Open Access and the Future of Scholarly Communication
Column Editor: Sanford G. Thatcher (Director, Penn State Press, USB 1, Suite C, 820 N. University Drive, University Park, PA 16802-1003; Phone: 814-865-1327; Fax: 814-863-1408) firstname.lastname@example.org, www.psupress.org
Column Editor’s Note: This article is the text of a talk given at the University of Rochester on September 23 under the auspices of its Andrew Neilly Lecture Series, which on this occasion was meant to help celebrate the 20th anniversary of the University of Rochester Press. The author would like to thank the press’s director, Suzanne Guiod, and the library’s director, Susan Gibbons, for extending the invitation and being such congenial hosts. — ST
As a term of art, “open access” (OA) has been widely used for not even a decade yet, probably gaining its popularity after a series of declarations identified with the cities of Budapest, Bethesda, and Berlin made it a familiar phrase to many early in this new century. It came into being in roughly the same period as the word “open source,” with which it is sometimes unfortunately confused but with which it has in common the inspiration of a democratic ideal of open and free communication and the sharing of knowledge.
As any librarian in this audience can tell you, the original and still sustaining impetus behind the open-access movement was the challenge for academic libraries that the costs of journal subscriptions posed to their budgets as serial prices increased at a rate greater than inflation, sometimes much greater. There are many reasons for this, and the pressure on commercial publishers to keep profit margins high is only one of them, though perhaps the most prominent in the rhetoric of OA advocates. Other reasons include the growing productivity of scientists (and the problem exists in its most extreme form for what are known as STM journals) as research funding increased dramatically after the end of World War II and the careers of scientists, especially in universities, came to depend on the number of publications even more than their quality; this growth led not only to the proliferation of journals but also to an increase in the number of issues published and in the length of each issue as it had to accommodate more articles per issue. On top of these exigencies of sheer growth came the temptations of new technology as the transition from print to electronic added substantial extra costs for publishers to provide all kinds of services beyond just the delivery of content online. Many journal publishers, including even small ones like our Press, now use very sophisticated editorial management systems that offer a wonderful resource for both the academic editors and the publishers’ staff (though it needs to be noted that many scholars who edit journals are not eager to learn how to use these systems properly and have to be coaxed, if not even sometimes forced, to go through the training). There are open-source alternatives like the Open Journals System of the Public Knowledge Project that are available for use, but in my experience they are not at the level of sophistication that the commercially developed products are, and may never be. At any rate, whatever one may think about corporate greed, it is not the only factor that explains the spiraling cost of serials acquisitions for libraries.
Where I began my publishing career, at Princeton University Press, this was a problem only indirectly since Princeton did not consider itself a major publisher of journals, having at that time only four in its portfolio. But the evidence of the impact of the so-called serials crisis on the breadand- butter monograph publishing that was that press’s core mission soon became clear. I trace my own awakening to the crisis and its baleful effects on scholarly book publishing to internal discussions at Princeton that led to the publication of a series of articles in the journal called Scholarly Publishing (later expanded to Journal of…) over a two-year period between April 1972 and April 1974. The titles of these articles were, in succession, “The Impending Crisis in University Publishing,” “The Crisis—One Year Later,” and “The Crisis—Is It Over?” Not long thereafter appeared an NSF-funded study that brought this crisis to the attention of librarians in a most dramatic way. In their report published in 1975, librarians Bernard Fry and Herbert White found, for the period 1969-1973, that the ratio of book to journal expenditures in the largest academic libraries had dropped over that five-year period from better than 2 to 1 to 1.16 to 1 (Fry/White 1975: 61), with every expectation that this trend would only get worse — as, indeed, it did. Fry and White’s prognosis for university presses was particularly gloomy: their situation, they said, “can be described, without exaggeration, as disastrous. Already heavily encumbered by operating deficits…, university presses appear…to be sliding even more rapidly toward financial imbalance” (Fry/White 1975: 11).
How, in light of this prediction of impending disaster, could Bill Becker, the CFO at Princeton University Press and author of that final article in the series, consider the crisis possibly to be past? His explanation was that, “except for the smaller ones, presses for the most part have managed to survive their financial difficulties quite well by making a host of adjustments, including radically increased book prices, substantially lower discounts, economies achieved in book production costs, slashing staffs, publishing more books with sales potential and fewer which cannot pay their own way, special inventory sales, and so forth.” Still, Becker went on to wonder, how much more could such methods be used without becoming at some point self-defeating? Ominously and — as we can now see with the wisdom of hindsight — presciently, he ended by pointing to “the increasing danger that presses will turn more and more to publishing books on the basis of saleability rather than scholarly merit.” And, while noting the temporary mitigating effects that a generous grant from the Mellon Foundation to presses for publishing books in the humanities might have, he asked: “But what then?” (Becker 1974: 202)
Indeed, one might ask that last question also in the wake of further grants that the Mellon Foundation has generously provided to fund other book-publishing projects, including (through the AHA and ACLS) the Gutenberg-e and Humanities E-Book Projects and, even more recently, cooperative ventures among groups of presses or between presses and other institutions on their own campuses to publish monographs in fields where—to borrow the language made famous through the Gutenberg-e project — books are “endangered species” (a phrase coined, by the way, by my mentor in publishing, Herbert Bailey, Jr., director of Princeton University Press at the time, and then put to effective use later by Robert Darnton, who had served on the press’s editorial board during this period and had many conversations with me and others at the press about the “crisis”). As I wrote in “A Post-Mortem for Gutenberg-e: Or, Why Ross Atkinson’s Dream Is Still a Dream” (Against the Grain, January 2009), none of us who served on the advisory board for these eBook initiatives before they were proposed to Mellon for funding believed that they could be self-sustaining beyond the point at which the original funding had been exhausted. Even though the Humanities E-Book Project claims to be paying its own way now, it is doing so only because of the addition to the core new monographs of a large corpus of scanned older titles, which make the overall package attractive enough to libraries to subscribe. In a mostly quite positive review of HEB in this month’s issue of the British online Reviews in History, Winthrop University library dean Mark Herring raises this question of sustainability over the long term even while recognizing that HEB is currently self-supporting. This is a problem, he says, that “must be worked out. Whether it will or not remains a mystery, and that mystery may well prove to be the undoing of many digital sources, not to mention the whole eBook enterprise.” He goes on to remind us that there is a second major problem that has mostly just been swept under the rug for the time being: digital preservation. People don’t want to talk about it these days because it is “unglamorous.” Yet “we have not solved the problem…, and it does not look like we will, at least in the short term. When I raised this matter at a cyber-conference it was not greeted with much more than ho-hum, not-so-polite throatclearing and a general, ‘Next question.’ Finally, a technician stood up and said he wanted to allay my fears. ‘We’ll fix all that,’ he said. ‘Consider it done.’ That was ten years ago.”
What has happened, though, is another development in technology that has occurred mostly out of public view and that has effectively bought some more time for university presses before they have to meet the grim reaper. Just in the same way that the changes university presses initiated in the early 1970s kept the sinking ship afloat for another three decades, digital printing came to the rescue in the late 1990s just as presses were beginning to despair of keeping their heads above water for much longer. What Cambridge sociologist and publishing entrepreneur John Thompson identified as “The Hidden Revolution” in his state-of-the-business study titled Books in the Digital Age (Polity, 2005), which I highly recommend to you as the most up-to-date and comprehensive treatment of academic publishing (even though Google only got mentioned in one footnote because its Book Search project was only just coming onto the scene when Thompson’s book went to press), turned out to be much more important to the finances of university presses than any revenue streams from eBook sales, which still are minuscule (in the 1-2% range for most presses, if even that much). You are all familiar with digital printing in its most basic form, as the photocopy machine, which has been around for decades now (and was the single most controversial element in determining how the Copyright Act of 1976 came to be written). Ever evolving into more sophisticated forms, this technology became revolutionary for the industry when it was integrated into the book distribution chain in the late 1990s when the major book wholesaler Ingram started up a new business under the name of Lightning Print, later changed to Lightning Source. The genius of this linkage was to solve academic publishers’ two major problems overnight: inventory and cash flow. Under the system prevailing almost since the days of Gutenberg, publishers always faced the temptation of decreasing unit costs by printing more copies, and ever the optimists, editors hoped against hope that their books would break out of the pack and become best sellers, against all odds (because the 80/20 rule pertains as much to book publishing as to other forms of commerce), persuading their marketing departments, which knew better, to go along with print runs that, as reality usually proved, would satisfy demand for the next 50 or 100 years instead of just the three years envisaged in the editors’ projected budget. So, warehouses bulged with excess inventory, which locked up precious capital that otherwise could be used for publishing new books coming down the pipeline. In one fell swoop, digital printers offered salvation: print runs could go as low as a single copy and still make economic sense. Lightning Source built its business originally on older backlist titles that publishers wanted to keep in print and wooed university presses by absorbing the costs of digitization in many instances. Its alliance with Amazon.com expanded its scope beyond just the retail bookstores that Ingram serviced and opened these older titles to a whole new audience — what came to be known, as Chris Anderson dubbed it, “the long tail.” A few years ago, Lightning Source set up shop in the United Kingdom as well, providing an opportunity for publishers like Penn State to make a decision, as we recently did, to supply that market entirely through POD rather than warehousing books in England, as we had done for many years. (This arrangement works for all of our books except highly illustrated art books for which the technology has not evolved yet to the point of being able to reproduce images at the level of quality expected in that field.) Even more recently, Amazon bought out the POD vendor BookSurge, pressuring publishers to use that company as the sole source for POD if they wanted to continue having their books listed on Amazon (a tactic that has spawned at least one law suit, from an association of small presses). And just earlier this month Lightning Source announced a new agreement to provide POD services to the giant French publisher Hachette Livre, which will not only make POD available for its own constituent units but for third parties as well, and Google made it known that two million of its digitized books could soon be printed out as paperbacks using the Espresso Book Machine, another version of this technology that is relatively cheap and can even be installed and used in libraries, as it has already in a few. So, this “revolution” is showing amazing signs of growth, no longer so hidden from the general public, who may come to appreciate that, despite Jeff Bezos’s confident predictions, the printed book is not about to die anytime soon, if ever.
It is just this revolution that has given hope to some of us in the university press community that we can eat our cake and have it, too. Many of us have long cherished the ideal of open access as basic to our missions as scholarly publishers. We share the same values as the rest of the academic community; we want what we publish to reach as many people as possible throughout the world at minimal expense. That is one of the most important reasons why we work for university presses and not for commercial publishers for whom, no matter how much they want to disseminate knowledge also, are ultimately driven by the imperatives of the market economy and the need to satisfy shareholders. At the same time, like our commercial counterparts, we run businesses that are obliged to pay their bills. If you want to know the single dominant reason that the AAUP differs from the ARL on most copyright issues, that is your answer. Our parent universities, so far at least, have not shown themselves willing to pay all of our costs up front so that we can cease worrying about who is stealing “our” property. (I put “our” in scare quotes because university presses, with a few rare exceptions, do not themselves own any intellectual property but simply manage it on behalf of their parent universities.) When we see state systems like Georgia’s play fast and loose with interpretations of fair use that allow for virtually unrestricted coursepack copying, we get alarmed because such practices, if they were to become widespread, would dry up a pool of revenue that all of us depend on today to help meet the goal of covering about 90% of our operating costs. Sure, this income may not much exceed 5% of our overall budget, but if any of you have been through exercises when you have to cut that much out of your budgets in already lean and efficiently run operations, you may appreciate more why we rush to use copyright law to defend our interests, which are defined for us by the budgetary strictures of our parent institutions.
But I digress. What I want to emphasize here is that POD affords an opportunity for presses to experiment with open access without losing their shirts economically. The National Academies Press pioneered in this experimentation when it began posting all of its books online beginning in the mid-1990s. It invented a mechanism that allowed people to browse an entire text online, using sophisticated search tools, but imposed certain constraints on their downloading and printing out the whole book. The dpi was fine for reading on screen but, when printed out, looked like old newspaper print, even harder on the eyes than prolonged reading on screen. And it required users who wanted a large number of pages to press a button to print each page, making this a very laborious and time-consuming task to get a print copy of a long book. What NAP discovered, and had hypothesized in advance, was that the online accessibility of the books would provide an analogue to the experience of browsing in a bricks-and-mortar bookstore, which many people like to do before making a decision to purchase. As a result, NAP experienced increases in sales of its books, compared with sales it had experienced for comparable books in the past, and besides POD versions, NAP also sold its books as PDFs, though this constituted only about 15% of overall sales, according to NAP’s Michael Jensen, who quoted this figure to me in a recent email. (There were also some problems with unauthorized superdistribution of some of these titles, I understand, especially the more trade-oriented books under its Joseph Henry Press imprint.) NAP thus had the happy outcome of being able to increase or at least sustain its revenue while being able to fulfill its mission of disseminating knowledge for its constituent scientific groups ever more widely throughout the whole world.
Envy can breed opportunity, so when we at Penn State were making plans to launch an Office of Digital Scholarly Publishing as a joint operation of the Press and the Libraries, one idea that came to mind immediately — with the extra stimulus provided by interested faculty from two departments — was to revive the dormant Penn State Studies in Romance Literatures, expand its scope, and relaunch it online as the Romance Studies series, following the model that NAP had provided a decade earlier. We did not know, of course, whether what had worked for the sciences would work for the humanities, so one goal was precisely to see if the combined online/POD approach could work in other areas. Another motivation for Penn State was to enable an extension of the DPubS open-source platform, which Penn State had joined with Cornell in developing further under a Mellon grant, from journal publishing into monograph publishing. The Libraries decided not to follow exactly the model used by NAP. Instead of the constraints on dpi and printing one page at a time, our site provides each chapter in the book as a PDF, with half of the chapters downloadable and printable at normal 300 dpi while the rest can be viewed online but not downloaded or printed, thus preserving the incentive to buy for some readers who might want the convenience of having the entire text in an affordable paperback format. I wasn’t initially happy with this decision, because I admired the NAP approach so much, but over time I reconciled myself to it and I don’t believe it has ultimately made any practical difference. Authors whose books get published in this series have the best of both worlds: they get a regular printed book that they can give to their P&T committees, their friends, and their families (and that the Press can send out to review media that are as yet still reluctant to assign people to review books just online), while anyone in the entire world with Internet access can consult their book, read it all for free online, or assign chapters to students at no charge, thus expanding their readership exponentially beyond the number who could find a copy in a few hundred academic libraries, mostly in the U.S. What scholar would not want to have a printed book to put on the shelf and at the same time have it exposed to readers all over the world at no cost to them?
This is, I think, the wave of the future — if it proves financially feasible. Whether it will do so remains to be seen. The early evidence is that we can sell almost as many books in POD form through this site as we did the regular offset-printed books from the series that was published during the decade of the 1990s. We had expected, and results have shown, that with the availability of the books in this series in either cloth or paper formats from the outset, sales of the cloth edition would not likely exceed 100 copies. University presses had come to learn by the late 1990s that more academic libraries were instructing their vendors to fill orders with paperbacks if published simultaneously, and many of our presses had therefore adapted by delaying publication of a paperback by a year or more. In this way we could still sell 300-400 copies of a book in hardback, a sufficiently large number that we could also have this run printed by traditional offset (since printers using that technology had reduced their prices in competition with the new digital printers) while using digital printing for the second stage of a book’s life-cycle in paperback, until sales dropped low enough to put the book into Lightning Source’s pure POD system. We don’t as yet know, however, how low we can go with paperback sales until we reach the point of no return. Experimenting with different pricing schemes is one of the aims of our Romance Studies series, and we used the opportunity provided by a near perfect test case, with two early books in the series both about women writers in 19th-century France by authors of the same academic rank, to see if differences in the cost between the hardback and paperback editions, i.e., the differential between the two prices, would affect sales very much. It didn’t make a huge difference, but interestingly the book that had the higher list price for the hardback and lower list price for the paperback outsold the other in both formats (85/308 for the book priced at $65/$25 and 64/143 for the book priced at $39/$30). But we also realize that there must be an upper limit to the pricing of a paperback before its sales are harmed. We just don’t know yet quite where that limit is. But when you consider that back in the 1960s university presses were regularly selling 3,000 copies of almost every monograph in hardback, and that even in the mid-1990s we were still able to sell 200 or more hardbacks simultaneously with 800 or 900 paperbacks, it is rather amazing that we are still in business when sales for a typical monograph now, even when the paperback is delayed, are about 300- 400 hardbacks and not more than about 400-500 paperbacks on average. The numbers are getting so low that every extra efficiency we can muster will not suffice to stave off the inevitable, barring some new technological innovation that can come to our rescue once again. And, as we all should know by now, trying to turn ourselves into mini-Random Houses is not the answer: just ask all those presses whose budgets were devastated this year by massive returns from the cash-strapped bookstores.
Open access in its most full-blown form, where there are no copyright or other restrictions on reuse of academic work of any kind (except, under the most popular Creative Commons license, “commercial” use — whatever that means), will come about only if universities decide to step up to the plate and move to Gold OA for monographs, as they seem to be beginning to do for journal articles, as evidenced by the just recently announced Compact for Open-Access Publication whereby Cornell, Dartmouth, Harvard, MIT, and UC-Berkeley commit themselves “to the timely establishment of durable mechanisms for underwriting reasonable publications fees for open-access journal articles written by its faculty for which other institutions would not be expected to provide funds.” This is of course just a promissory note, to be redeemed at some future point when these universities, perhaps joined then by others, actually do set up such “durable mechanisms.” But at least it is a step in the right direction and constitutes a recognition of the responsibility for universities to assume this burden at an institutional level and not just let each faculty member fend for himself or herself in paying the required fees. Another key term in this Compact is “reasonable.” One wonders who gets to define what is “reasonable” as a fee for any given journal, and how that is determined, especially in the absence of knowledge of the real costs the publisher incurs. (There is, I should note, not unanimous agreement that this is a step in the right direction. Stevan Harnad, an outspoken advocate for Green OA, thinks this is the wrong direction to move at this time, and that Green OA should logically precede Gold OA.)
Even if you take $3,500 as an emerging norm for an OA publication fee, based on the experience of PLoS journals, this is still a far cry from the $20,000 to $25,000 that it would cost to fully subsidize publication of a typical monograph (assuming that the costs associated with print publication amount to no more than 25% of the overall costs of publishing a book, with this being the “first copy” cost alone). So far I have seen no evidence of any university willing to take this leap from subsidizing OA faculty publishing in journals to OA faculty publishing in books. That is unfortunate as what I call the new “digital divide” between book and journal content will only grow wider as more of the latter is made available to OA while only a trickle of the former gets into that mode. And what sense does it make, intellectually, to have those two forms of scholarship so separated from each other in cyberspace? None.
What is happening is that there are more experiments with getting monographs into online systems, beyond those established by commercial entities like netLibrary, ebrary, Questia, MyiLibrary, etc., the eBook reader platforms that are beginning to extend well beyond the Kindle and Sony readers, and even Oxford Scholarship Online, which was a pioneering effort among university presses to break the logjam for getting books into the digital arena. In recent months there have been announcements by JSTOR, in cooperation with UC-Berkeley, and the Humanities E-Book Project of intentions to open their operations to become suppliers of large aggregations of eBooks. Also, Project Muse has begun a pilot project to test the feasibility of bringing monographs into its operation and making them integrated with the journal content already available. Meanwhile, five university presses have succeeded in persuading the Mellon Foundation to provide seed money to explore yet another cooperative non-profit venture in this arena, while Duke has now moved beyond beta testing of its online monographs program (using ebrary’s technology for its platform) into full operational mode. All of these proposed schemes, however, are simply non-profit variations of their commercial counterparts, and they all depend on the subscription model to sustain themselves. None is truly open access. More promising along these lines is the scholar-initiated Open Humanities Press, which a month ago announced the launching of five new OA monograph series in critical and cultural theory with the technical support of the University of Michigan Library’s Scholarly Publishing Office (which functions very much as our ODSP does). It is also at Michigan, of course, that the university rather dramatically announced that its press would in the future publish monographs online, misleading some to believe that the press would no longer provide any of its books in print form whereas, in fact, what Michigan intends to do is what we have already been doing with our Romance Studies series for four years and what Michigan’s press had already been doing for its Digital Culture books series. Give credit to Michigan, though, for pushing the envelope. Some of us presses would rather have more evidence of this approach’s economic feasibility before taking the full plunge with our entire lists. But if the Michigan administration is willing to put its money where its mouth is and support this venture even if losses mount, more power to it. Eventually, Michigan may find that it is just not worth the effort to maintain a market mechanism at all for cost recovery and just try going OA completely.
My hope is that if monographs do go the route of open access as journals already have begun to do on a large scale, it will be done with full funding so that we do not have to suffer the indignities that come with Green OA. The indignities I mean stem from the public availability of unedited though peer-reviewed manuscripts on institutional and personal Websites. Most people cheered when Harvard announced its plan to have the journal articles of its faculty in the arts and sciences posted online; I groaned. As a former copyeditor who used to edit some Harvard faculty writings, it did not come as happy news to think these scholars would be willing to have their work appear, grammatical infelicities, incorrect citations, clumsy prose, and all, for all the world to see. And it amazed me to hear Harvard administrators speak as though this would be for the public benefit and would enhance Harvard’s reputation all the more. Au contraire, I thought to myself, what better way to take the bloom off the rose than to expose the bugs inhabiting the plant, which are only exterminated when the unsung heroes of publishing, our copyeditors, do their work of restoration? Harvard’s Website has just recently been opened to the public and I am both anxious and fearful at what I will find when I take my copyeditor’s eye into its interior. Will I find, as I did as a copyeditor, that the expert on a famous religious figure will have systematically miscited the standard edition of his works, or that the authority on a 19th-century political philosopher will have badly transcribed quotation after quotation, or that so few really understand what a dangling modifier is? As much as I am fundamentally in favor of open access, I wish we could avoid the proliferation of versions of the same work, which is what is happening now in journal publishing, with many publishers (including our Press) reluctantly agreeing to Green OA just because we feel we cannot stand in the way of progress. But, at what cost, progress?
Leah was appointed Executive Director of the Charleston Conference in 2017, and has served in various roles with the Charleston Information Group, LLC, since 2004. Prior to working for the conference, she was Assistant Director of Graduate Admissions for the College of Charleston for four years. She lives in a small town near Columbia, SC, with her husband and two kids where they raise a menagerie of farm animals.