About a year and a half ago, in these pages, I posed the question: “Will the future of scholarly communication be pluralistic and democratic, or monocultural and authoritarian?”. I pointed out that significant players in the scholarly communication landscape are vigorously and explicitly working to establish the monocultural and authoritarian scenario, and suggesting that those of us who want to preserve pluralism and democracy in the system had better be prepared both to make our voices heard and to do the more difficult political work necessary to preserve a diversity of scholarly communication models.

A recent communication from the National Academy of Sciences, Engineering, and Medicine’s Roundtable on Aligning Incentives for Open Science (NASEM) brought home this reality to me again, and did so in a way that I think is worth examination and contemplation.

silhouettes of people with speech balloons,

To put the NASEM statement in perspective, imagine getting a group of faculty members to bring a resolution like this to your university or college president:

Jazz is perhaps the greatest musical invention of the 20th century. To a degree unlike any other genre of music, it combines rhythmic complexities and melodic influences from African traditions with the harmonic complexities of European traditions to create a musical whole that is uniquely American, broadly accessible, and in all ways greater than the sum of its parts. Given both the intrinsic qualities of this music and its cultural importance, we the undersigned recommend that, from now on, it be the kind of music taught primarily at our institution, while leaving open the possibility of instruction in other musical forms if needed.

If you balk at that statement, maybe it’s because you think jazz isn’t good music and isn’t worth the sustained attention and support of a university music program. More likely, though, you think jazz is just fine — you may even think it’s great — but you don’t think it’s the only great music there is, and therefore shouldn’t be treated as intrinsically superior to and more important than other kinds of music.

Now consider how you’d feel about bringing your president a statement like this one:

The subscription model is the one that spreads the costs and benefits of publishing most fairly and evenly across the scholarly ecosystem: those who want access to particular publications bear the cost rather than those who don’t, and authors get to retain their exclusive rights as content creators while never having to pay a cent for the publishing service — which means that grant funding can be used entirely to support research itself rather than being redirected to support dissemination. It provides for peer review, collaboration between researchers, and certification of research results. Given the merits of this model, we the undersigned recommend the creation of an institutional initiative supporting subscription-access publishing, while remaining flexible enough to accommodate other models where necessary.

Would a proposal like this make you uncomfortable? It would me. Not because I have anything against subscription-access publishing in principle, but because the proposal itself is so slanted and tendentious. It talks about real benefits of the subscription model, but studiously avoids any discussion of its downsides and costs — like the fact that it largely (though by no means entirely) shuts out those who can’t afford to buy access or who don’t have an institution that can broker access for them, and the fact that the monopoly characteristics of copyright lead to serious problems of price inelasticity in the access market (a fancier way of saying “If you want access to an article published in an Elsevier journal, you don’t have the option of buying it from any other source, which means Elsevier can charge more than they could if they had a direct competitor”). And it misleadingly refers to benefits like support for peer review, collaboration, and certification as if they were benefits of the subscription model itself, even though non-subscription publishing models can provide such benefits perfectly well, while also misleadingly oversimplifying the benefits it provides to authors as copyright holders.

OK, let’s look at one more statement, this one taken directly from the NASEM communication I referenced earlier:

Open scholarship is a key strategy for universities to fulfill their core missions of creating, disseminating, and preserving knowledge for the benefit of society. It provides transparency so that others can validate the quality, accuracy and reproducibility of research, thus building the public’s trust. It enables and expedites collaboration among researchers through sharing of data, methods and tools early in the discovery process. It promotes efficiency, by rapidly informing others of promising avenues of research as well as potential dead-ends… Presidents and provosts are encouraged to work with their academic senates to create an open scholarship initiative that promotes institution-wide actions supporting open scholarship practices, while remaining sufficiently flexible to accommodate disciplinary differences and norms.

Some readers may be bristling because by using the pros and cons of the subscription model as a sort of mirror analogy for the pros and cons of open models, I’m implying that the two models are morally equivalent. Allow me to correct that by stating my position more explicitly: I see the two models as morally equivalent. In both cases, real value is being provided to the world at a very real cost. Subscription and open access models spread the costs around in different ways; in all of these models there are individuals in the system who get something for nothing, and there are individuals who run the risk of being shut out due to a lack of resources. Some models require a shift of resources away from other worthy and important endeavors in order to provide free publishing services or free access (or both) to authors and readers; some models rely on what amounts an ongoing freewill offering from various members of the scholarly communication ecosystem. Open scholarship makes scientific findings quickly and broadly accessible to billions of people around the world — and as long as those findings are reliable and presented honestly, that’s a real benefit. But the lower the barriers to “publication,” the greater the risk that nonsense masquerading as legitimate scholarship will find its way into the bloodstream of scientific and scholarly discourse. And none of this even begins to address the issues of university patent development and technology transfer, both of which would be severely complicated (if not fatally undermined) by a comprehensive institutional turn towards open science and open scholarship principles.

In other words, none of these models fails to provide real and concrete benefits — and none of them is without serious weaknesses and flaws. This is the problem with picking one model or suite of models — such as open scholarship — talking up its benefits, discouraging discussion of its costs, and then urging campus leadership to adopt it as the model that should be universally encouraged (while leaving “sufficient flexibility” to allow others as well).

Of course, there isn’t a single kind of “open scholarship.” The question here is whether the spectrum of varying models of open scholarship, all of them characterized by immediate free access and unrestricted reusability, are always and inevitably morally and practically superior to models outside that category (ones that impose some degree of restriction on access and reuse) and should therefore be put forward as the consistently preferred models.

Allow me to venture my own proposal; unlike either of the examples above, this is a statement I would feel perfectly comfortable bringing to my campus leadership:

Every model of scholarship and publishing presents a mix of costs and benefits. Toll-access publishing models spread costs across the system relatively thinly and connect interest and access in a relatively rational way, and they avoid undercutting research by tying up grant funds in publication costs; scholarship that is “closed” in its early stages makes possible certain kinds of development and institutional revenue generation that are not possible with open models; etc. However, these models also limit access to content, delay the maximum exploitation of potentially valuable data, and locate fee-bearing services in a relatively noncompetitive area of the marketplace. Open models, on the other hand, create tremendous benefits in terms of access and reuse of data and content, making scholarship quickly and globally available and reusable. However, some of these models concentrate publishing costs in ways that put an increased burden on research-intensive institutions; other models marginalize researchers in less-privileged regions of the world, divert significant institutional resources from other priorities, or limit the institution’s ability to exploit locally-created intellectual property. Since no publishing model offers upsides with no downsides, presidents and provosts are encouraged to support a diverse scholarly communication ecosystem by working with their academic senates to ensure that faculty retain maximum flexibility to select the scholarship models that they believe will offer the best balance of local and global benefit, offering institutional support for those choices (including financial support where needed) without artificially narrowing faculty’s options.

Not everyone will agree with this proposal, of course. Some readers may feel perfectly comfortable bringing a proposal to their president or provost that says, in essence, “Scholarship Model X should always be preferred, due to its benefits (and let’s not talk about its costs and downsides).” In which case you should definitely bring that proposal to her. Be prepared, though, to answer tough questions.

Rick Anderson

Rick Anderson

Rick Anderson is University Librarian at Brigham Young University. He has worked previously as a bibliographer for YBP, Inc., as Head Acquisitions Librarian for the University of North Carolina, Greensboro, as Director of Resource Acquisition at the University of Nevada, Reno, and as Associate Dean for Collections & Scholarly Communication at the University of Utah.

Discussion

22 Thoughts on "Pluralism vs. Monoculture in Scholarly Communication, Part 2"

I don’t think it is fair to interpret Open Scholarship only as ‘gold OA access to publications’. I read the NASEM snippet not as a plea against publishers’ business models but as a call to support transparency throughout the academic process: be open about scientific methods, ensure the data is FAIR, use transparent and reproducible metrics to measure impact and enable anyone to use the work for further research. The way the final publication gets published may not even matter when all the scholarship behind it is open, transparent and accessible.

And btw: nonsense masquerading as legitimate scholarship is not a new problem and not related to Open Access. If anything, transparent open scholarship will reveal nonsense to be exactly that: nonsense.

I don’t believe I did “interpret Open Scholarship only as ‘gold OA access to publications.'” (In fact, I didn’t mention the Gold model of OA funding at all.) Can you point to something I said in my piece that seems to convey that narrow interpretation?

This is very good, Rick, and I believe it captures an essential point. But I fear that it’s practical effect will be small, as those not already predisposed to pluralism will view your comments through their own (monocultural) lens. One quibble: the opposite of “open” in scholarly communications is not “closed” or, in your phrase, “toll-access.” The opposite, if there is an opposite, is “edited.” Access is but one of many properties of scholarly communications, and arguably among the least important. This does not mean that open materials are not edited; in some instances they are edited very well. But editorial activity is epiphenomenal for open access publishing, whereas for the traditional model, it is the essence. I wrote about that here (to no avail): https://scholarlykitchen.sspnet.org/2018/09/17/how-traditional-publishing-works/.

Agreed re: the likely practical impact of this piece. Honestly, I’m not expecting to change anyone’s mind; my main goal is just to preserve some space in the public discourse for the pluralistic viewpoint. If no one defends that space, I worry that those who hold the viewpoint and would like to express it will become terminally discouraged.

Rick, I’ve enjoyed reading this as well as last year’s article and agree there needs to be a balance in access models. The thought that keeps running through my mind, is of how little importance all of this is to the vast majority of people in the world. The best estimate I’ve found is that 6.7% of the world’s population is college educated, 523M people. If we double that to 1,046B to cover any discrepancies we are still just over 1 billion people. On a day to day basis we know in heart that of that billion, most simply don’t access the scholarly research that’s being discussed here. So when we state that “Open Research makes scientific findings broadly and quickly accessible to billions”, while true in fact, it is certainly an exaggeration in useful practice and benefit to the vast majority of the world’s population. I think it would be better stated, Open Research makes scientific findings broadly and quickly available to the majority of the scholarly research community who can derive benefit from that access. My assumption being that most who want or require access are educated beyond the secondary level.

Thanks, John. I think part of the issue you raise here hinges on the variable meanings of “accessible.” It’s a simple statement of fact that “open research makes scientific findings broadly and quickly accessible to billions,” where “accessible” means “freely available to find and use.” But if by “accessible” you mean “comprehensible,” then you’re dealing with a whole different issue. As I pointed out a few years ago here in the Kitchen, all the openness and free access in the world won’t make an article of, say, Schenkerian analysis “accessible” to someone who doesn’t have foundational knowledge of music theory. (But obviously, this doesn’t reduce the value of that article for the significant though limited audience of readers who do have that foundational knowledge.)

Your proposal for a “diverse scholarly communication ecosystem” certainly sounds reasonable for striking a balance. I’m all for fairness and justice, but I’ve never heard good empirical evidence of why the free and/or open access system is really needed. The two common arguments are “rapid access” is better and “access for the poor” is better somehow. So here are two honest questions:
(1) Can you provide a real example of how the type of rapid access you are promoting created a key piece of knowledge to a single researcher so that he/she could then make a substantial contribution to science that wouldn’t have happened in due time under the status quo “slow” system?
(2) Can you provide a real example of a researcher working either outside an institution or in a poor country who did not have paid access to journals who made a substantial benefit to science because he/she had open access to research papers? Research is fairly expensive. Access to journals for researchers who don’t have access to research equipment and funding doesn’t sound like a convincing argument.
My field is psychiatry, but I’ll take examples on any topic from any field. No hypothetical examples. Need real-life examples.

Just to be clear, in case you’ve mistaken me for someone advocating for open access: I’m not doing that, any more than I’m advocating for toll access or any other particular model of access to scholarship. Because I can see both costs and benefits to every such model, I’m advocating for the preservation of diversity in the system.

But to answer your questions more directly: I can’t provide any examples of the type you propose. Of course, I also can’t provide a real-life example of someone who has benefitted from having fire insurance, or of someone who has avoided cervical cancer by receiving the HPV vaccine–or, for that matter, of someone who has overcome a phobia thanks to psychiatric intervention. As a medical professional, what conclusions do you think I should draw from this fact?

Yep, I understood you were not recommending only open access (OA). But recommending OA as one of the options because you believe it would have greater pros than cons for some hypothetical subgroup of researchers sure sounds the same as advocating for OA.
It would not be difficult to find real individuals who benefitted from fire insurance or psychiatric intervention. The numbers of beneficiaries from vaccines are easy to determine by comparing pre-vaccine to post-vaccine prevalence rates of diseases. Those seem like pretty easy examples to support.
To answer your question, here are the conclusions I draw. Advocating for OA seems like a belief system toward helping some hypothetical disadvantaged groups in search of true facts to support it. OA is being recommended for some hypothetical subgroup of researchers on some topic in some field or some country, but advocates can’t tell us who they are. You’re not the first person I’ve asked who can’t tell me. Maybe said subgroups don’t exist? If anybody else reading this thread knows the answers to my two original questions, please chime in.

Michael, you seem to be shifting the goalposts. Are you asking for aggregate data that would support the argument that OA models offer concrete benefits to science in addition to their costs, or are you looking for real-life examples of individuals who have benefitted from OA and/or provided a concrete benefit to science thanks to OA? Either one of those would require me (or anyone else you asked) to do a bit of research, but it could be done. I guess I’d be happy to help you with that research if you’d like, but I would imagine you could do it yourself if your interest extends beyond asking someone else to provide the information for you.

I notice also that you failed to answer my questions, which were (admittedly) designed to illustrate the frivolousness of your questions to me. In other words, neither of us could provide concrete answers to the others’ questions right off the top of our heads. You seem to believe that there’s a conclusion to be drawn from the fact that I’m “not the first person (you’ve) asked who can’t” do that, so again I ask: what conclusion do you think should be drawn from this fact?

I asked for “a real example.” Just one individual example. Asking for facts to support a theory is an expected part of the scholarly process. Whenever a new entity is believed to exist, whether it’s a new medical disease or a new social problem, it should always begin with one individual case report. If supporters of the entity can’t produce a single, real individual case, there’s no need to call for bigger studies or bigger solutions.
I already answered what your question about the conclusion I draw: “Advocating for OA seems like a belief system toward helping some hypothetical disadvantaged groups in search of true facts to support it. OA is being recommended for some hypothetical subgroup of researchers on some topic in some field or some country, but advocates can’t tell us who they are.” The shorter version: there is a theory that OA solves some problem, but no facts to support it.

If it’s helpful at all, here’s a recent example where open data was essential to a piece of research (note, the post is about “open science”, not just “open access”): https://www.nature.com/articles/s41598-021-93690-y
I can also provide anecdotal evidence for the utility of open access — at my former place of employment, we created an important policy for the business based on several open access papers that were published on the subject. Because I was not at a subscribing institution, I would not have been able to read those papers, nor implement the steps they suggested be implemented without those papers being freely available to me. Given that one of the main goals of governments funding research is to drive economic development, this would seem to be exactly the sort of success story you’re seeking, and one that is exceedingly common.

If supporters of the entity can’t produce a single, real individual case, there’s no need to call for bigger studies or bigger solutions.

So, by that logic, if someone who believes fire insurance is a good idea can’t come up with a single example of someone they know of who has benefitted from having fire insurance, there’s no need to investigate further. Clearly, there’s no evidence that fire insurance benefits anyone. Right?

If so, then OK: let’s posit that the benefits of OA are purely theoretical and haven’t been demonstrated to benefit science in the real world. In that case, we should then ask the same questions about toll access:

1. Can you provide an example of the toll-access system providing a key piece of knowledge to a single researcher so that he/she could then make a substantial contribution to science that wouldn’t have happened if the information had been made available to them freely?

2. Can you provide a real example of a researcher working in a well-funded institution or in a rich country who had paid access to journals and who made a substantial benefit to science solely because access to those journals cost money?

If you can’t provide such examples, then by your logic, shouldn’t we conclude that charging people for access to scholarship is just as unproven a system as providing access freely is?

(Incidentally, this conversation reminds me of one I had a few years ago with a guy who asked me “Can you provide scientific proof that peer review actually works?”. I was stumped, partly because it’s not reasonable to ask someone to provide “scientific proof” off the top of their head, but also because I couldn’t figure out how a scientifically valid study to prove the efficacy of peer review could even be designed. When I thought about it more later, I realized that the best answer I could have given him would probably have been “Well no, beyond the fact that in my many years of peer reviewing I’ve never yet seen a paper that didn’t have a problem that needed to be fixed, and in all my years of writing peer-reviewed research I’ve never yet had a paper published that wasn’t improved by the peer review process. I don’t know if that constitutes ‘scientific proof,’ but it’s certainly empirical evidence.”)

I’m not responding to your fire insurance, vaccine, and phobia questions because I don’t see the relevance of them. If you can explain why those examples are more informative than talking straightforwardly about open access, I’m happy to talk about them.
My view is that your “same questions in reverse” are not the right questions. It’s not relevant whether a researcher benefitted from a subscription journal article that wouldn’t have happened if the information had been made available to them openly. My view is that a subscription journal makes good sense as a business model. The consumers who want the products (journal articles) pay the businesses that produce them, just like any other business. The OA model wants some people to pay for the products so that other people receive the products for free. I haven’t heard a good argument about why dissemination of science knowledge should be subsidized for those who can’t afford it. Again, OS sounds like an ideology that believes it is somehow more fair or useful, but it’s not fair to the people doing the subsidizing and I’m not convinced it’s more useful to science.

Reply to David Crotty:
Thank you for the link to the article. For those who didn’t go read the article, it used publicly available seismological, hydroacoustic, infrasonic, and radar remote sensing data to estimate the explosive yield of the accidental ammonia nitrate explosion in Beirut, Lebanon. The information the authors needed did not come from research papers.
Going back to my two original questions: (1) I asked for a real example (a single, individual case) of how the type of rapid access provided by OA created a key piece of knowledge to a single researcher so that he/she could then make a substantial contribution to science that wouldn’t have happened in due time under the status quo, subscription model, “slow” system. In this Beirut example, the aim of the investigators was to show proof of concept that they could estimate explosive yield from “a novel approach” of public, remote sensing data. The investigators never articulated a clear conclusion of how this represented an important new piece of scientific knowledge.
Even if it did represent important new knowledge, it didn’t address the concern that publication of research is too slow. The aim of these investigators didn’t depend on research studies for their source data.
In addition, this example does not support the idea that open science would produce knowledge at a more rapid pace. The duration of time from the explosion to publication was eleven months. The explosion occurred August 4, 2020. The paper was published online July 8, 2021. At least a dozen videos of the actual explosion from different angles were freely available on YouTube in less than a month after the explosion. Workers were allowed to inspect the blast site the same day of the explosion. I’m no explosive expert, but it seems an explosive yield could have been estimated accurately enough from day one.
(2) I asked for a real example of a researcher working either outside an institution or in a poor country who did not have paid access to journals who would have made a substantial benefit to science if only he/she had open access to research papers. This example was published by researchers from the Federal Institute for Geosciences and Natural Resources in Hanover, Germany, the German Research Centre for Geosciences in Potsdam, Germany, and Christian Albrechts University in Kiel, Germany. I don’t know what arrangements those institutions have with journal publishers, but it seems moot; those researchers didn’t need journals to perform their study. I know many German universities have cut some ties with journal publishers who refuse to offer open access, but if these institutions lack access, it’s probably not because they can’t afford it.

Your place of employment example is difficult to judge without details. It sounds like having access to open access articles certainly made your task easier to craft your business policy. Most authors, however, are willing to share their articles when a peer sends them an email. I suppose you may have tried that and authors were not responsive. In my experience, unresponsive authors are rare. If authors were unresponsive to you, details might be interesting.
I suppose you could argue that their papers were only knowable to you at that point in time because open access made them knowable more rapidly than if published by slower subscription journals. That could be true, but it’s not clear that the time difference was important to your business. If it truly was a crucial time difference, how crucial was it? I may stipulate to your claim that it was crucial, but I would want to quantify it. If supporters of OA want to upend a reasonable subscription model, I would want to quantify it a bit more so we can all be informed about when, where, and just how important this might be.

I’m not responding to your fire insurance, vaccine, and phobia questions because I don’t see the relevance of them. If you can explain why those examples are more informative than talking straightforwardly about open access, I’m happy to talk about them.

As I explained before, the purpose of those examples was to illustrate, by analogy, the faulty logic underlying your question.

As for whether the subscription makes sense as a business model: it sounds like you and I agree on that. I believe it does. I don’t think it’s a perfect one (I’m not aware of any perfect business models), but I think it’s a generally logical one that mostly does a good job. Of course, it doesn’t do everything well. There are various OA models that, in my view, do some things better than the subscription model–but they also do some things worse. That’s why I don’t believe it makes sense to pick one model–any model–and say “This model is the only one authors and publishers should be allowed to use.”

(And by the way, asking the same question about the subscription model that you proposed for OA isn’t “asking the same questions in reverse.” It’s applying the same logic consistently.)

I think the point about patents and technology transfer are worth further exploration. There’s great movement toward making the stories written about research results open to all, but still seemingly little effort to make the actual results themselves openly usable. I suspect this is because doing so would (at least potentially) take money away from universities and researchers, rather than taking it away from much-hated publishers.

I wrote about this a while back here:
https://scholarlykitchen.sspnet.org/2013/08/06/is-access-to-the-research-paper-the-same-thing-as-access-to-the-research-results/

One unintended consequence of requiring true openness would be to drive a significant number of researchers out of academia and into the private sphere, where they would be able to reap the benefits of their discoveries.

Thanks for this article, Nick. It’s the kind of piece that makes us start the important conversations that help us avoid Manichean takes on the topic.

I just wanted to mention that I don’t think it is fair to describe the International Science Council as a “player that vigorously […] [works] to establish the monocultural and authoritarian scenario” by linking to its “Seven principles for scientific publishing”.
I want to cite some sections of the publication to illustrate my point:
– Principle 1 states that “Authors, regardless of their circumstance, whether funded or not, should not be denied access to the process of publication on the grounds of inability to pay.” And later on: “Scientific publishing should enable participation from less favoured individuals, institutions and regions […]”. This to me shows that there is a effort to acknowledge the existence of some models’ drawbacks in pricing out communities, and the importance in avoiding them.
– Principle 6 says: “The disciplines of science – natural, social, engineering and medical – and the humanities, tend to have their own principles of publication that reflect the history, values, cultures and practical norms of work of the discipline. They are individually valuable means of expressing contributions to learning and knowledge and part of the strength of each scholarly community’s inquiring, sceptical and analytic mindset.” This shows they are open to some diversity in publishing practices.

In regards to the lowering of the barrier to publication: I believe it would naturally exist hand in hand with increased transparency and accountability, public reviews, and the emergence of new kinds of metrics that would be used to judge the credibility of the authors, the publisher, the reviewers and the references the research is built upon. The historically dominating model hasn’t necessarily kept “bad research” out of publications (or stopped its spread), and the emerging models of Open Scholarship will not either, but they might very well offer further tools to weed it out more effectively. And very importantly: they would allow valuable experimental results to be published instead of being deemed not noteworthy enough in the traditional publishing process, contributing to build a less biased body of experimental data.

Stéphane, you’ve demonstrated that the ISC recognizes drawbacks to OA models and calls for those drawbacks to be mitigated, but you haven’t demonstrated that the ISC recommends the preservation of a diversity of models. Can you point to statements from the ISC that would indicate that organization’s openness to non-OA publishing models? If not, then I think my point stands: saying that we need to do things to mitigate the negative impacts of OA isn’t the same thing as saying that we should allow models other than OA to remain in the system.

As for the effectiveness of barriers to publication: it sounds like you and I agree that there are pros and cons to lowering those barriers. That’s my point: the imperfection of both “traditional” publishing and newer, more open models suggests that we need a scholarly communication ecosystem that allows multiple models to coexist.

Rick, how would your well-reasoned and thoughtful article change if the units of analysis were open educational resources (OER) versus for-fee textbooks and courseware, delivered in inclusive access and equitable access programs? I realize that this is the “Scholarly Kitchen,” not the “Learning Kitchen,” but I am running the thought experiment through my mind …

I think the arguments I’ve presented here would translate very directly. There are pros and cons to the OER model, and to the traditional textbook-buying model, and to the inclusive/equitable access textbook models. Rather than picking any one of those imperfect models and saying “This is the only one that should be allowed,” I think students, faculty, and institutions should continue having the option of picking the model (or, more likely, models) that work best for them. “Let a hundred flowers bloom and a hundred schools of thoughts contend,” as the Great Helmsman once (briefly) said.

Comments are closed.