Document Actions

No section

Authors’ rights vs. textual scholarship: a Portuguese overview

  1. Elsa Pereira


This article addresses the main restrictions that European textual and genetic scholars face when the literary works they study are not in the public domain. Using Portugal as an example, the essay illustrates the most relevant contours of copyright policy and licensing in countries with a legal tradition of Droit d’Auteur, which protects not only intellectual property but also the sensitive moral interests of authors. While considering a few limitations and exceptions for teaching and scientific research secured in the law, the paper refers to case studies that showcase legal shortcomings in balancing authors’ rights with the academic freedom of textual scholars, especially when digital editorial methodologies are involved. We argue that the dominant protection currently afforded to copyright holders in Europe undermines the research ecosystem of this disciplinary field, rendering knowledge production and scientific publication practically unfeasible for anyone investigating textual variance in the works of 20th- and 21st-century writers. After drawing attention to the problem, we plead for policy-making adjustments to allow greater freedom in using copyrighted works for humanities research and scholarship.


1. Introduction*


The primary goal of textual scholarship is to investigate how texts of the cultural heritage develop and change over time, whether due to authorial revision or corrupted transmission.  [1] For that purpose, scholars must examine the extant source documents of a written work and compare their materialised textual versions. This “interpretive criticism of variant readings”  [2] relies on a “range of discourses available to literary criticism”,  [3] such as textual and genetic criticism, which “does not have to” but usually involves “a form of scholarly editing”  [4] to display textual “variance in context”, using “a multilevel system of apparatus” with “a key function for interpretative discourse”.  [5]


Like other fields handling products of the human mind, textual scholarship is thus hampered by legal uncertainty, which the digital turn in humanities research has further accentuated due to cross-border dissemination and other intrinsic matters of the online environment. As Walter Scholger has already observed, “most humanists (or scholars in general, regardless of their respective domains) are not familiar with the legal implications of their work”, nor are they sufficiently conversant with the regulations that apply to their activity. “Unfortunately, there is also little to no support from universities’ legal offices”,  [6] which exposes researchers to potentially troubling consequences.


This article reviews the legal contours specifically affecting European textual and genetic scholars when the studied literary works are not in the public domain. While copyright laws grant authors exclusive rights of exploitation, third parties may use protected works by seeking authorisations from rightsholders and contracts providing for transfer or assignment. The law also includes exceptions and limitations to authors’ rights that allow specific unauthorised uses. As will be demonstrated, neither system (authorisations & assignments nor limitations & exceptions) functions properly with textual research and scholarship.


To illustrate the spectrum of complications arising in European jurisdictions, we will take the Portuguese legislation for reference and consider the main restrictions imposed on textual scholarship. The paper will refer to case studies in Portugal and other EU Member States, which showcase legal insufficiencies in balancing authors’ rights with some exceptions for teaching and scientific research and conclude that, regardless of recent provisions introduced by EU Directive 2019/790 on copyright and related rights in the Digital Single Market, the dominant protection afforded to rightsholders undermines the research ecosystem of this disciplinary field. After drawing attention to the problem, we ask for policy-making adjustments to allow greater freedom in using copyrighted works for humanities research and scholarship.

2. Authors’ rights in Portugal


Like other European countries, Portugal must abide by international treaties (such as the Berne Convention) and EU Directives for the harmonisation of copyright in the Member States, transposing those treaties and implementing these directives into national civil law.


Accordingly, the Code of Copyright and Related Rights  [7] protects all “intellectual works in the literary, scientific, and artistic fields, whatever their type, form of expression, merits, mode of communication, or objective”.  [8] The Portuguese legislator opted for presenting a non-exhaustive enumeration of protected works, introduced by the adverb “namely”,  [9] which in practice means that any written text (including those dictated by practical purposes, such as correspondence) is covered by copyright, as long as it constitutes an original intellectual creation. We should note, however, that although the list ambiguously mentions “books, pamphlets, magazines, and newspapers” among the examples,  [10] the object of legal protection should be the intangible content of the text, not its material record (either manuscript, print, or electronic):

“Copyright protects creations of the spirit, such as poems and paintings, which can be documented on a material record, such as books and canvases. However, any of these material records are distinct from the work. It is necessary to distinguish the work from the respective mechanical record or corpus mechanicum. […] We should also note that ownership of the material record does not confer any rights over the work. Therefore, whoever acquires a book or canvas has no copyright over the works contained in these materials”.  [11]


In line with the French tradition of Droit d’Auteur,  [12] such legal protection translates into two main types of provisions: “economic” as well as “personal rights”, also “termed moral rights”.  [13] We will examine each separately from a textual scholarly perspective to assess their constraining implications for our research field (parts I and II) before considering some exceptions and limitations to authors’ rights, which the law provides for allowing specific unauthorised uses of protected works (part III).

2.1. Economic rights


The Portuguese code generally defines economic rights as the “exclusive right” that authors have “to disclose, publish and exploit economically” their work “in any direct or indirect form”.  [14] According to the European Directive 2001/29/CE on the harmonisation of certain aspects of copyright in the information society, this should translate into three exclusive rights granted to authors and their representatives: the “exclusive right to authorise or prohibit direct or indirect, temporary or permanent reproduction by any means and in any form, in whole or in part” of the protected work;  [15] “the exclusive right to authorise or prohibit any communication to the public of their works, by wire or wireless means”;  [16] and “the exclusive right to authorise or prohibit any form of distribution to the public by sale or otherwise”.  [17]


Although the Berne Convention states that “the enjoyment and the exercise” of economic rights are not “subject to any formality”,  [18] until 1918 the property of literary or artistic works in Portugal was traditionally dependent on formal registration,  [19] and nowadays there is still an optional form managed by the General Inspection of Cultural Activities, which aims only to publicise the authorial property, upon requirement, but is not a pre-condition to holding copyright.


The duration of this protection has been changing throughout the years in Portugal. Since the 1990s, “in the absence of any special provision”, the author’s economic rights “lapse 70 years after the death of the creator of the work”.  [20] During that period, “the original owner of the copyright, as well as his successors […] may: (a) authorise the use of the work by a third party; (b) transfer or assign all or part of the economic content of the work’s copyright”,  [21] either in “temporary”  [22] or “permanent” terms.  [23] Rightsholders may, for instance, offer copyright as debt security  [24] or sell the exclusive right to reproduce an author’s works to a particular publishing house.


While transfer and assignment “may only be effected by public deed”  [25] and “contracts” “witnessed by a notary”,  [26] the Portuguese Code of Copyright holds that authorisations “to disclose, publish, use or exploit a work”  [27] should consist of a more straightforward procedure. They must only “be granted in writing”  [28] and “show specifically the authorised form of disclosure, publication and use, as well as the relevant conditions governing duration, place and remuneration”.  [29] Although the law states that authorisations and assignments are subject to remuneration, rightsholders may also waive any revenue in principle.  [30]


The daughters of Portuguese writer José Cardoso Pires (1925-1998), for example, are willing to waive any compensation after their mother’s death because they consider that “the best thing that can happen to an author’s work is being accessible in order to avoid oblivion”.  [31] They also think that heirs should not interfere with what scholars do about a literary legacy because “an author’s work does not belong to his descendants”, and an heir should not be “a writer’s zombie”.  [32] As we will see, this selfless attitude of Ana and Rita Cardoso Pires contrasts with some authors’ successors, who might be inclined to “make money from the shoes of the deceased” and “carve some visibility for themselves”,  [33] making the authorisation process held by art 41 of the Portuguese Code of Copyright highly random and uncertain.


To illustrate the depth of the problem, we must briefly refer to a few recent cases involving individual rightsholders and the Portuguese Society of Authors (SPA) – a limited liability cooperative established in 1925, which currently manages the rights of more than 20,000 affiliated authors.  [34]


In 2013, the author of this article obtained written authorisation from SPA to undertake a genetic-critical edition of poetry by Pedro Homem de Mello (1904-1984) within the scope of a post-doctoral project at the Center of Linguistics of the University of Lisbon. After approval by the Portuguese Foundation for Science and Technology, the scholar worked on this project in close collaboration with the author’s heirs, who approved and encouraged the project, giving her unlimited access to the poet’s manuscripts under their control. In 2019, however, after years of extensive research and project investment (both by funding institutions and by the researcher’s private expenditure), the heirs, represented by SPA, informed the scholar that the oeuvre of Pedro Homem de Mello was no longer available for the intended use. It turned out that they had sold Assírio & Alvim a commercial edition of the author’s complete works, led by a different editor chosen by the publishing house. Since the authorisation granted by SPA in 2013 did not mention the conditions of time and place for the authorised scholarly edition – as stipulated in art 41 (3) of the Portuguese Code of Copyright – or the term for paying the copyright fee, the heirs decided to act as if their personal commitment with the researcher never existed while SPA’s legal office repeatedly ignored a lawyer engaged by the University of Lisbon in finding room for negotiation. As a result, the scholarly edition had to be abandoned and the publication of essays avoided from then on, since SPA also ignored subsequent authorisation requests for using transcriptions of poems in articles, leaving the whole project paralysed in “a kind of perverse self-denial – perverse because not warranted by the porous nature of copyrights”.  [35]


No less disturbing is the case exposed recently by Federico Bertolazzi in the newspaper Nascer do Sol.  [36] For more than three years, this Italian researcher worked on a scholarly edition of the scattered non-fiction prose (essays, interviews, testimonies) of Portuguese poet Sophia de Mello Breyner Andresen (1919-2004). Supported by a sabbatical leave from the University of Rome Tor Vergata and a scholarship from Instituto Camões, he undertook extensive research to compile the texts that Sophia published in periodicals of the time. During that process, the author’s heirs authorised the scholar to access Sophia’s manuscripts at the Portuguese National Library, since these works, despite being deposited in an institution funded by public resources, can only be perused with “prior and personalised authorisations [from the heirs] depending on the researcher”.  [37] However, when the edition was finally concluded and proposed for publication, the heirs refused their authorisation, justifying the decision with the need to restrict publication to what they arguably consider the author’s best works. Eventually, the scholar had to convert his publication into a mere inventory of bibliographic descriptions without the texts,  [38] resorting to the same “art of designing around copyrights”  [39] followed by other European scholars to avoid “fiercely vigilant and obstructive”  [40] authors’ descendants.  [41]


Alarming as they are, reports of heirs and estates using copyright to hinder scholarship are not unique to Portugal, and the problem seems particularly accentuated in recent years with the digital turn in humanities research. James Joyce’s Estate, for instance, is renowned for refusing permission and making “difficult or impossible, numerous scholarly and creative projects – notably, an electronic multimedia version of Ulysses that an academic had spent years developing”.  [42] As a result, digital humanities research on contemporary authors is strongly discouraged:

“In the case of contemporary literary works, this inevitably means that the decision who to give the rights […] remains in the hands of the authors and the executors of their estates, until the day those rights expire and the materials enter the public domain. Should these limitations stop us from building DSEs [Digital Scholarly Editions] around their works? For Robinson, the answer seems to be: yes. […] Hans Walter Gabler’s unsuccessful attempts to win the favour of the Joyce estate […] rendered him unable to continue his work on a digital edition of Joyce’s Ulysses […]. ‘Events like these decided us’, Robinson explains, ‘We would never work on materials where someone else could, by fiat, render all our work worthless just by refusing publication permission’”.  [43]


Although “[t]raditional critical editing, defined by the paper and print limitations of the codex format, is now considered by many to be inadequate for the expression and interpretation of complex, multi-layered or multi-text works of the human imagination” – to such an extent “that in future all [scholarly] editions should be produced in digital and/or online form”  [44] – the truth is that digital editorial approaches are generally regarded with suspicion by heirs, legislators, and judges.  [45] As Valentina Moscon pointed out, the InfoSoc Directive itself “is based on the general assumption that, particularly in the online environment, right holders need effective and rigorous control over widespread forms of mass usage”  [46] due to the cross-border nature of the online environment, transcending the limits of national jurisdictions:

“Article 7(8) of the Berne Convention states that ‘the duration shall be governed by the law of the country in which protection is claimed’; however […] [t]he digital world is global. Cyberspace is a concept that goes beyond national borders. According to the US Supreme Court, […] cyberspace […] does not have a precise geographic location and any Internet user, anywhere in the world, can access it.

Since it is not easy to identify the territory from which the transmissions originate and where the contents are disseminated, the resolution of questions regarding the determination of the law applicable to the cross-border transmission of works in digital format, as well as the competent court, is compromised”.  [47]


This transnational inherent condition of the World Wide Web has been forcing digital scholarly projects to find ways of limiting access to online content in order to avoid liability for copyright infringement. The Association for Research and Access to Historical Texts and the Huygens Institute for the History of the Netherlands, for instance, opted to publish their 2021 scholarly edition of Anne Frank’s manuscripts (1929-1945)  [48] under geoblocking restrictions that allowed access only from IP addresses located in countries where Frank’s diaries were in the public domain – even though the restrictive measure was not enough to prevent the author’s heirs from taking legal action shortly after the online publication appeared:

“Because the copyright to a number of Anne Frank’s texts has not yet expired in the Netherlands, part of the research, such as the transcriptions of Anne Frank’s diaries, took place in Belgium. The online scholarly edition is only accessible in those countries where the copyright law on Anne Frank’s Texts so admits. In Belgium, Germany, the Netherlands Antilles and other countries, some 60 in all, this edition is available to everyone online at Through geo blocking the availability is limited to those countries. In the Netherlands and a number of other countries the online scholarly edition is not accessible due to copyright regulations. An English translation of this edition will be made available later in those countries where copyright law so permits”.  [49]


More cautious was the approach of the Beckett Digital Manuscript Project, developed by the Centre for Manuscript Genetics at the University of Antwerp, the Beckett International Foundation at the University of Reading, the Oxford Centre for Textual Editing and Theory at the University of Oxford, and the Harry Ransom Humanities Research Center at the University of Texas at Austin, with the permission of the Estate of Samuel Beckett and the support of a grant from the European Research Council.  [50] In this latter case, the research team was able to negotiate with the author’s heirs successfully and found a compromise between the parties by limiting access to the digital scholarly edition through a paywall, whose revenue reverts to the Beckett Estate:  [51]

“Thankfully, in the case of the BDMP, all parties involved have realised that the future of scholarly editing is digital, and that the scholarly augmentation of Beckett’s legacy will only increase the interest in his works – academic or otherwise. For this reason, the Beckett Estate agreed to give the directors of the BDMP the license to publish their genetic Editions of Beckett’s manuscripts, as long as this happens behind a pay-wall. […] each of the collaborating institutions are granted institutional access to the edition”.  [52]


Still, these national and international examples draw attention to the risks textual scholars face when working on 20th and 21st-century authors. Only through institutional protocols and formal contracts signed beforehand by all the parties involved (universities, holding libraries, authors’ heirs, and publishers) may a project in this research field be viable, and even when that is the case, there is no guarantee that unexpected complications will not arise.


Although negotiated and contracted by public authorities at the highest level, the critical-genetic edition of the works by Fernando Pessoa (1888-1935), for instance, was confronted with unexpected copyright issues that could end up jeopardising the entire project.  [53] As the Portuguese law at the time stipulated that copyright was in force for 50 years after an author’s death, Pessoa’s works entered the public domain for the first time in 1985. Therefore, in 1988, the Government entrusted a team of researchers led by Ivo Castro with creating a critical edition of this author’s works to be published by the national printing house. Five years later, however, the European Council Directive 93/98/EEC, sketched under the internal market provisions of the Rome Treaty, required all the member countries “to enact legislation extending copyright terms retroactively to seventy years post mortem auctoris. This meant that works which had been enjoying public-domain status […] were abruptly pulled back into copyright”.  [54] Thus armed with newly extended provisions, Pessoa’s heirs decided to transfer the economic content of their copyright to publishers Assírio & Alvim, which did not take long to exercise their property rights, requesting the immediate suspension of the ongoing critical edition. Fortunately, in this case (unlike what happened with the project on Pedro Homem de Mello), Assírio & Alvim and the authors’ heirs were open to negotiation.  [55] The institutional nature of the project, developed under the auspices of the Government, made it possible to reach an understanding that mirrored some mitigation actions taken in other European countries at the time.  [56] Despite the disruption and anxiety caused to everyone, the research team could thus proceed with their work, and the national printing house went on to publish eight volumes of the critical-genetic edition during Pessoa’s copyright repristination, only by paying a reasonable fee to Assírio & Alvim.  [57]


Even so, the cases above are illuminating evidence of the “distorting effects” that “extravagantly long monopolies” granted to author’s heirs and transferees “are having on culture”  [58] in general and textual scholarship in particular. In European jurisdictions, though, such dominant economic protection is further aggravated with additional personal rights granted to authors but exercised by heirs and estates alike. We will now focus on those moral rights described in the Portuguese law that have a major impact on textual scholarship.

2.2. Moral rights


Unlike authors’ economic rights, “personal rights, termed moral rights”,  [59] are “perpetual, inalienable, and imprescriptible”.  [60]


While the guardianship of these moral rights cannot be transmitted,  [61] the law holds that an author’s successors may exercise them upon the writer’s death until the work falls “within the public domain”,  [62] when the State takes over the responsibility.  [63] In practice, this allows heirs and estates to use authors’ moral rights to expand the sphere of power during and even beyond their period of economic exploitation and occasionally “suppress or control scholarship”  [64] in what has already been labelled as “copyright misuse”.  [65] We shall therefore look into the practical implications for textual scholarship.


In broad terms, the Portuguese Code of Copyright defines moral rights as the author’s “right to claim authorship of his work and to ensure its authenticity and integrity by opposing any mutilation, distortion or other modification thereof and, in general, opposing any action that might be prejudicial to his honour and reputation”.  [66] Simply put, this translates into two main provisions granted to rightsholders against the academic interest of textual scholarship: (a) the right to oppose any modifications to the authorial text; (b) the right to oppose the disclosure of unpublished or private writings that heirs may consider harmful to the author’s reputation.


Regarding the first provision, we should note that heirs may oppose all modifications performed by anyone but the author.  [67] Even where using a work “without […] consent is lawful”,  [68] scholarly editors cannot introduce any correction or alteration to a text without formal consent of the author’s successors, which “shall be requested by registered letter with acknowledgement of receipt”.  [69] The only exception is the “modernisation of spelling in accordance with the official rules in force”, provided that it does not constitute an aesthetic option of the author.  [70] This prescription runs against the principles of several schools of textual criticism – from the Italian filologia d’autore to the Anglo-American copytext editing approach – whose “editorial labor by controlled alterations” includes emendation to “critically recognisable” instances of the text,  [71] such as authorial errors  [72] and different types of corruptions introduced in the transmission process “when authors could not or did not read proof”.  [73] Although the spirit of the law behind art 56 (1) of the Code of Copyright is to protect the author against any harm caused to his intellectual creation, it also implies that authors and their heirs have the absolute power to block or derail the activity of a particular textual scholar in favour of another they trust. As often is the case, trusting personal acquaintances to exclusively make decisions on one’s behalf and preventing other scholars from carrying out their critical practice does not necessarily safeguard the authorial best interest in the long run. This assertion is especially true for works covered by art 58 of the Portuguese Code of Copyright.


According to this, “[w]here the author has partially or wholly revised his work and has effected or authorised” an edition carrying the legal expression “ne varietur”, not only is it forbidden to make alterations to that text, but no one, not even “his successors or third parties”, may ever “reproduce any of the previous versions” again.  [74] In this sense, a ne varietur edition, such as the one Maria Alzira Seixo and her team have been allowed to publish for António Lobo Antunes’ complete works since 2003, will indefinitely prohibit any critical or genetic editions of the author’s works, even after they fall into the public domain. Such condition is all the more disturbing if we think that a list of errors has already been identified in the definitive volumes,  [75] but “those typos in the ne varietur edition cannot be amended in reprintings”, since “any alterations introduced in a ne varietur, even with the best of intentions and for correct purposes, are under the purview of the Law and imply judicial action against any of the entities involved: authors, writer, and editor”.  [76] And whereas the editorial team trusted by the Portuguese writer and his family openly admits that “the previous editions […] constitute a valuable element for genetic studies”, they also acknowledge that after the ne varietur is published, “the official use of the author’s previous editions (in teaching, research, translations, citations, and other public purposes) is forbidden and subject to legal action”.  [77] Considering that Lobo Antunes was consulted but did not revise the ne varietur editions himself,  [78] that he apparently delegated many decisions to his daughters,  [79] and that new authorial – hence authentic  [80] – manuscripts may resurface in the future – clarifying obscure passages in the corrupted printed text  [81] or challenging the editorial decisions in the ne varietur printings – we can only agree that art 58 of the Portuguese Code of Copyright mainly protects “the commercial interests of the publishing house”  [82] instead of the “authenticity and integrity”  [83] of the literary work. Indeed, the absolute and definitive legal value of the moral protection granted to rightsholders does not take care of “the necessary elasticity” and “the desirable improvement of the established text”  [84] to safeguard the integrity of the author’s legacy, the ethics of cultural heritage for coming generations, and the independence of present and future scholarly research.


As for the second moral right with implications for textual scholarship, we will note that any person who “discloses or publishes a work not disclosed by its author or not destined to be disclosed or published, even where he presents it as the respective author’s work and whether or not he seeks to obtain economic benefits” shall “be guilty of the offence of illegal exercise of rights”.  [85] According to the Portuguese code, rightsholders have the exclusive “right to decide upon the use of undisclosed and unpublished works”,  [86] and anyone willing to use those texts must follow the authorisation process held in art 41, with its challenging contours. Federico Bertolazzi, for instance, reveals that Sophia de Mello Breyner’s daughter has been disclosing unpublished poems that the author left in manuscripts while prohibiting researchers from doing the same with the materials kept at the Portuguese National Library.  [87] One of the author’s grandsons was even allowed to continue a short story left unfinished by Sophia, which Porto Editora published with aplomb back in 2012.  [88] Although such decisions of the heirs were legally protected by articles 57 (1) and 70 (1) of the Code of Copyright, we should probably ask about the ethical legitimacy of those granted moral rights that authors’ successors have been executing.


To what extent should heirs and estates determine which unpublished works do or do not harm the reputation of the deceased? And to what extent should copyright protect literary drafts that the author did not even properly finish? In 2011, the Lisbon Court of Appeal, analysing a case that discussed whether a sculptor’s model could be protected by copyright, determined that, similarly to a sketch, the artefact in question constituted a mere stage on the path to the final work that embodied the original idea and, therefore, did not deserve protection:  [89]

“By dictating its incompleteness and classifying it as a ‘common thing’, the Court concluded [...] that the production in question was far from the concept of ‘work’ as a creation of the spirit or intellect, or that it lacked originality”.  [90]


For literary works, however, courts never applied the same interpretation, and, in general, authors’ heirs have been especially protective of unpublished texts, claiming “family privacy”  [91] or personal reputation to block the use of such writings as letters and diaries, but also poems, short stories, and novels in draft manuscripts. James Joyce’s Estate is renownedly zealous in this regard, taking legal action against many researchers who publish the author’s papers found in archives and libraries open to the community:

“Nearly all of the documents that the Estate has declared off-limits to publishing scholars — letters by James Joyce and Joyce family members, essays and memoirs by Lucia Joyce and Helen Kastor Fleischman Joyce — are either already published or held in archives and collections that are generally open to the public. So these documents are not ‘private’ in the sense that they are physically or legally inaccessible. We can learn any of their secrets; we just cannot quote our findings in articles and books or on the Internet. We can kiss but not tell. And how is our silence enforced? Through the climate of fear that many copyright holders have cultivated […]. And when copyrights are used as scarecrows to obtain ‘effective control over information,’ […] we are witnessing something that is increasingly being recognised by lawyers and judges as ‘copyright misuse’ — an attempt to extend copyright protection beyond its appropriate sphere”.  [92]


As such, and since obtaining authorisations from both custodians of the material and copyright owners  [93] “may be difficult and sometimes hopeless”,  [94] it seems imprudent to develop any scholarly research on unpublished works before they fall into the public domain, which “[i]n the absence of any special provision” should be “70 years after the death of the creator of the work, even in the case of works disclosed or published posthumously”.  [95]


Still, copyright policy is “blessedly porous”  [96] by incorporating several measures to permit unauthorised uses of protected works and limit copyright control exerted by rightsholders. The period legally in force for controlling works left unpublished by a writer is among those limitations to authors’ rights.

2.3. Limitations and exceptions to authors’ rights


While “the economic rationale for copyright is based on a public policy objective of encouraging creation for the benefit of society”, and the moral rationale protects “the author’s private right to control her expression”,  [97] many, if not all legislatures have fashioned ways to balance authors’ rights with the “public access to creative works […] in situations where the social costs of copyright restrictions outweigh the benefits”:  [98]

“Copyright protection must be broad enough to provide authors adequate incentives to produce and disseminate creative works, but not so broad that an author’s ability to extract monopoly rents for access chills the production and dissemination of, and access to, creative works”.  [99]


As a country with a Civil Law system, Portugal avoids the monopolistic protection of rightsholders by legally providing some limitations to authorial rights (legal licenses and compulsory licenses, where authors and successors cannot prohibit specific uses but are still entitled to financial compensation) and exceptions covering specific unauthorised uses (not subject to compensation).

2.3.1. Limitations


Presently, the Portuguese Code of Copyright incorporates two types of legal (or statutory) licences that allow for specific unauthorised uses of copyrighted material, even though they are subject to fair compensation to authors or their successors. For legal licenses, it is the law itself permitting the use, whereas for compulsory licenses, the authorial consent is to be replaced by a court decision.  [100]


Covered by legal licence situations are “the right to translate or transform the work in any way necessary for its use”  [101] and also the right to disclose works left unpublished by a writer, “where the successors do not use the work within a period of 25 years from the date of the author’s death”.  [102] However, the latter shall not apply “in the case of impossibility or delay in disclosure or publication for serious moral considerations that shall be decided upon by the courts”.  [103] In practice, this caveat leaves much latitude for litigation since, as we have seen, heirs often call on moral grounds, such as authorial reputation, to oppose the disclosure of unpublished works. Besides, significant discrepancies exist across jurisdictions and apply, depending on the country where publication takes place,  [104] which makes it particularly unwise for scholars to rely on this legal license for working with unpublished writings:

“for unpublished works such as manuscripts, letters, diary entries, etc., the waters are particularly muddy. In Canada, for example, copyright of unpublished materials expires 50 years after the calendar year of the author’s death […]. In the USA, this period is extended to the author’s life plus 70 years […]. In the UK, on the other hand, ‘[w]orks that were unpublished at the author’s death and remained so until 1 August 1989 […] are protected by copyright […]’ until the year 2040. And in Australia, copyright can be enforced for 70 years after the unpublished work has been ‘disclosed’, meaning that it will differ from work to work, and on the purview of what it means to legally ‘disclose’ an unpublished work”.  [105]


Additionally, we may count as legal licence the possibility of including “short excerpts or parts of another author’s work in works used for teaching”.  [106] Although the InfoSoc Directive allowed a similar optional provision for both teaching and scientific research,  [107] art 5 of the European CDSM Directive opted to exclude mention of research and scholarship.  [108] Accordingly, the Portuguese legislator excluded research publications from art 75 (2) (i) of the Code of Copyright, inhibiting textual scholars to use excerpts of copyrighted material without seeking permission and paying the required royalties – even though academics are generally not paid for their essays, since the incentive “to publish research results is mostly reputational rather than economic”.  [109]


As for compulsory licences, the Portuguese Code of Copyright holds that “[w]here the owner of the right to re-edit refuses to use his right or to authorise another edition after the work has become out of print, any interested party, including the State”,  [110] may obtain the authorisation through courts, “provided that re-edition of the work is in the public interest and that the refusal was not based on justified moral or material reasons, excluding financial reasons”.  [111] In practice, however, this extreme situation is hardly applicable in litigation with heirs because the attainable moral reason is an elastic notion, and, in general, Court disputes tend to be settled in favour of the author and his representatives.  [112]

2.3.2. Exceptions


In addition to these legal and compulsory licences, art 75 of the Portuguese Code of Copyright lists a series of exceptions regarding the use of copyrighted material, which are considered as fair and which are not subject to authorisation or payment to rightsholders.


Among the exceptions more directly implied in the activity of textual scholars, we shall highlight the possibility of libraries, archives, and educational establishments (including universities) digitising orphan works – that is “copyrighted works whose owners cannot be located”  [113] – and making them available to the public, for preservation.  [114] This provision has just been aligned with the recommendation held in art 6 of the European CDSM Directive for preserving cultural heritage in the Digital Single Market, but, nonetheless, falls far short of other measures implemented in North America to facilitate the use of orphan works to scholarly research.  [115]


Also relevant for textual scholars is the new exception for text and data mining (TDM)  [116] – recently transposed from the European CDSM Directive – since computational literary studies have been considering the publication of derived data or extracted features as a possible solution to navigate copyright restrictions. Besides metadata and ancillary data,  [117] researchers refer to information extracted through TDM, such as classification and clustering of texts (e.g. for authorship attribution and stylometry), extraction of distinctive features, semantic analysis with topic modelling, analysis of polarity with sentiment analysis, character relationships with network analysis, and analysis of relationships between texts (e.g. in text reuse).  [118] However, we should note that only those materials to which scholars have lawful access can be mined, and experiences in countries where TDM exceptions have been in force show that copyright issues will subsist:

“Despite the TDM exception in German copyright law, Text and Data Mining (TDM) with copyrighted texts is still subject to restrictions, including those concerning the storage, publication and follow-up use of the resulting corpora, leaving the full potential of TDM in the Digital Humanities untapped”.  [119]


Finally, we shall point out that art 75 of the Portuguese Code of Copyright also allows for the “inclusion of quotations or summaries from another author’s work, whatever their type or nature, in support of one’s own opinions or for purposes of criticism, discussion or teaching”.  [120] However, citations of protected texts must be fully integrated into a critical argument and their length “shall not be so extensive that they prejudice interest”  [121] or purchase of the original literary work. The Portuguese code does not state the precise extent or the proportion of a work that can be used for citation, but an amendment of the Berne Convention by the Brussels Act (1948) explicitly required that quotations be “short”,  [122] and while a few lines of a novel would be a small percentage of the work overall, the same amount of text may be considered rather substantial in shorter works such as poems. For that reason, some international publishing groups have been requiring in their regulations that any scholarly article quoting poems or song lyrics protected by copyright should be submitted with formal authorisation from rightsholders, regardless of the fair use clause found in most European and Anglo-American copyright laws:

“As a warranty in the Journal Author Publishing Agreement you make with us, you must obtain the necessary written permission to include material in your article that is owned and held in copyright by a third party, including – but not limited to – any proprietary text, illustration, table, or other material, including data, audio, video, film stills, screenshots, musical notation, and any supplemental material. It is the custom and practice in academic publishing that the reproduction of short extracts of text and some other types of material may be permitted on a limited basis for the purposes of criticism and review without securing formal permission, on the basis that: the purpose of quotation or use is objective and evidenced scholarly criticism or review (not merely illustration); a quotation is reproduced accurately, either within quotation marks or as displayed text; full attribution is given. However, a quotation from a song lyric or a poem, whether used as an epigraph or within the text, will always require written permission from a copyright holder. Our publishing agreement with you requires that you must obtain written permission to reproduce any content, especially image content, in your article, when that content is owned and held in copyright by a third party”.  [123]

“Do I need permission to use poems and songs? Yes, permission should always be obtained. Please be aware that some poets will not allow changes to the layout of the poem or allow you to use a small number of lines. Poem fees are normally charged per line. With song lyrics you should be aware that even if you only use one line you may be charged the same price as you would for the complete song. Rightsholders for song lyrics require people intending to reproduce lyrics to apply for permission for each reuse, and a fee may be charged”.  [124]


Moreover, art 5 (3) (d) of the European InfoSoc Directive recommends that “quotations for purposes such as criticism or review” should only be legal “provided that they relate to a work […] which has already been lawfully made available to the public”.  [125] Such a requirement opens the door for litigation by heirs and estates,  [126] forcing literary critics into “carefully designing around any impulse to quote from” unpublished material  [127] or “paraphrasing it nearly out of existence”.  [128] But while this kind of “design-around scholarship”  [129] may be attempted in some literary studies, it is usually not viable in textual and genetic criticism because assessing the “macrogenesis (the genesis of the work in its entirety across multiple versions)” requires collation of “large textual units along the syntagmatic axis with the development along the paradigmatic axis on a macrolevel”.  [130]


For all that, the narrow scope of the exceptions currently provided in national and communitarian laws, in accordance with the so-called three-step test,  [131] results in ineffective counterbalancing of “the monopolistic protection that copyright affords to authors […] against the limitation on public access to creative works”.  [132]

3. Conclusion and outlook


The case studies analysed in this article showcase several shortcomings in balancing authors’ rights with the academic freedom of textual scholars, especially when digital editorial methodologies are involved.


As others have noted before, literary “copyrights are coming to resemble closely guarded patents”, whose usage restrictions constitute “a deadweight loss to society”.  [133] Extremely long economic and moral provisions “premised on a neoclassical theory of copyright”  [134] have been placing “monopoly control in the hands of heirs and transferees who […] become privileged and sometimes arbitrary custodians of culture”.  [135] Such dominant protection afforded to European rightsholders undermines the production of new knowledge in the humanities and the textual scholarship research ecosystem, rendering scientific publication practically unfeasible or reduced to few derived data at best.  [136]


This circumstance compromises “the free exchange and criticism of ideas […] at the core of academic freedom”,  [137] which is “recognised as a fundamental right by several national constitutions and international treaties”.  [138] The examples and scenarios presented above demonstrate that the academic freedom of literary and textual scholars has often been challenged by interferences and obstructions that perversely amount to submitting scholarly research to the agendas of copyright owners. Therefore, in the interest of research and study, we shall join in pleading for policy-making adjustments to ensure more effective limitations to the lengthy copyrights executed by heirs and successors.


Firstly, the provisions in art 59 (3) and art 58 of the Portuguese Code of Copyright need to be revised to adequately safeguard the author’s literary legacy and the community’s cultural heritage. Secondly, we need to extend the scope of the available exceptions in art 75 of the Code of Copyright to allow for scholarly publication in the digital age – or otherwise, a legal license designed with scholarship in mind so that academic researchers may work with published texts and holographic materials in public archive librarie, disclosing research results (in person, on paper, and online) without interference from heirs or successors. Moreover, we also need national or European management systems led by independent copyright boards to facilitate the clearance of orphan works for different uses and reduce the randomness of our current authorisation system.  [139] Until these and further measures are implemented to solve the problem, academic research will remain perilous and risky for anyone investigating textual variance in the works of 20th- and 21st-century writers.

* By Elsa Pereira; University of Lisbon, School of Arts and Humanities, CLUL, Lisbon, Portugal. Email: <>. ORCID: <>. The research leading to this article was supported by Portuguese national funds through FCT – Fundação para a Ciência e a Tecnologia, I.P., within the scope of DL57/2016/CP1443/CT0033 and UIDB/00214/2020.

[1] Barbara Bordalejo, ‘What is textual scholarship?’ (2007) <> accessed 2 November 2022.

[2] Hans Walter Gabler, ‘Textual Criticism’ in M. Groden and M. Kreiswirth (eds), The Johns Hopkins Guide to Literary Theory & Criticism (1st ed, The Johns Hopkins University Press, Baltimore and London 1994) p. 709 <> accessed 7 May 2023.

[3] Hans Walter Gabler, Text Genetics in Literary Modernism and Other Essays (Open Book Publishers, Cambridge 2018), p. 209.

[4] Dirk Van Hulle and Peter Shillingsburg, ‘Orientations to text, revisited’ (2015) 59 Studies in Bibliography 37 <;;toc.depth=1;;brand=default> accessed 2 November 2022.

[5] Gabler, ‘Textual Criticism’, p. 713.

[6] Walter Scholger, ‘Intellectual Property Rights vs. Freedom of Research: Tripping stones in international IPR law’ (Abstracts of DH2014, Lausanne, 2014) <> accessed 2 November 2022.

[7] Código do Direito de Autor e dos Direitos Conexos (CDADC). Law no. 45/85, of 17 September 1985, as amended up to Law-Decree no. 47/2023, of 19 June 2023 <> accessed 2 September 2023. Throughout this article, all quotations are taken from the English translation provided by WIPO – World Intellectual Property Organization: Code of Copyright and Related Rights (as amended up to Law no. 45/85 of 17 September 1985) <> accessed 2 November 2022. However, the numbering of the articles will be updated, according to Law-Decree no. 47/2023.

[8] CDADC, art 2 (1).

[9] CDADC, art 2 (1).

[10] CDADC, art 2 (1) (a).

[11] Patricia Akester, Direito de Autor em Portugal, nos PALOP, na União Europeia e nos Tratados Internacionais (Almedina, Lisboa 2013), p. 68. Translated from the Portuguese by the author of this article.

[12] Whereas countries adhering to the Common Law Copyright system (such as the UK and its former colonies) have been focusing protection on the commercial exploitation of literary works, countries with a Civil Law tradition of Droit d’Auteur (historically rooted in the French laws of 1791 and 1793) tend to see copyright as a “dualistic right” (Akester, Direito de Autor em Portugal, p. 28) or a monistic right with two components (Germany), protecting the economic value of intellectual property as well as the moral interests of authors.

[13] CDADC, art 9 (1).

[14] CDADC, art 67 (1).

[15] Directive 2001/29/EC of the European Parliament and of the Council of 22 May 2001 on the harmonisation of certain aspects of copyright and related rights in the information society (InfoSoc Directive, art 2). Current consolidated version of 06 June 2019 <> accessed 2 November 2022.

[16] InfoSoc Directive, art 3.

[17] InfoSoc Directive, art 4.

[18] Berne Convention for the Protection of Literary and Artistic Works (Paris Act of July 24, 1971, as amended on September 28, 1979), art 5 (2) <> accessed 16 May 2023.

[19] Instituto da Biblioteca Nacional e do Livro – Direcção Geral dos Espectáculos, Direito de Autor em Portugal: Um Percurso Histórico (Biblioteca Nacional: Lisboa 1994), pp. 46-47.

[20] CDADC, art 31.

[21] CDADC, art 40.

[22] CDADC, art 43 (4).

[23] CDADC, art 44.

[24] CDADC, art 46.

[25] CDADC, art 44.

[26] CDADC, art 43 (2).

[27] CDADC, art 41 (1).

[28] CDADC, art 41 (2).

[29] CDADC, art 41 (3).

[30] While the Portuguese Code of Copyright states that authorisations “shall be considered nonexclusive and subject to payment” (CDACD, art 41 [3]), it also says that transfer deeds must show the amount of the remuneration “where payment is involved” (CDADC, art 43 [3]), suggesting that acts covered by chapter V (authorisation and transfer & assignment) are not necessarily subject to remuneration. In this respect, the CDSM Directive, for instance, clarifies that “[n]othing in this Directive should be interpreted as preventing holders of exclusive rights under Union copyright law from authorising the use of their works or other subject matter for free, including through non-exclusive free licences for the benefit of any users” – Directive 2019/790 of the European Parliament and of the Council of 17 April 2019 on copyright and related rights in the Digital Single Market (CDSM Directive, para 82) <> accessed 16 May 2023.

[31] Rui Couceiro, ‘Herdeiros ou zombies dos escritores?’ Visão (Lisboa, 10 January 2022) <> accessed 31 May 2023. Translated from the Portuguese by the author of this article.

[32] Ibid. Translated from the Portuguese by the author of this article.

[33] Ibid. Translated from the Portuguese by the author of this article.

[34] Information stated on the official webpage <> accessed 16 May 2023.

[35] Robert Spoo, ‘Copyrights and “Design-Around” Scholarship’ (2007) 44-3 James Joyce Quarterly 567 <> accessed 2 may 2023. Using full-text transcriptions or even excerpts in research articles is not covered by the exceptions specified in art 75 of the Portuguese Code of Copyright. See infra part 3 of this essay.

[36] Teresa Carvalho, Interview with Federico Bertolazzi, ‘Até que ponto os herdeiros de Sophia podem bloquear uma obra?’ Nascer do Sol (Lisboa, 27 December 2022) <> accessed 2 January 2023. Federico Bertolazzi, ‘Carta de resposta à Professora Maria Andresen’ Nascer do Sol (Lisboa, 8 January 2023) <> accessed 9 January 2023.

[37] Fátima Lopes, ‘Como se trabalha no Arquivo de Cultura Portuguesa Contemporânea’ in Luiz Fagundes Duarte and António Braz de Oliveira (org), As Mãos da Escrita (Biblioteca Nacional de Portugal, Lisboa 2007), pp. 53-54 <> accessed 9 January 2023. Translated from the Portuguese by the author of this article.

[38] Federico Bertolazzi, No Reino Terrível da Pureza: Bibliografia da Prosa Dispersa Não Ficcional de Sophia de Mello Breyner Andresen e Três Ensaios (Documenta, Lisboa 2022).

[39] Spoo, ‘Copyrights and “Design-Around” Scholarship’ 567.

[40] Robert Spoo, ‘Ezra Pound’s Copyright Statute: Perpetual Rights and the Problem of Heirs’ (2009) 56 UCLA Law Review 1825 <> accessed 2 may 2023.

[41] After opposition from the Joyce Estate to most research projects in textual scholarship and scholarly editing, Michael Groden, “noted for his close textual study of the genetic development of Ulysses, has written several articles containing general descriptions and inventories of the recent manuscript discoveries but has scrupulously avoided offering extracts” (Spoo, ‘Copyrights and “Design-Around” Scholarship’ 567).

[42] Spoo, ‘Ezra Pound’s Copyright Statute: Perpetual Rights and the Problem of Heirs’ 1826. See also Spoo, ‘Copyrights and “Design-Around” Scholarship’ 563-585, for mention of other scholarly projects affected by decisions of the Joyce estate.

[43] Wout Dillen and Vincent Neyt, ‘Digital scholarly editing within the boundaries of copyright restrictions’ (2016) 31-4 Digital Scholarship in the Humanities 787 <> accessed 2 January 2017.

[44] Marilyn Deegan and Kathryn Sutherland, ‘Introduction’ in Marilyn Deegan and Kathryn Sutherland (ed), Text Editing: Print and the Digital World (Ashgate, Farnham 2009), p. 1.

[45] “Judges engaging in a fair use analysis more often than not expect scholarship to come packaged in print monographs written in academic language aimed at an audience of disciplinary specialists. When they encounter scholarly artefacts that depart from those formal expectations and draw from pre-existing work, judges are less likely to find the use of pre-existing work is fair and therefore non-infringing” – Robin Wharton, ‘Digital Humanities, Copyright Law, and the Literary’ (2013) 7-1 Digital Humanities Quarterly <> accessed 11 February 2022.

[46] Valentina Moscon, ‘Academic Freedom, Copyright, and Access to Scholarly Works: A Comparative Perspective’ in Roberto Caso and Federica Giovanella (ed.), Balancing Copyright Law in the Digital Age: Comparative Perspectives (Springer, Berlin-Heidelberg 2015), p. 102.

[47] Akester, Direito de Autor em Portugal, pp. 103, 161. Translated from the Portuguese by the author of this article.

[48] Peter de Bruijn, Elli Bleeker and Marielle Scherer (eds), Anne Frank Manuscripten (2021) <> accessed 7 February 2023.

[49] Peter de Bruijn, Elli Bleeker and Marielle Scherer, ‘For the first time all Anne Frank’s manuscripts digitised’ (28-09-2021) <> accessed 7 February 2023.

[50] Dirk Van Hulle and Mark Nixon (ed), Samuel Beckett Digital Manuscript Project (2021) <> accessed 7 February 2023.

[51] Ibid <> accessed 7 February 2023.

[52] Dillen and Neyt, ‘Digital scholarly editing within the boundaries of copyright restrictions’ 789.

[53] Simone Celani, O Espólio Pessoa: Para Uma História das Edições e dos Critérios Adotados (Imprensa Nacional, Lisboa 2020), pp. 40, 42-45.

[54] Spoo, ‘Copyrights and “Design-Around” Scholarship’ 568.

[55] A copy of the intimation, dated 16 May 1997, is kept among the project’s documentation at the Library of the School of Arts and Humanities of the University of Lisbon. In the letter, addressed to the national printing house, Assírio & Alvim and Pessoa’s heirs state that they were willing to discuss any commitment previously assumed, but would not fail to take legal action and protect their interests (xerocopy of a letter from Manuel Hermínio Monteiro and Manuela Nogueira to Imprensa Nacional-Casa da Moeda, ref. 210/DA. Universidade de Lisboa – Faculdade de Letras – Biblioteca, Espólio Equipa Pessoa. Uncatalogued documentation).

[56] In the United Kingdom, for instance, a compulsory-license exception was issued to Danis Rose’s revised edition of Joyce’s Ulysses (Spoo, ‘Copyrights and “Design-Around” Scholarship’ 568-569).

[57] Between 1997 and 2006, the Portuguese national printing house published the following volumes of the scholarly edition: Poemas de Fernando Pessoa – Quadras (1997); Poemas Ingleses II: Poemas de Alexander Search (1997); Poemas Ingleses III: The Mad Fiddler (1999); Poemas de Fernando Pessoa: 1934-1935 (2000); Poemas de Fernando Pessoa: 1921-1930 (2001 ); Obras de António Mora (2002); Poemas de Fernando Pessoa: 1931-1933 (2004); Poemas de Fernando Pessoa: 1915-1920 (2005).

[58] Spoo, ‘Copyrights and “Design-Around” Scholarship’ 568.

[59] CDADC, art 9 (1).

[60] CDADC, art 56 (2).

[61] CDADC, art 42.

[62] CDADC, art 57 (1).

[63] CDADC, art 57 (2).

[64] Spoo, ‘Ezra Pound’s Copyright Statute: Perpetual Rights and the Problem of Heirs’ 1822.

[65] Spoo, ‘Copyrights and “Design-Around” Scholarship’ 575.

[66] CDADC, art 56 (1).

[67] CDADC, art 56 (1).

[68] CDADC, art 59 (1).

[69] CDADC, art 59 (3).

[70] CDADC, art 93.

[71] Gabler, ‘Textual Criticism’, p. 710.

[72] Based on Roncaglia’s and Cunha’s previous considerations on this matter, João Dionísio identifies two main types of authorial errors: errors by execution and conception errors (João Dionísio, Doença Bibliográfica [Imprensa Nacional, Lisboa 2021], p. 118). The former result from momentary or mechanical distractions while writing, while the latter correspond to what Roncaglia called errors of fact or language – i.e. inaccuracies caused by memory lapses, cultural limitations, or the author’s non-compliance with current linguistic norms. Only authorial errors by execution should deserve obvious correction by textual scholars (Ibid 107).

[73] Gabler, ‘Textual Criticism’, p. 710.

[74] CDADC, art 58.

[75] Maria Alzira Seixo, Graça Abreu, Eunice Cabral, Agripina Carriço Vieira, Memória Descritiva: Da Fixação do Texto para a Edição ne varietur da Obra de António Lobo Antunes (D. Quixote, Lisboa 2010), pp. 145-149.

[76] Ibid, p. 168. Translated from the Portuguese by the author of this article.

[77] Ibid, p. 30. Translated from the Portuguese by the author of this article.

[78] Ibid, p. 27.

[79] Ibid, p. 25.

[80] Dionísio, Doença Bibliográfica, p. 26.

[81] Ibid, p. 43.

[82] Ibid, p. 43. Translated from the Portuguese by the author of this article.

[83] CDADC, art 56 (1).

[84] Dionísio, Doença Bibliográfica, p. 43. Translated from the Portuguese by the author of this article.

[85] CDADC, art 195 (2).

[86] CDADC, art 70 (1).

[87] Carvalho, Interview with Federico Bertolazzi, ‘Até que ponto os herdeiros de Sophia podem bloquear uma obra?’. Bertolazzi, ‘Carta de resposta à Professora Maria Andresen’.

[88] ‘Conto inédito de Sophia terminado pelo seu neto’ (02/10/2012) Diário de Notícias <> accessed 4 May 2023.

[89] Judgment of the Court of Appeal of Lisbon [Acórdão do Tribunal da Relação de Lisboa], Process Nr. 323/07.8TVLSB.L1-2, 30-06-2011. Apud Akester, Direito de Autor em Portugal, pp. 64-65.

[90] Akester, Direito de Autor em Portugal, p. 65. Translated from the Portuguese by the author of this article.

[91] Spoo, ‘Copyrights and “Design-Around” Scholarship’ 573.

[92] Ibid 574-575.

[93] As warned by several holding libraries, “[t]he owner of copyright for material in the Manuscripts Collection is the writer or creator of the material, or the creator’s legal heir(s). Note that the donor of the material is not always the copyright owner. In addition, many collections contain a variety of letters, diaries, documents owned by multiple copyright owners. […] Should you wish to publish material from the Library’s Manuscript Collection, you will need to declare your intention to the Library as custodian of the material. You will also need to obtain copyright clearance from the copyright holder(s)” (National Library of Australia, ‘Rights and the Manuscripts Collection’ <> accessed 15 November 2019).

[94] Spoo, ‘Copyrights and “Design-Around” Scholarship’ 579.

[95] CDADC, art 31.

[96] Spoo, ‘Copyrights and “Design-Around” Scholarship’ 579.

[97] Matt Jackson, ‘Copyright’ in Wolfgang Donsbach (ed), The Concise Encyclopedia of Communication (Wiley Backwell, Sussex 2015), p. 115.

[98] Maureen Ryan, ‘Fair Use and Academic Expression: Rhetoric, Reality, and Restriction on Academic Freedom’ (1999) 8-3 Cornell Journal of Law and Public Policy 541-542 <> accessed 8 January 2023.

[99] Ibid 548-549.

[100] Luís Manuel Teles de Menezes Leitão, Direito de Autor (2 ed, Almedina, Lisboa 2018), p. 172.

[101] CDADC, art 71.

[102] CDADC, art 70 (3).

[103] CDADC, art 70 (3).

[104] Authors’ rights are protected by the law of the country where the use of a work occurs. Therefore, the law of the country where publication takes place should apply. In academic publications, however, a journal, based at the university of one country, is often published by an international editorial group from another country. Besides, digital publication is especially problematic due to the cross-border nature of the online environment transcending the limits of national jurisdictions.

[105] Dillen and Neyt, ‘Digital scholarly editing within the boundaries of copyright restrictions’ 788-789.

[106] CDADC, art 75 (2) (i).

[107] InfoSoc Directive, art 5 (3) says that “Member States may provide for exceptions or limitations” in the following case: “(a) use for the sole purpose of illustration for teaching or scientific research, as long as the source, including the author’s name, is indicated, unless this turns out to be impossible and to the extent justified by the non-commercial purpose to be achieved”.

[108] CDSM Directive, title II, art 5 (1) says that “Member States shall provide for an exception or limitation […] in order to allow the digital use of works and other subject matter for the sole purpose of illustration for teaching, to the extent justified by the non-commercial purpose to be achieved, on the condition that such use (a) takes place under the responsibility of an educational establishment, on its premises or at other venues, or through a secure electronic environment accessible only by the educational establisment’s pupils or students and teaching staff”. This provision has just been implemented in art 75 (2) (g) and art 76 (7) of the Portuguese Code of Copyright. Art 5 of the CDSM Directive also establishes that “Member States may provide for fair compensation for rightholders for the use of their works”, which the Portuguese CDADC has now incorporated into art 76 (1) (c), providing for “equitable remuneration to be paid to the author and publisher”.

[109] Moscon, ‘Academic Freedom, Copyright, and Access to Scholarly Works’ 101.

[110] CDADC, art 52 (1).

[111] CDADC, art 52 (2).

[112] As Valentina Moscon notes, the judicial power and the legislature currently seem “to care more about right holders’ than users’ interests” (Moscon, ‘Academic Freedom, Copyright, and Access to Scholarly Works’, p. 117).

[113] David R. Hansen, ‘Orphan Works: Mapping the Possible Solution Spaces’ (2012) Berkeley Digital Library Copyright Project White Paper 2 <> accessed 4 June 2023.

[114] CDADC, art 75 (2) (u).

[115] See David R. Hansen’s essay for a range of proposed orphan works solutions: “[r]emedy-limitation approaches, such as the one advocated in the 2006 U.S. Copyright office proposal, that are predicated on a user’s good-faith, reasonable search for rights holders; administrative systems, such as the one adopted in Canada, that allow users to petition a centralized copyright board to license specific reuses of orphan works; access and reuse solutions that are tailored to rely upon the existing doctrine of fair use; and extended collective licensing schemes, which permit collective management organizations ('CMOs') to license the use of works that are not necessarily owned by CMO members, but that are representative of the CMO members’ works” (Hansen, ‘Orphan Works: Mapping the Possible Solution Spaces’).

[116] CDADC, art 75 (2) (v) (w); art 75 (6); art 76 (4) (5) (6).

[117] Dillen and Neyt, ‘Digital scholarly editing within the boundaries of copyright restrictions’ 790-791.

[118] Christof Schöch, Frédéric Döhl, Achim Rettinger, Evelyn Gius, Peer Trilcke, Peter Leinen, Fotis Jannidis, Maria Hinzmann, Jörg Röpke, ‘Abgeleitete Textformate: Text und Data Mining mit urheberrechtlich geschützten Textbeständen’ (2020) Zeitschrift für Digitale Geisteswissenschaften <> accessed 25 August 2023; José Calvo Tello and Nanette Rißler-Pipka, ‘¿Qué hacer con textos que no se pueden publicar? Datos derivados, criterios FAIR y TEI’ (2023) 16 Journal of the Text Encoding Initiative <> accessed 25 August 2023.

[119] Schöch et al., ‘Abgeleitete Textformate: Text und Data Mining mit urheberrechtlich geschützten Textbeständen’.

[120] CDADC, art 75 (2) (h).

[121] CDADC, art 76 (2).

[122] Meanwhile, the Stockholm revision replaced this adjective by the expression “compatible with fair practice” (Berne Convention for the Protection of Literary and Artistic Works. Stockholm Act, 1967, art 10 [1] <> accessed 4 May 2023).

[123] Taylor & Francis, ‘Author Services’ <> accessed 15 November 2019.

[124] Taylor & Francis, ‘Author Publishing Agreement’ <> accessed 15 November 2019.

[125] Unlike art 5 of the new CDSM Directive, this art 5 of the InfoSoc Directive was a non-compulsory exception, stating that “Member States may provide for limitations” to allow “quotations for purposes such as criticism or review”. In the UK (where many scholarly publishing groups are based), quotations for criticism or review are also only allowed when “the work has been made available to the public” (Copyright, Designs and Patents Act 1988, Section 30 <> accessed 4 May 2023).

[126] There are several such cases abroad, involving the Joyce Estate. According to Robert Spoo, “Professor Carol Loeb Shloss of Stanford University’s English Department […] had spent years researching a biography of Joyce’s talented and troubled daughter, Lucia, and at last published it in 2003 […], but not before she and her publisher deleted many quotations from unpublished material after receiving multiple threats from Mr. Joyce. When Shloss informed the estate that she intended to create a website that would contain the deleted quotations placed within a scholarly context, the Estate forbade the project as unauthorized and infringing. Having engaged legal counsel […], Shloss filed an action against the estate in a California federal court, seeking a declaration that her proposed website made fair use of the copyrighted materials and that the estate’s actions with respect to her and other scholars over the years constituted copyright misuse. […] After losing its motion to dismiss, […] the Joyce estate agreed to a settlement whereby Shloss was able to place on her website all of the quoted material she had planned to include, and to make additional uses of the material that she had not sought in her complaint” (Spoo, ‘Ezra Pound’s Copyright Statute: Perpetual Rights and the Problem of Heirs’ 1826).

[127] Spoo, ‘Copyrights and “Design-Around” Scholarship’ 576.

[128] Ibid 566.

[129] Ibid 564.

[130] Dirk Van Hulle, ‘Modelling a Digital Scholarly Edition for Genetic Criticism: A Rapprochement’ (2016) 12-13 Variants 34-56 <> accessed 10 July 2023.

[131] “In international copyright law, the ‘three-step test’ restricts the ability of states to introduce, and maintain, exceptions to the exclusive rights of authors and other right-holders. Under its well-known terms, exceptions are only permitted (1) in certain special cases; (2) which do not result in a conflict with the normal exploitation of a work and (3) which do not unreasonably prejudice the legitimate interests of the author (or other right-holder). Originating in the 1967 Stockholm Conference revision of the Berne Convention, this formula now forms an integral part of several international agreements concerning copyright and related rights and has been applied as a constraint on the availability of exceptions to the exercise of other forms of intellectual property right at international level. The ‘test’ has also recently come to play a significant role in domestic copyright laws” – Jonathan Griffiths, ‘The “Three-Step Test” in European Copyright Law: Problems and Solutions’ (2009) 4 Intellectual Property Quarterly 428-457 <> accessed 27 August 2023.

[132] Ryan, ‘Fair Use and Academic Expression’ 541.

[133] Spoo, ‘Copyrights and “Design-Around” Scholarship’ 564.

[134] Ryan, ‘Fair Use and Academic Expression’ 590.

[135] Spoo, ‘Ezra Pound’s Copyright Statute: Perpetual Rights and the Problem of Heirs’ 1827.

[136] It goes without saying that the suppression of the words contained in the document or other scholarly use of “ultra-safe substitutions for literary art […] does not inject a functional equivalent into the intellectual” activity of our disciplinary field (Spoo, ‘Copyrights and Design-Around Scholarship’ 578).

[137] Spoo, ‘Copyrights and Design-Around Scholarship’ 590. Although “there is little consensus as to what academic freedom means” (Moscon, ‘Academic Freedom, Copyright, and Access to Scholarly Works’ 103), according to Michael W. McConnell “[t]he term refers both to the freedom of the individual scholar to teach and research without interference (except for the requirement of adherence to professional norms, which is judged by fellow scholars in the discipline) and to the freedom of the academic institution from outside control” – Michael W. McConnell, ‘Academic Freedom in Religious Colleges and Universities (1990) 53 Law and Contemporary Problems 305 <> accessed 26 August 2023. For an introduction to the theory of academic freedom, see also Ryan, ‘Fair Use and Academic Expression’ 573-576.

[138] Moscon, ‘Academic Freedom, Copyright, and Access to Scholarly Works’ 103. Moscon notes that art 13 of the Charter of Fundamental Rights of the European Union, for instance, establishes that “[t]he arts and scientific research shall be free of constraint” (Ibid 104), and “[a]t the European national level, academic freedom […] is usually afforded separate protection in the Constitution” (Ibid 105), as happens in Portugal (Constitution of the Portuguese Republic, 7th Revision [2005], pt I, title III, ch III, art 73 [4] <> accessed 25 August 2023).

[139] See the essays by Robert Spoo – ‘Ezra Pound’s Copyright Statute: Perpetual Rights and the Problem of Heirs’ 1828-1831 – and David Hansen – ‘Orphan Works: Mapping the Possible Solution Spaces’ – for more specific proposals on balancing long copyrights with the needs of the public.



Any party may pass on this Work by electronic means and make it available for download under the terms and conditions of the Digital Peer Publishing License. The text of the license may be accessed and retrieved at

JIPITEC – Journal of Intellectual Property, Information Technology and E-Commerce Law
Article search
Extended article search
Subscribe to our newsletter
Follow Us