The first-ever international IP treaty that touched on the copyright aspects of exhaustion was the TRIPS Agreement of 1994.  [33] Although the “Chairman’s Text” of 1991 still recommended for the introduction of both a general right of distribution and the principle of exhaustion related thereto,  [34] these plans failed after a lengthy preparatory work.  [35] The final text of the TRIPS Agreement referred to the principle from a neutral aspect when it stressed that “[f]or the purposes of dispute settlement under this Agreement, subject to the provisions of Articles 3 and 4 nothing in this Agreement shall be used to address the issue of the exhaustion of intellectual property rights.”  [36] Under these rules, all that the signatories have to keep in mind is that they shall apply the principle of national treatment and the most-favoured-nation treatment when designating the reach of exhaustion; furthermore, that any dispute related to a domestic regulation shall not be the subject to a dispute settlement procedure under the WTO law. The latter does not forbid, however, the initiation of legal proceedings in front of domestic courts.  [37]

The limited solution of the TRIPS Agreement is partially due to the fact that the agreement was finally concluded without any substantive provision on a distinct right of distribution.  [38] On the other hand, by 1994, only a few countries had a settled regulation and case law on exhaustion/first sale doctrine. These countries set the frames of the principle quite differently. Countries like the Netherlands  [39] or Switzerland  [40] codified the doctrine of international exhaustion. Likewise, several developed countries, for example Australia or New Zealand,  [41] and developing nations were interested in a broad reach of the principle.  [42] Others, like the United States  [43] or Germany  [44] , advocated for domestic/national exhaustion. The European Economic Community accepted two directives before 1994 that envisaged a Community-wide, regional exhaustion.  [45] The contracting parties opined differently on the principle, and the final text turned out to be the best compromise for the signatories. Therefore the TRIPS Agreement did not introduce any substantive obligation and provided absolute freedom to the contracting parties to decide whether they are willing to introduce a principle of exhaustion, and if yes, whether it should have a domestic, regional or international reach.  [46]

The two Internet treaties of the World Intellectual Property Organization (WIPO) from 1996 already included positive norms on exhaustion.  [47] This is partially due to the fact that, for the first time in international copyright law history, these treaties granted for a general right of distribution.  [48] Article 6(2) of the WIPO Copyright Treaty (WCT) – and Article 8(2) of the WIPO Performances and Phonograms Treaty (WPPT) with a really similar wording  [49] – noted that “[n]othing in this Treaty shall affect the freedom of Contracting Parties to determine the conditions, if any, under which the exhaustion of the right [of distribution] applies after the first sale or other transfer of ownership of the original or a copy of the work with the authorization of the author”. Under the above terms, members of the WIPO Treaties shall comply with some substantive provisions of the WCT/WPPT.

First, WCT and WPPT granted great flexibility to signatories to regulate the content of the principle by their domestic law. The treaties allowed a freedom to national legislators to choose from introducing domestic, regional or international reach of the doctrine, if any.  [50]

Second, it prescribed the sale/transfer of ownership and the lawful (authorized) nature of the sale of the work as elements of the definition of the principle.  [51]

Third, the expression “nothing in this Treaty shall affect” highlighted that no other provision of the treaty, including WCT Article 10 and WPPT Article 16 on the three-step test, forbids signatories to regulate on exhaustion.  [52] This technically means that the freedom of lawful acquirers to dispose of the property of the work is absolutely in accordance with the law; and that the resale does not per se conflict with the normal exploitation of the work and does not prejudice the legitimate interests of the respective author.  [53]

Finally, the Agreed Statement attached to Articles 6 and 7 made it clear that “[a]s used in these Articles, the expressions ’copies’ and ’original and copies,’ being subject to the right of distribution and the right of rental under the said Articles, refer exclusively to fixed copies that can be put into circulation as tangible objects.”

It seems that in light of the preamble of both the WCT  [54] and the WPPT  [55] – that is, to set the frames of an effective and working copyright law in the age of internet – the most important element of the definition is the one contained by the Agreed Statement; that is to say that exhaustion does only apply to tangible objects of works. As we will see it later, this statement is the main obstacle of the development of digital exhaustion. As Professor Sterling noted it: “[t]hus, as far as the treaty text is concerned, the debate as to whether the distribution right should cover dissemination of works by electronic means is resolved; the intention in the Treaty is to cover only ‘hard copies’ such as books, discs and tapes”.  [56] Others view it differently. Ruffler stresses that “what [the Agreed Statement] actually says is that the copies can be put into circulation as tangible objects. That is, that it must be possible to fix them in a tangible medium, and not that they must already be fixed as tangible objects. What is obviously meant is that it should not be a question of quick, one-off acts as is the case with on-demand online services but one of works which are capable of being permanently transferred.”  [57]

Related to the above, we need to touch upon another issue when talking about the distribution of works and objects of neighbouring rights protection. As soon as the use of internet became a daily routine, the question naturally emerged, whether copyright law should react on that phenomenon as well. The answer was a clear yes. The method of granting protection to the right holders regarding online uses was, however, much more difficult. Domestic copyright statutes showed significant differences regarding the rights granted to copyright holders. The two possible rights that were affected were the right of distribution (promoted greatly by the U.S.) and the right of communication to the public (supported mainly by European countries). None of these rights fit perfectly to the said Internet uses. Distribution was originally meant to cover the transfer of physical copies. Communication to the public generally covered simultaneous transmissions from one place to multiple places either by wire or wireless means (for example broadcasting, cable services, satellite transmission). Further, the international framework for both of these rights was incomplete, to say the best.

Mihály Ficsor, than Assistant Director General of the WIPO, explained “[b]etween the two opposing alternatives – that is, between basing the international regulation on interpretation and, as a maximum, on slight modification, of existing rights, on the one hand, and introducing a new on-demand transmission/delivery right, on the other – a third compromise alternative would also be possible (and, actually, would seem, for the time being, to have the best chance to get a sufficient general acceptance). This could consist of an umbrella provision under which it would be an obligation to grant an exclusive right or exclusive rights to authorize either any use of works and other protected productions – with some possible reasonable exceptions – through on-demand transmission/delivery, or to authorize certain acts – described in such a neutral way (that is, without involving any specific legal characterization) as possible – carried out in the case of such transmission/delivery, but the legal characterization of the right or rights granted (whether it results in the combination – and possible extension – of existing rights, and/or in a new specific right) would be left to national legislation”.  [58] This third compromise option was finally accepted by the members of the WIPO, and finally regulated the new economic right for both authors and neighbouring right holders.  [59] Member States of the WIPO have, since that time, implemented the umbrella solution in various ways. As it will be introduced below, the European Union and the United States of America have followed a different path in this perspective.

To sum up: although the right of making available to the public was formulated in a technology-neutral way, it was clearly designed to cover all on-demand transmissions of data via the Internet. The right of making available to the public only applies, therefore, where the copyrighted subject matter is made accessible to every member of the public (having Internet connection), where members of the public may access these works from a place and at a time individually chosen by them. Consequently, predetermined programs, including webcasting, pay-per-use, pay-tv or pay-radio services are excluded from the scope of this distinct right and shall be covered by other rights, especially communication to the public or broadcasting.  [60] Similarly, where tangible copies of contents (like CDs) are sold via the internet, the right of distribution applies.

The question that the present paper addresses is a special, third option: what if intangible copies of works are sold – and, from a technological perspective, are transferred – to the users via the internet?

The fact that TRIPS and WCT/WPPT were the first international treaties touching upon the doctrine of exhaustion does not mean that they were the first international norms at all that regulated the issue. The European Court of Justice (ECJ) and the legislature of the European Economic Community have taken serious steps to harmonize the rules of a Community-wide/regional exhaustion doctrine in the field of copyright law since the 1970’s.  [61] Schovsbo called the harmonization by the ECJ as “1.-phase” development of exhaustion or negative harmonization, and the creation of directives by the competent bodies of the EEC (and later the EU) as “2.-phase” development or positive harmonization.  [62]

The first-ever decision on the exhaustion of distribution rights was handed over in the famous Deutsche Grammophon case. Here, the ECJ based its decision on different objectives of the EEC Treaty: the prohibition of partitioning of the market, free movement of goods, as well as the prohibition of distortions of competition in the common market.  [63] The ECJ highlighted that prohibitions and restrictions on trade might be applied by Member States, also in cases of copyright law, if they do not constitute a means of arbitrary discrimination or a disguised restriction on trade between Member States.  [64] Based upon these, the ECJ concluded that “[i]f a right related to copyright is relied upon to prevent the marketing in a Member State of products distributed by the holder of the right or with his consent on the territory of another Member State on the sole ground that such distribution did not take place on the national territory, such a prohibition, which would legitimize the isolation of national markets, would be repugnant to the essential purpose of the Treaty, which is to unite national markets into a single market. That purpose could not be attained if, under the various legal systems of the Member States, nationals of those States were able to partition the market and bring about arbitrary discrimination or disguised restrictions on trade between Member States. Consequently, it would be in conflict with the provisions prescribing the free movement of products within the common market for a manufacturer of sound recordings to exercise the exclusive right to distribute the protected articles, conferred upon him by the legislation of a Member State, in such a way as to prohibit the sale in that State of products placed on the market by him or with his consent in another Member State solely because such distribution did not occur within the territory of the first Member State.”  [65]

The precedence of the basic principles and purposes of EEC Treaty – especially the free movement of goods – over the protection of copyright holders was later reaffirmed by the ECJ in the Membran case. The judges highlighted that “[i]t follows from the foregoing considerations that neither the copyright owner or his licensee, nor a copyright management society acting in the owner's or licensee's name, may rely on the exclusive exploitation right conferred by copyright to prevent or restrict the importation of sound recordings which have been lawfully marketed in another Member State by the owner himself or with his consent. (…) It should be observed next that no provision of national legislation may permit an undertaking which is responsible for the management of copyrights and has a monopoly on the territory of a Member State by virtue of that management to charge a levy on products imported from another Member State where they were put into circulation by or with the consent of the copyright owner and thereby cause the Common Market to be partitioned. Such a practice would amount to allowing a private undertaking to impose a charge on the importation of sound recordings which are already in free circulation in the Common Market on account of their crossing a frontier; it would therefore have the effect of entrenching the isolation of national markets which the Treaty seeks to abolish.”  [66]

The Luxembourg court later developed a second prong of Community-wide exhaustion under which it does not apply to services. In the Coditel I case the ECJ confirmed that it is not a mean of arbitrary discrimination or a disguised restriction to grant a territorial exclusivity to “show” cinematographic works, and consequently to prohibit the broadcasting of the said works by an unlicenced cable television company who received the signals from a different Member State.  [67] The ECJ noted that “the right of a copyright owner and his assigns to require fees for any showing of a film is part of the essential function of copyright in this type of literary and artistic work”.  [68] To put it differently: cinematographic works might be repeated (commercially exploited) without any limitations via performances/broadcasting.  [69]

Similarly, the EEC-compatibility of the rental right granted to copyright holders by the Danish Copyright Act was confirmed by the ECJ in the Warner Brothers v. Christiansen case. As the judges noted: “[t]he existence of [rental] market was made possible by various factors such as the improvement of manufacturing methods for video-cassettes which increased their strength and life in use, the growing awareness amongst viewers that they watch only occasionally the video-cassettes which they bought and, lastly, their relatively high purchase price. The market for the hiring-out of video-cassettes reaches a wider public than the market for their sale and, at present, offers great potential as a source of revenue for makers of films. However, it is apparent that, by authorizing the collection of royalties only on sales to private individuals and to persons hiring out video-cassettes, it is impossible to guarantee to makers of films a remuneration which reflects the number of occasions on which the video-cassettes are actually hired out and which secures for them a satisfactory share of the rental market. That explains why, as the Commission points out in its observations, certain national laws have recently provided specific protection of the right to hire out video-cassettes.”  [70]

To sum up: the ECJ has set a basic definition of the exhaustion doctrine, under which the copyright holder cannot object to the resale of the physical data carrier put into circulation by the copyright holders or their assignees in any of the Member States. This is reasoned by the basic principles of the EEC, especially the free movement of goods. The copyright holder retains, however, an exclusive right to use and allow copyrighted subject matter to be used via services, including broadcasting and rental.  [71]

Only a few weeks after the ECJ published its Warner Brothers v. Christiansen decision did the European Commission also revealed its Green Paper on Copyright and the Challenge of Technology. Here the Commission concluded that “[t]he Court has held that in all these fields reliance on an exclusive right to exclude goods lawfully marketed in other Member States would be incompatible with the fundamental principles of the Community Treaty providing for the free circulation of goods since it would legitimize the isolation of national markets. While Article 36 EEC authorizes the Member States to maintain restrictions on imports justified on the grounds of the protection of industrial and commercial property, it does not permit a right holder to prevent the free circulation of goods once, with the right holder’s consent, they have been placed on the market within the Community.”  [72] The fact that the ECJ has developed quite a detailed precedent law on this territory was enough for the Commission to conclude that “national copyright laws will not have adverse or divergent effects on the functioning of the common market”.  [73] The introduction of a right of distribution and consequently a principle of exhaustion of the distribution right was therefore declared to be unnecessary in 1988.  [74]

At the same time, the European Commission fairly realized that several urgent issues need to be addressed by community law. Unlike public lending that the Commission found unnecessary to regulate yet,  [75] provisions on the rental of sound recordings, audio-visual contents and computer programs turned out to be timely. Before the appearance of digital technologies the commercial rental seemed to be an unattractive business due to the massive deterioration of data carriers, and consequently the worsening of enjoyment of contents.  [76] As soon as CDs were standardized by Sony in 1983 and digital audio tapes – that for the first time allowed for the digital reproduction of contents at home – were put into circulation by Sony in 1987, the quality of the copies of works turned out to be stable enough, and “rental outlets have mushroomed”.  [77] The same was true for the audio-visual sector, since the public had greater interest in renting video contents, rather than purchasing them. The popularity of rental was partially due to the expensive nature of the video tapes and the related equipment as well.  [78] Sadly, the EU Green Paper missed to include any well-founded argumentation in respect of the rental of computer programs within the European Economic Community. The document only urged for the acceptance of broad use rights for the benefit of software producers.  [79]

It is therefore not a surprise that commercial actors of the music, video and software industry expressed their interest in introducing a new rental right on an EEC level. The European Commission accepted their opinion. It noted that “[p]resent trends in the distribution and marketing of sound and video recordings suggest that commercial rental will constitute an increasingly important means by which such recordings will be made available to the public. Furthermore, given the links between rental and the problems of piracy and private copying, this development implies significant economic consequences for those whose works and performances are recorded. In the absence of a firm legal basis for right holders to authorize the commercial exploitation of their works through rental, it seems likely that those responsible for creating recorded works will receive a much lower return for their efforts and investment than would otherwise be the case, while middlemen could profit disproportionately from the efforts of other.”  [80]

The European Commission recommended an action to introduce a new rental right for the benefit of the right holders of the music, video and software industry. The EU Green Paper also highlighted that the introduction of this right would not be contrary to the case law of the ECJ, since the latter developed the principle of exhaustion only in respect of the sale of tangible copies of works, and excluded the operation of the principle in respect of “service-style” rights, like performance.  [81]

The Software Directive was finally enacted in 1991.  [82] It introduced a new rental right for the producers of computer programs. In light of the case law of the ECJ introduced above it seemed to be wise enough to regulate the principle of exhaustion in respect of the rental right. Therefore, the Software Directive provided that “[t]he first sale in the Community of a copy of a program by the right holder or with his consent shall exhaust the distribution right within the Community of that copy, with the exception of the right to control further rental of the program or a copy thereof.”  [83]

The above wording of the principle has some specific features that differentiate it from the norms set by the TRIPS Agreement and the WIPO Internet Treaties. First, it is a special law on computer programs, not any other subject matter. Second, there is clearly no restriction on the form of the copy. That is, the resale of intangible copies is not excluded per se from the doctrine.  [84] Third, the Software Directive envisaged a Community-wide exhaustion, under which the sale of the computer program within any of the member states – that time only 12, nowadays 28 – exhausts the right of distribution in every other member states as well. Finally, the directive – following the ECJ’s case law – differentiated between “sale-style” and “service-style” forms of distribution, where the latter (more precisely rental of computer programs) was excluded from the scope of the doctrine. Oddly, the recitals of the directive include no further guidance on this issue.

A year later, the Council accepted the next directive, which has been referred to as Rental Directive.  [85] The short (or nick) name of the directive is quite misleading, since it harmonized the rights granted to related rights holders as well. As a part of the letter form of protection, Article 9(1) introduced a right of distribution for the benefit of performers, with regards to fixations of their performances; for phonogram producers, in respect of their phonograms; for producers of the first fixations of films, in respect of the original and copies of their films; and for broadcasting organizations, with regards to fixations of their broadcast. Article 9(2) followed the idea of Community-wide exhaustion.  [86] Further, the Rental Directive excluded the right of rental and lending – as services – from the theory of exhaustion.  [87] Similar to the Software Directive, the Rental Directive lacks further guidance on exhaustion in the recitals.

Shortly after the acceptance of this directive, a German corporation that was involved in the rental of sound recordings questioned the legitimacy of the newly erected rental right, especially its exclusion from the scope of the doctrine of exhaustion. The CJEU refused to accept these claims and stressed that “the release into circulation of a sound recording cannot therefore, by definition, render lawful other forms of exploitation of the protected work, such as rental, which are of a different nature from sale or any other lawful form of distribution. (…) Thus, the distinction drawn in the Directive between the effects of the specific rental and lending right (…) and those of the distribution right (…) is justified. The former is not exhausted by the sale or any other act of distribution of the object, whereas the latter may be exhausted, but only and specifically upon the first sale in the Community by the right holder or with his consent. The introduction by the Community legislation of an exclusive rental right cannot therefore constitute any breach of the principle of exhaustion of the distribution right, the purpose and scope of which are different.”  [88]

The Court continued its ruling as follows: “the inclusion, challenged by the defendant in the main proceedings, of phonogram producers among the beneficiaries of the exclusive rental right appears justified by the protection of the extremely high and risky investments which are required for the production of phonograms and are essential if authors are to go on creating new works. As the Advocate General explained in point 26 of his Opinion, the grant of an exclusive right to producers certainly constitutes the most effective form of protection, having regard in particular to the development of new technologies and the increasing threat of piracy, which is favoured by the extreme ease with which recordings can be copied. In the absence of such a right, it is likely that the remuneration of those who invest in the creation of those products would cease to be properly guaranteed, with inevitable repercussions for the creation of new works.”  [89]

In 1996, a further directive was passed by the European Parliament and the Council on the protection of database. Here, the directive also granted for a right of distribution for the benefit of authors of databases. This right was limited by the doctrine of exhaustion. According to the Database Directive, “the first sale in the Community of a copy of the database by the right holder or with his consent shall exhaust the right to control resale of that copy within the Community.”  [90] Such basic rule is limited in two major ways. First, Recital 43 of the Database Directive rules that “in the case of on-line transmission, the right to prohibit re-utilization is not exhausted either as regards the database or as regards a material copy of the database or of part thereof made by the addressee of the transmission with the consent of the right holder”. Second, the rental and lending of database is governed by the Rental Directive, and consequently there is no exhaustion related to these rights in respect of databases.  [91]

The Information Society (InfoSoc) – or as many call it, the Copyright – Directive harmonized a general right of distribution (granted to authors as well), as well as a general exhaustion doctrine covering all protectable subject matter in 2001. The InfoSoc Directive aimed the implementation of the respective provisions of the WIPO Internet Treaties. Under Article 4(2) of the latter directive, “[t]he distribution right shall not be exhausted within the Community in respect of the original or copies of the work, except where the first sale or other transfer of ownership in the Community of that object is made by the right holder or with his consent.” Unlike in the case of the Software and Rental Directive, the normative text of the InfoSoc Directive was supplemented by two recitals on the relevant subject matter. Recital 28 stressed that the right of distribution only affects works incorporated in a tangible article or object; further, it reaffirmed the Community-wide exhaustion.  [92] Recital 29 excluded services as a default and those material copies produced via any on-line service from the scope of exhaustion.  [93] Such an exclusion is similarly mirrored by Article 3(3) of the InfoSoc Directive.  [94]

Three important notes need to be added here. First, the doctrine of exhaustion has been completely regulated at the level of directives. It might be important to recall, therefore, a paragraph of an ECJ decision, according to which “recourse to Article 36 is nonetheless no longer possible where Community directive provide for harmonization of the measures necessary to achieve the specific objective which would be furthered by reliance on this provision.”  [95]

Second, the EU legislation has settled some remarkable limitations to the doctrine of exhaustion. The principle might only restrict the resale of tangible copies, as envisaged by Recital 28 of the InfoSoc Directive that incorporated Agreed Statement attached to Article 6 and 7 of the WCT. As a consequence, uses of protected subject matter that are not connected to any tangible copy, or those that are offered as a service, shall not exhaust the right holders’ exclusive rights.  [96] Such rights are communication to the public, including broadcasting, retransmission and making available to the public, rental and lending.  [97]

The other limitation is connected to the territorial nature of copyright law under which domestic provisions might only apply within the physical boundaries of the given country. This practically means that a use abroad authorized by the right holder in the given country does not exhaust the right to control any domestic distribution of the copies of the work. The ban on parallel importation is therefore a general tool to prevent the flow of – sometimes cheaper – copies produced or sold abroad to another markets. This ban was, however, partially overruled by the EEC and later the EU, when it introduced the concept of Community-wide or regional exhaustion. The territory of the member states has been envisaged as a single market, and consequently the resale of goods might be only blocked if originally sold outside of the economic community.  [98] This means that if the copy of the work is put into circulation in any of the 28 EU Member States or the three further European Economic Area members (Lichtenstein, Iceland or Norway) that copy shall freely circulate within the EEA.  [99] Corbet called the freedom of users to resell lawfully purchased copies within the EEA as “Eurodefense”.  [100] At the same time, the introduction of the doctrine of Community-wide or regional exhaustion effectively led to the acceptance that Member States of the European Union and the EEA are excluded from providing a broader scope of exhaustion, that is, an international one.  [101] As the CJEU stressed it in its ruling in the Laserdisken case: “[i]t follows from the clear wording of Article 4(2) of Directive 2001/29, in conjunction with the twenty-eighth recital in the preamble to that directive, that that provision does not leave it open to the Member States to provide for a rule of exhaustion other than the Community-wide exhaustion rule. (…) This, moreover, is the only interpretation which is fully consistent with the purpose of Directive 2001/29 which, according to the first recital in the preamble thereto, is to ensure the functioning of the internal market. A situation in which some Member States will be able to provide for international exhaustion of distribution rights whilst others will provide only for Community-wide exhaustion of those rights will inevitably give rise to barriers to the free movement of goods and the freedom to provide services.”  [102]

Finally, although it is not common to discuss it as a limit of the principle of exhaustion, the droit de suite  [103] similarly puts barriers to the uncontrolled flow of works within the European Union.  [104] Under the relevant EU directive, “[t]he resale right is a right of a productive character which enables the author/artist to receive consideration for successive transfers of the work. The subject-matter of the resale right is the physical work, namely the medium in which the protected work is incorporated”.  [105] Contrary to the scope of the Berne Union Convention, the EU directive grants the resale right only for visual artists. Literary authors and composers are excluded from the protection.  [106] The droit de suite is an inalienable and unwaivable right that provides for the receipt of a royalty “based on the sale price obtained for any resale of the work, subsequent to the first transfer of the work by the author”.  [107] The royalty shall be payable if the resale involves an art market professional, either as a seller, purchaser or intermediary.  [108] The droit de suite is not an unlimited right. According to the Article 1(3) of the directive “Member States may provide that the right referred to in paragraph 1 shall not apply to acts of resale where the seller acquired the work directly from the author less than three years before that resale and where the resale price does not exceed EUR 10000”. The subject matter of the directive is similarly limited to “works of graphic or plastic art such as pictures, collages, paintings, drawings, engravings, prints, lithographs, sculptures, tapestries, ceramics, glassware and photographs”.  [109] Member States are likewise obliged to set a threshold – minimum maximum 3000€ – under which the resale shall be royalty free.  [110] Still, as it is clearly visible from the above, the resale royalty right excludes exactly what the principle of exhaustion envisages: the free disposal of the property of a tangible copy.

Thirdly, due to the fact that the InfoSoc Directive has not only implemented a separate right of distribution but also added a distinct right of making available to the public (introduced as a subcategory of the communication to the public right), the EU legislation has expressed its clear willingness to deal with on-demand uses outside of the scope of the right of distribution. The question remains whether such a solution shall be deemed as a proper, working solution under the current technological realities.

Although the Bobbs-Merrill v. Straus case  [111] has undeniably become the leading historical precedent of the first sale doctrine, there were a few earlier federal court decisions dealing with the issue.  [112] For the purposes of the present article, however, there is no need to introduce other than the above mentioned Straus case. There, the Bobbs-Merrill Company registered the copyright on Hallie Erminie Rives’ book titled “The Castaway” by the U.S. Copyright Office on May 18, 1904. The book was published immediately thereafter. On the first inside page of the book the publisher printed an expressed warning that the copies of the work shall not be distributed for a retail price of less than one dollar. Naturally, the Bobbs-Merrill Company sold the books on wholesale for a reduced price. The R. H. Macy & Company owned by Isidor and Nathan Straus purchased from these books by the gross for less than a half dollar per item. Contrary to the warning described above, the R. H. Macy & Company resold the copies for a retail price of $0.89.  [113] The Bobbs-Merrill Company deemed this practice an infringement of its copyright and therefore sued the Straus brothers.

As the Supreme Court of the United States highlighted, the main question of the case was, whether “[the right to vend] intended to create a right which would permit the holder of the copyright to fasten by notice in a book or upon one of the articles mentioned within the statute, a restriction upon the subsequent alienation of the subject-matter of copyright after the owner had parted with the title to one who had acquired full dominion over it and had given a satisfactory price for it?”  [114] The justices answered in the negative as follows: “In our view the copyright statutes, while protecting the owner of the copyright in his right to multiply and sell his production, do not create the right to impose, by notice, such as is disclosed in this case, a limitation at which the book shall be sold at retail by future purchasers, with whom there is no privity of contract. (…) To add to the right of exclusive sale the authority to control all future retail sales, by a notice that such sales must be made at a fixed sum, would give a right not included in the terms of the statute, and, in our view, extend its operation, by construction, beyond its meaning, when interpreted with a view to ascertaining the legislative intent in its enactment.”  [115]

The Supreme Court first noted that in light of the aims of the Congress the right to vend (currently: right to distribute) – as regulated by the Copyright Act of 1831  [116] – did not allow the right holder to control the resale of the copy of the work by its lawful acquirer via a written stipulation, supposed that the right holder passed on the ownership of the said copy in exchange of a fair remuneration.  [117] Second, the stipulation that prohibited the resale of the copy contrary to the conditions set by the right holder only affected those distributors who contracted the right holder. That is, all those who accepted the conditions as binding upon himself committed a breach of contract by the resale of the books (for retail price of less than a dollar).  [118]

Congressmen who were involved in the drafting of the new Copyright Act at that very moment recognized the validity of the Supreme Court’s logic  [119] and enacted the doctrine in the Copyright Act of 1909. According to the statute “[n]othing in this title shall be deemed to forbid, prevent, or restrict the transfer of any copy of a copyrighted work, the possession of which has been lawfully obtained.”  [120] The creators of the current Copyright Act (enacted in 1976) transferred the previous statutory provision with some minor – but not insignificant – changes. USCA §109(a) regulates that “[n]otwithstanding the provisions of section 106(3), the owner of a particular copy or phonorecord lawfully made under this title, or any person authorized by such owner, is entitled, without the authority of the copyright owner, to sell or otherwise dispose of the possession of that copy or phonorecord.”

Shortly after the entry into force of the USCA, right holders faced unexpected challenges. New business models appeared – originally on the Japanese market – for the rental of sound recordings, computer programs and video tapes. The number of music rental stores grew to 1600 by 1983 in Japan,  [121] whilst Rent-A-Record – that opened its first store in Providence, Rhode Island, in 1981 – ran 250 stores in the United States by 1983.  [122] If we combine these numbers with the fact that the CD appeared on the market in 1983, we can easily understand why the record industry initiated an overwhelming attack against music rental. Lobbyists envisioned the death of the music industry, if Congress had missed the opportunity to take steps in this situation.  [123] The Senate reported surprisingly fast that the rental of sound recordings infringes on the right of distribution. The report concluded that “the purpose and result of record rentals is to enable and encourage customers to tape their rented albums at home. (…) This, a record rental and a blank tape purchase is now an alternative way of obtaining a record without having to buy one. The rental is a direct replacement of a sale.”  [124] As a consequence, Congress enacted USCA §109(b)(1). The original wording of the text said that “[n]otwithstanding the provisions of subsection (a) [the first sale doctrine], unless authorized by the owners of copyright in the sound recording and in the musical works embodied therein, the owner of a particular phonorecord may not, for purposes of direct or indirect commercial advantage dispose of, or authorize the disposal of, the possession of that phonorecord by rental, lease, or lending, or by any other act or practice in the nature of rental, lease, or lending.”  [125]

The social receptivity of rental similarly endangered the interests of the American software industry. Computer programs, especially computer games, became available for an approximately $1-20 daily rental fee. Some estimated that the software industry lost around 1.3 billion USD due to unauthorized rental of computer programs between 1981 and 1984.  [126] A bill of statute was submitted to the Congress in 1986, however, the law was enacted only in 1990. The USCA excluded the rental, leasing and lending of computer programs from the scope of the first sale doctrine.  [127]

Rental of video tapes has become a daily routine, as well, from the beginning of the 1980’s. There were times when the ratio of rental and sale has reached almost 9 to 1.  [128] A bill of statute was submitted to Congress in 1983. The proposal recommended the exclusion of video tape rentals from the scope of the first sale doctrine.  [129] There are multiple reasons why this bill failed. First, the Betamax decision was handed over by the U.S. Supreme Court in 1984. Although that case was not focusing on the first sale doctrine, however, the justices favoured private users and Sony by declaring “time-shifting” of broadcasts by Betamax video tape recorders as fair use.  [130] Some argue that this was a reason why members of the Congress left audio-visual contents outside of the scope of the Record Rental Amendment Act in 1984.  [131] Second, during the “golden age” of video rental the price of the video tape recorder and the video tapes was extremely high.  [132] It was therefore quite difficult and expensive to build a “library of videotapes”. Thirdly and the foremost, Hollywood studios succeeded in concluding contractual agreements with the franchise owners of the rental stores to rent only pre-recorded video tapes purchased legally and directly from the studios.  [133] In several cases studios acquired bigger profit than box office sales under these agreements.  [134]

The United States of America signed the North American Free Trade Agreement (NAFTA) with Canada and Mexico in 1992. One of the objectives of NAFTA is to “eliminate barriers to trade in, and facilitate the cross-border movement of, goods and services between territories of the Parties”.  [135] The Agreement regulates intellectual property rights as well.  [136] According to Article 1705-1706 contracting parties are obliged to guarantee the exercise of right holders’ economic rights.  [137] Looking carefully at these provisions we might discover that NAFTA prohibits “the importation into the Party's territory of copies of the work made without the right holder's authorization”. This sentence means that contracting states introduced a “limited regional exhaustion” doctrine. First, right holders are not allowed to control the importation of the lawfully made copies. Second, however, any further use of these copies within the importing country (including sale, rental and other forms of distribution) might be overseen by the right holder. The effectiveness of the above regulation is questionable, since there is no sanction that might be used in case the provisions are infringed.  [138] It is similarly squealing that there is no court decision on the above referred Articles of NAFTA.

It is not only NAFTA that speaks about parallel importation under American law. The USCA declares the importation of copies of works acquired abroad as an infringement of the distribution right.  [139] The statute, however, doesn’t include anything about the interplay of the first sale doctrine and the prohibition of parallel imports. It is left to the courts to decide about the legality of those copies that were lawfully produced, put into circulation and acquired abroad or those copies that were lawfully produced in the U.S., later exported and then acquired abroad, finally imported back to the U.S. again.

U.S. federal district and circuit courts have developed quite a complicated and sometimes inconsistent practice on the above question,  [140] which are topped by three separate rulings from the U.S. Supreme Court. The present article is going to summarize only the latter trio.

The first decision of the U.S. Supreme Court on parallel importation was handed over in the Quality King v. L’anza case. Here, several tons of shampoo were produced by L’anza in the United States and exported to a British distributor who in turn sold them to a Maltese company. Quality King purchased these products from the latter company for approximately 35-40% cheaper than the U.S. retail price. Quality King later imported the shampoo back to the United States.  [141] Both the district court  [142] and the Ninth Circuit  [143] favoured L’anza, however, the Supreme Court overturned those decisions. The justices stressed that the works – here, quite paradoxically, the copyrighted text printed on the cover of the shampoo boxes  [144] – produced within the United States and exported abroad were covered by USCA §109(a).  [145] Further, USCA §602(a) is subordinated to §106(3) and §109(a).  [146] The Supreme Court also noted that the general ban on parallel importation remains important, since “the first sale doctrine would not provide a defense to a §602(a) action against any non-owner such as a bailee, a licencee, a consignee, or one whose possession of the copy was unlawful.”  [147] As, however, Justice Ginsburg noted in her concurring opinion, “[t]his case involves a ’round trip’ journey, travel of copies in question from the United States to places abroad, than back again. I join the Court’s opinion recognizing that we do not today resolve cases in which the allegedly infringing imports were manufactured abroad”.  [148]

The next major issue was evolving around the importation of luxurious Omega watches that were decorated by trademarked and at the same time copyrighted designs (the “Omega Globe Design”). The U.S.-based wholesale company, Costco, offered for sale 117 Seamaster watches 35% cheaper than Omega’s original retail price. The said watches were originally produced in the Switzerland and sold in Paraguay and Egypt.  [149] Although the district court rejected plaintiff’s claims,  [150] the Ninth Circuit reversed that decision, concluding that – contrary to the round trip journey of L’anza’s shampoo – the Omega watches were not lawfully made under this title, that is, they were not lawfully produced on U.S. soil.  [151] The Ninth Circuit relied both on the dicta of the Quality King decision quoted above and Justice Ginsburg’s concurring opinion, when it confirmed the applicability of §602(a).  [152] The circuit court’s ruling was affirmed by an equally divided (4-4) Supreme Court,  [153] that was handed over per curiam.  [154] It is worth mentioning that after remanding the case to the trial court, Costco started to rely on an absolutely new argument in its defence. The company claimed that Omega misused its copyrights  [155] when it relied on the copyrighted design to prohibit the importation of the Seamaster watches to the United States. The district court accepted this argument, and concluded that “[h]ere, Omega concedes that a purpose of the copyrighted Omega Globe Design was to control the importation and sale of its watches containing the design, as the watches could not be copyrighted. Accordingly, Omega misused its copyright of the Omega Globe Design by leveraging its limited monopoly in being able to control the importation of that design to control the importation of its Seamaster watches.”  [156]

The Supreme Court continued to interpret the issue of parallel imports in the John Wiley & Sons v. Kirtsaeng case. Here, defendant Supap Kirtsaeng, national of Thailand, systematically imported textbooks to the United States printed by the Asian branch of John Wiley & Sons. Although the Asian versions were identical in their content with those printed in the United States, however, they were produced on worse paper, they were only published in paperback and they lacked multimedia supplements. Kirtsaeng profited over $37.000 during the years of his business. The publishing house finally sued Kirtsaeng in 2008 on copyright, trademark and unfair competition claims.  [157]

The district court favoured the plaintiff,  [158] and the majority of the Second Circuit panel affirmed the trial court decision.  [159] Circuit Judge Garvan Murtha expressed his criticism towards the majority’s decision in his dissenting opinion. He refused to accept that USCA §109(a) has a geographical connotation,  [160] and that any contrary decision would lead to the depreciation of the ban on parallel imports.  [161] As Murtha concluded “[s]uch a result would provide greater copyright protection to copies manufactured abroad than those manufactured domestically. (…) I do not believe Congress intended to provide an incentive for U.S. copyright holders to manufacture copies of their work abroad”.  [162]

The importance of Murtha’s dissenting opinion is evidenced by the decision of the Supreme Court that allowed for the application of the first sale doctrine to the importation of works lawfully produced abroad, irrespectively of the place where the works were first put into circulation. Justices noted that the expression “lawfully made under this title” does not have any geographical meaning. It refers to the prerequisite of authorization by the right holder and production under the rules of the USCA.  [163] The Supreme Court similarly accepted Murtha’s opinion on the equal treatment of right holders,  [164] and similarly refused that their interpretation would depreciate USCA §602(a).  [165] The majority of the Supreme Court also stressed that any geographical interpretation would inevitably lead to the increase of the costs of running the socially important services of public libraries, second hand book stores, IT companies or museums..  [166] Finally, the majority also took the view that “the Constitution's language nowhere suggests that its limited exclusive right should include a right to divide markets or a concomitant right to charge different purchasers different prices for the same book, say to increase or to maximize gain. (…) To the contrary, Congress enacted a copyright law that (through the ‘first sale’ doctrine) limits copyright holders' ability to divide domestic markets. And that limitation is consistent with antitrust laws that ordinarily forbid market divisions.”  [167]

The consequence of the Kirtsaeng decision  [168] might be that any work produced and put into circulation under the authorization of the right holder (or its subordinates) and in accordance with the copyright law of the relevant country shall be subject to importation, and there is no legal remedy against such activity. To sum up: the doors are open for a doctrine of international exhaustion in the United States. The consequences of such a decision are still unknown. Besides several positive effects (as emphasized by the justices above), it is almost sure that significant negative consequences will be visible soon, especially related to the pricing of books printed cheaper outside of the boundaries of the U.S., and related to the licensing practices of the U.S. companies.

Another crucial question needing to yet be discussed is that the USCA does not include any provision on the making available to the public right,  [169] although both the WCT and the WPPT entered into force in the United States of America in May 2002. The lack of such right is not per se against the Internet Treaties, since, as it was introduced above, this right was envisaged as an “umbrella solution” for the on-demand uses. Historically, the U.S. has showed a great interest in addressing the digital transmission of copyrightable subject matter as distribution.  [170] Since, however, neither the WCT nor the WPPT is a self-executing treaty in the U.S., the lack of clear wording on the issue in the USCA makes in difficult to know under what conditions U.S. judges might provide protection to right holders in respect of the online use of their copyrighted works. There are several federal courts that did not accept the concept of making available to the public at all,  [171] and there are other decisions with the opposite result.  [172] The uncertainty related to this issue was noticed by the Register of Copyrights, Maria A. Pallante, as well. Ms. Pallante stressed the importance to clarify the boundaries of distribution right, although she did not explicitly refer to the codification of right of making available to the public.  [173] The current academic literature supports this idea. David Nimmer recently amended his treaties on U.S. copyright law, where he noted that the making available to the public right shall be covered by the right of distribution.  [174] Professor Menell reached the same conclusion based upon the historical interpretation of the USCA.  [175]

Digital market places – especially the one introduced by ReDigi – were born into the previously introduced status quo of the first sale doctrine. As it will be visible below, the initial answers given to the challenges posed by the online resale of digital contents are strictly limited under the traditional interpretation of the USCA.

The first major court proceeding that shed light on the issue of digital exhaustion was the UsedSoft case.  [177] Here, Oracle, a renowned software producer sued a German corporation, UsedSoft, for reselling “used software licences”. Oracle develops and markets several valuable computer programs, including the famous Java. The affected database computer programs were offered for download from the internet in 85% of the cases. Oracle signs end user licence agreements (EULA) with the purchasers of its programs. In the current case, the respective section of the EULA provided as follows: “With the payment for services you receive, exclusively for your internal business purposes, for an unlimited period a non-exclusive non-transferable user right free of charge for everything that Oracle develops and makes available to you on the basis of this agreement.”  [178] Oracle similarly offered so-called volume licences to its computer programs, under which 25 end-users had the right to use the same licence. UsedSoft acquired volume licences, where the original licencee did not install the computer program in the available number offered by Oracle.  [179] That is, the business model of UsedSoft was partially based on the split of volume licences. UsedSoft at the same time did not transfer any copies of the computer program. Rather, it directed its clients to Oracle’s website to download the respective program from that web page. UsedSoft launched an Oracle Special Offer in October 2005, where it offered up-to-date (that is, the maintenance agreement was still in force) software licences for resale. The company testified to the validity of the original purchase of the licence key by a notarial certificate. Oracle initiated court proceedings to stop the above Special Offer.  [180]

Both the trial court  [181] and the appellate court  [182] favoured Oracle’s arguments. Consequently, the case reached the German Federal Court of Justice (Bundesgerichtshof, BGH). Although the BGH seemed to accept the lower courts’ decisions, it turned to the CJEU with three questions. In its first question, the BGH asked for an opinion from the CJEU, whether any second or later acquirer of computer program who did not sign an EULA with Oracle shall be deemed as a “lawful acquirer”, where the latter relies on the theory of exhaustion of distribution right when purchasing the software licence key. The BGH also needed guidance whether the distribution right of software producer exhausts at all where the computer program was offered for download (that is in an intangible form) rather than on any tangible medium (CD-ROM/DVD).  [183] Finally, the BGH raised a third question that reflected the specific fact patterns of the case, and asked for guidance whether “a person who has acquired a used licence may, for making a copy of the program (…) rely on exhaustion of the right of distribution of the copy of the program made by the first acquirer, with the consent of the right holder, by downloading it from the internet, if the first acquirer has deleted his copy or no longer uses it.”  [184]

The CJEU answered the three questions in a different order. Focusing first on the second question, the judges provided a bright-line rule on the exhaustion of software distribution rights. The Grand Chamber recalled that under Article 4(2) of the Software Directive the right of distribution exhausts if a copy of the computer program is sold within the EEA by the right holders (or under their authorization).  [185] It was consequently important to decide, whether the conclusion of an EULA and the download of the computer program from Oracle’s website leads to a first sale of the program or not.  [186] Since the term “sale” used by the Software Directive does not refer to Member States’ law, it was interpreted in an independent and uniform way.  [187] The CJEU concluded that “[a]ccording to a commonly accepted definition, a ‘sale’ is an agreement by which a person, in return for payment, transfers to another person his rights of ownership in an item of tangible or intangible property belonging to him.”  [188]

Oracle argued that it did not sell its computer programs; indeed, it signed an EULA with users, that is, only permitted the use of the computer programs and the ownership on the said works was not transferred to the users.  [189] The CJEU did not accept this position. It noted that the downloading of the computer program and the conclusion of the EULA form an indivisible whole. The two steps only function with the other. The CJEU similarly stressed that there is no difference, whether the computer program is offered for download or by means of a material medium, since downloading a copy of the computer program (the source code) from the data carrier or from the internet to the user’s computer and concluding a licence agreement remain inseparable from the point of view of the acquirer. Furthermore, Oracle’s EULA allowed for the permanent use of the software in exchange for the payment of a fee that was directly designed to correspond to the economic value of the said computer program.  [190] The CJEU concluded that Oracle’s practice led to a sale, rather than a licence. This way, the principle of exhaustion could not be evaded simply through calling the contract as a licence.  [191]

The CJEU refused the claim of Oracle and of the Commission that offering computer program for download on a website shall be deemed as making available to the public.  [192] The importance of this question is huge, since the right of making available to the public is not subject to the principle of exhaustion, unlike the right of distribution. The judges concluded that since Article 1(2) of the InfoSoc Directive notes that the directive does not affect in any way the rules of the Software Directive, the latter shall be deemed as lex specialis, and consequently the sale of the computer program leads to the exhaustion of the distribution right.  [193]

The CJEU also spent an enormous amount of time on discussing whether the principle of exhaustion applies to intangible copies of computer programs as well, or – as Oracle, the Commission and the Irish, Spanish, French and Italian governments indicated – only to tangible copies. The CJEU first noted that the Software Directive is referring to the first sale of the computer program without specifying the form of the copy sold. Therefore the principle of exhaustion shall cover the sale of both tangible and intangible copies of a computer program, including works that were downloaded from the web.  [194] Here, the CJEU returned to its previous argument that the Software Directive acts as lex specialis, and therefore the interpretation of Article 4(2) shall be independent from the said international and EU norms.  [195] The CJEU supported its point of view with two policy arguments. First, it claimed that “[t]he on-line transmission method is the functional equivalent of the supply of a material medium”.  [196] Second, the limitation of the principle of exhaustion to the copies sold on a tangible medium “would allow the copyright holder to control the resale of copies downloaded from the internet and to demand further remuneration on the occasion of each new sale, even though the first sale of the copy had already enabled the right holder to obtain an appropriate remuneration.”  [197]

Oracle also claimed that since the updates offered by the company transformed the original computer program into new copies the licencees are not allowed to resell the “current” computer programs, since those are not identical with the original ones. The CJEU refuted this argument, when noted that “the conclusion of a maintenance agreement, such as those at issue in the main proceedings, on the occasion of the sale of an intangible copy of a computer program has the effect that the copy originally purchased is patched and updated. Even if the maintenance agreement is for a limited period, the functionalities corrected, altered or added on the basis of such an agreement form an integral part of the copy originally downloaded and can be used by the acquirer of the copy for an unlimited period, even in the event that the acquirer subsequently decides not to renew the maintenance agreement.”  [198]

The CJEU accepted Oracle’s arguments on the legal classification of the partial resale of volume licences. The judges noted that volume licences are sold as a block by Oracle, and in the case that the original purchaser is willing to get rid of parts of that, he shall deactivate the remaining copies of the computer program.  [199] Naturally, this would be impractical, and that is exactly against the original purchaser’s will.

Finally, the CJEU provided a joint answer to the first and third questions, when it concluded that the second (and any later) purchaser of the licence key shall be deemed as a lawful acquirer, who has the right to refer to the principle of exhaustion as a limitation of the right holder’s exclusive right of distribution. On the other hand, the reseller of the computer program is obliged to make unusable the copy originally installed on its own computer, and the right holder is allowed to ensure the deactivation by all technical means.  [200]

The ruling of the CJEU was confirmed by the German Federal Court of Justice.  [201] The BGH refused two novel arguments of the plaintiff that tried to devaluate the binding nature of the preliminary ruling. Oracle first claimed that Member States are free to regulate their property system.  [202] Under the German civil law, property might only exist on tangible goods rather than intangibles. Therefore, when the CJEU discussed the definition of (first) sale, it intruded into the freedom of Member States, since as a consequence it accepted property interests over intangible computer program as well. The BGH resisted this argument and stressed that the CJEU interpreted the meaning of the sale (and concluded it is a transfer of ownership), rather than the meaning of ownership at all, and the latter shall be interpreted under domestic law.  [203] Second, Oracle also noted that the CJEU ruling interfered with the WCT, when it accepted the exhaustion on intangible computer program, contrary to the content of the Agreed Statement. According to the BGH, however, the WCT introduced the right of distribution as a minimum right, and therefore the EU is not prohibited from broadening the scope of the doctrine of exhaustion in order to include the redistribution of intangibles (computer program) as well.  [204]

The BGH finally vacated the appellate court’s decision and ordered it to conduct a new proceeding. Here, the appellate court has to keep in mind first that plaintiff did obtain a “remuneration corresponding to the economic value of the copy of the work of which it is the proprietor”.  [205] Second, the defendant has to prove that UsedSoft only resold software licences that permitted a perpetual use; and that the maintenance agreements were still in force in respect of the resold computer program.  [206] The judges also stressed that UsedSoft has to demonstrate that the original purchasers made the resold computer program unworkable on their computers. Here a notarial certificate shall not be deemed per se as a proper evidence of the uninstallation. Additionally, the split of the volume licence is not acceptable, as discussed by the CJEU, and therefore the defendant has to prove that its clients sold the company only undivided volume licences.  [207] Finally, since secondary purchasers are only allowed to use the computer program for the purposes envisaged by the original licence provided by the right holder, it is the defendant’s duty to inform these secondary purchasers to comply with the provisions of the licence.  [208]

Shortly after the publication of the publishing, the UsedSoft ruling the District Court for the Southern District of New York decided on a partially similar issue. Here, ReDigi, “the world’s first and only online marketplace for digital used music”  [210] was sued shortly after launching its service in October 2011. The original version of ReDigi allowed registered users to upload their legally purchased sound recordings to ReDigi’s Cloud Locker via the company’s Media Manager program. Media Manager detected the uploader’s computer and built a list of eligible files.  [211] Only those files were eligible for resale, which were lawfully purchased through iTunes  [212] or from another ReDigi user. This technically guaranteed that “pirate” copies of music files could not enter the system. Simultaneously, with the uploading of the file to the Cloud Locker, the content was erased from the source computer. This process was generally termed as “migration” (or “atomic transaction”) of the file. The other function of Media Manager was to continuously double-check whether the user retained any copy of the files they migrated to the Cloud Locker on their computers’ hard drive or any portable device synchronized with the computer. If Media Manager detected any file like that, the users were warned to erase the said copies. If the users missed to comply with the warning, their account was terminated by the company. After uploading the files to the Cloud Locker, the users had two options: they either accessed their music for personal use or sold them to other users. In the latter case, the files were kept at the same location in the Cloud Locker. However, the “file pointer” of the said contents was changed. Consequently, only the new purchaser had the right to access the sound recording and not the seller. The new users had the ability to access the files from their accounts or to download them to their computers. Users paid with credits purchased from ReDigi for each resale. Credits could not originally be “changed back” to cash (currently, it is possible); they could only be used for further purchases. ReDigi earned a high transaction fee on each and every sale. The price of the used files varied originally between 59 and 79 USD cents (currently there are cheaper tracks as well). ReDigi retained 60% of the said price. 20% was allocated to the seller, and 20% was retained on an “escrow” fund for the respective “artist”.  [213]

Capitol Records sued ReDigi on multiple counts, and sought several different permanent and preliminary injunctions, as well as damages, attorney’s fees, costs, interests and any other appropriate relief. The plaintiff applied for a summary judgment in July 2012, and the district court granted a partial summary judgment in March 2013.  [214] The present paper is only going to focus on those parts of the holding that is relevant to the discussion of the first sale doctrine.  [215]

As it was visible from the summary of ReDigi’s operation, the system aimed that the respective music file is always located on the hard drive rather than multiple locations. As the company stressed, files are migrated from the sellers’ computers to the Cloud Locker and between the purchasers’ account and their computers. There is no migration between the two different accounts, since in this case only the file pointer is changed to provide access to the content for the lawful user. This process is exactly what Capitol found infringing, claiming that each step of the migration leads to a new reproduction of the original sound recording. Consequently, Capitol argued, since first sale doctrine only covers the distribution but not the reproduction of copyrighted works, ReDigi’s users committed copyright infringements, and ultimately ReDigi’s service functions illegally.

The district court accepted Capitol’s claims. As the judge highlighted, “courts have not previously addressed whether the unauthorized transfer of a digital music file over the Internet – where only one file exists before and after the transfer – constitutes reproduction within the meaning of the Copyright Act. The Court holds that it does.”  [216] The district court noted that it is undisputed that sound recordings are protected under U.S. copyright law,  [217] and Capitol owned copyrights on several works that were transferred via ReDigi’s system. Second, sound recordings are fixed in phonorecords that are the material objects in which the sounds are fixed and “from which the sounds can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device”.  [218] Third, reproduction of the copyrighted sound recording occurs every time, when it is fixed in a new material object.

Based upon the above logic, the district court showed absolute rigidity, when it followed the case law on P2P file-sharing  [219] and noted that “when a user downloads a digital music file or ’digital sequence’ to his ’hard disk,’ the file is ’reproduce[d]’ on a new phonorecord within the meaning of the Copyright Act.”  [220] The migration of the file – that is the fact that there was only a single copy at each step of the process – was deemed to be irrelevant by the judge.  [221] Furthermore, the court found that the electronic file transfer is clearly within the meaning of the right of distribution.  [222] Consequently, ReDigi’s users infringed both the right of reproduction and the right of distribution when they used the company’s service. The only chance to escape liability was to rely on the theory of fair use and the first sale doctrine. Since both of these work as affirmative defences, the burden of proof rested on the shoulders of ReDigi to prove that its system fits into the realm of the two above doctrines.

ReDigi failed on both of the defences. The district court, however, erred when it addressed the defence on fair use doctrine.  [223] The judge started from the wrong point of view when it accepted the direct liability of ReDigi for the reproduction and distribution of Capitol’s sound recordings.  [224] Conversely, those were the users who uploaded, migrated, sold, purchased and, finally, downloaded the music files and not ReDigi. If ReDigi is liable for these acts in any way, its liability shall be based on secondary liability doctrines rather than direct liability.  [225] Fair use doctrine, however, only applies to direct infringements, that is, to the acts of the users, rather than ReDigi’s. A Second Circuit precedent confirmed that “space-shifting” – that is, reproduction of sound recordings from computers to portable devices and vice versa – of digital contents is fair use,  [226] since those are the users, who create copies in the “cloud”, rather than service providers.  [227]

This mistake of the district court needs review from the Second Circuit, especially with respect to the reasoning on the fourth statutory factor of the fair use doctrine. Here, the district court noted that the system of ReDigi is capable of interfering with the legitimate primary markets of the right holders.  [228] On appeal, however, circuit judges will have the ability to pay attention to the fact that the doctrine of exhaustion shall not be bound by the three-step test, and any resale of tangible copies of works by lawful acquirers shall be accepted, even if it is against the primary economic interest of the right holders.  [229] To sum up: should the concept of fair use be applicable to the upload and download of the sound recordings by private users (the reseller and the purchaser respectively), the main argument of the district court will become pointless.

The district court’s reasoning on the first sale doctrine similarly causes confusion. Relying on its own previous logic, the judge noted that “the first sale doctrine does not protect ReDigi's distribution of Capitol's copyrighted works [since] an unlawful reproduction, a digital music file sold on ReDigi is not ’lawfully made under this title.’ Moreover, the statute protects only distribution by ‘the owner of a particular copy or phonorecord ... of that copy or phonorecord.’ Here, a ReDigi user owns the phonorecord that was created when she purchased and downloaded a song from iTunes to her hard disk. But to sell that song on ReDigi, she must produce a new phonorecord on the ReDigi server. Because it is therefore impossible for the user to sell her ‘particular’ phonorecord on ReDigi, the first sale statute cannot provide a defense.”  [230]

As highlighted above, it is not ReDigi who reproduced or distributed the phonograms. Furthermore, the unlawfulness of the uploaded contents is far from clear (due to the misinterpretation of the fair use doctrine). There are some further concerns that the district court’s reasoning on “particular” and “that” copy is correct. Music files sold via iTunes are marked with a Persistent ID number that individually identifies the said content. The migration of the file via Media Manager and the Cloud Locker therefore leads to the duplication and transfer of an entirely identical file marked with the same ID number. From this perspective, the content sold via ReDigi is exactly “that particular copy”.

The district court further argued that “the first sale defense is limited to material items, like records, that the copyright owner put into the stream of commerce. Here, ReDigi is not distributing such material items; rather, it is distributing reproductions of the copyrighted code embedded in new material objects, namely, the ReDigi server in Arizona and its users' hard drives.”  [231]

If the purpose of the district court was to refer to the international norms introduced above, especially to the Agreed Statement of the WCT, its argument would be totally correct. What the judge practically said is, however, that although phonograms are deemed to be material objects under the sole definition of the USCA,  [232] and the right of distribution applies to the sale of phonograms in digital format (“electronic file transfer”) as well,  [233] the first sale doctrine does not apply to digital files, since they are not “material items”. That is a clear catch 22 situation, where the judge either tries to follow the norms of international copyright law or the vague (more properly: outdated) definition of the USCA on phonograms.

Finally, ReDigi tried to satisfy the judge by policy arguments that Capitol’s interpretation of the first sale doctrine would provide broader protection to the company as envisaged by the legislators. The district court refused to accept these arguments, especially since “amendment of the Copyright Act in line with ReDigi's proposal is a legislative prerogative that courts are unauthorized and ill suited to attempt. (…) [T]he first sale doctrine was enacted in a world where the ease and speed of data transfer could not have been imagined. There are many reasons, some discussed herein, for why such physical limitations may be desirable. It is left to Congress, and not this Court, to deem them outmoded.”  [234]

III. Further case law on audio books and e-books

Since both the CJEU and the BGH confirmed that the UsedSoft case discussed the doctrine of exhaustion solely from the perspective of computer programs, several other German courts have treated the doctrine differently with regards to other subject matter.

The first relevant decision was handed over before the CJEU’s preliminary ruling was published in 2011. Here, the Federal Appellate Court of Stuttgart concluded that the download of an audio book from a web shop does not lead to exhaustion, since the latter presupposes the transfer of ownership of the physical object of the copyrighted work.  [235] Furthermore, the general terms and conditions of the purchase via the web shop  [236] provided for a use licence and, at the same time, prohibited the resale of the digital copy. The court stressed that these shall be deemed as clearly formulated, enforceable terms. The court argued that digital contents are subject to easy reproduction, and therefore copyright holders deserve the right to authorize further uses of these contents.  [237] Furthermore, although the web shop used expressions on its website, like “purchase” (“Kauf”), “purchase price” (“Kaufpreis”) or “basket” (“Warenkorb”), none of these prove that the service offered shall lead to a transfer of ownership of the audio books. Indeed, a reasonable purchaser could properly understand that these terms are used functionally, and he could easily understand the contractual terms as well (especially the prohibition on resale).  [238]

In another case, an association designed for consumer protection sued an online bookstore to enforce the applicability of the doctrine of exhaustion on audio books sold and offered for download via the defendant’s website. The Federal Appellate Court of Hamm followed the lower court’s arguments,  [239] and refused the plaintiff’s claims.  [240]

The court analyzed defendant’s general terms and conditions  [241] from two perspectives. First, whether it is clear enough to be understood by an average client as it was designed for by the company. This question was answered positively when the wording of the terms was declared to be clear and understandable for anyone having a minimal technological knowledge, as well as declaratory and constitutive enough to make it clear that the contract allows only for a limited, non-alienable right to use the copyrighted work.  [242]

The court further discussed whether the general terms and conditions properly prohibited the resale of audio books, or on the contrary – as plaintiff asserted – if it contradicted CJEU’s UsedSoft ruling, consequently and unlawfully limiting purchasers’ freedom to transfer the ownership acquired on the copies of works. Here again, the appellate court refused plaintiff’s arguments. Irrespectively of the fact that the court improperly called the online transfer of audio books streaming,  [243] it followed the mainstream interpretation of WCT and the InfoSoc Directive by declaring the transmission of data as making available to the public rather than distribution. To reach that conclusion, the panel relied both on the existing case law on the exhaustion of distribution right related to audio books (including the OLG Stuttgart decision introduced above) and on the justification for the implementation of the making available to the public right by the German Parliament.  [244]

The appellate court further analyzed a draft bill that was submitted to the German Parliament by the parliamentary group “Die Linke” in 2012.  [245] This document called for the amendment of the German Copyright Act to allow for the resale of the copies of works by the lawful acquirer if no further copies are retained by the reseller, and if the copies are furthermore not made available to the public.  [246] The appellate court concluded that since the proposal was refused by the German Bundestag in 2013, the legislator has been unwilling to amend to status quo.  [247] Finally, the appellate court also excluded the application of the UsedSoft decision, since it referred to a different subject matter, namely computer programs that were specifically treated by both the European and the German legislators.  [248]

The most recent debate over digital resale of literary works originates from the Netherlands. There, a start-up called Tom Kabinet was launched in June 2014. The founders of the site aimed to support personal users to sell and purchase lawfully acquired “used”, DRM-free e-books.  [249] Only eight days after it started to operate, Tom Kabinet was already threatened with suit by the Dutch Trade Publishers Association. The latter treated the new service as illegal, even though Tom Kabinet planned to keep 20% of the purchase price of each and every e-book sold through its system on an escrow fund for the benefit of the specific author.  [250] After an unsuccessful negotiation, the association initiated a suit against Tom Kabinet and requested preliminary injunctions against the website. The competent court refused to order those preliminary injunctions, claiming that it is not self-evident under the UsedSoft decision that the resale of used e-books is precluded under European law.  [251] Bodewits noted that “[a] lot of emphasis was placed on the fact that Tom Kabinet adds a new watermark to the e-book after it has been purchased in an attempt to prevent trade in illegal copies. Although this may not be sufficient to prevent all illegal trade, the interim relief judge considered that further protective measures could not have been implemented without cooperation of the publishers. Moreover, the interim relief judge was clear that the behaviour of the publishers, by not replying to the invitation to discuss participation but instead initiating interim relief proceedings, was a step too far given the good intentions of Tom Kabinet.”  [252] The Dutch Court of Appeals (Hof Amsterdam) similarly refused to order the shut-down of Tom Kabinet. The panel concluded in its preliminary ruling that it is uncertain without a full trial whether the service offered by Tom Kabinet would be excluded under EU law.  [253]

The comparison of the court decisions shall begin with the analysis of the licence versus sale dichotomy. The relevance of this question is extremely important. Licence allows for the use of the original or the copy of the work other protected subject matter, and the conclusion of a licence contract does not involve the sale or any other form of transfer of ownership of the said work or other protected subject matter. Consequently, there is no way to apply the doctrine of exhaustion in the case of a licence/use contract.

The relevant court decisions show a significant difference in relation to this issue. One of the extremes was expressed by the CJEU in the UsedSoft case. There, the court highlighted that a licence might be characterized as a sale if the right to use a computer program (1) lasts for an indefinite period, and (2) “in return for payment of a fee designed to enable the copyright holder to obtain a remuneration corresponding to the economic value of the copy of the work of which he is the proprietor”.  [255] Furthermore, (3) merely calling a contract a licence is not enough “to circumvent the rule of exhaustion and divest it of all scope”.  [256]

The CJEU’s opinion has been criticized by many commentators.  [257] Stothers noted, for example, that it is not the judges’ task to limit the available remuneration for the right holders. As Stothers claimed: “intellectual property normally provides an exclusive right (an absolute right to exclude others from using the intellectual property) and just a right to ‘reasonable’ royalties. The actual value of the rights is then determined by negotiation in the marketplace.”  [258] Contrary to his opinion, however, the CJEU did not take any steps towards limiting the freedom of right holders to negotiate the value of their rights. What the CJEU said is that the right of distribution is exhausted as soon as the protected subject matter is put into circulation by or with the consent of the right holder in exchange for a reasonable remuneration. The CJEU noted in another notable preliminary ruling that right holders might demand reasonable remuneration rather than “the highest possible remuneration”.  [259] This theory has been described by the German copyright law as “reward theory” (“Belohnungstheorie”), and thus it is neither a rootless argument, nor is it irrational to rely on by the CJEU.  [260]

An important concurring opinion needs to be taken here. According to the CJEU, sale equals to the transfer of ownership rights in tangibles or intangibles.  [261] Such a bright line rule shall not be generally followed. What property means depends upon the legislation of the respective country. In Austria, for example, property might exist on intangibles, as well.  [262] Germany regulates it the opposite way.  [263] This is why Oracle claimed it correctly in front of the BGH in 2013 that the CJEU intruded into Germany’s freedom to regulate its property system.  [264] The verbal gymnastics used by the BGH, according to which the CJEU did not determine the definition of ownership but declared the transfer of ownership as a sale, is pointless. The CJEU specifically stressed that, under sale, someone “transfers to another person his rights of ownership in an item of tangible or intangible property belonging to him”.  [265]

On the other hand, German courts take the position that computer program might be subject to sale.  [266] German academia similarly confirms that computer programs, like any other intangibles might be sold, even though under the provisions of the German Civil Code this does not lead to the transfer of ownership, since ownership might only exist over physical goods rather than intangible data. Contrary to the above, the German audio book decisions have taken the stance that the doctrine of exhaustion might only be applied where the distribution involves the transfer of ownership on the protected subject matter.  [267] With respect to audio books as digital data, this requirement cannot be met.  [268] In summation, the CJEU improperly noted – at least in respect of German domestic law – that sale might include the transfer of ownership on intangibles.

The CJEU’s argumentation might remain acceptable, however, if we focus only on the expression of “sale” rather than “transfer of ownership”. The sale of rights (“Rechtskauf”) is clearly accepted under the German Civil Code,  [269] and therefore the sale of copyrighted content via the internet shall remain out of the scope of Recital 29 of the InfoSoc Directive.  [270]

Two further notes need to be added. German case law confirms that a licence to use a protected subject matter might be transferred by the licencee to a new licencee with the permission of the original licensor.  [271] Since Oracle’s original licence contracts explicitly precluded the transfer of the respective licence and UsedSoft ultimately purchased and resold licenses rather than copies of computer programs without proper authorization, CJEU’s argumentation seems to be totally misleading. UsedSoft’s practice might be declared ab ovo illegitimate under the relevant provision of the German Copyright Act.

U.S. federal courts have similarly confirmed that the conclusion of use contracts might ultimately lead to the sale of computer programs.  [272] The Ninth Circuit introduced a balanced analysis of several factors in its Wise decision when it interpreted several contracts for the use of film reels. These factors included the designation of an agreement as a license; the reservation of title in the copyright holder; a single, upfront payment by the transferee for a particular copy; the requirement to destroy the transferred copies; significant restrictions on the terms of use; and the inability of the right holder to regain possession.  [273] The Ninth Circuit held that especially those contracts led to a sale rather than a licence, where the transferee paid a single, upfront payment for the film reels; and where it was impossible to return the said reels to the right holders, or it was not required to do so, the transferee’s possession over the copy was, consequently, for an indefinite period.  [274]

The opposite position is mirrored by the German audio books decisions, where the clear contractual prohibition on the resale of the audio books was declared to be the decisive factor in the analysis. Similar to the Wise holding, but directly with an opposite result, the expressions used by the online booksellers (“purchase” [“Kauf”], “purchase price” [“Kaufpreis”] or “basket” [“Warenkorb”]) turned out to be irrelevant.  [275]

Many U.S. court judges reached the same conclusion,  [276] and two of these rulings deserve special attention here. First, the Aftermath case focused on the exploitation of the rights to Marshall B. Mathers III.’s (a.k.a. Eminem’s) sound recordings. F.B.T. Records, the original sound recording producer, contracted with Aftermath Records about the sale of the sound recordings via multiple channels (including iTunes), especially as ringtones. Aftermath was allowed to sell the tracks either as “single records”  [277] or as copies produced from a licenced master copy.  [278] The royalty rate for each sold record was 12-20%, whilst Aftermath owed 50% for each copy produced from a master. The Ninth Circuit ultimately held that Aftermath provided only a master to each of its business partners, and allowed for the reproduction of that master in an unlimited number. Consequently, neither iTunes, nor end-users received a proprietary interest over the copy they received. The contract was categorized as a licence rather than a sale.  [279]

Only a week after the publication of the Aftermath decision, the Ninth Circuit handed over another important ruling in the Vernor v. Autodesk case. Timothy Vernor purchased a copy of the AutoCAD Release 14 program at a garage sale in 2005, and ten further copies from the Cardwell/Thomas & Associates in 2007. Vernor later sold several copies of the computer program on eBay. Purchasers of the CDs were expressly warned on the cover of the carrier that the program is only licenced for use and that a use contract is concluded by the installation of the computer program. Should the purchaser disagree with these terms, Autodesk – the producer of the computer program – guaranteed a full refund upon return of the CD. Importantly, Vernor did not ever install any copy of the computer program, nor did he open any of the CDs’ covers.  [280]

Under the software licence agreement, Autodesk retained title to all copies; granted customers a nonexclusive and non-transferable license to use the computer program; imposed transfer and significant use restrictions; provided for license termination, should the user produce an unauthorized copy of the computer program or infringe the above restrictions. Finally, the agreement required the destruction of the original copy of the computer program, should the user upgrade to a newer version of the program.  [281]

The district court followed the detailed analysis set by the Wise court and concluded that Vernor’s possession over the copy was for an indefinite period, and therefore the reservation of the title in Autodesk was invalid. More factors favoured Vernor and the applicability of the first sale doctrine than Autodesk and the exclusion of the doctrine.  [282]

The Ninth Circuit reversed the district court’s decision. First, the circuit judges noted that “we considered [in the Wise decision] whether the agreement (a) was labeled a license, (b) provided that the copyright owner retained title to the prints, (c) required the return or destruction of the prints, (d) forbade duplication of prints, or (e) required the transferee to maintain possession of the prints for the agreement's duration.”  [283] Quite interestingly, the circuit judges did not take into account two of those factors that the district court relied on and that clearly favoured Vernor: the single, upfront payment and the possession of the copy for an indefinite period. Second, the Ninth Circuit stressed that “[w]e read Wise and the MAI trio to prescribe three considerations that we may use to determine whether a software user is a licensee, rather than an owner of a copy. First, we consider whether the copyright owner specifies that a user is granted a license. Second, we consider whether the copyright owner significantly restricts the user's ability to transfer the software. Finally, we consider whether the copyright owner imposes notable use restrictions.”  [284] Based upon these three – quite restrictive – factors, the Ninth Circuit ultimately concluded that Autodesk did only licence the use of the AutoCAD program but, consequently, any resale of the copies was prohibited.  [285]

The Supreme Court refused to grant certiorari in the case,  [286] and therefore implicitly accepted the Ninth Circuit’s holding, that on the other hand received harsh criticism.  [287] First, the facts of the cases that the circuit judges referred to (that is, Wise and MAI trio) were quite different than the facts in the Vernor case. For example, two prongs of the MAI trio – the MAI Systems and the Triad Systems decisions – related to the maintenance or repair of the computers that the protected computer program was installed on. During the first half of the 1990’s, only lawful owners, not licencees, of the computer programs were allowed to reproduce the computer programs for maintenance purposes. Without entering into any deep discussion of the contract that the software producers and the companies concluded, the Ninth Circuit held that the defendants were only licencees, and therefore they illegally produced copies of the program. The irrationality of this restriction was later acknowledged by the U.S. Congress when it enacted the Computer Maintenance Competition Assurance Act in 1998. Under the amendment, lawful licencees are allowed to use the computer program for maintenance and repair purposes as well.  [288] To put it differently, two of those decisions that the Ninth Circuit relied on in the Vernor case were already overruled by the Congress by 1998.

Another compelling reason for the criticism might be deducted from another decision handed over in the same circuit (although not by the Ninth Circuit itself). The SoftMan v. Adobe decision confirmed that the terms of the licence agreement do not bind the purchaser of the computer program if he misses to install the software.  [289] It might be important to recall that Vernor never installed any of the AutoCAD program on his computer. Furthermore, the SoftMan court stressed that “the circumstances surrounding the transaction strongly suggests that the transaction is, in fact, a sale rather than a license. For example, the purchaser commonly obtains a single copy of the computer program, with documentation, for a single price, which the purchaser pays at the time of the transaction, and which constitutes the entire payment for the ’license’. The license runs for an indefinite term without provisions for renewal. In light of these indicia, many courts and commentators conclude that a 'shrink-wrap license’ transaction is a sale of goods rather than a license.”  [290]

Oddly, the ReDigi court did not analyze the licence versus sale dichotomy in detail, although this issue could have had clear relevance, especially in light of the Aftermath holding (and the involvement of iTunes in the debate). There is slight indicium that the court took the view – as defendant claimed and as commentators stressed it as well – that iTunes originally sold rather than licenced the tracks that were later uploaded to ReDigi’s Cloud Locker. The court observed that “[h]ere, a ReDigi user owns the phonorecord that was created when she purchased and downloaded a song from iTunes to her hard disk.”  [291]

All in all, the CJEU did not commit a failure when it declared Oracle’s licence agreement as a sale. A contract that allows for a possession/use of the work for an indefinite period, in exchange for a single, upfront payment results in a sale.  [292] Similarly, it is not the expressions used by the seller but the content of the agreement which shall be decisive. As a consequence, the present paper opines that the above factors prevail over the reservation of the title or any use restrictions applied by the seller. Ultimately, the above argument means that digital content might be sold (even though it will not lead to the transfer of ownership over physical goods) and, consequently, might fall under the concept of distribution.

Since exhaustion only applies to the right of distribution, it is crucial to decide how we should categorize the sale of digital contents via the internet from the point of view of copyright law.

It might be important to recall that the USCA reserves the right to “distribute copies or phonorecords of the copyrighted work to the public by sale or other transfer of ownership, or by rental, lease, or lending” to the owner of the copyright.  [293] Similarly, under the InfoSoc Directive, “Member States shall provide for authors, in respect of the original of their works or of copies thereof, the exclusive right to authorise or prohibit any form of distribution to the public by sale or otherwise”.  [294] All of these provisions are backed by the WCT’s definition of distribution. Under this, “[a]uthors of literary and artistic works shall enjoy the exclusive right of authorizing the making available to the public of the original and copies of their works through sale or other transfer of ownership.”  [295] Historically, the right of distribution was going to mean the commercialization of physical, tangible copies.

As Chapter B.I. introduced above, with the rise of the internet age, a solution was needed to handle online uses. The so-called umbrella solution was codified by the WCT and WPPT. Proponents of this theory envisaged the general, technology-neutral making available to the public right to cover each and every on-demand transmission of data via the internet. However, they left Member States’ freedom unaffected to implement that right in their domestic laws. Consequently, signatories of the treaties were allowed to introduce that right within the frames of the right of communication to the public (the European Union did so) or the right of distribution (the United States of American did – most probably – so). That is, the content, rather than the form was important. The only requirement was to provide protection for right holders with respect to such uses where the end-user might access the content from a place and at a time freely chosen by him, no matter which way.  [296]

Since the U.S. did not implement the treaty language, the most probable scenario is that the right of distribution covers the making available to the public right, as well. Any other interpretation would mean that the USCA is contrary to the WCT/WPPT. In light of this interpretation, the district court’s holding that an electronic file transfer is within the meaning of distribution shall be deemed correct.  [297] The Federal Appellate Court of Hamm followed a similar argument.  [298] The CJEU accepted that the transfer of data via the internet fits generally into the communication rights, and especially into the making available to the public right.  [299] There is one significant difference between the latter and the earlier decisions. As the CJEU noted: “the transfer by the copyright holder to a customer of a copy of a computer program, accompanied by the conclusion between the same parties of a user licence agreement, constitutes a ‘first sale … of a copy of a program’ (…) [T]he existence of a transfer of ownership changes an ‘act of communication to the public’ provided for in Article 3 of [the InfoSoc] directive into an act of distribution referred to in Article 4 of the [InfoSoc] directive.” (Emphasize added.)  [300]

In fact, the CJEU differentiated between two types of uses via the internet. In the first scenario, uses that do not lead to the permanent reproduction or sale of any copy of a protected subject matter shall be governed by the making available to the public right. Such an example might be the posting of content on a website (for example a journal article on The New York Times’ home page), on-demand streaming (services offered by YouTube, Pandora, Spotify etc.) or dissemination of files via P2P file-sharing applications. Under the second scenario, a permanent copy is received by the end-user in exchange for a purchase price and is retained on a permanent basis. The best example might be the purchase of a track from iTunes. Another example is exactly mirrored by the UsedSoft case. The CJEU declared this second category of uses to be sale and, consequently, distribution of copyrighted subject matter, rather than making available to the public.

Many commentators criticized the above logic.  [301] At first glance, they might be correct. The right of making available to the public was harmonized by the InfoSoc Directive (in accordance with WCT/WPPT) and covers every subject matter protected under the EU copyright law. This means that the InfoSoc Directive and the WCT/WPPT have a commanding role when interpreting the Software Directive and the online transmission of computer programs.

If we examine, however, both the exact wording of the making available to the public right, as well as the business model used by UsedSoft, ReDigi, Tom Kabinet and all the other companies, the standard logic of the umbrella solution deserves some criticism. The said right is formulated as follows: “making available to the public of their [works/phonograms/performances fixed in phonograms] in such a way that members of the public may access these [works/phonograms/performances fixed in phonograms] from a place and at a time individually chosen by them”.  [302] Although the business models of the affected companies are generally available to any member of the public, access to a specific content is only conditional, to put it differently, not on-demand. One important obstacle hampers users from accessing the contents “from a place and at a time individually chosen by them”. They need to accept the terms of the sale; that is, they have to purchase the subject matter under specific conditions. Ultimately, access to protected subject matter is not on-demand but “pay-walled”. It is therefore not irrational to draw a line between different uses of end-users conducted on the internet, as the CJEU envisaged that.  [303] Viewed from this perspective, the judicial activism of the CJEU deserves handshake rather than criticism.

Sadly, the CJEU’s ruling is not absolutely comparable with the other decisions introduced above, especially not with the German audio book cases. There, the decisive factor was the contractual interpretation of the agreement of the parties, rather than the copyright aspects of the supply of digital media. Furthermore, the ReDigi case focused on the distribution versus reproduction dichotomy, rather than the making available to the public right.

As stressed above: the doctrine of exhaustion applies only to the right of distribution and, consequently, not to the right of reproduction. Although the EU (and almost all Member States’) copyright law allows for the copying for private purposes, and it might be accepted under the fair use doctrine as well, this limitation of the reproduction right shall not in any way lead to the broadening of the first-sale doctrine’s reach. To put it differently, under the doctrine of exhaustion the lawful acquirer of a protected subject matter might only resell “that particular” copy that he owns/possesses, and consequently the creation of a “new copy” might lead to the exclusion of the doctrine’s applicability.

This latter “new copy theory” is nothing new under the sun. It has been articulated for quite some time that the transmission of digital contents via the internet leads to the reproduction of the protected subject matter on the end-users computer, and any further transmission of the bits – even within the same computer – results in a new copy. As the U.S. Green Paper highlighted: “the system encompassed by sections 106(3) and 109(a) appears to 'fit' only 'conventional' transactions in which possessory interests in tangible copies are conveyed in the first instance (…) Electronic disseminations, by contrast, typically involve the proliferation of copies, with the 'publisher' retaining its copy and the user acquiring a new one.”  [304] Similarly, the U.S. White Paper noted that “the first sale doctrine does not allow the transmission of a copy of a work (through a computer network, for instance) because, under current technology, the transmitter retains the original copy of the work while the recipient of the transmission obtains a reproduction of the original copy (i.e., a new copy), rather than the copy owned by the transmitter.”  [305]

The court decisions introduced above show a significant difference in opinion when reviewing this “new copy theory”.

The UsedSoft decision seems to sidestep this concept. The judges noted that “[s]ince the copyright holder cannot object to the resale of a copy of a computer program for which that right holder’s distribution right is exhausted under Article 4(2) of Directive 2009/24, it must be concluded that a second acquirer of that copy and any subsequent acquirer are ‘lawful acquirers’ of it within the meaning of Article 5(1) of Directive 2009/24. Consequently, in the event of a resale of the copy of the computer program by the first acquirer, the new acquirer will be able, in accordance with Article 5(1) of Directive 2009/24, to download onto his computer the copy sold to him by the first acquirer. Such a download must be regarded as a reproduction of a computer program that is necessary to enable the new acquirer to use the program in accordance with its intended purpose.”  [306] If we combine this argument with the CJEU’s note that the download (reproduction) of a computer program and the conclusion of a licence agreement (ultimately classified as sale) shall be examined as an “indivisible whole”,  [307] it seems that the CJEU broadened the scope of the exhaustion doctrine.

Such a claim is not entirely accurate First, in the UsedSoft case, no computer program was technically transferred between the clients of the company, and the new purchaser acquired a copy from Oracle’s freely accessible website. Only licence keys were transferred. Second, the CJEU referred to a specific provision of the Software Directive. Under Article 5(1): “[i]n the absence of specific contractual provisions, [permanent reproduction of a computer program] shall not require authorisation by the right holder where they are necessary for the use of the computer program by the lawful acquirer in accordance with its intended purpose, including for error correction.” Consequently, if CJEU’s “indivisible whole theory” is correct, and the second (and any later) acquirer of the licence key is a lawful acquirer of the computer program, then the lawful acquirer is allowed to reproduce the said computer program.

On the other hand, the “new copy theory” is followed in the German audio book cases  [308] and by the ReDigi holding.  [309]

Still, the complicated nature of the “new copy theory” is clearly mirrored by the following argument. The ReDigi court concluded that “[s]ection 109(a) still protects a lawful owner's sale of her ’particular’ phonorecord, be it a computer hard disk, iPod, or other memory device onto which the file was originally downloaded”.  [310] A similar argument was formulated by the Federal Appellate Court of Hamm in 2014.  [311] Likewise, the US White Paper opined in 1995 that “the first sale doctrine should apply if the particular copy involved is, in fact, the copy that is further distributed, even if the copy was first obtained by transmission. Furthermore, if the technology utilized allows the transmission of a copy without making an unlawful reproduction – i.e., no copy remains with the original owner – the first sale doctrine would apply and the transmission would not be an infringement.”  [312] Although these thoughts are based on the logical interpretation of the respective copyright law, they are far from reality in several cases. It is true that media contents might be directly downloaded to a memory stick, an exterior hard-drive, a Smartphone, or mp3 players. However, average users, who are absolutely strangers to the small nuances of copyright, quite often download the content first to their computer’s hard drive and reproduce the file on any device thereafter. Sometimes they first “move” the file to another folder of the computer. Conversely, some devices, especially those produced by Apple (iPod, iPad, iPhone), need to be connected to a computer first, in order to allow the iTunes to synchronize the user’s account (files kept in the user’s iTunes library) and the device. To put it differently, in several cases, portable devices might only be indirectly carrying digital contents that shall be ultimately categorized as “not that particular copies”.  [313]

Another problem arises if the original acquirer produces a copy of the lawfully purchased copy of the work, keeps the copy for himself, and later sells the original data carrier. This behaviour is clearly against the scope and purpose of the doctrine of exhaustion and has therefore been criticized by academia.  [314] Some statutes also explicitly prohibited such activities,  [315] and U.S. case law similarly suggests that copies made incidental to sale are excluded from the scope of the fair use (that otherwise might cover private copies).  [316]

All of these concerns lead us to quite an important question that is mirrored by the ReDigi case, where itwas called “migration”. This issue ultimately relates to the so-called forward-and-delete technologies. Under ReDigi’s original concept of migration, the legally purchased iTunes tracks were erased simultaneously with the transmission of the content (and all related metadata) to ReDigi’s Cloud Locker. To put it differently, there was always only one single copy of the said digital file. The question of forward-and-delete technologies was already refuted by the U.S. White Paper in 1995. According to the report: “[s]ome argue that the first sale doctrine should also apply to transmissions, as long as the transmitter destroys or deletes from his computer the original copy from which the reproduction in the receiving computer was made. The proponents of this view argue that, at the completion of the activity, only one copy would exist between the original owner who transmitted the copy and the person who received it – the same number of copies as at the beginning. However, this zero sum gaming analysis misses the point. The question is not whether there exists the same number of copies at the completion of the transaction or not. The question is whether the transaction, when viewed as a whole, violates one or more of the exclusive rights, and there is no applicable exception from liability. In this case, without doubt, a reproduction of the work takes place in the receiving computer. To apply the first sale doctrine in such a case would violate the reproduction right.”  [317]

Only a few years later, the U.S. Copyright Office expressed some openness towards the applicability of the forward-and-delete technology. Although the DMCA Section 104 Report stated that “[u]nless a ‘forward-and-delete’ technology is employed to automatically delete the sender’s copy, the deletion of a work requires an additional affirmative act on the part of the sender subsequent to the transmission. This act is difficult to prove or disprove, as is a person's claim to have transmitted only a single copy, thereby raising complex evidentiary concerns. There were conflicting views on whether effective forward and delete technologies exist today. Even if they do, it is not clear that the market will bear the cost of an expensive technological measure.”  [318] Ultimately the Register of Copyrights recommended no changes to the first-sale doctrine.  [319] The report especially stated that “[r]elying on a ‘forward-and-delete’ technology is not workable either. At present, such technology does not appear to be available. Even assuming that it is developed in the future, the technology would have to be robust, persistent, and fairly easy to use. As such, it would likely be expensive – an expense that would have to be borne by the copyright owner or passed on to the consumer. Even so, the technology would probably not be 100 percent effective. Conditioning a curtailment of the copyright owners’ rights on the employment of an expensive technology would give the copyright owner every incentive not to use it. In the alternative, it would be damaging to the market to expand section 109 in anticipation of the application of technological protection measures, thus giving the copyright owner a choice between significantly increased expenses, significantly increased exposure to online infringement, or not offering works online.”  [320]

Forward-and-delete technologies have been improved a lot since 2001. Although it is still a recurring argument to question the effectiveness of such methods,  [321] this might be superficial. There have been no technologies (neither analogue, nor digital) that turned out to be able to perfectly control the use of protected subject matter. The introduction of digital rights management or technological protection measures was not a success.  [322]

The development of forward-and-delete technologies might be evidenced through at least three different models. The one introduced by ReDigi might be claimed to be easy to circumvent. However, this argument is simply one-sided. It might be true that end-users kept the possibility to produce a copy on portable devices, but they might only use these tools effectively if they never ever synchronize them with the computer on which ReDigi’s Media Manager has been installed. The user might forget the fact that he copied the relevant work on a memory stick or an iPod (especially the latter), however, Media Manager won’t. Ultimately, the suspension of the user’s account until the erasure of the illegal copies represents quite a severe sanction, and has a clear deterring effect against any illicit activities.

Other companies, like Apple and Amazon took similarly steps towards introducing digital second-hand marketplaces.  [323] As a guarantee of the legality of these services, both companies submitted a patent application to the U.S. Patent and Trademark Office. The patent was issued to Amazon, with respect to its technology in January 2013.  [324] Apple’s patent application is still pending. Both of these systems are based on the technological control of the simultaneous erasure of contents once sold through the online marketplace; the payment of a specific percentage of the generated revenue to the right holders. Amazon’s patent similarly sets an artificial limitation on the number of copies that might be produced from the original copy.  [325] None of these services were launched yet. Furthermore, their economical effectiveness was also questioned. Nonetheless, the fact that Amazon successfully applied for a patent proves that these technologies are present and provide workable and – most probably – effective solutions for the resale of digital contents via the internet.

In light of these improvements on the field of forward-and-delete technologies, we need to agree with Karjala. He noted that “[w]hether erasure takes place immediately after transfer or whether the transfer takes place one byte at a time with erasure occurring as part of the ongoing process makes no difference to the end result or to the position of the copyright owner once the process has finished. Yet, if the byte-by-byte process, including erasure, is deemed the making of an unauthorized copy, the first-sale doctrine is for all practical purposes a dead letter in the digital age.”  [326] Similarly, Spedicato, whilst referring to the Dior/Evora ruling of the CJEU,  [327] claimed that “copyright holders may not invoke their exclusive rights for the purpose of preventing legitimate purchasers of a protected good from reselling that good and thus may not prevent purchasers from reproducing the work or making any other use of it as long as that use is only functional to the purpose of reselling the good”.  [328] Ultimately, Spedicato opined that “the moment a copy of the work is sold by the right holder or with the right holder’s consent, the exclusive right to control any further circulation of that copy is exhausted, regardless of whether the copy is tangible or intangible. This means that the first purchaser should be free to resell that copy to a subsequent purchaser while deleting his own copy, without having to obtain the right holder’s consent”.  [329]

Another outcome of the rigid application of the exclusive rights might be that enterprises will find alternative technological ways to circumvent the mere language of the statutes. Such a notable example arrived. The owners of ReDigi introduced an amended version of their system, called ReDigi 2.0. The most important difference, when compared to the earlier model, is that users of the service are able to download their legally purchased iTunes tracks directly to ReDigi’s servers, specifically to their Cloud Locker account. Such a “first download” was declared by most aforementioned sources (especially the U.S. Green and White Papers) as a lawful step of the transmission of the purchased content. Consequently, in ReDigi 2.0’s system, no reproduction takes place when the files are resold to a new user, since only the file pointer is overwritten. This way, the new user receives access to the content, whilst the seller loses that right. Clearly, such a model does not contradict the “new copy theory”. Should the users (either the original purchaser or the new one) download the content to their computers (and later synchronize the file to their iPods or other portable devices), this act shall be governed by the fair use doctrine  [330] or the private copying exception under InfoSoc Directive.

Should the users’ activity be deemed fair use, however, ReDigi would be clearly protected under the “staple article of commerce test” developed by the Supreme Court of the United States. The test, that was transplanted from the field of patent law into copyright law by the famous Sony v. Universal case, stresses that “[t]he sale of copying equipment, like the sale of other articles of commerce, does not constitute contributory infringement if the product is widely used for legitimate, unobjectionable purposes, or, indeed, is merely capable of substantial noninfringing uses”.  [331] Since that time, this doctrine has been followed multiple times by federal judges, including in digital copyright cases.  [332] Nonetheless, ReDigi 2.0 still does not offer any solution to the general concerns of right holders regarding users’ secondary reproduction activities. Under the new model, users are still able to produce substitute copies on portable devices or other data carriers before selling the files on ReDigi’s marketplace.  [333]

The subject matter and the right holders affected by the court aforementioned proceedings show great diversity, as well. This diversity has been usually discussed from the perspective of whether rules on computer programs represent a special law (lex specialis), and especially whether computer programs deserve different treatment, with respect to the doctrine of exhaustion compared to all of the other subject matter (sound recordings of musical compositions; sound recordings of literary works, that is, audio books; e-books).

At first sight, the answer might be negative. Computer programs are protected like literary works under the leading international IP norms.  [334] On closer inspection, however, especially with respect to the law of the European Union, the answer is not that simple. The EU rulings on computer program protection show several major differences compared to those on literary works.  [335] This is similarly proven by the fact that the protection of computer programs has been introduced, separately from the norms of other types of works, by many legislators.  [336] Related to the doctrine of exhaustion, the Software Directive does not make any distinction between tangible and intangible copies of computer programs, unlike the InfoSoc Directive via Recital 29. This is clearly reasoned by the fact that the use of computer programs – as the archetype of digital contents – is absolutely independent from the initial carrier of the data. Accordingly, source and object code is equally protected by copyright law.  [337]

The CJEU correctly noted that “from an economic point of view, the sale of a computer program on CD‑ROM or DVD and the sale of a program by downloading from the internet are similar. The on-line transmission method is the functional equivalent of the supply of a material medium.”  [338] Indeed, the above argument is also true from a technological point of view. Due to the fact that the source code needs to be entered into the memory of a computer in order to use the computer program, it is irrelevant whether that code (and consequently the protected subject matter) is copied to the computer from a DVD or from a file made available to the public via the internet. The creation of a copy (installation) on the computer is inevitable and therefore lawful.  [339]

Still, the above logic is partially flawed. The reasoning of the CJEU, as it is continued in the same paragraph quoted above, highlights an important problem of the European acquis communautaire. According to the judges, “[i]nterpreting Article 4(2) of Directive 2009/24 in the light of the principle of equal treatment confirms that the exhaustion of the distribution right under that provision takes effect after the first sale in the European Union of a copy of a computer program by the copyright holder or with his consent, regardless of whether the sale relates to a tangible or an intangible copy of the program.”  [340]

Should the right of distribution apply to the online delivery of computer programs, the CJEU either wilfully blinded itself  [341] or made a mistake when it disregarded the WCT Agreed Statement. Since the WCT did not provide any specific right of distribution or exhaustion related to computer programs, the general rule should prevail for computer programs as well. Although the right of distribution represents a minimal right under the WCT, signatories of the treaty are able provide for stronger protection to right holders.  [342] The CJEU’s above logic envisaged a weaker protection for right holders, with respect to computer programs. Furthermore, any argument according to which the Agreed Statement of the WCT does not apply to computer programs because the Council introduced the original Software Directive with different content earlier than the WCT was accepted, making it lex specialis, is misleading.  [343] The European Union aimed at implementing the WCT by the InfoSoc Directive.  [344] Consequently, the respective provisions of the WCT have become part of the EU law. The latter directive regulates a general right of distribution and includes a general exhaustion doctrine. None of these provisions are differentiated regarding the affected subject matter. It is correct that the InfoSoc Directive left the “specific provisions on protection provided for by Directive 91/250/EEC” untouched.  [345] At the same time, however, the directive stressed that “[t]his Directive is based on principles and rules already laid down in the Directives currently in force in this area, in particular Directives 91/250/EEC (…), and it develops those principles and rules and places them in the context of the information society. The provisions of this Directive should be without prejudice to the provisions of those Directives, unless otherwise provided in this Directive.”  [346] (Emphasis added.) Recital 29 of the InfoSoc Directive is such an “other provision”. Consequently, the special application of the principle of exhaustion with regards to computer programs sold in intangible format is contrary to the general rules of international IP norms and the special rules of the EU.  [347] This inconsistency shall not be accepted even if the CJEU referred to the application of the principle of equal treatment.  [348]

The question arises, whether the functional equivalence theory might be applicable to copyrightable subject matter other than computer programs as well. Did the U.S. American and German courts correctly exclude the application of the doctrine of exhaustion in the ReDigi and the audio book cases in respect of the resale of digital sound recordings/audio books via the internet? Referring back to the CJEU’s point according to which the online transmission of computer program is from an economic (and at the same time from a technological) perspective functionally the same as selling a data carrier in a tangible format, we shall clearly agree with the outcome of the said decisions of the U.S. and German courts. Both sound recordings and audio books have multiple ways of exploitation, including the distribution of copies on tangible data carriers, making available to the public or selling a digital copy via the internet, communication to the public by wire or wireless means, public performance/display etc. We might paraphrase the CJEU’s logic in the following way: “from an economic point of view, the sale of a sound recording/audio-book on a physical data carrier and the sale of the said content by downloading from the internet are not similar. The on-line transmission method is not the functional equivalent of the supply of a material medium.” The outcome is absolutely the same from the technological point of view: unlike computer programs, sound recordings/audio books do not need to be permanently copied (installed) for the purpose of enjoyment. Commentators noted that “[t]he tangible medium on which software exists has a different function from a book: to effect the reproduction of the copyrighted work. In other words, with current software technology, one does not use the software through the medium on which it is initially fixed. Instead, the user is required to copy the software onto another location, like a hard drive.”  [349]

A notable decision of the Supreme Court of Canada from 2012 views it differently. The majority of the court noted in the ESA v. SOCAN case that “[i]n our view, there is no practical difference between buying a durable copy of the work in a store, receiving a copy in the mail, or downloading an identical copy using the Internet. The Internet is simply a technological taxi that delivers a durable copy of the same work to the end user.”  [350] The Supreme Court of Canada based its majority ruling on the policy claims of balance of interests of right holders and users,  [351] further on the requirement of technological neutrality. The panel stressed that “[t]he principle of technological neutrality requires that, absent evidence of Parliamentary intent to the contrary, we interpret the Copyright Act in a way that avoids imposing an additional layer of protections and fees based solely on the method of delivery of the work to the end user. To do otherwise would effectively impose a gratuitous cost for the use of more efficient, Internet-based technologies.”  [352]

To sum up: under the existing international copyright law it is not the subject matter that decides the application of the doctrine of exhaustion, but the medium in which it was put into circulation for the first time. It is therefore irrelevant, if the resale affects computer programs, literary works or sound recordings. The question is, whether the original or the copy of the work was distributed in a tangible or an intangible format. Consequently this is the point, where the CJEU clearly went beyond the boundaries set by international norms.  [353]

A further perspective was totally neglected by the CJEU in the UsedSoft ruling that might be summarized as follows: what if a computer program incorporates other protected subject matter as well? The present article is covering two related issues below: First, the inclusion of sound recordings, audiovisual contents or photographs/graphic works into the computer program, and Second the question of graphic user interfaces.

Irrespectively of the facts of the UsedSoft case several types of computer programs, especially computer games, include further works, for example sound recordings, audiovisual contents and photographs/graphic works. These pieces of the whole work are similarly protected by copyright law, as long as they are identifiable and surpass the threshold of originality requirement.  [354] As the CJEU stressed it in its PC Box ruling: „videogames (…) constitute complex matter comprising not only a computer program but also graphic and sound elements, which, although encrypted in computer language, have a unique creative value which cannot be reduced to that encryption. In so far as the parts of a videogame, in this case, the graphic and sound elements, are part of its originality, they are protected, together with the entire work, by copyright in the context of the system established by Directive 2001/29.”  [355]

The latter statement, however, leads us to the need of the cautious consideration of two distinct arguments. Let’s suppose that CJEU’s argument in the UsedSoft case was ultimately correct. In this situation intangible computer programs would be subject to the principle of exhaustion under the lex specialis provisions of the Software Directive. On the other hand, the resale of any other subject matter in intangible form would be treated under the lex generalis provisions of the InfoSoc Directive (and the WCT Agreed Statement), and they are ultimately excluded from the scope of exhaustion. In the case of video games, however, these two statements are irreconcilable. It would mean that the program might be resold, except the sound recordings, audiovisual contents and photographs. This solution is clearly absurd. Naturally, all these uncertainties disappear, if we follow the logic expressed above and accept that the ruling of the CJEU in the UsedSoft case regarding the applicability of the exhaustion principle to intangibles is erroneous.

Graphic user interfaces are almost uniformly excluded from the scope of computer program protection both under the law of the European Union and the United States of America, especially under the idea/expression dichotomy. This is clearly evidenced by the wording of the Software Directive that stresses in its Article 1(2) that “protection in accordance with this Directive shall apply to the expression in any form of a computer program. Ideas and principles which underlie any element of a computer program, including those which underlie its interfaces, are not protected by copyright under this Directive”.  [356] The general exclusion of ideas from the scope of copyright protection under USCA §102(b)  [357] leads to the same conclusion in the United States.  [358]

The case law of the CJEU affirmed the above policy. The Grand Chamber highlighted it in its BSA v. Ministerstvo kultury ruling that “the graphic user interface is an interaction interface which enables communication between the computer program and the user. In those circumstances, the graphic user interface does not enable the reproduction of that computer program, but merely constitutes one element of that program by means of which users make use of the features of that program. It follows that that interface does not constitute a form of expression of a computer program within the meaning of Article 1(2) of Directive 91/250 and that, consequently, it cannot be protected specifically by copyright in computer programs by virtue of that directive”.  [359] Decisions with the similar conclusion were published in the UK  [360] and Hungary.  [361] Such a conclusion has been applauded by academia as well.  [362]

Similarly, U.S. judges reached the same conclusions in multiple cases.  [363] In one of the leading precedents the Second Circuit noted that “functional elements and elements taken from the public domain do not qualify for copyright protection. With respect to the few remaining parameter lists and macros, the district court could reasonably conclude that they did not warrant a finding of infringement given their relative contribution to the overall program. (…) This aspect of the program's structure [that is, the list of services required for both ADAPTER and OSCAR 3.5] was dictated by the nature of other programs with which it was designed to interact and, thus, is not protected by copyright. Finally, in his infringement analysis, Judge Pratt accorded no weight to the similarities between the two programs' organizational charts, ‘because [the charts were] so simple and obvious to anyone exposed to the operation of the program[s].’ CA argues that the district court's action in this regard ‘is not consistent with copyright law’ – that ‘obvious’ expression is protected, and that the district court erroneously failed to realize this. However, to say that elements of a work are ‘obvious,’ in the manner in which the district court used the word, is to say that they ‘follow naturally from the work's theme rather than from the author's creativity.’ This is but one formulation of the scenes a faire doctrine, which we have already endorsed as a means of weeding out unprotectable expression”.  [364] Further, in the Apple v. Microsoft case the Ninth Circuit stressed that “Apple cannot get patent-like protection for the idea of a graphical user interface, or the idea of a desktop metaphor [under copyright law] which concededly came from Xerox. It can, and did, put those ideas together creatively with animation, overlapping windows, and well-designed icons; but it licensed the visual displays which resulted”.  [365] Similarly, the Hungarian Council of Copyright Experts noted in one of its expert opinions that “the graphic user interfaces conform to the functions to be executed and the data to be managed; and therefore they do not possess such individuality and originality that would be necessary to earn copyright protection as graphic works”.  [366]

Notwithstanding the above, graphic user interfaces might deserve copyright protection as graphic works, if they meet the requirements of originality. Under EU law only the Software Directive in respect of computer programs,  [367] the Database Directive in respect of databases,  [368] and the Copyright Term Directive in respect of photographs  [369] refer explicitly to originality as a prerequisite of protection. Nonetheless, the CJEU indirectly expanded the scope of the concept of originality to other protectable subject matters in its Infopaq ruling as well. There, the Fourth Chamber concluded that “copyright within the meaning of Article 2(a) of Directive 2001/29 is liable to apply only in relation to a subject-matter which is original in the sense that it is its author’s own intellectual creation.”  [370] The question whether such a constructive interpretation of lex specialis provisions of the three distinct directives listed above shall be treated as erroneous certainly overstrains the scope of the present paper. In any case the broad concept of originality expressed by the CJEU in the Infopaq case generally complies with the domestic treatment of the concept of originality. Indeed, it was followed by the CJEU in its BSA v. Ministerstvo kultury ruling as well: „the graphic user interface can, as a work, be protected by copyright if it is its author’s own intellectual creation. It is for the national court to ascertain whether that is the case in the dispute before it.”  [371]

With this respect the Navitaire decision deserves attention yet. There the High Court stressed that “the [Software] Directive is concerned only with the protection of computer programs as literary works, and I do not read it as having any impact on relevant artistic copyrights. It is certainly possible to view the GUI screens as tables, because they are 'drawn' by selecting from a palette of available objects things such as command buttons, toggle buttons, checkboxes, scrolling lists and so forth and moving them around on a form until a satisfactory layout is concerned. The 'interface builder' program provides 'stubs' for the routines that will be executed when the user selects or clicks on one of these objects, and it is the task of the programmer to provide the necessary code to ensure that the right thing happens when the user presses (for example) the OK button. Although composed of elements made available by the manufacturer of the interface builder program, I can see that the screen resulting from such an operation might properly be considered to be an artistic work. What the programmer ultimately produces is code that depends upon a large number of complex graphic routines that draw the background, the boxes and the shading in the places selected, and act appropriately when the mouse moves over them or they are selected. Programmers do not write this code: it is the scaffolding for their own window design. In my judgment, the better view is that the GUI screens are artistic works. They are recorded as such only in the complex code that displays them, but I think that this is strictly analogous to more simple digital representations of graphic works. The code constructs the screen from basic elements, and is so arranged to give a consistent appearance to the individual elements. I think, nonetheless, that to arrange a screen certainly affords the opportunity for the exercise of sufficient skill and labour for the result to amount to an artistic work. I consider that the GUI screens satisfy this requirement. There is force in the suggestion that they present a uniform appearance in layout of the elements, and so contribute to a uniformity of interface. On the whole this is sufficient skill and labour to entitle the screens sued on to artistic copyright”.  [372]

The U.S. Court of Appeals for the Federal Circuit overruled the previously introduced Atari precedent in the Oracle v. Google case. The panel concluded that “a set of commands to instruct a computer to carry out desired operations may contain expression that is eligible for copyright protection. (...) We agree with Oracle that, under Ninth Circuit law, an original work – even one that serves a function – is entitled to copyright protection as long as the author had multiple ways to express the underlying idea.”  [373] The Federal Circuit’s decision was immediately appealed by Google, and a very convincing amicus curiae was submitted by a group of U.S. copyright professors to the Supreme Court.

The professors supported the reversal of the decision on three main grounds. First, they claimed that the copyright protection does not extend to procedures, processes, systems, methods of operation, and other useful arts embodied in computer programs under USCA §102(b). The professors opined that APIs clearly fit into this provision.  [374] Second, the judicially developed doctrine on structure, sequence and organization (SSO) of computer programs – that has been otherwise discredited by multiple circuit decisions – should not be invoked in a way that negates Congress’s intent in USCA §102(b).  [375] Finally, the amicus curiae opined that when APIs constrain the design choices of subsequent programmers, the merger doctrine precludes copyright protection for the interface design. Under this doctrine, external factors like compatibility requirements limit the scope of copyright protection of programs, where these factors restrict the freedom of programmers in selecting the features of APIs. Contrary the Federal Circuit’s opinion quoted above, the amicus curiae noted that “courts often recognize that when there is only one or a small number of ways to express an idea, idea and expression will be considered to have merged, and no copyright protection is available to the merged elements”.  [376] The professors concluded that “the Federal Circuit has incorrectly applied binding Ninth Circuit precedents and issued an opinion that splits from rulings by the First, Second, Sixth, ninth, Tenth and Eleventh Circuit Courts of Appeals”.  [377]

Since the CJEU’s UsedSoft ruling is silent in this respect, this problem requires careful review by national judges. Until that point the stalemate mentioned above in respect of sound recordings, audiovisual contents and photographs/graphic works incorporated into the computer program equally applies to graphic user interfaces.

The first question that needs to be discussed here is whether the idea of digital exhaustion is relevant at all in the digital environment or just a fancy hype that supports the interests of several corporations and not the whole society. Indeed, this notion might be deemed as hype only as long as its meaning and relevance in the digital age is settled and undisputed on the one hand, and these doctrinal frames are unreasonably challenged, on the other hand.

With regard to the first aspect above, it needs to be stressed again that the first sale doctrine has originally rooted in the protection of personal property and of the owners over the economic interests of copyright holders. Since the emergence of digital technologies the concept of ownership has gone through significant changes, where the proprietary interests over intangible goods has been questioned in several countries. There is one condition that is generally admitted in the legal systems covered above: intangible goods might be subject to sale. Consequently, the digital file that contains a protectable subject matter shall be under the control of the lawful purchaser. It seems to be fair to claim broader abilities on the side of the customers (including the ability for further resale of the digital content), even in the lack of proprietary interests.

Such a claim might be justified from the perspective of general users; however, it might be similarly supported from the perspective of cultural organizations, like public libraries. The general aim of the latter is to distribute knowledge and information amongst members of the society.  [379] In the lack of a digital first sale doctrine these institutions are deprived of the possibility to participate in the downstream digital commerce.  [380] They are unable to rely on the benefits of digital society, and they are ultimately forced to spend enormous amounts of money on non-transferable digital copies of cultural goods (especially books).  [381] As a consequence, libraries are going to lose the chance to fulfil the demands of digital natives that could ultimately lead to the questioning of their existence at all.

Not everyone shows enthusiasm, however, towards the introduction of the doctrine of digital exhaustion, if it is about the possible effects upon public libraries. Necessarily, such a concept would not be the end result, but only the means to fulfil the said cultural aims of these institutions, including lending of digital contents to the patrons. As Chiarizio noted “[w]hile this form of digital lending is technically possible, it raises plenty of administrative problems. For example, patrons will inevitably ‘lose’ the digital copy on loan (through deletion, hardware failure, etc.), and libraries may then be required to buy a replacement copy. Libraries will also have a very difficult time policing patrons to be sure that all patron copies are deleted upon return. Besides administrative quandaries, publishers will notice that lack of wear and tear on digital books is resulting in decreased sales to libraries. Prices of e-books will have to increase in order to compensate for the decrease in replacement purchasing”.  [382] Indeed, Chiarizio stressed that there is a viable model for e-lending by public libraries in the United States, under which several major book publishers allow for the lending of e-books by the institutions, if they comply with the requirements set by management companies, like OverDrive, 3M or Baker & Taylor.  [383]

A further counter-argument might be stressed here, however. Commentators have constantly highlighted that the use of licence contracts is going to eclipse the use of sale contracts. This is especially true in respect of the use of works and other protected subject matters in the online environment.  [384] Unlike Oracle most of the licensors allow for the temporary use of contents the licensees, which ultimately allows for the ignorance of the UsedSoft ruling of the CJEU.

Such a great example is the subscription model of service providers like Netflix for audiovisual contents, Pandora or Spotify for sound recordings or Amazon’s Kindle Unlimited for e-books. Similarly, the fact that service providers “move to the cloud” might put an end to the ado related to this territory. Since access to contents via the cloud is granted on a temporary basis in exchange of a periodic payment of the licence fee, and users are not provided any permanent copy of the content, this model shall be qualified without any problem as a service.  [385] Under copyright law the above model is based upon communication to the public of the said contents rather than their distribution. Verbraecken interprets the above shift in the available models as a method of “exhaustion evasion”.  [386]

The fact that the available options of access to protected subject matter tend towards subscription based models offered via the cloud does not necessarily mean that users lost their – culturally and historically strong – interest in the possession of permanent copies of copyrighted works. Since the subscription based services mentioned above did not generate any profit until the end of 2014,  [387] it is clearly questionable, whether their systems will survive on a medium or long distance.

With regard to the second aspect mentioned above it is similarly clear that the concept of exhaustion in the digital age is far from settled or unquestionable. This is best evidenced by two distinct preliminary rulings of the CJEU and an opinion of the Advocate General in a third case.

First, the judgment of the CJEU in the Svensson case – and further its order in the following BestWater case  [388] – practically created new substantive law in the EU. The ratio of the two rulings is that the right of making available to the public under Article 3(3) of the InfoSoc Directive is practically exhausted, as long as the source material that is hyperlinked or embedded by a user on its own website has been available online before the use takes place.  [389] The CJEU expressly noted that “[t]he public targeted by the initial communication consisted of all potential visitors to the site concerned, since, given that access to the works on that site was not subject to any restrictive measures, all Internet users could therefore have free access to them. In those circumstances, it must be held that, where all the users of another site to whom the works at issue have been communicated by means of a clickable link could access those works directly on the site on which they were initially communicated, without the involvement of the manager of that other site, the users of the site managed by the latter must be deemed to be potential recipients of the initial communication and, therefore, as being part of the public taken into account by the copyright holders when they authorised the initial communication. Therefore, since there is no new public, the authorisation of the copyright holders is not required for a communication to the public such as that in the main proceedings. Such a finding cannot be called in question were the referring court to find, although this is not clear from the documents before the Court, that when Internet users click on the link at issue, the work appears in such a way as to give the impression that it is appearing on the site on which that link is found, whereas in fact that work comes from another site. That additional circumstance in no way alters the conclusion that the provision on a site of a clickable link to a protected work published and freely accessible on another site has the effect of making that work available to users of the first site and that it therefore constitutes a communication to the public. However, since there is no new public, the authorisation of the copyright holders is in any event not required for such a communication to the public.”  [390]

The above ruling was confirmed by the BestWater order in respect of embedding technology, even though the facts of the two cases showed a major difference in one significant aspect. Unlike in the Svensson case, where the online newspaper articles were originally posted by the right holders, the source material in the BestWater case was uploaded to and made available via YouTube without the prior approval of the right holder. Such a factual difference might be treated as a material one that allows for distinguishing in common law countries. The fact that the CJEU overlooked that aspect of the case  [391] led to a highly questionable practice.

The Art & Allposters case further mirrors that the application of the doctrine of exhaustion in the 21^st century is not without problems. There, the Dutch referring court (Hoge Raad) sought guidance from the CJEU whether the reprints of paintings on canvas run against Dutch and EU law. The facts of the case indicated that defendant Art & Allposters International BV purchased paper posters on the market that were printed and sold with the permission of the right holder. The defendant allowed for its own costumers to order reprints of the paintings on canvas. The plaintiff – the competent Dutch collective rights management association (Stichting Pictoright) – claimed payment for each copy sold on the ground that the original paper version of the painting was adapted to canvas format. The defendant argued that the right of distribution was exhausted when the original paper copies of the paintings were put on the market, and consequently Art & Allposters is allowed to reproduce the work irrespectively of the form of the new copies.  [392] Advocate General Pedro Cruz Villalón refused both arguments. The reference to the adaptation right is clearly flawless: the reproduction of the painting on canvas does clearly lack the creation of any derivative work.  [393] On the other hand, irrespectively of the fact that Art & Allposters purchased lawfully sold hard copies on the market, it did not acquire the rights from the original distributors.  [394] Consequently, all the reproductions made by the company are out of the scope of the original permission, and these copies cannot be covered by the doctrine of exhaustion either, since these new copies were not put on the market by or under the authorization of the original right holder.  [395] The fact that none of the parties viewed the case through these lenses evidence that the doctrinal frames of the principle of exhaustion are far from settled or clear under the European Union law.

Taking all the above arguments into account the present article takes the view that talking about the doctrine of digital exhaustion is clearly not hype. It reflects socially and economically significant questions. The future of access to protected subject matter under service-like models is not necessarily the only viable option for users to enjoy contents, and the aim to possess copies on a permanent basis, and ultimately the interest in the alienation of them remains a lively issue. Further there is a great chance that the frames of the doctrine of exhaustion might be stretched.

As the paper argued above the ultimate barrier to the acceptance of the notion of digital exhaustion is incorporated into the Agreed Statement of the WCT. One important question was not, however, answered in details by the case law introduced above. The Agreed Statement was implemented by the InfoSoc Directive in a partially different form, where a new doctrinal element was added to the definition of exhaustion. That issue is related to the services versus goods dichotomy.

Gaubiac noted it in the early 2000’s that „the dematerialization of works now taking place involves only the mode of public communication. But this dematerialization has the effect of transferring a mode of communication by material media to a mode of communication by means of computer storage. This leads to some doubt as to whether such a communication can be properly described as a good or service. The applicable system governing the exhaustion of rights will depend on how the medium of communication of the work is classed. If the dematerialized transmission is classed as reproduction in the various computer memories, the question arises of the survival of exhaustion, whereas if this phenomenon is classed as a communication taking the form of an online service provision, it shall not involve exhaustion in accordance with the generally recognized solutions.”  [396]

Several leading international and regional copyright norms evade giving a straight answer to the question raised by Gaubiac. The Agreed Statement attached to Article 6 of the WCT stresses that copies of protected works might only be subject to distribution (and consequently exhaustion) if they are fixed and can be put into circulation as tangible objects. Recitals 28-29 of the InfoSoc Directive equally exclude intangible copies, services (especially on-line services) and tangible copies produced with the help of services and on-line services from the scope of the doctrine of exhaustion. The model for InfoSoc Directive’s wording is to be found in the Database Directive that stressed in its Recital 33 that “the question of exhaustion of the right of distribution does not arise in the case of on-line databases, which come within the field of provision of services; whereas this also applies with regard to a material copy of such a database made by the user of such a service with the consent of the right holder; whereas, unlike CD-ROM or CD-I, where the intellectual property is incorporated in a material medium, namely an item of goods, every on-line service is in fact an act which will have to be subject to authorization where the copyright so provides”.  [397] Similarly, the E-Commerce Directive categorized on-line sale of goods as services.  [398]

The EU rules on value-added tax (VAT) similarly support the above treatment. The VAT Directive declares supply of services as “any transaction which does not constitute a supply of goods”,  [399] whereas supply of goods means “the transfer of the right to dispose of tangible property as owner”.  [400] Based upon the above terms the CJEU recently concluded that “the supply of electronic books cannot be regarded as a ‘supply of goods’ within the meaning of that provision, since an electronic book cannot qualify as tangible property”.  [401] Further, Implementing Regulation No. 282/2011 notes that electronically supplied services “shall include services which are delivered over the Internet or an electronic network and the nature of which renders their supply essentially automated and involving minimal human intervention, and impossible to ensure in the absence of information technology”.  [402] The CJEU concluded in the above decision that “the supply of electronic books clearly meets that definition”.  [403]

Such a separation of tangibles, intangibles and services (on-line services) with regard to the doctrine of exhaustion was treated justifiable by Espantaleon due to the distinct physical attributes and scope (both in terms of space and time) of goods and services, further the diverse language, pricing and technical barriers applied in connection with them.  [404] Others criticized the uncertainty mirrored by the InfoSoc Directive’s wording. Advocate General Bot noted in its Opinion to the UsedSoft case that “Recital 29 in the preamble to Directive 2001/29 is not without ambiguity either. While it appears to draw a distinction between the sale of goods, to which the exhaustion rule would apply, and the provision of services, to which that rule would be inapplicable, the fact remains that online services, as defined by EU law, include the sale of goods online. Thus, for example, by the standard of the wording of that recital, the exhaustion rule should not apply to an online purchase of a CD‑ROM in which the copy of the computer program is incorporated. To my mind, however, the distinction as to whether the sale takes place remotely or otherwise is irrelevant for the purposes of applying that rule.”  [405]

Wiebe points out that the doctrine of exhaustion and the goods versus services dichotomy have different purposes in law.  [406] He claims that the emphasis in respect of goods and services was misplaced in the EU copyright law. Under the WCT the exclusion of copies sold online from the scope of the doctrine of exhaustion is reasoned by the fact that no physical copy is provided by the seller to the purchaser. The EU legislature unnecessarily complicated the discussion of exhaustion with the introduction of the provisions on goods versus services.  [407] The European Commission stressed it in 1995 that “[w]hether a distribution right is capable of being exhausted by an exploiting act of the right holder, or a third party with the right holder’s consent, depends upon the form in which the protected work or related matter is exploited. If it is incorporated in a material form it is subject to the rules on free movement of goods and, in consequence, to the principle of Community exhaustion. (...) On the other hand, if the work or related matter is not incorporated in a material form but is used in the provision of services, the situation is entirely different. (...) In fact, given that the provision of services can in principle be repeated an unlimited number of times, the exhaustion rule cannot apply.” Wiebe argued, however, that “the assumption that online transmissions always involve a service is flawed”.  [408] Consequently, Wiebe’s view is that it is not the goods versus service dichotomy that leads to the exclusion of the doctrine of exhaustion in cases of online delivery of goods, where the purchasers are granted a permanent control over the copy of a work in digital format, but rather the fact that the seller is not obtaining control over a physical/tangible copy of the said work.  [409]

The final conclusion mirrored by Wiebe’s paper – that is ultimately in accordance with the literal reading of the WCT – has been criticized by other commentators. Spedicato noted that “the dichotomy that we should match to the one between goods and services in applying the principle of exhaustion is not that between tangible and intangible objects, as Recital 29 of the InfoSoc Directive would suggest, but rather that between selling a work and making at available (through an act other than a sale) or, on a more general level, that between forms of exploitation that entail a transfer of ownership of a copy of the work and forms that do not”.  [410] Spedicato argues that the UsedSoft ruling on the doctrine of sale shall be applicable to other subject matter than software as well, and consequently the sale of protected subject matter via the Internet shall not be per se excluded from the scope of the doctrine of exhaustion.  [411]

Tai noted that two U.S. reports from 2001 commissioned by the U.S. Department of Commerce, National Telecommunications and Information Administration  [412] and the U.S. Copyright Office  [413] both accepted a “limited digital first sale doctrine” and treated this concept as fully in accordance with the WCT’s Agreed Statement cited above. According to the NTIA Report “both the proponents and the copyright community seem to agree that if the files are downloaded with the consent of the copyright owner, a ‘lawfully made copy or phonorecord’ will have been created on the PC hard drive or tangible portable medium (such as a writeable CD). Thus, Section 109 would apply to the owner of that new digital copy or phonorecord. With respect to other applications of Section 109 to digitally downloaded files, however, there was considerable divergence between the stakeholders with respect to whether a copyright owner's interest could be adequately protected. There was significant information in the record to suggest, however, that DRM systems and other like developments hold some promise of offering a technological solution.”  [414]

The problematic of goods versus services viewed from the perspective of the doctrine of exhaustion seems to be a stalemate. Dreier correctly pointed out that “in times of offering copyrighted works in digital form, it is of secondary importance whether the offering is conducted offline or online. Hence, the distinction between ‘goods’ on the one hand and ’services’ on the other loses much if not all of its meaning. At best it is no longer technology-neutral. Moreover, due to increasing bandwidth, in the future most copyrighted works will be transmitted online anyway. If this is the case, then either all offerings of copyrighted material online will have to be considered as services. Or, if the distinction between freedom of movement or goods and services is to be maintained, the criteria for distinguishing between the two – and with it between ‘goods’ and ‘services’ – are to be found elsewhere. However, in this respect, the decisions handed down by the CJEU so far have not yet provided much guidance. What is needed is an appropriate definition of these meta-criteria, i.e. a coherent theory of when to treat online offerings of copyrighted works as ‘services’ and when to still treat them as ‘goods’, in spite of their intangible and immaterial nature.”  [415]

The present paper takes the view that the status quo related to the doctrine of exhaustion – especially in accordance with the Agreed statement of the WCT and Recitals 28-29 of the InfoSoc Directive – is outdated and do not properly reflect the economic, social and technological realities of our age. Therefore, in answering the question whether there is any need to reconsider the scope of the doctrine of exhaustion in order to cover digital transactions as well, economic, social and technological realities deserve priority.

Some commentators opined that the negative economic effects of digital exhaustion are tremendous. The most general claim – that might varies in form, but remains the same in its content – is that any market for digital resales necessarily decreases the need for “originals”, and consequently harms the interests of the right holders and the traditional intermediaries.  [416] A traditional counterargument is based on the vital premise of the doctrine of exhaustion that allows right holders to be remunerated once after the first sale of their creations. It has been stressed by the proponents of the German “Belohnungstheorie” as well. To put it differently: right holders are not allowed to control the future sales of their contents, if she asked for a fair price at the time of the first sale.  [417] WCT similarly makes it clear that the three-step test – especially the prong that relates to the economic effects of any limitation upon the economic rights – shall not cover resales covered by the doctrine of exhaustion. As such the form of the sale (digital or analogue) is not decisive; indeed, should the form be decisive, the right holders would be able to exclude others from the downstream market, and would receive an unfair advantage in this sphere.  [418] Other commentators further argued that the resale of a digital good is generally useful for the whole economy, since it leads to reinvestment into the system.  [419]

The other clear danger of the introduction of digital exhaustion is related to the rise of the prices of original copies. It is naturally unclear what “fair price” is going to mean so long as there is no second hand market in respect of any specific protected subject matter. The downstream commerce has to be, however, always “cheaper” than the original market. This can be extremely useful for the purchasers. Spedicato correctly noted that “the secondary markets will generally make lower prices available, thus making the work accessible to consumers with less purchasing power”.  [420] On the contrary, the emergence of second hand markets might require the increase of the prices of the original copies due to the loss of sales by the right holders.  [421] The amount of such increase might be marginal, where the number of items sold is extremely high. This shall be especially true for the music industry, and might be relevant for the audiovisual sector and e-book industry as well. For computer programs, databases and other subject matters (like photographs) there might be a clear danger that the increase of prices will be substantive that can have chilling effects upon the business of right holders.

The negative effects of the downstream commerce might be eased by the model that ReDigi or Tom Kabinet offered (and those models that Apple and Amazon patented, but did not launch yet  [422] ), where right holders are remunerated for each and every digital item sold by the clients of the service.  [423] Such a model might look, however, paradoxical. If we accept the view that exhaustion applies for the resale of digital goods as well, than any payment to the right holder is per se unnecessary. If, on the contrary, the service provider obliges itself to pay remuneration to the right holders it practically confirms the valid claims of the other party to the said money.

Thirdly, the fact that the downstream commerce is offered by new service providers (ReDigi, Tom Kabinet, UsedSoft and others) can clearly lead to the rearrangement of powers of the market. It is not surprising that world-wide leading corporation like Amazon and Apple replied to the new challenge. (They most probably wait for the final outcome of the ReDigi and Tom Kabinet cases, before entering the digital resales market with their systems.) A similar reaction might be mirrored by the introduction of Kindle Unlimited by Amazon.  [424] On the one hand, it is offered as a service, and therefore the first sale doctrine does not apply to the copies obtained (licensed) by the users. On the other hand, the pricing of the model is so low that it practically kills the second hand market for “used” e-books.  [425]

Does history repeat itself? It might be worth to recall what George Orwell envisioned in respect of the introduction of the “cheap” Penguin books in the first half the 20^th century. He opined that “it is, of course, a great mistake to imagine that cheap books are good for the book trade. Actually it is just the other way about. (…) Hence the cheaper books become, the less money is spent on books. This is an advantage from the reader’s point of view and doesn’t hurt trade as a whole, but for the publisher, the compositor, the author, and the bookseller it is a disaster.”  [426] Orwell’s vision might be theoretically correct, however, the history of the book industry has evidenced that publishers, editors, authors and bookseller (intermediaries) can and do survive, even if the price of items partially decreased. It seems to be more – rather than less – probable that the introduction of any digital exhaustion doctrine will not kill the traditional forms of copyright industry. Rather businesses will respond to it with new business models. All in all, such an outcome will be beneficial for members of the society. Indeed, access to culture is also vital for institutions, such as universities, libraries, archives or museums, but for individuals as well.  [427]

A classic argument against the introduction of any flexible limitation or exception in favour of the users regarding the use of digital copies of protected subject matter roots in the zero cost of reproduction of these contents.  [428] Any such claim is implicitly based on the assumption that members of the society are willing to copy protected contents for free. This opinion is correct on the one hand; and flawed on the other hand. Practically, copying a digital content takes extremely short time, marginal human and mechanic power is required, and costs virtually nothing. Indeed, humans have a deep-rooted willingness to possess as much as possible. To put it differently, there is nothing surprising, if users save the works onto their computers, if they can. Such a desire is present irrespectively of the form of the protected content. If the source material is analogue (like a paper book) or is bounded to any physical data carrier (like a sound recording on a CD), users are eager to transform (rip) the content into a digital file, if they cannot directly locate it over the internet. Dutch researchers stressed that “skilful consumers mastering information and communication technology have combined with the development of network capacity to increasingly squeeze the entertainment industry’s traditional business model. Digital consumers, wise to technological possibilities and new applications in the digital arena, are now making demands of products and services – demands that the entertainment industry, stuck in its traditional practices, has failed to meet sufficiently over the past few years.”  [429] As a consequence, the difference between the reproduction of an analogue and a digital content is clearly vanished with the help of digitization technologies (that has equally marginal costs, however, might last for a couple of minutes). Ultimately, there is no greater danger with respect to the reproduction of digital goods than the analogue ones.  [430]

It is reasonable to support the development of “copyright literacy” through the acceptance of digital exhaustion. In the latter case users would not simply sell their unused digital items (some might argue: their property), but would contribute to the development of a secondary market and reinvest in culture at the same time. Copyright law would be treated as a useful tool to support this system, rather than an obstacle that hinders downstream commerce, culture and personal property. The lack of workable alternative solutions to consume protected contents via the internet has already led to a tremendous gap between society and right holders. Peer-to-peer (P2P) file-sharing has appeared in 1999 partially due to the lack of lawful alternatives to consume digital goods online.  [431] The introduction of the doctrine of digital exhaustion might direct users’ attention to lawful digital retail stores.  [432]

A further argument might run against the doctrine of digital exhaustion. Although this paper claimed that the dangers of unauthorized reproduction is practically equal with respect to analogue and digital goods, it still accepts the fact that copies connected to a physical data carrier are subject to significantly faster deterioration than those that are saved on a hard drive or nowadays more often in the cloud. The indestructible nature of digital versions allow for an unlimited resale of the said copy. (For the sake of clarity, the paper focuses at this point on the sale of the copy in a single chain by multiple successive purchasers, rather than the clearly illegal method of multiple resales by one single user to more purchasers.) This shall not be, however, treated as an obstacle to the application of a concept of digital exhaustion. As highlighted several times above, the first sale doctrine controls each and every future transfers of a given content after the first lawful sale. No limitation on the number of these transactions is built into the doctrine. If right holders fear of unlimited future disposals of the given copy, they shall adopt proper pricing for the sale of their contents that meets the realities. If this opinion is correct, no such technological solutions are needed to control the “automatic aging” of digital files.  [433]

The present paper takes the view that the effective functioning of a digital first sale doctrine might be guaranteed with the help of two technological solutions. The first shall be the inclusion of effective forward-and-delete software into the model of the digital retail store. The other is the application of unique ID number (watermarking) for each and every contents sold.

Under this model users might only resell contents that were lawfully purchased from authorized corporations that apply specific ID-numbers or watermarking for these files.  [434] First, such a prerequisite excludes the possibility to swamp the system with copies acquired through P2P file-sharing sites (or any other illegal channels) or that were created by the users themselves from CDs or any other data carrier. Second, the ID-number or watermark shall work as rights management information that deserves protection under WCT Article 12. To put it differently, any attempt to erase or modify the ID-number or watermark ultimately leads to a copyright infringement.

Should the lawful acquirers sell their copies via the retail store’s system, the latter shall effectively control the removal of the original file from the user’s hard drive. If the user previously saved a copy of the said content to any portable device, the forward-and-delete software shall detect it at any time when the device is synchronized with the computer, and shall oblige the user to remove the said content from the device. All of these guarantee that there may always be only one copy of a lawfully purchased content having a unique ID-number or watermark.  [435] Berger stressed, and the present paper agrees with him, that a possible statutory description of the digital exhaustion doctrine shall explicitly refer to the obligation to erase the sold content from the users’ computer and that the doctrine only protects the lawful acquirers of digital files.  [436] Ultimately, such a model does not heavily or unnecessarily intrude into the private lives of the users. The control of the synchronized devices is absolutely reasonable in order to protect the interests of the right holders.

Taking all of the economic, social and technological arguments expressed above into account, there might be a valid claim to argue for the introduction of a digital first sale doctrine. Some commentators claim that the Agreed Statement of the WCT is cloudy in this respect. Ruffler stressed that “what [the Agreed Statement] actually says is that the copies can be put into circulation as tangible objects. That is to say that it must be possible to fix them in a tangible medium, and not that they must already be fixed as tangible objects”.  [437] If Ruffler is correct, it is only InfoSoc Directive Article 4(2) that needs to be clarified to cover the digital domain as well. If we, however, accept the more positivistic point of view, it becomes evident that it is the Agreed Statement that requires amendment first. The present paper votes for this second option.