Intellectual property is an area where there is a clash of Eastern and Western philosophies at a very basic level. Judeo-Christian religions worship a Creator God, and revere the creative element in man, created in His image. In Hinduism, Brahma, the God of creation, is by far the least important in the divine trinity. In a country containing hundreds of thousands of temples, hardly any are dedicated to Brahma. The material world is in any case a mirage, Maya, and so creating more Maya causes the soul to lose itself still further. While Eastern languages typically lack an equivalent of the verb "to have", property and possession are fundamental tenets of Western civilization, and it is interesting to note how these grew to encompass intellectual pursuits as well.

For much of Western history, might was right, particularly with regard to land, the prime source of wealth in early days. Slowly, as a landed gentry began to assert itself, the concept of property rights emerged as a way to put a stop to violent and expensive property disputes. This concept worked well with land, which could not be duplicated, unlike other sources of wealth that gradually gained importance.

With the industrial revolution, machines became a crucial source for the generation of wealth. Unlike land, machines could relatively easily be copied Ñ much easier than designed afresh. To prevent others from copying their designs, machine designers kept crucial information secret, so that they could reap greater benefits via a monopoly.

With a view to saving each designer the trouble and expense of reinventing the wheel, an incentive to share ideas without fear of monetary loss was created. The concept of property was widened to encompass the design of machines. Interestingly, the concept of patents is said to have originated during the Renaissance, when England was lagging far behind the rest of Europe in technological matters: to keep ahead in the arms race, merchants returning from France, Spain and Italy in particular, were given exclusive rights to any design or process that they brought back. The original purpose of patents was to encourage industrial espionage, contrary to the modern concept, which seeks to prevent it.

Patents gave a monopoly of design usage to the original designer for a limited time as compensation for making public design details of the machine in question. The designer obtained the desired monetary benefit via this monopoly for some time, while the ideas that went into it became available to everyone. This led to rapid advancement in machine design. Enforcement of the law relied on the relative ease with which it was possible to determine that a machine had been copied from another. However, with rapid technological progress, people soon found ways other than machines to make money, where matters were more complicated.

With pharmaceuticals for instance, which were manufactured not by a machine but by biochemical processes, it was not easy to patent the end product, which might even be a chemical already existing in nature. To provide protection to the innovator in such industries, the concept of process patenting was introduced. From the concrete and immovable, property slowly came to encompass ever more abstract concepts. The distinction between product and process patents is central to the current IPR controversy in the country.

Under the 1970 Indian Patents Act, only a 7 year duration for process patents was allowed. The drug itself was not patentable. Patents granted in the pharmaceutical, food and chemical sectors could be worked without the authorisation of the patent owner, under the "License of Right" and "Compulsory Licensing" provisions. This, as pointed out by former Commerce Secretary Mr. A.V. Ganesan, who was our Chief Negotiator in the Uruguay Round negotiations that led to the GATT agreement, amounted to overkill, for these licensing provisions have never actually been applied since the legislation came into force. In other words, much of the current criticism of GATT and of the government stand is based on non-issues.

Of the 250 drugs that the WHO considers essential for rational therapy, only 10 are covered by patents. Between 1983 and 1992, 433 drugs were introduced in the world market, of which only 33 were introduced in India. Of these, hardly four or five earn substantial revenue for their manufacturers. While providing little benefit to Indian consumers, our patent legislation provides hardly any protection to the inventor. The 7-year duration of the patent, for instance, is shorter that the time it normally takes to bring the drug to market after filing the patent application.

The only real beneficiary of the patent regime is the Indian pharmaceutical industry, which has strongly backed opposition to GATT. Experience in other areas of industry has clearly demonstrated that protection from competition severely hurts the protected industry by making it complacent and uncompetitive. Without providing much help to the consumer, this legislation is not in the long-term interest of Indian industry, and at the same time conveys the unfortunate impression that the country claiming to possess the world's second largest reservoir of scientists, engineers and skilled technical manpower has little respect for intellectual property. While Indian hesitation in the matter of intellectual property is understandable in the context of its civilization, its policies make little practical sense.

It is consequently not surprising that the Indian stand enjoys little international support. It would be appropriate to evaluate what our genuine concerns in the area are, what we need to do to put our own house in order, and what strategy we might employ to garner international support. Fortunately, the applicability of intellectual property rights is coming increasingly under attack in the West itself as being ill-suited to modern needs.

Firstly, the patenting process takes too long. Prior to the 1970 act, even though patents were available for a wider range of products and processes, the process was mandated to take not longer than two years. The period it took the patent office to process a patent application was relaxed under the new law, and even with a reduced load of patent applications, the time it takes the office to publish a notification that the patent has been accepted has significantly increased to an average of about 4 years. The average time taken to process a patent is also very different from one branch of the patent office to another, as an analysis of the patents granted in a random month shows. (see Table 1.) Delhi, typically, takes twice as long as Bombay.

Table 1: Number of years it took to process the patents accepted in January 1993, split up by processing branch. Source: Gazette of India, January 1993.

After notification of acceptance, other parties have up to five months in which to file an opposition to the granting of the patent. If someone does, the process drags on — there have been several cases where 14 years had passed, the patent period had already lapsed, yet no decision had been taken on the granting of the patent! Even if no opposition is filed, it usually takes a year after acceptance before a patent is finally granted Ñ a total of typically 5 years from the time the inventor first applies for a patent. (In a country like Australia, in comparison, a patent can be obtained in a matter of weeks.) This problem is likely to get a lot worse once, under the provisions of GATT, the range of patentable items expands.

Further, there is lack of clarity about what is patentable and what is not. While our current legislation excludes essentially biological processes from patent protection, patents have been issued in the area. For example, patent number 165830 was granted to an Australian company for the process for the preparation of a recombinant DNA molecule. Then again, very similar patents have been applied for in different branches of the Patent Office, where one branch has rejected it, and the other accepted.

And finally, new technologies need special attention when revising our patent legislation, as discussed in the matter of software below. Officially, India does not grant patents to software, though patenting does take place via the back door. When combined with some hardware, the entire system is patentable, which protects the software as well.

The classical approach for dealing with information is the concept of copyright. In keeping with the approach of separating the idea from the expression of it, a story could only be copyrighted after it was captured in a physical object, such as a book or a record. With information today moving at the speed of light and changing as various people interact with it, hardly setting foot in the physical world, such a distinction seems increasingly absurd.

There have been attempts to use copyrights to protect software, however, except in cases of blatant copying (such as software piracy), the approach hasn't really worked. For instance, Apple tried unsuccessfully to prevent Microsoft and Hewlett Packard from using graphic user interfaces resembling the one on the Macintosh, while Lotus lost its lawsuit seeking to protect the "look and feel" of its spreadsheet against Borland.

In India, software can only be copyrighted, it cannot be patented, which, until recently, was international practice. In the US, the situation changed with the case Diamond v/s Deihr, which related to a process for curing rubber. On the issue of whether the use of a computer program in the process was enough to render it unpatentable, the court ruled, sensibly, that it was not.

However, the US Patent Office took this to mean that software techniques as such were patentable. The flood-gates were opened, and thousands of all kinds of commonplace techniques were patented. For instance, Ronald Katz has 25 US. patents issued over six years covering a number of hardware and software interfaces that allow people to use a telephone to communicate with a computerized information service. Millions of people worldwide routinely infringe them, including government agencies.

This poses a dilemma. If unlimited patenting is allowed in the area, software writing will become impossible, because each program would infringe hundreds of patents. As the League for Programming Freedom put it, "imagine that each square of pavement on the sidewalk has an owner and that pedestrians must obtain individual licenses to step on particular squares. Think of the negotiations necessary to walk an entire block under this system. That is what writing a program will be like in the future if software patents continue." In 1994 alone, more than 4500 software patents were issued in the US, and there is a rising trend.

The problem is further complicated by the fact that the patent is granted from the date of filing. Until notification of acceptance, which may be a few years later, a software writer has no way of knowing that software under development is violating the patent. To continue the pavement analogy, the pedestrian discovers some years after taking a walk that a particular stone that she had stepped on is now owned by someone, and the action of stepping on it in the past must be paid for.

When a patent is granted, all software that violates it must be pulled off the market, or license fees paid. Small, and even medium-sized companies could easily go bankrupt as a result — for no fault of their own. Large companies, besides being able to engage a large number of lawyers to contest and postpone the matter in court, also have in their arsenal several patents themselves, and can usually get around having to pay substantial royalties by cross-licensing.

One of the few hi-tech areas in which Indian industry is potentially a major international player is software. Besides, software is now a part of virtually every modern process and it is also a major product of the information age. Developing a sensible approach towards software would help us deal with the increasingly information-based processes and products that will reach our patent offices.

If software is not patentable in the country, but it is in the countries we export to, the hardship for Indian software companies increases. Not possessing any patents of their own, they are at a severe disadvantage when negotiating with foreign companies for the rights to use their patents. Clearly then, the question of software patenting needs international consensus, and it is in India's interest as a software exporter that the issue be resolved quickly.

The problems the software example highlights are indicative of increasing problems that the patent office will face in dealing with hi-tech in all countries. Many problems are a result of a lack in expertise. This has many causes. Firstly, with increasing specialisation, there are now a large number of such areas, and it is hard to have that many experts in each office. Secondly, experts in emerging areas typically command premium remuneration in private industry, which government is not able to match. Thirdly, as the 1994 Annual Report of the European Patent Office pointed out, "Patent offices are aware that software-related inventions pose special problems for searches.

The major difficulty is that much of the relevant state of the art is to be found not in patent documents but in non-patent literature, eg articles in the specialist press, conference papers and company documents. Accessing these sources can be laborious and the material is often not ideally presented for search purposes. Moreover the patent classification system has not kept pace with the development of software technology." Finally, secrecy requirements prevent the patent office from making details of any patent application known till it has been accepted, thus preventing the office from consulting industry experts at large.

Without suitable expertise, the patent office is unable to decide which so-called inventions are "obvious" and thus undeserving of a patent. There have been cases of items people in the field considered so obvious that nobody bothered to publish a paper on them, yet they were accepted by the patent office. It does not help, of course, that patent applications are well clothed in legalese. The patent office traditionally considers even incremental improvements and uses of the same technique to new applications to be patentable. This attitude, in the case of software, has produced outrageous results.

Patent offices worldwide lack adequate resources to check patent applications against publications in the area in order to be able to judge uniqueness. All they can then fall back upon is the so-called "prior art" cited in the application itself. Greg Aharonian, who runs the Internet Patent News Service on the Internet says, "The majority of issued software patents still are citing no non-patent prior art, due to these companies not sending in any non-patent prior art of their own, and relying on the PTO not to find any non-patent prior art...IBM has the most questionable behavior of all software patenting companies. I estimate that IBM has at least 20,000 internal documents relating to software (its own patents, technical disclosure bulletins, IBM-published journal articles, IBM employee published articles, and government technical reports), and certainly the resources to organize such information and make it easily available to its patent lawyers (heck, I have about one third of these materials in my databases). Yet about half of all IBM software patents cite no non-patent prior art, not even any IBM prior art, let alone items they could find by spending a few bucks to search INSPEC." Other companies he considers guilty of similar behaviour include Motorola, Siemens, Intel, Matsushita, Hitachi, DEC and ATT Ñ a veritable who's who of the corporate world.

It is necessary that our patent offices have the facility to search international databases in order to properly judge the degree of uniqueness of a product or process. Via the Internet, this is possible at low or no cost. The public too could be provided access to such search facilities, so that researchers in India do not re-invent the wheel (the Patent Office has already taken some steps towards computerisation, however much more needs to be done, as well as public awareness of these activities increased). Once such facilities become commonly available, the law can require an appropriate level of citation of prior art (for instance, the average number of cited references in technical papers published in the area), so that abuses of the system are minimised. Further, patent applications should be opened to the public early in the processing phase, so that experts from the industry can point out obviousness where is occurs, and prevents the granting of frivolous patents.

Our experience with drugs and chemicals has showed that the Indian patent regime cannot be at odds with that of the rest of the world. Therefore, we should consider granting patents to software and other hi-tech products and processes when framing our new patent legislation. At the same time, we must meaningfully contribute to the formulation of international guidelines for the granting of software patents which are sufficiently tight so that the industry is helped, not crippled. For this to happen, it is necessary for technology experts, representatives of the patent office and government as well as some of our fine legal minds to come together on this issue, so that a national consensus can be obtained.