For the half-year to 30 June 2014, the IPKat's regular team is supplemented by contributions from guest bloggers Alberto Bellan, Darren Meale and Nadia Zegze.

Two of our regular Kats are currently on blogging sabbaticals. They are David Brophy and Catherine Lee.

Monday, 14 April 2008

Mental Acts at the High Court

In Rajesh Kapur v The Comptroller General of Patents, Designs and Trade Marks, a judgment issued at the end of last week, the issue of another of the exclusions from patentability under section 1(2) came to be decided. The judgment is, at the time of writing, not yet available from BAILII, but IPKat readers can download it for free via the IPKat's Google Groups file server.
[18 April update: it is now on BAILII here (thanks G)]

Mr Justice Floyd had the task of deciding whether the Patent Office had been correct in refusing Mr Kapur's applications on the grounds of being mental acts and computer programs as such. The eight similar inventions all related to document management systems, and a representative one of them was chosen for the argument, relating to a computer system for recovering deleted and overwritten files. The applicant had, in an (apparently misguided) attempt to overcome an objection relating to the computer program exclusion, drafted the claims to not explicitly state that the invention was to be carried out on a computer. In refusing the applications, the Patent Office hearing officer argued that, since the claims were not specifically directed to being implemented on a computer, the claimed method was nothing more than a mental act. Bearing in mind that the claimed invention related to a method for recycling document data in a system filestore, requiring various procedural steps including creating tables and separate filestores, Floyd J considered that the mental act exclusion was in this case being used in far too broad a way, and commented:

"In my judgment the narrow view of the [mental act] exclusion is the correct one. More specifically I think the correct view is that, provided the claim cannot be infringed by mental acts as such, its subject matter is not caught by the exclusion. It seems to me that if this were not so the scope of the exclusion would be unacceptably broad, as well as being uncertain in scope. It follows that the exclusion will not apply if there are appropriate non-mental limitations in the claim. In those circumstances it will not be possible to infringe the claim by mental acts alone, and the invention will not comprise a method for performing a mental act" (paragraph 26).

The judge considered that the inventions were simply applications of a computer's ability once programmed to manipulate and store data, and the hearing officer was therefore correct in deciding that the claims lay solely within the computer program exclusion. He did not, however, think that the hearing officer was right to rely on the mental act exclusion, and therefore ordered the Patent Office to "go on to examine them to see whether anything remains". The appeal was therefore allowed, but only to the extent that it related to the mental act exclusion.

The IPKat thinks this is a very odd decision, and not just because apparently irredeemable applications are being sent back to the Patent Office to be reconsidered. He certainly wishes Mr Kapur good luck in being able to persuade the Office that anything patentable does indeed remain. Given Mr Kapur's tenacity (he represented himself in this case, although "with great restraint and courtesy" according to the judge), the IPKat thinks it is not beyond the realms of possibility that we may be seeing a further judgment with his name on.

What makes the decision odder though is the reasoning behind overturning the Office's decision on the mental act exclusion. Applied to any of the other exclusions, this same reasoning would not seem to work at all. If a claimed invention can be infringed by an actual mental act, this seems like a good reason to consider it as being excluded. However, saying that the exclusion would not apply if appropriate "non-mental limitations" were included sounds dangerously close to the "any hardware" test that Jacob LJ derided in Aerotel/Macrossan, and not the sort of reasoning that would (normally) work for any of the other exclusions. This would not, of course, help Mr Kapur who will in any case have to argue that his inventions are something more than mere computer programs, but it does yet again illustrate how courts are forced into considering the different exclusions in different ways, even though they are all found under the same heading. Can there really be a valid overarching test to cover them all?


Anonymous said...

news flash: the valid overarching test has been found by the EPO Boards of Appeal. They called it "technical character".

David said...

Very funny. And I suppose you also think it's turtles all the way down?

Anonymous said...

But what would be the idea behind considering "a method for recycling document data in a system filestore, requiring various procedural steps including creating tables and separate filestores" a mental act? What does creating tables in a system filestore have to do with a mental act?

(And I do agree with the other anon.)

Anonymous said...

Just because David doesn't like the EPO test of technical character, it doesn't take away from the fact that they have developed a test that can be applied with some consistency across the differing categories of excluded subject matter. Got to be better than the haphazard approach to each category in the UK?

David said...

I don't agree that the technical character test is or can be applied with consistency across the differing categories, at least while the term technical remains undefined (and the whole point of the term is, as far as I can see, that it remains undefined). Under it, business methods are treated completely differently to computer programs. Whereas computer programs now are seen as being inherently technical, business methods are not. This is not consistent. The UK test might not be any better, but at least it tries to be honest in considering the exclusions equally, rather than bringing in a non-defined term to justify prejudice against certain categories over others. Unfortunately, the high court appears unable to follow the Aerotel test without mucking about with different distinctions for each category, as we see in this case and with the recent Symbian case too.

Anonymous said...

The concept of technicality as used in the EPO approach divides the problem in two parts:
1) what features or effects are excluded;
2) how to deal with these.

A difficult problem is cut into two subproblems. Divide and conquer.

Part 2) is pretty much solved: T 641/00, COMVIK.

Part 1) is more tricky. Some things are clearly "technical", some things clearly are not. The question is where to draw the line. As case law develops this will (hopefully) get clearer.

So indeed "technical" lacks a clear cut definition. However, it's not very helpful to infer from that alone that the whole EPO approach is useless. So many legal terms do not have a clear cut definition. It's often the whole point of having the term.


About computer program / business method. Imo the EPO treats the term computer program (as such) in Art. 52(2) as meaning "a pure algorithm". Once you have it embodied in a medium it's not a pure algorithm, once it's interacting with hardware it's not a pure algorithm. Imo the term "computer program" adds nothing to Art. 52(2). Remove it and a pure algorithm will continue to be "non-technical".

And if you ask me, it only makes sense to remove the term (or maybe replace it by "algoritm (as such)" as one of the clearly non-technical things in the list). It is silly to have technology-dependent terms in patent law. Mere algorithms are and have always been as (non)patentable as musical scores and horror stories (as such), even before computers were invented.

Anonymous said...

Quoting David: "Whereas computer programs now are seen [by the EPO] as being inherently technical, business methods are not. This is not consistent."

Yes it is! "Technical" might be ill defined, but taking any sensible possible definition of "technical" while a pure algorithm absent any physical constraints or considerations is fairly self evidently a computer program as such, once embodied in a medium or interacting with hardware it clearly has technical character and so is not a computer program as such. Of course, depending upon what it does, it might not be inventive if the new bit is (for instance) a business method).

Business methods, on the other hand, absent any physical constraints or considerations, are clearly not technical (according to any definition you care to mention).

As the previous anonymous mentioned, many (most?) legal terms (deliberately?) don't have a clear cut definition. Looking at the Aerotel test for potentially unclear terms: "3. ask whether it falls solely within the excluded subject matter". Well if the excluded subject matter includes "a program for a computer ... as such" that is hardly a model of clarity is it?

The reason "technical" remains so important in the UK is because the much vaunted Aerotel test must give the same result as the binding precedent of the Court of Appeal, e.g. Fujitsu "mere discoveries or ideas are not patentable, but those discoveries and ideas which have a technical aspect or make a technical contribution are". Or to put it another way, quoting Mr Justice Patten in Symbian at 58 "What is clear from the authorities is that the question whether the invention makes a relevant technical contribution has to be asked because it is the determinant of whether Art 52 (3) has any application. Whether it is asked as part of step 2, 3, or 4 matters much less than whether it is asked at all". Or, if I may be so bold as to paraphrase Patten J "you can run, but you can't hide from the question - is it technical". "Technical" is the determinant of "as such".

David said...

The above sort of argument just goes to show how ridiculous it is to rely on definitions of words that are not even in the relevant statute, let alone used in the context the above commenter needs. See my earlier post for a further explanation.

Anonymous said...

The requirement of technical character, although not explicit in the EPC, is inherent in the notion of "invention" in Art 52(1) EPC. A patentable invention must lie in a field of technology. Additionally, the rules make clear that the state of the technical art is the basis for the assessment of invention. Hence the invention must relate to a technical field R42(1)(a), it must concern a technical problem R42(1)(c) and it must have technical features in terms of which the matter for which protection is sought can be defined in the claim R43(1).

Even if the "technical" test is ridiculous:
1. It works.
2. It arguably is based on statue (S1(2) PA 77 must be interpreted in line with the EPC provisions, S130 or something like that)
3. The EPO say you must use it and you won't get anywhere (or serve your clients best interests) by trying to argue the EPO are wrong.
4. Binding precedent in the UK says you must use it.

You may shoot it down via fine intellectual arguments all you want, but why bother? It is what it is.

David said...

1. I can come up with any number of tests that will work, but they are all bound to have limitations.
2. There is no basis in A52 for the technical character test ("technology" does not equate with "technical"; it is a different word, in case you haven't noticed), just as there is no basis in A56 for the problem-solution approach. They are useful tools, but should not be a substitute for the actual law itself.
3. The EPO say we must use it, and so we must, but in order to know the boundaries of the test we need to know what it actually means. This does matter for advising clients. If my client wants to get a patent relating to a business method, I need to be able to advise him of his chances, and if necessary argue the case according to the test.
4. Binding precedent in the UK is actually now quite vague, if not contradictory, since Aerotel says the fourth step shouldn't be necessary.

Why bother? Why bother with anything? Shouldn't we all just give up and let the EPO decide for itself without any argument? If the EPO sticks with the technical character test, then why not try to extend its boundaries as much as possible? It's been done for computer programs, so why not for other excluded areas?

Anonymous said...

1. Agreed. Never said it was perfect.
2. Actually yes I had noticed that they were not the same word. Not quite sure why you feel the need to get personal. I also said that "technical" was not explicit in the EPC. As you say the technical character test, just like the problem-solution approah are useful tools. They are developed in order to help apply the actual law not replace. We could just bang our heads again and again by sticking to the wording of A52, but given that hasn't seemed to work in the past we probably should be prepared to think about it a little more and try something that does work. Also, I could point out that the Aerotel test finds just as little explicit textual basis in Art52.
3. Agreed. When I said "the EPO say you must use it" I meant precisely that as it is the test we must know the boundaries of it and what it means, and we must of course argue the case according to the test.
4. Binding precedent in the UK means that the Court of Appeal does not have the authority (normally) to overturn its own binding precedent. If it says in Fujitsu that discoveries and ideas which have a technical aspect or make a technical contribution are patentable then they are and always will be until this is overturned by the HoL or statute. Where Aerotel says that the fourth step may not be necessary, for consistency with Fujitsu I believe that this can only mean that whether the contribution is technical must have a bearing on whether the contribution falls solely within excluded subject matter.

By "why bother" I did not at all mean that we shouldn't be pushing the bounndaries of patentability (where appropriate). Rather, I think it is obvious I meant why bother trying to deny the existence and utility of the technical character test, rather than working with it.

One final point, going back to your comments earlier about definitions of legal terms, it is my understanding that the framers of the EPC deliberately left the definition of "invention" and excluded subject matter open in order to avoid tying the hands of the EPO and the courts to apply discretion in the light of new and evolving technology.

Anonymous said...

Another anonymous here, who agrees largely with the anonymice above.

In response to David's numbered points:

2) In my dictionary, one of the meanings of 'technical' is 'technological'. If an invention must be 'in a field of technology', then it must be 'technological'. Hence, it must be 'technical'.

There is no basis in Art 52 for the A/M test, in particular for taking account of the prior art.

Similarly, there is no basis in Art 56 for the Windsurfer test.

3) As far as advising clients is concerned, the EPO approach has been pretty much settled at least since Hitachi. I can't say I've had too much difficulty advsing clients in this respect.

On the other hand, advising a client as to the prospects for a UK application in recent years has been nigh on impossible.

4) Indeed. See 3) above.

David said...

"Technical" is something quite different from "technology", as I understand it (and I don't see how I can get personal with someone who remains anonymous - a philosophical question, perhaps). The former term, in its broadest sense, relates to specialist knowledge in a given field (which could, for example, be in financial instruments - without doubt a highly technical area, but one that the EPO does not regard as being "technical"), whereas the latter is usually defined as being the application of physical sciences. The EPO has not, as far as I can see, defined the technical character test in this way yet.

If I could summarise my main problem with the technical character test, it would be by analogy with scientific theories. Useful scientific theories are those that explain what can be observed, and predict future observations. Non-useful theories, like all religions, do not actually explain anything but replace an existing problem with a further, more complex or larger, one. I see the technical character test one that is more like religion than an explanation of how the law works based on how it is written. If you truly believe in the test it seems obvious and clear, but that is only because you are relying on an unstated presumption that you know what is the right outcome to start with. Some comments received by the IPKat verge on the religious in their fervent support of the test, which I find a bit disturbing. I am not trying to overturn the EPO approach (this is not the place to do it anyway), but just encourage a healthy degree of scepticism.

Anonymous said...

David, everyone here is agreeing with you that "technical" is ill-defined. What people don't agree with is your conclusion that the test or the EPO approach is "ridiculous". I don't see any anonymous here being religious about the test. Actually I see a lot of well-argued comments.

Also nobody in this thread said that one should not try to extend the boundaries of the test. Of course you should when that's in the interest of your client. You may even try to get the EPO approach overturned if that's what it takes to win a case (and you feel you have arguments and your client knows chances are small). Would certainly be interesting to see a board rejecting it.

Anonymous said...

Over in the USA, some people rant away, that the judges decline to define "obvious". And in the Protocol to Art 69 EPC, are we to demand definitions of "fair" or "reasonable". Seems to me that words like obvious, fair, reasonable, which have endless scope for provoking argument, are already down to a fundamental level of definition, to keep patent law going in the right overall direction. So too with "technical" no?

Anonymous said...

I'm not getting involved in this debate this time around, but I have seen that the judgment is now up on BAILII.


Greg Aharonian said...

"What does all of this mean? That UK/EPO software patenting policies is still nonsense because of the unethical judicial reliance on the undefined "technical" as a test for patentability, nonsense that flows from theunethical piece of pre-soiled piece of toilet paper otherwise known as the EPO Board's Pension Patent decision. In this scribbled piece of childish thinking, the EPO Board across 16 pages repeatedly asserted that computer programs aren't technical, don't achieve a technical effect, don't belong to a field of technology, blah blah - only on page 17 to admit that they didn't have a definition for "technical". Those EPO judges should have been impeached immediately.

And the result since then is a string of decisions playing more games with "technical" without defining it, which has lead to two major new idiocies, the (software incompetent) Lord Justice Jacob's Aerotel/Macrossan decision, and now this contradiction that Beauregard claims are/aren't allowable. (For the record, I still offer to debate Lord Jacob, handicapping myself with a few mugs of Guinness, as long as the UK government promises not to spike my beer with estrogen).

Too many judges and government bureacrats hurl the terms "technical" and "technological" to ban certain types of patents, on both sides of the Atlantic, without a) defining what they mean by these terms, b) confessing and/or bragging that they have no definition for these terms, and c) while having little experience in science or engineering to use the terms in a sentence.

It is long overdue for a trans-Atlantic committee of lawyers AND SCIENTISTS to define "technical/technological" in a patent context. Otherwise, we will keep on seeing new decisions and policies that make patenting issues more complicated, not less.

Six and half years ago, I gave talks at EPO and in the UK on this nonsense with "technical", and nothing has changed, as evidenced by Jacob's refusal to treat the undefined "technical/as-such" in Aerotel, and these conflicting
Beauregard claims decisions. Six years and no progress. Bollocks logic."

Anonymous said...

Although I'm not getting involved in the UKIPO v EPO debate, I have now read the judgment and would like to respond to David's concerns about the judgment's apparently odd treatment of mental acts.

Where the judge said that the exclusion would not apply if appropriate non-mental limitations were included, I do not think he was referring to computer hardware (or software). In other words, hardware/software are not "appropriate" limitations.

If he had meant this, David would be absoutely right to say that the judge was getting dangerously close to contradicting the CoA, if not actually doing so outright. However, if you look to paragraph 37, you see the judge is actually talking about physical records and the physical separation of documents as being the "appropriate" non-mental limitation and I think he's spot on.

Basically, the invention was to store copies of different documents in different places. It is difficult to see how that could be called a mental act.

I'm still mulling over the reasons for the computer program rejection.


Gobhicks said...

Man, can’t even go away for a week without another excluded matter decision coming down…

One of the anonymice said: “while a pure algorithm absent any physical constraints or considerations is fairly self evidently a computer program as such…”.

I would beg to differ. A “pure algorithm” surely only becomes a computer program once it is coded in some manner. Is an algorithm even expressed in “metacode” a computer program? Is a program in a high level language to be treated differently from a machine code equivalent compiled for a particular processor (“physical constraints or considerations”)?

And “… once embodied in a medium or interacting with hardware it clearly has technical character and so is not a computer program as such. Of course, depending upon what it does, it might not be inventive if the new bit is (for instance) a business method)”. Well, yes, that’s the EPO approach, we all know that, but it ain’t the UK approach.

And “…Business methods, on the other hand, absent any physical constraints or considerations, are clearly not technical (according to any definition you care to mention)”. There clearly are definitions of “technical” and “technology” that embrace any kind of specialist knowledge. The other week a TV programme on BBC2 called "Super Rich: The Greed Game" - a very interesting look at the financial dealings underlying the Credit Crunch – included some banker referring to the various novel ways in which debts were packaged as tradeable assets as "technologies" - one eye on patentability perhaps?

Also nice to hear from Mr Aharonian. I have a lot of sympathy for Greg’s frustration over the meaning of “technical” in EPO terms, but cannot agree with his often stated view that every aspect of commercial life should be potentially patentable (even given perfect examination for novelty and obviousness, which among other things, such as complete knowledge of the prior art, presupposes a bullet-proof test for obviousness).

One thing that Greg seems to miss is that the exclusion from patentability of computer programs etc. is written into the EPC (and by extension to the national laws of all contracting states), and changing that is akin to amending the US constitution. Further, the “technical” shenanigans at the EPO could really be said to have its origins in seeking to expand the scope of patentability by facilitating the patenting of some computer implemented inventions that had been falling foul of the exclusion.

I wonder how Mr Kapur’s applications would have fared at the EPO. It’ll be interesting to see how UKIPO will apply Symbian-thinking when they look at them again.

Anonymous said...


Mr Kapur seems to have 5 pending EP applications published on Register+, so a direct comparison should be possible at some point...

Gobhicks said...

Thanks anon - interesting. Of course, it's likely to be quite some time before the EPO examines these...

Anonymous said...

I thought it was decided in CFPH case that the various exclusions were individual and not expamples derived from some coherent principle. If so, maybe there will not be a valid overarching test?

In reply to Gobhicks comment: "
Further, the “technical” shenanigans at the EPO could really be said to have its origins in seeking to expand the scope of patentability by facilitating the patenting of some computer implemented inventions that had been falling foul of the exclusion.", the problem was not that at all. The problem was (and is) that inventions in fields which hitherto were perfectly patentable were falling foul of the exclusion merely because modern means in the form of computers were being used in their implementation. The EPO boards presumably took the view that the exclusion was written without intending that effect, and sought the logic to apply the exclusions narrowly.

Subscribe to the IPKat's posts by email here

Just pop your email address into the box and click 'Subscribe':