UK patent lawyer responses to GregNews on Technicity

[Hartmut Pilch <phm@a2e.de>]

All of this is highly relevant to the discussions of this forum,
but I'll try to abridge it as far as I can.

The discussants all agree that they want patents to be available at
least for very smart software and business method ideas, but some
of the patnews readers point out in quite a frank and insightful way
which political constraints exist and why under these constraints it
is difficult to do away with the concept of "technical".  Some of this
is worth quoting in the context of 

        Voided Terminology and its Contribution to Political Engineering 
        in the EU
        http://swpat.ffii.org/papers/eubsa-swpat0202/tech/#newspeak

The editor of this collection, Greg Aharonian

        http://swpat.ffii.org/players/aharonian/

makes an interesting distinction between legal terms which are difficult
to "ground" (= reduce to easily communicatable primitive concepts)  
(such as "love", "good", "obscene" etc) and terms such as "techn-", which
he thinks should be grounded in some highly sophisticated "scientific"
theories.  

Gert Kolle 1977 (http://swpat.ffii.org/papers/grur-kolle77/) also writes
that it is desirable to ground legal terms outside the area of law and 
that the classical definitions of the "technical invention" by reference 
to controllable forces of nature provides such grounding in a secure way.

Of course Aharonian and his patent lawyer friends also know this, so they 
gloss over the issue quickly, by dismissing this knowledge as 
old-fashioned, "extreme" or in contradiction with what they claim to be 
the latests rocket science (in reality just impermissible inferences from 
articles taken out of context, see discussions between Greg and 
the physicist Carsten Svaneborg earlier on this list).
 
Greg also endorses the CAFC fallacy, that "if you can not define 'business
method' you can not exclude anything for not being a 'business method'".  
One could as well say "If you can not define what is a 'technical
invention', then you can not grant any patent at all", and that position
would be more in line with liberal constitutions, which treat freedom as
the rule and exclusivity as the exception.  

Also, Greg and his friends agree on not wanting Art 52 EPC to mean
anything.  As usual, Greg says that the "as such" clause was inserted by
the legislator to make it meaningless.  See

        http://swpat.ffii.org/analysis/epc52/

for more about this debate.

I have tried to delete this propaganda and retain the new and insightful
parts of the text.  The final comment by a UK patent lawyer didn't need
any removal, and even Greg seems to have found that it put things in so
straightforward a manner that he didn't need to contradict or comment
anything.  It is written with a rare sense of objectivity and explains
well why inferences from dictionary definitions, as proposed by Aharonian,
are of limited relevance to the debate.

----------------------------

!20040315  UK patent lawyer critiques my "technical" comments

    For some years, I have been criticizing the Europeans for pretending
to be unable to define "technical" in IP matters while they definitively
use "technical" to exclude some software patents and all business method
patents.  Recently, the CIPA Journal (journal for the UK IP lawyers'
society) has published my more passionate critiques of this linguistic
nonsense (not being able to define 'technical').

    In the Feb. 2004 issue of the CIPA Journal, Simon Davies, an IP lawyer
at the UK firm D Young & Co. critiques my critiques.  Both Simon and I
agree on much, but where he disagrees with me below, well, let's just say
it is impossible for me to resist smashing the intellectual softballs he
throws my way (or the full tosses he crickets my way, a fulltoss for some
reason making me think of the Pension decision :-).


                              ====================


A REPLY TO MR AHARONIAN

By Dr Simon Davies
CIPA Journal, Feb 2004


As Chair of the Forum on Software Inventions at the EPO (reported in the
November CIPA [2003] 643) I am pleased that we have received such an
extensive analysis on the EPO presentations from Mr Aharonian, (January
CIPA [2004] 23) who is a well-known commentator on and critic of patent
systems, especially in relation to computer technology. After all, the
whole point of the Forum (as its name suggests) was to encourage dialogue
and debate. However, having been somewhat critical myself of the EPO on
the day of the Forum, I now feel compelled to defend them, at least in
part, against the remarks of Mr Aharonian.

Let us first turn to the meaning of `technical'. Perhaps the major
complaint of Mr Aharonian is that the EPO fails to provide any definition
of this term, yet it can play a decisive role in determining patentability.

Unlike Mr Aharonian, I am not a big fan of definitions. After all,
definitions simply express the meaning of one word in terms of other
words. But where do you get the meanings of these other words - from
other definitions? It seems to me that the whole thing risks degenerating
into (meaningless) circularity.

    [Greg note: since Simon is lawyer, he has a thick skin, so I have
    no problem saying this:  Simon's objection is pure nonsense with
    regards to "technical".  Yes, defining words is a complicated
    philsophical problem.  There is even a word for deciding when to
    stop defining a word in terms of other words - "grounding".  There
    are linguistic and computer conferences on this very subject,
    for example, in 1998 the American Association of Artificial
    Intelligence had a workshop titled "Grounding Word Meaning".

    [...]

[...]

Mr Aharonian continues by criticising the Pension Benefits decision
(T931/95). This is not a decision that I would wish to defend (and indeed
I criticised it myself at the Forum). In particular, it seems to me that
the Pension Benefits decision significantly shifted the interface between
the assessment of statutory subject-matter and of inventive step without
any consideration of whether such a shift was compatible with existing
case law. I find it regrettable therefore that the rationale of Pension
Benefits has been so rapidly adopted in both the EPO Guidelines and also
the proposed Directive on software patents. On the other hand, my
impression is that most European patent attorneys are not unduly troubled
by the ultimate decision in Pension Benefits, namely the rejection of the
application, given the absence of an `invention' (as the term is commonly
understood in Europe).

    [Greg note: talk about circularity. "given the absence of 'invention'"
    What's that mean, it isn't ......... technical?]

    Look, EPO had to evaluate the claims because they articulated an
    "apparatus"  whereas 52(2) only appears to bar a "method" of doing
    business.  And the only way they could reject the apparatus claims
    was to "manufacture" some rejection on the basis of inventive step
    - by redefining "technical contribution" - that is requiring the
    specific elements which contributed to the inventive step as
    requiring that they be "technical".  (wherein information was
    defined as non-technical.

    Which is based on the false assumption that information is not
    physical, so that information processing is not technical.  The
    EPO Board decided to ignore a growing body of European physics
    to the contrary.]


It is perhaps also worth mentioning that I am also no great fan of the
leading US decision on business method patents, namely State Street Bank
(State Street Bank & Trust Co. v Signature Financial Group, Inc, CAFC,
96-1327). One reason for this is that Circuit Judge Rich holds that the
invention is patentable because it defines a machine (one of the statutory
categories of patentability under U.S.C. 101). He supports this by
enumerating first arithmetic means, second arithmetic means and so on,
just as if these were separate wheels on a car, whereas (presumably) they
are all implemented by just a single component (a processor) performing
different sets of instructions. The decision does not, of itself, evidence
a great understanding of computer architecture.

In addition, while the State Street Bank decision notes that there is
no statutory basis for a `method of doing business' exclusion in US
national law, there is such an exclusion explicitly in Rule 39 of the
PCT - signed in Washington DC. One assumes that at the time the PCT was
signed and ratified, the relevant US parties, presumably including
Congress, were of an understanding that US law did incorporate such a
method of doing business exclusion. (Note: Certainly the PCT goes to some
length to ensure that it is compatible with US law.)  It is not the fault
of the EPO that some 28 years after the PCT, the US - or at least one
judge - changed its mind on this point.

                                      ====
[Comment by John Machek (?)]

[...]

So far I have concentrated on how to apply the law at it stands at
present, including the method of doing business exception. However, it
is clear from Mr Aharonian's article (and I have also heard him at a
conference in Brighton a couple of years ago on this same point) that he
is in favour of more extensive patentability than is currently permitted
by the EPO. In particular, Mr Aharonian supports the patentability of
business methods, where the `invention' lies in applied science, such as
economics. For example, one might discover that a particular form of
time series analysis is particularly good at predicting stock market
fluctuations, and want to protect this by a patent filing.

There is much to be said in favour of such a position. After all, if patent
protection is thought to stimulate investment for applying mathematics in
certain activities (such as electronic signal processing), then why should
the same not also apply in other activities of a more financial nature.

However, at present, I am not sure that Europe is ready to move in such
a direction. The consultation exercises performed by the UK Patent Office
and the Commission in preparation for the Directive came out largely in
favour of maintaining the status quo for patentability (some wanted to
narrow it, few to broaden it).

In fact, my understanding of Mr Aharonian's position is that he feels
that such inventions should already be patentable, that they are indeed
`technical'. This is in sharp contrast to the European Parliament,
which recently voted that `data processing' itself was not a form of
technology. Of course, the European Parliament's position is extreme,
and, I think, untenable.

    [Greg note: and not based on much science.  Politicians trying to
    ground scientific terms is about as productive as watching them
    trying to catch greased pigs.]

                              ====================


(Note: here is an UK lawyer's assessment of the politics of "technical".)

The problem in Europe is primarily a social, political, and cultural 
one rather than a legal one.  Among the plethora of substantive such 
issues which the EU Parliament and Council struggle with these days, 
this issue of the patentability of "computer-implemented inventions 
(read software based)" would appear to be one of the minor issues.  
However, it appears imbued with the same overall socio-political 
underpinnings: "in the EU we don't like too much IP protection,"  
"there is little or no proof that IP protection leads to increased 
production, profits, or industry growth in general," "we Europeans do 
not want to be like the Americans"  "the EU needs to be in control of 
IP rights, not the European Patent Office,"  "TRIPS does not mention 
software expressly so lets leave TRIPS out of this,"  "the European 
patent law has historically said computer programs and business methods 
are not patentable as such and the EPO has subverted these dictates by 
issuing such patents," "software patents are detrimental to European 
programmers and small enterprises," etc., etc. The many articles in the 
US which deride software patents, and decry the trivial patents issued 
by the USPTO and the quality problems which we acknowledge in the USPTO 
are capitalized upon to show that the US system is broken and certainly 
should not be followed by Europe.

Thus the positions articulated (which you attack on technical grounds) 
by the EPO really have nothing to do with "technical" as such.  They 
are saying "we are not like the USA" "we are tougher on computer 
implemented and especially business method inventions, and we will be 
so because the Europeans want it that way"  "our quality is immensely 
better than the USPTO and we are working to make it better"  etc., all 
arguments to hopefully say to the EU Parliament and Council, "look, 
make this directive follow our current rules on software patents as we 
are in sync with what you want, but do not try to kill them completely."

As to what is happening with non-European software patent holders and 
how they feel about all of this,  I believe most non-European software 
players fall into two categories: gorillas (IBM, Microsoft, Oracle, 
Sun, TI, Lucent, HP, Intel, Hitachi, and other Asian software biggies) 
and all-others.  The gorillas appear to be following this closely but 
generally (except for IBM, Microsoft, Lucent, TI, who actively license 
their patents) would seem to prefer a no-software patent regime in 
Europe for business strategic reasons, and therefore these companies 
appear to be silent at the moment as far as responses in protest of what
the Parliament has done so far.  The active licensers (IBM, Microsoft,
Lucent, TI) would lose a big royalty base under a no-software patent
regime in Europe but they too recognize that they are seen as biased
proponents of software patents and their arguments handily dismissed
by the European politicians.

In the "all-other" category I think these companies are not "ambivalent"
but rather "in the dark" as to what is happening and what could
potentially result in Europe.  I believe it is not on their radar
screens.  No one is talking about this publicly, partly because they are
not paying attention to what is happening, hope it will just go away,
or are afraid to take a stand one way or another.  I think this is a
looming disaster for the growth of a knowledge based economy in Europe,
and agree that a "no-software" patent regime in Europe would be opposed
by most knowledgeable European and non-European software players.  And
as you point out, the logical problems with software copyright make the
looming disaster even bigger.

                              ====================

Here are some comments from another UK patent lawyer:

    You wonder what the "political requirements" of the European
Parliament might be. I think the answer is in the observation that the
whole process is utterly irrational. Software activists, in particular
the FFII have managed to make a statement that caused the EP to react
to the Commission's proposal the way it did. Given this situation, I
believe the EP can not accept their decisions to be overruled just like
that, in whatever interest. I am not a politician but I could understand
that in the present situation, the EP is not prepared to accept any
argument simply to withdraw their (numerous) amendments, or even a
considerable portion. You are probably aware that the EP is not a very
respected institution - it's a lengthy process to create the "United
Europe" in the minds of the European Citizens. So logically they want
to demonstrate they have some power at least. (The number of people
showing up at EP elections continues to be very low.)

    But at least partly the problem is rational indeed, in my view. The
patent system is abused occasionally, and for instance the FTC hearings
on the Patent system have been organized because there were some serious
concerns, I guess. "Patent quality" remains a concern in the US, if I am
correctly informed. While the opinions on software patents are divided
in Europe (probably because everybody has his own view on what is meant
by a "computer-implemented invention"), the resistance against business
method patents is much more unanimous. I am aware that there are indeed
very advanced business methods, that definitely deserve a patent. But I
am afraid most people consider B patents as another ridiculous American
invention. Again, you may find that irrational. But there are irrational
realities to face ...

    You expect at least courts to act rationally and to adapt to changes
in society. I must disappoint you. I have studied many decisions of
the German Federal Patent Court and the German Supreme Court. They
still stick to the technicality requirement - which is however interpreted
in various ways (I am sure you are well aware of all the variations). It
is really a word game. They do not use economic arguments. For the EPO,
it is roughly similar. If they (occasionally) appear to become more
inclined to grant S/W and B/M patents, the reasoning typically is not
economically founded. They seem to stick to the old (German) paradigm of
"Begriffsjurisprudenz" - which was abandoned nearly a century ago in
other areas of law ... (the alternative is "Interessenjurisprudenz"
which might be more easy to understand if you are not familiar with the
German language).

    After studying patent law and the patent system now for some time,
I have noticed that really nobody knows whether patents are good for
"mankind" or not, if you consider the whole picture. Machlup in 1958
said "we don't know, we don't have facts", and I was amazed to see that
economists have made little progress since then. They make highly
sophisticated mathematical models all the time, but at the same time
they still admit the don't have facts (recently Adam Jaffe again). Of
course, you can't perform a test comparing "a world with patents" and
"a world without patents". But I am sure there is much more we could know
about the patent system. Your hear stories about startup firms making a
fortune because they own the patent on some highly creative innovation.
At the same time, I get told, software patents are really a "zero sum
game": it is just a matter of hygiene (at a high "transaction" cost
though) to collect a patent portfolio for cross-licensing purposes. New
entrants however have a tough time to enter into this game. I'm not saying
this is the truth, and relevant, but such rumors go on and on (as long as
we do not have more facts).

-- 
Hartmut Pilch, FFII & Eurolinux Alliance              tel. +49-89-18979927
Protecting Innovation against Patent Inflation       http://swpat.ffii.org/
300,000 votes 2000 firms against software patents    http://noepatents.org/