Skip navigation
SPECULATIVE DICTION

Open access and academic publishing – an interview with Dr. Martin Paul Eve

A frank discussion about open access publishing, Dr. Eve's own Open Library for the Humanities, and what future he sees for the academic publishing industry.

By MELONIE FULLICK | January 19, 2017

The current state of academic publishing is something we should all be thinking about, given that it’s a means of disseminating the knowledge generated by academic research — much of which is publicly funded yet inaccessible to the public. Publishing is also significant because of the key role it plays in academic careers, where it serves as a gatekeeping mechanism.

Changes to academic publishing both reflect, and contribute to, broader trends within academe; and they point to a range of critical questions. Within the contexts and constraints of established institutions and practices, what possibilities are opened up by recent technologies and new organisational forms? How could publishing change in (and differ across) the various academic areas and fields? How might the increased competition for academic jobs, and the smaller proportion of scholars with permanent positions, affect publishing and associated labour? Can we maintain the accessibility and quality of publications and the process of peer review while also ensuring compensation for the work involved? What about digital publication and circulation – will (paywalled) journal publishing eventually become obsolete?

This post explores some of those questions by way of an interview with Professor Martin Paul Eve, who is an all-round expert on, and active promoter of, open access academic publishing. Professor Eve is chair of literature, technology and publishing at Birkbeck, University of London, and the author of Open Access and the Humanities: Contexts, Controversies and the Future. Among many other things, Professor Eve also gave evidence to the U.K. government’s business, innovation and skills committee inquiry into open access, in 2013, and with Dr. Caroline Edwards he is a co-founder of the Open Library of the Humanities (OLH), in which 16 Canadian institutions are already participating. The following interview focuses on OLH and on the issues of access to knowledge (in the form of peer-reviewed research) that underlie its creation.

Dr. Martin Paul Eve.

For the benefit of readers who aren’t familiar with OLH, could you provide a bit of a description of what it does and how it works?

The Open Library of Humanities is a charitable, not-for-profit, open access publisher of peer-reviewed humanities research material that I co-founded with my colleague Dr. Caroline Edwards. The OLH currently publishes (or financially supports) 18 journals and in our first year of operation we made over 900 scholarly articles available at no charge to readers.

Unlike other open access publishers, though, the OLH has a different economic model. In a move to abandon subscriptions so that there is no cost to read academic research work on the internet, many publishers are turning to so-called Article Processing Charges (APCs); a fee to be paid by the author, their institutions, or their funders. This is not a fee to bypass peer review, but rather a fee to cover the publisher’s labour (and surplus/profits). It is a substitution for subscription revenue.

In our disciplines, however, it is far harder to see where this money will come from (Taylor & Francis for instance has an APC of about $2,000). Our model works differently. Instead of asking authors or funders to pay when work is accepted, we have over 200 libraries paying us an annual amount that looks like a subscription. The change though is that with the money from that fee, we make all our work openly available. So yes, even if libraries don’t pay, their academics can submit and read the work. We know, though, that this is not how libraries and universities think. Universities should exist for the good of society, not for exclusionary gain. The fact that so many libraries have supported this non-classical economic model so that research work can be broadly accessible is the proof.

You mentioned in another interview that this idea was coalescing at the end of 2012/early 2013, as part a broader discussion about scholarly publishing; three years later, OLH has been running for over a year and has participants at more than 200 institutions. How did you get started with a project of this size, let alone get it running within such a short period — and especially when the process involves so much consultation and buy-in from so many different groups?

I often wonder to myself: were I able to go back in time and warn my previous self how much work and how difficult the whole thing would be, would I still do it? I like to think the answer is “yes” because I believe in higher education and that research work should be accessible to anyone who is interested. But it is hard to say when speaking in such hypothetical terms.

We began by simply floating the idea online of creating something a bit like the Public Library of Science (PLOS) in the humanities. When I put up an initial web page, we had over 200 emails overnight from interested parties, so the demand seemed clear. We also realised early on, though, that we’d need to play the prestige game very carefully. The whole academic system of certification and accreditation is built on a symbolic economy of prestige. This makes it very hard for new players to enter the arena since they do not yet have a track record and many academics fear that they will lose out on career progression (or even jobs) if they do not play the game that hiring panels expect.

That said, some early supporters gave us a boost. We’re immensely grateful to David Armitage, for instance, the Lloyd C. Blankfein Professor of History at Harvard, who said of the OLH that “there is hardly a more important project in train for scholarship in the humanities today.” We also assembled committees of high-profile academics to ascertain what they felt would be needed to make an idea like OLH work. And we listened and we structured the platform around the advice we received. For instance, the first version of the OLH was going to be based on Article Processing Charges, as is PLOS. But the committees baulked at this and so we had to design a new, untested model for the ongoing economic support of academic journals. We also had to balance the desires of hard-line open access enthusiasts against a distaste for full open licensing among some of our disciplinary communities. We had to handle the more radical thinking of some technologists against the traditions of the humanities and the need to ensure their continuity.

We also had to work out how to establish and govern a charitable company; something with which I was totally unacquainted before we began the enterprise (I feel I should here mis-cite Bones McCoy from Star Trek: “Damn it Jim, I’m an academic not a businessman”). Generous support from the University of Lincoln; Birkbeck, University of London; and the Andrew W. Mellon Foundation made this all possible. Indeed, we spent those setup years travelling the world and telling academics, librarians, societies, and publishers what we wanted to do and gaining their trust, support, and feedback.

As a result we managed to go from rough idea at the end of 2012/2013 to a platform launch in September 2015. That’s not a bad timescale but we did nonetheless receive criticism that it was too slow (one person wrote, paraphrasing, that “the OLH has to-date not published a single article”). And this is the dilemma. If you spend time planning things, people criticize you for taking too long. If you rush into it, people criticize you for not planning.

You’ve said that “Cooperation can solve the budget crisis in scholarly communications where competition has failed us.” How does the OLH model (and maybe open access overall) function as cooperative?

The term “cooperative” has different meanings in different contexts. There’s a formal organizational structure called a “cooperative” that refers to co-ownership of an enterprise by its employees. I didn’t quite mean that but rather the underlying principle of working in concert rather than relying on free-market competition.

To explain this, I have to give a brief background to the economics of scholarly communications. Since 1986, according to ARL statistics, the cost for each institution of subscribing to the full range of journals that its academics and students require has risen by an estimated 300 percent above inflation. There are many reasons for this. One is that the mass expansion of higher education has led to greater publication volume. However, there is also the matter of vast profits and quasi-monopolistic practices among large publishers.

Indeed, the largest publisher of scientific journals, Elsevier (RELX Group), makes an approximate 35 percent profit. As I’ve pointed out before, this is double the profit rate of oil and pharmaceutical companies. Clearly – despite the efforts of Elsevier’s supporters to equate such exorbitant profits with charitable surpluses (the most frequent comeback I get to the preceding fact is: “even OLH has to make a surplus”… Yes, but it’s hardly the same as a 35 percent profit!) – something is going very wrong with the market here.

So why hasn’t free-market competition solved this problem? Well, goods in the scholarly communications market (academic journal articles and books, for example) are not substitutable goods. When an academic needs to read a specific piece of work, only that article or book will do because the content is, usually as a precondition, unique and novel. Companies like Elsevier who own the copyright on up to 40 percent of the supply in certain disciplinary spaces know that academics need the work they possess. So there is no competition in this sense; unique non-substitutable goods that are required for academic work are for sale at whatever rate the owner demands.

Open access and article processing charges do not necessarily resolve the crisis of library budgets, though. Hybrid open access refers to the conditions where publishers run a journal as a subscription entity but allow authors to pay a charge to make a specific article openly available. The problem here is that, without offsetting of the amount paid, the charges just mount and mount on top of subscription revenue. Because the people paying – the library – are not the same as the people submitting and reading – the academics – there is little sensitivity to such pricing from those who have the most agency within the system.

For me, this all amounts to a failure of scholarly communications practices to act like a market. And perhaps we don’t want it to be one. This is why I proposed cooperation as a way in which library acquisition staff can escape the failure of market competition. By pooling resources – as does the OLH by gathering small amounts from a moderate number of libraries – on a not-for-profit basis, we see a better use of budget than market competition was able to deliver. Indeed, across our 200 supporting institutions, we estimate that the approximate average paid was $1,000. This works out at just $55 per journal per institution. Alternatively, this is $1.10 per institution per article (many of the articles in our first year comprised cheaper back-content import though). However, if we flip this around and focus on the 118,686 unique users who visited article pages in our first year, this is just $0.008 per institution per reader. And this gets cheaper as more institutions join.

That’s why I advocated for a form of cooperation. Do not focus on the problem of “free riders” and the fact that even if one pays, others get access for nothing. Instead focus on what such non-classical economic thinking can actually deliver.

Some have described the academic publishing industry as a “racket, running on huge amounts of free labour while charging academic institutions, and individual researchers, exorbitant fees to read papers — and subsequently enjoying huge profit margins. How much labour do publishers actually perform and can they really justify charging as much as they do?

First, there’s a bit of a problem with referring to “publishers” as a homogeneous group. It spans the giants of Elsevier, Taylor & Francis, Wiley, Oxford University Press down to tiny university presses that barely break even. It runs from for-profits through charities. Publishers are many things.

Publishers perform, though, a variety of labour functions. There’s a (pretty eked-out) list of 96 things publishers do on the (pretty conservative) Scholarly Kitchen blog. However, I think that the labour function is mostly coordinating peer review, platform maintenance, typesetting, copyediting, proofreading, digital preservation, legal, accountancy, marketing (which is partly dissemination and partly business-model based) and general business overheads. List cultivation and actual editorial input might be placed here for some humanities disciplines, but whether this is a needed function or part of a business strategy remains hazy for me. Running OLH is a full-time day job for me and my colleagues at the moment. I tend to get up early and do my academic research reading and writing from 6:30 a.m. until 10 a.m. before then doing a full day of OLH work.

Publishers undoubtedly do things that we want and continue to require. However, I do not see why (as above) this should legitimatize extortionate levels of profit so that we can exclude others from reading research. One recent criticism I saw, for instance, was Daniel Allington comparing scholarly communications to a bar (as the old jokes always go) in a Twitter reply and deriding OA advocates for calling publishers “parasites.” Yet, Marx himself noted that “capital is dead labour” describing it as “vampire-like.” To call the profit-rates of mega companies a “racket” or “parasitical” is not to deride the labour of their employees or even the desired labour function. It is metaphorically to label such huge companies as a rentier group who use a variety of canny business strategies to feed off universities and achieve market dominance. Taylor & Francis, for instance, wrote in their annual shareholder report that their revenue stream was “dominated by subscription assets with high renewal rates, where customers generally pay us twelve months in advance. This provides strong visibility on revenue and allows the businesses to essentially fund themselves, with minimal external capital required.” They believe that “it is a uniquely attractive mode.”

One problem I see with the “Gold” OA model (explanation provided here) when author fees are involved, is that there’s an assumption that everyone publishing is affiliated with an institution and has access to institutional funding that can cover the cost. Many of those who want to publish are early career academics who may not have institutional support. To me it seems that the author-facing charges would inhibit access to publishing for a lot of scholars. OLH has avoided this by rejecting author fees; how does its funding model work? Would it be a viable option beyond the humanities?

Yes, I agree with your assessment here. I think this comes from the fact that many OA models have developed in the sciences, where much research work requires (and has) external funding that can budget for such charges. This is not universally the case, though. (And one criticism of OA is the fact that it grew from the sciences and is being applied elsewhere. I have long argued, though, that the benefits to the humanities of open research work are as great as those for the sciences; we just need a different implementation strategy.)

As above, OLH operates on a consortial funding model. Many libraries pay a relatively small amount into our central budget. We then make work openly available. There is precedent for this in arXiv, Knowledge Unlatched, and, to a lesser extent, the SCOAP3 purchasing consortium. It’s not a totally new idea. But it is the first time that a publisher has run its entire business model on this for front-matter serials publication.

The key part of what we’re doing, though, is participating in schemes to “flip” journals that were subscription to an open access model. In some cases, this involves an editorial board leaving a publisher – as did the editorial board of Lingua – while in others it involves us working with the university press to fund their journal; as with the University of Wales Press. We currently have two other university presses discussing flipping more substantial portions of their lists through the OLH model.

Certainly, though, there is no reason why what we’ve done could not be expanded beyond the disciplines in which we operate. However, we’d need to grow more rapidly and have more initial capacity if we were to run an Open Library of Science, for instance. Indeed, we’d need upfront funder investment to expand our consortium size at a swifter rate and to a higher level than we currently plan.

A lot of the discussion about “openness” has been about the sciences; your area is the humanities (and that’s the topic of your book). I have to apologise for not having read it yet — but what would you say are the specific challenges for the humanities (and perhaps for the social sciences as well) when it comes to open access scholarly publishing? How might those compare to what’s going on in STEM areas?

There are several challenges that are more pronounced in the humanities so I’ll just touch on a few here: the importance and prominence of monographs; the inclusion of third-party copyrighted material as an integral component of work (say in art history); and the different funding climate.

Monographs are tricky. They cost a lot more, they often have trade crossover, and they sometimes gain a broader audience by being available for sale in print. Certainly, I personally do not want to read 80,000 words on a screen for cover-to-cover reading. But that doesn’t mean that there isn’t merit in having books openly available; particularly as these can cost $100 or so per copy. In any case, a task force has been set up in the U.K. to plan a transition scheme to enable OA monographs by the mid-2020s. I’ve also sketched out some initial challenges in this space on my blog.

The third-party copyright matter is also a bit of sticking point. Without such inclusions work is often severely impoverished. When analysing images, if the image is not there, the evidence to back up the point is not clear; this is also an access issue. Getting clearance for images for unlimited online dissemination can be tough since galleries, libraries, archives, and museums are often not accustomed (yet) to such requests.

Finally, as above, the transition to OA in the humanities looks harder, since it is unclear whence the funding for APCs will emerge. Opponents of OA also argue that the journal costs in the humanities for subscriptions are lower than in the sciences, implying that they think the fuss isn’t worth it. Yet I have never worked at an institution that has subscriptions to every journal that I need for my research work, on the grounds of cost, so I’m not buying this.

None of this is to say that there aren’t specific challenges for the humanities but I prefer to think of the opportunities. Imagine a world where humanistic knowledge and inquiry – that is, research work about culture and humans – was available freely to every human on the planet. Is that not a goal worthy of overcoming a few hurdles?

What do you think are the most salient critiques of open access for academic publishing?

There are lots of very poor attacks on OA that I am tired of defending but the ones that are most salient seem to me to be the pragmatic critiques. How are we actually going to do this? How can we achieve that universal library for anyone who is interested in reading, without discriminating on their ability to pay? Usually, these critiques don’t come with answers. They point out the dire economic circumstances for the humanities, the difficulties of freeing subscription revenue etc. These are useful when they provide models to warn us of the challenges and dangers. But I am also interested in how we move from theory to practice. My book on OA (itself open access from Cambridge University Press) is designed as a theoretical basis on which the OLH or others can operate. It’s not a sales pitch for the model it’s just that I am not content with merely philosophising about the world; the point, after all, is to change it.

So much about the academic profession is shifting; I know this is always the case — and we are always acutely aware of our own time, particularly when it comes to technology and its effects. Given the nature of academic institutions, what changes to scholarly publishing (as knowledge dissemination) do you think might unfold within the span of your career?

This is the kind of question that can come back to haunt the interviewee and I’m very wary of such futurology! It’s also worth noting that things move at a glacial pace within the academy.

However, things I see happening already include a desire for open access among early career humanists. If at least some of them want it, then it is likely to happen. Data sharing in various ways is also on the up. I hope that we will see books available openly online for the good of all. I also hope that text and data mining provisions through the availability of semantically rich research articles can help us to more accurately understand the scholarly web and its contents.

ABOUT MELONIE FULLICK
Melonie Fullick
Melonie Fullick is a PhD candidate at York University. The topic of her dissertation is Canadian post-secondary education policy and its effects on the institutional environment in universities.
COMMENTS
Post a comment
University Affairs moderates all comments according to the following guidelines. If approved, comments generally appear within one business day. We may republish particularly insightful remarks in our print edition or elsewhere.

Your email address will not be published. Required fields are marked *

«
aborg