Earlier this week, cOAlition S released its draft Guidance on the Implementation of Plan S, which retains the requirement that publications resulting from cOAlition S funding be licensed under terms laid out in the Berlin Declaration.

For scholarly articles the public should be granted a worldwide, royalty-free, non-exclusive, irrevocable license to share (i.e. copy and redistribute the material in any medium or format) and adapt (i.e. remix, transform, and build upon the material) the work for any purpose, including commercially, provided proper attribution is given to the author.

Unfortunately, this guidance fails to respond adequately to the fact that many researchers – especially in the arts and humanities, but including also some scientists – would sometimes choose a license granting fewer reuse rights.

I know that my particular Twitter bubble doesn’t represent everyone; and it does include quite a few Open Access (OA) advocates, who tend to support the use of maximally open licenses (such as CC BY). Nonetheless, I’ve been surprised to find that most of the folks I interact with on Twitter seem either not to understand why anyone would object to mandating such an open license or have no problem supporting such a mandate over the objections of colleagues. Not understanding the objections is one thing. Not caring about the objections is quite another. I address each possibility, in turn. Here, I focus only on the problem of not understanding the objections. In a later post, I will address the problem of not caring.

First, though, a few points of clarification. The draft Guidance (at least with reference to licenses) explicitly concerns scholarly journal articles only, leaving guidance on books and book chapters until a later date.

The following guidance further specifies the principles of Plan S and provides paths for their implementation regarding scholarly articles. The guidance is directed at cOAlition S members and the wider international research community. cOAlition S will, at a later stage, issue guidance on Open Access monographs and book chapters.

I do think this makes some difference. Personally, I am much less likely to opt for a more restrictive license on scholarly articles than I would on a book. For me, book chapters are actually closer to journal articles in this regard. That doesn’t mean I fully support a license mandate for journal articles; but I do think I’d have less frequent or momentous worries about open-licensing an article or book chapter than I would a book.

For the sake of brevity, and since the draft Guidance will mandate CC BY 4.0 licenses for all scholarly articles, I use ‘CC BY’ to refer to the sort of open license mandated by the draft Guidance.

All scholarly articles that result from research funded by members of cOAlition S must be openly available immediately upon publication without any embargo period. They must be permanently accessible under an open license allowing for re-use for any purpose, subject to proper attribution of authorship. cOAlition S recommends using Creative Commons licenses (CC) for all scholarly publications and will by default require the CC BY Attribution 4.0 license for scholarly articles.

I also refer to ‘the author’ throughout this post, even though the author may not be the copyright holder of the work. Plan S initially mandated that authors retain copyright. The draft Guidance seems to have opened the door to the idea that there may be other rightsholders (such as the author’s institution, although perhaps even the publisher). I don’t address this question in this post.

For the record, I have no objection to the requirement to make journal articles immediately available and free to read (gratis). It is the option virtually all authors would choose, given the power. Insofar as Plan S “mandates” immediate gratis OA, it actually empowers authors. Mandating CC BY (or CC BY-SA or CC0 versions of libre) does not. Adopting a license mandate is a matter of domination, not empowerment. More on that, later.

One more preliminary: acknowledgements. In addition to the inhabitants of my Twittersphere, with whom I’ve had some excellent exchanges, I heard a great presentation as part of the NJIT Department of Humanities Fall Colloquium Series that spurred my thinking on this matter. On November 7, Lisa DeTora of Hofstra presented on “Not Quite Mythology: The Functional Limitations of Biomedical Language.” The views I outline below owe much to her talk. 

FAILURE TO UNDERSTAND OBJECTIONS TO CC BY

Mandating CC BY treats all articles as if they were scientific articles. But published articles mean different things in different fields. In the sciences, for instance, it is more typical to perceive the publication as simply describing the experiment or reporting on the results of the research. Although priority in publication of results is vital, scientists fully expect others to build upon, modify, and eventually abandon their results as science progresses. A CC BY or equivalent license makes some sense in the context of articles considered as merely containers for information to which the scientist is not wedded in a personal way.

But in the humanities, to paraphrase Nietzsche, nothing is impersonal. In the humanities, the publication itself actually constitutes the research. I still quote the Ancients in my own field of philosophy. That’s not because I haven’t read the latest literature. It’s because Plato and Aristotle have stood the test of time. At the risk of sounding as if I have delusions of grandeur, I hope the same for my own works.

In the sciences, articles might be thought to contain information that, like water, could be poured into various containers and retain its essential characteristics. Although one might prefer to be published in Science or Nature, ultimately, the container doesn’t matter. What matters most is the discovery, and to a lesser extent, the discoverer. Attribution is sufficient.

In the humanities, we have similar preferences for pet journals. I agree that this tendency should be corrected, and that the particular journal in which one is published should matter less than what’s published. But our conception of what’s published — of articles — is different. It’s not so easy to separate form and content in the humanities. To suggest that a philosophy article is simply information that could be transferred easily from place to place would be akin to suggesting we try to pour a marble statue — not water — into a container that could barely hold the statue’s volume of material. To manage it, one would have to destroy the statue.

One tends to attach a name to a theory or discovery in the sciences in one of a few circumstances: 1) when the discovery is new or big or a prize is awarded — as with the Higgs boson (though everyone seems uncomfortable focusing on Higgs alone, and credit is shared around); 2) when the theory has been surpassed — as with Aristotelian physics; or 3) when one wants to undermine a theory — as with creationist attacks on ‘Darwinism’.

In the humanities, attaching names to works or theories is important and routine.  Despite what you may have heard, there’s no such thing as ‘Utilitarianism’, but rather Bentham’s, Mill’s, or Singer’s versions thereof. It makes a difference whether one refers to a Jamesian or Deweyian pragmatism. The practice of naming in the humanities is not reserved for a flash in the pan or for theories that didn’t pan out. In the humanities, one’s name — including, but not limited to, one’s reputation as a good researcher — is important.

https://twitter.com/cofactoranna/status/1067774707917828100

Derivative works are a more sensitive matter in the humanities than in the sciences. One’s works present oneself — not just one’s findings — to the scholarly community. The focus of a translated work in the humanities is on the author, not on the translator. It is the author’s introduction to a new country’s readers. It is not by chance that most works in the humanities are single-authored.

Translating a work of philosophy, to pick my own field, is not something that can be done willy-nilly by an algorithm without destroying vital aspects of the work, thereby misrepresenting the author. Nor can a translation simply be thrown together by anyone who understands both languages. A good translation of a difficult text is more difficult than that.

A good translation depends on forming a relationship with the translator, such that he or she is actually familiar with the author’s work in the original language. They need to understand the subtleties of the argument and the choices the author made between terms in the original language in order to craft a good translation in the new language. Ideally, the author would have a relationship with more than one such person, so that the translation could be checked by someone the author trusts. All of this would involve a great deal of time, effort, and discussion.

A bad translation of a difficult text misrepresents the author’s views in ways that can severely damage their name. A good translation can help make it. Rumor has it that even native speakers of German read Norman Kemp Smith’s English translation of Kant’s Critique of Pure Reason. No one, however, is under the impression that the author of the work is anyone other than Kant.

On the link to the Google Books version of the Critique of Pure Reason, the preview is from Kemp Smith’s introduction, in which he describes the process of translating Kant. He was careful and meticulous and thought everything through. He studied and benefited from previous translations. He was writing commentaries on Kant — which is to say, he was a Kant scholar, not simply a Scotsman who spoke fair German. It took years. Kemp Smith’s is the sort of translation we need and generally aspire to in the humanities.

This is not to say that all translations in the humanities are good. Martin Paul Eve discusses the controversy over a particular translation of Michel Foucault, noting that perhaps some translation — even a poor or controversial one — is better than no translation at all. An author who holds such a view and has no objection to subjecting their work to rough translations can always choose a CC BY or equivalent license. Mandating that all authors explicitly grant permission for their works to be treated roughly, though, is unfair and fundamentally misunderstands the nature of research in the humanities.

The application of a more restrictive license — CC BY-ND, say, which would prevent others from making derivative works without the author’s permission — does not bar the author from granting permission for other uses in the future. Indeed, since the potential translator would actually need to contact the author to obtain permission to translate the work, such an arrangement could facilitate the formation of the sort of trusting relationship necessary for a good translation.

FAILURE TO CARE ABOUT OBJECTIONS TO CC BY

I hope that the vast majority of those currently willing to impose a CC BY mandate as part of Plan S suffer only from a failure to understand why someone might sometimes object to a CC BY license. I fear that some, however, fully understand why someone might sometimes opt for a different license and yet feel fully justified in mandating CC BY for everyone, always. Those who understand the objections to a CC BY license mandate, yet fail to take those objections into account, are guilty of embracing domination.

Since this is a strong charge, I want to take my time in making it. I also want to appeal relatively quickly to those who simply misunderstand objections to a CC BY license mandate. For these reasons, I will publish this post now and issue a promise to complete the second post later. [EDIT: part 2 now available HERE.]

I should add, briefly, that I am now rethinking my position on the CC BY license mandate. Heretofore, I had floated the possibility that cOAlition S could safely mandate a CC BY license without dominating authors as long as they were to offer a no-questions-asked waiver. Providing such an opt-out option would mirror the Harvard Open Access Policy, which I discuss in some detail here. Upon further review, however, I am leaning toward the position that even a mandate with a waiver would dominate authors. If anyone is interested in how my thinking has developed and cannot wait for the second post on this topic, here’s a hint.