The US Court of Appeals of the Federal Circuit, a specialist intellectual property (IP) court, handed down a recent decision (In re Bilski) that denied patents to a method of hedging risks in the trading of commodities. The result may seem fairly obvious to many — as such simple business ideas do not exactly require the incentive of a 20-year patent monopoly to materialize. And yet, it took the US courts some highly tortuous reasoning to reach this conclusion.
The courts qualified their rejection of the Bilski patent by noting that if a business method was “tied to a particular machine or apparatus”, or “transformed a particular article into a different state or thing”, it would merit patent protection. To this extent, the court overturned the infamous State Street case, which merely required that a method be useful, concrete and tangible in order to be patentable — criteria that almost any business method was likely to fulfil. No wonder, then, that the US has in the recent past granted a number of patents to several business methods, including a patent to IBM for a toilet reservation system on board a flight!
Thankfully, Indian judges need not venture down such a circuitous path. That a Bilski sort of invention is not patentable in India is crystal clear from section 3(k) of the Indian Patents Act, which prohibits a “business method” from patentability.
Section 3(k) also excludes “computer programs” per se and algorithms from patentability. In order to clarify the ambit of this exclusion, the government is currently evolving guidelines via a patent office manual; a process has spurred a fierce battle between proprietary software firms such as Microsoft Corp. and open-source evangelists such as Red Hat, which claim that the government is attempting to introduce software patents through the back door.
One wonders what the fuss is all about. The Indian Patents Act makes it amply clear that computer programs per se are not patentable. Period!
The term “per se” only means that if the invention involves something more than mere “computer software”, it is patentable. If the addition is merely cosmetic, such as an incorporation of hardware or other apparatus in a manner well known to a person skilled in the art, then it ought not to be granted a patent, since the inventive contribution resides solely in the computer program. We may require more fine-tuning on this, but let’s leave that to the courts rather than relying on the patent manual for conclusively interpreting this section.
Much like India, the UK also excludes software per se from the scope of patentability. And yet, owing to pressures from industry and in a bid to somehow harmonize with the European position, this exclusion has been all but eviscerated through clever legal sophistry. Consider a recent UK Court of Appeals case (Symbian Ltd v. Comptroller General of Patents), which states that if a software program makes some kind of a “technical contribution”, it ought to be patentable. One is hard-pressed to think of any software program that might fail this “technical contribution” test, particularly when a patent application covering such programs, is crafted by a clever patent attorney.
Indian judges ought to eschew such legal casuistry in favour of a simple straightforward reading of section 3(k), which excludes computer programs per se from patentability.
Some may argue that by not granting software patents, India violates Article 27 of TRIPS (Trade-Related Aspects of Intellectual Property Rights) which mandates that patents be granted to all “inventions” in all fields of technology. Given the fact that the US and the European Union (EU) have, at several times in their chequered patent jurisprudence history, found problems with conceptualising software patents as “inventions”, it would seem appropriate for India to insist that software is not an “invention” for the purposes of TRIPS.
If at all India wishes to grant software patents, it must do so explicitly through a parliamentary amendment to its patent law. Needless to state, any such amendment ought to come about only after carefully weighing the pros and cons of granting such patents. Recent scholarship, particularly a book by professors Bessen and Muerer, points to the myriad problems thrown up by software patents, including the creation of patent thickets that stultify the progress of technology, and the rather fuzzy boundaries of patent protection that trigger litigation at a rate higher than witnessed in most other areas of technology. Although this evidence might be insufficient for rejecting software patents outright, it might propel us to craft distinct rules for such patents, including more stringent norms of disclosure, shorter periods of protection and perhaps even an automatic compulsory licence.
However, till such time as the software patent exclusion remains on our statute book, it ought to be adhered to and respected. India must avoid buying into the tortuous interpretative process that has characterized American and European jurisprudence on this count.
Shamnad Basheer is the ministry of HRD chair in intellectual property law at the National University of Juridical Sciences, Kolkata. Comments are welcome at firstname.lastname@example.org