Publishing Embargoes and Versions of Preprints: Impact on the Dissemination of Information

1 Copyright Transfer and the Freedom to Disseminate Information

When a creative or literary work (mainly academic papers in the form of an article, letter, editorial, etc. that is submitted to a journal or preprint server) is born, the creator automatically holds copyright, although it can change hands. For instance, for US authors, the applicable law is 17 USC 201 (a) (U.S. Copyright Office, 2024). A common form of copyright in publishing that academics are likely to encounter is the transfer of copyright to a publisher (Oppenheim, 2020; Teixeira da Silva & Al-Khatib, 2021). Mostly, 10 of the world’s largest (by journal volume) academic publishers (Nishikawa-Pacher, 2022) request copyright transfer rather than a license to publish (Table A1). If authors transfer copyright to a publisher, does this stifle the dissemination of information? By assigning exclusive rights to a publisher, which seeks those rights to recoup its costs and investments, authors purportedly receive benefits such as scientific verification through peer review, indexing, the dissemination of information (Tennant et al., 2016; Ware & Mabe, 2015), and branding, via the journal’s imprimatur and prestige (Centivany, 2011), aspects that they would not achieve if they attempted to disseminate the same work or information on their own, i.e., via self-publishing. However, copyright transfer agreements can impose certain restrictions on how authors and their institutions may share their published works and what versions of it may be shared (Kohn & Lange, 2018; Tennant et al., 2019), and whether there are restrictions on the dissemination of knowledge (Lawson, 2017).

Legitimate publishers secure intellectual rights, as copyright, through their legal and voluntary transfer, as literary works, from authors. Subscription-based access is the “cost” of the transfer of copyright, limiting knowledge users or readers to those in institutions whose budgets can pay for such access, suggesting that the protection of intellectual property, in the form of copyright transfer, limits the dissemination of knowledge through limited access. The subscription model of access to scholarly content led to the “serials crisis” in the 1990s, caused by the rapid growth of subscription costs that exceeded the budgets and subscription capacity of libraries and the decreasing value of return of these “big deals” (McKenzie, 2018). Criticism arose of the value of “big deals,” which were designed to combat the same “serials crisis,” but instead allowed publishers to sell access to journal sets to library consortia (Rodríguez-Bravo, Fernández-Ramos, De-la-Mano, & Vianello-Osti, 2021). Dissatisfaction with this model expressed itself as boycotts of large commercial publishers because of their aggressive pricing policies (Larivière, Haustein, & Mongeon, 2015).

In this article, departing from the now somewhat antiquated copyright transfer-based subscription publishing model, we reflect on whether open access (OA) models, including preprints, offer authors a wider capacity to disseminate their work and information.

To identify relevant literature for this study, databases such as Scopus, Web of Science, and Google Scholar were used, employing search terms including “copyright,” “publishing embargoes,” and “preprint servers.” This non-systematic search methodology prioritized recent publications and incorporated snowball sampling by reviewing reference lists of key articles to uncover additional relevant sources. The selection of articles was based on their relevance to the research objectives and their theoretical and empirical contributions. This approach ensured a diverse and robust collection of interdisciplinary literature, with our expertise playing a pivotal role in this process.

2 Information Ownership and Dissemination: How do Different Publishing Models Stack Up?

In our opinion, a fundamental academic power struggle of our time is taking place that involves the ownership and dissemination of ideas, intellect, and information, i.e., knowledge, as well as academics’ freedom to engage in and control aspects of the latter process. There are debates on the hindrance of academic freedom through institutional OA policies and mandates, i.e., restrictions to the choice of publication venues (Anderson, 2021). Yet, there are also strong arguments for supporting OA in relation to the collective good of the public and broadening an understanding of positive forms of academic freedom (Moore, 2021). This may give the impression that there is a tug-of-war between the world of copyright and information restriction caused by the paid-to-access publishing model on the one hand, and the free dissemination of information (and to some extent open science) on the other hand, via the OA publishing model. In the rest of this article, focusing on OA and preprints, we examine a few topics that may bring more clarity to this issue. As our point of departure, we note that even though copyright and OA are not diametrically opposed concepts, there are fundamental aspects (e.g., subscription-based publishing models require that authors transfer complete and exclusive copyright to the publisher prior to publication) that do not always make them compatible in scholarly publishing practices (Dawson & Yang, 2016).

Gold open access (GOA), which emerged as a publishing model to liberalize information relative to the subscription model of access, represents a form of disruptive knowledge dissemination (Herman, Akeroyd, Bequet, Nicholas, & Watkinson, 2020). In GOA, authors retain their copyright through a creative commons license (Pollock & Michael, 2021). However, to achieve this, they must pay an article processing charge (APC) to flip from the subscription-based version to an OA version, i.e., unlike diamond OA, GOA involves a “tax” on the dissemination of information. Either authors pay the APC from their own pocket (least likely option due to personal financial restrictions) or their institute or funder pays it (Zhang, Wei, Huang, & Sivertsen, 2022).

The issue of dissemination can be often gauged indirectly via the number of citations a paper attracts. The OA “advantage” is thus often argued via its citation “advantage,” with the claim that OA papers are more highly cited than papers published in subscription journals (Dorta-González & Dorta-González, 2023; Mikki, 2017), although not all literature has reached the same conclusion (Langham-Putrow, Bakker, & Riegelman, 2021). That metrics-based citation advantage is then used as a marketing strategy to get academics (or their research institutes and/or funders) to pay APCs that might be proportional to the journal’s metrics (Zhang & Watson, 2017), which are also marketed. This switch might exacerbate inequalities and epistemic injustice in the dissemination of opinions and output for certain groups of researchers, such as early-career or independent researchers, those in specific disciplines, researchers from low-budget institutions, and/or low-income countries, among others (Hadad & Aharony, 2023; Klebel & Ross-Hellauer, 2023; Ma, Buggle, & O’Neill, 2023). This flips the inequality or injustice from readers (pay to read) to authors (pay to publish). In the diamond OA model, authors, their institutes, or their funders are not burdened with the cost of publication, but this is not an attractive business model for for-profit publishers (Fuchs & Sandoval, 2013), and is accompanied by other challenges such as long-term sustainability and selection bias (Normand, 2018).

From our perspective, three great forces in the ongoing struggle for the access, dissemination, and ownership of information are as follows: (1) black or pirate OA (i.e., illegally downloaded or pirated versions of an article), the most prominent portals being Sci-Hub (Himmelstein et al., 2018) and LibGen; (2) the rise of preprints in recent years; (3) the concomitant rise in the struggle for OA dominance, and the natural evolution of predatory OA publishing, as a competitive force within the APC-based OA journal market, but a model whose dynamics are not always easy to distinguish from exploitative publishing practices (Teixeira da Silva, Dobránszki, Tsigaris, & Al-Khatib, 2019).

3 Preprints as a Simple but Efficient Information Dissemination Model

Several preprint servers are vying for a greater share of the preprint market (Teixeira da Silva, 2017), seeking market dominance via the early “capture” of intellectual ideas, suggesting that an information power struggle may be taking place in the academic community and publishing industry. As evidence of this concept in practice, we point towards the acquisition, in December of 2022, of the preprint server Research Square (and associated editing services, American Journal Editors), by the publishing giant Springer Nature (Springer Nature Group, 2022), possibly as one way to consolidate and fortify its publishing market portfolio, thereby rekindling its desire to publicly trade the company (Teixeira da Silva & Fassin, 2022). The submission of any paper to a Springer Nature journal currently involves the mandatory sharing of metadata with Research Square, and in the process of submission as well as in the case of rejection, the possibility of posting that work as a preprint to Research Square.

There are ample articles debating the positive role of preprints in the dissemination of information, and offering an explanation of why preprints are gaining popularity (Moshontz, Binion, Walton, Brown, & Syed, 2021), although academics are cautioned about their risks and known inconsistencies in policies (Malički et al., 2020). As one example, variation in preprint policies would make some versions acceptable to some journals but not others (Penfold & Polka, 2020). Posting preprints is a priority practice in certain communities (e.g., higher energy physics) (Neylon, Pattinson, Bilder, & Lin, 2017; Smart, 2022), thus serving to amplify that information. There is evidence that this approach may have positive effects on the volume of citations (Fraser, Momeni, Mayr, & Peters, 2020; Wang, Glänzel, & Chen, 2020) and the diversity of citations (Huang et al., 2024), as signals of impact. Ni and Waltman (2024) recently explored geographical disparities in the adoption and perception of preprinting, revealing that familiarity with and utilization of preprints are considerably higher in the United States and Europe compared to China, where there are concerns about recognition and the risk of intellectual property theft.

4 Publishing Embargoes: A Thorn in Academics’ Sides, or Much Ado About Nothing?

5 Is Information Sharing in Academic Publishing in a State of Stagnation or at an Impasse?

While acknowledging that academic publishing has created unique niches with ancillary services (see our examples above with Springer Nature, and its consolidation of journal publishing, preprints, and English editing and science revision services as one set of integrated services), some of which also manage trivial requirements, such as formatting, we feel that academic publishing has reached some sort of a state of stagnation related to the novelty of information dissemination. A broad impression is created that much has changed over the years in academic publishing, but when viewed simplistically, there are really still only two broad choices for authors wishing to work with a publisher: subscription or OA (including its various forms) (Björk, 2017). Within these two main modalities, intricate intermediate models, concepts, terminologies, and platforms have been born, each in itself a potential market, either to capture intellect or with its financial derivatives (e.g., APCs).

Further emphasising our perception of the potential stagnation in scholarly communication is the issue of layered “guilt” and conformity. We offer two possible explanations. First, if an academic does not publish in an OA journal, or if their work is not OA, then a negative stigma may be attached to them and/or their work, or they may feel pressure to publish OA to match the patterns shown by their peers. Already, paying a low APC is labelled by some as one way of supporting “predatory” OA publishing (McCann & Polacsek, 2018; Mills & Inouye, 2021; Siler, 2020). Yet, paying a high APC may be a significant barrier for many authors, including a hindrance to the choices for disseminating their information. Yet, somehow, this aspect is not considered by the same academics as supporting an exploitative publishing model. This creates contradictory dynamics that contemporary authors need to navigate through. The second factor may be related to a social function and amplified by the surge in the open science philosophy, in which if one does not publish as OA, then one might not be supportive of the public “good.” To support the latter, as aptly stated by Lachapelle (2022) in the abstract of their thesis, “preprint servers participate in a profound reversal of epistemic evaluators and the logic of scientific capital,” and also “the early dissemination of preprint manuscripts propels scientific readers and other social actors (journalists, governmental officials, civil society) into the role of primary evaluators of scientific knowledge.” However, preprints may also contribute to the spread of misinformation (Heimstädt, 2020). Therefore, framing the conflict by labelling the user as what they are or what they are not, a characterization that is based on what they do/do not do or use/do not use is – as we see it – a novel (and potentially disturbing) aspect of the OA branding business model.

An outcome of this culture of “guilt” and conformity is the set of implications associated with terminology such as the “Global South” and its battle against the “Global North,” geographically irrelevant terms that introduce a cultural rift in the academic and financial status of academics (Teixeira da Silva, 2021). Another consideration is that the ability to publish OA does not necessarily reflect the desire of an author to publish OA, but is instead based on the institutional support infrastructure that covers OA costs (in the case of institutes that mandate that their researchers’ papers must all be published as OA). In other words, the ability to publish OA depends on the ability of the researcher’s institution to pay APCs if they are not from a low-income (or “Global South”) country that is eligible for an APC waiver (Hadad & Aharony, 2023). In essence, GOA is shifting the balance from “we can’t afford to read” to “we can’t afford to publish.” It is not uncommon to find, in the online submission systems of journals of some of the publishers listed in Table A1, a clause that asks authors whether they want to publish their papers as OA. That question is of course, not well phrased because the ability to publish as OA is not dependent exclusively on desire, but rather on means. In other words, the question should more realistically ask, “Can you (i.e., are you able to) publish your paper as OA?” This instead implies whether authors or their institutes and funders have the means to publish as OA, by paying the APC.

Finally, we caution readers to be vigilant of loosely used terms like “democratization” or “equity,” which might not necessarily represent a fair form of knowledge sharing, but rather a neoliberal marketing strategy to further commodify OA through the prism of “rights retention” (Moore, 2023). Such strategies might not fully align with the original intent of inclusivity and openness, potentially generating an entire class of marginalized academics who then seek alternative ways to express, publish and showcase their intellect, and sensing that such catch-phrases may liberalize their work’s information sharing. The scholarly publishing industry will equally adapt to new forms of needs and find their way to profit from new business models. There will always be two sides of the coin.

6 Conclusion

In this commentary, we acknowledge that there is value in a publisher’s imprimatur, recognizing that for-profit publishers would like to recoup their investments in, among other aspects, content branding, discoverability, quality assurance, typesetting, and proofreading, that they claim is imparted to papers whose copyright has been transferred to them in the subscription model, or in papers for which they have received an APC to publish as GOA. The ability of such publishers to showcase authors’ literature, primarily through database and indexing prowess, offers authors maximum visibility of their work relative to self-publishing, or who publish with competitors who do not promote their work as well. Visibility then equates to and translates into dissemination. Academics who might be tempted to use preprints, or who might heed the calls by preprint proponents to use preprints, would do well to appreciate some of the subtleties in the background of the OA economic models we debate in this article, reflecting on the availability of publishing options to advance and showcase their intellectual ideas. Even though we have no evidence to indicate that copyright transfer in any way is limiting the dissemination of scientific information, we recognize that this comes at a steep premium for academia and society. We do take aim at some restrictive measures in place by several of the largest or most prominent publishers (see 10 examples in Table A1) that request the transfer of copyright, or to a lesser extent, request an exclusive license, but at the same time place embargoes on self-archived versions, which we feel might restrict dissemination, although this claim is not yet based on evidence. Finally, our perception that the subscription model as a negative force (in terms of information dissemination) is in some sort of a battle against its positive OA counterpart was blunted by the flip-to-GOA-model, in which the “bad actor” was no longer perceived to be the publisher, who was providing a choice to revert to OA, but the authors (or their institute or funder) who are unwilling or unable to make knowledge open through paying the required APCs.