Saturday, 3 July 2010

Interpretative Spotlight: “consists”

It is generally held that the term “comprising” opens a claim whereas “consisting” closes it. The present decision shows that this cannot be taken for granted if the application suggests otherwise.

Claim 1 as granted read:

1. Exhaust gas recirculation system for an internal combustion engine, in particular a diesel or spark ignition engine, with an exhaust gas recirculation valve for recirculating exhaust gases into the air supply system of the internal combustion engine, the exhaust gases flowing through a cooling device, also with a bypass pipe (17) which bypasses the cooling device (10) and with a separating filter (18; 19), the inlet of which is connected to an exhaust gas recirculation line and has two passages, one of which is connected to the cooling device (10) and the other of which is connected to the bypass pipe (17) in such a way that the exhaust gases can flow controllably alternatively through the cooling device (10) or the bypass pipe (17), wherein the separating filter (18; 19) is spring-controlled (21) for one operating position and pressure-controlled for the other operating position and is held in the normal operating position by a spring (21), characterised in that the separating filter (18; 19) is arranged in front of the cooling device (10) and is an integral component of the cooling device (10) and consists [language of proceedings: besteht] of precisely one bypass valve (18) or precisely one bypass shutter (19), which is arranged in the junction in front of the cooling device (10) and the bypass pipe (17) and in a first operating position closes the cooling device (10) and in a second operating position the bypass pipe (17). 

** Translation from the German **

[2.2.1] The parties do not agree on whether the term “consists” (besteht) in claim 1 is to be understood as being a closed term or not […].


(a) In principle the terms used in patent documents are given the usual meaning they have in the prior art, unless the description unambiguously requires a special meaning (see, for example, T 1321/04). If a patent document explicitly uses terms in a meaning that is different from the usual meaning, then this meaning is decisive (geht vor).

(b) Claim 1 as originally filed discloses a separating filter the “inlet of which is connected to an exhaust gas recirculation line and which has two passages”. […] Moreover, it is disclosed that the valve 18 and the shutter 19 are bypass elements and form a separating filter. It follows that the separating filter has both an inlet and two passages as well as the valve 18 and the shutter 19.

The characterising part of claim 1 claims a separating filter having precisely one bypass valve or precisely one bypass shutter. The term “precisely” only determines the number of bypass valves or bypass shutter but does not exclude that the separating filter has more features, such as a housing, valve seats etc.

(c) Thus one may conclude from the application as filed that claim 1 does not use the term “consists” as a closed term (im abschliessenden Sinne).

To read the whole decision (T 523/08, in German), you may click here.


Anonymous said...

Well, I would say that "consists" is a closed term and if the application as filed does not support this interpretation, then the claim infringes Article 123(2) EPC. The Board's position here implies that the claim could be amended post-grant from "consists of" to "comprising" without offending against Art. 123(3) EPC. That seems too weird to be correct.

In any case, this decision seeems logically inconsistent with decision T 997/06, discussed here, which states that if the open term is disclosed, then automatically also the closed term is disclosed as a limiting case. Indeed, applying that logic to this case, the fact that the application as filed discloses the open term implies that it also discloses the closed term ("as a limiting case"). Therefore there is no need to interpret "consist of" as open, since the closed interpretation is already perfectly in line with the application as filed. (That is, if one accepts the logic of T 997/06, which I personally do not.)

Oliver G. Randl said...

I find these decisions quite stimulating. I have already had the occasion to say that in my opinion T 997/06 has it right. But more importantly, this decision and the discussion that followed the post dedicated to it made me think of the legal basis for our conceptions of “consisting” and “comprising”. And there is hardly any. These notions are part of the common general knowledge of the skilled patent drafter, but where do they creep into the EPC? As far as I can see, the sharp distinction comes from U.S. case law, but of course, this case law has no binding effect whatsoever in Europe.

The Guidelines C-III 4.21 state that “while in everyday language the word “comprise” may have both the meaning “include”, “contain” or “comprehend” and “consist of”, in drafting patent claims legal certainty normally requires it to be interpreted by the broader meaning “include”, “contain” or “comprehend”.” (which is a verbatim citation of T 759/91). T 711/90 is also cited, but this decision merely states that “as the terms “consists of” clearly excludes the possibility of the glass containing other oxides …” without further discussion of the basis for such an assertion. So the legal basis is quite shaky, isn’t it?

I would say that the relationship between the common understanding of “comprising” vs. “consisting” and the EPC is somewhat comparable to the relationship between customs and civil law. Now that we have decisions some of which accept the ‘custom’ and some of which do not, it may be wiser not to rely on the custom any more and make it clear in the specification how “comprising” or “consisting” are to be understood (and what they are meant to disclose).

Anonymous said...

So you agree with both T 997/06 and this decision (T 528/08)?

The point of T 997/06 is that any "open" collection of features implicitly discloses the "closed" collection as a limiting case.

In T 528/08, the Board considers that the application as filed discloses an "open" collection. By the logic of T 997/06, this means that the application as filed in T 528/08 also discloses the "closed" collection.

Prima facie, the use of "consists" in claim 1 of the main request in T 528/08 suggests that the "closed" collection is claimed. Since the application as filed discloses a "closed" collection (following T 997/06), there is no reason to not give "consists" its ordinary meaning.

The Board in T 528/08 however is of the view that the application as filed does NOT disclose the "closed" collection, and only for that reason assigns to "consists" a meaning different from its ordinary meaning.

So in my view T 997/06 and T 528/08 are contradictory. In a subtle way, but still contradictory.

Personally I disagree with both T 997/06 and T 528/08, but if you agree with one of them then at least (logically) you should disagree with the other.

Anonymous said...

(Oops, it's T 523/08, not T 528/08.)

Maybe to put it more simply:

If you believe disclosure of an "open" collection necessarily implies a disclosure of a "closed" collection as a limiting case, then you agree with T 997/06 but should disagree with T 523/08 (point 2.2.2(b), last sentence, and point 2.2.2(c)).

If you believe disclosure of an "open" collection does not necessarily imply a disclosure of a "closed" collection as a limiting case, then you disagree with T 997/06, point 29. The Board in that decision only looked at the claim language, without considering the description, and concluded that "comprising" (interpreted as "open") implied a disclosure of "consisting of" (interpreted as "closed") as a limiting case, without leaving any room for exception.

Anonymous said...

In my view, the legal basis for adhering to the conventional interpretation of "consisting of" and "comprising" is Art. 84 EPC and more generally the requirement of legal certainty.

If a claim states "consisting of", then it should not be necessary to delve into the description to see if that is really what it means. If the description attempts to redefine those basic patent terms, it should be objected to under Art. 84 EPC (in the spirit of GL C-III, 4.4).

Art. 123(2) permitting, any unfortunate confusion between "consisting of" and "comprising" in the application as filed can be fixed without any problem during the prosecution of the application. There is therefore no need that the applicant should benefit from any non-conventional (and therefore inherently ambiguous) interpretation of these terms.

After grant, it is of course more difficult to fix such problems. In my view, claim 1 of T 523/08 is first of all unclear: it allows for both an "open" and a "closed" interpretation. If either of those interpretations is not supported by the application as filed, the claim infringes Art. 123(2). Amendment to the "open" interpretation (consists -> comprises) infringes Art. 123(3), since it broadens the scope vis-à-vis the narrower "closed" interpretation. For the purpose of novelty and inventive step (in opposition), the broader "open" interpretation should be used.

Oliver G. Randl said...

Do I agree or disagree with T 523/08? We are dealing with complex matters, and it is not always easy to have a clear-cut opinion. I was quite surprised when reading some comments on T 997/06, because I really thought that it was hardly conceivable that “comprising” did not automatically disclose “consisting”. I now understand how one may come to this conclusion (although I remain convinced that this is too extreme a point of view). To some extent, T 997/06 prepared me for T 523/08 where the Board inverses things and accepts “consisting” to be understood as “comprising”. I would have been shocked not long ago, but T 997/06 has made me realise that my dogma on “consisting” and “comprising” relied on little legal basis. So, do I agree with T 523/08? I guess I would not have decided that way, but I can live with it. It is certainly a decision to be kept in mind: perhaps one day you will need to convince the EPO to have a larger understanding of a claim drafted in too closed a manner.

Anonymous said...

Do you not agree that T 523/08 implicitly disapproves of the idea that an "open" term implicitly discloses the "closed" term as a limiting case? Reason: otherwise, the "closed" interpretation of "consists" would have been disclosed in the application as filed (given that it discloses the "open" interpretation), while T 523/08 considers that the "closed" interpretation is not disclosed in the application as filed.

The point I'm trying to make is: although interpretation of natural language is inherently difficult and dependent on the facts, the principles applied should not be contradictory. Here, they are. Of course it is unavoidable that contradicting decisions occur, but then they can't both be considered "correct".

I agree that an attorney should be aware of the various decisions and refer to (only) those that help his case. In that sense, an attorney does not have to agree or disagree with any decision.

Btw, T 997/06 fully complies with your dogma on "consisting of" versus "comprising". T 997/06 categorically allows the amendment of "comprising" into "consisting of" *not* because they mean the same thing, but because the former implicitly discloses the latter (as an alternative, more or less).

Oliver G. Randl said...

I wonder if you do not read more into T 523/08 than there was in the mind of the Board members. But of course, I would agree that T 523/08 and T 997/06 are not coherent. If I understand T 523/08 correctly, the Board treats “consisting” as if it was a word like any other and not a word that has a very special (dare I say, technical) meaning when used in a claim. T 997/06, on the other hand, is fully aware of the very special meaning of the terms “comprising” and “consisting” and draws all the consequences.

I found this discussion on “comprising” and “consisting” very helpful. There are some forthcoming posts on A 123(2) related decisions which I trust will allow us to heartily disagree. ☺

Anonymous said...

Yes, that's also an incoherency between the two decisions, but not exactly what I meant.

I just thought of another angle to look at these issues.

Suppose claim 1 as filed states "consists of", but all embodiments in the description state "comprising".

What is now the proper interpretation of claim 1? Open or closed?

Anonymous said...

In my view these decisions show the limits of the case law as it now stands. We should not forget that the decisions are given by aprox 25 boards in different compositions (the full possible combinations are huge) and that a higher instance that can harmonize the principles does not exist.

Add to this that most decisions are not written by native speakers in any of the three official languages and you see that indeed the work done is quite a good one :-)

Oliver G. Randl said...

OK, I get you.

T 523/08 would say “open”, thereby excluding that “comprising” includes “consisting” and that the claim could be voluntarily limited to “consisting”. I think your reasoning is correct, but I doubt that the Board reasoned at so deep a level.

I wonder why the Board felt it necessary to interpret the term “consisting”. If “consisting” is a technical term indicating closedness, then the claim as such is clear and does not need any interpretation based on the description. Only if the term “consisting” as such is considered as requiring interpretation, then you would have a look at the embodiments and perhaps come to the conclusion of T 523/08. It somehow sounds as if the Board was not familiar with basic claim drafting conventions … or wanted to ignore them.

Anonymous said...

It seems the Board did not follow basic claim drafting conventions in order to "correct" for an Art. 123(2) problem (which btw did not exist according to T 997/06). But I don't believe that "claim 1 plus description" should be interpreted differently depending on whether claim 1 has been amended (as in T 523/08, if I understand correctly) or is as originally filed (as in my example). So indeed T 523/08 would stand in the way of this voluntary limitation.

I agree that the Board might not have reasoned at so deep a level, and that these decisions show the limits of the case law.

A question that is somewhat related. Suppose text B is "directly and unambiguously derivable" from text A, and text C is "directly and unambiguously derivable" from text B. Does it then necessarily hold that text C is "directly and unambigiously derivable" from text A?

In other words: is an "allowable amendment of an allowable amendment" of the application as filed itself by definition an allowable amendment of the application as filed?

I would say "yes", but it could be argued that this puts impractical limits on the allowability of amendments (since it has to be ensured that the amended text, when read without reference to the original text, cannot lead to interpretations not supported by the original application).

The alternative is that an amended application should always be read with reference to the application as filed, but that would seem to require inclusion of the text of the application as filed in the publication of the patent specification...