subtle approaches and led to uncertainty of computer-related inventions to some extent.

In the Vicom case [12] , the application related to the use of a normal computer for image processing purposes was first rejected on the ground both that it sought protection for a computer program and on the basis that it related to a mathematical method. But later the Board of Appeal held that “claims which can be considered as being directed to a computer set up to operate in accordance with a specified program (whether by means of hardware or software) for controlling or carrying out a technical process cannot be regarded as relating to a computer program…” “…an invention which would be patentable in accordance with conventional patentability criteria should not be excluded from protection by the mere fact that for its implementation modern technical means in the form of a computer program are used. Decisive is what technical contribution the invention as defined in the claim when considered as a whole makes to the known art. [13]

In the landmark UK decision of Merrill Lynch [14] , the Court of Appeal held that a data processing system operating to produce a novel technical result would normally be patentable although the claimed the computer program was to produce novel effects on a general-purpose computer.

In Gale’s Application [15] , The Mr. Gale had developed a new algorithm and sought a patent in respect of and invention entitled “Improvements and means whereby a binary manipulative system may derive a square root”. In order to outcome the obvious obstacle of unpatentability, Mr. Gale described the operation of the mathematical process as implemented on a ROM chip. Although rejected at first, Gale’s application were finally granted because “The effect of Mr. Gale’s chip was, he stated, to allow computers to calculate square roots both faster and to a great degree of accuracy than hitherto been possible.” [16] However, it might be quite inequitable that “the same effect might equally well have been achieved using software and would certainly have been unpatentable in this form prove fatal to the application.” [17]

In Fujitsu’s application [18] , the invention sought to virtually depict and manipulate crystal structures on a computer screen. The effect would be to allow the analysis of the properties of new compounds without the need to create these in the real world. Nevertheless, the application was finally rejected because the courts did not think the “virtual reality” depicted by the invention is technical. It is somehow quite a paradoxical decision because conventional machine which makes a process more efficient deserves patent protection while the more complicated “virtual reality” which did the same thing does not. In this application, distinction drawn with the situation in VICOM by the courts only exacerbates rather than clarifies the issues what is required to produce a technical effect. [19]

    As shown in the above cases, the current “technical contribution” requirement is so vague that it lacks the normal function of predictability and certainty of laws. Under current situations, the fate of a patent application will be improved if it can be shown that it leads to or produces a physical change in things since there has no clear guidance been given as to what the tribunals mean by a “technical” invention. [20]  

Third, the “technical contribution” requirement for the patentability of the world’s most important industry has a tendency to strangle the innovation and the competition ability of the software industry. In Europe, such requirement not only hinders the software invention application of titanic companies like IBM or Microsoft, it also unnecessarily elevate the software patent threshold of small and medium size enterprise, on which they may rely more than their counterparts of standard makers do to survive in the competitions.

After the collapse of the “dot com” economy, most operators of venture capitals become much more cautious in choosing their investment objects. A single idea or concept is not as attractive as what they did six years ago. Thus, to hold a profitable software patent becomes one of the few assets for small and medium size companies to fight for attracting venture capitals, which most of them desire for the further development of their enterprise. However, as the cases discussed above, the “technical contribution” requirement de facto has become an obstacle for many software per se inventions while the innovation in the European software industry is in the progress. But most of the innovation in the field cannot be protected in the patent law jurisdiction.

Fearing the refusal of their software patent application, many applicants had to conceal their computer program per se in a technically dressed form such as associating the computer program with a hardware add-on. However, this kind of methods is not always feasible although the EPC had become more liberal to the computer-related patent application. Then most applicants will have to experience long and cost consuming appeal process if they decide to stick on their innovation. As a result, many small and medium size companies are reluctant to apply for their innovation although this innovation appears to be patentable if only the novelty, inventiveness and applicability are considered. Many of them resort to the technical encryption or are reluctant to invest much in the developing seemingly unpatentable innovation.

While the US have been and are still improving their patent system to remove the business method exception from the USPTO Guidelines, and to have their patent statute extended to “anything under the sun that is made by man”, the Europe, despite some attempts and even an extensive interpretation of the EPC in the matter of computer programs, still refuses to abandon the “technical contribution” requirement. As was said in the US case of Lotus v Paperback: “It is no accident that the world’s strongest software industry is found in the United States, rather than in some other jurisdiction which provides weaker protection for computer programs. [21] ” If the “technical character” which was required in the 2002 Proposal for a directive becomes a directive after all, Europe will inevitably stand on a lag- behind position both in respect of software industry and its commitment in Article 27 of TRIPs.


[1] See Proposal For a Directive of The European Parliament and of The Council on the patentability of computer-implemented inventions 2002/0047 (COD).

[2]  US Supreme Court, Diamond v. Dier.

[3] See E. Kitch, “An Economic Review of the Patent System” (1977) 20 Journal of Law & Economics 265

[4] Asahi Kasei Kogyo [ 1991 ] RPC 485, 523 ( Lord Oliver ) (HL).

[5] Gottschalk v. Benson 409 U.S. 63 (1972)

[6] Paine, Webber, Jackson and Curtis, Inc. v Merill Lynch, Pierce, Fenner and Smith, Inc. 564 F Supp 1358 (1983)

[7] e.g. Badger Co Inc’s Application [1970] RPC 36

[8] [1987] 2 EPOR 74

[9] Intellectual Property Law, Lionel Berntly & Brad Sherman, 2001, Oxford

[10] EPC r. 27 (1) (b); EPC r 27 (1) (d); EPC r. 29 (1). EPO Guidelines C-IV, 1.2 (ii)

[11] IBM/Computer programs T1173/97 [2000] EPOR224-5

[12] [1987] 2EPOR 74

[13] Ibid

[14] [1989] RPC 561, reported at first instance at [1988] RPC 1

[15] [1991] RPC 305

[16] Ibid

[17] Ibid

[18] [1996] RPC 511

[19] Information Technology Law, Ian J Lloyd, Butterworths.

[20] Intellectual Property Law, Lionel Bently & Brad Sherman, Oxford

[21] 740 F Supp 37 (1990)

( 说明:文中的观点或信息与本网站主办单位无关)