Versió actual de l'article 52.

La conferència de revisió de l'CPE de 11/2000 proposava insertar "en tots els camps de la teconològia" al paragraf 1 i esborrar el paragraf 4.

Article 52: Invencions patentables

Les patents europees hauràn de ser concedides als invents [ en tots els camps de la tecnològia ], sempre i quan aquests siguin nous, representin un avenç o sig+uin susceptibles de la seva aplicació industrial.
Els següent particulars no poden ser reconeguts com a invencions dins dels significat del paragraf 1:
1. descobriments, teories cièntifiques i metodes matemàtics;
2. creacions estètiques;
3. esquemes, normes i mètodes per portar a terme actes mentals, jugar a jocs o fer negocis, i programes per ordinadors;
4. presentacions d'informació.
Les directrius del paragraf 2 han d'excloure la patentibilitat dels temes substancials o activitats referides en aquestes directrius només en tant que una sol·licitud de patent europea o patent europea tingui relació amb aquests temes o activitats com a tals.
Els metodes per al tractament del cos humà o animal, per a cirugia, terapia i els metodes diagnòstics practicats al cos humà o animal no podràn esser considerats com a invencions que siguin susceptibles de la seva aplicació en la industria en el sentit del paragraf 1. Aquesta directriu no pot èsser aplicada a productes, en particular substàncies o compostos, fets per utilitzar-se en els metodes anteriorment esmentats.

The Initial Interpretation and its Erosion

Until the late 80s, this was unanimously interpreted as clearly excluding software patents, as they are usually understood in the discussion today. E.g. in 1990 the Technical Board of Appeal of the EPO explains its refusal of 1984 to allow a document processing system on the basis of Art 52.2c:

The reason given for the refusal was that the contribution to the art resided solely in a computer program as such within the meaning of Article 52 EPC, paragraphs 2(c) and Consequently, this subject-matter was not a patentable invention within the meaning of Article 52(1) EPC, in whatever form it was claimed.

In arriving at this conclusion the Examining Division argued on the basis that Claims 1 and 2 related to a method for automatically abstracting and storing an input document in an information storage and retrieval system and Claims 3-6 to a corresponding method for retrieving a document from the system. The claims specifically referred to a dictionary memory, input means, a main memory and a processor. These hardware elements were classical elements of an information and retrieval system (...) and objectionable under Article 54(2) EPC as lacking novelty. According to the present description (...) the method steps were implemented by programming such a classical system. The claimed combination of steps did not imply an unusual use of the individual hardware elements involved. The claims merely defined a collocation of known hardware and new software concerned with document information to be stored but not with an unexpected or unconventional way of operating the known hardware. The differences between the prior art and the subject-matter of the present application were defined by functions to be realised by a computer program which was used to implement a particular algorithm, or mathematical method, for analysing a document. In other words the steps of the method defined operations which were based on the content of the information and were independent of the particular hardware used.

In other words, a collocation of standard computing hardware with new computing rules (algorithms), in whatever form it is presented in the claim, would be excluded from patentability.

This was also clearly expressed in the Examination Guidelines of the European Patent Office of 1978.

However, in 1985 the Guidelines were revised and in particular the limits of patentability with respect to programs for computers were blurred. In two decisions of 1986, the EPO's Technical Board of Appeal reinterpreted the list of exclusions to mean that only "non-technical" innovations should be excluded, but refused to define "technical" -- a concept that was not mentioned in the law. From thereon the EPO embarked on a slippery slope by gradually widening the meaning of what could be understood to be "technical".

The EPO's reinterpretation of 1985/1986 and the subsequent loosening were criticised by law scholars such as Krasser, Benkard i Vivant and have led to a schism of judicial practise, which a new EU directive is supposed to overcome.

Veieu també BPatG 2002-03-26: Suche fehlerhafter Zeichenketten, Melullis 2002: Zur Sonderrechtsfähigkeit von Computerprogrammen i BGH-Entscheidung Betriebssystem 1990 : Entscheidung des 1. Zivilsenats des BGH zur Abgrenzung von Urheber- und Patentrecht: ein Betriebssystem ist kein technisches Programm

The decisions at the EPO were understood to have been taken "in response to pressure from the computer industry and trends emerging in the US".

Hk-cityu-is-euswpat
Bernhardt & Kraßer 1986: Lehrbuch des Patentrechts: German manual of patent law of 1986, explains the correct interpretation of Art 52 EPC, as used by the German courts, and explains that the German courts are thereby resisting "pressure from the software industry". Krasser also mentions the revision of the EPO's examination guidelines in 1985 and explains that they, while still unclear, seem to be moving into the direction demanded by "the industry".

Nova versió de l'article 52 d'acord amb la proposta del l'OEP de 2000.

A la conferència diplomatica que va tenir lloc el novembre de l'any 2000, la OEP va intentar esborrar tot rastre de les definicions restrictives en termes com ara invenció, tecnicitat, aplicabilitat industrial, etc de la llei i en el seu lloc obrir el camí per a la patentatibilitat de totes les solucions pràctiques i repetibles de problemes. Això ha permés a la OEP de formular una proposta molt curta:

Invencions patentables

Les patents europees hauràn de ser concedides als invents [ en tots els camps de la tecnològia ], sempre i quan aquests siguin nous, representin un avenç o sig+uin susceptibles de la seva aplicació industrial.

As a result of an uproar of public opinion, politicians from major countries prevented this planned change of Art 52. Yet the "Base Proposal" version was accepted as a new wording for Art 52(1), and Art 52(4) was deleted (whereby the concept of "industrial application" was further weakened).

Thus the revised version of Art 52 EPC, which is not yet in force, contains the TRIPs formula "in all fields of technology", but fails to define the new term "technology", which doesn't exist in the old EPC. Thus clause (2) seems relativised by an indeterminate concept from an international treaty. It would have been in the interest of clarity and legal security to concretise this concept, e.g. by explaining clearly what is to be understood by a "technical invention" and why algorithms, business methods and rules for operating known data processing equipment do not belong into this category. Instead the legislators opted for introducing indeterminate concepts and potential contradictions into the law, which are then likely to be resolved by putting the appositive "as such" from 52(3) in quotes to make it appear mysterious and unclear, so as to allow the patent judiciary to rely on its own favored interpretation of "fields of technology" or even to point to alleged WTO constraints, thus giving up the clarity and integrity of national law in favor of arbitrary decisionmaking by the international patent lawyer community.

Art 52(4) about surgery on the human body was "only" reworded and moved to Art 53. This however implies that surgery methods are no longer considered to be non-inventions, non-technical or non-industrial. In this way, the Diplomatic Conference further weakened the TRIPs concepts on which it decided to rely for limiting patentability.

The Amended Directive and the EPC

La regulació legal actual sobre els límits de la patentatibilitat és clara i inequívoca. Hi ha, no obstant això, tribunals que consideren aquesta norma inadequada i l'han substituït per una regulació diferent en anticipació d'un canvi de la llei.

En réalité, les règles nationales et conventionnelles sont claires: elles posent sans équivoque un principe de non-brevetabilité du logiciel. Le jeu qui se joue aujourd'hui consiste à contourner d'une manière ou d'une autre celles-ci, par exemple en imaginant de considérer, comme on l'a vu, l'ensemble constitué par le matériel et le logiciel comme une machine virtuelle susceptible (demain ...) d'être breveteée. À ce compte-là, on peut parler brevets. Les brevets susceptibles d'être ainsi obtenus, par ce canal ou un autre, n'ont, toutefois, que la valeur qu'on leur prête - mais il ne faut pas écarter l'hypothèse selon laquelle on finirait par une sorte de consensus à ne pas vraiment la discuter. De fait, l'efficacité de ce countournement des règles légales sera largement fonction du fait qu'un tel consensus se dégagera pour accepter --- contre les règles positives --- que ce nouveau jeu se joue ou non. La question ne se situe plus sur le terrain juridique stricto sensu.

Després d'un intens debat públic resulta que la norma legal actual és l'adequada i que la jurisprudència recent de l'OEP contradiu tan la llei com l'interés públic. Cal cridar els tribunals a corregir la seva pràctica actual i aplicar la llei.

The European Parliament has passed an amended directive which reconfirms the system of Art 52 EPC and makes it more explicit. Frits Bolkestein and some people in the Council do not like this clarification and propose to opt instead for a revision of the EPC or some other kind of inter-governmental agreement. The UK Patent Office has proposed to rewrite Art 52(3) in a way that allows anything deemed "technical" to be patented. On the other hand it would also be possible to concretise Art 52 EPC itself further in the spirit of the amended directive. Putting positive definitions of "technical field", "technology", "industry" etc, as found in the amended directive, into Art 52ff EPC or its national versions could become a way of implementing the directive.

Proposem opcionalment d'esborrar l'apartat 3 de l'article 52 perquè és purament explicatiu i no agrega cap significat a l'article. El seu lloc és al reglament d'examen de sol·licituds de patents. Suprimir-lo de la llei seria una manera convenient de dir als tribunals que tornin a la interpretació correcta de la llei, que era la predominant durant els anys 70/80.

Enllaços anotats

Auslegung von Art. 52 des Europäischen Patentübereinkommens hinsichtlich der Frage, inwieweit Software patentierbar ist: Dr. Karl Friedrich Lenz, Professor für Deutsches Recht und Europarecht an der Universität Aoyama Gakuin in Tokio untersucht mit den allgemein anerkannten Methoden juristischer Auslegung, welche Bedeutung dem heute geltenden Text des Art 52 EPÜ beizumessen ist und gelangt zu dem Schluss, dass die Technischen Beschwerdekammern des EPA seit einiger Zeit regelmäßig Patente auf Programme für Datenverarbeitungsprogrammen als solche erteilen und eine beunruhigende Bereitschaft zeigen, die eigenen Wertungen an die Stelle der Wertungen des Gesetzgebers zu setzen.
Moses, the Ten Exclusions from Patentability and "stealing with a further ethical effect": Computer programs are both unpatentable and patentable in Europe. How did the European Patent Office's Technical Boards of Appeal gradually manage to patent the unpatentable? Where taboos and artificially induced complexity mine the road, satiric comparison is often the fastest way to a thourough understanding.
Stalin & Software Patents: In this contribution to the European Patent Office (EPO) mailing list, a European patent attorney cites the EPO Examination Guidelines of 1978 as clear documentary evidence for the intention of the legislator to keep computer programs on any storage medium free from any claims to the effect that their distribution, sale or use could infringe on a patent. But the European Patent Office's Technical Board of Appeal (TBA) apparently considered itself to be a kind of modern Stalin, an ultimate sources of wisdom in matters of whatever complexity, standing high above the legislator and the peoples of Europe, and even above the EPO's own institutions for judicial review. In this way the TBA risks to antagonise the public, to create harmful legal insecurity especially for small patent-holders and to severely damage the delicate process of building confidence in international institutions. The TBA should see itself as a conservator rather than an innovator.
Scandinavia: even without the "as such" clause, stealing can have a further legal effect: Art 52 of the European Patent Convention (EPC) stipulates that programs for computers as well as mental rules, mathematical methods, ways of presenting information etc are not patentable inventions and may therefore not be claimed as such. The wording "as such" from Art 52(3) has however been used to undo all explicit limits on patentability. The European Patent Office (EPO) has in 1997 begun to subdivide computer programs into two groups, "as-such programs" and "not-as-such programs", and has tried to justify this by historical claims about how art 52 EPC came about. This reasoning is at odds with grammar as well as with history. One particularly nice set of evidence comes from Scandinavia: the Danish and Swedish national versions of art 52 EPC do not literally render Art 52(3) but rather incorporate its meaning in the first line of their version of Art 52(2), coming to exactly the same common sense conclusions to which independent grammatical analyses have come and which are also stated in the early EPO examination guidelines: that a "program as such" is "something that constitutes only a program". It confirms that Art 52(3) merely exhorts examiners to look carefully where the novel achievement really lies, e.g. in a programming solution or in a chemical process which may happen to run under program control.
Why can't I patent my movie (as such)?: A comment on software patents, actually.
European Patent Convention: Text of EPO edition
Art 52 EPC: Patentable Inventions: full text from the EPO web server
EPC 172: Right of Governments to Act against EPO: The contracting states are responsible for shortcomings of the EPO.
Berichte der Münchener Diplomatischen Konferenz über die Einführung eines Europäischen Patenterteilungsverfahrens: In autumn 1973 patent experts of the european governments convened in Munich for 1 month in order to create a unified european patent examination system. This conference led to the conclusion of the European Patent Convention (EPC) and to the creation of the European Patent Office (EPO). Art 52 EPC excludes programs for computers, mental rules, mathematical methods etc from patentability. This principle was further elaborated by the EPO's examination guidelines of 1978 and the initial court practise. However starting in 1986 judges at the EPO and some national courts started to extend the scope of patentability and render the exclusions of Art 52 EPC meaningless. In order to justify this, they used a teleologic and historic method of law interpretation which makes frequent reference to what the legislators in 1973 allegedly meant or did not understand. Therefore we have dug out the relevant documents and taken a look at the (relatively short) account about the negotiations concerning Art 52 EPC. This text offers no support for the EPO's method of interpretation. Quite to the contrary.
EPO 1978: Examination Guidelines: Adopted by the President of the European Patent Office in accordance with EPC 10.2a with effect from 1978-06-01. Excerpts concerning the question of technical invention, limits of patentability, computer programs, industrial application etc.
EPO decision T 22/85 against IBM archival system: A Technical Board of Appeal of the European Patent Office (EPO) rejects a patent application which is directed to a program for computers. In 1984, the EPO's examiners had rejected the patents based on the original Examination Guidelines of 1978, saying that the claims referred to a "program for computers". The appellant argued on the basis of newer Guidelines and caselaw that his claims are directed to technical effects and not a program as such. The Board of Appeal rejects the appeal by arguing indirectly that the use of general-purpose computer hardware does not confer technicity on an abstract method: "Abstracting a document, storing the abstract, and retrieving it in response to a query falls as such within the category of schemes, rules and methods for performing mental acts and constitutes therefore non-patentable subject-matter under Article 52 EPC" and "The mere setting out of the sequence of steps necessary to perform an activity, excluded as such from patentability under Article 52 EPC, in terms of functions or functional means to be realised with the aid of conventional computer hardware elements does not import any technical considerations and cannot, therefore, lend a technical character to that activity and thereby overcome the exclusion from patentability."
EPO TBA 2002/03 T 49/99: information modelling not technical, computer-implementation not new: In March 2002, a Technical Board of Appeal at the European Patent Office (EPO) rejects a patent application for a computerised information modelling system on the grounds that the subject matter is not an invention according to Art 52 EPC. The Board argues largely in the original spirit of the EPO and differs significantly from some other recent EPO caselaw. This is an important reason why industrial patent lawyers are pressing for new patentability legislation. Under a CEC/McCarthy directive, EPO decisions such as this one would no longer be possible.
Regulació sobre el concepte d'invent dels sistema europeu de patents i la seva interpretació amb especial consideració als programes per a ordinadors: Proposem que el legislador projecti alguna regulació sobre patentabilitat de programari a través de les línies del següent curt i clar texte.
Patent Jurisprudence on a Slippery Slope: So far computer programs and other rules of organisation and calculation are not patentable inventions according to European law. This doesn't mean that a patentable manufacturing process may not be controlled by software. However the European Patent Office and some national courts have gradually blurred the formerly sharp boundary between material and immaterial innovation, thus risking to break the whole system and plunge it into a quagmire of arbitrariness, legal insecurity and dysfunctionality. This article offers an introduction and an overview of relevant research literature.
M. Vivant: Le Recours a la Propriété Industrielle: The article on patents analyses the history of French and European patent jurisdiction. It explains, why the European parliaments decided in the 60-70s against patentability of computer programs and how French courts supported this decision by some very clear verdicts even against software innovations related to the control of industrial production processes. It also explains how the European Patent Office since 1986 gradually deviated from these clear rules in 5 steps of logic-twisting. It warns however that the patents gratend by the EPO are of incertain value and could be negated by any national judge.
Dr. Swen Kiesewetter - Köbinger 2000: Über die Patentprüfung von Programmen für Datenverarbeitungsanlagen: Ein Patentprüfer zeigt die Ungereimtheiten der Prüfung von Software-Anmeldungen auf. In ihrem Bemühen, ein Gesetz umzuinterpretieren, welches unmissverständlich die Patentierung von Datenverarbeitungsprogrammen verbietet, hat die Rechtsprechung im Laufe der Jahre Funktionsansprüche zugelassen, die es dem Anmelder erlauben, ein Programm zu verkleiden. Aber diese Funktionsansprüche stellen eher Probleme als Lösungen dar, und die Lösung zu diesen Problemen besteht in einem (nicht patentierbaren) Datenverarbeitungsprogramm (als solchem). Probleme zu patentieren ist aber noch weniger zulässig und in seinen Auswirkungen noch bedenklicher als Programme zu patentieren.
Patentfähige Datenverarbeitungsprogramme - ein Widerspruch in sich: Dr. König, patent attorney from Düsseldorf, points out inconsistencies in the software patent caselaw of the EPO and BGH, criticises "circular conclusions" and argues that the EPO has "done violence to art 52 EPC". Through a "grammatical interpretation" of "programs for computers as such" he finds that this can refer only to all kinds of computer programs without exception, as far as they are claimed alone. The EPC of 1973, transcribed into German law in 1978, no longer allows a distinction between technical and untechnical programs. However it ist still possible to patent program-related combination inventions, which then have to be examined for technicity, novelty, non-obviousness and industrial applicability. Courts have often shown more imagination in "helping themselves over the obstacles of art 52". There is an elegant indirect way to effectively grant full patent protection for computer programs as such while avoiding the recent incoherences of the EPO and BGH jurisdiction. As parts of combination inventions, programs for computers may, just as is frequently the case with discoveries and scientific theories, enjoy full "usage protection", if their distribution can be construed as contributory infringement of a combination invention.
Europarl 2003-09-24: Amended Software Patent Directive: Consolidated version of the amended directive "on the patentability of computer-implemented inventions" for which the European Parliament voted on 2003-09-24.
Noël Mamère 2002-20-28: Let's just delete the "As Such" clause!: Proposes to delete Art 52(3) EPC.
CEC & BSA trying to impose unlimited patentability on Sweden: In a statement submitted to the Swedish Ministry of Justice on behalf of SSLUG, a group of 6100 programmers and users of free software in the area around Copenhagen and Malmö, Erik Josefsson shows how an influential group at the European Commission and the European Patent Office has eroded the standards of patentability and is trying to impose a regime of patentability on all achievements of the human mind that can help to solve some practical problem. This influential group has also, by overstretching the competence of the EPO's Technical Boards of Appeal, illegally overruled the Swedish courts and damaged the Swedish constitutional order. Even in their most recent decisions in the mid-nineties, the Swedish courts did not agree with the EPO's illegal practice, but now the European Commission is set to force this practice on Sweden by means of "european harmonisation". It was the duty of the EPO to abide by a role of "cold harmonisation" in the first place: act as a conservative follower and summarizer of national caselaw rather than as an innovative trendsetter pursuing its own agenda. Josefsson cites ample examples of patents granted by the EPO and rejected by Swedish courts.

Derzeitige Version des Art. 52

Die EPÜ-Revisionskonferenz von 2000/11 schlug vor, den Zusatz "auf allen Gebieten der Technik" in Abs 1 einzufügen und Abs 4 zu streichen.

Artikel 52: Patentfähige Erfindungen

Europäische Patente werden für Erfindungen [auf allen Gebieten der Technik] erteilt, sofern sie neu sind, auf einer erfinderischen Tätigkeit beruhen und gewerblich anwendbar sind.
Als Erfindungen im Sinn des Absatzes 1 werden insbesondere nicht angesehen:
1. Entdeckungen sowie wissenschaftliche Theorien und mathematische Methoden;
2. ästhetische Formschöpfungen;
3. Pläne, Regeln und Verfahren für gedankliche Tätigkeiten, für Spiele oder für geschäftliche Tätigkeiten sowie Programme für Datenverarbeitungsanlagen;
4. die Wiedergabe von Informationen.
Absatz 2 steht der Patentfähigkeit der in dieser Vorschrift genannten Gegenstände oder Tätigkeiten nur insoweit entgegen, als sich die europäische Patentanmeldung oder das europäische Patent auf die genannten Gegenstände oder Tätigkeiten als solche bezieht.
Verfahren zur chirurgischen oder therapeutischen Behandlung des menschlichen oder tierischen Körpers und Diagnostizierverfahren, die am menschlichen oder tierischen Körper vorgenommen werden, gelten nicht als gewerblich anwendbare Erfindungen im Sinn des Absatzes 1. Dies gilt nicht für Erzeugnisse, insbesondere Stoffe oder Stoffgemische, zur Anwendung in einem der vorstehend genannten Verfahren.

Die ursprüngliche Auslegung und ihre Aufweichung

Bis Ende der 80er Jahre wurde dies einhellig als klarer Ausschluss von Software-Patenten verstanden, wie sie heute diskutiert werden. Beispielsweise erklärt die technische Beschwerdekammer des EPA dessen Weigerung von 1984, ein Dokumentverarbeitungssystem zuzulassen, auf Grundlage von Art. 52.2c:

Als Begründung der Ablehnung wurde angeführt, dass der Beitrag zum Stand der Technik allein in einem Computerprogramm als solchem im Sinne von Artikel 52 Absatz 2(c) EPÜ bestand und dass demzufolge der Antragsgegenstand keine patentierbare Erfindung im Sinne von Artikel 52 Absatz 1 EPÜ sei, in welcher Form auch immer er beansprucht werde.

Zur Urteilsfindung ging die Prüfungsabteilung davon aus, dass Ansprüche 1 und 2 sich auf eine Methode bezogen, ein Dokument von dem System abzurufen. Die Ansprüche verwiesen insbesondere auf einen Wörterbuchspeicher, Eingabevorrichtungen, Hauptspeicher und einen Prozessor. Diese Rechnerbestandteile waren klassische Elemente eines Informations- und Ablagesystems [...] und gemäß Artikel 54(2) EPÜ wegen Mangels an Neuigkeit zu beanstanden. Nach der vorgelegten Beschreibung [...] wurden die Verfahrensschritte durch Programmierung eines solch klassischen Systems implementiert. Die beanspruchte Kombination der Schritte ergab keinen ungewöhnlichen Gebrauch der einzelnen Rechnerbestandteile. Die Ansprüche legen ein Zusammenspiel bekannter Hardware und neuer Software fest, das sich damit befasst, Dokumentinformationen zu speichern, aber nicht mit einer unerwarteten oder ungewöhnlichen Weise, die bekannte Hardware zu betreiben. Die Unterschiede zwischen dem Stand de r Technik und dem Anspruchsgegenstand der vorliegenden Anmeldung wurden durch Funktionen definiert, die durch ein Computerprogramm zu realisieren waren, das dazu verwendet wurde, einen bestimmten Algorithmus oder eine mathematische Methode zur Dokumentenanalyse zu implementieren. Mit anderen Worten, die Verfahrensschritte definierten Operationen, die auf dem Inhalt der Information beruhten und unabhängig von der jeweiligen Hardware waren.

Mit anderen Worten, ein Zusammenspiel gewohnter Rechner-Hardware mit neuen Rechenregeln (Algorithmen), in welcher Form auch immer sie in den Ansprüchen dargestellt werden, wäre von der Patentierbarkeit ausgeschlossen.

Dies wurde ebenfalls klar in den Prüfungsrichtlinien des Europäischen Patentamts von 1978 ausgedrückt.

Allerdings wurden 1985 die Prüfungsrichtlinien überarbeitet, und insbesondere die Grenzen der Patentierbarkeit hinsichtlich Computerprogrammen wurden verwischt. In zwei Entscheidungen von 1986 erklärte die technische Beschwerdekammer des EPA die Liste der Ausschlussgegenstände dahingehend, dass nur "nichttechnische" Neuerungen ausgeschlossen sein sollten, weigerte sich aber, "technisch" zu definieren -- ein Begriff, der im Gesetz nicht vorkam. Von da an begab sich das EPA auf einen Schlitterkurs, indem es den Bereich dessen, was als "technisch" verstanden werden könnte, immer weiter ausdehnte.

Die Neuauslegung durch das EPA von 1985/1986 und die nachfolgende Aufweichung wurden von Rechtsgelehrten wie Krasser, Benkard und Vivant kritisiert und haben zu einer Spaltung der Rechtspraxis geführt, deren Überwindung von einer neuen EU-Richtlinie erhofft wird.

Siehe auch BPatG 2002-03-26: Suche fehlerhafter Zeichenketten, Melullis 2002: Zur Sonderrechtsfähigkeit von Computerprogrammen und BGH-Entscheidung Betriebssystem 1990 : Entscheidung des 1. Zivilsenats des BGH zur Abgrenzung von Urheber- und Patentrecht: ein Betriebssystem ist kein technisches Programm

Die Entscheidungen am EPA wurden als "Antwort auf Druck der Computer-Industrie und der sich anbahnenden Entwicklungen in den USA" angesehen.

Hk-cityu-is-euswpat
Bernhardt & Kraßer 1986: Lehrbuch des Patentrechts: Deutsches Handbuch des Patentrechts von 1986, erklärt die korrekte Interpretation von Art. 52 EPÜ, wie sie von deutschen Gerichten verwendet wird, und erklärt, dass die deutschen Gerichte damit einem "Druck von der Software-Industrie" widerstehen würden. Krasser erwähnt auch die Überarbeitung der Prüfungsrichtlinien des EPA von 1985 und erklärt, dass sie, noch in unklarem Zustand, in die von "der Industrie" geforderte Richtung gehen würden.

Neue Version des Art. 52 laut EPA-Basisvorschlag von 2000

Das EPA möchte auf der Diplomatischen Konferenz im November 2000 alle Reste einer Definition der Begriffe Erfindung, Technizität, industrielle Anwendbarkeit usw. aus dem Gesetz tilgen und stattdessen die grenzenlose Patentierbarkeit aller praktischen und wiederholbaren Problemlösungen verankern. Das hat es dem EPA ermöglicht, seinen Vorschlag sehr kurz zu halten:

Patentfähige Erfindungen

Europäische Patente werden für Erfindungen [auf allen Gebieten der Technik] erteilt, sofern sie neu sind, auf einer erfinderischen Tätigkeit beruhen und gewerblich anwendbar sind.

Der darauf erfolgende Aufschrei in der Öffentlichkeit brachte Politiker der großen Länder dazu, diese vorgesehene Änderung von Art. 52 zu verhindern. Allerdings wurde die Version des "Basisvorschlags" zur Neuformulierung von Art. 52(1) akzeptiert, und Art. 52(4) wurde gestrichen (wodurch der Begriff der "gewerblichen Anwendbarkeit" an Substanz verlor).

Daher enthält die Neufassung von Art. 52 EPÜ, die noch nicht in Kraft ist, die TRIPs-Formel "auf allen Gebieten der Technik", lässt aber eine Definition des neuen Begriffs "Technik" vermissen, der im alten EPÜ nicht vorkommt. Somit scheint Absatz (2) relativiert zu werden durch einen unbestimmten Begriff aus einem internationalen Vertrag. Es wäre im Interesse an Klarheit und Rechtssicherheit, diesen Begriff zu konkretisieren, beispielsweise indem klar dargelegt wird, was unter einer "technischen Erfindung" zu verstehen ist und warum Algorithmen, Geschäftsmethoden und Regeln zum Betreiben bekannter Datenverarbeitungsanlagen nicht zu dieser Kategorie gehören. Stattdessen entschieden die Gesetzgeber sich dafür, unbestimmte Begriffe und denkbare Widersprüche in das Gesetz einzuführen, die dann eher dadurch aufgelöst werden, dass man die Beifügung "als solche" in 52(3) in Anführungszeichen setzt, um sie geheimnisvoll und unklar erscheinen zu lass en und so der Patentgerichtsbarkeit zu erlauben, ihre eigene Lieblingsinterpretation von "Gebiet der Technik" anzuwenden oder sogar auf angebliche WTO-Erfordernisse zu verweisen und damit die Klarheit und Stimmigkeit nationalen Rechts zu Gunsten willkürlicher Entscheidungen der internationalen Patentanwaltsgemeinschaft aufzugeben.

Art. 52(4) über Chirurgie am menschlichen Körper wurde "nur" umformuliert und in Art. 53 verschoben. Dies bedeutet allerdings, dass chirurgische Methoden nicht mehr als Nicht-Erfindungen, nichttechnisch oder nichtgewerblich angesehen werden. Auf diese Weise schwächte die diplomatische Konferenz wiederum die TRIPs-Begrifflichkeiten, auf die sie sich zur Begrenzung der Patentierbarkeit zu stützen entschieden hatte.

Die geänderte Richtlinie und Art. 52 EPÜ

Die derzeitige Gesetzesregel über die Grenzen der Patentierbarkeit ist klar und eindeutig. Es gibt jedoch einige Gerichte, welche diese Regel für unzweckmäßig halten und sie im Vorgriff auf eine angestrebte Gesetzesänderung verworfen haben. Wie Prof. Michel Vivant 1998 schreibt:

In Wirklichkeit sind die Gesetzesregeln des Übereinkommens und der nationalen Gesetze klar: Sie fordern unmissverständlich die Nicht-Patentierbarkeit von Software. Das Spiel, das heute gespielt wird, besteht darin, in einer oder der anderen Weise diese Regeln zu verdrehen, z.B. indem man sich, wie oben beschrieben, die Gesamtheit aus Hardware und Software als eine virtuelle Maschine denkt, die (künftig ...) patentierbar sein könnte. Unter dieser Voraussetzung kann man dann patentrechtlich argumentieren. Die auf diese Weise auf dem einen oder anderen Wege erhältlichen Patente haben allerdings nur denjenigen Wert, den man ihnen beimisst -- oder der sich durch einen Konsens ergibt, dieser Frage nicht genauer nachgehen zu wollen. Tatsächlich kann die Verdrehung der Gesetzesregeln nur insoweit Wirkung entfalten, wie sich ein Konsens darüber herstellen lässt, ob man dieses Spiel gegen die bestehenden Gesetzesregeln spielen soll oder nicht. Hierbei handelt es sich nicht mehr um eine juristische Frage im strengen Sinne.

Nach einer intensiven öffentlichen Debatte hat sich herausgestellt, dass die derzeit bestehende Gesetzesregel zweckmäßig ist und das davon abweichende neuere EPA-Fallrecht nicht dem öffentlichen Interesse entspricht. Die Gerichte sind aufgerufen, ihre Praxis zu korrigieren und das Gesetz anzuwenden.

Das Europäische Parlament hat eine geänderte Richtlinie angenommen, die das System von Art. 52 EPÜ bekräftigt und es genauer ausführt. Frits Bolkestein und einige Leute im Rat mögen diese Klarstellung nicht und schlagen vor, lieber das EPÜ abzuändern oder eine andere Art der zwischenstaatlichen Vereinbarung zu finden. Das UK-Patentamt hat vorgeschlagen, Art. 52(3) so umzuschreiben, dass alles, was man für "technisch" zu halten beliebt, patentiert werden kann. Andererseits wäre es auch möglich, Art. 52 EPÜ selbst im Sinne der vom Parlament geänderten Richtlinie zu konkretisieren. Positive Definitionen von "technischem Gebiet", "Technik", "Industrie" usw., wie sie in der geänderten Richtlinie zu finden sind, in Art. 52ff EPÜ oder dessen nationale Versionen einzubringen, könnte ein Weg sein, die Richtlinie umzusetzen.

Wir schlagen optional die Streichung von Art. 52 Abs (3) vor, weil dieser Absatz nur erläutert, was sich von selbst verstehen sollte. Er gehört nicht in das Gesetz sondern in die Prüfungsrichtlinien. Durch eine Streichung kann die Rechtsprechung auf relativ unkomplizierte Weise veranlasst werden, zu einer korrekten Gesetzesauslegung zurückzukehren, wie in den 1970/80er Jahren vorherrschte.

Kommentierte Verweise

Auslegung von Artikel 52 des Europäischen Patentübereinkommens hinsichtlich der Frage, inwieweit Software patentierbar ist: Dr. Karl Friedrich Lenz, Professor für Deutsches Recht und Europarecht an der Universität Aoyama Gakuin in Tokio untersucht mit den allgemein anerkannten Methoden juristischer Auslegung, welche Bedeutung dem heute geltenden Text des Art 52 EPÜ beizumessen ist und gelangt zu dem Schluss, dass die Technischen Beschwerdekammern des EPA seit einiger Zeit regelmäßig Patente auf Programme für Datenverarbeitungsprogrammen als solche erteilen und eine beunruhigende Bereitschaft zeigen, die eigenen Wertungen an die Stelle der Wertungen des Gesetzgebers zu setzen.
Moses, die Zehn Patentierungsverbote und das "Stehlen mit einem weiteren ethischen Effekt": Programme für Datenverarbeitungsanlagen sind heute in Europa unpatentierbar und patentierbar zugleich. Wie gelang es den Technischen Beschwerdekammern des Europäischen Patentamtes im Laufe der Zeit, das unpatentierbare zu patentieren? Bei komplizierten Themen bietet oft der satirische Vergleich den schnellsten Zugang zum gründlichen Verständnis.
Stalin & Software Patents: In this contribution to the European Patent Office (EPO) mailing list, a European patent attorney cites the EPO Examination Guidelines of 1978 as clear documentary evidence for the intention of the legislator to keep computer programs on any storage medium free from any claims to the effect that their distribution, sale or use could infringe on a patent. But the European Patent Office's Technical Board of Appeal (TBA) apparently considered itself to be a kind of modern Stalin, an ultimate sources of wisdom in matters of whatever complexity, standing high above the legislator and the peoples of Europe, and even above the EPO's own institutions for judicial review. In this way the TBA risks to antagonise the public, to create harmful legal insecurity especially for small patent-holders and to severely damage the delicate process of building confidence in international institutions. The TBA should see itself as a conservator rather than an innovator.
Skandinavien: auch ohne "als solches" Klausel kann Diebstahl einen weiteren rechtlichen Effekt haben: Art 52 of the European Patent Convention (EPC) stipulates that programs for computers as well as mental rules, mathematical methods, ways of presenting information etc are not patentable inventions and may therefore not be claimed as such. The wording "as such" from Art 52(3) has however been used to undo all explicit limits on patentability. The European Patent Office (EPO) has in 1997 begun to subdivide computer programs into two groups, "as-such programs" and "not-as-such programs", and has tried to justify this by historical claims about how art 52 EPC came about. This reasoning is at odds with grammar as well as with history. One particularly nice set of evidence comes from Scandinavia: the Danish and Swedish national versions of art 52 EPC do not literally render Art 52(3) but rather incorporate its meaning in the first line of their version of Art 52(2), coming to exactly the same common sense conclusions to which independent grammatical analyses have come and which are also stated in the early EPO examination guidelines: that a "program as such" is "something that constitutes only a program". It confirms that Art 52(3) merely exhorts examiners to look carefully where the novel achievement really lies, e.g. in a programming solution or in a chemical process which may happen to run under program control.
Why can't I patent my movie (as such)?: A comment on software patents, actually.
Europäisches Patentübereinkommen: Text der EPA-Version
Art. 52 EPÜ: Patentfähige Erfindungen: Volltext auf dem Webserver des EPA
EPÜ 172: Eispruchsrecht der Staaten: Die Mitgliedsstaaten sind für Fehler des EPA verantwortlich.
Berichte der Münchener Diplomatischen Konferenz von 1973: 1973 konferierten Patentreferenten der europäischen Regierungen einen ganzen Monat lang in München, um ein einheitliches europäisches Patenterteilungsverfahren einzuführen. Diese Konferenz führte zum Abschluss des Europäischen Patentübereinkommens (EPÜ) und zur Gründung des Europäischen Patentamtes (EPA). In Art 52 EPÜ werden Programme für Datenverarbeitungsanlagen von der Patentierbarkeit ausgeschlossen. Im Anschluss waren Softwareinnovationen einige Jahre lang tatsächlich nicht patentierbar, aber seit etwa 1986 haben Richter des EPA und einiger nationaler Gerichte den Ausschluss der Programme für Datenverarbeitungsanlagen Schritt für Schritt ausgehöhlt. Dabei übergingen sie den Gesetzeswortlaut und wandten eine gewagte teleologische und historische Methode der Gesetzesauslegung an. Zur Stützung der historischen Auslegung wurden im wesentlichen auf die Konferenzberichte verwiesen. Daher haben wir uns dieses Dokument per Fernleihe besorgt und uns den (relativ kurzen) Bericht über die Verhandlungen zu Artikel 52 näher angeschaut. Man lese und staune.
EPA 1978: Prüfungsrichtlinien: Adopted by the President of the European Patent Office in accordance with EPC 10.2a with effect from 1978-06-01. Excerpts concerning the question of technical invention, limits of patentability, computer programs, industrial application etc.
EPO 1990: T 0022/85: Eine technische Beschwerdekammer des Europäischen Patentamtes (EPA) weist eine Patentanmeldung zurück, die sich auf ein Programm für Datenverarbeitungsanlagen richtet. 1984 hatten die Prüfer des EPA die Anmeldung aufgrund der ursprünglichen Prüfungsrichtlinien von 1978 zurückgewiesen und festgestellt, dass die Ansprüche sich auf ein Programm für Datenverarbeitungsanlagen bezogen. Der Beschwerdeführer argumentierte auf der Grundlage neuerer Richtlinien und Gerichtsentscheidungen, dass seine Ansprüche sich auf technische Wirkungen und nicht auf ein Programm als solches richteten. Die Beschwerdekammer weist die Berufung zurück, indem sie indirekt argumentiert, dass der Gebrauch einer Allzweck-Datenverarbeitungsanlage einer abstrakten Methode keine Technizität verleiht: "Das Zusammenfassen eines Dokumentes, Speichern der Zusammenfassung und Abrufen derselben als Antwort auf eine Anfrage fällt als solche in die Kategorie der Pläne, Regeln und Verfahren für gedankliche Tätigkeiten und stellt damit einen nicht patentierbaren Gegenstand gemäß Artikel 52 EPÜ dar" und "Die bloße Darlegung der Folge von Schritten, die nötig sind, eine Handlung durchzuführen, die als solche von der Patentierbarkeit gemäß Artikel 52 EPÜ ausgenommen ist, in Begriffen von Funktionen oder funktionellen Vorrichtungen, die mit herkömmlichen Bestandteilen einer Datenverarbeitungsanlage realisiert werden sollen, führt keine technischen Betrachtungen ein und kann deswegen dieser Handlung keine technische Eigenschaft verleihen und damit den Ausschluss von der Patentierbarkeit überwinden."
EPO TBA 2002/03 T 49/99: information modelling not technical, computer-implementation not new: In March 2002, a Technical Board of Appeal at the European Patent Office (EPO) rejects a patent application for a computerised information modelling system on the grounds that the subject matter is not an invention according to Art 52 EPC. The Board argues largely in the original spirit of the EPO and differs significantly from some other recent EPO caselaw. This is an important reason why industrial patent lawyers are pressing for new patentability legislation. Under a CEC/McCarthy directive, EPO decisions such as this one would no longer be possible.
Gesetzesregel über den Erfindungsbegriff im Europäischen Patentwesen und seine Auslegung unter besonderer Berücksichtigung der Programme für Datenverarbeitungsanlagen: Wir schlagen dem Gesetzgeber vor, beim Entwurf einer Richtlinie zur Frage der Patentierbarkeit von Software auf dem folgenden kurzen und klaren Text aufzubauen.
Patentjurisprudenz auf Schlitterkurs -- der Preis für die Demontage des Technikbegriffs: Bisher gehören Computerprogramme ebenso wie andere Organisations- und Rechenregeln in Europa nicht zu den patentfähigen Erfindungen, was nicht ausschließt, dass ein patentierbares Herstellungsverfahren durch Software gesteuert werden kann. Das Europäische Patentamt und einige nationale Gerichte haben diese zunächst klare Regel jedoch immer weiter aufgeweicht. Dadurch droht das ganze Patentwesen in einem Morast der Beliebigkeit, Rechtsunsicherheit und Funktionsuntauglichkeit zu versinken. Dieser Artikel gibt eine Einführung in die Thematik und einen Überblick über die rechtswissenschaftliche Fachliteratur.
Lamy Droit Informatique 1998: Der Artikel über Patente analysiert die Geschichte der französischen und europäischen Patentrechtsprechung. Er erklärt, warum die europäischen Parlamente und Gerichte sich in den 60-70er Jahren gegen die Patentierbarkeit von Computerprogrammen und Programmierideen entschieden und wie das Europäische Patentamt diese Entscheidungen seit 1986 schrittweise rückgängig gemacht hat. Er warnt aber, dass auf diese Weise vom EPA gewährte Patente in der Praxis von ungewissem Wert sind.
Dr. Swen Kiesewetter - Köbinger 2000: Über die Patentprüfung von Programmen für Datenverarbeitungsanlagen: Ein Patentprüfer zeigt die Ungereimtheiten der Prüfung von Software-Anmeldungen auf. In ihrem Bemühen, ein Gesetz umzuinterpretieren, welches unmissverständlich die Patentierung von Datenverarbeitungsprogrammen verbietet, hat die Rechtsprechung im Laufe der Jahre Funktionsansprüche zugelassen, die es dem Anmelder erlauben, ein Programm zu verkleiden. Aber diese Funktionsansprüche stellen eher Probleme als Lösungen dar, und die Lösung zu diesen Problemen besteht in einem (nicht patentierbaren) Datenverarbeitungsprogramm (als solchem). Probleme zu patentieren ist aber noch weniger zulässig und in seinen Auswirkungen noch bedenklicher als Programme zu patentieren.
König 2001: Patentfähige Datenverarbeitungsprogramme - ein Widerspruch in sich: Der Düsseldorfer Patentanwalt Dr. König zeigt allerlei Ungereimtheiten in der Softwarepatent-Rechtsprechung des BGH und EPA auf, kritisiert "Zirkelschlüsse" und argumentiert, das EPA habe Art 52 EPÜ "Gewalt angetan". Durch eine "grammatische Auslegung des Begriffs "Datenverarbeitungsprogramme als solche"" gelangt er zu der Erkenntnis, dass damit nur alle Datenverarbeitungsprogramme ohne Ausnahme gemeint sein können, allerdings nur insoweit sie alleine Gegenstand des Patentbegehrens seien. Seit der Umsetzung des EPÜ in die PatG-Novelle von 1978 gibt es keine Grundlage mehr für eine Unterscheidung zwischen technischen und untechnischen Programmen. Beim Verständnis des Begriffes "DV-Programm" habe sich die Rechtsprechung nach allgemeinen außerjuristischen Definitionen zu richten, wie sie etwa von DIN-Fachleuten geleistet wurden. Danach ist ein Programm die Gesamtheit aus Sprachdefinitionen und Text und nicht etwa nur der urheberrechtlich schützbare Text. Das EPÜ/PatG erlaube es aber, "Kombinationserfindungen" zu patentieren, die als ganzes auf Technizität, Neuheit, Nichtnaheliegen und gewerbliche Anwendbarkeit zu prüfen seien. Gerichte hätten schon immer Kreativität entfaltet, wenn es darum ging, "sich über Patentierbarkeitsausschlüsse hinwegzuhelfen". Datenverarbeitungsprogrammen (als solchen) könne auf dem Wege über ein Kombinationspatent mittelbar der volle Schutz eines Sachpatents zukommen. Datenverarbeitungsprogramme seien ebenso wie etwa naturwissenschaftliche Entdeckungen (als solche) dem Verwendungsschutz zugänglich.
Europarl 2003-08-24: Geänderte Softwarepatent-Richtlinie: Konsolidierte Version der wesentlichen Bestimmungen (Art 1-6) der geänderten Richtlinie "über die Patentierbarkeit computer-implementierter Erfindungen", für die das Europäische Parlament am 24. September 2003 gestimmt hat.
Noël Mamère 2002-20-28: weg mit der "als-solche"-Klausel!: Schlägt Streichung von Art. 52(3) vor.
CEC & BSA trying to impose unlimited patentability on Sweden: In a statement submitted to the Swedish Ministry of Justice on behalf of SSLUG, a group of 6100 programmers and users of free software in the area around Copenhagen and Malmö, Erik Josefsson shows how an influential group at the European Commission and the European Patent Office has eroded the standards of patentability and is trying to impose a regime of patentability on all achievements of the human mind that can help to solve some practical problem. This influential group has also, by overstretching the competence of the EPO's Technical Boards of Appeal, illegally overruled the Swedish courts and damaged the Swedish constitutional order. Even in their most recent decisions in the mid-nineties, the Swedish courts did not agree with the EPO's illegal practice, but now the European Commission is set to force this practice on Sweden by means of "european harmonisation". It was the duty of the EPO to abide by a role of "cold harmonisation" in the first place: act as a conservative follower and summarizer of national caselaw rather than as an innovative trendsetter pursuing its own agenda. Josefsson cites ample examples of patents granted by the EPO and rejected by Swedish courts.

Current Version of Art 52

The European Patent Convention (EPC) was signed by its core member states in 1973 and went into force in 1978, when the European Patent Office (EPO) was established on its basis. At their revision conference in 2000, the EPC member states proposed to insert "in all fields of technology" into Art 52(1) and to delete paragraph Art 52(4).

Article 52: Patentable Inventions

European patents shall be granted for inventions [ in all fields of technology ], as far as they are new, involve an inventive step and are susceptible of industrial application.
The following in particular shall not be regarded as inventions within the meaning of paragraph 1:
1. discoveries, scientific theories and mathematical methods;
2. aesthetic creations;
3. schemes, rules and methods for performing mental acts, playing games or doing business, and programs for computers;
4. presentations of information.
The provisions of paragraph 2 shall exclude patentability of the subject-matter or activities referred to in that provision only to the extent to which a European patent application or European patent relates to such subject-matter or activities as such.
Methods for treatment of the human or animal body by surgery or therapy and diagnostic methods practised on the human or animal body shall not be regarded as inventions which are susceptible of industrial application within the meaning of paragraph 1. This provision shall not apply to products, in particular substances or compositions, for use in any of these methods.

The Initial Interpretation and its Erosion

In other words, a collocation of standard computing hardware with new computing rules (algorithms), in whatever form it is presented in the claim, would be excluded from patentability.

This was also clearly expressed in the Examination Guidelines of the European Patent Office of 1978.

The EPO's reinterpretation of 1985/1986 and the subsequent loosening were criticised by law scholars such as Krasser, Benkard and Vivant and have led to a schism of judicial practise, which a new EU directive is supposed to overcome.

The decisions at the EPO were understood to have been taken "in response to pressure from the computer industry and trends emerging in the US".

Hk-cityu-is-euswpat
Bernhardt & Kraßer 1986: Lehrbuch des Patentrechts -- Recht der Bundesrepublik Deutschland, Europäisches und Internationales Patentrecht: German manual of patent law of 1986, explains the correct interpretation of Art 52 EPC, as used by the German courts, and explains that the German courts are thereby resisting "pressure from the software industry". Krasser also mentions the revision of the EPO's examination guidelines in 1985 and explains that they, while still unclear, seem to be moving into the direction demanded by "the industry".

New Version of Art 52 according to the EPO's Base Proposal of 2000

At the Diplomatic Conference in november 2000, the EPO sought to remove all traces of restricting definitions of "invention" from the Law and instead open the way for patentability of all practical and repeatable problem solutions. This has allowed the EPO to formulate a very short proposal.

Patentable Inventions

European patents shall be granted for inventions in all fields of technology, as far as they are new, involve an inventive step and are susceptible of industrial application.

The Amended Directive and the EPC

The current legal regulation about the limits of patentability is, contrary to what its violators say, clear and unambiguous. There are however lawcourts which consider this regulation inadequate and have replaced it by a different regulation in anticipation of a change of law. As Prof. Michel Vivant writes in 1998:

In reality, the national and conventional rules are clear: they stipulate without ambiguity a principle of non-patentability of software. The game which is being played today consists in twisting these rules one way or another, e.g. by imagining to consider, as we have seen, the totality of software and hardware as a virtual machine which is potentially patentable (tomorrow ...). From that point on one can speak about software in patent language. The patents which may be obtained this way, by this channel or by another, however still do not have any value beyond what we lend to them - but of course it is possible that they will finally acquire a value simply through an informal consensus to stop discussing the question. In fact, the efficiency of this twisting of rules of law is largely dependent on whether this consensus evolves to take for granted -- against the rules of written law -- that we will play this game or not. This question is no longer a legal question in the strict sense of the term.

After an intensive public debate it has turned out that the current legal rule is adequate and that recent EPO caselaw is at odds with both the law and the public interest. The courts are called upon to correct their practise and apply the law.

Any rewriting would probably entail a change in Art 52(3), since this has been used to make the law appear unclear. If rewritten in the spirit of the amended directive, Art 52(3) might be removed altogether, because it is to be considered purely explanatory. A "program as such" is a "program as a program" (in contrast with "a program as implementation feature of an claim object whose inventive part is not a program"). It stresses a self-evident differentiation which belongs in the Examination Guidelines. Deleting it from the law would be a convenient way of telling lawcourts to return to the correct interpretation of the law, which was predominant during the 1970/80s.

Annotated Links

Interpretation of art 52 of the European Patent Convention in view of the question, to what extent software is patentable: Dr. Karl Friedrich Lenz, professor for German and European Law at Aoyama Gakuin University in Tokyo, investigates using the various universally accepted methods of law interpretation which meaning has to be attributed to the text of art 52 EPC today and reaches the conclusion that the Technical Boards of Appeal of the European Patent Office have for some time now regularly granted patents on programs for computers as such and are showing a disturbing willingness to substitute their own value judgements for those given by the legislator.
Moses, the Ten Exclusions from Patentability and "stealing with a further ethical effect": Computer programs are both unpatentable and patentable in Europe. How did the European Patent Office's Technical Boards of Appeal gradually manage to patent the unpatentable? Where taboos and artificially induced complexity mine the road, satiric comparison is often the fastest way to a thourough understanding.
Stalin & Software Patents: In this contribution to the European Patent Office (EPO) mailing list, a European patent attorney cites the EPO Examination Guidelines of 1978 as clear documentary evidence for the intention of the legislator to keep computer programs on any storage medium free from any claims to the effect that their distribution, sale or use could infringe on a patent. But the European Patent Office's Technical Board of Appeal (TBA) apparently considered itself to be a kind of modern Stalin, an ultimate sources of wisdom in matters of whatever complexity, standing high above the legislator and the peoples of Europe, and even above the EPO's own institutions for judicial review. In this way the TBA risks to antagonise the public, to create harmful legal insecurity especially for small patent-holders and to severely damage the delicate process of building confidence in international institutions. The TBA should see itself as a conservator rather than an innovator.
Scandinavia: even without the "as such" clause, stealing can have a further legal effect: Art 52 of the European Patent Convention (EPC) stipulates that programs for computers as well as mental rules, mathematical methods, ways of presenting information etc are not patentable inventions and may therefore not be claimed as such. The wording "as such" from Art 52(3) has however been used to undo all explicit limits on patentability. The European Patent Office (EPO) has in 1997 begun to subdivide computer programs into two groups, "as-such programs" and "not-as-such programs", and has tried to justify this by historical claims about how art 52 EPC came about. This reasoning is at odds with grammar as well as with history. One particularly nice set of evidence comes from Scandinavia: the Danish and Swedish national versions of art 52 EPC do not literally render Art 52(3) but rather incorporate its meaning in the first line of their version of Art 52(2), coming to exactly the same common sense conclusions to which independent grammatical analyses have come and which are also stated in the early EPO examination guidelines: that a "program as such" is "something that constitutes only a program". It confirms that Art 52(3) merely exhorts examiners to look carefully where the novel achievement really lies, e.g. in a programming solution or in a chemical process which may happen to run under program control.
Why can't I patent my movie (as such)?: A comment on software patents, actually.
European Patent Convention: Text of EPO edition
Art 52 EPC: Patentable Inventions: full text from the EPO web server
EPC 172: Right of Governments to Act against EPO: The contracting states are responsible for shortcomings of the EPO.
Reports about the Munich Diplomatic Conference of 1973: In autumn 1973 patent experts of the european governments convened in Munich for 1 month in order to create a unified european patent examination system. This conference led to the conclusion of the European Patent Convention (EPC) and to the creation of the European Patent Office (EPO). Art 52 EPC excludes programs for computers, mental rules, mathematical methods etc from patentability. This principle was further elaborated by the EPO's examination guidelines of 1978 and the initial court practise. However starting in 1986 judges at the EPO and some national courts started to extend the scope of patentability and render the exclusions of Art 52 EPC meaningless. In order to justify this, they used a teleologic and historic method of law interpretation which makes frequent reference to what the legislators in 1973 allegedly meant or did not understand. Therefore we have dug out the relevant documents and taken a look at the (relatively short) account about the negotiations concerning Art 52 EPC. This text offers no support for the EPO's method of interpretation. Quite to the contrary.
EPO 1978: Examination Guidelines: Adopted by the President of the European Patent Office in accordance with EPC 10.2a with effect from 1978-06-01. Excerpts concerning the question of technical invention, limits of patentability, computer programs, industrial application etc.
EPO 1990: T 0022/85: A Technical Board of Appeal of the European Patent Office (EPO) rejects a patent application which is directed to a program for computers. In 1984, the EPO's examiners had rejected the patents based on the original Examination Guidelines of 1978, saying that the claims referred to a "program for computers". The appellant argued on the basis of newer Guidelines and caselaw that his claims are directed to technical effects and not a program as such. The Board of Appeal rejects the appeal by arguing indirectly that the use of general-purpose computer hardware does not confer technicity on an abstract method: "Abstracting a document, storing the abstract, and retrieving it in response to a query falls as such within the category of schemes, rules and methods for performing mental acts and constitutes therefore non-patentable subject-matter under Article 52 EPC" and "The mere setting out of the sequence of steps necessary to perform an activity, excluded as such from patentability under Article 52 EPC, in terms of functions or functional means to be realised with the aid of conventional computer hardware elements does not import any technical considerations and cannot, therefore, lend a technical character to that activity and thereby overcome the exclusion from patentability."
EPO TBA 2002/03 T 49/99: information modelling not technical, computer-implementation not new: In March 2002, a Technical Board of Appeal at the European Patent Office (EPO) rejects a patent application for a computerised information modelling system on the grounds that the subject matter is not an invention according to Art 52 EPC. The Board argues largely in the original spirit of the EPO and differs significantly from some other recent EPO caselaw. This is an important reason why industrial patent lawyers are pressing for new patentability legislation. Under a CEC/McCarthy directive, EPO decisions such as this one would no longer be possible.
Regulation about the invention concept of the European patent system and its interpretation with special regard to programs for computers: We propose that the legislator draft any regulations on the question of software patentability along the lines of the following short and clear text.
Patent Jurisprudence on a Slippery Slope -- the price for dismantling the concept of technical invention: So far computer programs and other rules of organisation and calculation are not patentable inventions according to European law. This doesn't mean that a patentable manufacturing process may not be controlled by software. However the European Patent Office and some national courts have gradually blurred the formerly sharp boundary between material and immaterial innovation, thus risking to break the whole system and plunge it into a quagmire of arbitrariness, legal insecurity and dysfunctionality. This article offers an introduction and an overview of relevant research literature.
Lamy Droit Informatique 1998: The article on patents analyses the history of French and European patent jurisdiction. It explains, why the European parliaments decided in the 60-70s against patentability of computer programs and how French courts supported this decision by some very clear verdicts even against software innovations related to the control of industrial production processes. It also explains how the European Patent Office since 1986 gradually deviated from these clear rules in 5 steps of logic-twisting. It warns however that the patents gratend by the EPO are of incertain value and could be negated by any national judge.
Kiesewetter-Köbinger 2000: Über die Patentprüfung von Programmen für Datenverarbeitungsanlagen: A patent examiner analyses the inconsistencies of examining and granting patents on computer programs. In an attempt to reinterpret the law, which clearly prohibits the granting of patents on computer programs, jurisprudence has gradually allowed the granting of functional claims which allow the applicant to disguise a computer program. But these functional claims represent problems rather than solutions, the solution being a (non-patentable) computer program as such. Patenting problems is however even less permissible and even more harmful in its consequences.
König 2001: Patentfähige Datenverarbeitungsprogramme - ein Widerspruch in sich: Dr. König, patent attorney from Düsseldorf, points out inconsistencies in the software patent caselaw of the EPO and BGH, criticises "circular conclusions" and argues that the EPO has "done violence to art 52 EPC". Through a "grammatical interpretation" of "programs for computers as such" he finds that this can refer only to all kinds of computer programs without exception, as far as they are claimed alone. The EPC of 1973, transcribed into German law in 1978, no longer allows a distinction between technical and untechnical programs. However it ist still possible to patent program-related combination inventions, which then have to be examined for technicity, novelty, non-obviousness and industrial applicability. Courts have often shown more imagination in "helping themselves over the obstacles of art 52". There is an elegant indirect way to effectively grant full patent protection for computer programs as such while avoiding the recent incoherences of the EPO and BGH jurisdiction. As parts of combination inventions, programs for computers may, just as is frequently the case with discoveries and scientific theories, enjoy full "usage protection", if their distribution can be construed as contributory infringement of a combination invention.
Europarl 2003-09-24: Amended Software Patent Directive: Consolidated version of the amended directive "on the patentability of computer-implemented inventions" for which the European Parliament voted on 2003-09-24.
Noël Mamère 2002-20-28: Let's just delete the "As Such" clause!: Proposes to delete Art 52(3) EPC.
CEC & BSA trying to impose unlimited patentability on Sweden: In a statement submitted to the Swedish Ministry of Justice on behalf of SSLUG, a group of 6100 programmers and users of free software in the area around Copenhagen and Malmö, Erik Josefsson shows how an influential group at the European Commission and the European Patent Office has eroded the standards of patentability and is trying to impose a regime of patentability on all achievements of the human mind that can help to solve some practical problem. This influential group has also, by overstretching the competence of the EPO's Technical Boards of Appeal, illegally overruled the Swedish courts and damaged the Swedish constitutional order. Even in their most recent decisions in the mid-nineties, the Swedish courts did not agree with the EPO's illegal practice, but now the European Commission is set to force this practice on Sweden by means of "european harmonisation". It was the duty of the EPO to abide by a role of "cold harmonisation" in the first place: act as a conservative follower and summarizer of national caselaw rather than as an innovative trendsetter pursuing its own agenda. Josefsson cites ample examples of patents granted by the EPO and rejected by Swedish courts.

Current Version of Art 52

第52条: Patentable Inventions

European patents shall be granted for inventions [ in all fields of technology ], as far as they are new, involve an inventive step and are susceptible of industrial application.
The following in particular shall not be regarded as inventions within the meaning of paragraph 1:
1. discoveries, scientific theories and mathematical methods;
2. aesthetic creations;
3. schemes, rules and methods for performing mental acts, playing games or doing business, and programs for computers;
4. presentations of information.
The provisions of paragraph 2 shall exclude patentability of the subject-matter or activities referred to in that provision only to the extent to which a European patent application or European patent relates to such subject-matter or activities as such.
Methods for treatment of the human or animal body by surgery or therapy and diagnostic methods practised on the human or animal body shall not be regarded as inventions which are susceptible of industrial application within the meaning of paragraph 1. This provision shall not apply to products, in particular substances or compositions, for use in any of these methods.

The Initial Interpretation and its Erosion

In other words, a collocation of standard computing hardware with new computing rules (algorithms), in whatever form it is presented in the claim, would be excluded from patentability.

This was also clearly expressed in the Examination Guidelines of the European Patent Office of 1978.

The EPO's reinterpretation of 1985/1986 and the subsequent loosening were criticised by law scholars such as Krasser, Benkard 及び Vivant and have led to a schism of judicial practise, which a new EU directive is supposed to overcome.

又 BPatG 2002-03-26: Suche fehlerhafter Zeichenketten, Melullis 及び BGH-Entscheidung Betriebssystem 1990 : Entscheidung des 1. Zivilsenats des BGH zur Abgrenzung von Urheber- und Patentrecht: ein Betriebssystem ist kein technisches Programm も參照

The decisions at the EPO were understood to have been taken "in response to pressure from the computer industry and trends emerging in the US".

Dr Matthew Lee: Patentability of Software-Related Inventions
Bernhardt & Kraßer 1986: Lehrbuch des Patentrechts: German manual of patent law of 1986, explains the correct interpretation of Art 52 EPC, as used by the German courts, and explains that the German courts are thereby resisting "pressure from the software industry". Krasser also mentions the revision of the EPO's examination guidelines in 1985 and explains that they, while still unclear, seem to be moving into the direction demanded by "the industry".

Interpretation of art 52 of the European Patent Convention in view of the question, to what extent software is patentable: Dr. Karl Friedrich Lenz, professor for German and European Law at Aoyama Gakuin University in Tokyo, investigates using the various universally accepted methods of law interpretation which meaning has to be attributed to the text of art 52 EPC today and reaches the conclusion that the Technical Boards of Appeal of the European Patent Office have for some time now regularly granted patents on programs for computers as such and are showing a disturbing willingness to substitute their own value judgements for those given by the legislator.
モーセス，特許性の十戒及び「さらなる倫理的效果をもった窃盗行為」: 現在の欧州では計算機プログラムは特許性が無いが、有る、と見なされている。欧州特許庁の技術控訴部がどうなさって数十年間で法理上で特許できないものを事実上で許特できるものに変換することに成功したのでしょうか？難しい問題を理解するにあたって「諷刺的比較」はしばしば深い理解への近道である。
Stalin & Software Patents: In this contribution to the European Patent Office (EPO) mailing list, a European patent attorney cites the EPO Examination Guidelines of 1978 as clear documentary evidence for the intention of the legislator to keep computer programs on any storage medium free from any claims to the effect that their distribution, sale or use could infringe on a patent. But the European Patent Office's Technical Board of Appeal (TBA) apparently considered itself to be a kind of modern Stalin, an ultimate sources of wisdom in matters of whatever complexity, standing high above the legislator and the peoples of Europe, and even above the EPO's own institutions for judicial review. In this way the TBA risks to antagonise the public, to create harmful legal insecurity especially for small patent-holders and to severely damage the delicate process of building confidence in international institutions. The TBA should see itself as a conservator rather than an innovator.
Scandinavia: even without the "as such" clause, stealing can have a further legal effect: Art 52 of the European Patent Convention (EPC) stipulates that programs for computers as well as mental rules, mathematical methods, ways of presenting information etc are not patentable inventions and may therefore not be claimed as such. The wording "as such" from Art 52(3) has however been used to undo all explicit limits on patentability. The European Patent Office (EPO) has in 1997 begun to subdivide computer programs into two groups, "as-such programs" and "not-as-such programs", and has tried to justify this by historical claims about how art 52 EPC came about. This reasoning is at odds with grammar as well as with history. One particularly nice set of evidence comes from Scandinavia: the Danish and Swedish national versions of art 52 EPC do not literally render Art 52(3) but rather incorporate its meaning in the first line of their version of Art 52(2), coming to exactly the same common sense conclusions to which independent grammatical analyses have come and which are also stated in the early EPO examination guidelines: that a "program as such" is "something that constitutes only a program". It confirms that Art 52(3) merely exhorts examiners to look carefully where the novel achievement really lies, e.g. in a programming solution or in a chemical process which may happen to run under program control.
Why can't I patent my movie (as such)?: A comment on software patents, actually
European Patent Convention: Text of EPO edition
Art 52 EPC: Patentable Inventions: Text of EPO edition
EPC 172: Right of Governments to Act against EPO: The contracting states have failed miserably to supervise the EPO. This is mainly because their patent policy is determined by patent offices which share the same interests as the EPO.
Berichte der Münchener Diplomatischen Konferenz über die Einführung eines Europäischen Patenterteilungsverfahrens: In autumn 1973 patent experts of the european governments convened in Munich for 1 month in order to create a unified european patent examination system. This conference led to the conclusion of the European Patent Convention (EPC) and to the creation of the European Patent Office (EPO). Art 52 EPC excludes programs for computers, mental rules, mathematical methods etc from patentability. This principle was further elaborated by the EPO's examination guidelines of 1978 and the initial court practise. However starting in 1986 judges at the EPO and some national courts started to extend the scope of patentability and render the exclusions of Art 52 EPC meaningless. In order to justify this, they used a teleologic and historic method of law interpretation which makes frequent reference to what the legislators in 1973 allegedly meant or did not understand. Therefore we have dug out the relevant documents and taken a look at the (relatively short) account about the negotiations concerning Art 52 EPC. This text offers no support for the EPO's method of interpretation. Quite to the contrary.
EPO 1978: Examination Guidelines: Adopted by the President of the European Patent Office in accordance with EPC 10.2a with effect from 1978-06-01. Excerpts concerning the question of technical invention, limits of patentability, computer programs, industrial application etc.
EPO decision T 22/85 against IBM archival system: A Technical Board of Appeal of the European Patent Office (EPO) rejects a patent application which is directed to a program for computers. In 1984, the EPO's examiners had rejected the patents based on the original Examination Guidelines of 1978, saying that the claims referred to a "program for computers". The appellant argued on the basis of newer Guidelines and caselaw that his claims are directed to technical effects and not a program as such. The Board of Appeal rejects the appeal by arguing indirectly that the use of general-purpose computer hardware does not confer technicity on an abstract method: "Abstracting a document, storing the abstract, and retrieving it in response to a query falls as such within the category of schemes, rules and methods for performing mental acts and constitutes therefore non-patentable subject-matter under Article 52 EPC" and "The mere setting out of the sequence of steps necessary to perform an activity, excluded as such from patentability under Article 52 EPC, in terms of functions or functional means to be realised with the aid of conventional computer hardware elements does not import any technical considerations and cannot, therefore, lend a technical character to that activity and thereby overcome the exclusion from patentability."
EPO TBA 2002/03 T 49/99: information modelling not technical, computer-implementation not new: In March 2002, a Technical Board of Appeal at the European Patent Office (EPO) rejects a patent application for a computerised information modelling system on the grounds that the subject matter is not an invention according to Art 52 EPC. The Board argues largely in the original spirit of the EPO and differs significantly from some other recent EPO caselaw. This is an important reason why industrial patent lawyers are pressing for new patentability legislation. Under a CEC/McCarthy directive, EPO decisions such as this one would no longer be possible.
Regulation about the invention concept of the European patent system and its interpretation with special regard to programs for computers: We propose that the legislator draft any regulations on the question of software patentability along the lines of the following short and clear text.
Patent Jurisprudence on a Slippery Slope -- the price for dismantling the concept of technical invention: So far computer programs and other rules of organisation and calculation are not patentable inventions according to European law. This doesn't mean that a patentable manufacturing process may not be controlled by software. However the European Patent Office and some national courts have gradually blurred the formerly sharp boundary between material and immaterial innovation, thus risking to break the whole system and plunge it into a quagmire of arbitrariness, legal insecurity and dysfunctionality. This article offers an introduction and an overview of relevant research literature.
M. Vivant: Le Recours a la Propriété Industrielle: The article on patents analyses the history of French and European patent jurisdiction. It explains, why the European parliaments decided in the 60-70s against patentability of computer programs and how French courts supported this decision by some very clear verdicts even against software innovations related to the control of industrial production processes. It also explains how the European Patent Office since 1986 gradually deviated from these clear rules in 5 steps of logic-twisting. It warns however that the patents gratend by the EPO are of incertain value and could be negated by any national judge.
Dr. Swen Kiesewetter - Köbinger 2000: Über die Patentprüfung von Programmen für Datenverarbeitungsanlagen
Patentfähige Datenverarbeitungsprogramme - ein Widerspruch in sich: Dr. König, patent attorney from Düsseldorf, points out inconsistencies in the software patent caselaw of the EPO and BGH, criticises "circular conclusions" and argues that the EPO has "done violence to art 52 EPC". Through a "grammatical interpretation" of "programs for computers as such" he finds that this can refer only to all kinds of computer programs without exception, as far as they are claimed alone. The EPC of 1973, transcribed into German law in 1978, no longer allows a distinction between technical and untechnical programs. However it ist still possible to patent program-related combination inventions, which then have to be examined for technicity, novelty, non-obviousness and industrial applicability. Courts have often shown more imagination in "helping themselves over the obstacles of art 52". There is an elegant indirect way to effectively grant full patent protection for computer programs as such while avoiding the recent incoherences of the EPO and BGH jurisdiction. As parts of combination inventions, programs for computers may, just as is frequently the case with discoveries and scientific theories, enjoy full "usage protection", if their distribution can be construed as contributory infringement of a combination invention.
Europarl 2003-09-24: Amended Software Patent Directive: Consolidated version of the amended directive "on the patentability of computer-implemented inventions" for which the European Parliament voted on 2003-09-24.
Noël Mamère 2002-20-28: Let's just delete the "As Such" clause!: Proposes to delete Art 52(3) EPC.
CEC & BSA trying to impose unlimited patentability on Sweden: Proposes to delete Art 52(3) EPC.