Overview of Legal Aspects in the European Union
University of Karlsruhe, Germany
There is often much confusion about the workings and mechanisms of copyright protection. Copyright is a framework and not much more. Copyright can be used in order to favor access and it can also be used at the disadvantage of open access; it can have an enabling effect or a locked-in effect. However, copyright is often held responsible for blocking access.
COPYRIGHT AND OPEN ACCESS
Let us start with the idea of copyright and free access: are they friends or foes? We must consider some fundamentals of copyright. What is copyright? Copyright is an exclusive right. For nonlawyers one can imagine copyright along the same lines as a property right in tangibles. A property right is an absolute right over a good, which enables its holder to exclude others from using the particular good subject to the exclusive right and it gives its holder the right to use the good. Why do exclusive rights exist? Economists tell us that an immaterial, intangible good, such as data or information, has all the characteristics of public goods. Economists also tell us that without any incentive to take good care of public goods or create public goods there will be underinvestment in public goods. Hence, the legislature has decided to create an exclusive right, turning public goods into nonpublic goods.
What do exclusive rights enable right holders to do? Of course, an exclusive right allows its holder to exclude. However, in the field of copyright you generally do not want to exclude others arbitrarily, but rather you want to make sure that by allowing certain use and not others you are properly remunerated. The idea of getting money in return for the work you have created is an incentive to create. The author should be remunerated and the publisher, the producer who puts up the investment to disseminate the work, should recoup his investment. So, initially the idea of copyright is not to exclude others per se but only to exclude others in order to be able to grant an exclusive license to one particular producer. For example, authors may wish to grant the exclusive translation and publishing rights of their work to one particular producer or publisher and not to others, just to make sure that there is no illegal reprint of books, which would then transform the book as a product into a public good, which would deprive both authors and the producer of the original version of substantial income, which in turn would lead to an underinvestment into books.
The problem of exclusion of access may arise when intellectual property protection is granted to sole source products. A sole source product may be characterized as a product that has certain features that only come with this
one particular product. For example, we can all listen to music, which allows the user several choices. However, if you want to know the temperature in Paris yesterday between 9:30 a.m. and 9:45 a.m., that particular temperature might be stored in only one particular database. Possibly only one satellite recorded certain meteorological conditions a week ago over India and Bangladesh, which are found only in one particular source. If there is an exclusive right attached to such a sole source, you then have a locked-in effect.
The second problem that comes with the exclusive right is in the digital field. Producers are so afraid that their products will be copied, disseminated, and used without payment of adequate remuneration that they may feel tempted to block entire markets. Why is this a particular problem of the digital market? Well, digital goods can be repackaged, resold, and reformatted in many ways. As Jerry Reichman formulated: “Small incremental value may be added.” All these incremental values may be quite useful to the end user, but initial producers tend to keep the market free of would-be competitors who start in product niches where incremental value is added to the initial product and who might quickly grow out of that niche into the way of the initial producer. But if the initial producer keeps the market niches free, this deprives us for the time being from having these useful value-added services. Needless to point out, these control strategies produce certain locked-in effects.
An example is publishers of newspapers and the electronic press-clipping services. For the time being, in many countries the publishers themselves do not offer electronic press-clipping services. Everybody would agree that it is highly desirable to have electronic press-clipping services. First of all, we could save a lot of paper and we would not have to chop down so many forests. Second, some trucks would not drive on our highways and pollute the air. It is much more environmentally friendly to have electronic press-clipping services. Yet publishers are very reluctant to allow third parties to come in because these third parties offering electronic press-clipping services on the basis of the primary publisher’s product—namely the newspapers that contain these information services—apart from appropriating the market for digital exploitation of the initial product might even erode the basis for the primary product. Certainly we do not want our daily newspapers to disappear just because we gave too much freedom to third parties with electronic press-clipping services. Such services might be useful today, but if they then erode the basis of the primary product we do not have any value-added service anyway. But on the other hand, we want competition in markets of value-added services. To achieve this is a difficult task for copyright.
Third, let us briefly talk about free access and accessibility. The term “open” is very often confused not with mere accessibility but with “free” and this, moreover, sometimes in the sense of “cost-free” access. However, it is very clear that the production of information is cost intensive. Therefore, “free” in free access cannot mean free access as in free beer.
Fourth, even understood in the way just described, it should be noted that free access and accessibility are somehow linked to pricing. Users say, and the rights holders agree, that the price should be fair and reasonable. True, one has to pay a reasonable price in order to obtain access to certain information, but how high can the price be, or how low does the price have to be, so that we still can speak of open access or free access? It conveys dollar by dollar and if ultimately we pay $1,000 just to view one page, this very much looks like locking information away. Information would then no longer be free, no longer accessible. But what if the information is worth $1,000, or if it cost as much to produce the information? Who is to fix the price? In industrialized states, state authorities supervising prices were generally abolished after World War II, and the only legal instrument of at least indirectly controlling prices is antitrust law, which controls abuses of dominant market positions. But there is no such abuse as long as the right holder only makes use of his or her exclusive right. It follows that contrary to common opinion the distinction between free open access and information being locked away is a rather blurry one. Some energy should be devoted to clarifying these issues.
Another issue deals with access blocks by technological protection measures (TPM) and digital rights management (DRM) systems. It has often been said that the answer to the machine is in the machine. Initially TPMs were designed or understood as mere antipiracy devices, in essence, to ensure that works would not get copied. But there is more to it. Similarly, DRM has been understood as taking paper-negotiated contracts into the digital environment, but there too is more to it. The value added, so to speak, by TPMs and DRM is what economists call product diversification and price discrimination.
As an explanation, digital technology enables us to shape the use possibilities of one program in different ways. Let us take the example of music. If you receive a stream of musical data coming over the Internet, you are obtaining a listen-once-and-do-not-copy product. A CD, which is copy protected, allows you to listen as many times as you want but not to make copies. You could have another product further down the diversified product line that enables you to do more with that music, which would, for example, enable you to make one back up copy for use in your car. Ultimately, we still might have the “old-fashioned” CD as we know it today, which we could listen to and copy many times.
Again, economists tell us that with product diversification, producers can better capture the market. They can better answer to the market demand because there are some consumers who would be willing and able to pay more than they currently pay for the one-size-fits-all product, and there are others, who have not been able to pay the price for the current product, who could or might well be able to pay the (lower) price for a traditional CD. Product diversification by the user, and the ensuing price discrimination, raises the price for the high-use product and lowers the price for the low-use product. Economists tell us that this increases the overall public benefit. If TPMs are a good thing, we will have to make sure that consumers do not buy low-use products for a low price to circumvent the use-restricting TPM, and turn the product into a high-use product without paying. This, however, means that we need rather robust TPMs and strong legal protection against the circumvention of TPMs. This, of course, leads to the problem that effective technological protection measures can override the policy balancing that we find in our laws; again, we have a lock-in effect, this time unwanted by law.
Take the example of fair use. Fair use is permitted with regard to certain works, but if a work is protected by a TPM the system does not recognize whether the user wants to make commercial use of that particular work or just fair use as permitted by law. To this situation the law can respond in two ways. Either the law allows the user to circumvent the TPM in such a situation and thus risk that commercial use will be made without payment, or the legislature blocks the use completely, thus eliminating the freedom of fair use. In sum, introducing TPMs and DRM entails positive market effects, but it likewise results in rather strong lock-in effects.
THE LEGAL FRAMEWORK IN THE EUROPEAN UNION
Let us turn now to an overview of the legal framework in the European Union. First, a brief word on harmonization strategies in the field of copyright. The European Union has not created a community copyright similar to their community trademark, community design model, and community patent. Rather, an author obtains a bundle of national rights. These national rights are rather similar to each other so that there is no hindrance to the free movement of goods and services and no distortion of competition. It is the result of international conventions that makes this bundle seem as one unified right. An author creates a work in one country, and all the other countries that are parties to the international conventions promise to protect foreigners in the same way they protect their own nationals; they do so without asking for any formality requirements to be fulfilled. This makes copyright different from patents. For example, in order to get a British patent and a German patent you need two registrations. Of course, the process is simplified by the European Patent Office, where you can deposit one application, but the patentee still ends up with several national rights. With copyright you just create, and if you create in France you are treated in Germany, Great Britain, the Netherlands, and Belgium the same way as if a Belgian citizen had created a work in Belgium.
Within this bundle the copyrights are still distinct from each other. If there exists a limitation under U.K. law that declares certain use acts with regard to protected subject matter as fair dealing, then this only has its effects in Great Britain but not in Germany or France.
An interesting example is legal texts, which are free in Germany. There is some authority that even databases containing legal texts are free. In Germany I can download legislative material and make it available in my own database. However, the German author of legal texts could invoke his French copyright, since the acts I have performed in Germany, might also have—via the Internet—an effect in France and hence might violate the French copyright. Although the author of the material not protected in Germany cannot stop me from offering that
database for German Internet users, the author can still stop me for having that very same database made accessible to French users.
The purpose of copyright harmonization is to eliminate such discrepancies of national copyright laws. There is an E.U. directive on databases,1 which has tremendous repercussions for the scientific output. The E.U. database directive creates a copyright that protects the originality of a database, the way the data are collected and assembled, which is a right that can be violated only if the whole structure of a database is copied. On top of this the database directive also introduced a sui generis right that merely protects the investment made into a database and which makes it illegal to copy and reuse substantial parts of a database. It is even illegal to use insubstantial parts of a database provided the use is undertaken in a systematic way and interferes with the normal exploitation of the database in question.
Another point very often overlooked is the 1998 E.U. directive on access control.2 This started with a British case in which British tourists who wanted to watch encrypted BSkyB signals in Spain when they were on holiday purchased unauthorized decoders. BSkyB obtained legislation that made the production and sale of unauthorized decoders illegal. This ultimately led to the adoption of the European directive on access control, which protects any service that provides and allows access or is based on an access control, making it illegal to manufacture, import, or commercially advertise circumventing devices (it does not subject the act of circumventing itself to the prohibition).
In addition, the E.U. directive on copyright in the information society,3 as regards technical protection measures, also subjects the very act of circumventing to liability. Anyone who circumvents TPMs without authorization is subject to cease-and-desist orders, damages, and fines. That is potentially very far reaching for two reasons. The first is that in such an economy what do we do with the fair use? Do we allow it and eventually open up the doors for illegal commercial uses or do we close the doors to fair use on the assumption that it is better to block some fair use in order to make sure that no commercial use is being made if there is a TPM applied to the subject matter, rather then allowing fair use with regard to digital TPM-controlled subject matter? The European Commission has decided that in the online field, for databases that are contractually made available, TPMs override the balanced approach of limitations and exceptions of copyright. The European Commission has always argued that this is justified when you contractually agree to what you pay for. In practice the decisive question then is, who can pay for what? If prices charged are too high, then legal remedies are unlikely to break this situation. Of course, if the maker of the database abuses a dominant market position, antitrust violation might be applicable. But first, antitrust only comes in late (i.e., it only applies if the abuse has already taken place) and second, antitrust law is not really about price control. There has to be very strong evidence of an abuse of the dominant market position. The mere fact that the price charged for access is much higher than the actual production cost does not in itself make it an antitrust case. The only solution might then be to let the market decide. In rare cases such as the one in which Scientology tried to stop critics from making extensive quotes from its founder’s writings, the overriding effect of TPMs over statutorily granted exceptions, such as the citation right, might even result in a violation of the fundamental right of freedom of expression.
Another major problem with the European copyright information society directive is that it did not harmonize the limitations and exceptions. That is why I used that example with a database of legal text in Germany. While the database is perfectly free in Germany, it can be blocked in France by invoking French copyright. That is, of course, a big problem—we have in theory broad and far-reaching copyright harmonization, but it does not really harmonize copyright in Europe. The reason for this lack of harmonization is that member states do not like to give up
See Directive 96/9/E.C. of the European Parliament and of the Council of March 11, 1996, on the legal protection of databases at http://europa.eu.int/ISPO/infosoc/legreg/docs/969ec.html.
See Directive 98/84/E.C. of the European Parliament and of the Council of November 20, 1998, on the legal protection of services based on, or consisting of, conditional access, O.J.L. 320, November 28, 1998, at http://www.ebu.ch/departments/legal/pdf/leg_ref_ec_directive_conditional_access_201198.pdf.
See Directive 2001/29/E.C. of the European Parliament and of the Council of May 22, 2001, on the harmonization of certain aspects of copyright and related rights in the information society, O.J.L. 167/10, June 22, 2001, at http://www.patent.gov.uk/copy/notices/pdf/implement.pdf.
their national legal traditions. Moreover, any change in shaping the limitations invariably affects whole market segments that are so fiercely being fought over. The prospects for statutory compromise solutions in this respect look indeed rather dim.
First, copyright is not necessarily hostile to open access. For example, the open-source philosophy in software uses copyright in order to keep access open. The idea is to grant a nonexclusive license to anybody upon the condition that whoever uses the open code grants back any value that person may have added and upon condition that these users do not commercially exploit that particular software. In other words, users of open-source software under a general public license grant back to the open-source community a use right in any alterations and additions made, and they do not “proprietize” it by including open-source code in proprietary software. So open source is based on copyright and uses copyright in order to make sure that those who cheat on the conditions might be excluded.
Second, copyright can be used both to secure access and to block accessibility. This applies also to TPM and DRM. The real problems result from a too broad legal protection mainly in the field of sui generis database protection. There are also the economic concerns regarding sole sourcers, global players, and those who really want to block markets that they have either no intention to serve themselves or where they want to seek monopoly rents. The open question then is, when the price adequately reflects consumer demand, what is a monopoly rent?