Library budgets, the ‘serials crisis’ and over-supply

Various studies based on statistics from the Association of Research Libraries show that the cost to academic libraries of subscribing to journals has outstripped inflation by over 300% since 1986.²⁵ Meanwhile, total library expenditure (i.e. budget for staff, services, technology and books) has outpaced inflation by only 79% over the same period.²⁶ While the humanities’ expenditure accounts for a smaller portion of this ‘serials crisis’ than the natural sciences in absolute terms, this rise is reflected proportionately in humanities journals.²⁷ This budgetary problem has been fuelled not only by price increases, but also by an explosion of research output over the past half-century. The effect of this serials crisis is one of the core motivators for academic open-access advocates: as their libraries are unable to afford the subscriptions, academic researchers and students at many institutions come up against paywalls that hinder their ability to conduct research and to teach/learn efficiently. This effect is not, of course, felt uniformly: those at top, prestigious and wealthy institutions may not suffer from or notice this compared with their colleagues at the poorer end of the financial scale. However, even Harvard University, one of the wealthiest institutions in the world, has claimed that it cannot afford the material that it needs.²⁸ Some, like me, believe that this demonstrates that the economics of the system are broken, while others think such a claim to be overstated.²⁹ However, what certainly is true is that those without access to a well-endowed library, such as independent researchers or those at poorer institutions, find themselves locked out of a pay-to-read system if they cannot afford the fees. There are, of course, mitigating aspects that help with this. Although slow (thereby disadvantaging those without direct access in terms of productivity), inter-library loans are one good way in which a greater number of people can read work. This seems, however, to be more of a patch that is designed to hold together a system of subscription and purchase access, rather than an attempt to address the underlying economic problems that prevent direct access.

As is detailed more thoroughly in Chapter 2, the economics of scholarly publishing are complex. For instance, it is worth considering whether true markets can exist in journal subscriptions or book sales to academic institutions and whether the profit motive stands in fundamental opposition to the goal of academic research. After all, because academic work must be original, there should be no comparable ‘competition’ to an article or book when a researcher needs it, a fact that makes it difficult to construe conventional economic markets.³⁰ When a particular article or book is necessary, nothing else will do and the researcher must acquire access to it. From this fact, in every instance the publisher has a ‘mini-monopoly’ on an article or book, as Peter Suber frames it,³¹ an aspect made possible because, unlike many other commodities, books and journal articles ‘differ in that they are not substitutes for each other’.³²

Regardless of this, the demand-side system of subscriptions and sales was working relatively well until the late 1980s when a sudden mass expansion of higher education in the Anglophone academy (and elsewhere) triggered changes to the supply side of the ecosystem.³³ Combined with a growing credentialism in academia that focused on research output, the demand to be able to publish research in often niche areas overtook and outstripped the desire necessarily to read that same research.³⁴ This led to fierce competition to publish in prestigious journals or with respected publishers as they began to act as proxy measures for hiring committees, eventually replicating the scarcity in the job market and the high number of applicants for each post.

This is to say that, as much as there is a library budget crisis in being able to afford access to all research, there is also a supply-side crisis for all researchers to be able to publish their work; an aspect that becomes ever more crucial to holding a secure academic post.³⁵ This logic sounds callous when put in relation to academic research: humanities academics are, after all, most frequently used to viewing their niche work as holding esoteric, rather than popular or even utilitarian, value. It feels grimly utilitarian to specify that problems in library budgets could be driven by the presence of too much interesting work that should be published, but this is one way of viewing the problem.³⁶ To think firstly in terms of journals, as more researchers produce material in the ever more competitive quest for jobs, the need and desire to publish in top journals is increased. Because these journals will then have a continual supply of high-quality material, it is imperative that libraries subscribe to them. As this material overflows through rejection and cascades down to the next level of ‘mid-range’ publications, libraries find that there are also far more venues to which they must subscribe, a fact triggered by over-supply (not to mention material published in niche journals that may also be necessary for research). In short, paradoxically, there are both supply-side crises (too much competition for top journal slots) and demand-side crises (institutions’ inability to afford all material for students and researchers) in academic publishing. These are split across two forms of scholarly economics into which I delve in Chapter 2: the ‘economics’ of scholarly prestige and the economics of paying for the labour of publishing.

Similar phenomena exist in the monograph sphere and these are dealt with separately in Chapter 4 because of the importance of the book form to humanities disciplines. There are some differences, however, that are worth briefly highlighting here. Monographs are more clearly the gold standard for accreditation and reputation in the humanities. They also take an order of magnitude longer to write than their journal article counterparts, meaning that scholars expect a commensurate reputational return. The print runs for academic monographs in the humanities, though, are extremely low; around 200–250 is the figure that is usually cited. This means that presses often have returns on volumes and the margins are far lower than in scientific journal publishing, for instance. Presses then find themselves caught in a double bind. On the one hand, they cannot afford to raise prices as they are competing for the same library resources that have been dented by the serials crisis. On the other hand, presses must therefore stringently limit their intake on the basis of quality control (peer review) so that their authors see a reputational gain through exclusivity (prestige) and so that they do not expend labour on titles that will not sell to their target audience. Of course, there are other models: lowering review standards (and perhaps production quality) while opting for a wider list is an alternative tactic that relies on successful titles carrying others through cross-subsidy. Broadly speaking though, there are also supply-side and demand-side ‘crises’ in the monograph world.

This is where open access plays a role, at least in one respect. OA rests on a specific economic form called ‘nonrivalrous commodity exchange’.³⁷ This form, which requires digital dissemination, is one wherein the ‘use’ of a commodity does not entail somebody else's inability to use it, meaning that the costs no longer inhere in reproducing objects but instead in the labour required to create the first copy. This is the case for digital content. When one downloads a file this will not affect the next user's ability to likewise access that content in most normal circumstances because this is essentially an act of near-instantaneous copying in which the original remains.³⁸ This differs from the past systems of paper where, for example, printing a book again and again came with a material cost in each case because in selling a copy it was given away; one cannot re-sell the physical object that one has just sold (not to mention the costs of warehousing etc.). Under this new form of nonrivalrous commodity exchange in which replication costs almost disappear, if it were possible to allow anybody access to scholarly material at no charge, covering instead the labour costs to first copy, the demand-side problem would be eliminated. The ‘if’ in that sentence is somewhat large and the supply-side problem is far more difficult to broach without budget increases or cost reductions but, broadly speaking, the problem of demand-side economics and restricted access is one to which open access could be positioned as a partial solution, predicated upon nonrivalrous commodity exchange.

Free culture, copyright and an open ethos

Alongside the explosive growth of higher education and research output has been an enormous increase in technological capacity that it is also important to consider for an understanding of the emergence of open access. Indeed, the overwhelming assumption from the literature on open scholarship is that it has co-evolved with broader technological developments.³⁹ Although much of the recent history of computing technology is ensconced within commercial and military paradigms, it is also the birthplace of the free culture movement. Counter-intuitively, this stems from the fact that the historical rise of general-purpose computing to mainstream prominence took place within commercial paradigms. From Multics, through to Unix, Windows and Mac OS, there has been a raft of closed-source operating systems that are licensed to users for corporate profit and which have formed the basis of most people's interaction with computer systems. However, in response to this corporate practice, a counter-discourse of ‘free culture’ was also born. Free culture in this context does not exclusively mean monetarily free but more often refers to freedom of action; the freedom to reuse material. This movement finds its meeting point with academia in the proposed removal of reuse barriers under open access and the modifications to the practicalities of copyright that this would require.

One of the most important figures in the history of this movement is Richard M. Stallman who, in 1989, wrote a document called the GPL (the GNU Public License) that radically redefined thinking about copyright. Copyright is, in almost all global jurisdictions, an automatically conferred, time-limited, exclusive right to distribute an original work. Copyright, which is covered more thoroughly below in Chapter 3, is the legal mechanism through which any notion of control over one's academic or artistic work comes into being. Without copyright, anybody could do anything to anybody else's work, from redistributing to altering, and there would be no obligation to acknowledge the original source. While most software licenses are designed to use copyright to restrict the end-user's freedom to modify the underlying source code and/or redistribute the program, Stallman's license reverses this, using the authority of copyright to stipulate, explicitly, that the source code for applications must be made public to allow anybody else to view, redistribute and, most importantly, modify the program. The license further specifies that anybody else's modifications to the software must be redistributed under the same terms, thereby ensuring that this freedom is extended to future users. In other words, the GPL license is ‘viral’, sticking to future works, a phenomenon which is called CopyLeft.⁴⁰

Of perhaps more direct interest to those in the humanities, Stallman argues that in the past thirty years or so a tacit understanding of copyright has been adopted that sits badly with its original intent and that is now damaging the ability of others to create new, socially useful work. As Stallman notes from a 1932 Supreme Court case in the United States:

While authors generally think of copyright as protecting their intellectual property, Stallman disputes such a stance.⁴² This is not, he would argue, the original goal of legislation, which instead posits copyright as ‘a balance between a public goal and market forces’ in which ‘the government spends our freedom instead of our money’. For Stallman, ‘[f]reedom is more important than money’⁴³ and so there is, in his reading of the history of copyright, the impetus for the state to get the best bargain for the public and not for the individual creator.

This is where the lowering of permission barriers under open access begins to intersect with a history of copyright, technology and economics. In Stallman’s reading, the goal of the compromise of copyright is to give authors a limited time-period in which to sell their works to support themselves financially. Academics, however, are not generally understood to be economically dependent upon selling their research output (I will revisit this logic in more detail in Chapter 2). The question that arises for the Open Access movement from this, taking Stallman as a starting point, is: why should academics retain the economic protections of copyright if they are not dependent upon the system of remuneration that this is supposed to uphold?⁴⁴ As is clear, this mirrors the arguments for free access that I explored above: if authors are not required to sell their work, why can't they give it away? In parallel: if authors are not required to sell their work, why do they require all of the protections of copyright and specifically those protections that exist for financial benefit? Stallman’s reading of the history of copyright is not ubiquitously held and it is not the sole reason why advocates argue for (and sceptics argue against) the lowering of permission barriers to scholarly research, but it does form one cornerstone in the movement's history.

While it is wholly possible to dispute the health of the subscription-/sales-based economic model – and it certainly contributes to the financial problems of the serials crisis outlined above – an important point to register is that even under this mode there is usually no financial disincentive to a researcher in giving away the copyright to his or her work.⁴⁵ The same is not true, under a subscription or sale system, for publishers. To demonstrate this, consider that, under the current conditions of scholarly publishing, be that books or journals, copyright is usually assigned by the academic creator to a publisher. In most cases of journal publishing this is done without financial remuneration but is traded by the academic for symbolic reputational capital and the services that the publisher can provide. In the realm of books, publishers do pay royalties to academics, but because the majority of these monographs are not runaway trade successes, it is the reputational return that is most frequently desired in this sphere also. The publisher then retains that exclusive copyright and sells the packaged commodity object (an article or book), most often back to university libraries. Under the subscription economic model, therefore, it is publishers who exercise the rights enshrined in the time-limited exclusivity of the copyright monopoly to recover their labour costs and, in some cases, to make a profit.

The labours that must be compensated in publishing and that are currently protected by such copyright arrangements are many and varied. Publishers make a living through the sale of either journal subscriptions or books, to which they claim they add value. As has been most recently framed by Michael Bhaskar, at least three of these value-adding functions are ‘filtering’, ‘framing’ and ‘amplification’.⁴⁶ While it is impossible to recapitulate his entire argument here, one of the most potent examples of the types of labour involved in these processes lies in the age-old example of publishing as ‘making public’. Is an article or a book ‘published’ if only one single copy exists and it is put on a park bench? What about the printing of hundreds of copies of an article or a book that, then, nobody reads? Truly to publish, in these cases, requires some kind of amplification so that readers will ‘hear’ the content over the proliferation of noisy demands on their time. This requires labour. The term ‘publishing’, then, hides a multiplicity of labour activities that will be covered below, in Chapter 2. It is a mistake to think that publishing is the simple placement of material online and/or to think that it is labour-free, even in the digital age.

To this end, publishers currently require the protections of copyright for their subscription and/or sale business models. Scholars do not usually require such economic protections (instead, they need reputational protections), but they do need the labour of publishers. For open access advocates who see twofold difficulties in access and restrictive permission barriers, this adds up to a chiastic economic mess. While there remains no economic disincentive to researchers in allowing others to read their work for free, access is denied to scholarly research through paywalls that are necessary to remunerate publishers. On the other hand, while there is no financial disincentive to researchers in allowing others to redistribute and, in some cases, to reuse their work with attribution, permissions remain restricted in order to protect the paywall model through time-limited copyright. Under mechanisms where publishers could claim their remuneration elsewhere, which are the economic reconfigurations implied by gold open access, advocates of this economic stance believe that problems of both access and permissions would be resolved by new business models. Sceptics, conversely, believe that the new models for such a reconfiguration are unproven and could result in irreparable damage both to the economics of scholarly publishing houses and, consequently, to the circulation and discovery of high-quality research material in a useful form.

These difficulties notwithstanding and to return to the history, Lawrence Lessig, a Harvard-based lawyer, saw this logic unfurling in the wake of Stallman's and others’ successes with the GPL and began working on a series of licenses that would allow creators of any type of content (i.e. not just software) to extend such reuse rights to others. It was in this light that the Creative Commons (CC) licenses were unveiled. Coming in a variety of types corresponding to different levels of permissiveness, these licenses – although contentious, as I will explore in Chapter 3 – are designed to allow content creators to use their copyright protections to allow others to redistribute, modify, translate and computationally analyse works, among other activities. It is Lessig's Creative Commons licenses that are most frequently used to achieve the lowering of permission barriers to open-access material, predicated on the reasoning outlined above.

Creative Commons licensing for academic material, as with the GPL, is not a replacement of copyright, but a superstructure atop it that undoes many of the provisions that stop others from using work. Many aspects of Creative Commons licensing seem unproblematic if access to research is monetarily free. For instance, if work is freely available, it makes sense to allow others to redistribute it at will. Few also seem to have a problem with the insistence in almost all CC licenses on the necessity for attribution of the original author. However, as will be seen, it is the modification/derivatives clauses that have aggravated certain parties. Opponents of open licensing fear reputational damage, which is the core currency for academics, and the erosion of academic citation norms, let alone the potential economic consequences for publishers. Yet the core questions that I will address later are, from the advocate's perspective, whether the time-limited copyright monopoly, when free of financial gain, was ever intended to be used to protect the integrity of work.

Convergences

It is at the convergence point of these two narratives – problems of supply-/demand-side economics and the birth of the free culture movement (but certainly others, also) – that open access emerges. Open access was defined in three influential documents written around the turn of the millennium: the Budapest Open Access Initiative (2002), the Bethesda Statement on Open Access Publishing (2003) and the Berlin Declaration on Open Access to Knowledge in the Sciences and the Humanities (2003). Peter Suber refers to these documents collectively as the ‘BBB definition of OA’.⁴⁷ All three of these definitions outline the need for ‘user’ freedom beyond simply being able to read (i.e. they specify the lowering of permission barriers) but all three also specifically enshrine attribution of the author at the heart of their principles.

Over the last decade, OA has grown exponentially. Open Journal Systems, the free software project started by John Willinsky that provides the software needed to set up an OA academic journal, has clocked up over 11,000 worldwide installs. Various institutional, funder and national-level mandates have also been put into effect.⁴⁸ There have been legislative hearings in the UK, the States and elsewhere on open access. There are current panels around the world trying to ascertain strategies for OA monographs and projects working on new experimental business models for OA publication. Whatever else can be said of OA: governments want it, a portion of the research community wants it, some members of the public want it, funders want it, and it seems extremely probable that it will meet with widespread global adoption in some form over the next few years. What I hope this brief history has shown, though, is that the conditions of contemporary scholarly communication/publishing have been shaped by both legislative and financial mechanisms but that, from technological origins, a way of thinking arose that countered these norms. It is also clear that these technological and scientific origins pose a seemingly obvious follow-on question: if open access was born in the sciences and amid technologies, is OA, then, just for scientists? There are clear advantages for scientists in quickly and openly disseminating research and it has certainly been mooted that the humanities are being led to open access as part of a worldwide science-driven policy agenda. As will be seen, the answer to the question of whether OA is specific to the sciences is more complex than a simple ‘yes’ or ‘no’.

Open access: scientific origins

When the origin of the Free Culture movement within technological disciplines is coupled with the exponentially more challenging economic situation for scientific journals, a history of open access emerges that is firmly situated within the sciences. This is sometimes taken by sceptics to mean that the humanities should be excluded from funder requirements for open access because there are fundamental differences between these disciplines and the natural sciences. Disciplinary differences cannot be elided but, at the same time, there are ways in which open access could benefit, or at least alter, research practice for the humanities. To investigate this further, I want now to look at the differences a little more in terms of origins and policies to ascertain the exact points of tension.

It is true that a substantial degree of the development of open access has taken shape in the sciences. In fact, high-energy physics seems to be one of the strongest disciplines for OA. It is also wise to be cautious of the fact that the motivation of many governments pursuing open access is to allow industry to take the fruits of (often public) scientific research and to re-enclose it for commercial benefit. Depending upon your degree of market-orientation, this may or may not be a negative phenomenon. One's appraisal may also depend upon the type of enterprise appropriating the work; some would be happy with charities, for instance, using their work but not large corporations. Truly libre open access does not discriminate, however, on the type of activity undertaken by the reuser. This is a topic that will be explored in more detail in Chapter 3.

Despite the seemingly science-centric history of OA, however, open access also has both formal and impromptu roles and histories in the humanities. In terms of formality, the already-mentioned Peter Suber, a philosopher specialising in law, epistemology and ethics, was a principal drafter of the Budapest Open Access Initiative statement (the first formal statement on OA). Likewise, Jean-Claude Guédon, a professor of comparative literature at the University of Montreal, joined Suber in signing one of the other crucial formative documents of the Open Access movement, the 2003 Bethesda Statement on Open Access Publishing.⁴⁹ In this way, those working in the humanities were represented, even if under-represented, at the birth of the OA movement. That this representation of the humanities disciplines was proportionately minimal can be accounted for through two potential explanations: (1) that humanities disciplines are so distant from the sciences in the way in which they communicate research as to obviate the need for, or possibility of, open access – a line taken by some sceptics; or (2) that those working in the humanities have been less engaged in a critique of their own publication practices than the sciences and simply lag behind – the advocates’ stance. These are, of course, merely the most extreme poles. In reality, responses sit on a spectrum between these points.

Even if the majority were not part of these formal histories, however, this does not mean that people working in the humanities have not, in smaller clusters, just ‘done’ open access. In fact, excluding publisher-initiated moves, for those academics who opt for a Do It Yourself approach to starting new digital initiatives and journals, open access seems to be the default. Take, for the basis of an utterly non-systematic and far-from-inclusive survey, 19: Interdisciplinary Studies in the Long Nineteenth Century, an OA journal running since 2005; the NINES (Networked Infrastructure for Nineteenth-Century Electronic Scholarship) system; the Journal of Ethics & Social Philosophy (since 2005); Gamut: Online Journal of the Music Theory Society of the Mid-Atlantic (since 2008); Foucault Studies (since 2004); Culture Machine (since 1999) and other Open Humanities Press journals and books; Digital Humanities Quarterly (since 2007); Journal of Neo-Victorian Studies (since 2008); Open Book Publishers (since 2008, an open-access book publisher founded by scholars at Clare Hall and Trinity College, Cambridge); Punctum Books (since 2011, another scholar-led book project founded by Eileen Joy and Nicola Masciandaro); the list goes on. While such efforts are clearly significant, it is also evident that these scholar-led initiatives have not come to dominate the publishing landscape. There are also, on occasion, problems with their lack of publisher expertise. For instance, sometimes these venues do not have adequate digital preservation mechanisms to militate against disaster or do not adhere to commonly agreed upon standards for discoverability.

What these projects seem to indicate, more precisely, though, in their very existence, is that when academics design systems for communication from the ground up, independently, the idea of creating a subscription or sale model often doesn't seem to enter the equation (although it is more common as a form of supplementary income for OA book initiatives). Sceptics might point out that this is indicative of a fundamental unsustainability of these online academic resources; a lack of business sense. To my mind, however, it also signals that there cannot be an irreconcilable difference between the communication practices of the humanities and open dissemination on the internet – ‘open access’ – but that differences in research dissemination practice to the sciences must be found somewhere else, especially (but not only) in the journal sphere.

*

Even if, then, there is no fundamental tension between open online dissemination and the form of much humanities work, it is nonetheless no simple task to define precisely the role of research in the humanities subjects, as apart from the sciences. As Helen Small puts it, isolating distinctive traits of specific intellectual cultures is surprisingly difficult and even seemingly basic definitions very quickly become convoluted.⁵⁰ In Jerome McGann's formulation, however, the role of the humanities disciplines is ‘to preserve, to monitor, to investigate, and to augment our cultural life and inheritance’ and, in straightforward parlance, it is clear that those employed in universities’ humanities departments conduct ‘research’ in the service of these goals.⁵¹ It is also clear, though, that a single definition of research will likely be too loose to cover adequately the diversity of activities found within even this narrowed disciplinary spectrum. With that warning in mind, what does ‘research’ mean in broad strokes, what is its function and what does open access do to research activities in these disciplines?

Fundamentally, research – as opposed to ‘scholarship’, at least in some definitions – can be specified as the practice of either (1) discovering new aspects or interpretations of reality/cultures and communicating these findings; or (2) refuting previously communicated findings. This definition remains true across the sciences and the humanities and seems unaffected by methodological differences.⁵² In terms of differences, it is certainly the case that in the practice of literary criticism, for just one example, it is rare to work on the model of a hypothesis followed by a controlled experiment. It is also true that few scientists would adopt the practice-based research methodologies seen among live-art researchers, for instance. That said, although different methodologies in the humanities and sciences may be thought of as respectively more subjective or objective, each is also concerned with fostering intersubjective understandings through repetition, whether that be through persuasive argument or hypothesis-driven experimentation. As Hannah Arendt put it, ‘The reality of the public realm relies on the simultaneous presence of innumerable perspectives and aspects … Only where things can be seen by many in a variety of aspects without changing their identity can worldly reality truly and reliably appear.’⁵³ It is also worth noting that the interpretation of data in many parts of scientific practice relies on mediated subjective thinking and interpretation: the data do not speak for themselves.

None of this is to elide the very real differences in practice between the humanities and the sciences; the problem has not been referred to as the ‘Two Cultures’ for no reason.⁵⁴ It is instead, however, to note that the fundamental bases of conducting research are the same: the discovery and communication of new findings or interpretations. Furthermore, the ways in which such research is ‘used’ by others also has many overlapping characteristics between disciplinary fields, despite the problematic rhetoric of ‘utility’. For instance, the uses of preceding work through citation in both the humanities and the sciences remain broadly the same: (1) to inform the reader of the existing body of work upon which the new research rests, along with its applicability to or difference from the new material; (2) to refute existing work when inaccuracies of fact are alleged or disagreements over interpretation have arisen; (3) to credit the preceding work and author(s) with value and novelty or to discredit through dissent; (4) to create a chain of verification whereby the claims upon which the new work rests can be checked. It is worth unpacking these statements so that some of the purposes of scholarly communication and ‘publication’ of research can be defined for this discussion.

The first of these points is, essentially, informational about the extant literature. Those new to a field, if presented with no preceding research, would be likely simply to replicate existing findings and arguments; their work would not be novel. By requiring reference to existing literature, a network of citations is slowly built that acts as a map of the field. The newcomer can quickly gauge the central points of a particular field by cross-correlating citations. Furthermore, the importance of an author's works within particular niche sub-areas is revealed when a work is more frequently cited within such an area (although caution towards such a quantified ‘citation counting’ approach should be urged as measuring worth in such numerical terms is of dubious applicability to the educational enrichment provided by the humanities). In those disciplines where artistic practice forms a part of research and the output is, therefore, more akin to (or even is) art itself, reference to others serves to contextualise the work; it provides a constellation of other work within which the piece under discussion can be situated, read and understood.

The second point, of criticism and refutation, is one of public communication. As arguments or facts come under scrutiny, debate between scholars is committed to record and new understandings emerge from communication. While arguments in academia can often seem petty, the amount of research effort that goes into verifying findings (in archival work, for instance) and constructing viable arguments is substantial, so it is of little surprise that a lot is at stake for academics in these debates (Sayre's law notwithstanding).⁵⁵ Few would dispute, though, that the majority are motivated by intellectual curiosity and truth rather than malice in publicly refuting another's work.

The third point is one of reputation and novelty.⁵⁶ While this is not strictly necessary within a totally idealistic system under which people might work solely for the benefit of truth with no personal, ulterior motive, this is not the world that exists. Scholars’ reputations are positively founded and can negatively founder upon the basis of an idea that does or does not gain currency. This forms a crucial part of individualist contemporary systems of motivation and career accreditation and is enshrined in the legal and academic enforcement systems surrounding plagiarism (see Chapter 3 for more on this).⁵⁷

The final of these points, of verification, is the most important. In order to check the truth of a particular piece of work, when it is built on the foundation of others, one must be able to go back to appraise the current use of existing academic source material. As Anthony Grafton has put it in his history of the footnote, ‘the culturally contingent and eminently fallible footnote offers the only guarantee we have that statements about the past derive from identifiable sources. And that is the only ground we have to trust them.’⁵⁸ Without such an ability, one must simply place faith in the author not to have misrepresented, misattributed, misread or even misunderstood the piece that he or she is citing. With the best generosity in the world towards the character of fellow scholars, it is simply uncritical not to follow such practice in checking the assertions of others.

There are undoubtedly other areas of practice, but these seem to constitute at least some of the essence, and use values, of research in the humanities. Some of these are directly shared with the sciences. Of course, it is also clear that research in the humanities may not be purely ‘used’ in a practical sense and frequently exists to inform without a clear applicable use, at least in the sense of a market economy. However, within the spheres where use can be identified, it is important to consider how open access impinges upon each of these areas.

Taking these four points as a starting guide, it becomes possible to identify some of the ways in which open access interacts with research principles in the humanities. The first three of these notions – reporting upon preceding literature; refuting existing work; and crediting the preceding work and author(s) with novelty and value or discrediting through dissent – remain relatively unchanged in an open-access environment, although it is worth noting that various studies show that OA papers may be more widely cited, thus enhancing the citation map.⁵⁹ It may also be easier for researchers to undertake these activities (reporting on preceding literature etc.) if they have immediate, online, free access to work. In terms of value and credit, it is also important to remember that just because research is openly available to all (like the majority of websites on the world wide web) this does not mean that there are no quality controls (unlike the majority of the web).⁶⁰ Likewise, the need to cite the work of others remains in place.

The fourth of these points, however, the idea of creating a chain of verification whereby the claims upon which the new work rests can be checked, is potentially significantly enhanced in an open-access world. Although checking others’ use of sources is currently a far less common practice than might be hoped, if all research were open access and the necessary technological infrastructure was put into place, an environment could exist in which this kind of checking could be instantaneous: a linked click. Of course, much humanities writing requires a more totalised understanding of the work than just a link to a single paragraph – it requires the argument, the aesthetic and the context – but this does not impinge upon the potential supplementary benefits of such a system. This could be available not only to those established within universities, though, but rather to anybody with access to the internet. This could range from independent researchers through to those fresh out of their degree. In much the same way as it becomes easy to spend hours following links that look interesting on Wikipedia, a world could be possible where the same is true of an interlinked network of high-quality scholarly papers. Of course, just because OA might offer the possibility of such a system existing does not mean that it would spontaneously burst into existence; new publisher labour would be required to implement the linking, format migration and any supplemental technologies that might facilitate this.

Beyond the broader goal of access to research, such thinking as this represents the kind of benefits that advocates believe would enhance research practice and is among the reasons why OA is not purely applicable to the sciences. However, not all scholars and commentators feel that these changes are positive and they often point out that current research practices have evolved over a long timeframe while, by contrast, the switch to digital is happening (too) rapidly. The same critics also often feel that, even if these types of benefits were acknowledged (on top of the more general principles of broader access, equality etc.), the damage of a transition period is too economically dangerous to be practical. To that end, the next section of this chapter turns towards the reasons for dissent, especially in the humanities.

Academics’ oppositions to open access

As in every other stakeholder group, and as has already been seen, there are a number of academics in the humanities who totally support OA, even in its most liberal forms. There are also, however, two dissenting camps. The first group of academic dissenters support the principles of OA but object to the specific implementations that have been proposed, including concerns for the continued viability of humanities’ academic research labour as an activity. Those in this group might support only the green route, for example, or require more restrictive licensing. The second group object to the principle of open access in its entirety.

The latter standpoint – a seemingly complete objection to open access – was most explicitly set out in recent times by Robin Osborne, a Cambridge-based Professor of Ancient History, who argued that:

Osborne's basic arguments are: that ‘[i]f there are fees for access to teaching there should be fees for access to research’;⁶² that the publications that result from a research project ‘are only trivially a result of the research-funding’ provided by the broader public (and hence should not be subject to funder requirements to make this work available to taxpayers);⁶³ and that OA makes no sense because those who wish to have access ‘must invest in the education prerequisite to enable them to understand the language used’.⁶⁴

This book chapter, written by Osborne for the British Academy, caused some furore among OA advocates on its publication for obvious reasons. On the one hand, the thrust of Osborne's argument is clear: humanities research must be seen as more than simple Gradgrind-esque facticity and the provision of utilitarian databases for consultation. Osborne is correct that this is not the purpose of work in the humanities. On the other hand, though, this particular argument raises three specific questions in opposition: why should fees to access teaching (which is not the situation in every country, anyway) entail fees to access research, when teaching is that which provides the prerequisite to understanding research? If (taxpayer⁶⁵) funding only contributes such a small amount to the overall gain in knowledge and efforts of a project, should those in the humanities receive it?⁶⁶ Finally, are there not a large number of humanities graduates who do not work in the university but who would be able to understand this work? Osborne's argument is among the more extreme of objections to open access, even if rooted in a fairly accepted view of the function of humanities research; most arguments, as I will go on to discuss, take issue with specific aspects of open access implementation.

In the more moderate camps, there are several groups of academics who have objected, not to the basic principle of open access, but rather to the way in which it is to be implemented through article or book processing charges (for the removal of price barriers) and the various aspects of reuse permitted by the Creative Commons Attribution (‘CC BY’) license (the removal of permission barriers). I will try to give a flavour of these here.

One of the clearest statements with regard to the former stance was voiced by the editors of the Review of African Political Economy, who believe that ‘The potential gains of OA are fundamentally undermined by the “pay to publish” principle.’⁶⁷ Likewise, the Editors of History Journals’ written submission to the UK's Business Innovation and Skills (BIS) Select Committee Inquiry in 2013 noted that they were ‘concerned that the international reputation of UK journals is likely to suffer if scholars abroad begin to believe that they will have to pay to publish in UK journals’.⁶⁸ For now, I will let these statements stand on their own as concerns but will also note that in Chapter 2 below I address alternative business models for gold OA that may mitigate these problems. I will also note that there are no barriers of this sort to green OA which comes with no cost to authors. In other words, these objections were specifically addressed to the article/book processing charge implementation of gold open access. In turn, there is also concern that the pay-to-publish principle compromises academic freedom by tying finance to publication.⁶⁹ These economic worries are compounded by concerns that the destabilisation engendered by a switch to new models may hinder the ability of publishers to continue to operate as venues for the promulgation of research.

Academic objections to the lowering of permission barriers to humanities research have taken two different forms, which are explored much more thoroughly in Chapter 3. The first of these is an authorial worry over the potential corruption of scholarly integrity. In late 2013 this was aired by the Cambridge historian Peter Mandler in an article for the Journal of Victorian Culture. Mandler noted that, in the humanities, ‘Our form of words is unique to us and it cannot be dismembered and mixed with the words of others – which CC BY [the Creative Commons Attribution license] facilitates – without yielding what we tell our students is plagiarism: the mixing of their words and our words without specifying (through quotation) which is which.’⁷⁰

The second set of objections over permissive licensing of humanities work is made most forcefully by John Holmwood, whose concerns come in the wake of the UK government's decision to confer degree-awarding powers to entities that conduct no teaching, such as Pearson.⁷¹ The worry here is that, at present, academics have established a system of legitimation based upon community validation of standards. External bodies that seek to set examinations for higher education while not teaching themselves could financially undercut the research university by providing high-quality research material free of charge (through open access), followed by an examination of students upon that material. This would present a severe challenge not only to academic autonomy but to the continued financial viability of the research university.⁷² These concerns echo the political split that I outlined at the very outset of this book. Holmwood argues that:

Furthermore, according to Holmwood, the humanities and social sciences are also at risk:

Concerns such as these, centred in UK policy, cannot be wholly dismissed out of hand. They also have global consequences as the worldwide move to a financialised university continues. The most probable solution to this, in my mind, is to append a ‘share-alike’ clause to the license, as I outline below in Chapter 3.

Commercial publishers’ oppositions to open access

As has been noted, a substantial portion of the most recent wave of controversies surrounding open access arose in response to the national-level mandates enacted in the UK in 2013, and the prominent representation of commercial publishers in this debate gives some good evidence of the core concerns of this group. Although these are not solely related to the humanities, the issues that were raised at this point, even in a scientific context, form an important backdrop to an understanding of the different camps. This is because some of the clashes aired at various moments in this inquiry were between OA advocates, a market-orientated government panel and selected traditional scholarly publishers. Broadly speaking, as business entities, the primary concern of many commercial scholarly publishers pertains to the sustainability (or profitability) of their activities under new, untested, economic models for open access.

One of the oft-touted arguments by left-spectrum OA advocates is that commercial publishers extort their captive library clients.⁷⁴ At the parliamentary hearings, this was most explicit in the questioning of Alicia Wise, representing Reed Elsevier, a scientific publishing company known for its vocal opposition to, and legal lobbying against, OA in the States.⁷⁵ Wise confirmed that Reed Elsevier reported a 37% profit with ‘a revenue stream of £2.06 billion and a profit level of £780 million’ in 2012.⁷⁶ In the face of such evidence, it seems difficult to fault the argument of advocates that at least some resistance from these entities must come from the fact that they fare extremely well under the current subscription model. Advocates point out that the profit margins of major oil companies, for comparison, sit at around 6.5%. Big Pharma usually manages about 16%.⁷⁷ It was, in fact, specifically Elsevier's profit margins that triggered the Cambridge mathematician Timothy Gowers’ call for open access through a boycott of the publisher.⁷⁸

However, it is important to stress that not all commercial publishers are alike, particularly when speaking of the humanities. Indeed, there are many small, independent humanities presses with no profit-driven agenda who can only dream of Elsevier's margins. University presses, covered below, are also ‘commercial’ in some senses but often have different mission statements and levels of profitability; specifically, an obligation to publish on the basis of quality. Many commercial presses also doubtlessly act out of a motivation to facilitate scholarly communication. Some of these publishers, therefore, dissent from open access not because they will lose massive profits, but rather because they fear that their business model will collapse under OA and that the labour needed to support their mission will no longer be viable.

This important point aside, there are also, though, both mega-publishers who operate in the humanities sphere and ongoing campaigns of top-loading acquisitions of smaller publishers by large conglomerates that span the humanities and the sciences (take for instance the fact that Palgrave Macmillan is a sister company of Nature publishing group). One of these latter types seems to be Taylor & Francis/Routledge. This publisher, known to humanities scholars for its range of journals and book publications, has, in recent times, begun seeking the views of its scholar-base on open access. However, advocates have charged that the methodology of their surveys betrays an implicit bias against OA through leading poll statements on topics such as: ‘Open access journals are lower quality than subscription journals.’⁷⁹ Certainly, this publisher has a vested commercial interest in the subscription system. In terms of turnover, Informa Group, which owns Taylor & Francis and Routledge, posted a ‘record’ adjusted operating margin of 28.4% in 2012 with a £349.7m adjusted operating profit;⁸⁰ 38% of this came from Informa's publishing revenue, including their humanities division, which was, in the words of their annual shareholder report, ‘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.’ For publishers thriving in this environment, regardless of whether this limits those who can read research work, to use their own words: ‘It is a uniquely attractive mode.’⁸¹ With statements like these, it is hardly surprising that there is opposition to new models that might make it harder to achieve those margins and to continue a year-on-year 10% increase of dividends to shareholders.⁸²

The list goes on. Bloomsbury Academic is a humanities and social science publisher that seems to be using its trade success to buy up other academic publishers who are in the black, such as Continuum, an organisation that had itself previously acquired Cassell, and also T&T Clark, Berg Publishers, Methuen Drama, Arden Shakespeare, Bristol Classical Press, Fairchild Books and AVA.⁸³ In its 2012 financial report, the company noted an adjusted continuing profit of £12.1m on a continuing margin of 12.4%.⁸⁴ Of Bloomsbury's activities, ‘The Academic & Professional division grew the most year on year with a £2.9 million increase in continuing adjusted operating profit, due to both the acquisition of Continuum [a solely humanities-orientated publisher] and a significant increase in income from content licensing deals.’⁸⁵ Interestingly, the original intention of the Bloomsbury Academic imprint, when it was launched in 1998 under the stewardship of Frances Pinter (whose Knowledge Unlatched project is covered below), was one in which the ‘new publishing model [would consist] of releasing works for free online through a Creative Commons or other open license, and then offering print-on-demand (POD) copies at reasonable prices’.⁸⁶ This never came to long-term, mass-scale fruition, possibly because of fears for ongoing revenue. However, the publisher does continue to publish some books in an open-access form.

Now, it must be noted that there are advantages to having such financially healthy publishers. For one, especially in book markets, it allows commissioning editors to take risks with the list that would be untenable were they wholly reliant on a financial return on every single volume published. While, in a post-Friedman world, it seems that shareholder return must always be a chief concern of these entities, publishing is not a science and it is often impossible to know which volumes will succeed, thereby entailing the need for calculated risk.⁸⁷ Secondly, as John Thompson notes, good non-academic editors with strong disciplinary knowledge are valued by the academics they are serving.⁸⁸ It is only through having corporate entities with sufficient capital to retain staff that this remains possible, which, albeit in the university press sector, was the driving logic behind the accumulated surpluses of Cambridge and Oxford University Presses in the 1980s and 1990s, according to Thompson.⁸⁹ In another sense, though, it is problematic to have this tie to profit/surplus margins. It firstly discredits legitimate oppositions to aspects of open access when voiced by entities with (sometimes large) financial stakes in the current model. Secondly, there is the problematic nature of competition (the ‘mini-monopoly’) in scholarly publishing that seems oddly placed to function within a competition-centred, free-market environment. Thirdly and finally, there is the additional difficulty, at present, of the way in which academics are shielded from the economic consequences of their publishing decisions and how this benefits commercial organisations. In fact, advocate publishers in the sciences (such as PLOS – the Public Library of Science – originally a scientist-led enterprise that now runs the largest journal in the sciences) describe the present subscription scenario as a ‘systemic market failure’ because ‘authors have no price sensitivity when they choose a journal in which to publish’.⁹⁰ By this they mean that the price of the library subscription to a journal, or cost of a book, is usually not considered by researchers in their choice of where to publish. Of course, if one is to accuse subscription/sale-based publishers of having an economic motivation for perpetuating one model, PLOS has a similarly strong motivation from the obverse, open-access perspective.

This lack of researcher awareness of the price of journals is, though, a type of academic freedom: it is freedom from price impinging on the selection of where to publish. Whether this liberty is a positive force could be debated. How important is it for the communication of research (rather than for the career of the researcher) in the digital age that academics have the ability to choose exactly where to publish with limited financial awareness? Even if it remains important now, is it possible to foresee a time when it might not be? This phenomenon also represents, however, the way in which academics do not necessarily make for good rational market actors in this scenario. In the same way as students may not be rational agents in their choice of a university where they pay fees (what are their comparators if they've never been before and may never go again?), academics choose to publish on the basis of prestige (a form of symbolic capital) and often do not know (or care), in a subscription/sale environment, how much their choice of journal or book venue costs.⁹¹ This is due not just to publisher bundling practices and non-disclosure agreements, although these contribute, but rather to the fact that academics publish and consume but it is libraries that purchase. As non-rational actors will not get the best deal for themselves under market logic, this type of academic freedom – the freedom from knowing/directly bearing the financial consequences of one's actions – happens also to benefit commercial entities. Conversely, though, even the supposedly rational actors – libraries, that is – cannot get a good result out of this situation. Indeed, at least in the sciences, it has been shown that price does not correlate with quality but rather that ‘libraries typically must pay 4 to 6 times as much per page for journals owned by commercial publishers as for journals owned by non-profit societies’.⁹² Of course, commercial publishers are not always the stereotypical villains that some OA advocates make them out to be and so some of the arguments set out below on the oppositions of university presses, learned societies and even academics will also apply to commercial publishers. That said, at least a few of the oppositions of commercial publishers to OA, where they have been voiced, must be on grounds of finance, whether that be profit or sustainability.

University press and learned society publishers’ oppositions to open access

That this certainty of income stream has benefits can be seen in a different group of publishers: those of learned societies and university presses (UPs). While both can technically also be called ‘commercial’ in various senses, they often have very different agendas and histories and, where they object to open access, it can be for very different reasons. It is worth noting, upfront, that there are UPs and learned societies who have wholeheartedly embraced open access. The Linguistic Society of America (LSA), for instance, launched the open-access journal Semantics and Pragmatics in 2007. This journal does not levy article processing charges and instead receives sponsorship from the MIT Library, the MIT Department of Linguistics & Philosophy, the MIT School of Humanities, Arts, and Social Sciences and the University of Texas at Austin, Department of Linguistics.⁹³ In fact, as of April 2014, the ‘Societies and Open Access Research’ catalogue run by Caroline Sutton, Peter Suber and Amanda Page gives a list of 868 societies publishing 827 full-OA (i.e. not hybrid) journals.⁹⁴ Likewise, several smaller US-based academic presses have embraced OA. Amherst, Michigan, National Academies, Australian National University and Penn State all have affiliated presses who have expressed support for open access. In the case of Amherst's new enterprise, this is due to the belief that ‘when some presses … disseminate free literature, everybody (including those who run libraries) will enjoy access to that literature’.⁹⁵ Likewise, Patrick Alexander of Penn State UP notes that ‘Penn State University Press is philosophically and practically in favor of open access.’⁹⁶ Most interestingly, National Academies Press has provided free, full-text copies of its books online (open access) alongside priced print editions since 1994.⁹⁷

This positive stance on OA is not one that is universally shared among these entities and related university presses. Although there has been a shift of opinion in favour since 2013,⁹⁸ there have been objections from learned societies to the removal of both price and permission barriers.⁹⁹ The most frequently reiterated argument made by learned societies is that they depend on journal subscription revenue to fund their other activities. Indeed, a recent survey of thirty-three learned societies, across all disciplines, concluded that ‘Learned societies overwhelmingly agree that Open Access will inevitably place some learned societies’ journals into financial jeopardy.’¹⁰⁰ This is clearly a recognised problem as Jisc (formerly the UK's Joint Infrastructure and Services Council) commissioned research into the impact that a change to an open-access model would have upon learned societies’ business models. This report – which was commissioned in 2005 and is, therefore, somewhat out of sync with more recent developments – noted that while learned societies were interested in open-access models, at that point none of the publishers could see substantial cost advantages in the move to an OA publishing model and the majority highlighted the extra costs incurred in administering and collecting article processing charges.¹⁰¹ Although this report is now nine years old and suggested investigations of ‘pay for print’ and ‘freemium’ models that will be covered below, the arguments remain mostly unchanged today. This can be seen in the fact that Dame Janet Finch, who chaired the UK parliamentary report on OA that bears her name, warned of the implications of her own policy recommendation in early 2013, stating that there is ‘no doubt’ that some learned societies will face ‘some difficulty finding a business model that will work’.¹⁰² While a series of new business models are outlined below that might be applicable to learned societies, I will close these remarks only by noting that advocates feel that, in most cases, OA seems directly in line with the mission statements of societies to further and spread knowledge. If open access is to become widespread, however, those societies who do exist off subscription revenues have a potentially difficult road ahead to transition to alternative sources of revenue.

University presses are, again, another distinct group, but also one that has different sub-groupings. While there is, therefore, a need to speak broadly, it is also important that these differences should not be elided. That said, a good indication of UP interest to date has come from membership of the OAPEN-NL and OAPEN-UK projects. These projects, covered in more detail in the chapter on monographs below, have been investigating the effects of, and business models for, making monographs open access. While OAPEN-NL was proposed by Amsterdam University Press, OAPEN-UK has had Liverpool University Press and University of Wales Press as long-standing participants since its inception in 2011, with Oxford University Press joining in 2013.¹⁰³ From this involvement, especially in the monograph sphere, UP objections to open access can broadly be inferred as a conjoined worry about business models, concerns over unknown side effects and a lack of author demand for the mode. While noting that certain university presses are compelled to return their surpluses to their affiliated institution (particularly in the case of Oxford), on the whole, university presses are less vocally opposed to OA on profit (as opposed to sustainability) grounds as they do not have shareholders but rather exist to serve the academic community in the furtherance of their mission statements. As with learned societies, objections from university presses are, as a result, more often aligned with academic concerns concurrent with some clear worries about sustainability.

Librarians’ oppositions to open access

Finally, there is a substantial movement within the library community that favours open access. After all, librarians bear the brunt of frustration when researchers cannot access material, an aspect potentially solved by OA. Where objections exist, the primary anxieties raised by librarians with regard to OA concern the future status of the library. If a library no longer owns a collection, then what is its function? What is the role of a librarian in this new world? The answer that has been circulating at most recent library conferences has been a suggested move from ‘collecting to connecting’¹⁰⁴ – meaning that the library becomes a place that helps curate and find material. In some senses this is a return of the subject librarian, with an additional fresh role in digital preservation and access via institutional repositories. Of course, it is unclear whether these aspects might also be subjected to disintermediation by commercial entities in the future; what is to say that Google might not feel itself better placed to take this role?

There is a contingent of librarians whose constituents remain sceptical, however. The foremost of these figures is Jeffrey Beall. Beall is most widely known for his curation of a list (‘Beall's List’) that is designed to expose predatory open-access ‘publishers’. These predatory entities have disreputable review procedures and solicit material solely to collect article processing charges (thereby failing to filter material adequately for their supposed audience). While his curation of such a list is a valuable service, detractors feel that Beall should have done more to point out that the same is true of some types of ‘predatory’ publishers who work on the sale/subscription model, an aspect most clearly demonstrated by mass emails sent after conferences to solicit material for edited collections.¹⁰⁵ This finally spilled over into a full-scale revision of Beall's motivations when, in late 2013, he published an article that accused the OA movement of being an ‘anti-corporatist’, extreme-Leftist outfit ‘that wants to deny the freedom of the press to companies it disagrees with’, a radical opinion that separated Beall even from the usually conservative Scholarly Kitchen site (a popular weblog on scholarly communications known for its general scepticism towards the viability of open access).¹⁰⁶ Beall's article was not well received and sparked a series of responses that both decried the lack of evidence in his piece and noted that such a stance was unusual among librarians. One article even suggested that a ‘Randian worldview’ was the most plausible explanation for his stance.¹⁰⁷ That said, there are, of course, anti-corporatists who support OA (and, naturally, ‘anti-corporatist’ need not be a McCarthyite pejorative term) but there are also a large number of corporate publishers who do likewise. Where there are library-rooted objections to OA, then, they can intersect with concerns about quality and finance, but are also usually also related to anxieties surrounding the future role of the library and Beall remains a marginal, albeit loud, voice.

*

These form the core points of dissent with respect to open access in the humanities: fundamental objections to the principle; objections to specific implementations (including article or book processing charges and open licensing); objections on corporate-economic grounds; and objections around the future of the library. More broadly, this chapter has covered three specific background elements to open access: the history of the movement; the exceptionality (or otherwise) of the humanities; and the aforementioned objections. It should be apparent, I hope, to all readers that while open access is a theoretically simple idea, that simplicity hides a multitude of complexities. Whether these be financial, scholarly or even political, it is clear that OA is caught between stakeholders with a variety of motivations and levels of power. While these debates continue to rage, they can only be understood in totality through detailed examinations of the contexts within which they take place. In order to do so, over the next three chapters – on economics, open licensing and monographs – I further explore the terrain in the hope of more accurately charting the phenomenon of open access.