Nel 2012 l'Argentina si è dotata di un nuovo set di Guidelines per l'esame delle domande di brevetto per invenzioni chimiche e farmaceutiche, il cui intento sembra quello di consentire il rilascio del brevetto solo in favore delle invenzioni consistenti in una nuova molecola, e non anche in favore di invenzioni ottenute a partire da molecole già esistenti. L'obiettivo principale di queste Guidelines è favorire l'accesso ai farmaci.

L'articolo mostra che, per quanto il problema dell'accesso al farmaco  meriti ogni sforzo, le Guidelines dell'Argentina rischiano di provocare una riduzione dei flussi di investimenti nella ricerca di invenzioni incrementali, e quindi rischiano di ridurre il flusso di nuovi farmaci. L'ultima parte dell'articolo esamina il problema della compatibilità delle Guidelines dell'Argentina con gli obblighi derivanti dall'Accordo TRIPS, sollevando seri dubbi su detta compatibilità.

Sommario/Summary:

1. The main goal of the Argentine Guidelines. - 2. A personal hypothesis of the practical effects of the rules proposed by the Argentine Guidelines. - 3. Markush patents and selection inventions. - 4. Abolishing selection inventions (in the way the Argentine Guidelines do) does not broaden the public domain space. - 5. Unreasonableness of an exclusive right of the Markush patent holder on all the subsequent selection inventions. - 6. Per se non patentability of selection inventions and the TRIPS Agreement. - 7. The patentability of prodrugs. - 8. The impact of the Argentine Guidelines on the flow of innovation: a credible hypothesis. - 9. The Argentine Guidelines in one country only. - 10. The real tools for an acceptable solution of the problem of Universal Health Coverage. - 11. The Argentine Guidelines and the TRIPS compliance. - 12. No meaning for the words “invention”, “novelty”, “inventive step”, and “industrial application” in the TRIPS Agreement? - 13. The generally accepted meaning of the words. - 14. Flexibility and certainty. - 15. Conclusions. - NOTE

1. The main goal of the Argentine Guidelines.

In May 2012, Argentina issued a detailed and comprehensive set of Guidelines for the Patentability Examination of Patent Applications for Chemical and Pharmaceutical Inventions[1]. This was an important move and, in principle, a welcome one. Patent offices have a difficult task in applying the complex rules of today's patent law to a high number of patent applications for inventions in different fields of technology, coming from different countries, and posing a wide array of different patentability issues. Good guidelines give patent offices and applicants a valid tool for reducing uncertainties and increasing the consistency of decisions, and for ensuring similar treatment of similar cases. To begin with a brief presentation of what will be discussed in depth in this article, the Argentine Guidelines seem to grant patentability only to chemical and pharmaceutical inventions resulting in the creation of new molecules. Inventions resulting from research activity on existing and known compounds are not patentable, or are presumed to be not patentable: as a rule, they are denied the qualification as invention or are considered lacking one or more conditions of patentability. This presumption of non-patentability is not absolute, but leaves very little room for patentability, and it is not easy to understand whether this room is real or only virtual; but, frankly speaking, it seems more virtual than real. The Argentine Guidelines openly identify their raison d'être, stating that access to drugs, the protection of public health[2], is put at risk by the «proliferation of patent applications for matters not strictly constituting an invention or that are marginal developments»[3]. Obviously, the issue is not the mere proliferation of patent applications, but the proliferation of granted patents. And, arguably, the number of patents is not a problemper se, but in its possible effects, such as in the delay they may cause to the market entry of generics. It is not my intention to put in doubt the importance of the problem of access to drugs (in the new formula: the problem of Universal Health Coverage), since I am absolutely convinced that it is one of the main issues of today's world. Neither do I question that patent law can be better applied or even substantially modified to contribute to this goal[4]. However, I simply do not think that the approach chosen by the Argentine Guidelines can deliver good results. It is necessary to coordinate [...]

2. A personal hypothesis of the practical effects of the rules proposed by the Argentine Guidelines.

An exam of the content of the Argentine Guidelines shows that these rulesper sedo not embrace the patentability of innovations made in relation to a large number of subject matters: polymorphs, pseudo-polymorphs, enantiomers, Markush formulas, selections, salts, esters and other derivatives of known substances, active metabolites, prodrugs, formulations and compositions, dosages, and analogy processes. The Argentine Guidelines move from the assumption that, in said cases, invention is absent, or, if invention is present, novelty and/or inventive step are absent. These inventions cover the full field of inventions known in the practice of chemical research, not considering the case of inventions whose subject matter is a new molecule. The evident implication is that - though this is not expressly said - the Argentine Guidelines tend to consider as patentable inventions in the chemistry field (almost) only new molecules. This is the unavoidable consequence of the fact that all other possible inventions (different from new molecules) are expressly said to be, as a principle, not patentable. Each of the subject matters listed in the Guidelines deserves a specific and detailed analysis, backed up by a clear understanding of its technical aspects. For reasons of space, it is not possible to do it in this article, so I will discuss briefly only the Markush-type patents and selection patents, and the patentability of prodrugs. Before going ahead, I want to present a personal hypothesis of the practical effects of the rules proposed by the Argentine Guidelines. From 2008 to 2012, 154 New Molecular Entities (NME) were launched globally[7]. Roughly half of these can be considered "breakthrough" innovations, in the sense that each of them has a novel mechanism of action[8]. The other half has mechanisms of action already known in their approved indication. Many of these "incremental innovation" cases, which are likely to be considered not patentable by the Argentine Guidelines, have played a key role in medical advance in recent years. Consider enantiomers. The compound escitalopram (brand name Lexapro) is the S-enantiomer of the earlier marketed product citalopram (brand name Cilexa/Cipramil), which is a racemic mixture of the S- and R-enantiomers. The literature[9] demonstrates many significant therapeutic benefits of the S-enantiomer over the racemic mix. In the same way, the product esomeprazole (brand name Nexium) is the S-enantiomer of the earlier marketed [...]

3. Markush patents and selection inventions.

First, I want to discuss the Markush-type patents and selection patents. These two topics are closely linked and can be examined together. A Markush patent claims a genus of compounds by reference to a shared chemical structure. It encompasses all molecules that share the common structure and which have constituent elements matching the choices allowed for each point of variation. The Argentine Guidelines admit the grant of patents on applications structured in this way, saying only that the application must demonstrate the general conditions of patentability (unity of invention; novelty, inventive step, and industrial application; sufficient description). In this case, a «generic chemical structure that may have multiple chemical substituents linked to a central nucleus, covers a variety of compounds with properties which, in spite of not having been proved for all the claimed compounds, may be inferred for the whole group»[12]. The document goes on to say that «the protection of 'Markush' formulas must be restricted to what may be effectively reproduced by a person skilled in the art departing from what has been disclosed in the specification, and the industrial application of which undoubtedly results from the description provided»[13]. But it also says that a patent structured in this way «discloses all the components of that group which, in this way, become part of the art»[14]. Selection patents are patents having claims that fall within the generic claim scope of earlier filed patent applications. The selection claims may be one or several sub-genus of the earlier broader genus or may be related to specific embodiments falling within that earlier genus claim, but, of course, not disclosed in that earlier application. The Argentine Guidelines deny,per seand absolutely, the patentability of selection inventions, saying that a Markush claim effectively discloses all molecules encompassed by that claim, thus anticipating any later filed claim to a particular compound, even if there is no description of that molecule provided in the previous Markush patent, and the disclosure in the patent is inadequate to show how to produce that particular compound without undue experimentation. Selection inventions are said not to be  "novel", on the basis of the prior art disclosure of a genus encompassing, but not explicitly disclosing, that selection (or species). The Argentine Guidelines offer a detailed description [...]

4. Abolishing selection inventions (in the way the Argentine Guidelines do) does not broaden the public domain space.

The explicit goal of the document is to reduce the number of patents on drugs and broaden the scope of the public domain, on the grounds that the abolishing of "unworthy" patents (as the selection patents are considered to be) will increase the availability of (not patented) drugs and reduce their prices. But in reality, by abolishing selection patents, the Argentine Guidelines may reduce the number of patents, but do not broaden the public domain, as selection patents inevitably fall within the scope ("patent space") of pre-existing patent rights. Quite simply, the Argentine Guidelines expand the scope of the Markush patents far beyond their current scope in the countries admitting this kind of application.  In the countries admitting s.c. Markush claims, it is generally said that the Markush patent claim does not cover all the compounds that fall within its generically described scope, and, more exactly, does not cover the compounds (although falling within that earlier genus claim) not specifically disclosed in the patent application and that a person of ordinary skill in the art is not enabled to produce without undue experimentation. Correspondingly, it is possible to grant a patent (a selection patent) on a compound falling within that generic claim scope, but which has not been effectively disclosed in the earlier application and cannot be identified without inventive activity. If the rule denies the patentability of the selection inventions, stating that all the compounds falling within the scope of the Markush patent are not new, because they are covered by that patent (and this is the assumption of the Argentine Guidelines), all the compounds selected after the filing of the Markush patent application could not be the object of new patent applications, but will not be in public domain. Simply, they will be covered by the previous Markush patent. In conclusion, from the point of view of the extent of the public domain, there is no substantial difference between the generally accepted idea and the Argentine Guidelines. The number of patents is different, but the space covered by them, on the whole, is the same. In both cases, all the compounds falling within the scope of a Markush patent are patent protected. In the generally accepted rule, the compounds are covered by the Markush patent and one or more selection patents; in the Argentine Guidelines, one patent (the Markush patent) covers all the compounds that can be selected from it, [...]

5. Unreasonableness of an exclusive right of the Markush patent holder on all the subsequent selection inventions.

At a closer look, the Argentine Guidelines result in only one minor difference to the position they oppose, which is their different patent date. When selection invention is permitted, any (non-obvious) selection invention will have a different patent application with a different filing date, and any selection patent granted will have its own expiration date, obviously subsequent to the expiration date of the Markush patent. Under the Argentine Guidelines, all the selected compounds will have the filing date of the Markush patent, and the expiration date will be identical for all. Indeed, a selected compound will not enjoy the full term of the patent (twenty years), but only twenty years minus the time elapsed from the filing date of the Markush claim to the date of the selection. And this shorter patent life for the selected compound might seem a benefit for the public - at least from the point of view of the Argentine Guidelines. But, from a different perspective (and, I would say, from a more balanced perspective), the traditional, generally accepted position seems more reasonable. I do not see any reason for giving the holder of the Markush patent the exclusive right on all the compounds that could derive from its general formula, even with inventive activity, with undue experimentation, with investment in financial, technical, and human resources going far beyond what could be done by an "ordinary" technician or an "ordinary" research group. It seems to me much more reasonable that the exclusive right on a compound selected by third parties with inventive activity be given not to the holder of the previous Markush patent, but to the researcher that has effectively identified it. This is consistent both with the idea of the patent as a reward for an invention already made, and with the idea of the patent as an incentive for research.

6. Per se non patentability of selection inventions and the TRIPS Agreement.

It is worth making one more remark on the Argentine Guidelines regarding Markush claims and selection patents. In my opinion, the arguments of the Argentine Guidelines seem to be inconsistent with the generally accepted meaning of novelty - and, consequently, with the obligations created by the TRIPS Agreement. The problem of the consistency of the Guidelines with the TRIPS Agreement will be discussed later, but it is possible to make some preliminary remarks in this paragraph. In order for a prior art disclosure to destroy the novelty of a molecule, that disclosure must not only show the structure of the molecule, but also enable a person skilled in the art to make that molecule without undue experimentation. If it does not, then the prior art cannot be said to anticipate (i.e., destroy the novelty of) a claim limited to the specific molecule. The Guidelines impose this rule as an unconditional standard - even if it were shown that an individual molecule was neither described by the previous Markush patent nor shown to be producible using the disclosures in the Markush patent or available in the state of the art. In addition, the Argentine Guidelines consider the selection of a species or a group of species that is part of a larger group, as disclosed in prior art, even if not via a Markush-type patent application. They assume that disclosure of a genus always destroys the novelty of every species encompassed by that genus. According to this principle, even composition claims that include an element disclosed in the prior art are not novel. The Guidelines in effect redefine the concept of novelty and assume that a composition comprising several different molecules is the same as any single molecule found in that composition. Under practices followed before (and after) the adoption of the TRIPS Agreement, disclosure of a genus does not per se negate novelty of each species within that genus. For example, under U.S. law, disclosure of a generic chemical formula will anticipate a species covered by that generic formula only if the species can be "at once envisaged" from the generic formula[19]. Other countries follow a similar practice whereby disclosure of a genus does not necessarily destroy the novelty of every species falling within the genus. Similarly, the PCT Examination Guidelines hold that disclosure of a genus does not invariably destroy the novelty of every species within that genus[20]. In my opinion (and I think this is generally shared [...]

7. The patentability of prodrugs.

We can now move to a different aspect of the Argentine Guidelines regarding prodrugs. This term is used to identify compounds that, when administered to the body, can produce a therapeutically active ingredient, which is the product of the compound's metabolism in the body. The Argentine Guidelines recognize that molecules that can be classified as "prodrugs" can be novel, can involve an inventive step, and can be capable of industrial application. Consequently, they recognize that molecules that are classified as "prodrugs" can be patented if they satisfy the general conditions of patentability. However, the Guidelines impose an additional requirement for a prodrug to be patented. They specifically provide that «the application must contain evidence that the prodrug is inactive or less active than the originated compound, that the generation of the active compound (in the body) assures an effective level thereof, and also that it minimizes the direct metabolism of the prodrug»[21]. The Guidelines also distinguish these prodrug requirements from conventional patentability requirements. The text is clear in so far as an application for a prodrug, in addition tosatisfying the requirements for novelty, inventive step, industrial application and adequate disclosure, must provide evidence regarding the effects or properties of the prodrug in the body after administration. If the prodrug does not exhibit those properties, it will not be granted a patent. This additional requirement seems to be inconsistent with the TRIPS Agreement, which (and, again, this can be said even without the more precise analysis that will be conducted later on the consistency of the Argentine Guidelines with the obligations from TRIPS Agreement) simply does not permit WTO members to impose substantive patentability requirements in addition to those specified in the Agreement (novelty, inventive step, and industrial application). Similarly, the TRIPS Agreement does not allow member states to impose special disclosure requirements for an invention beyond those specified in Article 29.1. WTO members may only require that the patent applicant «disclose the invention in a manner sufficiently clear and complete for the invention to be carried out by a person skilled in the art and may require the applicant to indicate the best mode for carrying out the invention known to the inventor at the filing date or, where priority is claimed, at the priority date of the [...]

8. The impact of the Argentine Guidelines on the flow of innovation: a credible hypothesis.

As stated at the start of this article, what is really important is the impact that a set of guidelines for the examination of patent applications can have on inventive activity. The main goal of the patent system is to foster innovation through the flow of new inventions. The mission of any guidelines for patent examination should be consistent with the goal of the patent system. One question could be: If guidelines identical to the Argentine Guidelines had been in force in all countries in the last ten years, would we have had no change, an increase, or decrease in the flow of new drugs? The same question can be posed for the future. Would guidelines like the Argentine Guidelines give rise to a bigger flow of new drugs, or reduce the creation of new drugs? Would they encourage or discourage investment in research for new drugs? As to the past, in my view, most of the new drugs consisting in breakthrough NMEs[22] that have entered the market would likely still be in our pharmacies and drugstores, even if the Argentine Guidelines had been in effect everywhere in the world. Investment in research and development for breakthrough NMEs would still be fostered, as the patenting of these molecules is still admitted by the Argentine Guidelines. By the same token, there is no reason to think we would have, today, a larger number of breakthrough NMEs. The Argentine Guidelines do not encourage investment in the research for new molecules in comparison with the examination rules now in effect in developed countries. They do not interfere with the research activity for new molecules; they simply maintain the status quo. Nevertheless, it is possible to assume that in the context of the Argentine Guidelines the flow of breakthrough NMEs could decrease, given that pharmaceutical companies use a substantial part of the profits they make on "incremental innovation" drugs to pay for R&D of "breakthroughs". What about the flow of new drugs that are thought to be patentable, and have been patented in the last ten years, but could not have been patented if the Argentine Guidelines had been in effect? Here the answer is nuanced. Some of these "incremental innovation" drugs would have been invented anyway, because their invention was the natural conclusion of lines of research that were already underway. But at least some of these drugs would have been invented only as a result of investment that would have been undertaken only with the prospect of recouping the cost [...]

9. The Argentine Guidelines in one country only.

It is possible now to consider a more subtle issue that is also more sensitive in the perspective of international relations. What would the effect be of guidelines identical to the Argentine Guidelines entering into force in one country only? In principle, the Guidelines would lead, in that country, to a reduction of investment in research into new drugs and in the creation of new drugs. Knowing the pertinent data, it would be possible to quantify these effects. This impact could be considered tolerable if investment in research were already fairly low, and if the flow of new drugs were also low. From a radical point of view, reducing what is already low is not particularly compromising. Moreover, the country in question could count on a counter-effect that could be considered positive. Adopting these Guidelines would deny patents to new "incremental innovation" NMEs invented elsewhere, considering as patentable only new "breakthrough" NMEs drugs, both foreign and domestic. This would allow that country to produce freely all new "incremental innovation" NMEs invented around the world, without payment of royalties to the (foreign) patent holders. Obviously, these drugs could be produced in Argentine only for internal consumption, and not for export (being these drugs patented in other countries). The Argentine pharmaceutical sector, in the absence of a prospect of patenting the possible results, would be discouraged from conducting research on known compounds; the national industry will have no stimulus for investing in incremental innovation, and (if not strong enough to compete on the floor of research of new molecules, which is, perhaps, the case of Argentine), will give ground in research for incremental innovation to its competitors from different countries, whose investments in this kind of research will be fostered by patents in their countries of origin.  Moreover, for the reasons I will explain in the last part of this article, if the country in question is bound by the TRIPS Agreement, it is highly probable that this conduct would be considered in violation of its international obligations under that Agreement.

10. The real tools for an acceptable solution of the problem of Universal Health Coverage.

The explicit goal of the Argentine Guidelines is not to reduce the creation of new drugs (this could be, as I have said, their unintentional effect), or to increase the creation of new drugs (although this should be the main goal of any patent system). The explicit goal of the Argentine Guidelines is the reduction of the price of existing drugs on the Argentine market. The drug price issue (more generally, the issue of access to drugs) is, under any perspective, dramatic and multifaceted. It deserves effort, careful study, and strong action. But in principle, it is not directly connected with patent law, and it is not within patent law that the main tools (or the first steps) to resolve the problem can be found. Beyond the creation of a good system of drugs pricing, countries must develop a national  assistance for destitute citizens. Any country can decide whether to shoulder the cost of drugs for all citizens, or only for the indigent, and whether to pay for all medicines or only for essential medicines. But today no one questions that states should help the destitute to access at least the most essential medicines. This in turn needs a "good" state: a balanced fiscal structure, a reduction of economic inequalities, the development of a strong feeling of national solidarity, and much more. Notwithstanding the difficulties, this is the road states have to take to reach the goal of Universal Health Coverage and, more generally, the goal of a more developed health and welfare system and a society based on principles of justice and equality. This is not the place to discuss such moves and interventions, which are beyond the competence of the author and are not specifically linked to the issue considered by this article. However, one final question has still to be answered. Many countries today simply lack the elementary financial conditions for developing an acceptable system of public health. They cannot afford to buy drugs for their citizens, not even for the poorest. In this case, access to drugs should be assured by the development of a more integrated net of international solidarity, whose main actors should be the States, international organizations, as well as foundations and non-governmental organizations, which already play an important role. As a final remark, I do not think it is reasonable to try to impose mandatory contributions to private actors and pharmaceuticals manufacturers by neutralizing patent rights[23] because the effects [...]

11. The Argentine Guidelines and the TRIPS compliance.

The simple fact that it is possible to imagine that the effects of the Argentine Guidelines could be negative - because they could hinder the creation of new drugs by discouraging investments in research activities on known compounds, thus reducing technical progress and quality of life of future generations - suggests verifying whether the Argentine Guidelines are consistent with the existing international rules on patents. Hence, patent law, both at the national and international level, has the goal of fostering technical progress. If we assume that national patent rules have the opposite effect, it is easy to infer that these rules are not consistent with international patent law - for the simple reason that also international patent law has the goal of fostering innovation. The patent rules to be taken into account now are the rules proposed by the TRIPS Agreement. Although no WTO Member has yet submitted a complaint against the Argentine Guidelines, the results of the analysis set forth in the first and second part of this article suggest continuing with this inquiry.  The explicit premise of the Argentine Guidelines is that «the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS) - Annex 1C - determines the patentability requirements and grants Member Countries the authority to determine the standards of novelty and inventive step to be attained by the inventions in order to constitute patentable matter pursuant to their respective national law»[24]. In principle, this statement can be supported. However, after reading the entire text of the Guidelines, it can be inferred that their implicit premise is significantly more generous in interpreting the alleged prerogatives of member states. The real premise of the Argentine Guidelines is that the TRIPS Agreement does not bind countries to a precise definition of what constitutes a patentable invention, and does not propose a binding set of conditions of patentability, which are only indicated with words (novelty, inventive step and industrial application) absolutely lacking in content. The Argentine Guidelines regard the internal flexibility of the TRIPS Agreement as giving countries space for a national definition of "invention", "novelty", "inventive step", and "industrial application", divorced from the appropriate context. I disagree with this point of view. The interpretation of the TRIPS Agreement is not a free activity of WTO member states, but must [...]

12. No meaning for the words “invention”, “novelty”, “inventive step”, and “industrial application” in the TRIPS Agreement?

The TRIPS Agreement does not expressly define the terms "invention", "novelty", "inventive step", or "industrial application". As far as the term "invention" is concerned, this is by no means strange, as it is not generally defined in national patent laws or in international conventions. By contrast, it can seem strange at first sight that the Agreement does not define the terms "novelty", "inventive step", and "industrial application", which are defined in all modern patent law texts. In this study, I will not consider the meaning of the term "invention" in the TRIPS Agreement, as this is not the focus of this analysis of the Argentine Guidelines. I will only analyze the meaning that the TRIPS Agreement gives to the terms "novelty", "inventive step", and "industrial application". To understand what these words mean, WTO member states are obliged to use "the ordinary meaning" given to these terms "in the context" of the TRIPS Agreement and "in the light of its object and purpose". Strictly adhering to the provisions of the same Agreement, they cannot consider the meaning of these terms in other international treaties (in the sense of Article 31.2.a of the Vienna Convention), because there are no previous treaties "made between all the parties" of the TRIPS Agreement. This also applies to the Patent Cooperation Treaty (PCT), which gives a definition of the above-mentioned terms[28], but has not been signed by all the members of the TRIPS Agreement and especially not by Argentina[29]. Nevertheless, it would be strange to conclude that the TRIPS Agreement does not give a meaning to the conditions of patentability. The simple fact that it mentions the terms "novelty", "inventive step", and "industrial application" implies that these terms are held to have a commonly accepted meaning. If we assume that the TRIPS Agreement ignores the meaning of those terms and gives countries full freedom to define them as they prefer, their use becomes absurd and incomprehensible. Void of meaning, those words would be vacuous ghosts. Why would the TRIPS Agreement mention the words "invention", "novelty", "inventive step", and "industrial application" if these terms had no meaning at all? The meaning of the above-mentioned terms is to be looked for, first and foremost, in the "ordinary meaning" given to them by the global patent community, including in Argentina. Do those terms have a commonly accepted meaning? The conditions of patentability ("novelty", "inventive step", [...]

13. The generally accepted meaning of the words.

It is thus not true that countries are free to give their own meaning to the terms indicating the conditions of patentability in the TRIPS Agreement. Those terms are to be read in the generally accepted meaning they have in the patent world, which coincides with the definitions of the Patent Cooperation Treaty - even in countries like Argentina, which are not part of the PCT[34]. This does not imply that the meaning of those terms is rigidly defined once and for all; in addition, countries do have latitude (I would say: "some" latitude[35]) in defining the conditions of patentability. Indeed, the words expressing those conditions are themselves words that need interpreting. Their meaning, like the meaning of any word, can be described in a way frequently used by theories of the interpretation of law. Each word has a zone of light, a nucleus of meaning that can be considered clear for all (or most of) the people using that word and that language. This is what enables a word to act as an instrument for people to communicate with one another. Each word also has a zone of shadow, a further complex of meanings less generally and clearly shared. Using the well known words written by H. L. A. Hart, «a man with a shining smooth pate is clearly bald; another with a luxuriant mop clearly is not; but the question whether a third man, with a fringe of hair here and there, is bald might be indefinitely disputed, if it were thought worthwhile or any practical issue turned on it»[36].  To give one more example, the words "day" and "night" have a clear zone of light. Everyone agrees that it is day when the sun is up and night when the sun is down. Both words also have a zone of shadow: dawn and dusk, are they day or night? Is there an exact moment when day converts into night and vice versa? People can give different answers to these questions. The same is true for all other words, which differ only in the amplitude of the two zones, light and shadow. A word that is totally lacking a clear nucleus of generally accepted meaning has no reason to exist as a word; it is not a communication tool and should be expelled from language. By contrast, it is practically impossible, except for highly formalized languages (such as, in part, the mathematical language) to have a word lacking any (more or less ample) zone of shadow. Legal terms share the same properties as common language. Although they are subject to a more or less pronounced process of [...]

14. Flexibility and certainty.

The conclusion of this line of reflection is that the terms "novelty", "inventive step", and "industrial application" all have a meaning, including in the TRIPS Agreement. Their meaning presents a zone of light and a zone of shadow. The zone of light can be identified by looking at the text of national laws, at the jurisprudence of national legal systems and any relevant international panels, and at the practice of national and supranational patent offices. Divergent texts of law, divergent opinions of jurisprudence, doctrine, and practice in different countries and even within one country, indicate the zone of shadow. Countries have a "margin of appreciation[37]", have "flexibility", have "some" latitude[38], in the sense that they are obliged to accept the zone of light of terms, but are not obliged to accept their zone of shadow. However, this does not correspond to saying that states are free to give these terms the meaning they prefer. «Flexibility would nevertheless not appear to erode the legal certainty which jus scriptum is intended to provide[39]». To sum up, in the interpretation of the TRIPS Agreement, member states are not free to give the terms "novelty", "inventive step", and "industrial application" the meaning they want. They are positively bound by the generally accepted meaning of these terms, i.e. the "dominant pattern"[40] of these three conditions of patentability. The above remarks imply that, perhaps, countries can deny the existence of novelty and/or inventive step and/or industrial application in cases that have given rise to conflicting or different decisions in different countries. The novelty of a biological material (pre-existing in nature), the inventive step of a software implemented invention, the patentability of a second medical indication, are all widely debated problems, which do not have, as of now, a generally accepted solution. In these cases, countries have greater policy latitude, because the generally accepted meaning of the terms describing the conditions for patentability does not have a precise answer to those problems. However, countries cannota prioriandper sedeny the existence of novelty and/or inventive step in, for example, the invention of a new compound, since the patentability of this invention is generally accepted worldwide. By the same token, and coming back to one of the problems discussed in this article, countries cannota prioriandper sedeny the patentability of compounds that [...]

15. Conclusions.

My conclusions are simple. Access to existing drugs is and must continue to be one of the main goals of humankind, but thinking that patent law can give a great contribution to it is misleading. The mission of patent law is (mainly, if not only) to increase the possibility of having more drugs in the future. Trying to use patent law to increase the diffusion of existing drugs (as Argentine Guidelines seem to do) is, in my view, using the wrong tool. A substantial part of the most important new drugs of today is the fruit of research on existing compounds. Considering inventions resulting from research activity on existing compounds non patentableper semay seriously reduce the investments in research activity on existing compounds. The consequence could be (or would be) a reduction in the creation of new drugs. The above could be the effects of rules structured like the Argentine Guidelines. Hence, it is difficult to think that these rules comply with the TRIPS Agreement. Indeed, while this Agreement aims to foster innovation, these rules risk going the opposite way. Moreover, in light of the law of interpretation of treaties, the consistency of the Argentine Guidelines with the TRIPS Agreement is doubtful for the technical reasons explored above. Ultimately, the problem considered in this article seems to be linked, unpredictably, with the problem of the relations between generations. The patent is a cost that we bear for the sake of innovation. Do we want to pay less for existing drugs, and have fewer drugs for ourselves and our children in the future? Or do we want to pay a little more for existing drugs, with the hope of having more drugs tomorrow? This is the problem at stake. I am strongly in favor of the second choice. This does not mean to ignore the problem of access to drugs. For better access to existing drugs, we must turn to different, more appropriate and useful tools: good systems of drugs pricing, assistance for destitute citizens, a balanced fiscal structure, a reduction of economic inequalities, the development of a strong feeling of national and international solidarity. It takes a long way, but, as usual, a shortcut would be a dangerous mistake.

NOTE