Non-inventions of Article 52

Article 52 excludes expressly discoveries, scientific theories and mathematical methods; aesthetic creations; schemes, rules and methods for performing mental acts, playing games or doing business, and programs for computers; and presentations of information from patentability. Such "creations" or "ideas" which persuant to the provsions of Article 52 (2) are not entitled to patent protection are called "non patentable inventions" or just "non-inventions". 

The European Patent Office had to face in the sector of business methods an increase workload [1]. In order to ease this workload in certain cases, where the European Patent Office deems a European patent application as falling into the cathegory of Article 52 (2) it issued a declaration of no search under Rule 63 EPC (former Rule 45 EPC), usually with the following wording:

The claims relate to subject matter excluded from patentability under Art. 52(2) and (3) EPC. Given that the claims are formulated in terms of such subject matter or merely specify commonplace features relating to its technological implementation,the search examiner could not establish any technical problem which might potentially have required an inventive step to overcome. Hence it was not possible to carry out a meaningful search into the state of the art (Rule 63 EPC). See also Guidelines Part B Chapter VIII, 1-6.

"Notorious" technical features

In several decisions of the boards of appeal confirmed that this practice is only acceptable in the event that a search under Rule 45 EPC can be denied only if a search is not possible at all. Whether or not the search division believes that the result of search will be of use for the substantive examination is irrelevant. If nethertheless the situation occurs that no search has been performed although it could have been, an examination division is not obliged to perform an additional search for purely formal reasons if it considers a refusal of the application to be justified on the basis of prior art which is either so well known that its existence at the date of priority cannot be reasonable disputed (T 1411/08) or accepted by the applicant as known. Notorious prior art clearly does not require proof [in form of documentary evidence]. In all other cases an additional search should be performed. (T 1242/02 - Bereitstellung produktspezifischer Daten)

The term "notorious" however should be interpreted narrowly. (T 0690/06 - Financial records/AUKOL). As long as no search has been performed an examining division should normally not refuse an application for lack of inventive step if the invention as claimed contains at least one technical feature that is not notorious. In case there is a plurality of technical features which all in isolation are notorious it is not admissable without further reasoning to regard a subject matter that links notorious technical features in a usefull way as notorious as a whole (T 1287/04). It is the nature of most inventions to be assembled from well-known elements. The invention then resides in the teaching of a new interaction between the well-known elements.

In a "normal" situation where a search is carried out and the invention differs from the closest prior art by a certain technical feature and the examining division considers this technical feature obvious for a skilled person, the invention would be rejected for lack of inventive step. The same approach is not acceptable if no search was carried out and all but one technical features are notorious. In this case it is not acceptable to argue similar to the "normal case with a search" that the only not known feature is obvious for a person skilled in the art. The argument of the examining division that a search was not necessary as a decision could be reached anyway would lead to the conclusion that a substantive examination serves as a basis for the search and not vice versa (T 1515/07).

This differs from T 172/03 which stated that any piece of prior art which renders an invention obvious is relevant, whether or not there might be an "even better" one. This situation would be of course different if a patent is granted, in which case the best available piece of prior art normally has to be identified (T 967/97). In a case where the application had been refused by the examining devision due to obviousnes in view of [exclusively] "notorious" prior art the board of appeal saw its duty to decide on the refusal, i.e. the lack of inventive step and not if the examining division had violated its duty to search for the closest prior art (T 688/05).

The problem-and-solution approach always starts from the closest prior art. If a search is denied, then one cannot be sure if the priort art acknowledged by the applicant is the closest prior art or if a search would unveil prior art that is more pertinant than the prior art that has been admitted, or prior art would be found that even destroys the novelty of the claimed invention. This could lead to the unfortunate situtation that the examining division rejects a patent application to due to the admitted prior art for lack of inventive step and the applicant wins the appeal. In a normal case with a search the Board of appeal would be able to grant a patent. In the event no search was carried out then the case has to be remitted to the examining division which then has to carry out a search before a substantive examination could be carried out.

"Notorious" has alternatively also been interpreted that a claimed technical feature is a generic feature implying that technical detail [of the claimed feature] is not significant. Thus a "computer for sending a message" is notorious, whereas a "32-bit computer for sending a message" would not be, when for example, questions of advantages and disadvantages of 32-bit computer arise. (T 1411/08)

Examples of notorious technical features

  • Conncted terminals, databases and servers.
    (T 1000/08)

  • The use of a computer infrastructure for the automatisation of data administration is a notorious technical feature.
    (T 0290/07)

  • Assigning a unique identification attribute for accounting purposes of used resources or services to a token, which is exchanged between computers in a computernet is notorious from bank notes or coupons, which have unique serial numbers.
    (T 0823/08)

  • Using a cryptographic signature for rendring a token forgery-proof was notorious at the priority date of the application (04.04.2003) so that no documentary evidence was needed.
    (T 0823/08)

  • Attaching payload data, which consists of information of purely commercial or adminstative relevance, to a token is notorious from [paper] coupons or receipts which comprise details of a documented business transcations, such as date, value, or signatures.
    (T 0823/08)

  • The attribution of an identification code to a stored data record for the purpose of unique access is a notorious classification system in data base techniques.
    (T 0290/07)

Examples for non-notorious features

  • Co-operation of Super-Peers in a decentralized computer network for issuing tokens with distributed signatures.
    (T 0823/08)


  • Regardless of wheter keeping earlier search results in a local database for predicting the result of an acutal search query can be distinguished from caching operation (keeping an exact copy of selected data in a memory space with faster access) the distinguishing features of the claimed invention (submitting queries from a server to a computer network, storing received information in a local data base, determining either producing a prediction of an actual query from data in the local data base or submitting a new query to the computer network) goes far beyond what is "notorious" or essentially refutable.
    (T 0279/05)


  • Means for receiving case related data from a medical device are beyond what is indisputably the expertise in the field of data processing, this means that this feature is not notorious as it is not generic.
    (T 1707/08)


  • Websites that store information about the user's behaviour (If this was known or not known on the priority date is certainly open to question, and it is certainly something a search can be directed.
    (T 1411/08)


  • A packaging machine with an interface to  a [networked general-purpose computer for ] data storage does not constitute a commonplace peripheral of such a computer that the notariety of the computer system propagates ot the packaging system. Even though the application refers to a packaging machines only in a generci way, the Board does not regards packaging machines as notorious components. A packaging machine is a technical device which is not usual computer peripheral, such as a printer, keyboard, data storage, display, I/O modem or the like.
    (T 0698/11)


  • Apparatus and method for storing data (relating to offer and search profiles) and for the targeted control of access to the data comprising a data storage means, a user interface, comparing means and access control means.
    (T 1287/04)

It should be noted that in a lot of cases, where the Board of Appeal in contrast to the examining division assessed that there was at least one non-notorious feature and remitted the case to the examining division, the examining division after having carried out a search for the non-notorious feature found prior art, that destroyed in the end novelty or inventive step of the application. On various decisions the Board of Appeal expressively emphasized that in taking this decision no assessment of any substantive issues of the case has been made (for example T 1515/07) or that the remisson of the case does not imply any statement or assumption on the prospect of success of the patent application in the examination process that will be continued (for example T 0823/08).


[1] Daniel Closa "Nov" ??? indicates in a presentation dated 23.11.2005 that some thousand of "no-searches" have been sent [by the EPO] in the last 4 years.

[2] Daniel Closa states in his presentation "Communication under Rule 45 EPC (Non Search) in Directorates 1238 & 2221", dated 27/11/2003, that the problem of non-search is not directly related to business methods. However, the content of recently filed applications has changed dramatically due to the fact that in the US, it is now possible to have patents for application having very little (almost no) technical content. The same type of application is then filed in Europe and in particular at the EPO. Suddenly (May 2001-March 2002) the number of applications to be searched at the EPO in that field increased from 20 a month to 600 per month. 80% of these applications (in particular US-PCT) were drafted according to the US legal criteria, not [respecting criteria that comply with the European Patent Convention]. Daniel Closa argues that as it is of no value to the Public to spend a lot of time on searching applications which will, according to the present state of the [European Patent] law, never result in a European grant, it is logical to issue a communication of non-search.