A Review of the “Problem and Solution” Approach to Inventive Step under Article 56 EPC

A. Kennington (GB)

This is the second part of a long article, which will be published in three parts in successive issues of epi Information.

Origins of the Comvik Approach

The idea that the problem should include non-technical features from the claim does not originate with decision T 0641/00 Comvik. It appears to have been proposed initially in decision T 1053/98 Canon (22 October 1999), which was decided only 15 months after the initial break with the "technical contribution" approach in decision T 1173/97.

In T 1053/98 (Canon), the application related to a fax machine. It was already known for a fax machine, intended to be shared between several different users, to pre-store a number of names and allow a pre-stored name to be selected as the sender of the fax. The alleged invention involved recording the name selected for each fax sent together with the cost of sending the fax, so that a report could be output that allowed fax costs to be allocated between the different users. The appellant and the Board disagreed over the correct formulation of the problem. The Board considered that the overall aim of the invention was the economic aim of controlling charges, and stated that non-technical aims are usually not taken into account when formulating the technical problem.

The appellant's formulation of the problem was regarded as being almost entirely non-technical, such that a solution to the problem would have to include the idea of monitoring the costs incurred by each user of the fax machine. The Board stated (part 3.4 of the Reasons) "It is this non-technical part of the solution which provides an incentive for the technical part ... Consequently, in order to assess inventive step it would be necessary to consider in particular – and above all – the non-technical part of the solution."

The Board then continued (part 3.5 of the reasons) "It is exactly to avoid this situation that a technical problem has to be formulated in such a way that there is no possibility of an inventive step being involved by purely non-technical features. Such a formulation of the problem could refer to the non-technical aspect of the invention as a given framework within which the technical problem is posed." The Board then proposed a problem that included monitoring costs. Thus a non-technical aspect, that the Board considered would have to be part of the solution according to the appellant's formulation of the problem, was moved into the problem itself.

The appellant objected that the Board's formulation of the problem included parts of the solution. In part 3.7 of the reasons, the Board defended itself, saying "The Board agrees that its formulation of the technical problem to be solved by the present invention indeed contains elements of a solution, namely a non-technical solution (monitoring costs) to a non-technical problem (charge control). However, for the reasons already stated, it appears that a more general wording of the technical problem to be solved is not possible in cases such as the present one. The formal starting point should be the closest prior art, and this must be a technical document. Therefore the technical problem will relate to such technical prior art."

Several aspects of the Board's reasoning are worth commenting on, and parts of it read strangely to modern eyes.

In the passage quoted from part 3.4 of the Reasons, the Board stated that because the non-technical part of the solution provided the incentive for the technical part, it would be necessary to consider the non-technical part in order to assess inventive step. It is not clear why the Board thought this. Under the subsequently-adopted Comvik approach the non-technical features could simply have been ignored in the consideration of inventive step.

In part 3.5 of the Reasons, the Board does not say why or how the inclusion of a non-technical aspect of the invention in the problem would avoid the possibility of an inventive step being involved by purely non-technical features. Again, under the Comvik approach this concern would not exist, since such non-technical features would be ignored in the assessment of inventive step. Thus the Board's motivation for adding the non-technical aspects to the problem no longer applies under modern case law.

The Board seems to have made no attempt to establish the legality of its decision to include non-technical elements of the solution into the problem. The only justification given by the Board is the practical one given at the beginning of part 3.5 of the Reasons. This is unfortunate, since a tribunal must always follow the law. However desirable a particular outcome may seem in practice, this cannot justify a departure from the requirements of the EPC.

I believe that if the facts of this case are reconsidered in the light of the current treatment of non-technical features, it can be seen that there is no need to include non-technical features from the solution in the statement of the problem. This reconsideration can be based on two legal principles.

First, an invention must be technical, and the invention must have an inventive step. This follows from Article 52(1), which reads as follows.

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

Since inventive step must be a feature of the invention, and the invention must be technical, non-technical features in a claim cannot contribute to inventive step unless they combine with technical features to contribute to a technical effect.

Second, Article 56 defines inventive step as not being obvious "having regard to the state of the art", and the state of the art can be regarded as everything that has been made available to the public anywhere in the world before the priority date of the claim. Therefore the problem can incorporate anything that is known having regard to the state of the art and anything that is obvious having regard to the state of the art. The word "everything" requires that nothing made available to the public can be excluded from the state of the art, and therefore non-technical features must be included in the state of the art. Since the state of the art necessarily includes all known non-technical features as well as all known technical features, the problem does not have to be seen as an exclusively "technical" problem (decision T 1784/06, of 21 September 2012, has recently stated explicitly that a technical problem is not an absolute requirement of the problem and solution approach).

Combining these two principles, we can say that the claimed solution must involve a non-obvious technical invention, but the problem may be technical, non-technical or a mixture of technical and non-technical. If the solution to an obvious problem is technical but obvious, there is no inventive step. If the solution is not obvious, but is also not technical, there is again no inventive step since inventive step has to be a property of the invention and the non-technical aspects of the solution cannot provide an invention in accordance with Article 52(1) EPC. However, if the solution to an obvious problem is technical and not obvious, there is an inventive step (even if the problem is entirely non-technical) since the claim includes non-obvious technical features. To put it another way, when the problem and solution approach is used, the (known or obvious) problem may be technical or non-technical, but there is only an invention if the solution is both non-obvious and technical.

Turning to the facts of decision T 1053/98, it was well known long before the priority date to split a telephone bill between different users on the basis of how much each person had used the telephone, and it was also well known that manual records of who used the telephone were unreliable. In the case of a facsimile machine to be used by multiple different departments, as in prior art D2 of T 1053/98, it would be obvious to want to split the telephone bill for use of the machine between the departments on the basis of the amount that each department had used the machine. Once the skilled person sees that the machine of D2 already has means for selecting the name of the user sending the fax on any particular occasion, it must be obvious to add the feature of storing usage data against each name so as to provide a more reliable report of how much each department has used the machine.

Therefore the Board in decision T 1053/98 could have avoided adopting a new procedure, of adding features from the solution to the formulation of the problem, if it had adopted a more robust approach to combining the starting point prior art with non-technical common general knowledge. Additionally, its concern that it could be required to give weight to non-technical features of the solution no longer arises since it is now accepted, especially in the light of the revised wording of Article 51(1) EPC in EPC 2000, that non-technical features of the claim may simply be ignored as incapable of contributing to an inventive step.

On the other hand, it should be accepted that if the technical aspects of the claim are not obvious over what is already known, there is an inventive step even if the novel technical features are provided purely for a non-technical purpose. This was confirmed in Board of Appeal decision T 1689/07 Proctor & Gamble, Colour-changing absorbent article (6 November 2009).

Consequently, it appears that there has been development of the jurisprudence since decision T 1053/98, such that there is no need to follow its proposal to incorporate non-technical but new and non-obvious features of the claim into the formulation of the problem.

T 0931/95 (Pension Benefits Systems, dated 8 September 2000) does not make any reference to earlier decision T 1053/98, but this decision appears to adopt a similar approach to non-technical features of the invention.

In this case, the applicant filed a main request and auxiliary requests. Claim 1 of the main request was directed to a method of controlling a pension benefits programme. The claimed steps all related to performing pension-related data-processing operations on personal or pension cost data, using a computer. Claim 1 of the 1st auxiliary request (which was claim 5 of the main request) related to apparatus for controlling a pension benefits system comprising data processing means arranged to receive and process various items of personal and pension-related data.

Claim 1 of the main request was considered to relate to a method of doing business. Each of its steps was a purely business step, and the fact that it was carried out on a computer did not stop the steps from relating to a business method "as such". Therefore this claim was rejected as being non-technical.

Claim 1 of the 1st auxiliary request was an apparatus claim that required the presence of data processing means. The data processing means included a processor that comprised computing means for determining an average age, determining a periodic cost, and estimating expenses. This claim was held to be technical because the processor comprising computing means was a technical feature. Specifically, the Board held that it was implicit in the EPC that an invention had to have a technical character, but that there was no basis for distinguishing between new features and known features in a claim when considering the requirement for technical character (contrary to the previous "technical contribution" test).

The Board then ruled that the claim lacked inventive step. In part 8 of the reasons for the decision, the Board pointed out that the improvement envisaged by the invention was essentially economic, which cannot contribute to inventive step. It continued as follows.

The regime of patentable subject-matter is only entered with programming of a computer system for carrying out the invention. The assessment of inventive step has thus to be carried out from the point of view of a software developer or application programmer, as the appropriate person skilled in the art, having the knowledge of the concept and structure of the improved pension benefits system and of the underlying schemes of information processing as set out for example in the present method claims.

The Board then stated that the technical claimed apparatus are defined functionally by precisely the details of the improved system that it considered the skilled person should be regarded as knowing. Consequently those features were all obvious.

In the passage quoted above, the Board stated that the assessment of inventive step has to be carried out from the point of view of a skilled person having knowledge of the improved pension benefits system. In other words, the Board stated that it was necessary to assess inventive step starting from knowledge that was not in the prior art. On the face of it, this appears to be contrary to Article 56 EPC. Unfortunately, the Board in this case provided no discussion of the legality of this approach, nor of why it felt compelled to treat the skilled person as having knowledge not in the prior art.

Judging from the way in which the inventive step of the computing means was considered, it appears that the Board felt that the only way to prevent the non-technical features from being considered in the assessment of inventive step was to treat them as already known by the skilled person. However, if we follow modern jurisprudence in stating that non-technical features cannot contribute to inventive step unless they combine with technical features to produce a technical effect, it appears to be possible to arrive at the same conclusion as the Board in T 0931/95 without needing treat the skilled person as knowing the new non-technical features.
The claim in this case recited a processor comprising computing means. The computing means were for carrying out non-technical steps. There was no suggestion that the non-technical steps combined with the technical computing means to have a technical effect. On the contrary, the result of the steps appeared to be entirely non-technical information. Therefore the purposes or uses of the computing means, as recited in the claim, were non-technical and it would be correct to ignore them when carrying out the assessment of inventive step. This would have the effect of reducing the claim to data processing means including a processor comprising computing means. This is not new over any known general purpose computer. Thus it is possible to reach a finding of lack of inventive step (or even a finding of lack of novelty) simply by ignoring the non-technical features, and there was no need to treat the non-technical features as if they were known to the skilled person.

Therefore if this case was decided today, using the modern approach to non-technical features, it would not be necessary to add the new non-technical features to the knowledge of the skilled person. The result, that claim 1 of the 1st auxiliary request lacked inventive step, could be achieved starting from a definition of the problem that only included matter that was already known from the state of the art and did not include any novel features.

T 0641/00 (Comvik, dated 26 September 2002) concerned identities in a cellular mobile telephone phone. In a mobile phone, the SIM card stores a subscriber identity known as an IMSI. This is transmitted when the telephone is used, to identify the user. Comvik's application claimed a method in a GSM mobile telephone system in which a SIM was allocated at least two IMSIs, with corresponding information being stored in a system database, and the user selectively activating a desired one of the identities to allow for selective distribution of call costs between the identities. Claim 1 included the feature that the method was used to distribute costs between different users.

The Board considered that the claim contained a mix of technical and non-technical features. The Reasons for the Decision starts with a lengthy discussion of the relevant law and practice. In discussing the problem and solution approach in part 5 of the Reasons for the Decision, the Board stated that the problem must be a technical problem, and if no technical problem can be derived from the application then there is no invention within the meaning of Article 52 EPC, citing decision T 0026/81. It then stated that if a feature does not contribute to the solution of a technical problem by providing a technical effect, it is not relevant for inventive step (Reasons for the Decision, part 6).

When discussing the formulation of the problem, the Board stated:

"...the problem must be one that the skilled person in the particular technical field might be asked to solve at the priority date." (Reasons for the Decision, part 5)

and

"The technical problem should not be formulated to refer to matters of which the skilled person would only have become aware by knowledge of the solution now claimed. ... Thus a problem should not contain pointers to the solution or partially anticipate it." (Reasons for the Decision, part 7)

but continued

"However, in the Board's view this principle applies to those aspects of the subject matter claimed which contribute to the technical character of the invention and hence are part of the technical solution. Merely because some feature appears in the claim does not automatically exclude it from appearing in the formulation of the problem. In particular where the claim refers to an aim to be achieved in a non-technical field, this aim may legitimately appear in the formulation of the problem as part of the framework of the technical problem that is to be solved, in particular as a constraint that has to be met." (Reasons for the Decision, part 7)

The Board then justified this view by referring to earlier decisions T 1053/98 and T 0931/95 (discussed above). It did not discuss the compatibility of this approach with Article 56 EPC. The Board appears to have considered that, by putting non-technical features from the claim into the statement of the problem, it ensured that only technical aspects of the invention were regarded as contributing to the solution. The Board went on to state that this approach "is actually a method of construing the claim to determine the technical features of the claimed invention".

The Board determined that the novel features of claim 1 were (i) allocating at least two identities to the SIM, (ii) the two identities being selectively usable, and (iii) the selective activation being used for distributing costs. However, it considered that selectively distributing costs according to specific schemes (features (ii) and (iii)) was not technical. In accordance with its stated view about incorporating non-technical features in the problem, the Board reformulated the problem to include the feature of allowing user-selectable discrimination, between calls for different purposes or different users, for the purpose of cost distribution. In the light of this formulation of the problem, the Board considered that it would be
obvious to allocate multiple IMSIs to a SIM (feature (i)), on the grounds that this was the only way to allocate costs for one mobile telephone between multiple identities. Therefore the claim lacked inventive step.

Specifically, the Board stated

"Discriminating between calls originating from one and the same mobile station, therefore, requires the allocation of different IMSI numbers, or in other terms, the implementation of a corresponding number of GSM applications (feature (i)). Faced with this technical requirement, the skilled person finds a solution in document D8 ...." (Reasons for the Decision, part 15)

Thus the novel technical features of the claim became obvious because the novel non-technical feature was included in the formulation of the problem and was therefore regarded as known to the skilled person even though it was not available to the public at the priority date of the claim.

It is worth noting that, in its discussion of earlier decision T 1053/98, the Board referred to an approach that "accepts it as correct to formulate the technical problem to include non-technical aspects whether novel or not". Therefore the Board was clearly aware that it might be including in the problem of a feature that was not available to the public.

It is difficult to understand some of the thinking that underlies the decision.

In the discussion of earlier decision T 0931/95, the Board stated that this "approach, which is actually a method of construing the claim to determine the technical features of the claimed invention, allows separating the technical from the non-technical aspects of the invention even if they are intermingled in a mixed type claim feature". It is hard to make sense of this statement.

The act of formulating the problem, or attributing certain knowledge to the person skilled in the art, does not contribute to the act of construing a claim. The act of construing a claim is one in which the meaning and scope of the claim are determined. When doing this, there is no need to identify which features of the claim are known, and no need to consider how to formulate the problem. In general, the act of construing the claim will need to be undertaken, at least partially, before attempting to define the problem or identify what features are known to the skilled person. There is no point in trying to formulate the problem, or identify what relevant knowledge the skilled person has, until features of the claim have been identified.

Equally, it does not seem to be correct to say that the approach, of treating the skilled person as knowing non-technical features of the claim, allows the separation of the technical from the non-technical aspects of the invention. In order to treat the skilled person as knowing the non-technical aspects, it is necessary first to identify what those non-technical aspects are. Therefore the non-technical aspects must have been identified before this approach can be put into practice. However, once the non-technical aspects have been identified, they can be separated from the technical features and ignored when considering whether the technical aspects of the invention solve the problem. It does not seem to be necessary to add the (already identified) non-technical aspects to the stated problem in order to separate them from the technical features.

Perhaps the Board in T 0641/00 Comvik felt that where a new technical feature is adopted in order to achieve a non-technical aim, the non-technical nature of the aim or motivation behind the invention somehow pollutes any new technical feature that is obvious over the non-technical aim and renders such features not eligible for a patent. In order to achieve this effect within the problem and solution approach, the non-technical aim is incorporated in the statement of the problem used for the analysis of inventive step. However, if this was the motivation behind the practice of adding novel non-technical features to the problem, it was not clearly stated in the decision and appears to be incorrect. If an invention includes technical features that are not obvious over everything that has been made available to the public, it is inventive according to the EPC. This test relies only on the presence of those features in the claim. The additional presence of other features in the claim, which may be known or may be non-technical, is irrelevant. The aim or motivation for adopting the new and non-obvious technical features is also irrelevant. A technical invention is patentable regardless of whether it is aimed at solving a technical problem or a non-technical problem, and it does not matter if it is the obvious solution to the problem that the inventor set himself or herself provided that the problem set by the inventor is itself unknown and the invention is not obvious over what is known.

The point, that a non-technical motivation does not prevent patentability, is made in decision T 1689/07 Proctor & Gamble, Colour-changing absorbent article (6 November 2009). In that case the examining division refused the application on the grounds that the claimed article was an aesthetic creation since the technical feature distinguishing the claimed article from the closest prior art had solely an aesthetic effect. The Board in that case disagreed. It stated that "Whether the final aim of the technical effect achieved by the claimed absorbent article ... is only aesthetic ... has no impact on the technical nature of the claimed absorbent article itself. It is in fact common practice that inventions such as dyes or hair colouring compositions, although having only an aesthetic goal, are nevertheless patentable inventions in the sense of Article 52 EPC since ... dyes and hair colouring compositions[,] are not per se aesthetic creations."

Accordingly, if a technical feature in a claim is obvious over the inventor's motivation, but that motivation is itself unknown and the technical feature is not obvious over what is known, that technical feature should confer patentability on a claim and it should make no difference whether the inventor's motivation was technical or non-technical.

Compatibility of the Comvik Approach with the Proposed Rule on the Formulation of the Problem

It is apparent from the preceding discussion that the rule for formulating the problem in the Comvik approach is unlikely to be compatible with the proposed rule that the problem should be formulated so as to be known or obvious over the state of the art. This can also be seen by considering whether the outcome in T 0641/00 Comvik would have been different if the proposed rule had been followed. I have argued above that in earlier cases T 1053/98 and T 0931/95 there was no need to include any novel non-technical features in the statement of the problem, and the same outcome could have been reached by entirely conventional reasoning. This does not appear to be the case with T 0641/00 Comvik. When considering T 0641/00 Comvik, the following possible situations may be considered.

First, the (non-technical) idea of user-selectable discrimination between calls originating from one and the same mobile station, e.g. for selectively distributing costs, might have been known or obvious at the priority date of the claim. In this case, the problem as formulated by the Board would have been known or obvious from the state of the art. Alternatively, the technical requirement of allocating different IMSI numbers to the same mobile station might have been known or obvious for some other reason. In either case, the technical requirement would have been known or obvious from the state of the art, and therefore the claimed invention would have been obvious over a problem that was itself known or obvious having regard to the state of the art. Under these circumstances the outcome of decision T 0641/00 Comvik would have been consistent with the proposal that the problem should be known or obvious from the state of the art. However, this does not appear to be the case. On the contrary, in decision T 0641/00 the Board clearly considered that both the non-technical idea and the technical requirement were new, and there is no suggestion that either of them was obvious from the state of the art.

Second, it might be possible that the claim was an obvious solution to the problem, even if the problem did not contain the non-technical idea of user-selectable discrimination between calls originating from one and the same mobile station. In this case, there would have been no need to include the non-technical idea in the statement of the problem. However, there is nothing in the decision to suggest that the Board took this view. On the contrary, the fact that the Board relied on the skilled person's knowledge of the cost distribution concept in order to find the claim obvious, suggests that the Board did not consider the claim to provide an obvious solution to the problem in the case that the problem omitted the non-technical idea.

Finally, it may be the case that the claim was not obvious over the problem unless the problem included the novel non-technical idea of user-selectable discrimination between calls originating from one and the same mobile station, and it may also be the case that this non-technical idea was itself not obvious over the state of the art. This appears to have been the view of the Board. However, this must imply that the claim was not obvious over the state of the art alone, and only became obvious by including in the formulation of the problem a (non-technical) idea that did not exist in the state of the art (defined as everything made available to the public in any way before the priority date). Consequently it does not seem possible to reconcile the Board's approach with the proposal that the problem, as formulated for the problem and solution approach to inventive step, should be known or obvious based on the prior art.

Many subsequent decisions have followed the Comvik manner of formulating the problem, and there is not space to review all of these in detail. Nor is it necessary to do so. The circumstances of the Comvik decision itself, as discussed above, demonstrate clearly that this manner of formulating the problem cannot always be reconciled with the rule that the problem must be known or obvious having regard to the state of the art.

Therefore it appears that either it is necessary to abandon the proposed rule that the problem should always be formulated so as to be known or obvious over the state of the art, or it is necessary to modify the "Comvik approach" so as to abandon the practice of including non-technical features in the problem where those features are not known or obvious. In general, I am reluctant to propose a change to well-established practice of the Boards of Appeal. However I think that in this particular case there are good reasons for considering that jurisprudence has moved on and it is now appropriate to modify the Comvik approach in this way.