In the position paper from November 2000, V.K. Gupta—spokesperson for the Task Force on Conservation and Sustainable Use of Medical Plants and the TKDL’s first director—explains why India needs a traditional knowledge database. Gupta (2000, p. 307) begins by stating that: ‘Globally, there are two distinct and potentially conflicting knowledge systems. The knowledge system of the formal sector, of both private and public institutions, and the knowledge system of the informal sector of communities and individuals.’ While the former is well documented, institutionalized and often protected by intellectual property rights, the latter is largely oral, mostly undocumented and often hard to protect.
As the TKDL came to demonstrate, these systems are not incommensurable, but bridging the gap between them calls for new approaches to classification. One of the architects behind the TKDL explains: ‘if you want patent examiners to take this prior knowledge into account when looking at applications claiming the novelty of therapeutic formulations, you have to make it comprehensible, you have to translate it into a language they understand’ (Gaudillière 2014, p. 392). This was in line with the position of the World Intellectual Property Organization (WIPO), which also concluded that the misappropriation of traditional knowledge was partly caused by patent examiners’ inability to identify ‘relevant traditional knowledge as prior art’ and called for initiatives that could bridge the gap between patent offices and traditional knowledge documentation (WIPO/GRTKF/IC/2/6). Consequently, the TKDL needed to provide a translation not only between languages, but also between knowledge systems and make traditional medical knowledge available in a format that is accessible to patent examiners across the world, or as Gupta puts it: to prepare a ‘format to suit’ WIPO (Gupta 2001, p. 122). The first step towards creating the TKDL was thus to build the architecture and the software for sorting and storing the data in a searchable format. This was done through the so-called Traditional Knowledge Resource Classification system (TKRC), which provides a language-independent format especially suited for categorizing metadata about traditional medical knowledge. Once this was in place a team of 30 Ayurvedic experts, five IT experts, four scientists and two patent examiners began collecting, translating and classifying content sourced from a wide range of texts on traditional Indian medicine (WIPO/IPC/CE/31/6). A large number of traditional formulas were codified and by March 2003 a first batch of data was in place. Six months later a demo of the TKDL was released (CSIR 2021).
The TKRC was created specifically for the TKDL, but it was not created without precedence as it had to relate to existing standards of classification that govern the global patent systems. The position paper compares the TKDL to WIPO’s Intellectual Property Digital Library (IPDL)—an electronic catalogue of data on patents and trademarks that was also under construction at the time (WIPO/SCIT/5/5)—but with the major difference that the IPDL contains information on already existing patents, while the TKDL collects information on non-patented knowledge that can serve as prior art (Gupta 2000). Gupta concludes that the ‘[f]irst level of search in an IP office begins with the sheet of the patent application, therefore it may be prudent to create TKDL based on information similar to that of the first sheet of a patent document’ (Gupta 2000, p. 310). So, even though the content of the TKDL is not patented innovations, it is best presented in a way that mimics that of patents. Consequently, the posts in the TKRC were described according to key attributes taken from the IPDL, such as ‘title, knowledge resource, date since known, country, contact organization, abstract on usage, key words, IPC’ (Gupta 2000, p. 311).
The last key attribute—IPC—refers to the classification code provided by the International Patent Classification system (IPC): a documentation system launched by WIPO in 1971 to make patents internationally searchable (Kang 2012; WIPO 2019). In order to make its content easily accessible all over the world, TKRC based its classification system on the IPC (Gupta 2001; Thomas 2010; Sen and Chakraborty 2014; Suwapan 2016; Anilkumar 2018). The IPC is a hierarchical classification system that categorizes all kinds of inventions into five levels of subdivisions defined by a combination of letters and numbers. The classification begins with a single letter—from A to H which represents a section. This is followed by a two digit number representing a class; a single letter representing a sub-class; a one to three digit number representing a group and finally a two to five digit number representing a sub group (Adams 2000). The result is a system where all different kinds of inventions—from hat pins (A44B 9/06) to driving mechanisms for harvesters (A01D 69/00)—are classified into subgroups such as A61K 31/43 (medicinal preparations containing penicillin). The 2020 edition of the IPC contained approximately 70,000 such subgroups. The classification system is maintained by WIPO and continuously updated by the IPC committee of experts. While the signatory states are required to adapt their national patent documentation to the IPC system, they are also invited to take part in the annual revision of the classification system (Kang 2012). The TKRC adopted the same system of classification as the IPC, but developed its own set of categories that were particularly suited to its specific content. Consequently, India’s traditional medical knowledge was categorized into 25,000 subgroups that were compatible with the IPC (CSIR 2021).
Jean Paul Gaudillière (2014, p. 399) describes how TKDL draws its content from a large body of ancient texts and documents that are being deciphered, translated and classified by a group of, mostly female, practitioners of Ayurveda, Siddha, Unani and Yoga:
each of them typing in information on a classical formulation read in Sanskrit texts and using a whole set of English, Hindi, Malayalam, and Urdu dictionaries. The translations they were crafting supposed a set of scarcely obvious equivalences between the vernacular denominations of medical materials and the ‘modern’ botanical denominations, on the one hand, or between the etiological and nosological categories of biomedicine and those of the Indian medical system considered, on the other.
A standardized form was visible on each screen […], displaying the basic structure that had been adopted for the database. Coded in numbers defined in a new general classification scheme […] The final product, as displayed on the TKDL website, is a form taking the formulation as a unit but presenting it in a format any patent-document reader will find familiar.
Gaudillière’s account gives a glimpse of how this traditional medical knowledge is processed through the TKRC, in ways that translate and fit that knowledge into a format—a documentation template—which is prescribed by WIPO and the IPC.
According to the CSIR this strategy was successful on the basis that the TKDL team had, as of 2011, managed to get more than 200 patent applications based on traditional Indian medicine withdrawn or rejected (Gupta 2011; CSIR 2021). Furthermore, the mere existence of the TKDL might have had a pre-emptive effect on the misappropriation of traditional knowledge; a study conducted by the CSIR suggests that the number of patent applications based on Indian traditional medicine, filed at the EPO, had declined by 44% by 2011 (Gupta 2011). In that sense the TKDL comes across as a more systematic and efficient way to prevent misappropriation of Indian traditional medical knowledge than to merely challenge individual patents retrospectively.
Another important outcome of the TKDL is its impact on the global patent system. The TKRC is not unilaterally shaped by the IPC; the IPC has also been updated and adapted to the TKRC. Like most classification systems, the IPC is a dynamic system that is annually updated and revised to incorporate new forms of content (Kang 2012). In that regard changes to the IPC reflect developments in the body of knowledge that the patent system needs to address. Hyo Yoon Kang gives an example of how the IPC was adapted to better incorporate the recent growth of innovations within the field of chemistry, resulting in the inclusion of a new subclass for combinatorial chemistry (C40B) in the 2006 revision of the IPC (Kang 2012). Similarly, the expanding field of ethnopharmacology and the growing importance of traditional medicine for the pharmaceutical industry made traditional medical knowledge increasingly relevant for the IPC. Already before the formation of the TKDL, there was an awareness within WIPO that the IPC classification system was insufficient to document traditional knowledge, particularly regarding medicinal plants (Suwapan 2016). Consequently, in February 2001, WIPO’s IPC committee of experts appointed a Task Force on Classification of Traditional Knowledge (WIPO/GRTF/IC/2/6, § 19). In a report the following year, the task force concluded that:
the most efficient way of developing classification tools for traditional knowledge would be their integration into the IPC on the basis of its revision, in particular in the area of traditional medicine. The material for such revision could be provided by TKRC and other classification systems for traditional knowledge available in various countries. (IPC/CE/31/6, annex § 15)
Until now medicinal plants had been classified in one single subgroup, but the work of the Task Force resulted in the inclusion of a new main group called A61K 36/00 (referring to ‘Medical preparations of undetermined constitution containing material from algae, lichens, fungi or plants, or derivatives thereof, e.g. traditional herbal medicines’) in the 2003 revision of the IPC. AK61K 36/00 consists of 207 subgroups covering different categories of plants, which enable a more effective search and examination process regarding traditional medicine (Suwapan 2016).
This willingness to adapt the IPC to the TKRC jars with the common conception of international intellectual property rights regulations as a tool for a neocolonial ‘information feudalism’: a term coined by Drahos and Braithwaite and evoked by Oguamanam earlier in this text to characterize a ‘Western knowledge hegemony’ where intellectual property regimes are unilaterally imposed on the developing world by developed countries and multinational corporations (Drahos and Braithwaite 2002; Oguamanam 2008). Peter Drahos (2008, p. 153) describes how this affects the work at the patent offices in developing countries:
developing country patent offices have been integrated into a system of international patent administration, in which the grant of low quality patents by major patent offices is a daily occurrence. Developing countries for the most part have only had modest success in influencing the evolution of standards at the international level. They have little prospect of influencing the standards of patent examination in the EPO, JPO, and the USPTO, even though those standards impact on the work of their own patent offices.Footnote 1
Seen in the context of a global information feudalism, it seems remarkable that India’s TKRC has had a direct influence on the standards of patent classification globally. Oguamanam (2008) interprets this readiness to revise the IPC to accommodate traditional knowledge as a turning point in the relation between patent law and traditional knowledge, since it discards the common assumption that intellectual property rights and traditional knowledge are inherently incompatible because they rely on different knowledge traditions. The revision of the IPC could thus imply that the TKDL not only changes how we can view and approach traditional medical knowledge, but also that the administration of patents is opening up to alternative knowledge systems.