Monday, November 12, 2018

Plan S and Researchers’ Rights: (Re)Framing Academic Freedom

When in 1915 the American Association of University Professors (AAUP) was formed, and published its Declaration of Principles on Academic Freedom and Academic Tenure, it was responding to a specific historical situation.

Today the historical situation researchers find themselves in is different and academic freedom faces new challenges. What complicates the picture is that there is considerable disagreement over what exactly academic freedom is, and what aspects of academic life it covers.

This lack of consensus has become all too apparent in the current discussions about Plan S. Indeed, some of those who support Plan S maintain that academic freedom is no longer relevant, or has ceased to exist in any meaningful way.

The key principle of Plan S – which a growing number of funders are signing up to – is that “After 1 January 2020 scientific publications on the results from research funded by public grants provided by national and European research councils and funding bodies, must be published in compliant Open Access Journals or on compliant Open Access Platforms.”

Most (if not all) researchers would surely agree that this is a laudable goal. But for some (we don’t know how many) the way in which it is currently envisaged that Plan S will be implemented raises a red flag with regard to academic freedom. It does not help that those who devised the initiative dismiss these concerns out of hand. The main architect of Plan S –  Robert-Jan Smits – has even bemoaned the fact that people are citing academic freedom in the context of Plan S, “because it stifles a lot of debate.”

Be that as it may, it is clear that some researchers believe Plan S does have implications for academic freedom. And a growing number of them (currently approaching 1,000) are signing a petition that asserts, amongst other things, that Plan S “is a serious violation of academic freedom”.

Given the apparent disenchantment with Plan S amongst at least some in the research community, and given that researchers find themselves increasingly subjected to ever more demanding OA policies like it (in which new duties, new restrictions and limitations, and new responsibilities are imposed on them), it is surely time to look again at what academic freedom does and does not mean, and what it should and should not mean in today’s context, and try to redefine and/or refine it for today’s historical situation; or at least to, as Marc Couture puts it in his guest post below, seek to “reframe” it?

What seems clear to me is that there is a pressing need for a debate about the relevance, role and responsibilities of academic freedom in the context of the growing list of open access policies, with a view to arriving at some kind of consensus.

What also seems clear to me is that this discussion ought to take place amongst researchers before funders seek to impose radical initiatives like Plan S on them. After all, if academic freedom means anything, it surely means that it is researchers and their institutions who should be in the driving seat over this, not funders.

But why not read what Marc Couture has to say below and think about the questions he asks before reaching your own conclusion on these matters. It may well be that Marc’s views on whether and how Plan S has implications for academic freedom are different to mine. But I am not a member of the research community, so my views are not what counts here. What I think does count is what the majority of researchers think. And without a meaningful debate, we will never know that.

Plan S and Researchers’ Rights: (Re)Framing Academic Freedom

By Marc Couture
The announcement of Plan S has generated many much needed (and much heated) discussions. I’m pleased to observe that these don’t concern the relevance of open access, whose wide-ranging benefits now seem to be almost universally acknowledged, but only potentially negative side-effects of the massive, if not global, shift to open access that the plan hopes to bring about.

Though many aspects of the plan are somewhat unclear, and most details of its implementation are still being drafted, what we do know already raises various worries. One is a possible conflict with academic freedom.

Two requirements of the plan are specifically targeted: (1) publication only in compliant journals (full OA, no hybrid) and (2) dissemination under a licence compliant with the Berlin OA definition, which would require authors to accept ceding generous usage rights in their works to others.

I must say first that I firmly believe that academic freedom is important; it is at the very heart of higher education. Drawing from the numerous available definitions, I would formulate its basic, most general definition as follows:

Academic freedom (AF) is the right, for individual academics as well as their institutions, to decide by themselves the subjects and ideas they wish to investigate, disseminate or teach upon without fear of reprisal or censorship, with the ultimate purpose of benefitting both scholarship and the common good.

Some definitions of AF include further rights, for instance the right – individually or as a community – to decide not only what is investigated, disseminated or taught, but how it is done.

Others add more specific freedoms, the most relevant in relation to Plan S being the freedom to choose the publication venue for one’s research results. One also remarks that the language used in some definitions can be very strong, qualifying such freedoms as “full”, “complete” or “without constraints”.

A quid pro quo

It must be pointed out that AF is neither an absolute, divine-like right, nor a “blank check” granted to researchers. It’s rather a privilege, bestowed to academia by society. Moreover, a quid pro quo is at work here.

On the one hand, society acknowledges that scholarship is the most powerful and trustworthy way to explain the world and solve its problems, as long as it is protected from ideological, political and economic vested interests.

On the other hand, scholars, both individually and collectively, must act in a responsible way, by living up to the values underpinning scholarship (rationality, critical thinking, honesty and respect in debates) and the common good (equity, inclusivity, human dignity).

So far, the discussions on actual or potential conflicts between Plan S and AF have involved specific freedoms found in one or another definition, with participants arguing that:

(1) This or that requirement of Plan S does (or doesn’t) concern, or conflict with, AF

(2) This or that measure proposed in Plan S may concern or conflict with AF, but other things as problematic, if not more, have been going on for years in academia without anybody complaining

(3) (Generally, in response to #1 or #2) this or that definition of AF is unclear, outdated or irrelevant, and must be dismissed, or interpreted in this or that way

My stance is that while more freedom is a priori preferable in the academy, the exercise of rights included in AF, especially the more specific ones found in its various definitions, may or may not contribute, in a given situation or context, to AF’s ultimate purpose of benefitting scholarship and/or the common good. In fact, it could even do the exact opposite, or entail both benefits and harms.

Thus, when a measure or a policy seems to conflict with those rights, one should not simply brandish AF as a “trump card” as an argument to counter it, but see this conflict as a warning, inviting one to examine to what extent the exercise of AF, in the situation at hand, effectively benefits – or hinders – scholarship and the common good.

Let me illustrate this approach by looking at the two issues I mentioned in relation to Plan S where the topic of AF has been brought forth.

(1) Constraints on publishing venues

Plan S puts severe constraints on the venues where research results may be published. This will reduce the choices available to researchers. However, to what extent it will do so is currently hard to assess, as it depends on publishers’ response to Plan S.

Here, any conflict with AF would be of utmost concern, since the right to disseminate one’s results is viewed as part of the basic definition of AF. One such conflict I could foresee in this context is if, when Plan S is finally implemented, researchers choosing marginal or unpopular, yet legitimate, topics or methods found they were unable to publish their results in peer-reviewed scholarly journals altogether, because not one, or too few of those complying with Plan S would consider, let alone accept, their papers.

However, I haven’t seen any mention of such a dire scenario in the discussions about Plan S; considering the ever-increasing number of OA journals in all fields, this is understandable. Rather, the major complaint is that Plan S will impede researchers from publishing in very specific journals, meaning highly selective, high impact ones.

This is viewed as conflicting with AF, on the grounds that definitions of AF either imply researchers must be the only ones to have a say in the way research is done (this may or may not include publication), or with reference to more specific freedoms, like “full freedom in publication” or the right to choose the publishing venue.

It is thus not surprising that critics have invoked specific rights associated to AF to oppose measures that could either (1) hurt leading, generally (for the time being) non-compliant journals or (2) impede researchers from publishing in them.

However, researchers wouldn’t lose the possibility of disseminating their results. Most papers submitted to these highly selective journals are rejected not for being faulty, but because they are deemed unimportant, not original enough, or unfit for the venue.

Nevertheless, most of them are subsequently published in other journals, often journals that focus less on criteria like relevance and potential impact and more on the general quality of the paper and the soundness of reasoning and analysis.

What those researchers who succeed in having their papers published in prestigious journals really fear is losing the ability to benefit, career-wise, from the so-called “prestige economy”, where journals are evaluated and ranked according to qualitative (perceived prestige) and, increasingly, quantitative criteria (rejection rates, impact factors), and where individual researchers are judged by their peers in large part on the number of papers and the journals in which the papers appeared.

But there are strong reasons to believe, and mounting evidence to that effect, that this prestige-based evaluation system, as deeply ingrained in academia as it may be, is deeply flawed on many accounts.

In addition, the “publishing oligopoly” of large publishers has skillfully used it to increase its stranglehold on scholarly publication and to justify its skyrocketing costs.

One can even argue that the current system conflicts with the freedom of choice in research included in AF’s basic definition. After all, giving prominence to the number of papers and restricting journals that really “count” towards evaluation to leading, high impact ones may in practice make researchers, if they hope to remain in academia, avoid research topics or methods that take too much time to produce results, or that won’t be considered by these highly selective journals.

Here we face a peculiar situation where, in the very name of AF, the research community hesitates to let go of, or significantly reform, a system that is far from optimal and that, quite paradoxically, may well limit the AF of individual researchers.

The Preamble of Plan S alludes to this prestige economy, by speaking of a “misdirected reward system” and by endorsing the DORA Declaration, which proposes that researchers are judged upon the quality and impact of their research, both multi-faceted notions not reducible to numbers and journal names. And while its principles don’t refer explicitly to this system, Plan S has certainly the potential to disturb, or even disrupt it, again to an extent difficult to assess.

Thus, one has to put into balance two consequences of the drastic change Plan S hopes to achieve.

On the one side, there could be drawbacks for some researchers, though one can think that under a new evaluation paradigm, many if not most of them would still see their valuable work recognized.

The importance of high-impact journals, possibly imperilled by Plan S, for assessing the quality of scholarship has also been highlighted, although studies (like this one) draw a more nuanced picture of the efficiency of the gatekeeping they provide.

On the other side, there are potential benefits to scholarship, in terms of efficiency (e.g., the overall burden of peer-review being reduced, publication delays shortened), and to society (e.g. fewer public funds diverted towards publishers’ revenues and profits, universal access to research results).

(2) The requirement to publish with a liberal licence

Plan S also requires that the main research results are disseminated under a licence with minimal restrictions (CC BY or similar). Since deciding the way in which one’s works are used may be part of “full freedom in publication”, this requirement may also be viewed as conflicting with AF.

True, some researchers don’t like to see their works used by others to “make money”; many are also reluctant to let others adapt them, fearing that this may endanger both the quality of scholarship and the author’s own reputation.

Again, we might wonder to what extent allowing researchers to decide how their works are used (this includes, ironically, letting publishers become the ones to decide) is beneficial to scholarship and society.

What contributes more to AF’s ultimate purpose: protecting works against uses researchers dislike or fear, or allowing those works to be easily used and adapted?

There are numerous studies and discussions on this topic, covering issues like fear vs actual risk, the fuzziness of the various usage restrictions, and their effects on potential users, that could feed the discussion (see for instance here, here and here).

In short, what I propose here is to reframe the role that academic freedom plays in discussions about the proposals put forth in Plan S, by assessing both their potential consequences on scholarship and the common good as well as the nature of their actual or potential conflict with AF, keeping in mind that the raison d’être of AF is, after all, to benefit scholarship and the common good.

While this leaves much room for interpretation and discussion, notably how we define quality in scholarship and what constitutes the common good, I believe it would help answer questions like: Is Plan S a sound, or reasonable proposal? How can it be improved and implemented in order to reach its stated goal, that is flipping to OA a sizable part of scholarly communication, while bringing benefits to both scholarship and society?

These are crucial issues, and I see a real danger that a strict, almost dogmatic stance on AF could cause Plan S to fall short of its potential to pave the way towards a less costly, yet more fair, full-OA scholarly publication system.

The author thanks Marie-Josée Legault and Richard Poynder for their helpful comments.


Marc Couture was until recently science professor at TELUQ, University of Quebec distance education arm. He views himself as a generalist, having over his 30-year career had his areas of interest and activity evolve from physics (his PhD was in optics and lasers) to the application of IT in science education, to intellectual property in academia and, above all, open access, for which he remains a dedicated and shameless advocate.


Clarke Iakovakis said...

Great article and an important reminder to keep the intent of academic freedom--benefits to the common good as well as scholarship--in the forefront of our conversations on the topic. As Richard indicates in the preface, this should lead to an attempt to reach a consensus definition of AF, including how we define "common good," and specification of the ways in which scholarship benefits it.

The author makes a compelling point that the current evaluation system itself restricts not just choice of publication venue, but also choice of research topic. However, I am interested to hear more about the line he draws between the implementation of Plan S and the reform of academic evaluation processes. In what ways will Plan S lead academic departments away from valuing high-impact journals? What is to keep them from merely shifting from high-impact toll access journals to high-impact OA journals? How does the economic and access model of the publication venue influence the ways in which researchers are evaluated? As the author mentions, faculty career advancement is contingent on the prestige they extract from their publications, and this is increasingly measured with reference to citation metrics of the journal and citations to the article itself. Thus the debate about academic freedom must go deeper than choice of publication venue, and extend into evaluation as well.

Richard Poynder said...

Thank you for commenting Clarke.

Leaving aside for a moment the issue of academic freedom and considering evaluation alone: it strikes me that the architects of Plan S (and indeed the OA movement more generally) have got things back to front.

If the intention was to get researchers to embrace open access, then surely the first step should have been to change the evaluation system in order to encourage them to do so. However, today (in Europe at least) we see researchers being given contradictory commands from funders and institutions: make your work open access but make sure you publish in high-impact, prestigious journals. This is contradictory because the vast majority of high-impact, prestigious journals are legacy journals, so it is usually only possible to meet both demands by means of either green OA or hybrid OA. Yet Plan S looks set to ban hybrid OA and make green OA nigh impossible (by demanding immediate deposit plus CC BY or equivalent).

This mirrors what appears to have become the consensus within the OA movement, where advocates argue that only gold OA is “true OA”.

The end result is that in Europe researchers are now receiving tyrannical and contradictory commands from funders and research institutions.

How can this possibly make sense?

Marc Couture said...


Clark and Richard: you both raise excellent questions. Let me first address Clarke’s last one: “How does the economic and access model of the publication venue influence the ways in which researchers are evaluated?”.

I would say that this should rather be viewed the other way around. The idea of personal reputation based upon publishing in esteemed and highly-selective journals is far from new, as Aileen Fyfe explains. But what we have seen in the recent pas is commercial publishers (and I include here large high-revenue generating non-profit society publishers like ACS and a few other) succeed in harnessing this “prestige economy”.

When APCs where first introduced in the early 2000s, and even when they had been envisioned in the still earlier discussions on the very idea of OA, in the mid 1990s, there was this idea of creating an efficient market (from the point of view of its consumers, namely authors or, most likely, institutions and funders taking the tab), where publishers would have to limit, and most probably lower, their prices to avoid being thrown out of business by more efficient competitors.

But this didn’t happen: the journal-based reputation system was too much ingrained, and publishers had clever idea: transforming the impact factor into a pervasive, powerful prestige indicator; introducing Big Deals that tied the fleet of all their journal (prestigious or not); enlarging them to (still more costly) schemes like Read-and-Publish; offering hybrid OA, that only took flight when funders (especially in UK) started simultaneously to require or favour Gold OA and put money in it without constraints as to the costs, significantly higher than other OA options.

As to your other questions, that touch the way Plan S could, or should favour, induce or bring a major shift in researchers evaluation, I will try if not to provide answers (no easy task, nor one that anybody can pretend to do alone), at least to help frame and clarify some of the main issues I perceive, notably in the various critical comments on Plan S I’ve read in the last weeks.

Before, a few words on the impact factor (IF) and its irrelevance or inadequacy for most of its current uses. Scientometricians and information science experts, among them Eugene Garfield, the very inventor of IF, have repeatedly shown – in the scholarly sense of the word – that all its uses beyond its original one (helping librarians make subscription decisions), and even outside of its original context of print publishing, are unwarranted. This is especially true of using it to evaluate individual papers or researchers. This is a message that must be repeated again and again, until it becomes part common knowledge in the research community.

Now, one has to acknowledge that this prestige economy, and the journal hierarchy associated to it must fulfil purposes deemed important by researchers. Any plan to disrupt it should therefore take these purposes seriously, but also question them using the same approach I propose in my post. To that effect, one must wonder if these purposes are important enough, or if the can be achieved otherwise, considering the deleterious effects of the current scholarly publication system on both scholarship and society.

I see at least four such purposes: evaluation, motivation, dissemination and quality of scholarship.

[to be continued]

Marc Couture said...


Evaluation. Institutions and researchers (individually or collectively) make decisions affecting the careers of individual researchers (hiring, tenure, promotion, grants). The respective roles of institutions and researchers vary among contexts and institutions, but researchers often play a decisive part. In doing so, they often have to judge many competing researchers, or a huge scholarly output from an individual, without necessarily being experts on the topics covered.

Counting the number of papers and weighting them according to journal prestige (or IF) is an easy, fast way to go, but certainly not the most fair and accurate. This what the DORA Declaration is all about. But what could be the alternatives? Plan S should lead the way here, by not just formulating wishes or suggestions, but implementing very concrete measures in its evaluation process, for instance asking candidates to submit – and reviewers to consider only – a limited list of contributions with an explanation of their relevance, importance and usefulness. Not doing so would be perfectly incoherent, as Richard rightfully points out.

Should it go farther, for instance requiring that institutions commit to DORA-like principles to be eligible? After all, funders already impose constraints to these institutions, for instance for ethical or financial considerations.

Motivation. Researchers need ways to see the quality of their research recognized. And the prestige economy covers a lot more than publishing: its currency also includes prizes and awards, invitations, nominations, media (social or traditional) visibility, and grants. Arguably, all these reflect, in principle at least, the quality and/or usefulness of their research.

Are these other sources sufficient, or should new ones should be found or created? Basing recognition on journal prestige, directly or indirectly (through criteria used to make decisions for awards or funding, for instance) is not the way to go. New metrics have been proposed, but there is still work to do to assess their limitations and, ultimately, define truly responsible metrics.

Dissemination. Researchers want their results to be known to those who can benefit from them. They usually think foremost of their colleagues working in the same domain. I’ve repeatedly heard of results having to be published in prestigious venues in order to be known or considered. I’m really troubled by this perspective, that seems to me a relic from a time when the primary way to check the literature was to leaf through the journals that departments or libraries subscribed to. Do modern researchers really miss, or discard a paper relevant to their subject just because it was not published in a top journal? If true, this would be a matter of real concern.

Quality of scholarship. Despite all its pitfalls, peer-review is still largely considered as improving the overall quality of scholarship. It does so both in a positive manner (pointing to authors errors or weaknesses in their papers as well as where and how they can be improved) and a negative one (filtering out manuscripts that don’t meet basic criteria of scholarship and/or communication, or that raise ethical concerns). The hierarchy of journals is often seen to play a crucial role in this regard. As I wrote in my post, there is mounting evidence going against this belief.

More fundamentally, one can wonder how the high selectivity of top journals, which imply rejecting perfectly sound and legitimate papers, but deemed unimportant, not original enough, or unfit for the venue, and seeing them published elsewhere, contributes to the overall quality of scholarship. What is important is to be able to have journals we can trust to both help improve the papers researchers submit and filter out the really problematic ones. Plan S could certainly help here, for instance supporting the various already existing initiatives that share this goal (and I don’t mean lists, of any colour, here).

Sylvain Ribault said...

The invocation of academic freedom is one of the most common arguments against Plan S (or academic publishing reform in general), and I am glad that it is demolished so carefully in this post. I have cited this post in my effort to discuss the rest of the open letter's objections to Plan S: