Editor’s Note: Today’s post is by Elizabeth Gadd, Research Policy Manager (Publications), Loughborough University.
The recent law suit against ResearchGate brought by Elsevier and the American Chemical Society follows hard upon the $15 million damages awarded to Elsevier in their recent case against Sci-Hub. These are just the latest actions in a long line of scholarly copyright wars. As far back as 2001, Stephen Pinfield found that academics were illegally posting publisher’s copyright content on ArXiv — a practice since dubbed ‘Black Open Access’ (Bjork, 2017). A series of subsequent studies have reported a similar disregard for publisher copyright policies (e.g., Antelmann, 2006; Troll Covey, 2009) culminating in Jamali’s 2017 study that suggested just over half of all papers on ResearchGate were illegally posted publisher PDFs. This has led to a perception that academics just don’t care about or understand copyright and that someone needs to take responsibility for getting those academics a copyright education. There is no doubt some truth in this, but I think the situation is slightly more complex than that. In fact, I think that the keys to resolving this may rest in the hands of publishers, but in using those keys publishers may pay a high price.
Let me explain what I mean by walking through a number of theories I have come to over the years regarding academics’ understanding of copyright.
Theory 1: Academics occupy a broad spectrum of copyright knowledge, beliefs and confusion
It is no longer fair to say that academics know nothing about copyright. Twenty years ago you may have seen very little variation among academics in terms of their understanding as to who owns copyright. Indeed, a study by the ALPSP in 2002 found that 79% of respondents quite confidently claimed to own the copyright in their research papers (this was unlikely as around 90% of publishers asked for copyright transfer at the time). However the open access (OA) movement has brought with it greater awareness of copyright issues as OA advocates have called for academics to retain their copyright so they can self-archive their work; and funders have demanded increasingly liberal reuse licenses (e.g., CC-BY) which have not been universally supported. However, this has also brought with it higher levels of uncertainty and a greater range of beliefs. A recent survey by Kudos showed a very mixed bag when it came to academics’ views on publisher copyright policies. There were both those who thought they were fair (39%) and unfair (27%), straightforward (48%) and complicated (34%), restrictive (42%) and not restrictive (34%), and both fit (44%) and unfit (23%) for purpose. Only 20% didn’t think it was necessary to check a publisher copyright policy before posting to a Social Networking Site (SNS), suggesting the majority had a greater understanding that copyright may not actually belong to them. There is confusion as there has always been, but not all academics are confused, and the confused ones are not confused in the same way.
Theory 2: Academics display cognitive dissonance in relation to copyright practices
There is no shortage of studies that show that academics believe that copyright is rightfully theirs and they should retain it (Swan, 1999; Swan,2002; Pila,2010; Rahmatian, 2014; Davies, 2015). However, academics do assign copyright to publishers in large numbers, some freely (41%) but more reluctantly (49%) (Gadd et al, 2003a). It has been suggested to me that this is a form of cognitive dissonance. Cognitive dissonance describes the practice of believing one thing (e.g., smoking kills) and doing another (e.g., smoking). There are many commentators who put academics’ reluctant copyright transfer activity down to a desperation to get published at any cost. There is no doubt some truth in this. However there are other examples of cognitive dissonance in relation to copyright practices (such as posting a paper on ResearchGate knowing you have assigned copyright to the publisher) which lead me to wondering whether something else is at play. I go on to expound on this in theories 3 and 4.
Theory 3: Scholarly culture is different to copyright culture
When academics are asked about how they want to protect their OA papers, and use others’ OA papers, they do not demand the full range of protection or usage rights offered by copyright law or e-journal license agreements (Gadd et al, 2003b,c; Rowley et al, 2017). However, they are principally concerned to protect what copyright wonks would call their “moral rights”, i.e., their right to author attribution (paternity right) and the right to object to derogatory treatment of their works (integrity right) (Gadd, et al, 2003b; Gadd et al, 2007; Rowley et al, 2017). Unfortunately, depending on your circumstances and where you are in the world, moral rights may or may not apply to scholarly works. Interestingly, recent demands by funders to make journal articles available under CC-BY licenses have met with objections for entirely the same reason: CC-BY promises authors’ attribution, but not the right to prevent unauthorized derivative works, some of which they might consider derogatory. What we see at work here is the tension between what I would call ‘copyright culture’ and scholarly culture. I use the term ‘copyright culture’ to refer to the range of copyright legislation, licenses, policies and practices that impact on scholarly activity. I believe the two cultures exist in a sort of Venn diagram relationship with areas of overlap and areas of distinct separateness.
Thus, you have: 1) practices that are legal under copyright but are contrary to scholarly culture; 2) practices that are accepted scholarly culture, but are not supported by copyright; and 3) practices in the middle where copyright supports or overlaps with scholarly culture. An example of 1) might be the taking of a CC-BY licensed work and selling it: definitely legal but definitely contrary to accepted scholarly norms. An example of 2) might be attributing 500 authors on a journal paper. Copyright law has clear guidelines as to what constitutes authorship and you’d struggle to argue that 500 individuals were joint authors (and therefore copyright owners) of 5,000 words. However, it is accepted scholarly culture to attribute large research groups on research papers. An example of 3) might be where a work is plagiarized (infringement of accepted scholarly culture) and copyright law allows the copyright owner to bring a court case based on infringement of copyright.
Theory 4: For academics, scholarly culture takes priority over copyright culture
So, going back to our cognitive dissonance theory, the way that individuals ease the tension caused by cognitive dissonance is to either change their practices (stop smoking) or change their beliefs (perhaps the ‘smoking kills’ evidence is wrong?). It would appear that academics deal with their copyright cognitive dissonance through giving priority to the beliefs and practices of scholarly culture, over and above that of copyright culture. We see this at work in the Kudos survey where 83% of academics felt that copyright policies should be respected, but at the same time 63% felt that despite such policies, academics should be allowed to post their papers on SNS. From recent evidence, it looks like that for the majority of academics, given the choice of aligning themselves with scholarly culture (sharing papers with their peers) or copyright culture (adhering to copyright policies) the former wins.
Theory 5: Academics believe there is greater overlap between copyright culture and scholarly culture than there actually is
This leads me on to my most important theory, and one which I think throws the most light on academics’ copyright practices. I would suggest that one of the key reasons that academics allow themselves to give greater priority to scholarly culture over copyright culture is that they believe that copyright supports (overlaps with) scholarly culture to a greater extent than it actually does. Thus, academics continue to assign copyright to publishers despite preferring not to because they see copyright assignment as an integral part of scholarly culture — the scholarly publication process — and it is to scholarly culture that academics are the most loyal. So, how have academics come to the conclusion that copyright culture supports scholarly culture, when open access advocates and librarians have been saying that it does not, and have been calling on them to retain copyright for so long? Well, to a large extent I think these beliefs have their roots in publisher’s efforts to convince academics that they both share the same commitment to scholarly culture and have developed a copyright framework that will support scholarly activity.
Thus, an analysis of publisher copyright transfer agreements in 2003 found that the primary justification for such agreements was to enable the publication of, and greater copyright protection for, the resulting work (Gadd et al, 2003d). In some cases copyright transfer was presented as a legal requirement. Certainly, most manuscript submission systems now make copyright transfer or exclusive license agreements an integral and unavoidable step in the publication process. However, when academics started objecting to the idea of copyright transfer, in the mid-2000’s many larger commercial publishers moved away from copyright transfer agreements and instead started introducing ‘exclusive licenses’ (Inger & Gardner,2013). These have almost exactly the same effect as copyright transfer agreements (they are an exclusive transfer of all rights under copyright) but they are more palatable to academic authors as they think they are retaining “copyright” and thereby all the rights that copyright bestows. Thus on exclusive licenses you will see phrases such as “copyright remains mine as the author…” (Sage). This blurring of the facts can leave academics with the view that they have more rights than they really do.
I believe it is for this reason we see academics posting their research papers onto SNS. Yes, academics are confused about copyright. However, their commitment to scholarly culture is the dominant force, and their belief that copyright supports this culture is shored up by publisher messages. So, on publisher web pages you will see statements such as “authors retain significant rights to use and share their own published articles. Elsevier supports the need for authors to share, disseminate and maximize the impact of their research“ or “Where possible, Emerald seeks to obtain copyright for the material it publishes, without authors giving up their moral or scholarly rights to reuse their work.” When academics are being sent such messages, it is not surprising that they believe they have the necessary rights to engage with scholarly culture in whatever way they see fit. To accuse academics of copyright ignorance or infringement in such circumstances is tantamount to tripping someone up and then telling them off for falling over.
To accuse academics of copyright ignorance or infringement in such circumstances is tantamount to tripping someone up and then telling them off for falling over.
So why don’t publishers send stronger and clearer messages to individual academics around what constitutes copyright infringement? Well, it would seem that by doing so, they would also be sending a message to academics that their interests don’t align: that when it comes down to it, publishers are primarily supportive of copyright culture and the exclusive rights it gives them, rather than scholarly culture which is something quite different. Indeed when the American Psychological Association recently issued take-down notices to various sites — including 80 university websites — which hosted ‘illegal’ copies of their papers, such was the outcry from academics that they quickly re-focused their attention onto ‘commercial piracy sites’ instead. However, it is not the sites that post these papers, it is academics themselves. By focusing on the sites as infringing rather than individual academics it further obfuscates the fact that academics’ copyright practices are not in line with publishers’ copyright policies. As a result academics are insulated from the sting in the tail of copyright culture and they can continue in their belief that copyright culture supports their scholarly practice.
I am not defending copyright infringement. I seek only to offer an explanation for it. And to suggest publishers may actually be contributing to the problem that they believe is affecting them so badly. Until academics recognize that the current copyright culture is not in alignment with their preferred scholarly communication practices, they will not change their copyright infringing behaviors. However, there is a risk with copyright education of this nature. If academics become aware that publisher interests do not align with their own, and the term ‘illegal sharing’ is not an oxymoron, but an actual possibility, they may not only change their copyright infringing behavior, but their copyright assignment behavior. This is what librarians and OA advocates have been calling for all along as they believe it would leave academics in a much stronger position to engage with scholarly culture — the sharing of their research papers — on their own terms.
In actual fact, relying on academics to retain copyright may no longer be necessary to achieve this end with the increase in ‘Harvard-style’ licenses. These are non-exclusive, worldwide, perpetual license grants from an author to their institution to enable them to make their journal articles and conference papers available on Institutional Repositories, often under a Creative Commons license. There are currently about 70 institutions in the US (Fruin & Sutton, 2016) with such a license, and the UK, led by Imperial College London, is developing a version of their own called the UK Scholarly Communication Licence (UK-SCL) (Banks, 2016). On the face of it, this will align copyright culture with scholarly communication culture to a far greater extent and will allow academics to share their papers on SNS as well as using them in other ways. However, there is clearly confusion among some academics around this and significant concerns from publishers. Whatever the outcome of this initiative, the UK-SCL is at least forcing some of these issues to the surface and seems to be resulting in an open (if tense) dialogue regarding the overlap and differences between copyright culture and scholarly practice.
The author is grateful to Rick Anderson, Charlie Rapple, Charles Oppenheim, Jane Secker and Chris Morrison for useful discussions around earlier drafts of this piece, and to the editorial team at Learned Publishing.
This post is made available under the terms of the Creative Commons Attribution-ShareAlike License.