Lead Author: Siva Thambisetty
Organization: London School of Economics and Political Science, Law Department
Country: UK


It is widely accepted by legal commentators that there is often a misalignment in the private rights of inventors and the public interest, reflected in controversies around the access to patented medicines and other technologies. This short paper argues that the resultant policy incoherence can often be traced to the distinctly self-referential way in which patent law functions, with remarkably little interpretational space for consequential analysis of the effect of rights once granted. Certain features of the Patent system including techno-legal terminology and the frequently specialized nature of courts create a perverse situation in which policy coherence is undermined by the nature of textual coherence in patent law - a form of coherence devoid of substantive or purposive constructions, and often achieved through highly specific terms of reference unmoored from other areas of law and governance. This dynamic of ‘textualisation’ that I describe is endemic to patent law. In addition to being the driver of a steady accretion of patent rights that may not be in the public interest, it also severely inhibits the internal ability of the law to reform itself. To this end I propose a three step legal test exogenous to the patent system that would apply internationally and in domestic jurisdictions, to improve policy coherence by facilitating a more consequential approach to the granting and use of exclusive patent rights. This test could legally and politically represent the strengths of the Berne Convention and the so-called ‘Doha solution’ respectively. Here I propose the wording for such a test and provide a brief explanatory note on how it may be applied or interpreted.


Why Policy Incoherence is Endemic in Patent Law: Proposal for an Antidote

Over time, most leading jurisdictions have seen an accretion of rights in patent law – both in terms of kinds of innovations that can be owned and what that ownership means in terms of exclusive use and exploitation. This accretion in UK, European and US law – is often realized through specialized features of the patent system. One such feature is the common use of ‘technolaw’ (1) - an intermediary between legal and technical standards that allows the operationalisation of legal doctrine within a given technological context. It is material to a number of standards, such as novelty, disclosure and infringement. Since the language of the law and language of technology are not designed to be in easy concert, technolaw involves a greater degree of constructed meanings than other kinds of law, particularly in rapidly moving technologies where technical facts are prone to revision or where new terminology is emerging. The opportunity for manipulation in meaning is therefore considerable.

The institutional and linguistic architecture of the patent system creates a perverse situation in which policy coherence is undermined by the nature of textual coherence in patent law; the latter is a form of coherence devoid of substantive or purposive constructions, and often achieved through highly specific terms of reference unmoored from other areas of law and governance. I define policy incoherence as the inability to take a balanced and comprehensive account of several public interest issues simultaneously while resolving any question of rights over inventions, or limitations to patent rights. It is the preoccupation with textual coherence that fuels the misalignment in the private rights of inventors and other public interest concerns like international human rights law, trade rules, public health, and the regulation of the competitive process. (2)

Close scrutiny of the text of patent in US, UK and EU laws, as seen in legislation as well as patent examination guides, evidence mechanisms of ‘textualisation’ of the law where substantive meaning is compacted to become a tool of persuasion that communicates and influences through linguistic modalities. The modalities that contribute to this dynamic of textualisation as a meaning-making enterprise include ‘stratification’ and ‘incrementalisation’. Textualisation also enables the easy transplantation of patentability standards to other jurisdictions through networks that legitimize and entrench such outcomes. (3) Typically, textualisation can be demonstrated using the examples of diagnostic methods in European law, Swiss-Type claims in UK and European law and limitedly through the subject matter eligibility rules recently developed in the US Supreme Court (4).

Many have suggested that the problem of access to patented medicines can be solved through a more reasonable interpretation of both patentability as well as the limitations such as they exist in patent law (5); however textualisation as a process makes it very difficult to bring in alternate or reformed interpretations of the law. The example of the interpretation of the exclusion of diagnostic methods in European law shows how endogenous textual coherence is prioritized over policy coherence.

An Example of How Textualisation Works in Patent Law

Diagnostic methods for humans and animals have a long history of being prohibited from patentability. (6) Strong socio-economic policy reasons for the exclusion have facilitated statutory inclusion in a number of jurisdictions. (7) Yet European patent Examination Guidelines based on the Enlarged Board of Appeal’s decision (Enlarged Board) in Diagnostic Methods G1/04 set out a very narrow interpretation of the exclusion that appears to reserve the entire weight of the scope of this exclusion only to the most unskilled patent applicant. (8) They state that the term ‘diagnostic methods’ does not cover all methods related to diagnosis; and that ‘accordingly, methods for merely obtaining information (data, physical quantities) from the living human or animal body (e.g. X-ray investigations, MRI studies, and blood pressure measurements) are not excluded from patentability under Art. 53(c).’ In other words, many of the methods that an ordinary person would regard as contributing towards a ‘diagnostic method’ are in fact patentable.

In presenting the controversial position as final outcomes, the Guidelines paper over the intense disagreements that preceded this final position. (9) The final position– namely that a number of different kinds of diagnostic methods are in fact patentable – is expressed in bite-sized increments. The increments themselves go either towards or away from patentability, although taken as a whole, there can be no doubt that the position is one that makes it fairly easy to obtain a patent on a diagnostic method.

The Examination Guidelines based on G1/04 state that for a ‘diagnostic method’ to be practiced on ‘the human or animal body’, direct physical contact with the body is not required. But it does so by saying that ‘each of the multiple technical steps’ in the diagnostic methods must be ‘performed’ on a human or animal body, and that for each such step we must ascertain if there has been ‘an interaction’ with the human or animal body. Such interaction is not determined by type and intensity, only by the presence of a human or animal body.

The substantive statutory requirement of ‘practiced on the human or animal body’ in Art 53 (c) EPC is overwhelmed by a successive dilution in terminology – practiced to performed to interaction to mere presence – such that no actual physical contact with the body is necessary. In effect, the type of interaction with the human or animal body is irrelevant to patentability such that visual observation, X-rays and invasive techniques would all equally qualify. Each of these terms is collated in G1/04 from a number of previous Technical Board of Appeal (TBA) decisions where least common denominators of ‘interaction’ appear to have been set out in no particular coherent order. So framed, the stipulation that the diagnostic method be ‘practiced on human or animal body’ has the potential to filter out a greater number of methods than if the actual presence of a human or animal body were required. This increment therefore makes it more difficult to patent a diagnostic method.

The text of the Guidelines is stratified which is typical of a technical document where the text is often arranged in layers. The Guidelines for Article 53(c) of the EPC state (the bold highlighting is part of these Guidelines):

The claim must include method steps relating to all of the following phases:
(i) the examination phase, involving the collection of data,
(ii) the comparison of these data with standard values,
(iii) the finding of any significant deviation, i.e. a symptom, during the comparison,
(iv) the attribution of the deviation to a particular clinical picture, i.e. the deductive medical or veterinary decision phase (diagnosis for curative purposes stricto sensu).

If features pertaining to any of these phases are missing and are essential for the definition of the invention, those features are to be included in the independent claim (see Example 9 in Annex II of F-IV). Due account should be taken of steps which may be considered to be implicit: for example, steps relating to the comparison of data with standard values (phase (ii)) may imply the finding of a significant deviation (phase (iii) - see T1197/02). The deductive medical or veterinary decision phase (iv), i.e. the "diagnosis for curative purposes stricto sensu", is the determination of the nature of a medical or veterinary medicinal condition intended to identify or uncover a pathology; the identification of the underlying disease is not required (see T125/02).

One of the ways in which stratification manifests in this paragraph is by the splitting up of a claim into several ‘diagnostic method steps’ itself not a term of art in the EPC. These ‘phases’ of diagnostic method claims are presented in technical language redolent of medical or veterinarian phraseology that function as technical markers of credibility.

Stipulating that the final phase does not need to relate to a disease, but must merely be intended to uncover a pathology appears to be strengthening the impact of this exclusion, by including within its ambit ‘mere pathologies’. The reader is then directed towards citations to legal decisions which are supportive decisions of different Boards of the same tribunal - the equivalent of repeating oneself to gain authority – while ignoring a similar number of Technical Board of Appeal decisions that lead to the opposite conclusion.

The overall directionality of stratification becomes clear in the first sentence – when the Guidelines stipulate that the claims must include all of these phases in order to be a true ‘diagnostic method’ (and therefore not patentable). There is a triple negative here that is also put to great use: ‘If all the phases are not present, then it is not a diagnostic method, and is therefore not excluded.’ In other words, to patent a diagnostic method all you have to do is avoid these phases in sequence in the claims, which will not be hard to do as they are in fact a construct that can be side-stepped by the patent applicant either by showing that not all the stipulated phases are reflected in his claim, or by showing that even though all the phases are present in his claim, one or more of them are ‘not practiced on/performed on/do not interact with a human or animal body’. The result is the creation of multiple loopholes in the law that can be exploited even by the most un-opportunistic patent applicant who seeks to patent a diagnostic method.

There are at least three observations one can make about the mode of textualisation in this example. Firstly, we see no allusion to the purpose or function of the exclusion of diagnostic methods on patentability. This has the effect of muting the long history of policy reasoning behind this exclusion. Secondly, these guidelines present as settled, a deeply controversial area of the law, and one that was subject to opposite Technical Board of Appeal decisions not so long ago. There is no hint of this controversy in the Guidelines. Thirdly, the law still says ‘diagnostic methods’ are excluded, and therefore potential dissenters are subdued if not silenced, except now we have a newly constructed meaning of ‘diagnostic method’ that has a life only in the Examination Guidelines of the European Patent Office. The overall impact is of text that is staggeringly self-assured and persuasive. Reaching into the legal, policy and interpretational reasoning in order to place this text in the context of the disagreements it arose from requires the reader to be an insider – not just to the law, but also to the particularities of the technical language that forms the scaffold to the final interpretative position. The result is a self-referential text that is coherent on it’s own terms and clearly increases the possibility of patents on diagnostic methods.

An Exogenous Antidote

The mechanics of textualisation shows us that patent law, specifically the language in which the invention is claimed and denoted, is suffused with axioms constructed from a set of rules that are too complex to work with routinely, yet are simultaneously presented as uncontroversial legal facts. There are many areas where the resultant policy incoherence is particularly intense – for instance in the case of the patentability of software in the EU and the UK (11); and in the interpretation of Swiss use claims with severe implications of such claims on the exclusive use and control over important medications, as I discuss in detail in Textualization (3). The prevalence of textualisation means that we need an exogenous mechanism that is not similarly constrained by textual coherence, and that can open an interpretational space for consequential reasoning. This will strengthen the hand of courts trying to address the many instances of policy incoherence we see in the domestic interpretation of the law as well as in the way legal standards can be transplanted across jurisdictions.

Since the first mention of the three-step test for copyright in the 1886 Berne Convention, the legal significance of this test has grown far beyond the original international legal context. It has endured in newer digital contexts, and multiple international treaties. (12) The generality of the wording provides a good example of a legal instrument that can help cut through some of the policy incoherence in intellectual property law. Similarly the Doha declaration coalesced a remarkable degree of political support as a solution to the affordable access to patented medications for public health crises (13).

I believe a mechanism that combines the political capital of the Doha declaration with the doctrinal legitimacy of the three-step Berne test is now essential both within domestic interpretation of patent law, as well as to temper the legal transplantation of such rights through patent office cooperation. This antidote to policy incoherence will require similar support to be built up by exposing some of the current and acute challenges of access to patented medicines and other medical technologies – a task that is particularly suited to this UN High-Level Panel. Problems of access and exorbitant pricing that is a direct result of exclusive rights is not just confined to developing countries but can also be very problematic in developed countries whether or not there is a publicly funded health service.

The wording of such a test will have to be discussed further, subject to political interests and goodwill, but I would like to propose the following:

The grant and exploitation of patent rights
a) Must facilitate the enjoyment of the reasonable rights of the inventor
b) Should not pre-empt competition
c) Should not inhibit innovative improvements.

Rights demonstrably exercised in prejudice of any of the above may be curtailed or revoked.

Such a provision will open up a path for the consequential analysis of the impact of patent rights; and will function as an antidote to the textual pre-occupation in patent law. The key mischief that can be remedied with such a test will be the unreasonable exploitation of patent rights including exorbitant pricing, monopoly licensing of essential technologies and the prevention of reasonable research and development efforts. ‘Reasonableness’ as a test seen in (a) will allow for the consideration of international human rights principles including those raised by public health needs. The legal test will apply to the analysis of the extension of patent rights to new or unprecedented subject matter (for example, new uses of known substances) as well as apply to the post grant exploitation of such rights (such as in the case of exorbitant pricing of medicines). The fear that such a provision may begin to whittle away at the incentive effect of patents for inventors may be tempered by establishing guidelines for robust evidentiary standards, directed by an explanation of the rationale for such a provision by the UN High-Level Panel.

Bibliography and References

(1) Similar in function to Latour’s ‘technoscience’. See B Latour Science in Action (Harvard University Press (1997)

(2) For some examples see Commission on Intellectual Property Rights (2002) http://www.iprcommission.org/; Margaret Chon Slouching Towards Development in International Intellectual Property, (2007) Mich St Law Review 71; Patent Hold-Up and the Limits of Competition Law: A Trans-Atlantic Perspective’, U Petrovcic, (2013) 50 Common Market Law Review 1363

(3) S Thambisetty ‘Textualisation as Mode of Persuasion for Patent Law and What it Means for Legitimacy' LSE Law Working Paper Series, 07-2015 [Hereafter referred to as Textualization]; Also see S Thambisetty Legal transplants in patent law: why "utility" is the new "industrial applicability" (2009) Jurimetrics, 49 (2) 155-201

(4) For a discussion of recent US decisions and their impact on patent eligibility see Dan L Burk ‘Dolly and Alice’ (2015) Journal of Law and Biosciences, Available here <http://jlb.oxfordjournals.org/content/early/2015/10/16/jlb.lsv042.full?sid=1e0d27bb-5e06-40ab-b991-3a9b6d40671e>

(5) See ‘Declaration on Patent Protection: Regulatory Sovereignty under TRIPS Available here< https://www.mpg.de/8132986/Patent-Declaration.pdf>; Rochelle Dreyfuss, TRIPS and Essential Medicines: Must One Size Fit All? Making the WTO Responsive to the Global Health Crisis in Incentives for Global Public Health: Patent Law and Access to Essential Medicines 35 (Thomas Pogge, Matthew Rimmer and Kim Rubenstein, eds., Cambridge: Cambridge University Press, 2010) and S Thambisetty ‘Novartis v Union of India and the Person Skilled in the Art: A Missed Opportunity' Q.M.J.I.P. 2014, 4 (1) 79-94

(6) T. Piper, ‘A Common Law Prescription for a Medical Malaise’ in C. Ng, L. Bently and G. D’Agostino, The Common Law of Intellectual Property: Essays in Honour of David Vaver (Hart Publishing, 2010).

(7) TRIPS Agreement, Art 27, Patents Act 1977, s 4A(1), European Patent Convention, Art 53(c).

(8) G1/04 OJ 2006, 334

(9) See for instance T385/86 (BRUKER/Non-invasive measurement) and T964/99 (CYGNUS/Diagnostic method)

(10) Paragraph of the Examination Guidelines. Available here <http://www.epo.org/law-practice/legaltexts/html/guidelines/e/g_ii_4_2_1_3.htm>

(11) Justine Pila ‘Software Patents, Separation of Powers, And Failed Syllogisms: A Cornucopia from the Enlarged Board of Appeal Of The European Patent Office’ (2011) Cambridge Law Journal 70(1)

(12) Seen in the TRIPS Agreement, the WIPO Copyright Treaty, the WIPO Performances and Phonograms Treaty and the EU Copyright Directive. Also see Gervais, D. (2005), ‘Towards a New Core International Copyright Norm: The Reverse Three-Step Test’, Marquette Intellectual Property Law Review, vol. 9, no. 1, pp. 1- 37 (

13) See Frederik Abbott and Jerome Reichman ‘The Doha Round’s Public Health Legacy: Strategies for the Production and Diffusion of Patented Medicines under the Amended TRIPS Provisions’ Journal of International Economic Law 10(4), 921–987