The team is joined by Guest Kats Rosie Burbidge, Stephen Jones, Mathilde Parvis, and Eibhlin Vardy, and by InternKats Verónica Rodríguez Arguijo, Hayleigh Bosher, Tian Lu and Cecilia Sbrolli.

Monday, 24 April 2017

Once upon a time: Inventive step argumentation as storytelling

The IPKat crafting an invention story, but struggling
to identify a suitable protagonist....
It is all about the story. A few years ago, a judge commented to the AmeriKat that they felt that sometimes patent litigators can underestimate the benefit of telling a convincing invention story when opening a case. The story, after all, is used as a tool to indicate value. "Judge, my invention has value! It is worthy of protection!", the story is meant to convey. So when the AmeriKat learnt that Dai Rees, former chairman of the Technical Board of Appeal, had written a paper on the very topic entitled "Inventive Step: The Stories We Tell" she asked him if he would be willing to explain to IPKat readers how one could go about telling a convincing invention story.  Dai explains:  
"In my paper I argue that inventive step in patent cases is of necessity in telling a story, and as such should be judged by appropriate standards. After exploring the reasons for taking this position, the standard approaches used by the English and German courts and the European Patent Organisation Boards of Appeal are examined to see how well they stand up as story-telling. 
For the English and German courts and the EPO Boards of Appeal, determining whether there is an inventive step in patent cases boils down to deciding on the truth of a “counterfactual conditional”, “If the skilled person had had knowledge of prior art X at the date of filing, he would have created (or done or considered), without the exercise of inventiveness, something which fell within the invention claimed”. Defining the formal truth of counterfactual conditionals is difficult; defining a formal proof of their truth is even more so. Nonetheless every day advocates make their cases for and against inventive step and judges decide. How? If they cannot use formal criteria they must use informal ones, and at least for this particular form of counterfactual conditional there is an obvious way to do it: let the advocate tell a story of how the skilled person would have arrived at the invention, and judge whether that story is convincing or not.

To be convincing a story must have realistic characters with reasonable motivation and no unlikely coincidences. We only have one character, the skilled person. Unfortunately, he (or she) is not realistic for two reasons: 1) he has access to all the public prior art, wherever it is to be found. There is nothing to be done about this; it is a necessary legal fiction resulting from policy choices. 2) The skilled person is strange in that he does not have “any spark of inventive capacity”. In my opinion this is an unnecessary restriction – we only need to tell a story in which he doesn’t use any inventiveness, which is not to deny him any inventive capacity.

Both the English (following Pozzoli) and the German courts regularly begin their inventive step discussion by defining the relevant skilled person, who then chooses, finds or possibly is presented with a piece of prior art to start his considerations from. This is clearly a gaping hole in the story. An alternative approach would be to start with the particular prior art and suppose that for every piece of prior art there is a skilled person, an engineer with the education and experience to develop on it. This is a reasonable supposition; after all the prior art was also developed by somebody, who can more or less take the place of the skilled person. Moreover this idea is in line with the position which the courts and the EPO BoA all take in practice, that inventive step arguments can start from any prior art which works. 
If we take this view, we further have to see the first step of the problem and solution approach (PSA) adopted by the German courts and the EPO BoA, namely “determining the closest prior art”, not as an instruction to the skilled person, but as an instruction helping us to find the right story to tell. The next step of the PSA is to determine the problem solved. This is carried out by comparing the features of the invention with those of the prior art chosen. Frequently in decisions there is no mention of how the skilled person becomes aware of the problem, but this is clearly essential for the motivation of our character. This is a significant weakness in many German and EPO BoA decisions.

Undoubtedly there is more that could be said however, for example applying story criteria to cases where the skilled person is a team, or where the invention involves a mixture of “technical” and “non-technical” considerations."
Dai's paper discussing these ideas in more detail (complete with citations and footnotes) is found here.  The AmeriKat is struck by the point on "instruction to a skilled person", and more generally the PSA.  As soon as you instruct a skilled person to do anything (including reading a piece of prior art with interest or comparing the closest prior art with the invention), you already in danger of giving them some indication as to where you want them to end up or at least what the focus of the dispute is (plainly the invention, especially if you are on one side of the debate).  At that point the story seems to have already been written. Could defining your skilled person not based on the patent but on the prior art help avoid this?  You are therefore not necessarily (and knowingly) already imbuing the skilled person or team with the realm of the subject matter.  It would also lend more naturally to allowing the skilled person to identify whether they would reach out to another skilled person in an abutting profession instead of presupposing a team from the start and having to navigate those issues. Or do we all end up at the same place? What do readers think?


MaxDrei said...

I have some thoughts, but I'm wondering whether I am just not seeing the subtlety of the Rees argumentation.

In any First to File jurisdiction, the date on which obviousness falls to be judged is AFTER the inventor has drafted her patent application. The drafter of the patent application in suit already knew the invention and the prior art before they started drafting, and set the contribution to the art in the context of the prior art most relevant to the inventor's subjective perceptions of what is the technical field and the contribution their invention makes to it. The drafter will further the ambitions of their client by manipulating the field and the problem to show off the claimed subject matter in the best light. The "story" of its patentability should be as convincing and persuasive as the drafter/salesman can make it.

In judging obviousness, one has to be "fair", as between the inventor and the public. And one strives to be objective. So the notional person of ordinary skill (the legal fiction we call the "PHOSITA") is deemed to be in the technical field chosen by the drafter, and addressing the technical problem identified by the drafter, and aware of all the art. To do anything else is to load the scales in favour of the inventor.

But conversely, it isn't fair to the inventor to imbue the PHOSITA with creativity. And as soon as we attribute any given "level" of creativity to the PHOSITA (like the USA's "ordinary" level) we are into measuring amounts of creativity. That is as futile a pursuit as measuring "degrees" of obviousness (or inventiveness). Germany had that once upon a time (20 year vs 10 year patent levels) but has now given it up as unworkable). So we set their level of creativity at zero.

By now we have a system of debate (the paradigm case is EPO-PSA) for deciding "Obvious Y/N" which is workable, impartial, objective and predictable. That the PHOSITA is not like any real human being doesn't bother me at all, provided that with his help we can argue obviousness in a structured way, objectively, and with fairness to both the Opponent and the Patent Owner.

What am I not seeing?

Resp PI said...

I remember that in a recent decision by BoA, it was said the selection of the closest priori art must be consistent with the definition of the person skilled in the art. When you select a prior art document, it must be in a domain close to the one of the person skilled in the art. Then this person has the prejudice and knowledge of this field when considering the rest of the inventiveness approach.

CK said...

Dai's paper is beautifully written and touches home with the aspirations that led me through more than two decades of examining patents: you need to tell a good story to "prove" lack of inventive step in order to make it convincing and satisfying to pan narrans. In contrast, a votum for grant very often is anticlimactic, ending with something like "this idea is not suggested by the (available) prior art".
However, I cannot follow the "multiverse" scenario there are imaginary worlds, with their corresponding skilled persons, in which each promising path will be followed for choosing one of several problems or solutions - isn't this exactly the could/would problem, elevated to a meta-level? A bit like the chimpanzees hacking on a typewriter to produce the Bible? The essential storytelling element appears to be missing here, namely the motivation why the skilled person would pursue the chosen path.

Subscribe to the IPKat's posts by email here

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